Table of Contents
This appendix lists the changes from version to version in the MySQL source code through the latest version of MySQL 5.0, which is currently MySQL 5.0.84. Starting with MySQL 5.0, we began offering a new version of the Manual for each new series of MySQL releases (5.0, 5.1, and so on). For information about changes in previous release series of the MySQL database software, see the corresponding version of this Manual. For information about legacy versions of the MySQL software through the 4.1 series, see MySQL 3.23, 4.0, 4.1 Reference Manual.
We update this section as we add new features in the 5.0 series, so that everybody can follow the development process.
This section documents all enhancements, changes, and bug fixes made to MySQL Enterprise Server and MySQL Community Server.
Releases in MySQL Enterprise Server are divided into the following release packs:
Rapid Update Service Packs are issued once a month and incorporate all the bug fixes and security updates introduced since the previous MySQL Enterprise Server release. A single Service Pack can be used to update MySQL Enterprise Server; it is not necessary to install intervening service packs to bring your system up to date.
Quarterly Service Packs are issued each quarter and incorporate all the bug fixes and security updates introduced up to the Rapid Update that the QSP it is based on, and possibly some critical bug fixes and security updates from later releases. A single Service Pack can be used to update MySQL Enterprise Server; it is not necessary to install intervening service packs to bring your system up to date.
Hot-fix releases incorporate fixes for bugs that caused significant issues that are not released as part of a Service Pack.
Note that we tend to update the manual at the same time we make changes to MySQL. If you find a recent version of MySQL listed here that you can't find on our download page (http://dev.mysql.com/downloads/), it means that the version has not yet been released (and will normally be marked so in the appropriate Release Note section).
The date mentioned with a release version is the date of the last Bazaar commit on which the release was based, not the date when the packages were made available. The binaries are usually made available a few days after the date of the tagged ChangeSet, because building and testing all packages takes some time.
For information on how to determine your current version and release type, see Section 2.2, “Determining your current MySQL version”.
The manual included in the source and binary distributions may not be fully accurate when it comes to the release changelog entries, because the integration of the manual happens at build time. For the most up-to-date release changelog, please refer to the online version instead.
The following list summarizes what has been done in the 5.0 tree. For a full list of changes, please refer to the changelog sections for each individual 5.0 release.
VARCHAR and
VARBINARY columns retain trailing
spaces. A VARCHAR() or
VARBINARY column can contain up
to 65,535 characters or bytes, respectively.
Strict SQL mode, which causes an error instead of a warning when inserting an incorrect value into a column. See Section 5.1.7, “Server SQL Modes”.
A new fixed-point math library supports precision math,
resulting in more accurate results when working with the
DECIMAL and
NUMERIC data types. For details,
see Section 11.13, “Precision Math”.
Basic support for views. See Section 18.4, “Using Views”.
Basic support for stored procedures and functions (SQL:2003 style). See Section 18.2, “Using Stored Routines (Procedures and Functions)”.
Basic support for triggers. See Section 18.3, “Using Triggers”.
Basic support for read-only server side cursors. For information
about using cursors within stored routines, see
Section 12.8.5, “Cursors”. For information about using cursors
from within the C API, see
Section 20.9.7.3, “mysql_stmt_attr_set()”.
Support for SELECT INTO
, where the
variables can be a mix of global and local types. See
Section 12.8.3.3, “list_of_varsSELECT ... INTO Statement”.
MEMORY (HEAP) tables can
have VARCHAR columns.
When using a constant string or a function that generates a
string result in CREATE ... SELECT, MySQL
creates the result column based on the maximum length of the
string or expression:
| Maximum Length | Data type |
| = 0 | CHAR(0) |
| < 512 | VARCHAR( |
| >= 512 | TEXT |
Removed the update log. It is fully replaced by the binary log.
If the MySQL server is started with
--log-update, it is translated to
--log-bin (or ignored if the
server is explicitly started with
--log-bin), and a warning message
is written to the error log. Setting
sql_log_update silently sets
sql_log_bin instead (or do
nothing if the server is explicitly started with
--log-bin).
Removed support for the ISAM storage engine.
If you have ISAM tables, you should convert
them before upgrading. See
Section 2.18.1.2, “Upgrading from MySQL 4.1 to 5.0”.
Removed support for RAID options in
MyISAM tables. If you have tables that use
these options, you should convert them before upgrading. See
Section 2.18.1.2, “Upgrading from MySQL 4.1 to 5.0”.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server and MySQL Community Server release (5.0.82). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
The time zone tables for Windows available at http://dev.mysql.com/downloads/timezones.html have been updated. (Bug#39923)
Bugs fixed:
Replication:
When stopping and restarting the slave while it was replicating
temporary tables, the slave server could crash or raise an
assertion failure. This was due to the fact that, although
temporary tables were saved between slave thread restarts, the
reference to the thread being used
(table->in_use) was not being properly
updated when restarting, continuing to reference the old thread
instead of the new one. This issue affected statement-based
replication only.
(Bug#41725)
UNCOMPRESSED_LENGTH() returned a
garbage result when passed a string shorter than 5 bytes. Now
UNCOMPRESSED_LENGTH() returns
NULL and generates a warning.
(Bug#44796)
Several Valgrind warnings were silenced. (Bug#44774, Bug#44792)
Incorrect time was reported at the end of mysqldump output. (Bug#44424)
EXPLAIN
EXTENDED could crash for
UNION queries in which the last
SELECT was not parenthesized and
included an ORDER BY clause.
(Bug#43612)
SELECT ... INTO
@var could produce values different from
SELECT ...
without the INTO clause.
(Bug#42009)
Using --hexdump together
with
--read-from-remote-server
caused mysqlbinlog to crash.
(Bug#41943)
For some queries, an equality propagation problem could cause
a = b and b = a to be
handled differently.
(Bug#40925)
For views created with a column list clause, column aliases were
not substituted when selecting through the view using a
HAVING clause.
(Bug#40825)
A multiple-table DELETE involving
a table self-join could cause a server crash.
(Bug#39918)
Creating an InnoDB table with a comment
containing a '#' character caused foreign key
constraints to be omitted.
(Bug#39793)
The mysql option
--ignore-spaces was nonfunctional.
(Bug#39101)
If a query was such as to produce the error 1054
Unknown column '...' in 'field list', using
EXPLAIN
EXTENDED with the query could cause a server crash.
(Bug#37362)
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server and MySQL Community Server release (5.0.80). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Bugs fixed:
Replication:
Restarting the replication slave — either by using
STOP SLAVE plus
START SLAVE, or by restarting the
slave mysqld process — could sometimes
cause the slave to crash when using a debug version of the
server.
(Bug#38694)
Replication: Killing the thread executing a DDL statement, after it had finished its execution but before it had written the binlog event, caused the error code in the binlog event to be set (incorrectly) to ER_SERVER_SHUTDOWN or ER_QUERY_INTERRUPTED, which caused replication to fail. (Bug#37145)
Replication: Column aliases used inside subqueries were ignored in the binary log. (Bug#35515)
Replication:
The statements
DROP PROCEDURE
IF EXISTS and
DROP FUNCTION IF
EXISTS were not written to the binary log if the
procedure or function to be dropped did not exist.
(Bug#13684)
See also Bug#25705.
Use of HANDLER statements with
INFORMATION_SCHEMA tables caused a server
crash. Now HANDLER is prohibited
with such tables.
(Bug#44151)
myisamchk could display a negative
Max keyfile length value.
(Bug#43950)
mysqld_multi incorrectly passed
--no-defaults to
mysqld_safe.
(Bug#43876)
On Windows, a server crash occurred for attempts to insert a
floating-point value into a CHAR
column with a maximum length less than the converted
floating-point value length.
(Bug#43833)
InnoDB uses random numbers to
generate dives into indexes for calculating index cardinality.
However, under certain conditions, the algorithm did not
generate random numbers, so ANALYZE
TABLE did not update cardinality estimates properly. A
new algorithm has been introduced with better randomization
properties, together with a system variable,
innodb_use_legacy_cardinality_algorithm,
that controls which algorithm to use. The default value of the
variable is 1 (ON), to use the original
algorithm for compatibility with existing applications. The
variable can be set to 0 (OFF) to use the new
algorithm with improved randomness.
(Bug#43660)
UNION of floating-point numbers
did unnecessary rounding.
(Bug#43432)
Certain statements might open a table and then wait for an
impending global read lock without noticing whether they hold a
table being waiting for by the global read lock, causing a hang.
Affected statements are
SELECT ... FOR
UPDATE,
LOCK TABLES ...
WRITE,
TRUNCATE
TABLE, and
LOAD DATA
INFILE.
(Bug#43230)
The InnoDB
btr_search_drop_page_hash_when_freed()
function had a race condition.
(Bug#42279)
Compressing a table with the myisampack utility caused the server to produce Valgrind warnings when it opened the table. (Bug#41541)
For a MyISAM table with
DELAY_KEY_WRITE enabled, the index file could
be corrupted without the table being marked as crashed if the
server was killed.
(Bug#41330)
Multiple-table UPDATE statements
did not properly activate triggers.
(Bug#39953)
The functions listed in Section 11.12.4.2.3, “Creating Geometry Values Using MySQL-Specific Functions”, previously accepted WKB arguments and returned WKB values. They now accept WKB or geometry arguments and return geometry values.
The functions listed in Section 11.12.4.2.2, “Creating Geometry Values Using WKB Functions”, previously accepted WKB arguments and returned geometry values. They now accept WKB or geometry arguments and return geometry values. (Bug#38990)
An UPDATE statement that updated
a column using the same
DES_ENCRYPT() value for each row
actually updated different rows with different values.
(Bug#35087)
For shared-memory connections, the read and write methods did
not properly handle asynchronous close events, which could lead
to the client locking up waiting for a server response. For
example, a call to
mysql_real_query() would block
forever on the client side if the executed statement was aborted
on the server side. Thanks to Armin Schöffmann for the bug
report and patch.
(Bug#33899)
CHECKSUM TABLE was not killable
with KILL QUERY.
(Bug#33146)
myisamchk and myisampack
were not being linked with the library that enabled support for
* filename pattern expansion.
(Bug#29248)
COMMIT did not delete savepoints
if there were no changes in the transaction.
(Bug#26288)
Several memory allocation functions were not being checked for out-of-memory return values. (Bug#25058)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.77.
Functionality added or changed:
Security Enhancement:
To enable stricter control over the location from which
user-defined functions can be loaded, the
plugin_dir system variable has
been backported from MySQL 5.1. If the value is nonempty,
user-defined function object files can be loaded only from the
directory named by this variable. If the value is empty, the
behavior that is used prior to the inclusion of
plugin_dir applies: The UDF
object files must be located in a directory that is searched by
your system's dynamic linker.
(Bug#37428)
A new status variable,
Queries, indicates the number
of statements executed by the server. This includes statements
executed within stored programs, unlike the
Questions variable which
includes only statements sent to the server by clients.
(Bug#41131)
Previously, index hints did not work for
FULLTEXT searches. Now they work as follows:
For natural language mode searches, index hints are silently
ignored. For example, IGNORE INDEX(i) is
ignored with no warning and the index is still used.
For boolean mode searches, index hints are honored. (Bug#38842)
Bugs fixed:
Important Change: Security Fix: Additional corrections were made for the symlink-related privilege problem originally addressed in MySQL 5.0.60. The original fix did not correctly handle the data directory path name if it contained symlinked directories in its path, and the check was made only at table-creation time, not at table-opening time later. (Bug#32167, CVE-2008-2079)
See also Bug#39277.
Security Enhancement:
The server consumed excess memory while parsing statements with
hundreds or thousands of nested boolean conditions (such as
OR (OR ... (OR ... ))). This could lead to a
server crash or incorrect statement execution, or cause other
client statements to fail due to lack of memory. The latter
result constitutes a denial of service.
(Bug#38296)
Incompatible Change:
There were some problems using DllMain()
hook functions on Windows that automatically do global and
per-thread initialization for
libmysqld.dll:
Per-thread initialization: MySQL internally counts the
number of active threads, which causes a delay in
my_end() if not all threads have
exited. But there are threads that can be started either by
Windows internally (often in TCP/IP scenarios) or by users.
Those threads do not necessarily use
libmysql.dll functionality but still
contribute to the open-thread count. (One symptom is a
five-second delay in times for PHP scripts to finish.)
Process-initialization:
my_init() calls
WSAStartup that itself loads DLLs and
can lead to a deadlock in the Windows loader.
To correct these problems, DLL initialization code now is not
invoked from libmysql.dll by default. To
obtain the previous behavior (DLL initialization code will be
called), set the LIBMYSQL_DLLINIT environment
variable to any value. This variable exists only to prevent
breakage of existing Windows-only applications that do not call
mysql_thread_init() and work
okay today. Use of LIBMYSQL_DLLINIT is
discouraged and is removed in MySQL 6.0.
(Bug#37226, Bug#33031)
Incompatible Change:
SHOW STATUS took a lot of CPU
time for calculating the value of the
Innodb_buffer_pool_pages_latched
status variable. Now this variable is calculated and included in
the output of SHOW STATUS only if
the UNIV_DEBUG symbol is defined at MySQL
build time.
(Bug#36600)
Incompatible Change:
In connection with view creation, the server created
arc directories inside database directories
and maintained useless copies of .frm files
there. Creation and renaming procedures of those copies as well
as creation of arc directories has been
discontinued.
This change does cause a problem when downgrading to older server versions which manifests itself under these circumstances:
Create a view v_orig in MySQL 5.0.72 or
higher.
Rename the view to v_new and then back to
v_orig.
Downgrade to an older 5.0.x server and run mysql_upgrade.
Try to rename v_orig to
v_new again. This operation fails.
As a workaround to avoid this problem, use either of these approaches:
Dump your data using mysqldump before downgrading and reload the dump file after downgrading.
Instead of renaming a view after the downgrade, drop it and recreate it.
The downgrade problem introduced by the fix for this bug has been addressed as Bug#40021. (Bug#17823)
Replication: When rotating relay log files, the slave deletes relay log files and then edits the relay log index file. Formerly, if the slave shut down unexpectedly between these two events, the relay log index file could then reference relay logs that no longer existed. Depending on the circumstances, this could when restarting the slave cause either a race condition or the failure of replication. (Bug#38826, Bug#39325)
In example option files provided in MySQL distributions, the
thread_stack value was
increased from 64K to 128K.
(Bug#41577)
SET PASSWORD caused a server
crash if the account name was given as
CURRENT_USER().
(Bug#41456)
The
INFORMATION_SCHEMA.SCHEMA_PRIVILEGES
table was limited to 7680 rows.
(Bug#41079)
In debug builds, obsolete debug code could be used to crash the server. (Bug#41041)
CHECK TABLE ... FOR
UPGRADE did not check for incompatible collation
changes made in MySQL 5.0.48 (Bug#27562, Bug#29461, Bug#29499).
This also affects mysqlcheck and
mysql_upgrade, which cause that statement to
be executed. See Section 2.18.3, “Checking Whether Table Indexes Must Be Rebuilt”.
(Bug#40984)
See also Bug#39585.
Some queries that used a “range checked for each record” scan could return incorrect results. (Bug#40974)
Certain SELECT queries could fail
with a Duplicate entry error.
(Bug#40953)
The FEDERATED handler had a memory
leak.
(Bug#40875)
IF(..., CAST( as
an argument to an aggregate function could cause an assertion
failure.
(Bug#40761)longtext_val AS
UNSIGNED), signed_val)
Prepared statements allowed invalid dates to be inserted when
the ALLOW_INVALID_DATES SQL
mode was not enabled.
(Bug#40365)
mc.exe is no longer needed to compile MySQL on Windows. This makes it possible to build MySQL from source using Visual Studio Express 2008. (Bug#40280)
Support for the revision field in
.frm files has been removed. This addresses
the downgrading problem introduced by the fix for Bug#17823.
(Bug#40021)
If the operating system is configured to return leap seconds
from OS time calls or if the MySQL server uses a time zone
definition that has leap seconds, functions such as
NOW() could return a value having
a time part that ends with :59:60 or
:59:61. If such values are inserted into a
table, they would be dumped as is by
mysqldump but considered invalid when
reloaded, leading to backup/restore problems.
Now leap second values are returned with a time part that ends
with :59:59. This means that a function such
as NOW() can return the same
value for two or three consecutive seconds during the leap
second. It remains true that literal temporal values having a
time part that ends with :59:60 or
:59:61 are considered invalid.
For additional details about leap-second handling, see Section 9.7.2, “Time Zone Leap Second Support”. (Bug#39920)
The server could crash during a sort-order optimization of a dependent subquery. (Bug#39844)
With the ONLY_FULL_GROUP_BY
SQL mode enabled, the check for nonaggregated columns in queries
with aggregate functions, but without a GROUP
BY clause was treating all the parts of the query as
if they were in the select list. This is fixed by ignoring the
nonaggregated columns in the WHERE clause.
(Bug#39656)
The server crashed if an integer field in a CSV file did not have delimiting quotes. (Bug#39616)
Creating a table with a comment of 62 characters or longer caused a server crash. (Bug#39591)
CHECK TABLE failed for
MyISAM
INFORMATION_SCHEMA tables.
(Bug#39541)
InnoDB could hang trying to open an adaptive
hash index.
(Bug#39483)
For a TIMESTAMP column in an
InnoDB table, testing the column with
multiple conditions in the WHERE clause
caused a server crash.
(Bug#39353)
The server returned a column type of
VARBINARY rather than
DATE as the result from the
COALESCE(),
IFNULL(),
IF(),
GREATEST(), or
LEAST() functions or
CASE expression if the result was
obtained using filesort in an anonymous
temporary table during the query execution.
(Bug#39283)
References to local variables in stored procedures are replaced
with
NAME_CONST( when written to the
binary log. However, an “illegal mix of collation”
error might occur when executing the log contents if the value's
collation differed from that of the variable. Now information
about the variable collation is written as well.
(Bug#39182)name,
value)
Some recent releases for Solaris 10 were built on Solaris 10 U5,
which included a new version of libnsl.so
that does not work on U4 or earlier. To correct this, Solaris 10
builds now are created on machines that do not have that
upgraded libnsl.so, so that they will work
on Solaris 10 installations both with and without the upgraded
libnsl.so.
(Bug#39074)
With binary logging enabled CREATE
VIEW was subject to possible buffer overwrite and a
server crash.
(Bug#39040)
Queries of the form SELECT ... REGEXP BINARY
NULL could lead to a hung or crashed server.
(Bug#39021)
Statements of the form INSERT ... SELECT .. ON
DUPLICATE KEY UPDATE could result in a server crash.
(Bug#39002)col_name =
DEFAULT
Column names constructed due to wild-card expansion done inside a stored procedure could point to freed memory if the expansion was performed after the first call to the stored procedure. (Bug#38823)
Repeated CREATE
TABLE ... SELECT statements, where the created table
contained an AUTO_INCREMENT column, could
lead to an assertion failure.
(Bug#38821)
If delayed insert failed to upgrade the lock, it did not free
the temporary memory storage used to keep newly constructed
BLOB values in memory, resulting
in a memory leak.
(Bug#38693)
A server crash resulted from concurrent execution of a
multiple-table UPDATE that used a
NATURAL or USING join
together with FLUSH
TABLES WITH READ LOCK or ALTER
TABLE for the table being updated.
(Bug#38691)
On ActiveState Perl, mysql-test-run.pl --start-and-exit started but did not exit. (Bug#38629)
Server-side cursors were not initialized properly, which could cause a server crash. (Bug#38486)
Stored procedures involving substrings could crash the server on certain platforms due to invalid memory reads. (Bug#38469)
A server crash or Valgrind warnings could result when a stored procedure selected from a view that referenced a function. (Bug#38291)
Incorrect handling of aggregate functions when loose index scan was used caused a server crash. (Bug#38195)
Queries containing a subquery with DISTINCT
and ORDER BY could cause a server crash.
(Bug#38191)
Queries with a HAVING clause could return a
spurious row.
(Bug#38072)
Use of spatial data types in prepared statements could cause memory leaks or server crashes. (Bug#37956, Bug#37671)
The server crashed if an argument to a stored procedure was a subquery that returned more than one row. (Bug#37949)
When analyzing the possible index use cases, the server was incorrectly reusing an internal structure, leading to a server crash. (Bug#37943)
A SELECT with a NULL NOT
IN condition containing a complex subquery from the
same table as in the outer select caused an assertion failure.
(Bug#37894)
For InnoDB tables, ORDER BY ...
DESC sometimes returned results in ascending order.
(Bug#37830)
If a table has a BIT NOT NULL column
c1 with a length shorter than 8 bits and some
additional NOT NULL columns
c2, ..., and a
SELECT query has a
WHERE clause of the form (c1 =
, the
query could return an unexpected result set.
(Bug#37799)constant) AND c2 ...
Nesting of IF() inside of
SUM() could cause an extreme
server slowdown.
(Bug#37662)
The MONTHNAME() and
DAYNAME() functions returned a
binary string, so that using
LOWER() or
UPPER() had no effect. Now
MONTHNAME() and
DAYNAME() return a value in
character_set_connection
character set.
(Bug#37575)
TIMEDIFF() was erroneously
treated as always returning a positive result. Also,
CAST() of
TIME values to
DECIMAL dropped the sign of
negative values.
(Bug#37553)
See also Bug#42525.
mysqlcheck used
SHOW FULL
TABLES to get the list of tables in a database. For
some problems, such as an empty .frm file
for a table, this would fail and mysqlcheck
then would neglect to check other tables in the database.
(Bug#37527)
The <=>
operator could return incorrect results when comparing
NULL to DATE,
TIME, or
DATETIME values.
(Bug#37526)
Updating a view with a subquery in the CHECK
option could cause an assertion failure.
(Bug#37460)
Statements that displayed the value of system variables (for
example, SHOW VARIABLES) expect
variable values to be encoded in
character_set_system. However,
variables set from the command line such as
basedir or
datadir were encoded using
character_set_filesystem and
not converted correctly.
(Bug#37339)
For a MyISAM table with CHECKSUM =
1 and ROW_FORMAT = DYNAMIC table
options, a data consistency check (maximum record length) could
fail and cause the table to be marked as corrupted.
(Bug#37310)
The max_length result set metadata value was
calculated incorrectly under some circumstances.
(Bug#37301)
CREATE INDEX could crash with
InnoDB plugin 1.0.1.
(Bug#37284)
Certain boolean-mode FULLTEXT searches that
used the truncation operator did not return matching records and
calculated relevance incorrectly.
(Bug#37245)
The NO_BACKSLASH_ESCAPES SQL
mode was ignored for
LOAD DATA
INFILE and SELECT INTO ... OUTFILE.
The setting is taken into account now.
(Bug#37114)
On a 32-bit server built without big tables support, the offset
argument in a LIMIT clause might be truncated
due to a 64-bit to 32-bit cast.
(Bug#37075)
If the server failed to expire binary log files at startup, it could crash. (Bug#37027)
The code for the ut_usectime() function in
InnoDB did not handle errors from the
gettimeofday() system call. Now it retries
gettimeofday() several times and updates
the value of the
Innodb_row_lock_time_max
status variable only if ut_usectime() was
successful.
(Bug#36819)
Use of CONVERT() with
GROUP BY to convert numeric values to
CHAR could return truncated
results.
(Bug#36772)
A query which had an ORDER BY DESC clause
that is satisfied with a reverse range scan could cause a server
crash for some specific CPU/compiler combinations.
(Bug#36639)
Dumping information about locks in use by sending a
SIGHUP signal to the server or by invoking
the mysqladmin debug command could lead to a
server crash in debug builds or to undefined behavior in
production builds.
(Bug#36579)
The mysql client, when built with Visual Studio 2005, did not display Japanese characters. (Bug#36279)
When the fractional part in a multiplication of
DECIMAL values overflowed, the
server truncated the first operand rather than the longest. Now
the server truncates so as to produce more precise
multiplications.
(Bug#36270)
A read past the end of the string could occur while parsing the
value of the
--innodb-data-file-path option.
(Bug#36149)
Host name values in SQL statements were not being checked for
'@', which is illegal according to RFC952.
(Bug#35924)
The UUID() function returned
UUIDs with the wrong time; this was because the offset for the
time part in UUIDs was miscalculated.
(Bug#35848)
SHOW CREATE TABLE did not display
a printable value for the default value of
BIT columns.
(Bug#35796)
mysql_install_db failed on machines that had
the host name set to localhost.
(Bug#35754)
Dynamic plugins failed to load on i5/OS. (Bug#35743)
Freeing of an internal parser stack during parsing of complex stored programs caused a server crash. (Bug#35577, Bug#37269, Bug#37228)
The max_length metadata value was calculated
incorrectly for the FORMAT()
function, which could cause incorrect result set metadata to be
sent to clients.
(Bug#35558)
Index scans performed with the sort_union()
access method returned wrong results, caused memory to be
leaked, and caused temporary files to be deleted when the limit
set by sort_buffer_size was
reached.
(Bug#35477, Bug#35478)
If the server crashed with an InnoDB error
due to unavailability of undo slots, errors could persist during
rollback when the server was restarted: There are two
UNDO slot caches (for
INSERT and
UPDATE). If all slots end up in
one of the slot caches, a request for a slot from the other slot
cache would fail. This can happen if the request is for an
UPDATE slot and all slots are in
the INSERT slot cache, or vice
versa.
(Bug#35352)
For InnoDB tables, ALTER TABLE
DROP failed if the name of the column to be dropped
began with “foreign”.
(Bug#35220)
perror on Windows did not know about Win32 system error codes. (Bug#34825)
EXPLAIN EXTENDED evaluation of aggregate
functions that required a temporary table caused a server crash.
(Bug#34773)
Queries of the form SELECT ... WHERE
failed
when the server used a single-byte character set and the client
used a multi-byte character set.
(Bug#34760)string = ANY(...)
See also Bug#20835.
Using OPTIMIZE TABLE as the first
statement on an InnoDB table with an
AUTO_INCREMENT column could cause a server
crash.
(Bug#34286)
mysql_install_db failed if the server was
running with an SQL mode of
TRADITIONAL. This program now
resets the SQL mode internally to avoid this problem.
(Bug#34159)
Changes to build files were made to enable the MySQL distribution to compile on Microsoft Visual C++ Express 2008. (Bug#33907)
The mysql client incorrectly parsed statements containing the word “delimiter” in mid-statement.
This fix is different from the one applied for this bug in MySQL 5.0.66. (Bug#33812)
See also Bug#38158.
For a stored procedure containing a SELECT * ... RIGHT
JOIN query, execution failed for the second call.
(Bug#33811)
Previously, use of index hints with views (which do not have indexes) produced the error ERROR 1221 (HY000): Incorrect usage of USE/IGNORE INDEX and VIEW. Now this produces ERROR 1176 (HY000): Key '...' doesn't exist in table '...', the same error as for base tables without an appropriate index. (Bug#33461)
Cached queries that used 256 or more tables were not properly
cached, so that later query invalidation due to a
TRUNCATE
TABLE for one of the tables caused the server to hang.
(Bug#33362)
Some division operations produced a result with incorrect precision. (Bug#31616)
mysql_upgrade attempted to use the
/proc file system even on systems that do
not have it.
(Bug#31605)
mysqldump could fail to dump views containing a large number of columns. (Bug#31434)
Queries executed using join buffering of
BIT columns could produce
incorrect results.
(Bug#31399)
ALTER TABLE CONVERT TO CHARACTER SET did not
convert TINYTEXT or
MEDIUMTEXT columns to a longer
text type if necessary when converting the column to a different
character set.
(Bug#31291)
For installation on Solaris using pkgadd
packages, the mysql_install_db script was
generated in the scripts directory, but the
temporary files used during the process were left there and not
deleted.
(Bug#31052)
Several MySQL programs could fail if the HOME
environment variable had an empty value.
(Bug#30394)
On NetWare, mysql_install_db could appear to execute normally even if it failed to create the initial databases. (Bug#30129)
The Serbian translation for the
ER_INCORRECT_GLOBAL_LOCAL_VAR
error was corrected.
(Bug#29738)
XA transaction rollbacks could result in corrupted transaction states and a server crash. (Bug#28323)
The BUILD/check-cpu build script failed if gcc had a different name (such as gcc.real on Debian). (Bug#27526)
On Windows, Visual Studio does not take into account some x86
hardware limitations, which led to incorrect results converting
large DOUBLE values to unsigned
BIGINT values.
(Bug#27483)
SSL support was not included in some “generic” RPM packages. (Bug#26760)
In some cases, the parser interpreted the ;
character as the end of input and misinterpreted stored program
definitions.
(Bug#26030)
The Questions status variable
is intended as a count of statements sent by clients to the
server, but was also counting statements executed within stored
routines.
(Bug#24289)
For access to the
INFORMATION_SCHEMA.VIEWS table, the
server did not check the SHOW
VIEW and SELECT
privileges, leading to inconsistency between output from that
table and the SHOW CREATE VIEW
statement.
(Bug#22763)
The FLUSH
PRIVILEGES statement did not produce an error when it
failed.
(Bug#21226)
A race condition between the mysqld.exe server and the Windows service manager could lead to inability to stop the server from the service manager. (Bug#20430)
mysqld_safe would sometimes fail to remove
the pid file for the old mysql process after
a crash. As a result, the server would fail to start due to a
false A mysqld process already exists...
error.
(Bug#11122)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.79). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Support Ending for AIX 5.2: Per the MySQL Support Lifecycle policy regarding ending support for OS versions that have reached vendor end of life, we plan to discontinue building or supporting MySQL binaries for AIX 5.2 as of April 30, 2009. This release of MySQL 5.0 (5.0.80) is the last MySQL 5.0 release with support for AIX 5.2. For more information, see the March 24, 2009 note at MySQL Product Support EOL Announcements.
Functionality added or changed:
The MD5 algorithm now uses the Xfree implementation. (Bug#42434)
Bugs fixed:
Replication:
An INSERT
DELAYED into a
TIMESTAMP column issued
concurrently with an insert on the same column not using
DELAYED, but applied after the other insert,
was logged using the same timestamp as generated by the other
(non-DELAYED) insert.
(Bug#41719)
An attempt by a user who did not have the
SUPER privilege to kill a system
thread could cause a server crash.
(Bug#43748)
Use of USE INDEX hints could cause
EXPLAIN
EXTENDED to crash.
(Bug#43354)
mysql crashed if a request for the current
database name returned an empty result, such as after the client
has executed a preceding SET
sql_select_limit=0 statement.
(Bug#43254)
The strings/CHARSET_INFO.txt file was not
included in source distributions.
(Bug#42937)
mysqldump included views that were excluded
with the --ignore-table
option.
(Bug#42635)
Passing an unknown time zone specification to
CONVERT_TZ() resulted in a memory
leak.
(Bug#42502)
With more than two arguments,
LEAST(),
GREATEST(), and
CASE could unnecessarily return
Illegal mix of collations errors.
(Bug#41627)
The mysql client could misinterpret its input if a line was longer than an internal buffer. (Bug#41486)
In the help command output displayed by
mysql, the description for the
\c (clear) command was
misleading.
(Bug#41268)
The use of NAME_CONST() can
result in a problem for
CREATE TABLE ...
SELECT statements when the source column expressions
refer to local variables. Converting these references to
NAME_CONST() expressions can
result in column names that are different on the master and
slave servers, or names that are too long to be legal column
identifiers. A workaround is to supply aliases for columns that
refer to local variables.
Now a warning is issued in such cases that indicate possible problems. (Bug#35383)
CHECK TABLE,
REPAIR TABLE,
ANALYZE TABLE, and
OPTIMIZE TABLE erroneously
reported a table to be corrupt if the table did not exist or the
statement was terminated with
KILL.
(Bug#29458)
The Time column for SHOW
PROCESSLIST output now can have negative values.
Previously, the column was unsigned and negative values were
displayed incorrectly as large positive values. Negative values
can occur if a thread alters the time into the future with
SET TIMESTAMP =
or the thread is
executing on a slave and processing events from a master that
has its clock set ahead of the slave.
(Bug#22047)value
Restoring a mysqldump dump file containing
FEDERATED tables failed because the file
contained the data for the table. Now only the table definition
is dumped (because the data is located elsewhere).
(Bug#21360)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.78). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
The libedit library was upgraded to version
2.11.
(Bug#42433)
The query cache now checks whether a
SELECT statement begins with
SQL_NO_CACHE to determine whether it can skip
checking for the query result in the query cache. This is not
supported when SQL_NO_CACHE occurs within a
comment.
(Bug#37416)
Bugs fixed:
Replication: Server IDs greater than 2147483647 (232 – 1) were represented by negative numbers in the binary log. (Bug#37313)
Replication:
The
--replicate-
options were not evaluated correctly when replicating
multi-table updates.
*-table
As a result of this fix, replication of multi-table updates no longer fails when an update references a missing table but does not update any of its columns. (Bug#37051)
Replication:
When its disk becomes full, a replication slave may wait while
writing the binary log, relay log or
MyISAM tables, continuing after
space has been made available. The error message provided in
such cases was not clear about the frequency with which checking
for free space is done (once every 60 seconds), and how long the
server waits after space has been freed before continuing (also
60 seconds); this caused users to think that the server had
hung.
These issues have been addressed by making the error message clearer, and dividing it into two separate messages:
The error message Disk is full writing
'filename' (Errcode:
error_code). Waiting for someone
to free space... (Expect up to 60 secs delay for server to
continue after freeing disk space) is printed
only once.
The warning Retry in 60 secs, Message reprinted in 600 secs is printed once every for every 10 times that the check for free space is made; that is, the check is performed once each 60 seconds, but the reminder that space needs to be freed is printed only once every 10 minutes (600 seconds).
On 32-bit Windows, mysqld could not use large buffers due to a 2GB user mode address limit. (Bug#43082)
The use by libedit of the
__weak_reference() macro caused compilation
failure on FreeBSD.
(Bug#42817)
Tables could enter open table cache for a thread without being properly cleaned up, leading to a server crash. (Bug#42419)
mysqldumpslow parsed the
--debug and
--verbose options
incorrectly.
(Bug#42027)
String reallocation could cause memory overruns. (Bug#41868)
Queries that used the loose index scan access method could return no rows. (Bug#41610)
In InnoDB recovery after a server crash,
rollback of a transaction that updated a column from
NULL to NULL could cause
another crash.
(Bug#41571)
If InnoDB reached its limit on the number of
concurrent transactions (1023), it wrote a descriptive message
to the error log but returned a misleading error message to the
client, or an assertion failure occurred.
(Bug#41529)
Use of SELECT * allowed users with rights to
only some columns of a view to access all columns.
(Bug#41354)
The server did not robustly handle problems hang if a table
opened with HANDLER needed to be
re-opened because it had been altered to use a different storage
engine that does not support
HANDLER. The server also failed
to set an error if the re-open attempt failed. These problems
could cause the server to crash or hang.
(Bug#41110, Bug#41112)
For prepared statements, multibyte character sets were not
taking into account when calculating
max_length for string values and
mysql_stmt_fetch() could return
truncated strings.
(Bug#41078)
The mysql_change_user() C API
function changed the value of the
sql_big_selects session
variable.
(Bug#40363)
See also Bug#20023.
For a view that references a table in another database, mysqldump wrote the view name qualified with the current database name. This makes it impossible to reload the dump file into a different database. (Bug#40345)
For an InnoDB table,
DROP TABLE or
ALTER TABLE ...
DISCARD TABLESPACE could take a long time or cause a
server crash.
(Bug#39939)
perror did not produce correct output for error codes 153 to 163. (Bug#39370)
Comparisons between row constructors, such as (a, b) =
(c, d) resulted in unnecessary Illegal mix of
collations errors for string columns.
(Bug#37601)
An argument to the MATCH()
function that was an alias for an expression other than a column
name caused a server crash.
(Bug#36737)
For DROP FUNCTION with names that
were qualified with a database name, the database name was
handled in case-sensitive fashion even with
lower_case_table_names set to
1.
(Bug#33813)
mysqldump --compatible=mysql40 emitted
statements referring to the
character_set_client system
variable, which is unknown before MySQL 4.1. Now the statements
are enclosed in version-specific comments.
(Bug#33550)
Use of MBR spatial functions such as
MBRTouches() with columns of
InnoDB tables caused a server crash rather
than an error.
(Bug#31435)
The mysql client mishandled input parsing if
a delimiter command was not first on the
line.
(Bug#31060)
For installation on Solaris using pkgadd
packages, the mysql_install_db script was
generated in the scripts directory, but the
temporary files used during the process were left there and not
deleted.
(Bug#31052)
SHOW PRIVILEGES listed the
CREATE ROUTINE privilege as
having a context of Functions,Procedures, but
it is a database-level privilege.
(Bug#30305)
SHOW TABLE STATUS could fail to
produce output for tables with non-ASCII characters in their
name.
(Bug#25830)
Floating-point numbers could be handled with different numbers of digits depending on whether the text or prepared-statement protocol was used. (Bug#21205)
Incorrect length metadata could be returned for LONG
TEXT columns when a multibyte server character set was
used.
(Bug#19829)
ROUND() sometimes returned
different results on different platforms.
(Bug#15936)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.76). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Bugs fixed:
MySQL Cluster: Packaging:
Packages for MySQL Cluster were missing the
libndbclient.so and
libndbclient.a files.
(Bug#42278)
An optimization introduced for Bug#37553 required an explicit
cast to be added for some uses of
TIMEDIFF() because automatic
casting could produce incorrect results. (It was necessary to
use TIME(TIMEDIFF(...)).)
(Bug#42525)
The SSL certficates included with MySQL distributions were regenerated because the previous ones had expired. (Bug#42366)
Dependent subqueries such as the following caused a memory leak proportional to the number of outer rows:
SELECT COUNT(*) FROM t1, t2 WHERE t2.b IN (SELECT DISTINCT t2.b FROM t2 WHERE t2.b = t1.a);
Some queries using NAME_CONST(.. COLLATE
...) led to a server crash due to a failed type cast.
(Bug#42014)
On Mac OS X, some of the universal client libraries were not actually universal and were missing code for one or more architectures. (Bug#41940)
DATE_FORMAT() could cause a
server crash for year-zero dates.
(Bug#41470)
When substituting system constant functions with a constant
result, the server was not expecting NULL
function return values and could crash.
(Bug#41437)
For a TIMESTAMP NOT
NULL DEFAULT ... column, storing
NULL as the return value from some functions
caused a “cannot be NULL” error.
NULL returns now correctly cause the column
default value to be stored.
(Bug#41370)
The Windows installer displayed incorrect product names in some images. (Bug#40845)
The query cache stored only partial query results if a statement failed while the results were being sent to the client. This could cause other clients to hang when trying to read the cached result. Now if a statement fails, the result is not cached. (Bug#40264)
The expression ROW(...) IN (SELECT ... FROM
DUAL) always returned TRUE.
(Bug#39069)
The greedy optimizer could cause a server crash due to improper handling of nested outer joins. (Bug#38795)
Use of COUNT(DISTINCT) prevented
NULL testing in the HAVING
clause.
(Bug#38637)
Enabling the sync_frm system
variable had no effect on the handling of
.frm files for views.
(Bug#38145)
The query cache stored packets containing the server status of the time when the cached statement was run. This might lead to an incorrect transaction status on the client side if a statement was cached during a transaction and later served outside a transaction context (or vice versa). (Bug#36326)
If the system time was adjusted backward during query execution, the apparent execution time could be negative. But in some cases these queries would be written to the slow query log, with the negative execution time written as a large unsigned number. Now statements with apparent negative execution time are not written to the slow query log. (Bug#35396)
For mysqld_multi, using the
--mysqld=mysqld_safe option
caused the --defaults-file
and --defaults-extra-file
options to behave the same way.
(Bug#32136)
Attempts to open a valid MERGE table sometimes resulted in a
ER_WRONG_MRG_TABLE error. This
happened after failure to open an invalid MERGE table had also
generated an ER_WRONG_MRG_TABLE
error.
(Bug#32047)
The mysql_change_user() C API
function caused global
Com_ status
variable values to be incorrect.
(Bug#31222)xxx
For Solaris package installation using
pkgadd, the postinstall script failed,
causing the system tables in the mysql
database not to be created.
(Bug#31164)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.67 (binary) and 5.0.75 (source-only).
Functionality added or changed:
Security Enhancement:
To enable stricter control over the location from which
user-defined functions can be loaded, the
plugin_dir system variable has
been backported from MySQL 5.1. If the value is nonempty,
user-defined function object files can be loaded only from the
directory named by this variable. If the value is empty, the
behavior that is used prior to the inclusion of
plugin_dir applies: The UDF
object files must be located in a directory that is searched by
your system's dynamic linker.
(Bug#37428)
A new status variable,
Queries, indicates the number
of statements executed by the server. This includes statements
executed within stored programs, unlike the
Questions variable which
includes only statements sent to the server by clients.
(Bug#41131)
Previously, index hints did not work for
FULLTEXT searches. Now they work as follows:
For natural language mode searches, index hints are silently
ignored. For example, IGNORE INDEX(i) is
ignored with no warning and the index is still used.
For boolean mode searches, index hints are honored. (Bug#38842)
Bugs fixed:
Important Change: Security Fix: Additional corrections were made for the symlink-related privilege problem originally addressed in MySQL 5.0.60. The original fix did not correctly handle the data directory path name if it contained symlinked directories in its path, and the check was made only at table-creation time, not at table-opening time later. (Bug#32167, CVE-2008-2079)
See also Bug#39277.
Security Enhancement:
The server consumed excess memory while parsing statements with
hundreds or thousands of nested boolean conditions (such as
OR (OR ... (OR ... ))). This could lead to a
server crash or incorrect statement execution, or cause other
client statements to fail due to lack of memory. The latter
result constitutes a denial of service.
(Bug#38296)
Incompatible Change:
There were some problems using DllMain()
hook functions on Windows that automatically do global and
per-thread initialization for
libmysqld.dll:
Per-thread initialization: MySQL internally counts the
number of active threads, which causes a delay in
my_end() if not all threads have
exited. But there are threads that can be started either by
Windows internally (often in TCP/IP scenarios) or by users.
Those threads do not necessarily use
libmysql.dll functionality but still
contribute to the open-thread count. (One symptom is a
five-second delay in times for PHP scripts to finish.)
Process-initialization:
my_init() calls
WSAStartup that itself loads DLLs and
can lead to a deadlock in the Windows loader.
To correct these problems, DLL initialization code now is not
invoked from libmysql.dll by default. To
obtain the previous behavior (DLL initialization code will be
called), set the LIBMYSQL_DLLINIT environment
variable to any value. This variable exists only to prevent
breakage of existing Windows-only applications that do not call
mysql_thread_init() and work
okay today. Use of LIBMYSQL_DLLINIT is
discouraged and is removed in MySQL 6.0.
(Bug#37226, Bug#33031)
Incompatible Change:
SHOW STATUS took a lot of CPU
time for calculating the value of the
Innodb_buffer_pool_pages_latched
status variable. Now this variable is calculated and included in
the output of SHOW STATUS only if
the UNIV_DEBUG symbol is defined at MySQL
build time.
(Bug#36600)
Incompatible Change:
In connection with view creation, the server created
arc directories inside database directories
and maintained useless copies of .frm files
there. Creation and renaming procedures of those copies as well
as creation of arc directories has been
discontinued.
This change does cause a problem when downgrading to older server versions which manifests itself under these circumstances:
Create a view v_orig in MySQL 5.0.72 or
higher.
Rename the view to v_new and then back to
v_orig.
Downgrade to an older 5.0.x server and run mysql_upgrade.
Try to rename v_orig to
v_new again. This operation fails.
As a workaround to avoid this problem, use either of these approaches:
Dump your data using mysqldump before downgrading and reload the dump file after downgrading.
Instead of renaming a view after the downgrade, drop it and recreate it.
The downgrade problem introduced by the fix for this bug has been addressed as Bug#40021. (Bug#17823)
Replication: When rotating relay log files, the slave deletes relay log files and then edits the relay log index file. Formerly, if the slave shut down unexpectedly between these two events, the relay log index file could then reference relay logs that no longer existed. Depending on the circumstances, this could when restarting the slave cause either a race condition or the failure of replication. (Bug#38826, Bug#39325)
In example option files provided in MySQL distributions, the
thread_stack value was
increased from 64K to 128K.
(Bug#41577)
SET PASSWORD caused a server
crash if the account name was given as
CURRENT_USER().
(Bug#41456)
The
INFORMATION_SCHEMA.SCHEMA_PRIVILEGES
table was limited to 7680 rows.
(Bug#41079)
In debug builds, obsolete debug code could be used to crash the server. (Bug#41041)
CHECK TABLE ... FOR
UPGRADE did not check for incompatible collation
changes made in MySQL 5.0.48 (Bug#27562, Bug#29461, Bug#29499).
This also affects mysqlcheck and
mysql_upgrade, which cause that statement to
be executed. See Section 2.18.3, “Checking Whether Table Indexes Must Be Rebuilt”.
(Bug#40984)
See also Bug#39585.
Some queries that used a “range checked for each record” scan could return incorrect results. (Bug#40974)
Certain SELECT queries could fail
with a Duplicate entry error.
(Bug#40953)
The FEDERATED handler had a memory
leak.
(Bug#40875)
IF(..., CAST( as
an argument to an aggregate function could cause an assertion
failure.
(Bug#40761)longtext_val AS
UNSIGNED), signed_val)
Prepared statements allowed invalid dates to be inserted when
the ALLOW_INVALID_DATES SQL
mode was not enabled.
(Bug#40365)
mc.exe is no longer needed to compile MySQL on Windows. This makes it possible to build MySQL from source using Visual Studio Express 2008. (Bug#40280)
Support for the revision field in
.frm files has been removed. This addresses
the downgrading problem introduced by the fix for Bug#17823.
(Bug#40021)
If the operating system is configured to return leap seconds
from OS time calls or if the MySQL server uses a time zone
definition that has leap seconds, functions such as
NOW() could return a value having
a time part that ends with :59:60 or
:59:61. If such values are inserted into a
table, they would be dumped as is by
mysqldump but considered invalid when
reloaded, leading to backup/restore problems.
Now leap second values are returned with a time part that ends
with :59:59. This means that a function such
as NOW() can return the same
value for two or three consecutive seconds during the leap
second. It remains true that literal temporal values having a
time part that ends with :59:60 or
:59:61 are considered invalid.
For additional details about leap-second handling, see Section 9.7.2, “Time Zone Leap Second Support”. (Bug#39920)
The server could crash during a sort-order optimization of a dependent subquery. (Bug#39844)
With the ONLY_FULL_GROUP_BY
SQL mode enabled, the check for nonaggregated columns in queries
with aggregate functions, but without a GROUP
BY clause was treating all the parts of the query as
if they were in the select list. This is fixed by ignoring the
nonaggregated columns in the WHERE clause.
(Bug#39656)
The server crashed if an integer field in a CSV file did not have delimiting quotes. (Bug#39616)
Creating a table with a comment of 62 characters or longer caused a server crash. (Bug#39591)
CHECK TABLE failed for
MyISAM
INFORMATION_SCHEMA tables.
(Bug#39541)
InnoDB could hang trying to open an adaptive
hash index.
(Bug#39483)
For a TIMESTAMP column in an
InnoDB table, testing the column with
multiple conditions in the WHERE clause
caused a server crash.
(Bug#39353)
The server returned a column type of
VARBINARY rather than
DATE as the result from the
COALESCE(),
IFNULL(),
IF(),
GREATEST(), or
LEAST() functions or
CASE expression if the result was
obtained using filesort in an anonymous
temporary table during the query execution.
(Bug#39283)
References to local variables in stored procedures are replaced
with
NAME_CONST( when written to the
binary log. However, an “illegal mix of collation”
error might occur when executing the log contents if the value's
collation differed from that of the variable. Now information
about the variable collation is written as well.
(Bug#39182)name,
value)
Some recent releases for Solaris 10 were built on Solaris 10 U5,
which included a new version of libnsl.so
that does not work on U4 or earlier. To correct this, Solaris 10
builds now are created on machines that do not have that
upgraded libnsl.so, so that they will work
on Solaris 10 installations both with and without the upgraded
libnsl.so.
(Bug#39074)
With binary logging enabled CREATE
VIEW was subject to possible buffer overwrite and a
server crash.
(Bug#39040)
Queries of the form SELECT ... REGEXP BINARY
NULL could lead to a hung or crashed server.
(Bug#39021)
Statements of the form INSERT ... SELECT .. ON
DUPLICATE KEY UPDATE could result in a server crash.
(Bug#39002)col_name =
DEFAULT
Column names constructed due to wild-card expansion done inside a stored procedure could point to freed memory if the expansion was performed after the first call to the stored procedure. (Bug#38823)
Repeated CREATE
TABLE ... SELECT statements, where the created table
contained an AUTO_INCREMENT column, could
lead to an assertion failure.
(Bug#38821)
If delayed insert failed to upgrade the lock, it did not free
the temporary memory storage used to keep newly constructed
BLOB values in memory, resulting
in a memory leak.
(Bug#38693)
A server crash resulted from concurrent execution of a
multiple-table UPDATE that used a
NATURAL or USING join
together with FLUSH
TABLES WITH READ LOCK or ALTER
TABLE for the table being updated.
(Bug#38691)
On ActiveState Perl, mysql-test-run.pl --start-and-exit started but did not exit. (Bug#38629)
Server-side cursors were not initialized properly, which could cause a server crash. (Bug#38486)
Stored procedures involving substrings could crash the server on certain platforms due to invalid memory reads. (Bug#38469)
A server crash or Valgrind warnings could result when a stored procedure selected from a view that referenced a function. (Bug#38291)
Incorrect handling of aggregate functions when loose index scan was used caused a server crash. (Bug#38195)
Queries containing a subquery with DISTINCT
and ORDER BY could cause a server crash.
(Bug#38191)
Queries with a HAVING clause could return a
spurious row.
(Bug#38072)
Use of spatial data types in prepared statements could cause memory leaks or server crashes. (Bug#37956, Bug#37671)
The server crashed if an argument to a stored procedure was a subquery that returned more than one row. (Bug#37949)
When analyzing the possible index use cases, the server was incorrectly reusing an internal structure, leading to a server crash. (Bug#37943)
A SELECT with a NULL NOT
IN condition containing a complex subquery from the
same table as in the outer select caused an assertion failure.
(Bug#37894)
For InnoDB tables, ORDER BY ...
DESC sometimes returned results in ascending order.
(Bug#37830)
If a table has a BIT NOT NULL column
c1 with a length shorter than 8 bits and some
additional NOT NULL columns
c2, ..., and a
SELECT query has a
WHERE clause of the form (c1 =
, the
query could return an unexpected result set.
(Bug#37799)constant) AND c2 ...
Nesting of IF() inside of
SUM() could cause an extreme
server slowdown.
(Bug#37662)
The MONTHNAME() and
DAYNAME() functions returned a
binary string, so that using
LOWER() or
UPPER() had no effect. Now
MONTHNAME() and
DAYNAME() return a value in
character_set_connection
character set.
(Bug#37575)
TIMEDIFF() was erroneously
treated as always returning a positive result. Also,
CAST() of
TIME values to
DECIMAL dropped the sign of
negative values.
(Bug#37553)
See also Bug#42525.
mysqlcheck used
SHOW FULL
TABLES to get the list of tables in a database. For
some problems, such as an empty .frm file
for a table, this would fail and mysqlcheck
then would neglect to check other tables in the database.
(Bug#37527)
The <=>
operator could return incorrect results when comparing
NULL to DATE,
TIME, or
DATETIME values.
(Bug#37526)
Updating a view with a subquery in the CHECK
option could cause an assertion failure.
(Bug#37460)
Statements that displayed the value of system variables (for
example, SHOW VARIABLES) expect
variable values to be encoded in
character_set_system. However,
variables set from the command line such as
basedir or
datadir were encoded using
character_set_filesystem and
not converted correctly.
(Bug#37339)
For a MyISAM table with CHECKSUM =
1 and ROW_FORMAT = DYNAMIC table
options, a data consistency check (maximum record length) could
fail and cause the table to be marked as corrupted.
(Bug#37310)
The max_length result set metadata value was
calculated incorrectly under some circumstances.
(Bug#37301)
CREATE INDEX could crash with
InnoDB plugin 1.0.1.
(Bug#37284)
Certain boolean-mode FULLTEXT searches that
used the truncation operator did not return matching records and
calculated relevance incorrectly.
(Bug#37245)
The NO_BACKSLASH_ESCAPES SQL
mode was ignored for
LOAD DATA
INFILE and SELECT INTO ... OUTFILE.
The setting is taken into account now.
(Bug#37114)
On a 32-bit server built without big tables support, the offset
argument in a LIMIT clause might be truncated
due to a 64-bit to 32-bit cast.
(Bug#37075)
If the server failed to expire binary log files at startup, it could crash. (Bug#37027)
The code for the ut_usectime() function in
InnoDB did not handle errors from the
gettimeofday() system call. Now it retries
gettimeofday() several times and updates
the value of the
Innodb_row_lock_time_max
status variable only if ut_usectime() was
successful.
(Bug#36819)
Use of CONVERT() with
GROUP BY to convert numeric values to
CHAR could return truncated
results.
(Bug#36772)
A query which had an ORDER BY DESC clause
that is satisfied with a reverse range scan could cause a server
crash for some specific CPU/compiler combinations.
(Bug#36639)
Dumping information about locks in use by sending a
SIGHUP signal to the server or by invoking
the mysqladmin debug command could lead to a
server crash in debug builds or to undefined behavior in
production builds.
(Bug#36579)
The mysql client, when built with Visual Studio 2005, did not display Japanese characters. (Bug#36279)
When the fractional part in a multiplication of
DECIMAL values overflowed, the
server truncated the first operand rather than the longest. Now
the server truncates so as to produce more precise
multiplications.
(Bug#36270)
A read past the end of the string could occur while parsing the
value of the
--innodb-data-file-path option.
(Bug#36149)
Host name values in SQL statements were not being checked for
'@', which is illegal according to RFC952.
(Bug#35924)
The UUID() function returned
UUIDs with the wrong time; this was because the offset for the
time part in UUIDs was miscalculated.
(Bug#35848)
SHOW CREATE TABLE did not display
a printable value for the default value of
BIT columns.
(Bug#35796)
mysql_install_db failed on machines that had
the host name set to localhost.
(Bug#35754)
Dynamic plugins failed to load on i5/OS. (Bug#35743)
Freeing of an internal parser stack during parsing of complex stored programs caused a server crash. (Bug#35577, Bug#37269, Bug#37228)
The max_length metadata value was calculated
incorrectly for the FORMAT()
function, which could cause incorrect result set metadata to be
sent to clients.
(Bug#35558)
Index scans performed with the sort_union()
access method returned wrong results, caused memory to be
leaked, and caused temporary files to be deleted when the limit
set by sort_buffer_size was
reached.
(Bug#35477, Bug#35478)
If the server crashed with an InnoDB error
due to unavailability of undo slots, errors could persist during
rollback when the server was restarted: There are two
UNDO slot caches (for
INSERT and
UPDATE). If all slots end up in
one of the slot caches, a request for a slot from the other slot
cache would fail. This can happen if the request is for an
UPDATE slot and all slots are in
the INSERT slot cache, or vice
versa.
(Bug#35352)
For InnoDB tables, ALTER TABLE
DROP failed if the name of the column to be dropped
began with “foreign”.
(Bug#35220)
perror on Windows did not know about Win32 system error codes. (Bug#34825)
EXPLAIN EXTENDED evaluation of aggregate
functions that required a temporary table caused a server crash.
(Bug#34773)
Queries of the form SELECT ... WHERE
failed
when the server used a single-byte character set and the client
used a multi-byte character set.
(Bug#34760)string = ANY(...)
See also Bug#20835.
Using OPTIMIZE TABLE as the first
statement on an InnoDB table with an
AUTO_INCREMENT column could cause a server
crash.
(Bug#34286)
mysql_install_db failed if the server was
running with an SQL mode of
TRADITIONAL. This program now
resets the SQL mode internally to avoid this problem.
(Bug#34159)
Changes to build files were made to enable the MySQL distribution to compile on Microsoft Visual C++ Express 2008. (Bug#33907)
The mysql client incorrectly parsed statements containing the word “delimiter” in mid-statement.
This fix is different from the one applied for this bug in MySQL 5.0.66. (Bug#33812)
See also Bug#38158.
For a stored procedure containing a SELECT * ... RIGHT
JOIN query, execution failed for the second call.
(Bug#33811)
Previously, use of index hints with views (which do not have indexes) produced the error ERROR 1221 (HY000): Incorrect usage of USE/IGNORE INDEX and VIEW. Now this produces ERROR 1176 (HY000): Key '...' doesn't exist in table '...', the same error as for base tables without an appropriate index. (Bug#33461)
Cached queries that used 256 or more tables were not properly
cached, so that later query invalidation due to a
TRUNCATE
TABLE for one of the tables caused the server to hang.
(Bug#33362)
Some division operations produced a result with incorrect precision. (Bug#31616)
mysql_upgrade attempted to use the
/proc file system even on systems that do
not have it.
(Bug#31605)
mysqldump could fail to dump views containing a large number of columns. (Bug#31434)
Queries executed using join buffering of
BIT columns could produce
incorrect results.
(Bug#31399)
ALTER TABLE CONVERT TO CHARACTER SET did not
convert TINYTEXT or
MEDIUMTEXT columns to a longer
text type if necessary when converting the column to a different
character set.
(Bug#31291)
For installation on Solaris using pkgadd
packages, the mysql_install_db script was
generated in the scripts directory, but the
temporary files used during the process were left there and not
deleted.
(Bug#31052)
Several MySQL programs could fail if the HOME
environment variable had an empty value.
(Bug#30394)
On NetWare, mysql_install_db could appear to execute normally even if it failed to create the initial databases. (Bug#30129)
The Serbian translation for the
ER_INCORRECT_GLOBAL_LOCAL_VAR
error was corrected.
(Bug#29738)
XA transaction rollbacks could result in corrupted transaction states and a server crash. (Bug#28323)
The BUILD/check-cpu build script failed if gcc had a different name (such as gcc.real on Debian). (Bug#27526)
On Windows, Visual Studio does not take into account some x86
hardware limitations, which led to incorrect results converting
large DOUBLE values to unsigned
BIGINT values.
(Bug#27483)
SSL support was not included in some “generic” RPM packages. (Bug#26760)
In some cases, the parser interpreted the ;
character as the end of input and misinterpreted stored program
definitions.
(Bug#26030)
The Questions status variable
is intended as a count of statements sent by clients to the
server, but was also counting statements executed within stored
routines.
(Bug#24289)
For access to the
INFORMATION_SCHEMA.VIEWS table, the
server did not check the SHOW
VIEW and SELECT
privileges, leading to inconsistency between output from that
table and the SHOW CREATE VIEW
statement.
(Bug#22763)
The FLUSH
PRIVILEGES statement did not produce an error when it
failed.
(Bug#21226)
A race condition between the mysqld.exe server and the Windows service manager could lead to inability to stop the server from the service manager. (Bug#20430)
mysqld_safe would sometimes fail to remove
the pid file for the old mysql process after
a crash. As a result, the server would fail to start due to a
false A mysqld process already exists...
error.
(Bug#11122)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.74). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Bugs fixed:
Replication: When rotating relay log files, the slave deletes relay log files and then edits the relay log index file. Formerly, if the slave shut down unexpectedly between these two events, the relay log index file could then reference relay logs that no longer existed. Depending on the circumstances, this could when restarting the slave cause either a race condition or the failure of replication. (Bug#38826, Bug#39325)
In example option files provided in MySQL distributions, the
thread_stack value was
increased from 64K to 128K.
(Bug#41577)
SET PASSWORD caused a server
crash if the account name was given as
CURRENT_USER().
(Bug#41456)
The
INFORMATION_SCHEMA.SCHEMA_PRIVILEGES
table was limited to 7680 rows.
(Bug#41079)
In debug builds, obsolete debug code could be used to crash the server. (Bug#41041)
Some queries that used a “range checked for each record” scan could return incorrect results. (Bug#40974)
Certain SELECT queries could fail
with a Duplicate entry error.
(Bug#40953)
IF(..., CAST( as
an argument to an aggregate function could cause an assertion
failure.
(Bug#40761)longtext_val AS
UNSIGNED), signed_val)
The server crashed if an integer field in a CSV file did not have delimiting quotes. (Bug#39616)
Creating a table with a comment of 62 characters or longer caused a server crash. (Bug#39591)
InnoDB could hang trying to open an adaptive
hash index.
(Bug#39483)
Use of spatial data types in prepared statements could cause memory leaks or server crashes. (Bug#37956, Bug#37671)
The MONTHNAME() and
DAYNAME() functions returned a
binary string, so that using
LOWER() or
UPPER() had no effect. Now
MONTHNAME() and
DAYNAME() return a value in
character_set_connection
character set.
(Bug#37575)
Certain boolean-mode FULLTEXT searches that
used the truncation operator did not return matching records and
calculated relevance incorrectly.
(Bug#37245)
The code for the ut_usectime() function in
InnoDB did not handle errors from the
gettimeofday() system call. Now it retries
gettimeofday() several times and updates
the value of the
Innodb_row_lock_time_max
status variable only if ut_usectime() was
successful.
(Bug#36819)
A read past the end of the string could occur while parsing the
value of the
--innodb-data-file-path option.
(Bug#36149)
SHOW CREATE TABLE did not display
a printable value for the default value of
BIT columns.
(Bug#35796)
The max_length metadata value was calculated
incorrectly for the FORMAT()
function, which could cause incorrect result set metadata to be
sent to clients.
(Bug#35558)
EXPLAIN EXTENDED evaluation of aggregate
functions that required a temporary table caused a server crash.
(Bug#34773)
The mysql client incorrectly parsed statements containing the word “delimiter” in mid-statement.
This fix is different from the one applied for this bug in MySQL 5.0.66. (Bug#33812)
See also Bug#38158.
Queries executed using join buffering of
BIT columns could produce
incorrect results.
(Bug#31399)
ALTER TABLE CONVERT TO CHARACTER SET did not
convert TINYTEXT or
MEDIUMTEXT columns to a longer
text type if necessary when converting the column to a different
character set.
(Bug#31291)
On Windows, Visual Studio does not take into account some x86
hardware limitations, which led to incorrect results converting
large DOUBLE values to unsigned
BIGINT values.
(Bug#27483)
SSL support was not included in some “generic” RPM packages. (Bug#26760)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.67.
Functionality added or changed:
Security Enhancement:
To enable stricter control over the location from which
user-defined functions can be loaded, the
plugin_dir system variable has
been backported from MySQL 5.1. If the value is nonempty,
user-defined function object files can be loaded only from the
directory named by this variable. If the value is empty, the
behavior that is used prior to the inclusion of
plugin_dir applies: The UDF
object files must be located in a directory that is searched by
your system's dynamic linker.
(Bug#37428)
Previously, index hints did not work for
FULLTEXT searches. Now they work as follows:
For natural language mode searches, index hints are silently
ignored. For example, IGNORE INDEX(i) is
ignored with no warning and the index is still used.
For boolean mode searches, index hints are honored. (Bug#38842)
Bugs fixed:
Important Change: Security Fix: Additional corrections were made for the symlink-related privilege problem originally addressed in MySQL 5.0.60. The original fix did not correctly handle the data directory path name if it contained symlinked directories in its path, and the check was made only at table-creation time, not at table-opening time later. (Bug#32167, CVE-2008-2079)
See also Bug#39277.
Security Enhancement:
The server consumed excess memory while parsing statements with
hundreds or thousands of nested boolean conditions (such as
OR (OR ... (OR ... ))). This could lead to a
server crash or incorrect statement execution, or cause other
client statements to fail due to lack of memory. The latter
result constitutes a denial of service.
(Bug#38296)
Incompatible Change:
There were some problems using DllMain()
hook functions on Windows that automatically do global and
per-thread initialization for
libmysqld.dll:
Per-thread initialization: MySQL internally counts the
number of active threads, which causes a delay in
my_end() if not all threads have
exited. But there are threads that can be started either by
Windows internally (often in TCP/IP scenarios) or by users.
Those threads do not necessarily use
libmysql.dll functionality but still
contribute to the open-thread count. (One symptom is a
five-second delay in times for PHP scripts to finish.)
Process-initialization:
my_init() calls
WSAStartup that itself loads DLLs and
can lead to a deadlock in the Windows loader.
To correct these problems, DLL initialization code now is not
invoked from libmysql.dll by default. To
obtain the previous behavior (DLL initialization code will be
called), set the LIBMYSQL_DLLINIT environment
variable to any value. This variable exists only to prevent
breakage of existing Windows-only applications that do not call
mysql_thread_init() and work
okay today. Use of LIBMYSQL_DLLINIT is
discouraged and is removed in MySQL 6.0.
(Bug#37226, Bug#33031)
Incompatible Change:
SHOW STATUS took a lot of CPU
time for calculating the value of the
Innodb_buffer_pool_pages_latched
status variable. Now this variable is calculated and included in
the output of SHOW STATUS only if
the UNIV_DEBUG symbol is defined at MySQL
build time.
(Bug#36600)
Incompatible Change:
In connection with view creation, the server created
arc directories inside database directories
and maintained useless copies of .frm files
there. Creation and renaming procedures of those copies as well
as creation of arc directories has been
discontinued.
This change does cause a problem when downgrading to older server versions which manifests itself under these circumstances:
Create a view v_orig in MySQL 5.0.72 or
higher.
Rename the view to v_new and then back to
v_orig.
Downgrade to an older 5.0.x server and run mysql_upgrade.
Try to rename v_orig to
v_new again. This operation fails.
As a workaround to avoid this problem, use either of these approaches:
Dump your data using mysqldump before downgrading and reload the dump file after downgrading.
Instead of renaming a view after the downgrade, drop it and recreate it.
The downgrade problem introduced by the fix for this bug has been addressed as Bug#40021. (Bug#17823)
CHECK TABLE ... FOR
UPGRADE did not check for incompatible collation
changes made in MySQL 5.0.48 (Bug#27562, Bug#29461, Bug#29499).
This also affects mysqlcheck and
mysql_upgrade, which cause that statement to
be executed. See Section 2.18.3, “Checking Whether Table Indexes Must Be Rebuilt”.
(Bug#40984)
See also Bug#39585.
The FEDERATED handler had a memory
leak.
(Bug#40875)
Prepared statements allowed invalid dates to be inserted when
the ALLOW_INVALID_DATES SQL
mode was not enabled.
(Bug#40365)
mc.exe is no longer needed to compile MySQL on Windows. This makes it possible to build MySQL from source using Visual Studio Express 2008. (Bug#40280)
Support for the revision field in
.frm files has been removed. This addresses
the downgrading problem introduced by the fix for Bug#17823.
(Bug#40021)
If the operating system is configured to return leap seconds
from OS time calls or if the MySQL server uses a time zone
definition that has leap seconds, functions such as
NOW() could return a value having
a time part that ends with :59:60 or
:59:61. If such values are inserted into a
table, they would be dumped as is by
mysqldump but considered invalid when
reloaded, leading to backup/restore problems.
Now leap second values are returned with a time part that ends
with :59:59. This means that a function such
as NOW() can return the same
value for two or three consecutive seconds during the leap
second. It remains true that literal temporal values having a
time part that ends with :59:60 or
:59:61 are considered invalid.
For additional details about leap-second handling, see Section 9.7.2, “Time Zone Leap Second Support”. (Bug#39920)
The server could crash during a sort-order optimization of a dependent subquery. (Bug#39844)
With the ONLY_FULL_GROUP_BY
SQL mode enabled, the check for nonaggregated columns in queries
with aggregate functions, but without a GROUP
BY clause was treating all the parts of the query as
if they were in the select list. This is fixed by ignoring the
nonaggregated columns in the WHERE clause.
(Bug#39656)
CHECK TABLE failed for
MyISAM
INFORMATION_SCHEMA tables.
(Bug#39541)
For a TIMESTAMP column in an
InnoDB table, testing the column with
multiple conditions in the WHERE clause
caused a server crash.
(Bug#39353)
The server returned a column type of
VARBINARY rather than
DATE as the result from the
COALESCE(),
IFNULL(),
IF(),
GREATEST(), or
LEAST() functions or
CASE expression if the result was
obtained using filesort in an anonymous
temporary table during the query execution.
(Bug#39283)
References to local variables in stored procedures are replaced
with
NAME_CONST( when written to the
binary log. However, an “illegal mix of collation”
error might occur when executing the log contents if the value's
collation differed from that of the variable. Now information
about the variable collation is written as well.
(Bug#39182)name,
value)
Some recent releases for Solaris 10 were built on Solaris 10 U5,
which included a new version of libnsl.so
that does not work on U4 or earlier. To correct this, Solaris 10
builds now are created on machines that do not have that
upgraded libnsl.so, so that they will work
on Solaris 10 installations both with and without the upgraded
libnsl.so.
(Bug#39074)
With binary logging enabled CREATE
VIEW was subject to possible buffer overwrite and a
server crash.
(Bug#39040)
Queries of the form SELECT ... REGEXP BINARY
NULL could lead to a hung or crashed server.
(Bug#39021)
Statements of the form INSERT ... SELECT .. ON
DUPLICATE KEY UPDATE could result in a server crash.
(Bug#39002)col_name =
DEFAULT
Column names constructed due to wild-card expansion done inside a stored procedure could point to freed memory if the expansion was performed after the first call to the stored procedure. (Bug#38823)
Repeated CREATE
TABLE ... SELECT statements, where the created table
contained an AUTO_INCREMENT column, could
lead to an assertion failure.
(Bug#38821)
If delayed insert failed to upgrade the lock, it did not free
the temporary memory storage used to keep newly constructed
BLOB values in memory, resulting
in a memory leak.
(Bug#38693)
A server crash resulted from concurrent execution of a
multiple-table UPDATE that used a
NATURAL or USING join
together with FLUSH
TABLES WITH READ LOCK or ALTER
TABLE for the table being updated.
(Bug#38691)
On ActiveState Perl, mysql-test-run.pl --start-and-exit started but did not exit. (Bug#38629)
Server-side cursors were not initialized properly, which could cause a server crash. (Bug#38486)
Stored procedures involving substrings could crash the server on certain platforms due to invalid memory reads. (Bug#38469)
A server crash or Valgrind warnings could result when a stored procedure selected from a view that referenced a function. (Bug#38291)
Incorrect handling of aggregate functions when loose index scan was used caused a server crash. (Bug#38195)
Queries containing a subquery with DISTINCT
and ORDER BY could cause a server crash.
(Bug#38191)
Queries with a HAVING clause could return a
spurious row.
(Bug#38072)
The server crashed if an argument to a stored procedure was a subquery that returned more than one row. (Bug#37949)
When analyzing the possible index use cases, the server was incorrectly reusing an internal structure, leading to a server crash. (Bug#37943)
A SELECT with a NULL NOT
IN condition containing a complex subquery from the
same table as in the outer select caused an assertion failure.
(Bug#37894)
For InnoDB tables, ORDER BY ...
DESC sometimes returned results in ascending order.
(Bug#37830)
If a table has a BIT NOT NULL column
c1 with a length shorter than 8 bits and some
additional NOT NULL columns
c2, ..., and a
SELECT query has a
WHERE clause of the form (c1 =
, the
query could return an unexpected result set.
(Bug#37799)constant) AND c2 ...
Nesting of IF() inside of
SUM() could cause an extreme
server slowdown.
(Bug#37662)
TIMEDIFF() was erroneously
treated as always returning a positive result. Also,
CAST() of
TIME values to
DECIMAL dropped the sign of
negative values.
(Bug#37553)
See also Bug#42525.
mysqlcheck used
SHOW FULL
TABLES to get the list of tables in a database. For
some problems, such as an empty .frm file
for a table, this would fail and mysqlcheck
then would neglect to check other tables in the database.
(Bug#37527)
The <=>
operator could return incorrect results when comparing
NULL to DATE,
TIME, or
DATETIME values.
(Bug#37526)
Updating a view with a subquery in the CHECK
option could cause an assertion failure.
(Bug#37460)
Statements that displayed the value of system variables (for
example, SHOW VARIABLES) expect
variable values to be encoded in
character_set_system. However,
variables set from the command line such as
basedir or
datadir were encoded using
character_set_filesystem and
not converted correctly.
(Bug#37339)
For a MyISAM table with CHECKSUM =
1 and ROW_FORMAT = DYNAMIC table
options, a data consistency check (maximum record length) could
fail and cause the table to be marked as corrupted.
(Bug#37310)
The max_length result set metadata value was
calculated incorrectly under some circumstances.
(Bug#37301)
CREATE INDEX could crash with
InnoDB plugin 1.0.1.
(Bug#37284)
The NO_BACKSLASH_ESCAPES SQL
mode was ignored for
LOAD DATA
INFILE and SELECT INTO ... OUTFILE.
The setting is taken into account now.
(Bug#37114)
On a 32-bit server built without big tables support, the offset
argument in a LIMIT clause might be truncated
due to a 64-bit to 32-bit cast.
(Bug#37075)
If the server failed to expire binary log files at startup, it could crash. (Bug#37027)
Use of CONVERT() with
GROUP BY to convert numeric values to
CHAR could return truncated
results.
(Bug#36772)
A query which had an ORDER BY DESC clause
that is satisfied with a reverse range scan could cause a server
crash for some specific CPU/compiler combinations.
(Bug#36639)
Dumping information about locks in use by sending a
SIGHUP signal to the server or by invoking
the mysqladmin debug command could lead to a
server crash in debug builds or to undefined behavior in
production builds.
(Bug#36579)
The mysql client, when built with Visual Studio 2005, did not display Japanese characters. (Bug#36279)
When the fractional part in a multiplication of
DECIMAL values overflowed, the
server truncated the first operand rather than the longest. Now
the server truncates so as to produce more precise
multiplications.
(Bug#36270)
Host name values in SQL statements were not being checked for
'@', which is illegal according to RFC952.
(Bug#35924)
The UUID() function returned
UUIDs with the wrong time; this was because the offset for the
time part in UUIDs was miscalculated.
(Bug#35848)
mysql_install_db failed on machines that had
the host name set to localhost.
(Bug#35754)
Dynamic plugins failed to load on i5/OS. (Bug#35743)
Freeing of an internal parser stack during parsing of complex stored programs caused a server crash. (Bug#35577, Bug#37269, Bug#37228)
Index scans performed with the sort_union()
access method returned wrong results, caused memory to be
leaked, and caused temporary files to be deleted when the limit
set by sort_buffer_size was
reached.
(Bug#35477, Bug#35478)
If the server crashed with an InnoDB error
due to unavailability of undo slots, errors could persist during
rollback when the server was restarted: There are two
UNDO slot caches (for
INSERT and
UPDATE). If all slots end up in
one of the slot caches, a request for a slot from the other slot
cache would fail. This can happen if the request is for an
UPDATE slot and all slots are in
the INSERT slot cache, or vice
versa.
(Bug#35352)
For InnoDB tables, ALTER TABLE
DROP failed if the name of the column to be dropped
began with “foreign”.
(Bug#35220)
perror on Windows did not know about Win32 system error codes. (Bug#34825)
Queries of the form SELECT ... WHERE
failed
when the server used a single-byte character set and the client
used a multi-byte character set.
(Bug#34760)string = ANY(...)
See also Bug#20835.
Using OPTIMIZE TABLE as the first
statement on an InnoDB table with an
AUTO_INCREMENT column could cause a server
crash.
(Bug#34286)
mysql_install_db failed if the server was
running with an SQL mode of
TRADITIONAL. This program now
resets the SQL mode internally to avoid this problem.
(Bug#34159)
Changes to build files were made to enable the MySQL distribution to compile on Microsoft Visual C++ Express 2008. (Bug#33907)
For a stored procedure containing a SELECT * ... RIGHT
JOIN query, execution failed for the second call.
(Bug#33811)
Previously, use of index hints with views (which do not have indexes) produced the error ERROR 1221 (HY000): Incorrect usage of USE/IGNORE INDEX and VIEW. Now this produces ERROR 1176 (HY000): Key '...' doesn't exist in table '...', the same error as for base tables without an appropriate index. (Bug#33461)
Cached queries that used 256 or more tables were not properly
cached, so that later query invalidation due to a
TRUNCATE
TABLE for one of the tables caused the server to hang.
(Bug#33362)
Some division operations produced a result with incorrect precision. (Bug#31616)
mysql_upgrade attempted to use the
/proc file system even on systems that do
not have it.
(Bug#31605)
mysqldump could fail to dump views containing a large number of columns. (Bug#31434)
Several MySQL programs could fail if the HOME
environment variable had an empty value.
(Bug#30394)
On NetWare, mysql_install_db could appear to execute normally even if it failed to create the initial databases. (Bug#30129)
The Serbian translation for the
ER_INCORRECT_GLOBAL_LOCAL_VAR
error was corrected.
(Bug#29738)
XA transaction rollbacks could result in corrupted transaction states and a server crash. (Bug#28323)
The BUILD/check-cpu build script failed if gcc had a different name (such as gcc.real on Debian). (Bug#27526)
In some cases, the parser interpreted the ;
character as the end of input and misinterpreted stored program
definitions.
(Bug#26030)
The Questions status variable
is intended as a count of statements sent by clients to the
server, but was also counting statements executed within stored
routines.
(Bug#24289)
For access to the
INFORMATION_SCHEMA.VIEWS table, the
server did not check the SHOW
VIEW and SELECT
privileges, leading to inconsistency between output from that
table and the SHOW CREATE VIEW
statement.
(Bug#22763)
The FLUSH
PRIVILEGES statement did not produce an error when it
failed.
(Bug#21226)
A race condition between the mysqld.exe server and the Windows service manager could lead to inability to stop the server from the service manager. (Bug#20430)
mysqld_safe would sometimes fail to remove
the pid file for the old mysql process after
a crash. As a result, the server would fail to start due to a
false A mysqld process already exists...
error.
(Bug#11122)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.74).
If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
The libedit library was upgraded to version
2.11.
(Bug#42433)
Bugs fixed:
An attempt by a user who did not have the
SUPER privilege to kill a system
thread could cause a server crash.
(Bug#43748)
Tables could enter open table cache for a thread without being properly cleaned up, leading to a server crash. (Bug#42419)
The SSL certficates included with MySQL distributions were regenerated because the previous ones had expired. (Bug#42366)
Some queries using NAME_CONST(.. COLLATE
...) led to a server crash due to a failed type cast.
(Bug#42014)
DATE_FORMAT() could cause a
server crash for year-zero dates.
(Bug#41470)
SET PASSWORD caused a server
crash if the account name was given as
CURRENT_USER().
(Bug#41456)
When substituting system constant functions with a constant
result, the server was not expecting NULL
function return values and could crash.
(Bug#41437)
Creating a table with a comment of 62 characters or longer caused a server crash. (Bug#39591)
EXPLAIN EXTENDED evaluation of aggregate
functions that required a temporary table caused a server crash.
(Bug#34773)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.72). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Previously, index hints did not work for
FULLTEXT searches. Now they work as follows:
For natural language mode searches, index hints are silently
ignored. For example, IGNORE INDEX(i) is
ignored with no warning and the index is still used.
For boolean mode searches, index hints are honored. (Bug#38842)
Bugs fixed:
CHECK TABLE ... FOR
UPGRADE did not check for incompatible collation
changes made in MySQL 5.0.48 (Bug#27562, Bug#29461, Bug#29499).
This also affects mysqlcheck and
mysql_upgrade, which cause that statement to
be executed. See Section 2.18.3, “Checking Whether Table Indexes Must Be Rebuilt”.
(Bug#40984)
See also Bug#39585.
The FEDERATED handler had a memory
leak.
(Bug#40875)
Prepared statements allowed invalid dates to be inserted when
the ALLOW_INVALID_DATES SQL
mode was not enabled.
(Bug#40365)
Support for the revision field in
.frm files has been removed. This addresses
the downgrading problem introduced by the fix for Bug#17823.
(Bug#40021)
If the operating system is configured to return leap seconds
from OS time calls or if the MySQL server uses a time zone
definition that has leap seconds, functions such as
NOW() could return a value having
a time part that ends with :59:60 or
:59:61. If such values are inserted into a
table, they would be dumped as is by
mysqldump but considered invalid when
reloaded, leading to backup/restore problems.
Now leap second values are returned with a time part that ends
with :59:59. This means that a function such
as NOW() can return the same
value for two or three consecutive seconds during the leap
second. It remains true that literal temporal values having a
time part that ends with :59:60 or
:59:61 are considered invalid.
For additional details about leap-second handling, see Section 9.7.2, “Time Zone Leap Second Support”. (Bug#39920)
With the ONLY_FULL_GROUP_BY
SQL mode enabled, the check for nonaggregated columns in queries
with aggregate functions, but without a GROUP
BY clause was treating all the parts of the query as
if they were in the select list. This is fixed by ignoring the
nonaggregated columns in the WHERE clause.
(Bug#39656)
CHECK TABLE failed for
MyISAM
INFORMATION_SCHEMA tables.
(Bug#39541)
With binary logging enabled CREATE
VIEW was subject to possible buffer overwrite and a
server crash.
(Bug#39040)
Queries with a HAVING clause could return a
spurious row.
(Bug#38072)
TIMEDIFF() was erroneously
treated as always returning a positive result. Also,
CAST() of
TIME values to
DECIMAL dropped the sign of
negative values.
(Bug#37553)
See also Bug#42525.
mysqlcheck used
SHOW FULL
TABLES to get the list of tables in a database. For
some problems, such as an empty .frm file
for a table, this would fail and mysqlcheck
then would neglect to check other tables in the database.
(Bug#37527)
Updating a view with a subquery in the CHECK
option could cause an assertion failure.
(Bug#37460)
Statements that displayed the value of system variables (for
example, SHOW VARIABLES) expect
variable values to be encoded in
character_set_system. However,
variables set from the command line such as
basedir or
datadir were encoded using
character_set_filesystem and
not converted correctly.
(Bug#37339)
CREATE INDEX could crash with
InnoDB plugin 1.0.1.
(Bug#37284)
Use of CONVERT() with
GROUP BY to convert numeric values to
CHAR could return truncated
results.
(Bug#36772)
The mysql client, when built with Visual Studio 2005, did not display Japanese characters. (Bug#36279)
perror on Windows did not know about Win32 system error codes. (Bug#34825)
Queries of the form SELECT ... WHERE
failed
when the server used a single-byte character set and the client
used a multi-byte character set.
(Bug#34760)string = ANY(...)
See also Bug#20835.
For a stored procedure containing a SELECT * ... RIGHT
JOIN query, execution failed for the second call.
(Bug#33811)
Previously, use of index hints with views (which do not have indexes) produced the error ERROR 1221 (HY000): Incorrect usage of USE/IGNORE INDEX and VIEW. Now this produces ERROR 1176 (HY000): Key '...' doesn't exist in table '...', the same error as for base tables without an appropriate index. (Bug#33461)
Some division operations produced a result with incorrect precision. (Bug#31616)
A race condition between the mysqld.exe server and the Windows service manager could lead to inability to stop the server from the service manager. (Bug#20430)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.72).
If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Previously, index hints did not work for
FULLTEXT searches. Now they work as follows:
For natural language mode searches, index hints are silently
ignored. For example, IGNORE INDEX(i) is
ignored with no warning and the index is still used.
For boolean mode searches, index hints are honored. (Bug#38842)
Bugs fixed:
MySQL Cluster: Packaging:
Packages for MySQL Cluster were missing the
libndbclient.so and
libndbclient.a files.
(Bug#42278)
Support for the revision field in
.frm files has been removed. This addresses
the downgrading problem introduced by the fix for Bug#17823.
(Bug#40021)
If the operating system is configured to return leap seconds
from OS time calls or if the MySQL server uses a time zone
definition that has leap seconds, functions such as
NOW() could return a value having
a time part that ends with :59:60 or
:59:61. If such values are inserted into a
table, they would be dumped as is by
mysqldump but considered invalid when
reloaded, leading to backup/restore problems.
Now leap second values are returned with a time part that ends
with :59:59. This means that a function such
as NOW() can return the same
value for two or three consecutive seconds during the leap
second. It remains true that literal temporal values having a
time part that ends with :59:60 or
:59:61 are considered invalid.
For additional details about leap-second handling, see Section 9.7.2, “Time Zone Leap Second Support”. (Bug#39920)
Queries of the form SELECT ... WHERE
failed
when the server used a single-byte character set and the client
used a multi-byte character set.
(Bug#34760)string = ANY(...)
See also Bug#20835.
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.70). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Bugs fixed:
Incompatible Change:
In connection with view creation, the server created
arc directories inside database directories
and maintained useless copies of .frm files
there. Creation and renaming procedures of those copies as well
as creation of arc directories has been
discontinued.
This change does cause a problem when downgrading to older server versions which manifests itself under these circumstances:
Create a view v_orig in MySQL 5.0.72 or
higher.
Rename the view to v_new and then back to
v_orig.
Downgrade to an older 5.0.x server and run mysql_upgrade.
Try to rename v_orig to
v_new again. This operation fails.
As a workaround to avoid this problem, use either of these approaches:
Dump your data using mysqldump before downgrading and reload the dump file after downgrading.
Instead of renaming a view after the downgrade, drop it and recreate it.
The downgrade problem introduced by the fix for this bug has been addressed as Bug#40021. (Bug#17823)
mc.exe is no longer needed to compile MySQL on Windows. This makes it possible to build MySQL from source using Visual Studio Express 2008. (Bug#40280)
The server could crash during a sort-order optimization of a dependent subquery. (Bug#39844)
The server returned a column type of
VARBINARY rather than
DATE as the result from the
COALESCE(),
IFNULL(),
IF(),
GREATEST(), or
LEAST() functions or
CASE expression if the result was
obtained using filesort in an anonymous
temporary table during the query execution.
(Bug#39283)
References to local variables in stored procedures are replaced
with
NAME_CONST( when written to the
binary log. However, an “illegal mix of collation”
error might occur when executing the log contents if the value's
collation differed from that of the variable. Now information
about the variable collation is written as well.
(Bug#39182)name,
value)
Some recent releases for Solaris 10 were built on Solaris 10 U5,
which included a new version of libnsl.so
that does not work on U4 or earlier. To correct this, Solaris 10
builds now are created on machines that do not have that
upgraded libnsl.so, so that they will work
on Solaris 10 installations both with and without the upgraded
libnsl.so.
(Bug#39074)
Column names constructed due to wild-card expansion done inside a stored procedure could point to freed memory if the expansion was performed after the first call to the stored procedure. (Bug#38823)
If delayed insert failed to upgrade the lock, it did not free
the temporary memory storage used to keep newly constructed
BLOB values in memory, resulting
in a memory leak.
(Bug#38693)
A server crash resulted from concurrent execution of a
multiple-table UPDATE that used a
NATURAL or USING join
together with FLUSH
TABLES WITH READ LOCK or ALTER
TABLE for the table being updated.
(Bug#38691)
On ActiveState Perl, mysql-test-run.pl --start-and-exit started but did not exit. (Bug#38629)
Stored procedures involving substrings could crash the server on certain platforms due to invalid memory reads. (Bug#38469)
The server crashed if an argument to a stored procedure was a subquery that returned more than one row. (Bug#37949)
When analyzing the possible index use cases, the server was incorrectly reusing an internal structure, leading to a server crash. (Bug#37943)
A SELECT with a NULL NOT
IN condition containing a complex subquery from the
same table as in the outer select caused an assertion failure.
(Bug#37894)
On a 32-bit server built without big tables support, the offset
argument in a LIMIT clause might be truncated
due to a 64-bit to 32-bit cast.
(Bug#37075)
Host name values in SQL statements were not being checked for
'@', which is illegal according to RFC952.
(Bug#35924)
mysql_install_db failed on machines that had
the host name set to localhost.
(Bug#35754)
Dynamic plugins failed to load on i5/OS. (Bug#35743)
XA transaction rollbacks could result in corrupted transaction states and a server crash. (Bug#28323)
The Questions status variable
is intended as a count of statements sent by clients to the
server, but was also counting statements executed within stored
routines.
(Bug#24289)
For access to the
INFORMATION_SCHEMA.VIEWS table, the
server did not check the SHOW
VIEW and SELECT
privileges, leading to inconsistency between output from that
table and the SHOW CREATE VIEW
statement.
(Bug#22763)
mysqld_safe would sometimes fail to remove
the pid file for the old mysql process after
a crash. As a result, the server would fail to start due to a
false A mysqld process already exists...
error.
(Bug#11122)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.68). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Security Enhancement:
To enable stricter control over the location from which
user-defined functions can be loaded, the
plugin_dir system variable has
been backported from MySQL 5.1. If the value is nonempty,
user-defined function object files can be loaded only from the
directory named by this variable. If the value is empty, the
behavior that is used prior to the inclusion of
plugin_dir applies: The UDF
object files must be located in a directory that is searched by
your system's dynamic linker.
(Bug#37428)
Bugs fixed:
Important Change: Security Fix: Additional corrections were made for the symlink-related privilege problem originally addressed in MySQL 5.0.60. The original fix did not correctly handle the data directory path name if it contained symlinked directories in its path, and the check was made only at table-creation time, not at table-opening time later. (Bug#32167, CVE-2008-2079)
See also Bug#39277.
Incompatible Change:
There were some problems using DllMain()
hook functions on Windows that automatically do global and
per-thread initialization for
libmysqld.dll:
Per-thread initialization: MySQL internally counts the
number of active threads, which causes a delay in
my_end() if not all threads have
exited. But there are threads that can be started either by
Windows internally (often in TCP/IP scenarios) or by users.
Those threads do not necessarily use
libmysql.dll functionality but still
contribute to the open-thread count. (One symptom is a
five-second delay in times for PHP scripts to finish.)
Process-initialization:
my_init() calls
WSAStartup that itself loads DLLs and
can lead to a deadlock in the Windows loader.
To correct these problems, DLL initialization code now is not
invoked from libmysql.dll by default. To
obtain the previous behavior (DLL initialization code will be
called), set the LIBMYSQL_DLLINIT environment
variable to any value. This variable exists only to prevent
breakage of existing Windows-only applications that do not call
mysql_thread_init() and work
okay today. Use of LIBMYSQL_DLLINIT is
discouraged and is removed in MySQL 6.0.
(Bug#37226, Bug#33031)
For a TIMESTAMP column in an
InnoDB table, testing the column with
multiple conditions in the WHERE clause
caused a server crash.
(Bug#39353)
Queries of the form SELECT ... REGEXP BINARY
NULL could lead to a hung or crashed server.
(Bug#39021)
Statements of the form INSERT ... SELECT .. ON
DUPLICATE KEY UPDATE could result in a server crash.
(Bug#39002)col_name =
DEFAULT
Repeated CREATE
TABLE ... SELECT statements, where the created table
contained an AUTO_INCREMENT column, could
lead to an assertion failure.
(Bug#38821)
A server crash or Valgrind warnings could result when a stored procedure selected from a view that referenced a function. (Bug#38291)
Incorrect handling of aggregate functions when loose index scan was used caused a server crash. (Bug#38195)
If a table has a BIT NOT NULL column
c1 with a length shorter than 8 bits and some
additional NOT NULL columns
c2, ..., and a
SELECT query has a
WHERE clause of the form (c1 =
, the
query could return an unexpected result set.
(Bug#37799)constant) AND c2 ...
The <=>
operator could return incorrect results when comparing
NULL to DATE,
TIME, or
DATETIME values.
(Bug#37526)
For a MyISAM table with CHECKSUM =
1 and ROW_FORMAT = DYNAMIC table
options, a data consistency check (maximum record length) could
fail and cause the table to be marked as corrupted.
(Bug#37310)
The max_length result set metadata value was
calculated incorrectly under some circumstances.
(Bug#37301)
The NO_BACKSLASH_ESCAPES SQL
mode was ignored for
LOAD DATA
INFILE and SELECT INTO ... OUTFILE.
The setting is taken into account now.
(Bug#37114)
A query which had an ORDER BY DESC clause
that is satisfied with a reverse range scan could cause a server
crash for some specific CPU/compiler combinations.
(Bug#36639)
Dumping information about locks in use by sending a
SIGHUP signal to the server or by invoking
the mysqladmin debug command could lead to a
server crash in debug builds or to undefined behavior in
production builds.
(Bug#36579)
When the fractional part in a multiplication of
DECIMAL values overflowed, the
server truncated the first operand rather than the longest. Now
the server truncates so as to produce more precise
multiplications.
(Bug#36270)
Changes to build files were made to enable the MySQL distribution to compile on Microsoft Visual C++ Express 2008. (Bug#33907)
mysqldump could fail to dump views containing a large number of columns. (Bug#31434)
Several MySQL programs could fail if the HOME
environment variable had an empty value.
(Bug#30394)
The BUILD/check-cpu build script failed if gcc had a different name (such as gcc.real on Debian). (Bug#27526)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.66a). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Bugs fixed:
Security Enhancement:
The server consumed excess memory while parsing statements with
hundreds or thousands of nested boolean conditions (such as
OR (OR ... (OR ... ))). This could lead to a
server crash or incorrect statement execution, or cause other
client statements to fail due to lack of memory. The latter
result constitutes a denial of service.
(Bug#38296)
Incompatible Change:
SHOW STATUS took a lot of CPU
time for calculating the value of the
Innodb_buffer_pool_pages_latched
status variable. Now this variable is calculated and included in
the output of SHOW STATUS only if
the UNIV_DEBUG symbol is defined at MySQL
build time.
(Bug#36600)
Server-side cursors were not initialized properly, which could cause a server crash. (Bug#38486)
Queries containing a subquery with DISTINCT
and ORDER BY could cause a server crash.
(Bug#38191)
For InnoDB tables, ORDER BY ...
DESC sometimes returned results in ascending order.
(Bug#37830)
Nesting of IF() inside of
SUM() could cause an extreme
server slowdown.
(Bug#37662)
If the server failed to expire binary log files at startup, it could crash. (Bug#37027)
The UUID() function returned
UUIDs with the wrong time; this was because the offset for the
time part in UUIDs was miscalculated.
(Bug#35848)
Freeing of an internal parser stack during parsing of complex stored programs caused a server crash. (Bug#35577, Bug#37269, Bug#37228)
Index scans performed with the sort_union()
access method returned wrong results, caused memory to be
leaked, and caused temporary files to be deleted when the limit
set by sort_buffer_size was
reached.
(Bug#35477, Bug#35478)
If the server crashed with an InnoDB error
due to unavailability of undo slots, errors could persist during
rollback when the server was restarted: There are two
UNDO slot caches (for
INSERT and
UPDATE). If all slots end up in
one of the slot caches, a request for a slot from the other slot
cache would fail. This can happen if the request is for an
UPDATE slot and all slots are in
the INSERT slot cache, or vice
versa.
(Bug#35352)
For InnoDB tables, ALTER TABLE
DROP failed if the name of the column to be dropped
began with “foreign”.
(Bug#35220)
Using OPTIMIZE TABLE as the first
statement on an InnoDB table with an
AUTO_INCREMENT column could cause a server
crash.
(Bug#34286)
mysql_install_db failed if the server was
running with an SQL mode of
TRADITIONAL. This program now
resets the SQL mode internally to avoid this problem.
(Bug#34159)
Cached queries that used 256 or more tables were not properly
cached, so that later query invalidation due to a
TRUNCATE
TABLE for one of the tables caused the server to hang.
(Bug#33362)
mysql_upgrade attempted to use the
/proc file system even on systems that do
not have it.
(Bug#31605)
On NetWare, mysql_install_db could appear to execute normally even if it failed to create the initial databases. (Bug#30129)
The Serbian translation for the
ER_INCORRECT_GLOBAL_LOCAL_VAR
error was corrected.
(Bug#29738)
In some cases, the parser interpreted the ;
character as the end of input and misinterpreted stored program
definitions.
(Bug#26030)
The FLUSH
PRIVILEGES statement did not produce an error when it
failed.
(Bug#21226)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.51b.
Functionality added or changed:
Security Enhancement:
To enable stricter control over the location from which
user-defined functions can be loaded, the
plugin_dir system variable has
been backported from MySQL 5.1. If the value is nonempty,
user-defined function object files can be loaded only from the
directory named by this variable. If the value is empty, the
behavior that is used prior to the inclusion of
plugin_dir applies: The UDF
object files must be located in a directory that is searched by
your system's dynamic linker.
(Bug#37428)
Important Change: Incompatible Change:
The FEDERATED storage engine is now disabled
by default in the .cnf files shipped with
MySQL distributions (my-huge.cnf,
my-medium.cnf, and so forth). This affects
server behavior only if you install one of these files.
(Bug#37069)
Cluster API: Important Change:
Because NDB_LE_MemoryUsage.page_size_kb shows
memory page sizes in bytes rather than kilobytes, it has been
renamed to page_size_bytes. The name
page_size_kb is now deprecated and thus
subject to removal in a future release, although it currently
remains supported for reasons of backward compatibility. See
The Ndb_logevent_type Type, for more information about
NDB_LE_MemoryUsage.
(Bug#30271)
Important Change:
Some changes were made to
CHECK TABLE ... FOR
UPGRADE and REPAIR
TABLE with respect to detection and handling of tables
with incompatible .frm files (files created
with a different version of the MySQL server). These changes
also affect mysqlcheck because that program
uses CHECK TABLE and
REPAIR TABLE, and thus also
mysql_upgrade because that program invokes
mysqlcheck.
If your table was created by a different version of the
MySQL server than the one you are currently running,
CHECK TABLE ...
FOR UPGRADE indicates that the table has an
.frm file with an incompatible version.
In this case, the result set returned by
CHECK TABLE contains a line
with a Msg_type value of
error and a Msg_text
value of Table upgrade required. Please do "REPAIR
TABLE `
tbl_name`" to fix
it!
REPAIR TABLE without
USE_FRM upgrades the
.frm file to the current version.
If you use REPAIR TABLE ...USE_FRM and
your table was created by a different version of the MySQL
server than the one you are currently running,
REPAIR TABLE will not attempt
to repair the table. In this case, the result set returned
by REPAIR TABLE contains a
line with a Msg_type value of
error and a Msg_text
value of Failed repairing incompatible .FRM
file.
Previously, use of REPAIR TABLE
...USE_FRM with a table created by a different
version of the MySQL server risked the loss of all rows in
the table.
mysql_upgrade now has a
--tmpdir option to enable
the location of temporary files to be specified.
(Bug#36469)
mysql-test-run.pl now supports
--client-bindir and
--client-libdir options for specifying the
directory where client binaries and libraries are located.
(Bug#34995)
The ndbd and ndb_mgmd man pages have been reclassified from volume 1 to volume 8. (Bug#34642)
For binary .tar.gz packages,
mysqld and other binaries now are compiled
with debugging symbols included to enable easier use with a
debugger. If you do not need debugging symbols and are short on
disk space, you can use strip to remove the
symbols from the binaries.
(Bug#33252)
mysqldump produces a -- Dump
completed on comment
at the end of the dump if
DATE--comments is given. The date
causes dump files for identical data take at different times to
appear to be different. The new options
--dump-date and
--skip-dump-date
control whether the date is added to the comment.
--skip-dump-date
suppresses date printing. The default is
--dump-date (include the date
in the comment).
(Bug#31077)
mysqltest now has mkdir
and rmdir commands for creating and removing
directories.
(Bug#31004)
The mysql_odbc_escape_string() C API
function has been removed. It has multi-byte character escaping
issues, doesn't honor the
NO_BACKSLASH_ESCAPES SQL mode
and is not needed anymore by Connector/ODBC as of 3.51.17.
(Bug#29592)
The default value of the
connect_timeout system variable
was increased from 5 to 10 seconds. This might help in cases
where clients frequently encounter errors of the form
Lost connection to MySQL server at
'.
(Bug#28359)XXX', system error:
errno
The use of InnoDB hash indexes now can be
controlled by setting the new
innodb_adaptive_hash_index
system variable at server startup. By default, this variable is
enabled. See Section 13.2.10.4, “Adaptive Hash Indexes”.
The argument for the mysql-test-run.pl
--do-test and --skip-test
options is now interpreted as a Perl regular expression if there
is a pattern metacharacter in the argument value. This allows
more flexible specification of which tests to perform or skip.
Bugs fixed:
Important Change: Security Fix:
It was possible to circumvent privileges through the creation of
MyISAM tables employing the DATA
DIRECTORY and INDEX DIRECTORY
options to overwrite existing table files in the MySQL data
directory. Use of the MySQL data directory in DATA
DIRECTORY and INDEX DIRECTORY path
name is now disallowed.
Additional fixes were made in MySQL 5.0.70.
See also Bug#39277.
Security Fix: Three vulnerabilities in yaSSL versions 1.7.5 and earlier were discovered that could lead to a server crash or execution of unauthorized code. The exploit requires a server with yaSSL enabled and TCP/IP connections enabled, but does not require valid MySQL account credentials. The exploit does not apply to OpenSSL.
The proof-of-concept exploit is freely available on the Internet. Everyone with a vulnerable MySQL configuration is advised to upgrade immediately.
Security Fix:
Using RENAME TABLE against a
table with explicit DATA DIRECTORY and
INDEX DIRECTORY options can be used to
overwrite system table information by replacing the symbolic
link points. the file to which the symlink points.
MySQL will now return an error when the file to which the symlink points already exists. (Bug#32111, CVE-2007-5969)
Security Fix:
ALTER VIEW retained the original
DEFINER value, even when altered by another
user, which could allow that user to gain the access rights of
the view. Now ALTER VIEW is
allowed only to the original definer or users with the
SUPER privilege.
(Bug#29908)
Security Fix:
When using a FEDERATED table, the local
server could be forced to crash if the remote server returned a
result with fewer columns than expected.
(Bug#29801)
Security Enhancement: It was possible to force an error message of excessive length which could lead to a buffer overflow. This has been made no longer possible as a security precaution. (Bug#32707)
Incompatible Change:
It was possible to use FRAC_SECOND as a
synonym for MICROSECOND with
DATE_ADD(),
DATE_SUB(), and
INTERVAL; now, using
FRAC_SECOND with anything other than
TIMESTAMPADD() or
TIMESTAMPDIFF() produces a syntax
error.
It is now possible (and preferable) to use
MICROSECOND with
TIMESTAMPADD() and
TIMESTAMPDIFF(), and
FRAC_SECOND is now deprecated.
(Bug#33834)
Incompatible Change:
With ONLY_FULL_GROUP_BY SQL
mode enabled, queries such as SELECT a FROM t1 HAVING
COUNT(*)>2 were not being rejected as they should
have been.
This fix results in the following behavior:
There is a check against mixing group and nongroup columns
only when
ONLY_FULL_GROUP_BY is
enabled.
This check is done both for the select list and for the
HAVING clause if there is one.
This behavior differs from previous versions as follows:
Previously, the HAVING clause was not
checked when
ONLY_FULL_GROUP_BY was
enabled; now it is checked.
Previously, the select list was checked even when
ONLY_FULL_GROUP_BY was not
enabled; now it is checked only when
ONLY_FULL_GROUP_BY is
enabled.
Incompatible Change: The MySQL 5.0.50 patch for this bug was reverted because it changed the behavior of a General Availability MySQL release. (Bug#30234)
See also Bug#27525.
Incompatible Change: It was possible to create a view having a column whose name consisted of an empty string or space characters only.
One result of this bug fix is that aliases for columns in the
view SELECT statement are checked to ensure
that they are legal column names. In particular, the length must
be within the maximum column length of 64 characters, not the
maximum alias length of 256 characters. This can cause problems
for replication or loading dump files. For additional
information and workarounds, see
Section D.4, “Restrictions on Views”.
(Bug#27695)
See also Bug#31202.
Incompatible Change:
Several type-preserving functions and operators returned an
incorrect result type that does not match their argument types:
COALESCE(),
IF(),
IFNULL(),
LEAST(),
GREATEST(),
CASE. These now aggregate using the
precise SQL types of their arguments rather than the internal
type. In addition, the result type of the
STR_TO_DATE() function is now
DATETIME by default.
(Bug#27216)
Incompatible Change: It was possible for option files to be read twice at program startup, if some of the standard option file locations turned out to be the same directory. Now duplicates are removed from the list of files to be read.
Also, users could not override system-wide settings using
~/.my.cnf because
SYSCONFDIR/my.cnf~/.my.cnf can override system-wide
settings.
The fix for this problem had a side effect such that on Unix,
MySQL programs looked for options in
~/my.cnf rather than the standard location
of ~/.my.cnf. That problem was addressed as
Bug#38180.
(Bug#20748)
Important Change: MySQL Cluster:
AUTO_INCREMENT columns had the following
problems when used in NDB tables:
The AUTO_INCREMENT counter was not
updated correctly when such a column was updated.
AUTO_INCREMENT values were not
prefetched beyond statement boundaries.
AUTO_INCREMENT values were not handled
correctly with
INSERT
IGNORE statements.
After being set,
ndb_autoincrement_prefetch_sz
showed a value of 1, regardless of the value it had
actually been set to.
As part of this fix, the behavior of
ndb_autoincrement_prefetch_sz
has changed. Setting this to less than 32 no longer has any
effect on prefetching within statements (where IDs are now
always obtained in batches of 32 or more), but only between
statements. The default value for this variable has also
changed, and is now 1.
(Bug#25176, Bug#31956, Bug#32055)
Important Change: Replication:
When the master crashed during an update on a transactional
table while in autocommit mode,
the slave failed. This fix causes every transaction (including
autocommit transactions) to be
recorded in the binlog as starting with a
BEGIN and
ending with a COMMIT or
ROLLBACK.
(Bug#26395)
Important Change:
The server no longer issues warnings for truncation of excess
spaces for values inserted into
CHAR columns. This reverts a
change in the previous release that caused warnings to be
issued.
(Bug#30059)
Replication: Important Note: Network timeouts between the master and the slave could result in corruption of the relay log. This fix rectifies a long-standing replication issue when using unreliable networks, including replication over wide area networks such as the Internet. If you experience reliability issues and see many You have an error in your SQL syntax errors on replication slaves, we strongly recommend that you upgrade to a MySQL version which includes this fix. (Bug#26489)
MySQL Cluster:
When configured with NDB support,
MySQL failed to compile using gcc 4.3 on
64bit FreeBSD systems.
(Bug#34169)
MySQL Cluster: The failure of a DDL statement could sometimes lead to node failures when attempting to execute subsequent DDL statements. (Bug#34160)
MySQL Cluster:
Extremely long SELECT statements
(where the text of the statement was in excess of 50000
characters) against NDB tables
returned empty results.
(Bug#34107)
MySQL Cluster:
A periodic failure to flush the send buffer by the
NDB TCP transporter could cause a
unnecessary delay of 10 ms between operations.
(Bug#34005)
MySQL Cluster:
When all data and SQL nodes in the cluster were shut down
abnormally (that is, other than by using STOP
in the cluster management client), ndb_mgm
used excessive amounts of CPU.
(Bug#33237)
MySQL Cluster:
An improperly reset internal signal was observed as a hang when
using events in the NDB API but
could result in various errors.
(Bug#33206)
MySQL Cluster: Incorrectly handled parameters could lead to a crash in the Transaction Coordinator during a node failure, causing other data nodes to fail. (Bug#33168)
MySQL Cluster: The failure of a master node could lead to subsequent failures in local checkpointing. (Bug#32160)
MySQL Cluster:
An uninitialized variable in the
NDB storage engine code led to
AUTO_INCREMENT failures when the server was
compiled with gcc 4.2.1.
(Bug#31848)
This regression was introduced by Bug#27437.
MySQL Cluster:
An error with an if statement in
sql/ha_ndbcluster.cc could potentially lead
to an infinite loop in case of failure when working with
AUTO_INCREMENT columns in
NDB tables.
(Bug#31810)
MySQL Cluster:
The NDB storage engine code was not
safe for strict-alias optimization in gcc
4.2.1.
(Bug#31761)
MySQL Cluster:
Primary keys on variable-length columns (such as
VARCHAR) did not work correctly.
(Bug#31635)
MySQL Cluster: Transaction atomicity was sometimes not preserved between reads and inserts under high loads. (Bug#31477)
MySQL Cluster:
Numerous NDBCLUSTER test failures
occurred in builds compiled using icc on IA64
platforms.
(Bug#31239)
MySQL Cluster: Transaction timeouts were not handled well in some circumstances, leading to excessive number of transactions being aborted unnecessarily. (Bug#30379)
MySQL Cluster: Having tables with a great many columns could cause Cluster backups to fail. (Bug#30172)
MySQL Cluster:
Issuing an
INSERT ...
ON DUPLICATE KEY UPDATE concurrently with or following
a TRUNCATE statement on an
NDB table failed with
NDB error 4350
Transaction already aborted.
(Bug#29851)
MySQL Cluster: In some cases, the cluster managment server logged entries multiple times following a restart of mgmd. (Bug#29565)
MySQL Cluster: An interpreted program of sufficient size and complexity could cause all cluster data nodes to shut down due to buffer overruns. (Bug#29390)
MySQL Cluster:
It was possible in config.ini to define
cluster nodes having node IDs greater than the maximum allowed
value.
(Bug#28298)
MySQL Cluster:
UPDATE IGNORE could sometimes fail on
NDB tables due to the use of
unitialized data when checking for duplicate keys to be ignored.
(Bug#25817)
MySQL Cluster:
When inserting a row into an NDB
table with a duplicate value for a nonprimary unique key, the
error issued would reference the wrong key.
This improves on an initial fix for this issue made in MySQL 5.0.30 and MySQL 5.0.33 (Bug#21072)
Replication: Some kinds of internal errors, such as Out of memory errors, could cause the server to crash when replicating statements with user variables.
certain internal errors. (Bug#37150)
Replication:
CREATE PROCEDURE and
CREATE FUNCTION statements
containing extended comments were not written to the binary log
correctly, causing parse errors on the slave.
(Bug#36570)
See also Bug#32575.
Replication:
insert_id was not written to
the binary log for inserts into BLACKHOLE
tables.
(Bug#35178)
Replication: The character sets and collations used for constant identifiers in stored procedures were not replicated correctly. (Bug#34289)
Replication:
A CREATE USER,
DROP USER, or
RENAME USER statement that fails
on the master, or that is a duplicate of any of these
statements, is no longer written to the binlog; previously,
either of these occurrences could cause the slave to fail.
See also Bug#29749.
Replication:
SHOW BINLOG EVENTS could fail
when the binlog contained one or more events whose size was
close to the value of
max_allowed_packet.
(Bug#33413)
Replication:
An extraneous
ROLLBACK
statement was written to the binary log by a connection that did
not use any transactional tables.
(Bug#33329)
Replication:
When a stored routine or trigger, running on a master that used
MySQL 5.0 or MySQL 5.1.11 or earlier, performed an insert on an
AUTO_INCREMENT column, the
insert_id value was not
replicated correctly to a slave running MySQL 5.1.12 or later
(including any MySQL 6.0 release).
(Bug#33029)
See also Bug#19630.
Replication:
CREATE VIEW statements containing
extended comments were not written to the binary log correctly,
causing parse errors on the slave. Now, all comments are
stripped from such statements before being written to the binary
log.
(Bug#32575)
See also Bug#36570.
Replication:
SQL statements containing comments using --
syntax were not replayable by mysqlbinlog,
even though such statements replicated correctly.
(Bug#32205)
Replication: It was possible for the name of the relay log file to exceed the amount of memory reserved for it, possibly leading to a crash of the server. (Bug#31836)
See also Bug#28597.
Replication: Corruption of log events caused the server to crash on 64-bit Linux systems having 4 GB of memory or more. (Bug#31793)
Replication:
Use of the @@hostname system variable in
inserts in mysql_system_tables_data.sql did
not replicate. The workaround is to select its value into a user
variable (which does replicate) and insert that.
(Bug#31167)
Replication:
STOP SLAVE did not stop
connection attempts properly. If the IO slave thread was
attempting to connect, STOP SLAVE
waited for the attempt to finish, sometimes for a long period of
time, rather than stopping the slave immediately.
(Bug#31024)
See also Bug#30932.
Replication:
Issuing a DROP VIEW statement
caused replication to fail if the view did not actually exist.
(Bug#30998)
Replication: One thread could read uninitialized memory from the stack of another thread. This issue was only known to occur in a mysqld process acting as both a master and a slave. (Bug#30752)
Replication:
Replication of LOAD
DATA INFILE could fail when
read_buffer_size was larger
than max_allowed_packet.
(Bug#30435)
Replication:
Setting server_id did not
update its value for the current session.
(Bug#28908)
Replication: Due a previous change in how the default name and location of the binary log file were determined, replication failed following some upgrades. (Bug#28597, Bug#28603)
See also Bug#31836.
This regression was introduced by Bug#20166.
Replication:
MASTER_POS_WAIT() did not return
NULL when the server was not a slave.
(Bug#26622)
Replication:
Stored procedures having BIT
parameters were not replicated correctly.
(Bug#26199)
Replication:
Issuing SHOW SLAVE STATUS as
mysqld was shutting down could cause a crash.
(Bug#26000)
Replication:
An UPDATE statement using a
stored function that modified a nontransactional table was not
logged if it failed. This caused the copy of the
nontransactional table on the master have a row that the copy on
the slave did not.
In addition, when an
INSERT ...
ON DUPLICATE KEY UPDATE statement encountered a
duplicate key constraint, but the
UPDATE did not actually change
any data, the statement was not logged. As a result of this fix,
such statements are now treated the same for logging purposes as
other UPDATE statements, and so
are written to the binary log.
(Bug#23333)
See also Bug#12713.
Replication:
The inspecific error message Wrong parameters to
function register_slave resulted when
START SLAVE failed to register on
the master due to excess length of any the slave server options
--report-host,
--report-user, or
--report-password. An error
message specific to each of these options is now returned in
such cases. The new error messages are:
Failed to register slave: too long 'report-host'
Failed to register slave: too long 'report-user'
Failed to register slave; too long 'report-password'
See also Bug#19328.
Replication:
A replication slave sometimes failed to reconnect because it was
unable to run SHOW SLAVE HOSTS.
It was not necessary to run this statement on slaves (since the
master should track connection IDs), and the execution of this
statement by slaves was removed.
(Bug#21132)
Replication:
START SLAVE UNTIL
MASTER_LOG_POS=
issued on a slave that was using
position--log-slave-updates and that was
involved in circular replication would cause the slave to run
and stop one event later than that specified by the value of
position.
(Bug#13861)
Replication:
PURGE BINARY LOGS TO and PURGE
BINARY LOGS BEFORE did not handle missing binary log
files correctly or in the same way. Now for both of these
statements, if any files listed in the
.index file are missing from the file
system, the statement fails with an error.
Cluster API:
When reading a BIT(64) value using
NdbOperation:getValue(), 12 bytes were
written to the buffer rather than the expected 8 bytes.
(Bug#33750)
The fix for Bug#20748 caused a problem such that on Unix, MySQL
programs looked for options in ~/my.cnf
rather than the standard location of
~/.my.cnf.
(Bug#38180)
The fix for Bug#33812 had the side effect of causing the mysql client not to be able to read some dump files produced with mysqldump. To address this, that fix was reverted. (Bug#38158)
Some binary distributions had a duplicate “-64bit” suffix in the file name. (Bug#37623)
On Windows 64-bit systems, temporary variables of
long types were used to store
ulong values, causing key cache
initialization to receive distorted parameters. The effect was
that setting key_buffer_size to
values of 2GB or more caused memory exhaustion to due allocation
of too much memory.
(Bug#36705)
Multiple-table UPDATE statements
that used a temporary table could fail to update all qualifying
rows or fail with a spurious duplicate-key error.
(Bug#36676)
A REGEXP match could return
incorrect rows when the previous row matched the expression and
used CONCAT() with an empty
string.
(Bug#36488)
mysqltest ignored the value of
--tmpdir in one place.
(Bug#36465)
The mysql client failed to recognize comment
lines consisting of -- followed by a newline.
(Bug#36244)
Conversion of a FLOAT ZEROFILL value to
string could cause a server crash if the value was
NULL.
(Bug#36139)
On Windows, the installer attempted to use JScript to determine whether the target data directory already existed. On Windows Vista x64, this resulted in an error because the installer was attempting to run the JScript in a 32-bit engine, which wasn't registered on Vista. The installer no longer uses JScript but instead relies on a native WiX command. (Bug#36103)
An error in calculation of the precision of zero-length items
(such as NULL) caused a server crash for
queries that employed temporary tables.
(Bug#36023)
For EXPLAIN EXTENDED, execution of an
uncorrelated IN subquery caused a crash if
the subquery required a temporary table for its execution.
(Bug#36011)
The server crashed inside NOT IN subqueries
with an impossible WHERE or
HAVING clause, such as NOT IN
(SELECT ... FROM t1, t2, ... WHERE 0).
(Bug#36005)
Grouping or ordering of long values in unindexed
BLOB or
TEXT columns with the
gbk or big5 character set
crashed the server.
(Bug#35993)
SET GLOBAL debug='' resulted in a Valgrind
warning in DbugParse(), which was reading
beyond the end of the control string.
(Bug#35986)
An empty bit-string literal (b'') caused a
server crash. Now the value is parsed as an empty bit value
(which is treated as an empty string in string context or 0 in
numeric context).
(Bug#35658)
mysqlbinlog left temporary files on the disk after shutdown, leading to the pollution of the temporary directory, which eventually caused mysqlbinlog to fail. This caused problems in testing and other situations where mysqlbinlog might be invoked many times in a relatively short period of time. (Bug#35543)
There was a memory leak when connecting to a
FEDERATED table using a connection string
that had a host value of localhost or omitted
the host and a port value of 0 or omitted the
port.
(Bug#35509)
The code for detecting a byte order mark (BOM) caused mysql to crash for empty input. (Bug#35480)
Using LOAD DATA
INFILE with a view could crash the server.
(Bug#35469)
The combination of
GROUP_CONCAT(),
DISTINCT, and LEFT JOIN
could crash the server when the right table is empty.
(Bug#35298)
When a view containing a reference to DUAL
was created, the reference was removed when the definition was
stored, causing some queries against the view to fail with
invalid SQL syntax errors.
(Bug#35193)
Debugging symbols were missing for some executables in Windows binary distributions. (Bug#35104)
A query that performed a
ref_or_null join where the
second table used a key having one or columns that could be
NULL and had a column value that was
NULL caused the server to crash.
(Bug#34945)
This regression was introduced by Bug#12144.
Some binaries produced stack corruption messages due to being built with versions of bison older than 2.1. Builds are now created using bison 2.3. (Bug#34926)
mysqldump failed to return an error code when
using the --master-data option
without binary logging being enabled on the server.
(Bug#34909)
Under some circumstances, the value of
mysql_insert_id() following a
SELECT ... INSERT statement could return an
incorrect value. This could happen when the last SELECT
... INSERT did not involve an
AUTO_INCREMENT column, but the value of
mysql_insert_id() was changed by
some previous statements.
(Bug#34889)
Table and database names were mixed up in some places of the subquery transformation procedure. This could affect debugging trace output and further extensions of that procedure. (Bug#34830)
A malformed URL used for a FEDERATED
table's CONNECTION option value in a
CREATE TABLE statement was not
handled correctly and could crash the server.
(Bug#34788)
Queries such as SELECT ROW(1, 2) IN (SELECT t1.a, 2)
FROM t1 GROUP BY t1.a (combining row constructors and
subqueries in the FROM clause) could lead to
assertion failure or unexpected error messages.
(Bug#34763)
Using NAME_CONST() with a negative number and
an aggregate function caused MySQL to crash. This could also
have a negative impact on replication.
(Bug#34749)
A memory-handling error associated with use of
GROUP_CONCAT() in subqueries
could result in a server crash.
(Bug#34747)
For an indexed integer column
col_name and a value
N that is one greater than the
maximum value allowed for the data type of
col_name, conditions of the form
WHERE failed to return rows
where the value of col_name <
Ncol_name is
N - 1
Executing a TRUNCATE statement on
a table having both a foreign key reference and a
DELETE trigger crashed the
server.
(Bug#34643)
Some subqueries using an expression that included an aggregate function could fail or in some cases lead to a crash of the server. (Bug#34620)
A server crash could occur if
INFORMATION_SCHEMA tables built in memory
were swapped out to disk during query execution.
(Bug#34529)
CAST(AVG( produced incorrect results for
non-arg) AS
DECIMAL)DECIMAL arguments.
(Bug#34512)
mysql_explain_log concatenated multiple-line
statements, causing malformed results for statements that
contained SQL comments beginning with --.
(Bug#34339)
Executing an ALTER VIEW statement
on a table crashed the server.
(Bug#34337)
Several additional configuration scripts in the
BUILD directory now are included in source
distributions. These may be useful for users who wish to build
MySQL from source. (See
Section 2.16.3, “Installing from the Development Source Tree”, for information about
what they do.)
(Bug#34291)
Under some conditions, a SET GLOBAL
innodb_commit_concurrency or SET GLOBAL
innodb_autoextend_increment statement could fail.
(Bug#34223)
mysqldump attempts to set the
character_set_results system
variable after connecting to the server. This failed for pre-4.1
servers that have no such variable, but
mysqldump did not account for this and 1)
failed to dump database contents; 2) failed to produce any error
message alerting the user to the problem.
(Bug#34192)
mysql_install_db failed if the server was
running with an SQL mode of
TRADITIONAL. This program now
resets the SQL mode internally to avoid this problem.
(Bug#34159)
For a FEDERATED table with an index on a
nullable column, accessing the table could crash a server,
return an incorrect result set, or return ERROR 1030
(HY000): Got error 1430 from storage engine.
(Bug#33946)
Passing anything other than a integer to a
LIMIT clause in a prepared statement would
fail. (This limitation was introduced to avoid replication
problems; for example, replicating the statement with a string
argument would cause a parse failure in the slave). Now,
arguments to the LIMIT clause are converted
to integer values, and these converted values are used when
logging the statement.
(Bug#33851)
An internal buffer in mysql was too short. Overextending it could cause stack problems or segmentation violations on some architectures. (This is not a problem that could be exploited to run arbitrary code.) (Bug#33841)
A query using WHERE
(column1=', where
string1' AND
column2=constant1) OR
(column1='string2' AND
column2=constant2)col1 used a binary collation and
string1 matched
string2 except for case, failed to
match any records even when matches were found by a query using
the equivalent clause WHERE
column2=.
(Bug#33833)constant1 OR
column2=constant2
The mysql client incorrectly parsed statements containing the word “delimiter” in mid-statement.
The fix for this bug had the side effect of causing the problem reported in Bug#38158, so it was reverted in MySQL 5.0.67. (Bug#33812)
Large unsigned integers were improperly handled for prepared statements, resulting in truncation or conversion to negative numbers. (Bug#33798)
Reuse of prepared statements could cause a memory leak in the embedded server. (Bug#33796)
The server crashed when executing a query that had a subquery
containing an equality X=Y where Y referred to a named select
list expression from the parent select. The server crashed when
trying to use the X=Y equality for
ref-based access.
(Bug#33794)
Some queries using a combination of IN,
CONCAT(), and an implicit type
conversion could return an incorrect result.
(Bug#33764)
In some cases a query that produced a result set when using
ORDER BY ASC did not return any results when
this was changed to ORDER BY DESC.
(Bug#33758)
Disabling concurrent inserts caused some cacheable queries not to be saved in the query cache. (Bug#33756)
Use of uninitialized memory for filesort in a
subquery caused a server crash.
(Bug#33675)
The server could crash when
REPEAT
or another control instruction was used in conjunction with
labels and a
LEAVE
instruction.
(Bug#33618)
The parser allowed control structures in compound statements to have mismatched beginning and ending labels. (Bug#33618)
make_binary_distribution passed the
--print-libgcc-file option to the C compiler,
but this does not work with the ICC compiler.
(Bug#33536)
Certain combinations of views, subselects with outer references and stored routines or triggers could cause the server to crash. (Bug#33389)
SET GLOBAL myisam_max_sort_file_size=DEFAULT
set myisam_max_sort_file_size
to an incorrect value.
(Bug#33382)
See also Bug#31177.
SLEEP(0) failed to return on
64-bit Mac OS X due to a bug in
pthread_cond_timedwait().
(Bug#33304)
CREATE TABLE ...
SELECT created tables that for date columns used the
obsolete Field_date type instead of
Field_newdate.
(Bug#33256)
Granting the UPDATE privilege on
one column of a view caused the server to crash.
(Bug#33201)
For DECIMAL columns used with the
ROUND(
or
X,D)TRUNCATE(
function with a nonconstant value of
X,D)D, adding an ORDER
BY for the function result produced misordered output.
(Bug#33143)
Some valid SELECT statements
could not be used as views due to incorrect column reference
resolution.
(Bug#33133)
The fix for Bug#11230 and Bug#26215 introduced a significant input-parsing slowdown for the mysql client. This has been corrected. (Bug#33057)
When MySQL was built with OpenSSL, the SSL library was not properly initialized with information of which endpoint it was (server or client), causing connection failures. (Bug#33050)
Under some circumstances a combination of aggregate functions
and GROUP BY in a
SELECT query over a view could
lead to incorrect calculation of the result type of the
aggregate function. This in turn could lead to incorrect
results, or to crashes on debug builds of the server.
(Bug#33049)
For DISTINCT queries, 4.0 and 4.1 stopped
reading joined tables as soon as the first matching row was
found. However, this optimization was lost in MySQL 5.0, which
instead read all matching rows. This fix for this regression may
result in a major improvement in performance for
DISTINCT queries in cases where many rows
match.
(Bug#32942)
The server was built even when configure was
run with the --without-server
option.
(Bug#32898)
See also Bug#23973.
Repeated creation and deletion of views within prepared statements could eventually crash the server. (Bug#32890)
See also Bug#34587.
UNION constructs cannot contain
SELECT ... INTO except in the final
SELECT. However, if a
UNION was used in a subquery and
an INTO clause appeared in the top-level
query, the parser interpreted it as having appeared in the
UNION and raised an error.
(Bug#32858)
The correct data type for a NULL column
resulting from a UNION could be
determined incorrectly in some cases: 1) Not correctly inferred
as NULL depending on the number of selects;
2) Not inferred correctly as NULL if one
select used a subquery.
(Bug#32848)
An ORDER BY query using IS
NULL in the WHERE clause did not
return correct results.
(Bug#32815)
For queries containing GROUP_CONCAT(DISTINCT
, there was a
limitation that the col_list ORDER BY
col_list)DISTINCT columns had to
be the same as ORDER BY columns. Incorrect
results could be returned if this was not true.
(Bug#32798)
Incorrect assertions could cause a server crash for
DELETE triggers for transactional
tables.
(Bug#32790)
Use of the cp932 character set with
CAST() in an ORDER
BY clause could cause a server crash.
(Bug#32726)
Inserting strings with a common prefix into a table that used
the ucs2 character set corrupted the table.
(Bug#32705)
A subquery using an IS NULL check of a column
defined as NOT NULL in a table used in the
FROM clause of the outer query produced an
invalid result.
(Bug#32694)
Specifying a nonexistent column for an
INSERT DELAYED statement caused a
server crash rather than producing an error.
(Bug#32676)
Use of CLIENT_MULTI_QUERIES caused
libmysqld to crash.
(Bug#32624)
The INTERVAL() function
incorrectly handled NULL values in the value
list.
(Bug#32560)
Use of a NULL-returning GROUP
BY expression in conjunction with WITH
ROLLUP could cause a server crash.
(Bug#32558)
See also Bug#31095.
A SELECT ... GROUP BY
query failed
with an assertion if the length of the
bit_columnBIT column used for the
GROUP BY was not an integer multiple of 8.
(Bug#32556)
Using SELECT INTO OUTFILE with 8-bit
ENCLOSED BY characters led to corrupted data
when the data was reloaded using LOAD DATA INFILE. This was
because SELECT INTO OUTFILE failed to escape
the 8-bit characters.
(Bug#32533)
For FLUSH TABLES WITH
READ LOCK, the server failed to properly detect
write-locked tables when running with low-priority updates,
resulting in a crash or deadlock.
(Bug#32528)
A build problem introduced in MySQL 5.0.52 was resolved: The x86 32-bit Intel icc-compiled server binary had unwanted dependences on Intel icc runtime libraries. (Bug#32514)
Queries using LIKE on tables having indexed
CHAR columns using either of the
eucjpms or ujis character
sets did not return correct results.
(Bug#32510)
The rules for valid column names were being applied differently for base tables and views. (Bug#32496)
Sending several KILL
QUERY statements to target a connection running
SELECT SLEEP() could freeze the server.
(Bug#32436)
ssl-cipher values in option files were not
being read by libmysqlclient.
(Bug#32429)
Repeated execution of a query containing a
CASE
expression and numerous AND and
OR relations could crash the server. The root
cause of the issue was determined to be that the internal
SEL_ARG structure was not properly
initialized when created.
(Bug#32403)
Referencing within a subquery an alias used in the
SELECT list of the outer query
was incorrectly permitted.
(Bug#32400)
An ORDER BY query on a view created using a
FEDERATED table as a base table caused the
server to crash.
(Bug#32374)
Comparison of a BIGINT NOT NULL column with a
constant arithmetic expression that evaluated to NULL mistakenly
caused the error Column '...' cannot be
null (error 1048).
(Bug#32335)
Assigning a 65,536-byte string to a
TEXT column (which can hold a
maximum of 65,535 bytes) resulted in truncation without a
warning. Now a truncation warning is generated.
(Bug#32282)
The LAST_DAY() function returns a
DATE value, but internally the
value did not have the time fields zeroed and calculations
involving the value could return incorrect results.
(Bug#32270)
MIN() and
MAX() could return incorrect
results when an index was present if a loose index scan was
used.
(Bug#32268)
Executing a prepared statement associated with a materialized cursor sent to the client a metadata packet with incorrect table and database names. The problem occurred because the server sent the name of the temporary table used by the cursor instead of the table name of the original table.
The same problem occured when selecting from a view, in which case the name of the table name was sent, rather than the name of the view. (Bug#32265)
Memory corruption could occur due to large index map in
Range checked for each record status reported
by EXPLAIN
SELECT. The problem was based in an incorrectly
calculated length of the buffer used to store a hexadecimal
representation of an index map, which could result in buffer
overrun and stack corruption under some circumstances.
(Bug#32241)
Various test program cleanups were made: 1)
mytest and libmysqltest
were removed. 2) bug25714 displays an error
message when invoked with incorrect arguments or the
--help option. 3)
mysql_client_test exits cleanly with a proper
error status.
(Bug#32221)
The default grant tables on Windows contained information for
host production.mysql.com, which should not
be there.
(Bug#32219)
Under certain conditions, the presence of a GROUP
BY clause could cause an ORDER BY
clause to be ignored.
(Bug#32202)
For comparisons of the form date_col OP
datetime_const (where
OP is
=,
<,
>,
<=,
or
>=),
the comparison is done using
DATETIME values, per the fix for
Bug#27590. However that fix caused any index on
date_col not to be used and
compromised performance. Now the index is used again.
(Bug#32198)
DATETIME arguments specified in
numeric form were treated by
DATE_ADD() as
DATE values.
(Bug#32180)
InnoDB adaptive hash latches could be held
too long during filesort operations, resulting in a server
crash. Now the hash latch is released when a query on
InnoDB tables performs a filesort. This
eliminates the crash and may provide significant performance
improvements on systems on which many queries using filesorts
with temporary tables are being performed.
(Bug#32149)
InnoDB does not support
SPATIAL indexes, but could crash when asked
to handle one. Now an error is returned.
(Bug#32125)
The server crashed on optimizations involving a join of
INT and
MEDIUMINT columns and a system
variable in the WHERE clause.
(Bug#32103)
SHOW STATUS caused a server crash
if InnoDB had not been initialized.
(Bug#32083)
With lower_case_table_names
set, CREATE TABLE LIKE was treated
differently by libmysqld than by the
nonembedded server.
(Bug#32063)
Within a subquery, UNION was
handled differently than at the top level, which could result in
incorrect results or a server crash.
(Bug#32036, Bug#32051)
User-defined functions are not loaded if the server is started
with the --skip-grant-tables
option, but the server did not properly handle this case and
issued an Out of memory error message
instead.
(Bug#32020)
HOUR(),
MINUTE(), and
SECOND() could return nonzero
values for DATE arguments.
(Bug#31990)
A column with malformed multi-byte characters could cause the full-text parser to go into an infinite loop. (Bug#31950)
Changing the SQL mode to cause dates with “zero”
parts to be considered invalid (such as
'1000-00-00') could result in indexed and
nonindexed searches returning different results for a column
that contained such dates.
(Bug#31928)
Queries testing numeric constants containing leading zeroes
against ZEROFILL columns were not evaluated
correctly.
(Bug#31887)
In debug builds, testing the result of an IN
subquery against NULL caused an assertion
failure.
(Bug#31884)
mysql-test-run.pl sometimes set up test scenarios in which the same port number was passed to multiple servers, causing one of them to be unable to start. (Bug#31880)
Comparison results for BETWEEN were
different from those for operators like
< and
> for
DATETIME-like values with
trailing extra characters such as '2007-10-01 00:00:00
GMT-6'. BETWEEN treated
the values as DATETIME, whereas
the other operators performed a binary-string comparison. Now
they all uniformly use a DATETIME
comparison, but generate warnings for values with trailing
garbage.
(Bug#31800)
Name resolution for correlated subqueries and
HAVING clauses failed to distinguish which of
two was being performed when there was a reference to an outer
aliased field. This could result in error messages about a
HAVING clause for queries that had no such
clause.
(Bug#31797)
If an error occurred during file creation, the server sometimes did not remove the file, resulting in an unused file in the file system. (Bug#31781)
The server could crash during filesort for
ORDER BY based on expressions with
INET_NTOA() or
OCT() if those functions returned
NULL.
(Bug#31758)
For a fatal error during a filesort in
find_all_keys(), the error was returned
without the necessary handler uninitialization, causing an
assertion failure.
(Bug#31742)
The examined-rows count was not incremented for
const queries.
(Bug#31700)
The mysql_change_user() C API
function was subject to buffer overflow.
(Bug#31669)
For SELECT ... INTO
OUTFILE, if the ENCLOSED BY string
is empty and the FIELDS TERMINATED BY string
started with a special character (one of n,
t, r,
b, 0,
Z, or N), every occurrence
of the character within field values would be duplicated.
(Bug#31663)
SHOW COLUMNS and
DESCRIBE displayed
null as the column type for a view with no
valid definer. This caused mysqldump to
produce a nonreloadable dump file for the view.
(Bug#31662)
The mysqlbug script did not include the
correct values of CFLAGS and
CXXFLAGS that were used to configure the
distribution.
(Bug#31644)
ucs2 does not work as a client character set,
but attempts to use it as such were not rejected. Now
character_set_client cannot be
set to ucs2. This also affects statements
such as SET NAMES and SET CHARACTER
SET.
(Bug#31615)
The server returned the error message Out of memory; restart server and try again when the actual problem was that the sort buffer was too small. Now an appropriate error message is returned in such cases. (Bug#31590)
A buffer used when setting variables was not dimensioned to
accommodate the trailing '\0' byte, so a
single-byte buffer overrun was possible.
(Bug#31588)
HAVING could treat lettercase of table
aliases incorrectly if
lower_case_table_names was
enabled.
(Bug#31562)
The fix for Bug#24989 introduced a problem such that a
NULL thread handler could be used during a
rollback operation. This problem is unlikely to be seen in
practice.
(Bug#31517)
Killing a CREATE TABLE ... LIKE statement
that was waiting for a name lock caused a server crash. When the
statement was killed, the server attempted to release locks that
were not held.
(Bug#31479)
The length of the result from
IFNULL() could be calculated
incorrectly because the sign of the result was not taken into
account.
(Bug#31471)
Queries that used the ref
access method or index-based subquery execution over indexes
that have DECIMAL columns could
fail with an error Column
.
(Bug#31450)col_name cannot be null
SELECT 1 REGEX NULL caused an assertion
failure for debug servers.
(Bug#31440)
Executing RENAME while tables were open for
use with HANDLER statements could
cause a server crash.
(Bug#31409)
mysql-test-run.pl tried to create files in a
directory where it could not be expected to have write
permission. mysqltest created
.reject files in a directory other than the
one where test results go.
(Bug#31398)
For an almost-full MyISAM table, an insert
that failed could leave the table in a corrupt state.
(Bug#31305)
myisamchk --unpack could corrupt a table that when unpacked has static (fixed-length) row format. (Bug#31277)
CONVERT( would fail on invalid input, but processing
was not aborted for the val,
DATETIME)WHERE clause, leading
to a server crash.
(Bug#31253)
Allocation of an insufficiently large group-by buffer following creation of a temporary table could lead to a server crash. (Bug#31249)
Use of DECIMAL( in
n,
n) ZEROFILLGROUP_CONCAT() could cause a
server crash.
(Bug#31227)
When sorting privilege table rows, the server treated escaped
wildcard characters (\% and
\_) the same as unescaped wildcard characters
(% and _), resulting in
incorrect row ordering.
(Bug#31194)
Server variables could not be set to their current values on Linux platforms. (Bug#31177)
See also Bug#6958.
WIth small values of
myisam_sort_buffer_size,
REPAIR TABLE for
MyISAM tables could cause a server crash.
(Bug#31174)
If MAKETIME() returned
NULL when used in an ORDER
BY that was evaluated using
filesort, a server crash could result.
(Bug#31160)
Full-text searches on ucs2 columns caused a
server crash. (FULLTEXT indexes on
ucs2 columns cannot be used, but it should be
possible to perform IN BOOLEAN MODE searches
on ucs2 columns without a crash.)
(Bug#31159)
Data in BLOB or
GEOMETRY columns could be cropped when
performing a UNION query.
(Bug#31158)
An assertion designed to detect a bug in the
ROLLUP implementation would incorrectly be
triggered when used in a subquery context with noncacheable
statements.
(Bug#31156)
Selecting spatial types in a
UNION could cause a server crash.
(Bug#31155)
Use of GROUP_CONCAT(DISTINCT
caused an
assertion failure.
(Bug#31154)bit_column)
The server crashed in the parser when running out of memory. Memory handling in the parser has been improved to gracefully return an error when out-of-memory conditions occur in the parser. (Bug#31153)
MySQL declares a UNIQUE key as a
PRIMARY key if it doesn't have
NULL columns and is not a partial key, and
the PRIMARY key must alway be the first key.
However, in some cases, a nonfirst key could be reported as
PRIMARY, leading to an assert failure by
InnoDB. This is fixed by correcting the key
sort order.
(Bug#31137)
GROUP BY NULL WITH ROLLUP could cause a
server crash.
(Bug#31095)
See also Bug#32558.
REGEXP operations could cause a
server crash for character sets such as ucs2.
Now the arguments are converted to utf8 if
possible, to allow correct results to be produced if the
resulting strings contain only 8-bit characters.
(Bug#31081)
Internal conversion routines could fail for several multi-byte
character sets (big5,
cp932, euckr,
gb2312, sjis) for empty
strings or during evaluation of SOUNDS
LIKE.
(Bug#31069, Bug#31070)
Many nested subqueries in a single query could led to excessive memory consumption and possibly a crash of the server. (Bug#31048)
The MOD() function and the
% operator crashed the server for a divisor
less than 1 with a very long fractional part.
(Bug#31019)
On Windows, the pthread_mutex_trylock()
implementation was incorrect.
(Bug#30992)
A character set introducer followed by a hexadecimal or bit-value literal did not check its argument and could return an ill-formed result for invalid input. (Bug#30986)
CHAR( did not check its
argument and could return an ill-formed result for invalid
input.
(Bug#30982)str USING
charset)
The result from
CHAR() did not add a leading 0x00 byte for input
strings with an odd number of bytes.
(Bug#30981)str USING
ucs2
On Windows, SHOW PROCESSLIST
could display process entries with a State
value of *** DEAD ***.
(Bug#30960)
The GeomFromText() function could
cause a server crash if the first argument was
NULL or the empty string.
(Bug#30955)
MAKEDATE() incorrectly moved year
values in the 100–200 range into the 1970–2069
range. (This is legitimate for 00–99, but three-digit
years should be used unchanged.)
(Bug#30951)
When invoked with constant arguments,
STR_TO_DATE() could use a cached
value for the format string and return incorrect results.
(Bug#30942)
GROUP_CONCAT() returned
',' rather than an empty string when the
argument column contained only empty strings.
(Bug#30897)
ROUND(
or
X,D)TRUNCATE(
for nonconstant values of X,D)D could
crash the server if these functions were used in an
ORDER BY that was resolved using
filesort.
(Bug#30889)
For MEMORY tables, lookups for
NULL values in BTREE
indexes could return incorrect results.
(Bug#30885)
Calling NAME_CONST() with
nonconstant arguments triggered an assertion failure.
Nonconstant arguments are now disallowed.
(Bug#30832)
For a spatial column with a regular
(non-SPATIAL) index, queries failed if the
optimizer tried to use the index.
(Bug#30825)
Values for the --tc-heuristic-recover option
incorrectly were treated as values for the
--myisam-stats-method option.
(Bug#30821)
The optimizer incorrectly optimized conditions out of the
WHERE clause in some queries involving
subqueries and indexed columns.
(Bug#30788)
If an alias was used to refer to the value returned by a stored function within a subselect, the outer select recognized the alias but failed to retrieve the value assigned to it in the subselect. (Bug#30787)
Improper calculation of CASE
expression results could lead to value truncation.
(Bug#30782)
On Windows, the pthread_mutex_trylock()
implementation was incorrect. One symptom was that invalidating
the query cache could cause a server crash.
(Bug#30768)
A multiple-table UPDATE involving
transactional and nontransactional tables caused an assertion
failure.
(Bug#30763)
Under some circumstances,
CREATE TABLE ...
SELECT could crash the server or incorrectly report
that the table row size was too large.
(Bug#30736)
Using the MIN() or
MAX() function to select one part
of a multi-part key could cause a crash when the function result
was NULL.
(Bug#30715)
The optimizer could ignore ORDER BY in cases
when the result set is ordered by filesort,
resulting in rows being returned in incorrect order.
(Bug#30666)
MyISAM tables could not exceed 4294967295
(2^32 - 1) rows on Windows.
(Bug#30638)
mysql-test-run.pl could not run
mysqld with root
privileges.
(Bug#30630)
Binary logging for a stored procedure differed depending on whether or not execution occurred in a prepared statement. (Bug#30604)
For MEMORY tables,
DELETE statements that remove
rows based on an index read could fail to remove all matching
rows.
(Bug#30590)
Using GROUP BY on an expression of the form
timestamp_col DIV
number
The options available to the CHECK
TABLE statement were also allowed in
OPTIMIZE TABLE and
ANALYZE TABLE statements, but
caused corruption during their execution. These options were
never supported for these statements, and an error is now raised
if you try to apply these options to these statements.
(Bug#30495)
When expanding a * in a
USING or NATURAL join, the
check for table access for both tables in the join was done
using only the grant information of the first table.
(Bug#30468)
When casting a string value to an integer, cases where the input
string contained a decimal point and was long enough to overrun
the unsigned long long type were not handled
correctly. The position of the decimal point was not taken into
account which resulted in miscalculated numbers and incorrect
truncation to appropriate SQL data type limits.
(Bug#30453)
Versions of mysqldump from MySQL 4.1 or
higher tried to use START TRANSACTION WITH CONSISTENT
SNAPSHOT if the
--single-transaction and
--master-data options were
given, even with servers older than 4.1 that do not support
consistent snapshots.
(Bug#30444)
For CREATE ... SELECT ... FROM, where the
resulting table contained indexes, adding
SQL_BUFFER_RESULT to the
SELECT part caused index
corruption in the table.
(Bug#30384)
An orphaned PID file from a no-longer-running process could cause mysql.server to wait for that process to exit even though it does not exist. (Bug#30378)
The optimizer made incorrect assumptions about the value of the
is_member value for user-defined functions,
sometimes resulting in incorrect ordering of UDF results.
(Bug#30355)
Some valid euc-kr characters having the
second byte in the ranges [0x41..0x5A] and
[0x61..0x7A] were rejected.
(Bug#30315)
Simultaneous ALTER TABLE
statements for BLACKHOLE tables caused 100%
CPU use due to locking problems.
(Bug#30294)
Setting certain values on a table using a spatial index could cause the server to crash. (Bug#30286)
Tables with a GEOMETRY column could be marked
as corrupt if you added a non-SPATIAL index
on a GEOMETRY column.
(Bug#30284)
Some INFORMATION_SCHEMA tables are intended
for internal use, but could be accessed by using
SHOW statements.
(Bug#30079)
On some 64-bit systems, inserting the largest negative value
into a BIGINT column resulted in
incorrect data.
(Bug#30069)
Specifying the --without-geometry option for
configure caused server compilation to fail.
(Bug#29972)
Under some circumstances, a UDF initialization function could be passed incorrect argument lengths. (Bug#29804)
configure did not find nss
on some Linux platforms.
(Bug#29658)
The mysql_config command would output
CFLAGS values that were incompatible with C++
for the HP-UX platform.
(Bug#29645)
InnoDB had a race condition for an adaptive
hash rw-lock waiting for an X-lock. This fix may also provide
significant speed improvements on systems experiencing problems
with contention for the adaptive hash index.
(Bug#29560)
Views were treated as insertable even if some base table columns with no default value were omitted from the view definition. (This is contrary to the condition for insertability that a view must contain all columns in the base table that do not have a default value.) (Bug#29477)
The mysql client program now ignores Unicode byte order mark (BOM) characters at the beginning of input files. Previously, it read them and sent them to the server, resulting in a syntax error.
Presence of a BOM does not cause mysql to
change its default character set. To do that, invoke
mysql with an option such as
--default-character-set=utf8.
(Bug#29323)
For transactional tables, an error during a multiple-table
DELETE statement did not roll
back the statement.
(Bug#29136)
The log and
log_slow_queries system
variables were displayed by SHOW
VARIABLES but could not be accessed in expressions as
@@log and
@@log_slow_queries. Also, attempting to set
them with SET produced an incorrect
Unknown system variable message. Now these
variables can be accessed in expressions and attempting to set
their values produces an error message that the variable is read
only.
(Bug#29131)
Denormalized double-precision numbers cannot be handled properly by old MIPS pocessors. For IRIX, this is now handled by enabling a mode to use a software workaround. (Bug#29085)
SHOW VARIABLES did not display
the relay_log,
relay_log_index, or
relay_log_info_file system variables.
(Bug#28893)
The MySQL preferences pane did not work to start or stop MySQL on Mac OS X 10.5 (Leopard). (Bug#28854)
When doing a DELETE on a table
that involved a JOIN with
MyISAM or MERGE tables and
the JOIN referred to the same table, the
operation could fail reporting ERROR 1030 (HY000): Got
error 134 from storage engine. This was because scans
on the table contents would change because of rows that had
already been deleted.
(Bug#28837)
On Windows, mysql_upgrade created temporary
files in C:\ and did not clean them up.
(Bug#28774)
Index hints specified in view definitions were ignored when using the view to select from the base table. (Bug#28702)
Views do not have indexes, so index hints do not apply. Use of index hints when selecting from a view is now disallowed. (Bug#28701)
After changing the SQL mode to a restrictive value that would make already-inserted dates in a column be considered invalid, searches returned different results depending on whether the column was indexed. (Bug#28687)
For upgrading to a new major version using RPM packages (such as 4.1 to 5.0), if the installation procedure found an existing MySQL server running, it could fail to shut down the old server, but also erroneously removed the server's socket file. Now the procedure checks for an existing server package from a different vendor or major MySQL version. In such case, it refuses to install the server and recommends how to safely remove the old packages before installing the new ones. (Bug#28555)
The result from CHAR() was
incorrectly assumed in some contexts to return a single-byte
result.
(Bug#28550)
mysqlhotcopy silently skipped databases with names consisting of two alphanumeric characters. (Bug#28460)
The parser confused user-defined function (UDF) and stored
function creation for CREATE
FUNCTION and required that there be a default database
when creating UDFs, although there is no such requirement.
(Bug#28318, Bug#29816)
The SQL parser did not accept an empty
UNION=() clause. This meant that, when there
were no underlying tables specified for a
MERGE table, SHOW CREATE
TABLE and mysqldump both output
statements that could not be executed.
Now it is possible to execute a CREATE
TABLE or ALTER TABLE
statement with an empty UNION=() clause.
However, SHOW CREATE TABLE and
mysqldump do not output the
UNION=() clause if there are no underlying
tables specified for a MERGE table. This also
means it is now possible to remove the underlying tables for a
MERGE table using ALTER TABLE ...
UNION=().
(Bug#28248)
The result of a comparison between
VARBINARY and
BINARY columns differed depending
on whether the VARBINARY column
was indexed.
(Bug#28076)
The metadata in some MYSQL_FIELD members
could be incorrect when a temporary table was used to evaluate a
query.
(Bug#27990)
An ORDER BY at the end of a
UNION affected individual
SELECT statements rather than the
overall query result.
(Bug#27848)
comp_err created files with permissions such that they might be inaccessible during make install operations. (Bug#27789)
It was possible to exhaust memory by repeatedly running
index_merge queries and never
performing any FLUSH
TABLES statements.
(Bug#27732)
The anonymous accounts were not being created during MySQL installation. (Bug#27692)
When utf8 was set as the connection character
set, using SPACE() with a
non-Unicode column produced an error.
(Bug#27580)
See also Bug#23637.
A race condition between killing a statement and the thread executing the statement could lead to a situation such that the binary log contained an event indicating that the statement was killed, whereas the statement actually executed to completion. (Bug#27571)
Some queries using the
NAME_CONST() function failed to
return either a result or an error to the client, causing it to
hang. This was due to the fact that there was no check to insure
that both arguments to this function were constant expressions.
(Bug#27545, Bug#32559)
With the read_only system
variable enabled, CREATE DATABASE
and DROP DATABASE were allowed to
users who did not have the SUPER
privilege.
(Bug#27440)
resolveip failed to produce correct results for host names that begin with a digit. (Bug#27427)
In ORDER BY clauses, mixing aggregate
functions and nongrouping columns is not allowed if the
ONLY_FULL_GROUP_BY SQL mode is
enabled. However, in some cases, no error was thrown because of
insufficient checking.
(Bug#27219)
For the --record_log_pos
option, mysqlhotcopy now determines the slave
status information from the result of SHOW
SLAVE STATUS by using the
Relay_Master_Log_File and
Exec_Master_Log_Pos values rather than the
Master_Log_File and
Read_Master_Log_Pos values. This provides a
more accurate indication of slave execution relative to the
master.
(Bug#27101)
The MySQL Instance Configuration Wizard would not allow you to choose a service name, even though the criteria for the service name were valid. The code that checks the name has been updated to support the correct criteria of any string less than 256 character and not containing either a forward or backward slash character. (Bug#27013)
mysqld sometimes miscalculated the number of
digits required when storing a floating-point number in a
CHAR column. This caused the
value to be truncated, or (when using a debug build) caused the
server to crash.
(Bug#26788)
See also Bug#12860.
config-win.h unconditionally defined
bool as BOOL,
causing problems on systems where bool is 1
byte and BOOL is 4 bytes.
(Bug#26461)
The internal init_time() library function
was renamed to my_init_time() to avoid
conflicts with external libraries.
(Bug#26294)
On Windows, for distributions built with debugging support, mysql could crash if the user typed Control-C. (Bug#26243)
mysqlcheck -A -r did not correctly identify all tables that needed repairing. (Bug#25347)
On Windows, an error in configure.js caused
installation of source distributions to fail.
(Bug#25340)
Using mysqldump in MySQL 5.1 resulted in dump
files that could not be loaded in MySQL 5.0 because
USING
options in index definitions appeared after the index column
list, whereas 5.0 accepted only the old syntax that has
type_nameUSING before the column list. The parser in
5.0 now accepts USING following the column
list.
(Bug#25162)
The client library had no way to return an error if no
connection had been established. This caused problems such as
mysql_library_init() failing
silently if no errmsg.sys file was
available.
(Bug#25097)
On Mac OS X, the StartupItem for MySQL did not work. (Bug#25008)
For Windows 64-bit builds, enabling shared-memory support caused client connections to fail. (Bug#24992)
If the expected precision of an arithmetic expression exceeded the maximum precision supported by MySQL, the precision of the result was reduced by an unpredictable or arbitrary amount, rather than to the maximum precision. In some cases, exceeding the maximum supported precision could also lead to a crash of the server. (Bug#24907)
mysql did not use its completion table. Also, the table contained few entries. (Bug#24624)
If a user installed MySQL Server and set a password for the
root user, and then uninstalled and
reinstalled MySQL Server to the same location, the user could
not use the MySQL Instance Config wizard to configure the server
because the uninstall operation left the previous data directory
intact. The config wizard assumed that any
new install (not an upgrade) would have the default data
directory where the root user has no
password. The installer now writes a registry key named
FoundExistingDataDir. If the installer finds
an existing data directory, the key will have a value of 1,
otherwise it will have a value of 0. When
MySQLInstanceConfig.exe is run, it will
attempt to read the key. If it can read the key, and the value
is 1 and there is no existing instance of the server (indicating
a new installation), the Config Wizard will allow the user to
input the old password so the server can be configured.
(Bug#24215)
The MySQL header files contained some duplicate macro definitions that could cause compilation problems. (Bug#23839)
SHOW COLUMNS on a
TEMPOARY table caused locking issues.
(Bug#23588)
For distributions compiled with the bundled
libedit library, there were difficulties
using the mysql client to enter input for
non-ASCII or multi-byte characters.
(Bug#23097)
For Windows Vista, MySQLInstanceConfig.exe did not include a proper manifest enabling it to run with administrative privileges. (Bug#22563)
See also Bug#24732.
On Mac OS X, mysqld did not react to Ctrl-C
when run under gdb, even when run with the
--gdb option.
(Bug#21567)
mysql_config output did not include
-lmygcc on some platforms when it was needed.
(Bug#21158)
mysql-stress-test.pl and mysqld_multi.server.sh were missing from some binary distributions. (Bug#21023, Bug#25486)
mysqldumpslow returned a confusing error message when no configuration file was found. (Bug#20455)
Host names sometimes were treated as case sensitive in
account-management statements (CREATE
USER, GRANT,
REVOKE, and so forth).
(Bug#19828)
The readline library has been updated to
version 5.2. This addresses issues in the
mysql client where history and editing within
the client would fail to work as expected.
(Bug#18431)
The Aborted_clients status
variable was incremented twice if a client exited without
calling mysql_close().
(Bug#16918)
The parser used signed rather than unsigned values in some cases that caused legal lengths in column declarations to be rejected. (Bug#15776)
A SET column whose definition specified 64
elements could not be updated using integer values.
(Bug#15409)
Clients were ignoring the TCP/IP port number specified as the default port via the --with-tcp-port configuration option. (Bug#15327)
Zero-padding of exponent values was not the same across platforms. (Bug#12860)
Values of types REAL ZEROFILL,
DOUBLE ZEROFILL, FLOAT
ZEROFILL, were not zero-filled when converted to a
character representation in the C prepared statement API.
(Bug#11589)
mysql stripped comments from statements sent
to the server. Now the
--comments or
--skip-comments option can be
used to control whether to retain or strip comments. The default
is --skip-comments.
(Bug#11230, Bug#26215)
If an INSERT ...
SELECT statement is executed, and no automatically
generated value is successfully inserted, then
mysql_insert_id() returns the ID
of the last inserted row.
If no automatically generated value is successfully inserted,
then mysql_insert_id() returns
0.
(Bug#9481)
MySQLInstanceConfig.exe did not save the
innodb_data_home_dir value to
the my.ini file under certain
circumstances.
(Bug#6627)
Several buffer-size system variables were either being handled incorrectly for large values (for settings larger than 4GB, they were truncated to values less than 4GB without a warning), or were limited unnecessarily to 4GB even on 64-bit systems. The following changes were made:
For key_buffer_size, values
larger than 4GB are allowed on 64-bit platforms (except
Windows, for which large values are truncated to 4GB with a
warning).
For join_buffer_size,
sort_buffer_size, and
myisam_sort_buffer_size,
values are limited to 4GB on all platforms. Larger values
are truncated to 4GB with a warning.
In addition, settings for
read_buffer_size and
read_rnd_buffer_size are
limited to 2GB on all platforms. Larger values are truncated to
2GB with a warning.
(Bug#5731, Bug#29419, Bug#29446)
Executing DISABLE KEYS and ENABLE
KEYS on a nonempty table would cause the size of the
index file for the table to grow considerable. This was because
the DISABLE KEYS operation would only mark
the existing index, without deleting the index blocks. The
ENABLE KEYS operation would re-create the
index, adding new blocks, while the previous index blocks would
remain. Existing indexes are now dropped and recreated when the
ENABLE KEYS statement is executed.
(Bug#4692)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.66a).
If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Security Enhancement:
To enable stricter control over the location from which
user-defined functions can be loaded, the
plugin_dir system variable has
been backported from MySQL 5.1. If the value is nonempty,
user-defined function object files can be loaded only from the
directory named by this variable. If the value is empty, the
behavior that is used prior to the inclusion of
plugin_dir applies: The UDF
object files must be located in a directory that is searched by
your system's dynamic linker.
(Bug#37428)
Bugs fixed:
Important Change: Security Fix:
It was possible to circumvent privileges through the creation of
MyISAM tables employing the DATA
DIRECTORY and INDEX DIRECTORY
options to overwrite existing table files in the MySQL data
directory. Use of the MySQL data directory in DATA
DIRECTORY and INDEX DIRECTORY path
name is now disallowed.
Additional corrections were made to handle the data directory path name if it contains symlinked directories in its path, and to make the check both at table-creation time and at table-opening time later. (Bug#32167, CVE-2008-2079)
See also Bug#39277.
Security Enhancement:
The server consumed excess memory while parsing statements with
hundreds or thousands of nested boolean conditions (such as
OR (OR ... (OR ... ))). This could lead to a
server crash or incorrect statement execution, or cause other
client statements to fail due to lack of memory. The latter
result constitutes a denial of service.
(Bug#38296)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This is a bugfix release that replaces MySQL 5.0.66.
Bugs fixed:
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This release was withdrawn from production due to the side effect produced by Bug#20748. It has been replaced by MySQL 5.0.66a, which should be used instead.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.64). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
mysql-test-run.pl now supports
--client-bindir and
--client-libdir options for specifying the
directory where client binaries and libraries are located.
(Bug#34995)
Bugs fixed:
Incompatible Change:
An additional correction to the original MySQL 5.0.64 fix was
made to normalize directory names before adding them to the list
of directories. This prevents /etc/ and
/etc from being considered different, for
example.
(Bug#20748)
See also Bug#38180.
Replication: Some kinds of internal errors, such as Out of memory errors, could cause the server to crash when replicating statements with user variables.
certain internal errors. (Bug#37150)
Some binary distributions had a duplicate “-64bit” suffix in the file name. (Bug#37623)
The mysql client failed to recognize comment
lines consisting of -- followed by a newline.
(Bug#36244)
An empty bit-string literal (b'') caused a
server crash. Now the value is parsed as an empty bit value
(which is treated as an empty string in string context or 0 in
numeric context).
(Bug#35658)
mysqlbinlog left temporary files on the disk after shutdown, leading to the pollution of the temporary directory, which eventually caused mysqlbinlog to fail. This caused problems in testing and other situations where mysqlbinlog might be invoked many times in a relatively short period of time. (Bug#35543)
The code for detecting a byte order mark (BOM) caused mysql to crash for empty input. (Bug#35480)
The mysql client incorrectly parsed statements containing the word “delimiter” in mid-statement.
The fix for this bug had the side effect of causing the problem reported in Bug#38158, so it was reverted in MySQL 5.0.67. (Bug#33812)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.62). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Important Change: Incompatible Change:
The FEDERATED storage engine is now disabled
by default in the .cnf files shipped with
MySQL distributions (my-huge.cnf,
my-medium.cnf, and so forth). This affects
server behavior only if you install one of these files.
(Bug#37069)
Bugs fixed:
Replication:
CREATE PROCEDURE and
CREATE FUNCTION statements
containing extended comments were not written to the binary log
correctly, causing parse errors on the slave.
(Bug#36570)
See also Bug#32575.
On Windows 64-bit systems, temporary variables of
long types were used to store
ulong values, causing key cache
initialization to receive distorted parameters. The effect was
that setting key_buffer_size to
values of 2GB or more caused memory exhaustion to due allocation
of too much memory.
(Bug#36705)
Multiple-table UPDATE statements
that used a temporary table could fail to update all qualifying
rows or fail with a spurious duplicate-key error.
(Bug#36676)
A REGEXP match could return
incorrect rows when the previous row matched the expression and
used CONCAT() with an empty
string.
(Bug#36488)
For EXPLAIN EXTENDED, execution of an
uncorrelated IN subquery caused a crash if
the subquery required a temporary table for its execution.
(Bug#36011)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.60). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Important Change:
Some changes were made to
CHECK TABLE ... FOR
UPGRADE and REPAIR
TABLE with respect to detection and handling of tables
with incompatible .frm files (files created
with a different version of the MySQL server). These changes
also affect mysqlcheck because that program
uses CHECK TABLE and
REPAIR TABLE, and thus also
mysql_upgrade because that program invokes
mysqlcheck.
If your table was created by a different version of the
MySQL server than the one you are currently running,
CHECK TABLE ...
FOR UPGRADE indicates that the table has an
.frm file with an incompatible version.
In this case, the result set returned by
CHECK TABLE contains a line
with a Msg_type value of
error and a Msg_text
value of Table upgrade required. Please do "REPAIR
TABLE `
tbl_name`" to fix
it!
REPAIR TABLE without
USE_FRM upgrades the
.frm file to the current version.
If you use REPAIR TABLE ...USE_FRM and
your table was created by a different version of the MySQL
server than the one you are currently running,
REPAIR TABLE will not attempt
to repair the table. In this case, the result set returned
by REPAIR TABLE contains a
line with a Msg_type value of
error and a Msg_text
value of Failed repairing incompatible .FRM
file.
Previously, use of REPAIR TABLE
...USE_FRM with a table created by a different
version of the MySQL server risked the loss of all rows in
the table.
mysql_upgrade now has a
--tmpdir option to enable
the location of temporary files to be specified.
(Bug#36469)
Bugs fixed:
Important Change:
The server no longer issues warnings for truncation of excess
spaces for values inserted into
CHAR columns. This reverts a
change in the previous release that caused warnings to be
issued.
(Bug#30059)
Replication:
CREATE VIEW statements containing
extended comments were not written to the binary log correctly,
causing parse errors on the slave. Now, all comments are
stripped from such statements before being written to the binary
log.
(Bug#32575)
See also Bug#36570.
mysqltest ignored the value of
--tmpdir in one place.
(Bug#36465)
Conversion of a FLOAT ZEROFILL value to
string could cause a server crash if the value was
NULL.
(Bug#36139)
An error in calculation of the precision of zero-length items
(such as NULL) caused a server crash for
queries that employed temporary tables.
(Bug#36023)
The server crashed inside NOT IN subqueries
with an impossible WHERE or
HAVING clause, such as NOT IN
(SELECT ... FROM t1, t2, ... WHERE 0).
(Bug#36005)
Grouping or ordering of long values in unindexed
BLOB or
TEXT columns with the
gbk or big5 character set
crashed the server.
(Bug#35993)
SET GLOBAL debug='' resulted in a Valgrind
warning in DbugParse(), which was reading
beyond the end of the control string.
(Bug#35986)
The combination of
GROUP_CONCAT(),
DISTINCT, and LEFT JOIN
could crash the server when the right table is empty.
(Bug#35298)
Several additional configuration scripts in the
BUILD directory now are included in source
distributions. These may be useful for users who wish to build
MySQL from source. (See
Section 2.16.3, “Installing from the Development Source Tree”, for information about
what they do.)
(Bug#34291)
The internal init_time() library function
was renamed to my_init_time() to avoid
conflicts with external libraries.
(Bug#26294)
The parser used signed rather than unsigned values in some cases that caused legal lengths in column declarations to be rejected. (Bug#15776)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.60). For this release, there are no such changes or fixes.
If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.58). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Bugs fixed:
Important Change: Security Fix:
It was possible to circumvent privileges through the creation of
MyISAM tables employing the DATA
DIRECTORY and INDEX DIRECTORY
options to overwrite existing table files in the MySQL data
directory. Use of the MySQL data directory in DATA
DIRECTORY and INDEX DIRECTORY path
name is now disallowed.
Additional fixes were made in MySQL 5.0.70.
See also Bug#39277.
Incompatible Change:
It was possible to use FRAC_SECOND as a
synonym for MICROSECOND with
DATE_ADD(),
DATE_SUB(), and
INTERVAL; now, using
FRAC_SECOND with anything other than
TIMESTAMPADD() or
TIMESTAMPDIFF() produces a syntax
error.
It is now possible (and preferable) to use
MICROSECOND with
TIMESTAMPADD() and
TIMESTAMPDIFF(), and
FRAC_SECOND is now deprecated.
(Bug#33834)
Important Change:
The server handled truncation of values having excess trailing
spaces into CHAR,
VARCHAR, and
TEXT columns in different ways.
This behavior has now been made consistent for columns of all
three of these types, and now follows the existing behavior of
VARCHAR columns in this regard;
that is, a Note is always issued whenever
such truncation occurs.
This change does not affect columns of these three types when
using a binary encoding; BLOB
columns are also unaffected by the change, since they always use
a binary encoding.
(Bug#30059)
Replication:
insert_id was not written to
the binary log for inserts into BLACKHOLE
tables.
(Bug#35178)
Replication: The character sets and collations used for constant identifiers in stored procedures were not replicated correctly. (Bug#34289)
Replication:
An extraneous
ROLLBACK
statement was written to the binary log by a connection that did
not use any transactional tables.
(Bug#33329)
Replication:
When a stored routine or trigger, running on a master that used
MySQL 5.0 or MySQL 5.1.11 or earlier, performed an insert on an
AUTO_INCREMENT column, the
insert_id value was not
replicated correctly to a slave running MySQL 5.1.12 or later
(including any MySQL 6.0 release).
(Bug#33029)
See also Bug#19630.
Replication:
STOP SLAVE did not stop
connection attempts properly. If the IO slave thread was
attempting to connect, STOP SLAVE
waited for the attempt to finish, sometimes for a long period of
time, rather than stopping the slave immediately.
(Bug#31024)
See also Bug#30932.
Replication:
MASTER_POS_WAIT() did not return
NULL when the server was not a slave.
(Bug#26622)
Replication:
The inspecific error message Wrong parameters to
function register_slave resulted when
START SLAVE failed to register on
the master due to excess length of any the slave server options
--report-host,
--report-user, or
--report-password. An error
message specific to each of these options is now returned in
such cases. The new error messages are:
Failed to register slave: too long 'report-host'
Failed to register slave: too long 'report-user'
Failed to register slave; too long 'report-password'
See also Bug#19328.
Replication:
START SLAVE UNTIL
MASTER_LOG_POS=
issued on a slave that was using
position--log-slave-updates and that was
involved in circular replication would cause the slave to run
and stop one event later than that specified by the value of
position.
(Bug#13861)
Replication:
PURGE BINARY LOGS TO and PURGE
BINARY LOGS BEFORE did not handle missing binary log
files correctly or in the same way. Now for both of these
statements, if any files listed in the
.index file are missing from the file
system, the statement fails with an error.
On Windows, the installer attempted to use JScript to determine whether the target data directory already existed. On Windows Vista x64, this resulted in an error because the installer was attempting to run the JScript in a 32-bit engine, which wasn't registered on Vista. The installer no longer uses JScript but instead relies on a native WiX command. (Bug#36103)
There was a memory leak when connecting to a
FEDERATED table using a connection string
that had a host value of localhost or omitted
the host and a port value of 0 or omitted the
port.
(Bug#35509)
Using LOAD DATA
INFILE with a view could crash the server.
(Bug#35469)
When a view containing a reference to DUAL
was created, the reference was removed when the definition was
stored, causing some queries against the view to fail with
invalid SQL syntax errors.
(Bug#35193)
Debugging symbols were missing for some executables in Windows binary distributions. (Bug#35104)
A query that performed a
ref_or_null join where the
second table used a key having one or columns that could be
NULL and had a column value that was
NULL caused the server to crash.
(Bug#34945)
This regression was introduced by Bug#12144.
Some binaries produced stack corruption messages due to being built with versions of bison older than 2.1. Builds are now created using bison 2.3. (Bug#34926)
mysqldump failed to return an error code when
using the --master-data option
without binary logging being enabled on the server.
(Bug#34909)
Under some circumstances, the value of
mysql_insert_id() following a
SELECT ... INSERT statement could return an
incorrect value. This could happen when the last SELECT
... INSERT did not involve an
AUTO_INCREMENT column, but the value of
mysql_insert_id() was changed by
some previous statements.
(Bug#34889)
Table and database names were mixed up in some places of the subquery transformation procedure. This could affect debugging trace output and further extensions of that procedure. (Bug#34830)
A malformed URL used for a FEDERATED
table's CONNECTION option value in a
CREATE TABLE statement was not
handled correctly and could crash the server.
(Bug#34788)
Queries such as SELECT ROW(1, 2) IN (SELECT t1.a, 2)
FROM t1 GROUP BY t1.a (combining row constructors and
subqueries in the FROM clause) could lead to
assertion failure or unexpected error messages.
(Bug#34763)
Using NAME_CONST() with a negative number and
an aggregate function caused MySQL to crash. This could also
have a negative impact on replication.
(Bug#34749)
A memory-handling error associated with use of
GROUP_CONCAT() in subqueries
could result in a server crash.
(Bug#34747)
For an indexed integer column
col_name and a value
N that is one greater than the
maximum value allowed for the data type of
col_name, conditions of the form
WHERE failed to return rows
where the value of col_name <
Ncol_name is
N - 1
Executing a TRUNCATE statement on
a table having both a foreign key reference and a
DELETE trigger crashed the
server.
(Bug#34643)
Some subqueries using an expression that included an aggregate function could fail or in some cases lead to a crash of the server. (Bug#34620)
A server crash could occur if
INFORMATION_SCHEMA tables built in memory
were swapped out to disk during query execution.
(Bug#34529)
CAST(AVG( produced incorrect results for
non-arg) AS
DECIMAL)DECIMAL arguments.
(Bug#34512)
Under some conditions, a SET GLOBAL
innodb_commit_concurrency or SET GLOBAL
innodb_autoextend_increment statement could fail.
(Bug#34223)
mysqldump attempts to set the
character_set_results system
variable after connecting to the server. This failed for pre-4.1
servers that have no such variable, but
mysqldump did not account for this and 1)
failed to dump database contents; 2) failed to produce any error
message alerting the user to the problem.
(Bug#34192)
For a FEDERATED table with an index on a
nullable column, accessing the table could crash a server,
return an incorrect result set, or return ERROR 1030
(HY000): Got error 1430 from storage engine.
(Bug#33946)
A query using WHERE
(column1=', where
string1' AND
column2=constant1) OR
(column1='string2' AND
column2=constant2)col1 used a binary collation and
string1 matched
string2 except for case, failed to
match any records even when matches were found by a query using
the equivalent clause WHERE
column2=.
(Bug#33833)constant1 OR
column2=constant2
Reuse of prepared statements could cause a memory leak in the embedded server. (Bug#33796)
Some queries using a combination of IN,
CONCAT(), and an implicit type
conversion could return an incorrect result.
(Bug#33764)
In some cases a query that produced a result set when using
ORDER BY ASC did not return any results when
this was changed to ORDER BY DESC.
(Bug#33758)
Disabling concurrent inserts caused some cacheable queries not to be saved in the query cache. (Bug#33756)
Certain combinations of views, subselects with outer references and stored routines or triggers could cause the server to crash. (Bug#33389)
SLEEP(0) failed to return on
64-bit Mac OS X due to a bug in
pthread_cond_timedwait().
(Bug#33304)
Granting the UPDATE privilege on
one column of a view caused the server to crash.
(Bug#33201)
Under some circumstances a combination of aggregate functions
and GROUP BY in a
SELECT query over a view could
lead to incorrect calculation of the result type of the
aggregate function. This in turn could lead to incorrect
results, or to crashes on debug builds of the server.
(Bug#33049)
For DISTINCT queries, 4.0 and 4.1 stopped
reading joined tables as soon as the first matching row was
found. However, this optimization was lost in MySQL 5.0, which
instead read all matching rows. This fix for this regression may
result in a major improvement in performance for
DISTINCT queries in cases where many rows
match.
(Bug#32942)
Incorrect assertions could cause a server crash for
DELETE triggers for transactional
tables.
(Bug#32790)
Inserting strings with a common prefix into a table that used
the ucs2 character set corrupted the table.
(Bug#32705)
Queries using LIKE on tables having indexed
CHAR columns using either of the
eucjpms or ujis character
sets did not return correct results.
(Bug#32510)
Queries testing numeric constants containing leading zeroes
against ZEROFILL columns were not evaluated
correctly.
(Bug#31887)
If an error occurred during file creation, the server sometimes did not remove the file, resulting in an unused file in the file system. (Bug#31781)
The server returned the error message Out of memory; restart server and try again when the actual problem was that the sort buffer was too small. Now an appropriate error message is returned in such cases. (Bug#31590)
When sorting privilege table rows, the server treated escaped
wildcard characters (\% and
\_) the same as unescaped wildcard characters
(% and _), resulting in
incorrect row ordering.
(Bug#31194)
On Windows, SHOW PROCESSLIST
could display process entries with a State
value of *** DEAD ***.
(Bug#30960)
If an alias was used to refer to the value returned by a stored function within a subselect, the outer select recognized the alias but failed to retrieve the value assigned to it in the subselect. (Bug#30787)
Binary logging for a stored procedure differed depending on whether or not execution occurred in a prepared statement. (Bug#30604)
An orphaned PID file from a no-longer-running process could cause mysql.server to wait for that process to exit even though it does not exist. (Bug#30378)
The mysql_config command would output
CFLAGS values that were incompatible with C++
for the HP-UX platform.
(Bug#29645)
The SQL parser did not accept an empty
UNION=() clause. This meant that, when there
were no underlying tables specified for a
MERGE table, SHOW CREATE
TABLE and mysqldump both output
statements that could not be executed.
Now it is possible to execute a CREATE
TABLE or ALTER TABLE
statement with an empty UNION=() clause.
However, SHOW CREATE TABLE and
mysqldump do not output the
UNION=() clause if there are no underlying
tables specified for a MERGE table. This also
means it is now possible to remove the underlying tables for a
MERGE table using ALTER TABLE ...
UNION=().
(Bug#28248)
It was possible to exhaust memory by repeatedly running
index_merge queries and never
performing any FLUSH
TABLES statements.
(Bug#27732)
When utf8 was set as the connection character
set, using SPACE() with a
non-Unicode column produced an error.
(Bug#27580)
See also Bug#23637.
In ORDER BY clauses, mixing aggregate
functions and nongrouping columns is not allowed if the
ONLY_FULL_GROUP_BY SQL mode is
enabled. However, in some cases, no error was thrown because of
insufficient checking.
(Bug#27219)
For the --record_log_pos
option, mysqlhotcopy now determines the slave
status information from the result of SHOW
SLAVE STATUS by using the
Relay_Master_Log_File and
Exec_Master_Log_Pos values rather than the
Master_Log_File and
Read_Master_Log_Pos values. This provides a
more accurate indication of slave execution relative to the
master.
(Bug#27101)
The MySQL Instance Configuration Wizard would not allow you to choose a service name, even though the criteria for the service name were valid. The code that checks the name has been updated to support the correct criteria of any string less than 256 character and not containing either a forward or backward slash character. (Bug#27013)
config-win.h unconditionally defined
bool as BOOL,
causing problems on systems where bool is 1
byte and BOOL is 4 bytes.
(Bug#26461)
On Windows, for distributions built with debugging support, mysql could crash if the user typed Control-C. (Bug#26243)
On Windows, an error in configure.js caused
installation of source distributions to fail.
(Bug#25340)
Using mysqldump in MySQL 5.1 resulted in dump
files that could not be loaded in MySQL 5.0 because
USING
options in index definitions appeared after the index column
list, whereas 5.0 accepted only the old syntax that has
type_nameUSING before the column list. The parser in
5.0 now accepts USING following the column
list.
(Bug#25162)
The client library had no way to return an error if no
connection had been established. This caused problems such as
mysql_library_init() failing
silently if no errmsg.sys file was
available.
(Bug#25097)
On Mac OS X, the StartupItem for MySQL did not work. (Bug#25008)
For Windows 64-bit builds, enabling shared-memory support caused client connections to fail. (Bug#24992)
If a user installed MySQL Server and set a password for the
root user, and then uninstalled and
reinstalled MySQL Server to the same location, the user could
not use the MySQL Instance Config wizard to configure the server
because the uninstall operation left the previous data directory
intact. The config wizard assumed that any
new install (not an upgrade) would have the default data
directory where the root user has no
password. The installer now writes a registry key named
FoundExistingDataDir. If the installer finds
an existing data directory, the key will have a value of 1,
otherwise it will have a value of 0. When
MySQLInstanceConfig.exe is run, it will
attempt to read the key. If it can read the key, and the value
is 1 and there is no existing instance of the server (indicating
a new installation), the Config Wizard will allow the user to
input the old password so the server can be configured.
(Bug#24215)
The MySQL header files contained some duplicate macro definitions that could cause compilation problems. (Bug#23839)
SHOW COLUMNS on a
TEMPOARY table caused locking issues.
(Bug#23588)
For distributions compiled with the bundled
libedit library, there were difficulties
using the mysql client to enter input for
non-ASCII or multi-byte characters.
(Bug#23097)
On Mac OS X, mysqld did not react to Ctrl-C
when run under gdb, even when run with the
--gdb option.
(Bug#21567)
mysql-stress-test.pl and mysqld_multi.server.sh were missing from some binary distributions. (Bug#21023, Bug#25486)
A SET column whose definition specified 64
elements could not be updated using integer values.
(Bug#15409)
MySQLInstanceConfig.exe did not save the
innodb_data_home_dir value to
the my.ini file under certain
circumstances.
(Bug#6627)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.56). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Cluster API: Important Change:
Because NDB_LE_MemoryUsage.page_size_kb shows
memory page sizes in bytes rather than kilobytes, it has been
renamed to page_size_bytes. The name
page_size_kb is now deprecated and thus
subject to removal in a future release, although it currently
remains supported for reasons of backward compatibility. See
The Ndb_logevent_type Type, for more information about
NDB_LE_MemoryUsage.
(Bug#30271)
The ndbd and ndb_mgmd man pages have been reclassified from volume 1 to volume 8. (Bug#34642)
mysqltest now has mkdir
and rmdir commands for creating and removing
directories.
(Bug#31004)
Bugs fixed:
MySQL Cluster:
When configured with NDB support,
MySQL failed to compile using gcc 4.3 on
64bit FreeBSD systems.
(Bug#34169)
MySQL Cluster: The failure of a DDL statement could sometimes lead to node failures when attempting to execute subsequent DDL statements. (Bug#34160)
MySQL Cluster:
Extremely long SELECT statements
(where the text of the statement was in excess of 50000
characters) against NDB tables
returned empty results.
(Bug#34107)
MySQL Cluster:
A periodic failure to flush the send buffer by the
NDB TCP transporter could cause a
unnecessary delay of 10 ms between operations.
(Bug#34005)
MySQL Cluster:
When all data and SQL nodes in the cluster were shut down
abnormally (that is, other than by using STOP
in the cluster management client), ndb_mgm
used excessive amounts of CPU.
(Bug#33237)
MySQL Cluster: Transaction atomicity was sometimes not preserved between reads and inserts under high loads. (Bug#31477)
MySQL Cluster:
Numerous NDBCLUSTER test failures
occurred in builds compiled using icc on IA64
platforms.
(Bug#31239)
MySQL Cluster: Having tables with a great many columns could cause Cluster backups to fail. (Bug#30172)
MySQL Cluster:
Issuing an
INSERT ...
ON DUPLICATE KEY UPDATE concurrently with or following
a TRUNCATE statement on an
NDB table failed with
NDB error 4350
Transaction already aborted.
(Bug#29851)
MySQL Cluster:
It was possible in config.ini to define
cluster nodes having node IDs greater than the maximum allowed
value.
(Bug#28298)
Cluster API:
When reading a BIT(64) value using
NdbOperation:getValue(), 12 bytes were
written to the buffer rather than the expected 8 bytes.
(Bug#33750)
mysql_explain_log concatenated multiple-line
statements, causing malformed results for statements that
contained SQL comments beginning with --.
(Bug#34339)
Executing an ALTER VIEW statement
on a table crashed the server.
(Bug#34337)
Passing anything other than a integer to a
LIMIT clause in a prepared statement would
fail. (This limitation was introduced to avoid replication
problems; for example, replicating the statement with a string
argument would cause a parse failure in the slave). Now,
arguments to the LIMIT clause are converted
to integer values, and these converted values are used when
logging the statement.
(Bug#33851)
An internal buffer in mysql was too short. Overextending it could cause stack problems or segmentation violations on some architectures. (This is not a problem that could be exploited to run arbitrary code.) (Bug#33841)
Large unsigned integers were improperly handled for prepared statements, resulting in truncation or conversion to negative numbers. (Bug#33798)
make_binary_distribution passed the
--print-libgcc-file option to the C compiler,
but this does not work with the ICC compiler.
(Bug#33536)
When MySQL was built with OpenSSL, the SSL library was not properly initialized with information of which endpoint it was (server or client), causing connection failures. (Bug#33050)
Repeated creation and deletion of views within prepared statements could eventually crash the server. (Bug#32890)
See also Bug#34587.
Executing a prepared statement associated with a materialized cursor sent to the client a metadata packet with incorrect table and database names. The problem occurred because the server sent the name of the temporary table used by the cursor instead of the table name of the original table.
The same problem occured when selecting from a view, in which case the name of the table name was sent, rather than the name of the view. (Bug#32265)
InnoDB adaptive hash latches could be held
too long during filesort operations, resulting in a server
crash. Now the hash latch is released when a query on
InnoDB tables performs a filesort. This
eliminates the crash and may provide significant performance
improvements on systems on which many queries using filesorts
with temporary tables are being performed.
(Bug#32149)
SHOW STATUS caused a server crash
if InnoDB had not been initialized.
(Bug#32083)
The MySQL preferences pane did not work to start or stop MySQL on Mac OS X 10.5 (Leopard). (Bug#28854)
For upgrading to a new major version using RPM packages (such as 4.1 to 5.0), if the installation procedure found an existing MySQL server running, it could fail to shut down the old server, but also erroneously removed the server's socket file. Now the procedure checks for an existing server package from a different vendor or major MySQL version. In such case, it refuses to install the server and recommends how to safely remove the old packages before installing the new ones. (Bug#28555)
mysqlhotcopy silently skipped databases with names consisting of two alphanumeric characters. (Bug#28460)
mysql did not use its completion table. Also, the table contained few entries. (Bug#24624)
mysql_config output did not include
-lmygcc on some platforms when it was needed.
(Bug#21158)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied in MySQL 5.0.56sp1 since the previous MySQL Enterprise Server Quarterly Service Pack release (5.0.50sp1a). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
mysqldump produces a -- Dump
completed on comment
at the end of the dump if
DATE--comments is given. The date
causes dump files for identical data take at different times to
appear to be different. The new options
--dump-date and
--skip-dump-date
control whether the date is added to the comment.
--skip-dump-date
suppresses date printing. The default is
--dump-date (include the date
in the comment).
(Bug#31077)
The mysql_odbc_escape_string() C API
function has been removed. It has multi-byte character escaping
issues, doesn't honor the
NO_BACKSLASH_ESCAPES SQL mode
and is not needed anymore by Connector/ODBC as of 3.51.17.
(Bug#29592)
The default value of the
connect_timeout system variable
was increased from 5 to 10 seconds. This might help in cases
where clients frequently encounter errors of the form
Lost connection to MySQL server at
'.
(Bug#28359)XXX', system error:
errno
The use of InnoDB hash indexes now can be
controlled by setting the new
innodb_adaptive_hash_index
system variable at server startup. By default, this variable is
enabled. See Section 13.2.10.4, “Adaptive Hash Indexes”.
The argument for the mysql-test-run.pl
--do-test and --skip-test
options is now interpreted as a Perl regular expression if there
is a pattern metacharacter in the argument value. This allows
more flexible specification of which tests to perform or skip.
Bugs fixed:
Security Fix:
Using RENAME TABLE against a
table with explicit DATA DIRECTORY and
INDEX DIRECTORY options can be used to
overwrite system table information by replacing the symbolic
link points. the file to which the symlink points.
MySQL will now return an error when the file to which the symlink points already exists. (Bug#32111, CVE-2007-5969)
Security Fix:
ALTER VIEW retained the original
DEFINER value, even when altered by another
user, which could allow that user to gain the access rights of
the view. Now ALTER VIEW is
allowed only to the original definer or users with the
SUPER privilege.
(Bug#29908)
Security Fix:
When using a FEDERATED table, the local
server could be forced to crash if the remote server returned a
result with fewer columns than expected.
(Bug#29801)
Security Enhancement: It was possible to force an error message of excessive length which could lead to a buffer overflow. This has been made no longer possible as a security precaution. (Bug#32707)
Incompatible Change:
With ONLY_FULL_GROUP_BY SQL
mode enabled, queries such as SELECT a FROM t1 HAVING
COUNT(*)>2 were not being rejected as they should
have been.
This fix results in the following behavior:
There is a check against mixing group and nongroup columns
only when
ONLY_FULL_GROUP_BY is
enabled.
This check is done both for the select list and for the
HAVING clause if there is one.
This behavior differs from previous versions as follows:
Previously, the HAVING clause was not
checked when
ONLY_FULL_GROUP_BY was
enabled; now it is checked.
Previously, the select list was checked even when
ONLY_FULL_GROUP_BY was not
enabled; now it is checked only when
ONLY_FULL_GROUP_BY is
enabled.
Incompatible Change: The MySQL 5.0.50 patch for this bug was reverted because it changed the behavior of a General Availability MySQL release. (Bug#30234)
See also Bug#27525.
Incompatible Change: It was possible to create a view having a column whose name consisted of an empty string or space characters only.
One result of this bug fix is that aliases for columns in the
view SELECT statement are checked to ensure
that they are legal column names. In particular, the length must
be within the maximum column length of 64 characters, not the
maximum alias length of 256 characters. This can cause problems
for replication or loading dump files. For additional
information and workarounds, see
Section D.4, “Restrictions on Views”.
(Bug#27695)
See also Bug#31202.
Incompatible Change:
Several type-preserving functions and operators returned an
incorrect result type that does not match their argument types:
COALESCE(),
IF(),
IFNULL(),
LEAST(),
GREATEST(),
CASE. These now aggregate using the
precise SQL types of their arguments rather than the internal
type. In addition, the result type of the
STR_TO_DATE() function is now
DATETIME by default.
(Bug#27216)
Incompatible Change: It was possible for option files to be read twice at program startup, if some of the standard option file locations turned out to be the same directory. Now duplicates are removed from the list of files to be read.
Also, users could not override system-wide settings using
~/.my.cnf because
SYSCONFDIR/my.cnf~/.my.cnf can override system-wide
settings.
The fix for this problem had a side effect such that on Unix,
MySQL programs looked for options in
~/my.cnf rather than the standard location
of ~/.my.cnf. That problem was addressed as
Bug#38180.
(Bug#20748)
Important Change: MySQL Cluster:
AUTO_INCREMENT columns had the following
problems when used in NDB tables:
The AUTO_INCREMENT counter was not
updated correctly when such a column was updated.
AUTO_INCREMENT values were not
prefetched beyond statement boundaries.
AUTO_INCREMENT values were not handled
correctly with
INSERT
IGNORE statements.
After being set,
ndb_autoincrement_prefetch_sz
showed a value of 1, regardless of the value it had
actually been set to.
As part of this fix, the behavior of
ndb_autoincrement_prefetch_sz
has changed. Setting this to less than 32 no longer has any
effect on prefetching within statements (where IDs are now
always obtained in batches of 32 or more), but only between
statements. The default value for this variable has also
changed, and is now 1.
(Bug#25176, Bug#31956, Bug#32055)
Important Change: Replication:
When the master crashed during an update on a transactional
table while in autocommit mode,
the slave failed. This fix causes every transaction (including
autocommit transactions) to be
recorded in the binlog as starting with a
BEGIN and
ending with a COMMIT or
ROLLBACK.
(Bug#26395)
Replication: Important Note: Network timeouts between the master and the slave could result in corruption of the relay log. This fix rectifies a long-standing replication issue when using unreliable networks, including replication over wide area networks such as the Internet. If you experience reliability issues and see many You have an error in your SQL syntax errors on replication slaves, we strongly recommend that you upgrade to a MySQL version which includes this fix. (Bug#26489)
MySQL Cluster:
An improperly reset internal signal was observed as a hang when
using events in the NDB API but
could result in various errors.
(Bug#33206)
MySQL Cluster: Incorrectly handled parameters could lead to a crash in the Transaction Coordinator during a node failure, causing other data nodes to fail. (Bug#33168)
MySQL Cluster: The failure of a master node could lead to subsequent failures in local checkpointing. (Bug#32160)
MySQL Cluster:
An uninitialized variable in the
NDB storage engine code led to
AUTO_INCREMENT failures when the server was
compiled with gcc 4.2.1.
(Bug#31848)
This regression was introduced by Bug#27437.
MySQL Cluster:
An error with an if statement in
sql/ha_ndbcluster.cc could potentially lead
to an infinite loop in case of failure when working with
AUTO_INCREMENT columns in
NDB tables.
(Bug#31810)
MySQL Cluster:
The NDB storage engine code was not
safe for strict-alias optimization in gcc
4.2.1.
(Bug#31761)
MySQL Cluster:
Primary keys on variable-length columns (such as
VARCHAR) did not work correctly.
(Bug#31635)
MySQL Cluster: Transaction timeouts were not handled well in some circumstances, leading to excessive number of transactions being aborted unnecessarily. (Bug#30379)
MySQL Cluster: In some cases, the cluster managment server logged entries multiple times following a restart of mgmd. (Bug#29565)
MySQL Cluster: An interpreted program of sufficient size and complexity could cause all cluster data nodes to shut down due to buffer overruns. (Bug#29390)
MySQL Cluster:
UPDATE IGNORE could sometimes fail on
NDB tables due to the use of
unitialized data when checking for duplicate keys to be ignored.
(Bug#25817)
MySQL Cluster:
When inserting a row into an NDB
table with a duplicate value for a nonprimary unique key, the
error issued would reference the wrong key.
This improves on an initial fix for this issue made in MySQL 5.0.30 and MySQL 5.0.33 (Bug#21072)
Replication:
A CREATE USER,
DROP USER, or
RENAME USER statement that fails
on the master, or that is a duplicate of any of these
statements, is no longer written to the binlog; previously,
either of these occurrences could cause the slave to fail.
See also Bug#29749.
Replication:
SHOW BINLOG EVENTS could fail
when the binlog contained one or more events whose size was
close to the value of
max_allowed_packet.
(Bug#33413)
Replication:
SQL statements containing comments using --
syntax were not replayable by mysqlbinlog,
even though such statements replicated correctly.
(Bug#32205)
Replication: It was possible for the name of the relay log file to exceed the amount of memory reserved for it, possibly leading to a crash of the server. (Bug#31836)
See also Bug#28597.
Replication: Corruption of log events caused the server to crash on 64-bit Linux systems having 4 GB of memory or more. (Bug#31793)
Replication:
Use of the @@hostname system variable in
inserts in mysql_system_tables_data.sql did
not replicate. The workaround is to select its value into a user
variable (which does replicate) and insert that.
(Bug#31167)
Replication:
Issuing a DROP VIEW statement
caused replication to fail if the view did not actually exist.
(Bug#30998)
Replication: One thread could read uninitialized memory from the stack of another thread. This issue was only known to occur in a mysqld process acting as both a master and a slave. (Bug#30752)
Replication:
Replication of LOAD
DATA INFILE could fail when
read_buffer_size was larger
than max_allowed_packet.
(Bug#30435)
Replication:
Setting server_id did not
update its value for the current session.
(Bug#28908)
Replication: Due a previous change in how the default name and location of the binary log file were determined, replication failed following some upgrades. (Bug#28597, Bug#28603)
See also Bug#31836.
This regression was introduced by Bug#20166.
Replication:
Stored procedures having BIT
parameters were not replicated correctly.
(Bug#26199)
Replication:
Issuing SHOW SLAVE STATUS as
mysqld was shutting down could cause a crash.
(Bug#26000)
Replication:
An UPDATE statement using a
stored function that modified a nontransactional table was not
logged if it failed. This caused the copy of the
nontransactional table on the master have a row that the copy on
the slave did not.
In addition, when an
INSERT ...
ON DUPLICATE KEY UPDATE statement encountered a
duplicate key constraint, but the
UPDATE did not actually change
any data, the statement was not logged. As a result of this fix,
such statements are now treated the same for logging purposes as
other UPDATE statements, and so
are written to the binary log.
(Bug#23333)
See also Bug#12713.
Replication:
A replication slave sometimes failed to reconnect because it was
unable to run SHOW SLAVE HOSTS.
It was not necessary to run this statement on slaves (since the
master should track connection IDs), and the execution of this
statement by slaves was removed.
(Bug#21132)
The server crashed when executing a query that had a subquery
containing an equality X=Y where Y referred to a named select
list expression from the parent select. The server crashed when
trying to use the X=Y equality for
ref-based access.
(Bug#33794)
Use of uninitialized memory for filesort in a
subquery caused a server crash.
(Bug#33675)
The server could crash when
REPEAT
or another control instruction was used in conjunction with
labels and a
LEAVE
instruction.
(Bug#33618)
The parser allowed control structures in compound statements to have mismatched beginning and ending labels. (Bug#33618)
SET GLOBAL myisam_max_sort_file_size=DEFAULT
set myisam_max_sort_file_size
to an incorrect value.
(Bug#33382)
See also Bug#31177.
CREATE TABLE ...
SELECT created tables that for date columns used the
obsolete Field_date type instead of
Field_newdate.
(Bug#33256)
For DECIMAL columns used with the
ROUND(
or
X,D)TRUNCATE(
function with a nonconstant value of
X,D)D, adding an ORDER
BY for the function result produced misordered output.
(Bug#33143)
Some valid SELECT statements
could not be used as views due to incorrect column reference
resolution.
(Bug#33133)
The fix for Bug#11230 and Bug#26215 introduced a significant input-parsing slowdown for the mysql client. This has been corrected. (Bug#33057)
UNION constructs cannot contain
SELECT ... INTO except in the final
SELECT. However, if a
UNION was used in a subquery and
an INTO clause appeared in the top-level
query, the parser interpreted it as having appeared in the
UNION and raised an error.
(Bug#32858)
The correct data type for a NULL column
resulting from a UNION could be
determined incorrectly in some cases: 1) Not correctly inferred
as NULL depending on the number of selects;
2) Not inferred correctly as NULL if one
select used a subquery.
(Bug#32848)
An ORDER BY query using IS
NULL in the WHERE clause did not
return correct results.
(Bug#32815)
For queries containing GROUP_CONCAT(DISTINCT
, there was a
limitation that the col_list ORDER BY
col_list)DISTINCT columns had to
be the same as ORDER BY columns. Incorrect
results could be returned if this was not true.
(Bug#32798)
Use of the cp932 character set with
CAST() in an ORDER
BY clause could cause a server crash.
(Bug#32726)
A subquery using an IS NULL check of a column
defined as NOT NULL in a table used in the
FROM clause of the outer query produced an
invalid result.
(Bug#32694)
Specifying a nonexistent column for an
INSERT DELAYED statement caused a
server crash rather than producing an error.
(Bug#32676)
Use of CLIENT_MULTI_QUERIES caused
libmysqld to crash.
(Bug#32624)
The INTERVAL() function
incorrectly handled NULL values in the value
list.
(Bug#32560)
Use of a NULL-returning GROUP
BY expression in conjunction with WITH
ROLLUP could cause a server crash.
(Bug#32558)
See also Bug#31095.
A SELECT ... GROUP BY
query failed
with an assertion if the length of the
bit_columnBIT column used for the
GROUP BY was not an integer multiple of 8.
(Bug#32556)
Using SELECT INTO OUTFILE with 8-bit
ENCLOSED BY characters led to corrupted data
when the data was reloaded using LOAD DATA INFILE. This was
because SELECT INTO OUTFILE failed to escape
the 8-bit characters.
(Bug#32533)
For FLUSH TABLES WITH
READ LOCK, the server failed to properly detect
write-locked tables when running with low-priority updates,
resulting in a crash or deadlock.
(Bug#32528)
A build problem introduced in MySQL 5.0.52 was resolved: The x86 32-bit Intel icc-compiled server binary had unwanted dependences on Intel icc runtime libraries. (Bug#32514)
The rules for valid column names were being applied differently for base tables and views. (Bug#32496)
Sending several KILL
QUERY statements to target a connection running
SELECT SLEEP() could freeze the server.
(Bug#32436)
ssl-cipher values in option files were not
being read by libmysqlclient.
(Bug#32429)
Repeated execution of a query containing a
CASE
expression and numerous AND and
OR relations could crash the server. The root
cause of the issue was determined to be that the internal
SEL_ARG structure was not properly
initialized when created.
(Bug#32403)
Referencing within a subquery an alias used in the
SELECT list of the outer query
was incorrectly permitted.
(Bug#32400)
An ORDER BY query on a view created using a
FEDERATED table as a base table caused the
server to crash.
(Bug#32374)
Comparison of a BIGINT NOT NULL column with a
constant arithmetic expression that evaluated to NULL mistakenly
caused the error Column '...' cannot be
null (error 1048).
(Bug#32335)
Assigning a 65,536-byte string to a
TEXT column (which can hold a
maximum of 65,535 bytes) resulted in truncation without a
warning. Now a truncation warning is generated.
(Bug#32282)
The LAST_DAY() function returns a
DATE value, but internally the
value did not have the time fields zeroed and calculations
involving the value could return incorrect results.
(Bug#32270)
MIN() and
MAX() could return incorrect
results when an index was present if a loose index scan was
used.
(Bug#32268)
Memory corruption could occur due to large index map in
Range checked for each record status reported
by EXPLAIN
SELECT. The problem was based in an incorrectly
calculated length of the buffer used to store a hexadecimal
representation of an index map, which could result in buffer
overrun and stack corruption under some circumstances.
(Bug#32241)
Various test program cleanups were made: 1)
mytest and libmysqltest
were removed. 2) bug25714 displays an error
message when invoked with incorrect arguments or the
--help option. 3)
mysql_client_test exits cleanly with a proper
error status.
(Bug#32221)
The default grant tables on Windows contained information for
host production.mysql.com, which should not
be there.
(Bug#32219)
Under certain conditions, the presence of a GROUP
BY clause could cause an ORDER BY
clause to be ignored.
(Bug#32202)
For comparisons of the form date_col OP
datetime_const (where
OP is
=,
<,
>,
<=,
or
>=),
the comparison is done using
DATETIME values, per the fix for
Bug#27590. However that fix caused any index on
date_col not to be used and
compromised performance. Now the index is used again.
(Bug#32198)
DATETIME arguments specified in
numeric form were treated by
DATE_ADD() as
DATE values.
(Bug#32180)
InnoDB does not support
SPATIAL indexes, but could crash when asked
to handle one. Now an error is returned.
(Bug#32125)
The server crashed on optimizations involving a join of
INT and
MEDIUMINT columns and a system
variable in the WHERE clause.
(Bug#32103)
With lower_case_table_names
set, CREATE TABLE LIKE was treated
differently by libmysqld than by the
nonembedded server.
(Bug#32063)
Within a subquery, UNION was
handled differently than at the top level, which could result in
incorrect results or a server crash.
(Bug#32036, Bug#32051)
User-defined functions are not loaded if the server is started
with the --skip-grant-tables
option, but the server did not properly handle this case and
issued an Out of memory error message
instead.
(Bug#32020)
HOUR(),
MINUTE(), and
SECOND() could return nonzero
values for DATE arguments.
(Bug#31990)
A column with malformed multi-byte characters could cause the full-text parser to go into an infinite loop. (Bug#31950)
Changing the SQL mode to cause dates with “zero”
parts to be considered invalid (such as
'1000-00-00') could result in indexed and
nonindexed searches returning different results for a column
that contained such dates.
(Bug#31928)
In debug builds, testing the result of an IN
subquery against NULL caused an assertion
failure.
(Bug#31884)
mysql-test-run.pl sometimes set up test scenarios in which the same port number was passed to multiple servers, causing one of them to be unable to start. (Bug#31880)
Comparison results for BETWEEN were
different from those for operators like
< and
> for
DATETIME-like values with
trailing extra characters such as '2007-10-01 00:00:00
GMT-6'. BETWEEN treated
the values as DATETIME, whereas
the other operators performed a binary-string comparison. Now
they all uniformly use a DATETIME
comparison, but generate warnings for values with trailing
garbage.
(Bug#31800)
Name resolution for correlated subqueries and
HAVING clauses failed to distinguish which of
two was being performed when there was a reference to an outer
aliased field. This could result in error messages about a
HAVING clause for queries that had no such
clause.
(Bug#31797)
The server could crash during filesort for
ORDER BY based on expressions with
INET_NTOA() or
OCT() if those functions returned
NULL.
(Bug#31758)
For a fatal error during a filesort in
find_all_keys(), the error was returned
without the necessary handler uninitialization, causing an
assertion failure.
(Bug#31742)
The examined-rows count was not incremented for
const queries.
(Bug#31700)
The mysql_change_user() C API
function was subject to buffer overflow.
(Bug#31669)
For SELECT ... INTO
OUTFILE, if the ENCLOSED BY string
is empty and the FIELDS TERMINATED BY string
started with a special character (one of n,
t, r,
b, 0,
Z, or N), every occurrence
of the character within field values would be duplicated.
(Bug#31663)
SHOW COLUMNS and
DESCRIBE displayed
null as the column type for a view with no
valid definer. This caused mysqldump to
produce a nonreloadable dump file for the view.
(Bug#31662)
The mysqlbug script did not include the
correct values of CFLAGS and
CXXFLAGS that were used to configure the
distribution.
(Bug#31644)
ucs2 does not work as a client character set,
but attempts to use it as such were not rejected. Now
character_set_client cannot be
set to ucs2. This also affects statements
such as SET NAMES and SET CHARACTER
SET.
(Bug#31615)
A buffer used when setting variables was not dimensioned to
accommodate the trailing '\0' byte, so a
single-byte buffer overrun was possible.
(Bug#31588)
HAVING could treat lettercase of table
aliases incorrectly if
lower_case_table_names was
enabled.
(Bug#31562)
The fix for Bug#24989 introduced a problem such that a
NULL thread handler could be used during a
rollback operation. This problem is unlikely to be seen in
practice.
(Bug#31517)
Killing a CREATE TABLE ... LIKE statement
that was waiting for a name lock caused a server crash. When the
statement was killed, the server attempted to release locks that
were not held.
(Bug#31479)
The length of the result from
IFNULL() could be calculated
incorrectly because the sign of the result was not taken into
account.
(Bug#31471)
Queries that used the ref
access method or index-based subquery execution over indexes
that have DECIMAL columns could
fail with an error Column
.
(Bug#31450)col_name cannot be null
SELECT 1 REGEX NULL caused an assertion
failure for debug servers.
(Bug#31440)
Executing RENAME while tables were open for
use with HANDLER statements could
cause a server crash.
(Bug#31409)
mysql-test-run.pl tried to create files in a
directory where it could not be expected to have write
permission. mysqltest created
.reject files in a directory other than the
one where test results go.
(Bug#31398)
For an almost-full MyISAM table, an insert
that failed could leave the table in a corrupt state.
(Bug#31305)
myisamchk --unpack could corrupt a table that when unpacked has static (fixed-length) row format. (Bug#31277)
CONVERT( would fail on invalid input, but processing
was not aborted for the val,
DATETIME)WHERE clause, leading
to a server crash.
(Bug#31253)
Allocation of an insufficiently large group-by buffer following creation of a temporary table could lead to a server crash. (Bug#31249)
Use of DECIMAL( in
n,
n) ZEROFILLGROUP_CONCAT() could cause a
server crash.
(Bug#31227)
Server variables could not be set to their current values on Linux platforms. (Bug#31177)
See also Bug#6958.
WIth small values of
myisam_sort_buffer_size,
REPAIR TABLE for
MyISAM tables could cause a server crash.
(Bug#31174)
If MAKETIME() returned
NULL when used in an ORDER
BY that was evaluated using
filesort, a server crash could result.
(Bug#31160)
Full-text searches on ucs2 columns caused a
server crash. (FULLTEXT indexes on
ucs2 columns cannot be used, but it should be
possible to perform IN BOOLEAN MODE searches
on ucs2 columns without a crash.)
(Bug#31159)
Data in BLOB or
GEOMETRY columns could be cropped when
performing a UNION query.
(Bug#31158)
An assertion designed to detect a bug in the
ROLLUP implementation would incorrectly be
triggered when used in a subquery context with noncacheable
statements.
(Bug#31156)
Selecting spatial types in a
UNION could cause a server crash.
(Bug#31155)
Use of GROUP_CONCAT(DISTINCT
caused an
assertion failure.
(Bug#31154)bit_column)
The server crashed in the parser when running out of memory. Memory handling in the parser has been improved to gracefully return an error when out-of-memory conditions occur in the parser. (Bug#31153)
MySQL declares a UNIQUE key as a
PRIMARY key if it doesn't have
NULL columns and is not a partial key, and
the PRIMARY key must alway be the first key.
However, in some cases, a nonfirst key could be reported as
PRIMARY, leading to an assert failure by
InnoDB. This is fixed by correcting the key
sort order.
(Bug#31137)
GROUP BY NULL WITH ROLLUP could cause a
server crash.
(Bug#31095)
See also Bug#32558.
REGEXP operations could cause a
server crash for character sets such as ucs2.
Now the arguments are converted to utf8 if
possible, to allow correct results to be produced if the
resulting strings contain only 8-bit characters.
(Bug#31081)
Internal conversion routines could fail for several multi-byte
character sets (big5,
cp932, euckr,
gb2312, sjis) for empty
strings or during evaluation of SOUNDS
LIKE.
(Bug#31069, Bug#31070)
Many nested subqueries in a single query could led to excessive memory consumption and possibly a crash of the server. (Bug#31048)
The MOD() function and the
% operator crashed the server for a divisor
less than 1 with a very long fractional part.
(Bug#31019)
On Windows, the pthread_mutex_trylock()
implementation was incorrect.
(Bug#30992)
A character set introducer followed by a hexadecimal or bit-value literal did not check its argument and could return an ill-formed result for invalid input. (Bug#30986)
CHAR( did not check its
argument and could return an ill-formed result for invalid
input.
(Bug#30982)str USING
charset)
The result from
CHAR() did not add a leading 0x00 byte for input
strings with an odd number of bytes.
(Bug#30981)str USING
ucs2
The GeomFromText() function could
cause a server crash if the first argument was
NULL or the empty string.
(Bug#30955)
MAKEDATE() incorrectly moved year
values in the 100–200 range into the 1970–2069
range. (This is legitimate for 00–99, but three-digit
years should be used unchanged.)
(Bug#30951)
When invoked with constant arguments,
STR_TO_DATE() could use a cached
value for the format string and return incorrect results.
(Bug#30942)
GROUP_CONCAT() returned
',' rather than an empty string when the
argument column contained only empty strings.
(Bug#30897)
ROUND(
or
X,D)TRUNCATE(
for nonconstant values of X,D)D could
crash the server if these functions were used in an
ORDER BY that was resolved using
filesort.
(Bug#30889)
For MEMORY tables, lookups for
NULL values in BTREE
indexes could return incorrect results.
(Bug#30885)
Calling NAME_CONST() with
nonconstant arguments triggered an assertion failure.
Nonconstant arguments are now disallowed.
(Bug#30832)
For a spatial column with a regular
(non-SPATIAL) index, queries failed if the
optimizer tried to use the index.
(Bug#30825)
Values for the --tc-heuristic-recover option
incorrectly were treated as values for the
--myisam-stats-method option.
(Bug#30821)
The optimizer incorrectly optimized conditions out of the
WHERE clause in some queries involving
subqueries and indexed columns.
(Bug#30788)
Improper calculation of CASE
expression results could lead to value truncation.
(Bug#30782)
On Windows, the pthread_mutex_trylock()
implementation was incorrect. One symptom was that invalidating
the query cache could cause a server crash.
(Bug#30768)
A multiple-table UPDATE involving
transactional and nontransactional tables caused an assertion
failure.
(Bug#30763)
Under some circumstances,
CREATE TABLE ...
SELECT could crash the server or incorrectly report
that the table row size was too large.
(Bug#30736)
Using the MIN() or
MAX() function to select one part
of a multi-part key could cause a crash when the function result
was NULL.
(Bug#30715)
The optimizer could ignore ORDER BY in cases
when the result set is ordered by filesort,
resulting in rows being returned in incorrect order.
(Bug#30666)
MyISAM tables could not exceed 4294967295
(2^32 - 1) rows on Windows.
(Bug#30638)
mysql-test-run.pl could not run
mysqld with root
privileges.
(Bug#30630)
For MEMORY tables,
DELETE statements that remove
rows based on an index read could fail to remove all matching
rows.
(Bug#30590)
Using GROUP BY on an expression of the form
timestamp_col DIV
number
The options available to the CHECK
TABLE statement were also allowed in
OPTIMIZE TABLE and
ANALYZE TABLE statements, but
caused corruption during their execution. These options were
never supported for these statements, and an error is now raised
if you try to apply these options to these statements.
(Bug#30495)
When expanding a * in a
USING or NATURAL join, the
check for table access for both tables in the join was done
using only the grant information of the first table.
(Bug#30468)
When casting a string value to an integer, cases where the input
string contained a decimal point and was long enough to overrun
the unsigned long long type were not handled
correctly. The position of the decimal point was not taken into
account which resulted in miscalculated numbers and incorrect
truncation to appropriate SQL data type limits.
(Bug#30453)
Versions of mysqldump from MySQL 4.1 or
higher tried to use START TRANSACTION WITH CONSISTENT
SNAPSHOT if the
--single-transaction and
--master-data options were
given, even with servers older than 4.1 that do not support
consistent snapshots.
(Bug#30444)
For CREATE ... SELECT ... FROM, where the
resulting table contained indexes, adding
SQL_BUFFER_RESULT to the
SELECT part caused index
corruption in the table.
(Bug#30384)
The optimizer made incorrect assumptions about the value of the
is_member value for user-defined functions,
sometimes resulting in incorrect ordering of UDF results.
(Bug#30355)
Some valid euc-kr characters having the
second byte in the ranges [0x41..0x5A] and
[0x61..0x7A] were rejected.
(Bug#30315)
Simultaneous ALTER TABLE
statements for BLACKHOLE tables caused 100%
CPU use due to locking problems.
(Bug#30294)
Setting certain values on a table using a spatial index could cause the server to crash. (Bug#30286)
Tables with a GEOMETRY column could be marked
as corrupt if you added a non-SPATIAL index
on a GEOMETRY column.
(Bug#30284)
Some INFORMATION_SCHEMA tables are intended
for internal use, but could be accessed by using
SHOW statements.
(Bug#30079)
On some 64-bit systems, inserting the largest negative value
into a BIGINT column resulted in
incorrect data.
(Bug#30069)
Specifying the --without-geometry option for
configure caused server compilation to fail.
(Bug#29972)
Under some circumstances, a UDF initialization function could be passed incorrect argument lengths. (Bug#29804)
configure did not find nss
on some Linux platforms.
(Bug#29658)
InnoDB had a race condition for an adaptive
hash rw-lock waiting for an X-lock. This fix may also provide
significant speed improvements on systems experiencing problems
with contention for the adaptive hash index.
(Bug#29560)
Views were treated as insertable even if some base table columns with no default value were omitted from the view definition. (This is contrary to the condition for insertability that a view must contain all columns in the base table that do not have a default value.) (Bug#29477)
The mysql client program now ignores Unicode byte order mark (BOM) characters at the beginning of input files. Previously, it read them and sent them to the server, resulting in a syntax error.
Presence of a BOM does not cause mysql to
change its default character set. To do that, invoke
mysql with an option such as
--default-character-set=utf8.
(Bug#29323)
For transactional tables, an error during a multiple-table
DELETE statement did not roll
back the statement.
(Bug#29136)
The log and
log_slow_queries system
variables were displayed by SHOW
VARIABLES but could not be accessed in expressions as
@@log and
@@log_slow_queries. Also, attempting to set
them with SET produced an incorrect
Unknown system variable message. Now these
variables can be accessed in expressions and attempting to set
their values produces an error message that the variable is read
only.
(Bug#29131)
Denormalized double-precision numbers cannot be handled properly by old MIPS pocessors. For IRIX, this is now handled by enabling a mode to use a software workaround. (Bug#29085)
SHOW VARIABLES did not display
the relay_log,
relay_log_index, or
relay_log_info_file system variables.
(Bug#28893)
When doing a DELETE on a table
that involved a JOIN with
MyISAM or MERGE tables and
the JOIN referred to the same table, the
operation could fail reporting ERROR 1030 (HY000): Got
error 134 from storage engine. This was because scans
on the table contents would change because of rows that had
already been deleted.
(Bug#28837)
On Windows, mysql_upgrade created temporary
files in C:\ and did not clean them up.
(Bug#28774)
Index hints specified in view definitions were ignored when using the view to select from the base table. (Bug#28702)
Views do not have indexes, so index hints do not apply. Use of index hints when selecting from a view is now disallowed. (Bug#28701)
After changing the SQL mode to a restrictive value that would make already-inserted dates in a column be considered invalid, searches returned different results depending on whether the column was indexed. (Bug#28687)
The result from CHAR() was
incorrectly assumed in some contexts to return a single-byte
result.
(Bug#28550)
The parser confused user-defined function (UDF) and stored
function creation for CREATE
FUNCTION and required that there be a default database
when creating UDFs, although there is no such requirement.
(Bug#28318, Bug#29816)
The result of a comparison between
VARBINARY and
BINARY columns differed depending
on whether the VARBINARY column
was indexed.
(Bug#28076)
The metadata in some MYSQL_FIELD members
could be incorrect when a temporary table was used to evaluate a
query.
(Bug#27990)
An ORDER BY at the end of a
UNION affected individual
SELECT statements rather than the
overall query result.
(Bug#27848)
comp_err created files with permissions such that they might be inaccessible during make install operations. (Bug#27789)
The anonymous accounts were not being created during MySQL installation. (Bug#27692)
A race condition between killing a statement and the thread executing the statement could lead to a situation such that the binary log contained an event indicating that the statement was killed, whereas the statement actually executed to completion. (Bug#27571)
Some queries using the
NAME_CONST() function failed to
return either a result or an error to the client, causing it to
hang. This was due to the fact that there was no check to insure
that both arguments to this function were constant expressions.
(Bug#27545, Bug#32559)
With the read_only system
variable enabled, CREATE DATABASE
and DROP DATABASE were allowed to
users who did not have the SUPER
privilege.
(Bug#27440)
resolveip failed to produce correct results for host names that begin with a digit. (Bug#27427)
mysqld sometimes miscalculated the number of
digits required when storing a floating-point number in a
CHAR column. This caused the
value to be truncated, or (when using a debug build) caused the
server to crash.
(Bug#26788)
See also Bug#12860.
mysqlcheck -A -r did not correctly identify all tables that needed repairing. (Bug#25347)
If the expected precision of an arithmetic expression exceeded the maximum precision supported by MySQL, the precision of the result was reduced by an unpredictable or arbitrary amount, rather than to the maximum precision. In some cases, exceeding the maximum supported precision could also lead to a crash of the server. (Bug#24907)
For Windows Vista, MySQLInstanceConfig.exe did not include a proper manifest enabling it to run with administrative privileges. (Bug#22563)
See also Bug#24732.
mysqldumpslow returned a confusing error message when no configuration file was found. (Bug#20455)
Host names sometimes were treated as case sensitive in
account-management statements (CREATE
USER, GRANT,
REVOKE, and so forth).
(Bug#19828)
The readline library has been updated to
version 5.2. This addresses issues in the
mysql client where history and editing within
the client would fail to work as expected.
(Bug#18431)
The Aborted_clients status
variable was incremented twice if a client exited without
calling mysql_close().
(Bug#16918)
Clients were ignoring the TCP/IP port number specified as the default port via the --with-tcp-port configuration option. (Bug#15327)
Zero-padding of exponent values was not the same across platforms. (Bug#12860)
Values of types REAL ZEROFILL,
DOUBLE ZEROFILL, FLOAT
ZEROFILL, were not zero-filled when converted to a
character representation in the C prepared statement API.
(Bug#11589)
mysql stripped comments from statements sent
to the server. Now the
--comments or
--skip-comments option can be
used to control whether to retain or strip comments. The default
is --skip-comments.
(Bug#11230, Bug#26215)
If an INSERT ...
SELECT statement is executed, and no automatically
generated value is successfully inserted, then
mysql_insert_id() returns the ID
of the last inserted row.
If no automatically generated value is successfully inserted,
then mysql_insert_id() returns
0.
(Bug#9481)
Several buffer-size system variables were either being handled incorrectly for large values (for settings larger than 4GB, they were truncated to values less than 4GB without a warning), or were limited unnecessarily to 4GB even on 64-bit systems. The following changes were made:
For key_buffer_size, values
larger than 4GB are allowed on 64-bit platforms (except
Windows, for which large values are truncated to 4GB with a
warning).
For join_buffer_size,
sort_buffer_size, and
myisam_sort_buffer_size,
values are limited to 4GB on all platforms. Larger values
are truncated to 4GB with a warning.
In addition, settings for
read_buffer_size and
read_rnd_buffer_size are
limited to 2GB on all platforms. Larger values are truncated to
2GB with a warning.
(Bug#5731, Bug#29419, Bug#29446)
Executing DISABLE KEYS and ENABLE
KEYS on a nonempty table would cause the size of the
index file for the table to grow considerable. This was because
the DISABLE KEYS operation would only mark
the existing index, without deleting the index blocks. The
ENABLE KEYS operation would re-create the
index, adding new blocks, while the previous index blocks would
remain. Existing indexes are now dropped and recreated when the
ENABLE KEYS statement is executed.
(Bug#4692)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.54). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Bugs fixed:
Important Change: MySQL Cluster:
AUTO_INCREMENT columns had the following
problems when used in NDB tables:
The AUTO_INCREMENT counter was not
updated correctly when such a column was updated.
AUTO_INCREMENT values were not
prefetched beyond statement boundaries.
AUTO_INCREMENT values were not handled
correctly with
INSERT
IGNORE statements.
After being set,
ndb_autoincrement_prefetch_sz
showed a value of 1, regardless of the value it had
actually been set to.
As part of this fix, the behavior of
ndb_autoincrement_prefetch_sz
has changed. Setting this to less than 32 no longer has any
effect on prefetching within statements (where IDs are now
always obtained in batches of 32 or more), but only between
statements. The default value for this variable has also
changed, and is now 1.
(Bug#25176, Bug#31956, Bug#32055)
Important Change: Replication:
When the master crashed during an update on a transactional
table while in autocommit mode,
the slave failed. This fix causes every transaction (including
autocommit transactions) to be
recorded in the binlog as starting with a
BEGIN and
ending with a COMMIT or
ROLLBACK.
(Bug#26395)
Replication: Important Note: Network timeouts between the master and the slave could result in corruption of the relay log. This fix rectifies a long-standing replication issue when using unreliable networks, including replication over wide area networks such as the Internet. If you experience reliability issues and see many You have an error in your SQL syntax errors on replication slaves, we strongly recommend that you upgrade to a MySQL version which includes this fix. (Bug#26489)
MySQL Cluster:
An improperly reset internal signal was observed as a hang when
using events in the NDB API but
could result in various errors.
(Bug#33206)
MySQL Cluster: Incorrectly handled parameters could lead to a crash in the Transaction Coordinator during a node failure, causing other data nodes to fail. (Bug#33168)
MySQL Cluster: The failure of a master node could lead to subsequent failures in local checkpointing. (Bug#32160)
MySQL Cluster:
Primary keys on variable-length columns (such as
VARCHAR) did not work correctly.
(Bug#31635)
MySQL Cluster:
When inserting a row into an NDB
table with a duplicate value for a nonprimary unique key, the
error issued would reference the wrong key.
This improves on an initial fix for this issue made in MySQL 5.0.30 and MySQL 5.0.33 (Bug#21072)
Replication:
A CREATE USER,
DROP USER, or
RENAME USER statement that fails
on the master, or that is a duplicate of any of these
statements, is no longer written to the binlog; previously,
either of these occurrences could cause the slave to fail.
See also Bug#29749.
Replication:
SHOW BINLOG EVENTS could fail
when the binlog contained one or more events whose size was
close to the value of
max_allowed_packet.
(Bug#33413)
Replication:
SQL statements containing comments using --
syntax were not replayable by mysqlbinlog,
even though such statements replicated correctly.
(Bug#32205)
Replication:
Issuing a DROP VIEW statement
caused replication to fail if the view did not actually exist.
(Bug#30998)
Replication:
Replication of LOAD
DATA INFILE could fail when
read_buffer_size was larger
than max_allowed_packet.
(Bug#30435)
Replication:
Setting server_id did not
update its value for the current session.
(Bug#28908)
The server crashed when executing a query that had a subquery
containing an equality X=Y where Y referred to a named select
list expression from the parent select. The server crashed when
trying to use the X=Y equality for
ref-based access.
(Bug#33794)
Use of uninitialized memory for filesort in a
subquery caused a server crash.
(Bug#33675)
The server could crash when
REPEAT
or another control instruction was used in conjunction with
labels and a
LEAVE
instruction.
(Bug#33618)
The parser allowed control structures in compound statements to have mismatched beginning and ending labels. (Bug#33618)
SET GLOBAL myisam_max_sort_file_size=DEFAULT
set myisam_max_sort_file_size
to an incorrect value.
(Bug#33382)
See also Bug#31177.
CREATE TABLE ...
SELECT created tables that for date columns used the
obsolete Field_date type instead of
Field_newdate.
(Bug#33256)
For DECIMAL columns used with the
ROUND(
or
X,D)TRUNCATE(
function with a nonconstant value of
X,D)D, adding an ORDER
BY for the function result produced misordered output.
(Bug#33143)
Some valid SELECT statements
could not be used as views due to incorrect column reference
resolution.
(Bug#33133)
The fix for Bug#11230 and Bug#26215 introduced a significant input-parsing slowdown for the mysql client. This has been corrected. (Bug#33057)
UNION constructs cannot contain
SELECT ... INTO except in the final
SELECT. However, if a
UNION was used in a subquery and
an INTO clause appeared in the top-level
query, the parser interpreted it as having appeared in the
UNION and raised an error.
(Bug#32858)
The correct data type for a NULL column
resulting from a UNION could be
determined incorrectly in some cases: 1) Not correctly inferred
as NULL depending on the number of selects;
2) Not inferred correctly as NULL if one
select used a subquery.
(Bug#32848)
For queries containing GROUP_CONCAT(DISTINCT
, there was a
limitation that the col_list ORDER BY
col_list)DISTINCT columns had to
be the same as ORDER BY columns. Incorrect
results could be returned if this was not true.
(Bug#32798)
HOUR(),
MINUTE(), and
SECOND() could return nonzero
values for DATE arguments.
(Bug#31990)
mysql-test-run.pl sometimes set up test scenarios in which the same port number was passed to multiple servers, causing one of them to be unable to start. (Bug#31880)
Name resolution for correlated subqueries and
HAVING clauses failed to distinguish which of
two was being performed when there was a reference to an outer
aliased field. This could result in error messages about a
HAVING clause for queries that had no such
clause.
(Bug#31797)
ROUND(
or
X,D)TRUNCATE(
for nonconstant values of X,D)D could
crash the server if these functions were used in an
ORDER BY that was resolved using
filesort.
(Bug#30889)
Views were treated as insertable even if some base table columns with no default value were omitted from the view definition. (This is contrary to the condition for insertability that a view must contain all columns in the base table that do not have a default value.) (Bug#29477)
An ORDER BY at the end of a
UNION affected individual
SELECT statements rather than the
overall query result.
(Bug#27848)
With the read_only system
variable enabled, CREATE DATABASE
and DROP DATABASE were allowed to
users who did not have the SUPER
privilege.
(Bug#27440)
resolveip failed to produce correct results for host names that begin with a digit. (Bug#27427)
mysqlcheck -A -r did not correctly identify all tables that needed repairing. (Bug#25347)
For Windows Vista, MySQLInstanceConfig.exe did not include a proper manifest enabling it to run with administrative privileges. (Bug#22563)
See also Bug#24732.
mysqldumpslow returned a confusing error message when no configuration file was found. (Bug#20455)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This is a bugfix release that replaces MySQL 5.0.54.
Bugs fixed:
Security Fix: Three vulnerabilities in yaSSL versions 1.7.5 and earlier were discovered that could lead to a server crash or execution of unauthorized code. The exploit requires a server with yaSSL enabled and TCP/IP connections enabled, but does not require valid MySQL account credentials. The exploit does not apply to OpenSSL.
The proof-of-concept exploit is freely available on the Internet. Everyone with a vulnerable MySQL configuration is advised to upgrade immediately.
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.52). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
The mysql_odbc_escape_string() C API
function has been removed. It has multi-byte character escaping
issues, doesn't honor the
NO_BACKSLASH_ESCAPES SQL mode
and is not needed anymore by Connector/ODBC as of 3.51.17.
(Bug#29592)
The argument for the mysql-test-run.pl
--do-test and --skip-test
options is now interpreted as a Perl regular expression if there
is a pattern metacharacter in the argument value. This allows
more flexible specification of which tests to perform or skip.
Bugs fixed:
Security Enhancement: It was possible to force an error message of excessive length which could lead to a buffer overflow. This has been made no longer possible as a security precaution. (Bug#32707)
Incompatible Change: The MySQL 5.0.50 patch for this bug was reverted because it changed the behavior of a General Availability MySQL release. (Bug#30234)
See also Bug#27525.
Incompatible Change: It was possible for option files to be read twice at program startup, if some of the standard option file locations turned out to be the same directory. Now duplicates are removed from the list of files to be read.
Also, users could not override system-wide settings using
~/.my.cnf because
SYSCONFDIR/my.cnf~/.my.cnf can override system-wide
settings.
The fix for this problem had a side effect such that on Unix,
MySQL programs looked for options in
~/my.cnf rather than the standard location
of ~/.my.cnf. That problem was addressed as
Bug#38180.
(Bug#20748)
Replication: It was possible for the name of the relay log file to exceed the amount of memory reserved for it, possibly leading to a crash of the server. (Bug#31836)
See also Bug#28597.
Replication: Corruption of log events caused the server to crash on 64-bit Linux systems having 4 GB of memory or more. (Bug#31793)
Replication: One thread could read uninitialized memory from the stack of another thread. This issue was only known to occur in a mysqld process acting as both a master and a slave. (Bug#30752)
Replication: Due a previous change in how the default name and location of the binary log file were determined, replication failed following some upgrades. (Bug#28597, Bug#28603)
See also Bug#31836.
This regression was introduced by Bug#20166.
Replication:
Stored procedures having BIT
parameters were not replicated correctly.
(Bug#26199)
Replication:
Issuing SHOW SLAVE STATUS as
mysqld was shutting down could cause a crash.
(Bug#26000)
Replication:
An UPDATE statement using a
stored function that modified a nontransactional table was not
logged if it failed. This caused the copy of the
nontransactional table on the master have a row that the copy on
the slave did not.
In addition, when an
INSERT ...
ON DUPLICATE KEY UPDATE statement encountered a
duplicate key constraint, but the
UPDATE did not actually change
any data, the statement was not logged. As a result of this fix,
such statements are now treated the same for logging purposes as
other UPDATE statements, and so
are written to the binary log.
(Bug#23333)
See also Bug#12713.
Replication:
A replication slave sometimes failed to reconnect because it was
unable to run SHOW SLAVE HOSTS.
It was not necessary to run this statement on slaves (since the
master should track connection IDs), and the execution of this
statement by slaves was removed.
(Bug#21132)
An ORDER BY query using IS
NULL in the WHERE clause did not
return correct results.
(Bug#32815)
Use of the cp932 character set with
CAST() in an ORDER
BY clause could cause a server crash.
(Bug#32726)
A subquery using an IS NULL check of a column
defined as NOT NULL in a table used in the
FROM clause of the outer query produced an
invalid result.
(Bug#32694)
Specifying a nonexistent column for an
INSERT DELAYED statement caused a
server crash rather than producing an error.
(Bug#32676)
Use of CLIENT_MULTI_QUERIES caused
libmysqld to crash.
(Bug#32624)
The INTERVAL() function
incorrectly handled NULL values in the value
list.
(Bug#32560)
Use of a NULL-returning GROUP
BY expression in conjunction with WITH
ROLLUP could cause a server crash.
(Bug#32558)
See also Bug#31095.
A SELECT ... GROUP BY
query failed
with an assertion if the length of the
bit_columnBIT column used for the
GROUP BY was not an integer multiple of 8.
(Bug#32556)
Using SELECT INTO OUTFILE with 8-bit
ENCLOSED BY characters led to corrupted data
when the data was reloaded using LOAD DATA INFILE. This was
because SELECT INTO OUTFILE failed to escape
the 8-bit characters.
(Bug#32533)
For FLUSH TABLES WITH
READ LOCK, the server failed to properly detect
write-locked tables when running with low-priority updates,
resulting in a crash or deadlock.
(Bug#32528)
Sending several KILL
QUERY statements to target a connection running
SELECT SLEEP() could freeze the server.
(Bug#32436)
ssl-cipher values in option files were not
being read by libmysqlclient.
(Bug#32429)
Repeated execution of a query containing a
CASE
expression and numerous AND and
OR relations could crash the server. The root
cause of the issue was determined to be that the internal
SEL_ARG structure was not properly
initialized when created.
(Bug#32403)
Referencing within a subquery an alias used in the
SELECT list of the outer query
was incorrectly permitted.
(Bug#32400)
An ORDER BY query on a view created using a
FEDERATED table as a base table caused the
server to crash.
(Bug#32374)
Comparison of a BIGINT NOT NULL column with a
constant arithmetic expression that evaluated to NULL mistakenly
caused the error Column '...' cannot be
null (error 1048).
(Bug#32335)
Assigning a 65,536-byte string to a
TEXT column (which can hold a
maximum of 65,535 bytes) resulted in truncation without a
warning. Now a truncation warning is generated.
(Bug#32282)
The LAST_DAY() function returns a
DATE value, but internally the
value did not have the time fields zeroed and calculations
involving the value could return incorrect results.
(Bug#32270)
MIN() and
MAX() could return incorrect
results when an index was present if a loose index scan was
used.
(Bug#32268)
Memory corruption could occur due to large index map in
Range checked for each record status reported
by EXPLAIN
SELECT. The problem was based in an incorrectly
calculated length of the buffer used to store a hexadecimal
representation of an index map, which could result in buffer
overrun and stack corruption under some circumstances.
(Bug#32241)
Various test program cleanups were made: 1)
mytest and libmysqltest
were removed. 2) bug25714 displays an error
message when invoked with incorrect arguments or the
--help option. 3)
mysql_client_test exits cleanly with a proper
error status.
(Bug#32221)
For comparisons of the form date_col OP
datetime_const (where
OP is
=,
<,
>,
<=,
or
>=),
the comparison is done using
DATETIME values, per the fix for
Bug#27590. However that fix caused any index on
date_col not to be used and
compromised performance. Now the index is used again.
(Bug#32198)
DATETIME arguments specified in
numeric form were treated by
DATE_ADD() as
DATE values.
(Bug#32180)
InnoDB does not support
SPATIAL indexes, but could crash when asked
to handle one. Now an error is returned.
(Bug#32125)
With lower_case_table_names
set, CREATE TABLE LIKE was treated
differently by libmysqld than by the
nonembedded server.
(Bug#32063)
Within a subquery, UNION was
handled differently than at the top level, which could result in
incorrect results or a server crash.
(Bug#32036, Bug#32051)
Changing the SQL mode to cause dates with “zero”
parts to be considered invalid (such as
'1000-00-00') could result in indexed and
nonindexed searches returning different results for a column
that contained such dates.
(Bug#31928)
ucs2 does not work as a client character set,
but attempts to use it as such were not rejected. Now
character_set_client cannot be
set to ucs2. This also affects statements
such as SET NAMES and SET CHARACTER
SET.
(Bug#31615)
Killing a CREATE TABLE ... LIKE statement
that was waiting for a name lock caused a server crash. When the
statement was killed, the server attempted to release locks that
were not held.
(Bug#31479)
myisamchk --unpack could corrupt a table that when unpacked has static (fixed-length) row format. (Bug#31277)
Server variables could not be set to their current values on Linux platforms. (Bug#31177)
See also Bug#6958.
Data in BLOB or
GEOMETRY columns could be cropped when
performing a UNION query.
(Bug#31158)
The server crashed in the parser when running out of memory. Memory handling in the parser has been improved to gracefully return an error when out-of-memory conditions occur in the parser. (Bug#31153)
MySQL declares a UNIQUE key as a
PRIMARY key if it doesn't have
NULL columns and is not a partial key, and
the PRIMARY key must alway be the first key.
However, in some cases, a nonfirst key could be reported as
PRIMARY, leading to an assert failure by
InnoDB. This is fixed by correcting the key
sort order.
(Bug#31137)
REGEXP operations could cause a
server crash for character sets such as ucs2.
Now the arguments are converted to utf8 if
possible, to allow correct results to be produced if the
resulting strings contain only 8-bit characters.
(Bug#31081)
Many nested subqueries in a single query could led to excessive memory consumption and possibly a crash of the server. (Bug#31048)
The optimizer incorrectly optimized conditions out of the
WHERE clause in some queries involving
subqueries and indexed columns.
(Bug#30788)
Improper calculation of CASE
expression results could lead to value truncation.
(Bug#30782)
A multiple-table UPDATE involving
transactional and nontransactional tables caused an assertion
failure.
(Bug#30763)
mysql-test-run.pl could not run
mysqld with root
privileges.
(Bug#30630)
The options available to the CHECK
TABLE statement were also allowed in
OPTIMIZE TABLE and
ANALYZE TABLE statements, but
caused corruption during their execution. These options were
never supported for these statements, and an error is now raised
if you try to apply these options to these statements.
(Bug#30495)
When casting a string value to an integer, cases where the input
string contained a decimal point and was long enough to overrun
the unsigned long long type were not handled
correctly. The position of the decimal point was not taken into
account which resulted in miscalculated numbers and incorrect
truncation to appropriate SQL data type limits.
(Bug#30453)
For CREATE ... SELECT ... FROM, where the
resulting table contained indexes, adding
SQL_BUFFER_RESULT to the
SELECT part caused index
corruption in the table.
(Bug#30384)
The optimizer made incorrect assumptions about the value of the
is_member value for user-defined functions,
sometimes resulting in incorrect ordering of UDF results.
(Bug#30355)
Some valid euc-kr characters having the
second byte in the ranges [0x41..0x5A] and
[0x61..0x7A] were rejected.
(Bug#30315)
Simultaneous ALTER TABLE
statements for BLACKHOLE tables caused 100%
CPU use due to locking problems.
(Bug#30294)
Tables with a GEOMETRY column could be marked
as corrupt if you added a non-SPATIAL index
on a GEOMETRY column.
(Bug#30284)
On some 64-bit systems, inserting the largest negative value
into a BIGINT column resulted in
incorrect data.
(Bug#30069)
InnoDB had a race condition for an adaptive
hash rw-lock waiting for an X-lock. This fix may also provide
significant speed improvements on systems experiencing problems
with contention for the adaptive hash index.
(Bug#29560)
The mysql client program now ignores Unicode byte order mark (BOM) characters at the beginning of input files. Previously, it read them and sent them to the server, resulting in a syntax error.
Presence of a BOM does not cause mysql to
change its default character set. To do that, invoke
mysql with an option such as
--default-character-set=utf8.
(Bug#29323)
For transactional tables, an error during a multiple-table
DELETE statement did not roll
back the statement.
(Bug#29136)
Denormalized double-precision numbers cannot be handled properly by old MIPS pocessors. For IRIX, this is now handled by enabling a mode to use a software workaround. (Bug#29085)
When doing a DELETE on a table
that involved a JOIN with
MyISAM or MERGE tables and
the JOIN referred to the same table, the
operation could fail reporting ERROR 1030 (HY000): Got
error 134 from storage engine. This was because scans
on the table contents would change because of rows that had
already been deleted.
(Bug#28837)
A race condition between killing a statement and the thread executing the statement could lead to a situation such that the binary log contained an event indicating that the statement was killed, whereas the statement actually executed to completion. (Bug#27571)
Some queries using the
NAME_CONST() function failed to
return either a result or an error to the client, causing it to
hang. This was due to the fact that there was no check to insure
that both arguments to this function were constant expressions.
(Bug#27545, Bug#32559)
mysqld sometimes miscalculated the number of
digits required when storing a floating-point number in a
CHAR column. This caused the
value to be truncated, or (when using a debug build) caused the
server to crash.
(Bug#26788)
See also Bug#12860.
If the expected precision of an arithmetic expression exceeded the maximum precision supported by MySQL, the precision of the result was reduced by an unpredictable or arbitrary amount, rather than to the maximum precision. In some cases, exceeding the maximum supported precision could also lead to a crash of the server. (Bug#24907)
Zero-padding of exponent values was not the same across platforms. (Bug#12860)
If an INSERT ...
SELECT statement is executed, and no automatically
generated value is successfully inserted, then
mysql_insert_id() returns the ID
of the last inserted row.
If no automatically generated value is successfully inserted,
then mysql_insert_id() returns
0.
(Bug#9481)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.50). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
mysqldump produces a -- Dump
completed on comment
at the end of the dump if
DATE--comments is given. The date
causes dump files for identical data take at different times to
appear to be different. The new options
--dump-date and
--skip-dump-date
control whether the date is added to the comment.
--skip-dump-date
suppresses date printing. The default is
--dump-date (include the date
in the comment).
(Bug#31077)
The default value of the
connect_timeout system variable
was increased from 5 to 10 seconds. This might help in cases
where clients frequently encounter errors of the form
Lost connection to MySQL server at
'.
(Bug#28359)XXX', system error:
errno
The use of InnoDB hash indexes now can be
controlled by setting the new
innodb_adaptive_hash_index
system variable at server startup. By default, this variable is
enabled. See Section 13.2.10.4, “Adaptive Hash Indexes”.
Bugs fixed:
Security Fix:
Using RENAME TABLE against a
table with explicit DATA DIRECTORY and
INDEX DIRECTORY options can be used to
overwrite system table information by replacing the symbolic
link points. the file to which the symlink points.
MySQL will now return an error when the file to which the symlink points already exists. (Bug#32111, CVE-2007-5969)
Security Fix:
ALTER VIEW retained the original
DEFINER value, even when altered by another
user, which could allow that user to gain the access rights of
the view. Now ALTER VIEW is
allowed only to the original definer or users with the
SUPER privilege.
(Bug#29908)
Security Fix:
When using a FEDERATED table, the local
server could be forced to crash if the remote server returned a
result with fewer columns than expected.
(Bug#29801)
Incompatible Change:
With ONLY_FULL_GROUP_BY SQL
mode enabled, queries such as SELECT a FROM t1 HAVING
COUNT(*)>2 were not being rejected as they should
have been.
This fix results in the following behavior:
There is a check against mixing group and nongroup columns
only when
ONLY_FULL_GROUP_BY is
enabled.
This check is done both for the select list and for the
HAVING clause if there is one.
This behavior differs from previous versions as follows:
Previously, the HAVING clause was not
checked when
ONLY_FULL_GROUP_BY was
enabled; now it is checked.
Previously, the select list was checked even when
ONLY_FULL_GROUP_BY was not
enabled; now it is checked only when
ONLY_FULL_GROUP_BY is
enabled.
Incompatible Change: It was possible to create a view having a column whose name consisted of an empty string or space characters only.
One result of this bug fix is that aliases for columns in the
view SELECT statement are checked to ensure
that they are legal column names. In particular, the length must
be within the maximum column length of 64 characters, not the
maximum alias length of 256 characters. This can cause problems
for replication or loading dump files. For additional
information and workarounds, see
Section D.4, “Restrictions on Views”.
(Bug#27695)
See also Bug#31202.
Incompatible Change:
Several type-preserving functions and operators returned an
incorrect result type that does not match their argument types:
COALESCE(),
IF(),
IFNULL(),
LEAST(),
GREATEST(),
CASE. These now aggregate using the
precise SQL types of their arguments rather than the internal
type. In addition, the result type of the
STR_TO_DATE() function is now
DATETIME by default.
(Bug#27216)
MySQL Cluster:
An uninitialized variable in the
NDB storage engine code led to
AUTO_INCREMENT failures when the server was
compiled with gcc 4.2.1.
(Bug#31848)
This regression was introduced by Bug#27437.
MySQL Cluster:
An error with an if statement in
sql/ha_ndbcluster.cc could potentially lead
to an infinite loop in case of failure when working with
AUTO_INCREMENT columns in
NDB tables.
(Bug#31810)
MySQL Cluster:
The NDB storage engine code was not
safe for strict-alias optimization in gcc
4.2.1.
(Bug#31761)
MySQL Cluster: Transaction timeouts were not handled well in some circumstances, leading to excessive number of transactions being aborted unnecessarily. (Bug#30379)
MySQL Cluster: In some cases, the cluster managment server logged entries multiple times following a restart of mgmd. (Bug#29565)
MySQL Cluster: An interpreted program of sufficient size and complexity could cause all cluster data nodes to shut down due to buffer overruns. (Bug#29390)
MySQL Cluster:
UPDATE IGNORE could sometimes fail on
NDB tables due to the use of
unitialized data when checking for duplicate keys to be ignored.
(Bug#25817)
Replication:
Use of the @@hostname system variable in
inserts in mysql_system_tables_data.sql did
not replicate. The workaround is to select its value into a user
variable (which does replicate) and insert that.
(Bug#31167)
A build problem introduced in MySQL 5.0.52 was resolved: The x86 32-bit Intel icc-compiled server binary had unwanted dependences on Intel icc runtime libraries. (Bug#32514)
The rules for valid column names were being applied differently for base tables and views. (Bug#32496)
The default grant tables on Windows contained information for
host production.mysql.com, which should not
be there.
(Bug#32219)
Under certain conditions, the presence of a GROUP
BY clause could cause an ORDER BY
clause to be ignored.
(Bug#32202)
The server crashed on optimizations involving a join of
INT and
MEDIUMINT columns and a system
variable in the WHERE clause.
(Bug#32103)
User-defined functions are not loaded if the server is started
with the --skip-grant-tables
option, but the server did not properly handle this case and
issued an Out of memory error message
instead.
(Bug#32020)
A column with malformed multi-byte characters could cause the full-text parser to go into an infinite loop. (Bug#31950)
In debug builds, testing the result of an IN
subquery against NULL caused an assertion
failure.
(Bug#31884)
Comparison results for BETWEEN were
different from those for operators like
< and
> for
DATETIME-like values with
trailing extra characters such as '2007-10-01 00:00:00
GMT-6'. BETWEEN treated
the values as DATETIME, whereas
the other operators performed a binary-string comparison. Now
they all uniformly use a DATETIME
comparison, but generate warnings for values with trailing
garbage.
(Bug#31800)
The server could crash during filesort for
ORDER BY based on expressions with
INET_NTOA() or
OCT() if those functions returned
NULL.
(Bug#31758)
For a fatal error during a filesort in
find_all_keys(), the error was returned
without the necessary handler uninitialization, causing an
assertion failure.
(Bug#31742)
The examined-rows count was not incremented for
const queries.
(Bug#31700)
The mysql_change_user() C API
function was subject to buffer overflow.
(Bug#31669)
For SELECT ... INTO
OUTFILE, if the ENCLOSED BY string
is empty and the FIELDS TERMINATED BY string
started with a special character (one of n,
t, r,
b, 0,
Z, or N), every occurrence
of the character within field values would be duplicated.
(Bug#31663)
SHOW COLUMNS and
DESCRIBE displayed
null as the column type for a view with no
valid definer. This caused mysqldump to
produce a nonreloadable dump file for the view.
(Bug#31662)
The mysqlbug script did not include the
correct values of CFLAGS and
CXXFLAGS that were used to configure the
distribution.
(Bug#31644)
A buffer used when setting variables was not dimensioned to
accommodate the trailing '\0' byte, so a
single-byte buffer overrun was possible.
(Bug#31588)
HAVING could treat lettercase of table
aliases incorrectly if
lower_case_table_names was
enabled.
(Bug#31562)
The fix for Bug#24989 introduced a problem such that a
NULL thread handler could be used during a
rollback operation. This problem is unlikely to be seen in
practice.
(Bug#31517)
The length of the result from
IFNULL() could be calculated
incorrectly because the sign of the result was not taken into
account.
(Bug#31471)
Queries that used the ref
access method or index-based subquery execution over indexes
that have DECIMAL columns could
fail with an error Column
.
(Bug#31450)col_name cannot be null
SELECT 1 REGEX NULL caused an assertion
failure for debug servers.
(Bug#31440)
Executing RENAME while tables were open for
use with HANDLER statements could
cause a server crash.
(Bug#31409)
mysql-test-run.pl tried to create files in a
directory where it could not be expected to have write
permission. mysqltest created
.reject files in a directory other than the
one where test results go.
(Bug#31398)
For an almost-full MyISAM table, an insert
that failed could leave the table in a corrupt state.
(Bug#31305)
CONVERT( would fail on invalid input, but processing
was not aborted for the val,
DATETIME)WHERE clause, leading
to a server crash.
(Bug#31253)
Allocation of an insufficiently large group-by buffer following creation of a temporary table could lead to a server crash. (Bug#31249)
Use of DECIMAL( in
n,
n) ZEROFILLGROUP_CONCAT() could cause a
server crash.
(Bug#31227)
WIth small values of
myisam_sort_buffer_size,
REPAIR TABLE for
MyISAM tables could cause a server crash.
(Bug#31174)
If MAKETIME() returned
NULL when used in an ORDER
BY that was evaluated using
filesort, a server crash could result.
(Bug#31160)
Full-text searches on ucs2 columns caused a
server crash. (FULLTEXT indexes on
ucs2 columns cannot be used, but it should be
possible to perform IN BOOLEAN MODE searches
on ucs2 columns without a crash.)
(Bug#31159)
An assertion designed to detect a bug in the
ROLLUP implementation would incorrectly be
triggered when used in a subquery context with noncacheable
statements.
(Bug#31156)
Selecting spatial types in a
UNION could cause a server crash.
(Bug#31155)
Use of GROUP_CONCAT(DISTINCT
caused an
assertion failure.
(Bug#31154)bit_column)
GROUP BY NULL WITH ROLLUP could cause a
server crash.
(Bug#31095)
See also Bug#32558.
Internal conversion routines could fail for several multi-byte
character sets (big5,
cp932, euckr,
gb2312, sjis) for empty
strings or during evaluation of SOUNDS
LIKE.
(Bug#31069, Bug#31070)
The MOD() function and the
% operator crashed the server for a divisor
less than 1 with a very long fractional part.
(Bug#31019)
On Windows, the pthread_mutex_trylock()
implementation was incorrect.
(Bug#30992)
A character set introducer followed by a hexadecimal or bit-value literal did not check its argument and could return an ill-formed result for invalid input. (Bug#30986)
CHAR( did not check its
argument and could return an ill-formed result for invalid
input.
(Bug#30982)str USING
charset)
The result from
CHAR() did not add a leading 0x00 byte for input
strings with an odd number of bytes.
(Bug#30981)str USING
ucs2
The GeomFromText() function could
cause a server crash if the first argument was
NULL or the empty string.
(Bug#30955)
MAKEDATE() incorrectly moved year
values in the 100–200 range into the 1970–2069
range. (This is legitimate for 00–99, but three-digit
years should be used unchanged.)
(Bug#30951)
When invoked with constant arguments,
STR_TO_DATE() could use a cached
value for the format string and return incorrect results.
(Bug#30942)
GROUP_CONCAT() returned
',' rather than an empty string when the
argument column contained only empty strings.
(Bug#30897)
For MEMORY tables, lookups for
NULL values in BTREE
indexes could return incorrect results.
(Bug#30885)
Calling NAME_CONST() with
nonconstant arguments triggered an assertion failure.
Nonconstant arguments are now disallowed.
(Bug#30832)
For a spatial column with a regular
(non-SPATIAL) index, queries failed if the
optimizer tried to use the index.
(Bug#30825)
Values for the --tc-heuristic-recover option
incorrectly were treated as values for the
--myisam-stats-method option.
(Bug#30821)
On Windows, the pthread_mutex_trylock()
implementation was incorrect. One symptom was that invalidating
the query cache could cause a server crash.
(Bug#30768)
Under some circumstances,
CREATE TABLE ...
SELECT could crash the server or incorrectly report
that the table row size was too large.
(Bug#30736)
Using the MIN() or
MAX() function to select one part
of a multi-part key could cause a crash when the function result
was NULL.
(Bug#30715)
The optimizer could ignore ORDER BY in cases
when the result set is ordered by filesort,
resulting in rows being returned in incorrect order.
(Bug#30666)
MyISAM tables could not exceed 4294967295
(2^32 - 1) rows on Windows.
(Bug#30638)
For MEMORY tables,
DELETE statements that remove
rows based on an index read could fail to remove all matching
rows.
(Bug#30590)
Using GROUP BY on an expression of the form
timestamp_col DIV
number
When expanding a * in a
USING or NATURAL join, the
check for table access for both tables in the join was done
using only the grant information of the first table.
(Bug#30468)
Versions of mysqldump from MySQL 4.1 or
higher tried to use START TRANSACTION WITH CONSISTENT
SNAPSHOT if the
--single-transaction and
--master-data options were
given, even with servers older than 4.1 that do not support
consistent snapshots.
(Bug#30444)
Setting certain values on a table using a spatial index could cause the server to crash. (Bug#30286)
Some INFORMATION_SCHEMA tables are intended
for internal use, but could be accessed by using
SHOW statements.
(Bug#30079)
Specifying the --without-geometry option for
configure caused server compilation to fail.
(Bug#29972)
Under some circumstances, a UDF initialization function could be passed incorrect argument lengths. (Bug#29804)
configure did not find nss
on some Linux platforms.
(Bug#29658)
The log and
log_slow_queries system
variables were displayed by SHOW
VARIABLES but could not be accessed in expressions as
@@log and
@@log_slow_queries. Also, attempting to set
them with SET produced an incorrect
Unknown system variable message. Now these
variables can be accessed in expressions and attempting to set
their values produces an error message that the variable is read
only.
(Bug#29131)
SHOW VARIABLES did not display
the relay_log,
relay_log_index, or
relay_log_info_file system variables.
(Bug#28893)
On Windows, mysql_upgrade created temporary
files in C:\ and did not clean them up.
(Bug#28774)
Index hints specified in view definitions were ignored when using the view to select from the base table. (Bug#28702)
Views do not have indexes, so index hints do not apply. Use of index hints when selecting from a view is now disallowed. (Bug#28701)
After changing the SQL mode to a restrictive value that would make already-inserted dates in a column be considered invalid, searches returned different results depending on whether the column was indexed. (Bug#28687)
The result from CHAR() was
incorrectly assumed in some contexts to return a single-byte
result.
(Bug#28550)
The parser confused user-defined function (UDF) and stored
function creation for CREATE
FUNCTION and required that there be a default database
when creating UDFs, although there is no such requirement.
(Bug#28318, Bug#29816)
The result of a comparison between
VARBINARY and
BINARY columns differed depending
on whether the VARBINARY column
was indexed.
(Bug#28076)
The metadata in some MYSQL_FIELD members
could be incorrect when a temporary table was used to evaluate a
query.
(Bug#27990)
comp_err created files with permissions such that they might be inaccessible during make install operations. (Bug#27789)
The anonymous accounts were not being created during MySQL installation. (Bug#27692)
Host names sometimes were treated as case sensitive in
account-management statements (CREATE
USER, GRANT,
REVOKE, and so forth).
(Bug#19828)
The readline library has been updated to
version 5.2. This addresses issues in the
mysql client where history and editing within
the client would fail to work as expected.
(Bug#18431)
The Aborted_clients status
variable was incremented twice if a client exited without
calling mysql_close().
(Bug#16918)
Clients were ignoring the TCP/IP port number specified as the default port via the --with-tcp-port configuration option. (Bug#15327)
Values of types REAL ZEROFILL,
DOUBLE ZEROFILL, FLOAT
ZEROFILL, were not zero-filled when converted to a
character representation in the C prepared statement API.
(Bug#11589)
mysql stripped comments from statements sent
to the server. Now the
--comments or
--skip-comments option can be
used to control whether to retain or strip comments. The default
is --skip-comments.
(Bug#11230, Bug#26215)
Several buffer-size system variables were either being handled incorrectly for large values (for settings larger than 4GB, they were truncated to values less than 4GB without a warning), or were limited unnecessarily to 4GB even on 64-bit systems. The following changes were made:
For key_buffer_size, values
larger than 4GB are allowed on 64-bit platforms (except
Windows, for which large values are truncated to 4GB with a
warning).
For join_buffer_size,
sort_buffer_size, and
myisam_sort_buffer_size,
values are limited to 4GB on all platforms. Larger values
are truncated to 4GB with a warning.
In addition, settings for
read_buffer_size and
read_rnd_buffer_size are
limited to 2GB on all platforms. Larger values are truncated to
2GB with a warning.
(Bug#5731, Bug#29419, Bug#29446)
Executing DISABLE KEYS and ENABLE
KEYS on a nonempty table would cause the size of the
index file for the table to grow considerable. This was because
the DISABLE KEYS operation would only mark
the existing index, without deleting the index blocks. The
ENABLE KEYS operation would re-create the
index, adding new blocks, while the previous index blocks would
remain. Existing indexes are now dropped and recreated when the
ENABLE KEYS statement is executed.
(Bug#4692)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.51.
Bugs fixed:
On Windows, the installer attempted to use JScript to determine whether the target data directory already existed. On Windows Vista x64, this resulted in an error because the installer was attempting to run the JScript in a 32-bit engine, which wasn't registered on Vista. The installer no longer uses JScript but instead relies on a native WiX command. (Bug#36103)
The MySQL preferences pane did not work to start or stop MySQL on Mac OS X 10.5 (Leopard). (Bug#28854)
On Mac OS X, the StartupItem for MySQL did not work. (Bug#25008)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.51.
Bugs fixed:
Security Fix: Three vulnerabilities in yaSSL versions 1.7.5 and earlier were discovered that could lead to a server crash or execution of unauthorized code. The exploit requires a server with yaSSL enabled and TCP/IP connections enabled, but does not require valid MySQL account credentials. The exploit does not apply to OpenSSL.
The proof-of-concept exploit is freely available on the Internet. Everyone with a vulnerable MySQL configuration is advised to upgrade immediately.
Security Fix:
ALTER VIEW retained the original
DEFINER value, even when altered by another
user, which could allow that user to gain the access rights of
the view. Now ALTER VIEW is
allowed only to the original definer or users with the
SUPER privilege.
(Bug#29908)
Security Fix:
When using a FEDERATED table, the local
server could be forced to crash if the remote server returned a
result with fewer columns than expected.
(Bug#29801)
When running the MySQL Instance Configuration Wizard, a race condition could exist that would fail to connect to a newly configured instance. This was because mysqld had not completed the startup process before the next stage of the installation process. (Bug#28628)
For Windows Vista, MySQLInstanceConfig.exe did not include a proper manifest enabling it to run with administrative privileges. (Bug#22563)
See also Bug#24732.
MySQLInstanceConfig.exe failed to grant
certain privileges to the 'root'@'%' account.
(Bug#17303)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.45.
Functionality added or changed:
Incompatible Change:
The parser accepted statements that contained /* ...
*/ that were not properly closed with
*/, such as SELECT 1 /* +
2. Statements that contain unclosed
/*-comments now are rejected with a syntax
error.
This fix has the potential to cause incompatibilities. Because
of Bug#26302, which caused the trailing */
to be truncated from comments in views, stored routines,
triggers, and events, it is possible that objects of those types
may have been stored with definitions that now will be rejected
as syntactically invalid. Such objects should be dropped and
re-created so that their definitions do not contain truncated
comments. If a stored object definition contains only a single
statement (does not use a BEGIN ... END
block) and contains a comment within the statement, the comment
should be moved to follow the statement or the object should be
rewritten to use a BEGIN ... END block. For
example, this statement:
CREATE PROCEDURE p() SELECT 1 /* my comment */ ;
Can be rewritten in either of these ways:
CREATE PROCEDURE p() SELECT 1; /* my comment */ CREATE PROCEDURE p() BEGIN SELECT 1 /* my comment */ ; END;
MySQL Cluster:
Mapping of NDB error codes to MySQL
storage engine error codes has been improved.
(Bug#28423)
MySQL Cluster:
auto_increment_increment and
auto_increment_offset are now
supported for NDB tables.
(Bug#26342)
MySQL Cluster: The output from the cluster management client showing the progress of data node starts has been improved. (Bug#23354)
Replication:
The sql_mode,
foreign_key_checks,
unique_checks, character
set/collations, and
sql_auto_is_null session
variables are written to the binary log and honored during
replication. See Section 5.2.3, “The Binary Log”.
Server parser performance was improved for expression parsing by lowering the number of state transitions and reductions needed. (Bug#30625)
Server parser performance was improved for boolean expressions. (Bug#30237)
If a MyISAM table is created with no
DATA DIRECTORY option, the
.MYD file is created in the database
directory. By default, if MyISAM finds an
existing .MYD file in this case, it
overwrites it. The same applies to .MYI
files for tables created with no INDEX
DIRECTORY option. To suppress this behavior, start the
server with the new --keep_files_on_create
option, in which case MyISAM will not
overwrite existing files and returns an error instead.
(Bug#29325)
MySQL source distributions are now available in Zip format. (Bug#27742)
If a MERGE table cannot be opened or used
because of a problem with an underlying table,
CHECK TABLE now displays
information about which table caused the problem.
(Bug#26976)
The EXAMPLE storage engine is now enabled by
default.
Bugs fixed:
Security Fix:
Using RENAME TABLE against a
table with explicit DATA DIRECTORY and
INDEX DIRECTORY options can be used to
overwrite system table information by replacing the symbolic
link points. the file to which the symlink points.
MySQL will now return an error when the file to which the symlink points already exists. (Bug#32111, CVE-2007-5969)
Incompatible Change:
The file mysqld.exe was mistakenly included
in binary distributions between MySQL 5.0.42 and 5.0.48. You
should use mysqld-nt.exe.
(Bug#32197)
Incompatible Change:
Multiple-table DELETE statements
containing ambiguous aliases could have unintended side effects
such as deleting rows from the wrong table. Example:
DELETE FROM t1 AS a2 USING t1 AS a1 INNER JOIN t2 AS a2;
This bug fix enables alias declarations to be declared only in
the table_references part. Elsewhere
in the statement, alias references are allowed but not alias
declarations.
(Bug#30234)
See also Bug#27525.
Incompatible Change:
Failure to consider collation when comparing space characters
could result in incorrect index entry order, leading to
incorrect comparisons, inability to find some index values,
misordered index entries, misordered ORDER BY
results, or tables that CHECK
TABLE reports as having corrupt indexes.
As a result of this bug fix, indexes must be rebuilt for columns
that use any of these character sets:
eucjpms, euc_kr,
gb2312, latin7,
macce, ujis. See
Section 2.18.3, “Checking Whether Table Indexes Must Be Rebuilt”.
(Bug#29461)
MySQL Cluster: Packaging:
Some commercial MySQL Cluster RPM packages included support for
the InnoDB storage engine.
(InnoDB is not part of the standard
commercial MySQL Cluster offering.)
(Bug#31989)
MySQL Cluster: Attempting to restore a backup made on a cluster host using one endian to a machine using the other endian could cause the cluster to fail. (Bug#29674)
MySQL Cluster: When restarting a data node, queries could hang during that node's start phase 5, and continue only after the node had entered phase 6. (Bug#29364)
MySQL Cluster: Replica redo logs were inconsistently handled during a system restart. (Bug#29354)
MySQL Cluster:
Reads on BLOB columns were not
locked when they needed to be to guarantee consistency.
(Bug#29102)
See also Bug#31482.
MySQL Cluster:
A query using joins between several large tables and requiring
unique index lookups failed to complete, eventually returning
Uknown Error after a very long period of
time. This occurred due to inadequate handling of instances
where the Transaction Coordinator ran out of
TransactionBufferMemory, when the cluster
should have returned NDB error code 4012 (Request
ndbd time-out).
(Bug#28804)
MySQL Cluster:
The description of the --print option provided
in the output from ndb_restore --help
was incorrect.
(Bug#27683)
MySQL Cluster:
The management client's response to START BACKUP
WAIT COMPLETED did not include the backup ID.
(Bug#27640)
MySQL Cluster:
An invalid subselect on an NDB
table could cause mysqld to crash.
(Bug#27494)
MySQL Cluster:
An attempt to perform a SELECT ... FROM
INFORMATION_SCHEMA.TABLES whose result included
information about NDB tables for
which the user had no privileges crashed the MySQL Server on
which the query was performed.
(Bug#26793)
MySQL Cluster:
Warnings and errors generated by ndb_config
--config-file=
were sent to filestdout, rather than to
stderr.
(Bug#25941)
MySQL Cluster: Large file support did not work in AIX server binaries. (Bug#10776)
Replication:
The thread ID was not reset properly after execution of
mysql_change_user(), which could
cause replication failure when replicating temporary tables.
(Bug#29734)
Replication: Operations that used the time zone replicated the time zone only for successful operations, but did not replicate the time zone for errors that need to know it. (Bug#29536)
Replication:
INSERT DELAYED statements on a
master server are replicated as non-DELAYED
inserts on slaves (which is normal, to preserve serialization),
but the inserts on the slave did not use concurrent inserts. Now
INSERT DELAYED on a slave is
converted to a concurrent insert when possible, and to a normal
insert otherwise.
(Bug#29152)
Replication:
DROP USER statements that named
multiple users, only some of which could be dropped, were
replicated incorrectly.
(Bug#29030)
Replication:
An error that happened inside
INSERT,
UPDATE, or
DELETE statements performed from
within a stored function or trigger could cause inconsistency
between master and slave servers.
(Bug#27417)
Replication: Slave servers could incorrectly interpret an out-of-memory error from the master and reconnect using the wrong binary log position. (Bug#24192)
When a TIMESTAMP with a nonzero
time part was converted to a DATE
value, no warning was generated. This caused index lookups to
assume that this is a valid conversion and was returning rows
that match a comparison between a
TIMESTAMP value and a
DATE keypart. Now a warning is
generated so that TIMESTAMP with
a nonzero time part will not match
DATE values.
(Bug#31221)
A server crash could occur when a
non-DETERMINISTIC stored function was used in
a GROUP BY clause.
(Bug#31035)
For an InnoDB table if a
SELECT was ordered by the primary
key and also had a WHERE field = value clause
on a different field that was indexed, a DESC
order instruction would be ignored.
(Bug#31001)
A failed HANDLER ... READ operation could
leave the table in a locked state.
(Bug#30632)
The optimization that uses a unique index to remove
GROUP BY did not ensure that the index was
actually used, thus violating the ORDER BY
that is implied by GROUP BY.
(Bug#30596)
SHOW STATUS LIKE 'Ssl_cipher_list' from a
MySQL client connected via SSL returned an empty string rather
than a list of available ciphers.
(Bug#30593)
Memory corruption occurred for some queries with a top-level
OR operation in the WHERE
condition if they contained equality predicates and other
sargable predicates in disjunctive parts of the condition.
(Bug#30396)
Issuing a DELETE statement having
both an ORDER BY clause and a
LIMIT clause could cause
mysqld to crash.
(Bug#30385)
The Last_query_cost status
variable value can be computed accurately only for simple
“flat” queries, not complex queries such as those
with subqueries or UNION.
However, the value was not consistently being set to 0 for
complex queries.
(Bug#30377)
Queries that had a GROUP BY clause and
selected COUNT(DISTINCT
returned
incorrect results.
(Bug#30324)bit_column)
The server created temporary tables for filesort operations in
the working directory, not in the directory specified by the
tmpdir system variable.
(Bug#30287)
The query cache does not support retrieval of statements for which column level access control applies, but the server was still caching such statements, thus wasting memory. (Bug#30269)
Using DISTINCT or GROUP BY
on a BIT column in a
SELECT statement caused the
column to be cast internally as an integer, with incorrect
results being returned from the query.
(Bug#30245)
GROUP BY on
BIT columns produced incorrect
results.
(Bug#30219)
Using KILL QUERY
or KILL
CONNECTION to kill a
SELECT statement caused a server
crash if the query cache was enabled.
(Bug#30201)
Prepared statements containing
CONNECTION_ID() could be written
improperly to the binary log.
(Bug#30200)
When a thread executing a DROP
TABLE statement was killed, the table name locks that
had been acquired were not released.
(Bug#30193)
Short-format mysql commands embedded within
/*! ... */ comments were parsed incorrectly
by mysql, which discarded the rest of the
comment including the terminating */
characters. The result was a malformed (unclosed) comment. Now
mysql does not discard the
*/ characters.
(Bug#30164)
When mysqldump wrote
DROP DATABASE statements within
version-specific comments, it included the terminating semicolon
in the wrong place, causing following statements to fail when
the dump file was reloaded.
(Bug#30126)
Use of local variables with non-ASCII names in stored procedures crashed the server. (Bug#30120)
On Windows, client libraries lacked symbols required for linking. (Bug#30118)
--myisam-recover='' (empty option value) did
not disable MyISAM recovery.
(Bug#30088)
The IS_UPDATABLE column in the
INFORMATION_SCHEMA.VIEWS table was
not always set correctly.
(Bug#30020)
Statements within stored procedures ignored the value of the
low_priority_updates system
variable.
(Bug#29963)
See also Bug#26162.
For MyISAM tables on Windows,
INSERT,
DELETE, or
UPDATE followed by
ALTER TABLE within
LOCK TABLES could cause table
corruption.
(Bug#29957)
With auto-reconnect enabled, row fetching for a prepared statement could crash after reconnect occurred because loss of the statement handler was not accounted for. (Bug#29948)
LOCK TABLES did not pre-lock
tables used in triggers of the locked tables. Unexpected locking
behavior and statement failures similar to failed:
1100: Table 'xx' was not locked with
LOCK TABLES could result.
(Bug#29929)
INSERT ... VALUES(CONNECTION_ID(), ...)
statements were written to the binary log in such a way that
they could not be properly restored.
(Bug#29928)
Adding DISTINCT could cause incorrect rows to
appear in a query result.
(Bug#29911)
Using the DATE() function in a
WHERE clause did not return any records after
encountering NULL. However, using
TRIM or CAST produced the
correct results.
(Bug#29898)
Very long prepared statements in stored procedures could cause a server crash. (Bug#29856)
If query execution involved a temporary table,
GROUP_CONCAT() could return a
result with an incorrect character set.
(Bug#29850)
If one thread was performing concurrent inserts, other threads reading from the same table using equality key searches could see the index values for new rows before the data values had been written, leading to reports of table corruption. (Bug#29838)
Repeatedly accessing a view in a stored procedure (for example, in a loop) caused a small amount of memory to be allocated per access. Although this memory is deallocated on disconnect, it could be a problem for a long running stored procedures that make repeated access of views. (Bug#29834)
mysqldump produced output that incorrectly
discarded the
NO_AUTO_VALUE_ON_ZERO value of
the sql_mode variable after
dumping triggers.
(Bug#29788)
An assertion failure occurred within yaSSL for very long keys. (Bug#29784)
For MEMORY tables, the
index_merge union access
method could return incorrect results.
(Bug#29740)
Comparison of TIME values using
the BETWEEN operator led to string
comparison, producing incorrect results in some cases. Now the
values are compared as integers.
(Bug#29739)
For a table with a DATE column
date_col such that selecting rows
with WHERE yielded
a nonempty result, adding date_col =
'date_val 00:00:00'GROUP BY
caused the result
to be empty.
(Bug#29729)date_col
In some cases, INSERT INTO ... SELECT ... GROUP
BY could insert rows even if the
SELECT by itself produced an
empty result.
(Bug#29717)
For the embedded server, the
mysql_stmt_store_result() C API
function caused a memory leak for empty result sets.
(Bug#29687)
EXPLAIN produced
Impossible where for statements of the form
SELECT ... FROM t WHERE c=0, where
c was an ENUM
column defined as a primary key.
(Bug#29661)
On Windows, ALTER TABLE hung if
records were locked in share mode by a long-running transaction.
(Bug#29644)
A left join between two views could produce incorrect results. (Bug#29604)
Certain statements with unions, subqueries, and joins could result in huge memory consumption. (Bug#29582)
Clients using SSL could hang the server. (Bug#29579)
A slave running with
--log-slave-updates would fail to
write INSERT DELAY IGNORE statements to its
binary log, resulting in different binary log contents on the
master and slave.
(Bug#29571)
An incorrect result was returned when comparing string values
that were converted to TIME
values with CAST().
(Bug#29555)
gcov coverage-testing information was not written if the server crashed. (Bug#29543)
In the ascii character set, conversion of DEL
(0x7F) to Unicode incorrectly resulted in
QUESTION MARK (0x3F) rather than DEL.
(Bug#29499)
A field packet with NULL fields caused a
libmysqlclient crash.
(Bug#29494)
When using a combination of HANDLER... READ
and DELETE on a table, MySQL
continued to open new copies of the table every time, leading to
an exhaustion of file descriptors.
(Bug#29474)
This regression was introduced by Bug#21587.
On Windows, the mysql client died if the user entered a statement and Return after entering Control-C. (Bug#29469)
Corrupt data resulted from use of SELECT ... INTO
OUTFILE ', where
file_name' FIELDS ENCLOSED
BY 'c'c is a digit or minus sign, followed
by LOAD DATA INFILE
'.
(Bug#29442)file_name' FIELDS ENCLOSED BY
'c'
Killing an INSERT DELAYED thread
caused a server crash.
(Bug#29431)
Use of SHOW BINLOG EVENTS for a
nonexistent log file followed by PURGE
BINARY LOGS caused a server crash.
(Bug#29420)
Assertion failure could occur for grouping queries that employed
DECIMAL user variables with
assignments to them.
(Bug#29417)
For CAST(,
the limits of 65 and 30 on the precision
(expr AS
DECIMAL(M,D))M) and scale
(D) were not enforced.
(Bug#29415)
If a view used a function in its
SELECT statement, the columns
from the view were not inserted into the
INFORMATION_SCHEMA.COLUMNS table.
(Bug#29408)
Results for a select query that aliases the column names against
a view could duplicate one column while omitting another. This
bug could occur for a query over a multiple-table view that
includes an ORDER BY clause in its
definition.
(Bug#29392)
mysqldump created a stray file when a given a too-long file name argument. (Bug#29361)
The special “zero”
ENUM value was coerced to the
normal empty string ENUM value
during a column-to-column copy. This affected CREATE
... SELECT statements and
SELECT statements with aggregate
functions on ENUM columns in the
GROUP BY clause.
(Bug#29360)
Optimization of queries with DETERMINISTIC
stored functions in the WHERE clause was
ineffective: A sequential scan was always used.
(Bug#29338)
MyISAM corruption could occur with the
cp932_japanese_ci collation for the
cp932 character set due to incorrect
comparison for trailing space.
(Bug#29333)
The mysql_list_fields() C API
function incorrectly set
MYSQL_FIELD::decimals for some view columns.
(Bug#29306)
FULLTEXT indexes could be corrupted by
certain gbk characters.
(Bug#29299)
SELECT ... INTO
OUTFILE followed by LOAD
DATA could result in garbled characters when the
FIELDS ENCLOSED BY clause named a delimiter
of '0', 'b',
'n', 'r',
't', 'N', or
'Z' due to an interaction of character
encoding and doubling for data values containing the enclosed-by
character.
(Bug#29294)
Sort order of the collation wasn't used when comparing trailing
spaces. This could lead to incorrect comparison results,
incorrectly created indexes, or incorrect result set order for
queries that include an ORDER BY clause.
(Bug#29261)
If an ENUM column contained
'' as one of its members (represented with
numeric value greater than 0), and the column contained error
values (represented as 0 and displayed as
''), using ALTER
TABLE to modify the column definition caused the 0
values to be given the numeric value of the nonzero
'' member.
(Bug#29251)
Calling mysql_options() after
mysql_real_connect() could cause
clients to crash.
(Bug#29247)
CHECK TABLE for
ARCHIVE tables could falsely report table
corruption or cause a server crash.
(Bug#29207)
Mixing binary and utf8 columns in a union
caused field lengths to be calculated incorrectly, resulting in
truncation.
(Bug#29205)
AsText() could fail with a buffer overrun.
(Bug#29166)
InnoDB refused to start on some versions of
FreeBSD with LinuxThreads. This is fixed by enabling file
locking on FreeBSD.
(Bug#29155)
LOCK TABLES was not atomic when
more than one InnoDB tables were locked.
(Bug#29154)
A network structure was initialized incorrectly, leading to embedded server crashes. (Bug#29117)
An assertion failure occurred if a query contained a conjunctive
predicate of the form
in
the view_column = constantWHERE clause and the GROUP
BY clause contained a reference to a different view
column. The fix also enables application of an optimization that
was being skipped if a query contained a conjunctive predicate
of the form view_column =
constantWHERE clause and
the GROUP BY clause contained a reference to
the same view column.
(Bug#29104)
A maximum of 4TB InnoDB free space was
reported by SHOW TABLE STATUS, which is
incorrect on systems with more than 4TB space.
(Bug#29097)
If an INSERT INTO
... SELECT statement inserted into the same table that
the SELECT retrieved from, and
the SELECT included
ORDER BY and LIMIT
clauses, different data was inserted than the data produced by
the SELECT executed by itself.
(Bug#29095)
Queries that performed a lookup into a
BINARY index containing key
values ending with spaces caused an assertion failure for debug
builds and incorrect results for nondebug builds.
(Bug#29087)
The semantics of BIGINT depended
on platform-specific characteristics.
(Bug#29079)
A byte-order issue in writing a spatial index to disk caused bad index files on some systems. (Bug#29070)
If one of the queries in a UNION
used the SQL_CACHE option and another query
in the UNION contained a
nondeterministic function, the result was still cached. For
example, this query was incorrectly cached:
SELECT NOW() FROM t1 UNION SELECT SQL_CACHE 1 FROM t1;
Creation of a legal stored procedure could fail if no default database had been selected. (Bug#29050)
REPLACE,
INSERT IGNORE,
and UPDATE IGNORE did not work for
FEDERATED tables.
(Bug#29019)
Inserting into InnoDB tables and executing
RESET MASTER in multiple threads
cause assertion failure in debug server binaries.
(Bug#28983)
For a ucs2 column,
GROUP_CONCAT() did not convert
separators to the result character set before inserting them,
producing a result containing a mixture of two different
character sets.
(Bug#28925)
Queries using UDFs or stored functions were cached. (Bug#28921)
For a join with GROUP BY and/or
ORDER BY and a view reference in the
FROM list, the query metadata erroneously
showed empty table aliases and database names for the view
columns.
(Bug#28898)
Coercion of ASCII values to character sets that are a superset of ASCII sometimes was not done, resulting in illegal mix of collations errors. These cases now are resolved using repertoire, a new string expression attribute (see Section 9.1.7, “String Repertoire”). (Bug#28875)
Non-utf8 characters could get mangled when
stored in CSV tables.
(Bug#28862)
ALTER VIEW is not supported as a
prepared statement but was not being rejected.
ALTER VIEW is now prohibited as a
prepared statement or when called within stored routines.
(Bug#28846)
In strict SQL mode, errors silently stopped the SQL thread even
for errors named using the --slave-skip-errors
option.
(Bug#28839)
Fast ALTER TABLE (that works
without rebuilding the table) acquired duplicate locks in the
storage engine. In MyISAM, if
ALTER TABLE was issued under
LOCK
TABLE, it caused all data inserted after
LOCK
TABLE to disappear.
(Bug#28838)
Killing an SSL connection on platforms where MySQL is compiled
with -DSIGNAL_WITH_VIO_CLOSE (Windows, Mac OS
X, and some others) could crash the server.
(Bug#28812)
Runtime changes to the
log_queries_not_using_indexes
system variable were ignored.
(Bug#28808)
Tables using the InnoDB storage engine
incremented AUTO_INCREMENT values incorrectly
with ON DUPLICATE KEY UPDATE.
(Bug#28781)
Selecting a column not present in the selected-from table caused
an extra error to be produced by SHOW
ERRORS.
(Bug#28677)
For a statement of the form CREATE t1 SELECT
, the
server created the column using the
integer_constantDECIMAL data type for large
negative values that are within the range of
BIGINT.
(Bug#28625)
For InnoDB tables, MySQL unnecessarily sorted
records in certain cases when the records were retrieved by
InnoDB in the proper order already.
(Bug#28591)
A SELECT in one connection could
be blocked by
INSERT ...
ON DUPLICATE KEY UPDATE in another connection even
when low_priority_updates is
set.
(Bug#28587)
mysql_install_db could fail to find script files that it needs. (Bug#28585)
When one thread attempts to lock two (or more) tables and
another thread executes a statement that aborts these locks
(such as REPAIR TABLE,
OPTIMIZE TABLE, or
CHECK TABLE), the thread might
get a table object with an incorrect lock type in the table
cache. The result is table corruption or a server crash.
(Bug#28574)
mysql_upgrade could run binaries dynamically linked against incorrect versions of shared libraries. (Bug#28560)
If a stored procedure was created and invoked prior to selecting
a default database with USE, a
No database selected error occurred.
(Bug#28551)
On Mac OS X, shared-library installation path names were incorrect. (Bug#28544)
Using the
--skip-add-drop-table
option with mysqldump generated
incorrect SQL if the database included any views. The recreation
of views requires the creation and removal of temporary tables.
This option suppressed the removal of those temporary tables.
The same applied to --compact
since this option also invokes
--skip-add-drop-table.
(Bug#28524)
mysqlbinlog --hexdump generated incorrect
output due to omission of the “ #
” comment character for some comment lines.
(Bug#28293)
A race condition in the interaction between
MyISAM and the query cache code caused the
query cache not to invalidate itself for concurrently inserted
data.
(Bug#28249)
Indexing column prefixes in InnoDB tables
could cause table corruption.
(Bug#28138)
Index creation could fail due to truncation of key values to the maximum key length rather than to a mulitiple of the maximum character length. (Bug#28125)
The LOCATE() function returned
NULL if any of its arguments evaluated to
NULL. Likewise, the predicate,
LOCATE(, erroneously evaluated to
str,NULL)
IS NULLFALSE.
(Bug#27932)
On Windows, symbols for yaSSL and taocrypt were missing from
mysqlclient.lib, resulting in unresolved
symbol errors for clients linked against that library.
(Bug#27861)
SHOW COLUMNS returned
NULL instead of the empty string for the
Default value of columns that had no default
specified.
(Bug#27747)
The modification of a table by a partially completed multi-column update was not recorded in the binlog, rather than being marked by an event and a corresponding error code. (Bug#27716)
With recent versions of DBD::mysql, mysqlhotcopy generated table names that were doubly qualified with the database name. (Bug#27694)
The anonymous accounts were not being created during MySQL installation. (Bug#27692)
Some SHOW statements and
INFORMATION_SCHEMA queries could expose
information not allowed by the user's access privileges.
(Bug#27629)
A stack overrun could occur when storing
DATETIME values using repeated
prepared statements.
(Bug#27592)
Dropping a user-defined function could cause a server crash if the function was still in use by another thread. (Bug#27564)
Some character mappings in the ascii.xml
file were incorrect.
As a result of this bug fix, indexes must be rebuilt for columns
that use the ascii_general_ci collation for
columns that contain any of these characters:
'`', '[',
'\', ']',
'~'. See
Section 2.18.3, “Checking Whether Table Indexes Must Be Rebuilt”.
(Bug#27562)
The parser rules for the SHOW
PROFILE statement were revised to work with older
versions of bison.
(Bug#27433)
Unsafe aliasing in the source caused a client library crash when compiled with gcc 4 at high optimization levels. (Bug#27383)
A SELECT with more than 31 nested
dependent subqueries returned an incorrect result.
(Bug#27352)
Index-based range reads could fail for comparisons that involved
contraction characters (such as ch in Czech
or ll in Spanish).
(Bug#27345)
Aggregations in subqueries that refer to outer query columns were not always correctly referenced to the proper outer query. (Bug#27333)
INSERT INTO ...
SELECT caused a crash if
innodb_locks_unsafe_for_binlog
was enabled.
(Bug#27294)
Error returns from the time() system call
were ignored.
(Bug#27198)
Phantom reads could occur under InnoDB
SERIALIZABLE isolation level.
(Bug#27197)
The SUBSTRING() function returned
the entire string instead of an empty string when it was called
from a stored procedure and when the length parameter was
specified by a variable with the value “
0 ”.
(Bug#27130)
ALTER TABLE ... ENABLE KEYS could cause
mysqld to crash when executed on a table
containing on a MyISAM table containing
billions of rows.
(Bug#27029)
FEDERATED tables had an artificially low
maximum of key length.
(Bug#26909)
Binary content 0x00 in a
BLOB column sometimes became
0x5C 0x00 following a dump and reload, which
could cause problems with data using multi-byte character sets
such as GBK (Chinese). This was due to a
problem with SELECT INTO OUTFILE whereby
LOAD DATA later incorrectly
interpreted 0x5C as the second byte of a
multi-byte sequence rather than as the
SOLIDUS (“\”) character, used by
MySQL as the escape character.
(Bug#26711)
Index creation could corrupt the table definition in the
.frm file: 1) A table with the maximum
number of key segments and maximum length key name would have a
corrupted .frm file, due to incorrect
calculation of the total key length. 2)
MyISAM would reject a table with the maximum
number of keys and the maximum number of key segments in all
keys. (It would allow one less than this total maximum.) Now
MyISAM accepts a table defined with the
maximum.
(Bug#26642)
After the first read of a TEMPORARY table,
CHECK TABLE could report the
table as being corrupt.
(Bug#26325)
If an operation had an InnoDB table, and two
triggers, AFTER UPDATE and AFTER
INSERT, competing for different resources (such as two
distinct MyISAM tables), the triggers were
unable to execute concurrently. In addition,
INSERT and
UPDATE statements for the
InnoDB table were unable to run concurrently.
(Bug#26141)
ALTER DATABASE did not require at
least one option.
(Bug#25859)
Using HANDLER to open a table
having a storage engine not supported by
HANDLER properly returned an
error, but also improperly prevented the table from being
dropped by other connections.
(Bug#25856)
The index merge union access algorithm could produce incorrect
results with InnoDB tables. The problem could
also occur for queries that used DISTINCT.
(Bug#25798)
When using a FEDERATED table, the value of
LAST_INSERT_ID() would not
correctly update the C API interface, which would affect the
autogenerated ID returned both through the C API and the MySQL
protocol, affecting Connectors that used the protocol and/or C
API.
(Bug#25714)
The server was blocked from opening other tables while the
FEDERATED engine was attempting to open a
remote table. Now the server does not check the correctness of a
FEDERATED table at
CREATE TABLE time, but waits
until the table actually is accessed.
(Bug#25679)
Under ActiveState Perl, mysql-test-run.pl
could kill itself when attempting to kill other processes.
(Bug#25657)
Several InnoDB assertion failures were
corrected.
(Bug#25645)
A query with DISTINCT in the select list to
which the loose-scan optimization for grouping queries was
applied returned an incorrect result set when the query was used
with the SQL_BIG_RESULT option.
(Bug#25602)
For a multiple-row insert into a FEDERATED
table that refers to a remote transactional table, if the insert
failed for a row due to constraint failure, the remote table
would contain a partial commit (the rows preceding the failed
one) instead of rolling back the statement completely. This
occurred because the rows were treated as individual inserts.
Now FEDERATED performs bulk-insert handling
such that multiple rows are sent to the remote table in a batch.
This provides a performance improvement and enables the remote
table to perform statement rollback properly should an error
occur. This capability has the following limitations:
The size of the insert cannot exceed the maximum packet size between servers. If the insert exceeds this size, it is broken into multiple packets and the rollback problem can occur.
Bulk-insert handling does not occur for
INSERT
... ON DUPLICATE KEY UPDATE.
The FEDERATED storage engine failed silently
for INSERT
... ON DUPLICATE KEY UPDATE if a duplicate key
violation occurred. FEDERATED does not
support ON DUPLICATE KEY UPDATE, so now it
correctly returns an ER_DUP_KEY
error if a duplicate key violation occurs.
(Bug#25511)
For InnoDB tables, CREATE TABLE a AS
SELECT * FROM A would fail.
(Bug#25164)
In a stored function or trigger, when InnoDB
detected deadlock, it attempted rollback and displayed an
incorrect error message (Explicit or implicit commit
is not allowed in stored function or trigger). Now
InnoDB returns an error under these
conditions and does not attempt rollback. Rollback is handled
outside of InnoDB above the function/trigger
level.
(Bug#24989)
A too-long shared-memory-base-name value
could cause a buffer overflow and crash the server or clients.
(Bug#24924)
Dropping a temporary InnoDB table that had
been locked with LOCK TABLES
caused a server crash.
(Bug#24918)
On Windows, executables did not include Vista manifests. (Bug#24732)
See also Bug#22563.
If MySQL/InnoDB crashed very quickly after
starting up, it would not force a checkpoint. In this case,
InnoDB would skip crash recovery at next
startup, and the database would become corrupt. Now, if the redo
log scan at InnoDB startup goes past the last
checkpoint, crash recovery is forced.
(Bug#23710)
The server deducted some bytes from the
key_cache_block_size option
value and reduced it to the next lower 512 byte boundary. The
resulting block size was not a power of two. Setting the
key_cache_block_size system
variable to a value that is not a power of two resulted in
MyISAM table corruption.
(Bug#23068, Bug#28478, Bug#25853)
SHOW INNODB STATUS caused an
assertion failure under high load.
(Bug#22819)
SHOW BINLOG EVENTS displayed
incorrect values of End_log_pos for events
associated with transactional storage engines.
(Bug#22540)
A statement of the form CREATE TABLE IF NOT EXISTS t1
SELECT f1() AS i failed with a deadlock error if the
stored function f1() referred to a table with
the same name as the to-be-created table. Now it correctly
produces a message that the table already exists.
(Bug#22427)
Read lock requests that were blocked by a pending write lock request were not allowed to proceed if the statement requesting the write lock was killed. (Bug#21281)
Under heavy load with a large query cache, invalidating part of the cache could cause the server to freeze (that is, to be unable to service other operations until the invalidation was complete). (Bug#21074)
mysql-stress-test.pl and mysqld_multi.server.sh were missing from some binary distributions. (Bug#21023, Bug#25486)
On Windows, the server used 10MB of memory for each connection thread, resulting in memory exhaustion. Now each thread uses 1MB. (Bug#20815)
Worked around an icc problem with an incorrect machine instruction being generated in the context of software pre-fetching after a subroutine got in-lined. (Upgrading to icc 10.0.026 makes the workaround unnecessary.) (Bug#20803)
InnoDB produced an unnecessary (and harmless)
warning: InnoDB: Error: trying to
declare trx to enter InnoDB, but
InnoDB: it already is declared
Under ActiveState Perl, mysql-test-run.pl
would not run.
(Bug#18415)
The server crashed when the size of an
ARCHIVE table grew larger than 2GB.
(Bug#15787)
SQL_BIG_RESULT had no effect for
CREATE TABLE ... SELECT SQL_BIG_RESULT ...
statements.
(Bug#15130)
On 64-bit Windows systems, the Config Wizard failed to complete
the setup because 64-bit Windows does not resolve dynamic
linking of the 64-bit libmysql.dll to a
32-bit application like the Config Wizard.
(Bug#14649)
mysql_setpermission tried to grant global-only privileges at the database level. (Bug#14618)
Parameters of type DATETIME or
DATE in stored procedures were
silently converted to VARBINARY.
(Bug#13675)
For the general query log, logging of prepared statements
executed via the C API differed from logging of prepared
statements performed with PREPARE
and EXECUTE. Logging for the
latter was missing the Prepare and
Execute lines.
(Bug#13326)
The server returned data from SHOW CREATE
TABLE statement or a
SELECT statement on an
INFORMATION_SCHEMA table using the binary
character set.
(Bug#10491)
Backup software can cause
ERROR_SHARING_VIOLATION or
ERROR_LOCK_VIOLATION conditions during file
operations. InnoDB now retries forever until
the condition goes away.
(Bug#9709)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This is a bugfix release that replaces MySQL 5.0.50sp1.
Bugs fixed:
Security Fix: Three vulnerabilities in yaSSL versions 1.7.5 and earlier were discovered that could lead to a server crash or execution of unauthorized code. The exploit requires a server with yaSSL enabled and TCP/IP connections enabled, but does not require valid MySQL account credentials. The exploit does not apply to OpenSSL.
The proof-of-concept exploit is freely available on the Internet. Everyone with a vulnerable MySQL configuration is advised to upgrade immediately.
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.50). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Bugs fixed:
Security Fix:
Using RENAME TABLE against a
table with explicit DATA DIRECTORY and
INDEX DIRECTORY options can be used to
overwrite system table information by replacing the symbolic
link points. the file to which the symlink points.
MySQL will now return an error when the file to which the symlink points already exists. (Bug#32111, CVE-2007-5969)
Security Fix:
ALTER VIEW retained the original
DEFINER value, even when altered by another
user, which could allow that user to gain the access rights of
the view. Now ALTER VIEW is
allowed only to the original definer or users with the
SUPER privilege.
(Bug#29908)
Security Fix:
When using a FEDERATED table, the local
server could be forced to crash if the remote server returned a
result with fewer columns than expected.
(Bug#29801)
A build problem introduced in MySQL 5.0.52 was resolved: The x86 32-bit Intel icc-compiled server binary had unwanted dependences on Intel icc runtime libraries. (Bug#32514)
InnoDB does not support
SPATIAL indexes, but could crash when asked
to handle one. Now an error is returned.
(Bug#32125)
mysql-test-run.pl could not run
mysqld with root
privileges.
(Bug#30630)
InnoDB had a race condition for an adaptive
hash rw-lock waiting for an X-lock. This fix may also provide
significant speed improvements on systems experiencing problems
with contention for the adaptive hash index.
(Bug#29560)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.48). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
Incompatible Change:
The parser accepted statements that contained /* ...
*/ that were not properly closed with
*/, such as SELECT 1 /* +
2. Statements that contain unclosed
/*-comments now are rejected with a syntax
error.
This fix has the potential to cause incompatibilities. Because
of Bug#26302, which caused the trailing */
to be truncated from comments in views, stored routines,
triggers, and events, it is possible that objects of those types
may have been stored with definitions that now will be rejected
as syntactically invalid. Such objects should be dropped and
re-created so that their definitions do not contain truncated
comments. If a stored object definition contains only a single
statement (does not use a BEGIN ... END
block) and contains a comment within the statement, the comment
should be moved to follow the statement or the object should be
rewritten to use a BEGIN ... END block. For
example, this statement:
CREATE PROCEDURE p() SELECT 1 /* my comment */ ;
Can be rewritten in either of these ways:
CREATE PROCEDURE p() SELECT 1; /* my comment */ CREATE PROCEDURE p() BEGIN SELECT 1 /* my comment */ ; END;
MySQL Cluster:
Mapping of NDB error codes to MySQL
storage engine error codes has been improved.
(Bug#28423)
MySQL Cluster: The output from the cluster management client showing the progress of data node starts has been improved. (Bug#23354)
Server parser performance was improved for expression parsing by lowering the number of state transitions and reductions needed. (Bug#30625)
Server parser performance was improved for boolean expressions. (Bug#30237)
Bugs fixed:
Incompatible Change:
The file mysqld.exe was mistakenly included
in binary distributions between MySQL 5.0.42 and 5.0.48. You
should use mysqld-nt.exe.
(Bug#32197)
Incompatible Change:
Multiple-table DELETE statements
containing ambiguous aliases could have unintended side effects
such as deleting rows from the wrong table. Example:
DELETE FROM t1 AS a2 USING t1 AS a1 INNER JOIN t2 AS a2;
This fix enables alias declarations to be made only in the
table_references part. Elsewhere in
the statement, alias references are allowed but not alias
declarations. However, this patch was reverted in MySQL 5.0.54
because it changed the behavior of a General Availability MySQL
release.
(Bug#30234)
See also Bug#27525.
MySQL Cluster: Packaging:
Some commercial MySQL Cluster RPM packages included support for
the InnoDB storage engine.
(InnoDB is not part of the standard
commercial MySQL Cluster offering.)
(Bug#31989)
MySQL Cluster: Attempting to restore a backup made on a cluster host using one endian to a machine using the other endian could cause the cluster to fail. (Bug#29674)
MySQL Cluster:
Reads on BLOB columns were not
locked when they needed to be to guarantee consistency.
(Bug#29102)
See also Bug#31482.
MySQL Cluster:
A query using joins between several large tables and requiring
unique index lookups failed to complete, eventually returning
Uknown Error after a very long period of
time. This occurred due to inadequate handling of instances
where the Transaction Coordinator ran out of
TransactionBufferMemory, when the cluster
should have returned NDB error code 4012 (Request
ndbd time-out).
(Bug#28804)
MySQL Cluster:
The description of the --print option provided
in the output from ndb_restore --help
was incorrect.
(Bug#27683)
MySQL Cluster:
An invalid subselect on an NDB
table could cause mysqld to crash.
(Bug#27494)
MySQL Cluster:
An attempt to perform a SELECT ... FROM
INFORMATION_SCHEMA.TABLES whose result included
information about NDB tables for
which the user had no privileges crashed the MySQL Server on
which the query was performed.
(Bug#26793)
When a TIMESTAMP with a nonzero
time part was converted to a DATE
value, no warning was generated. This caused index lookups to
assume that this is a valid conversion and was returning rows
that match a comparison between a
TIMESTAMP value and a
DATE keypart. Now a warning is
generated so that TIMESTAMP with
a nonzero time part will not match
DATE values.
(Bug#31221)
A server crash could occur when a
non-DETERMINISTIC stored function was used in
a GROUP BY clause.
(Bug#31035)
For an InnoDB table if a
SELECT was ordered by the primary
key and also had a WHERE field = value clause
on a different field that was indexed, a DESC
order instruction would be ignored.
(Bug#31001)
A failed HANDLER ... READ operation could
leave the table in a locked state.
(Bug#30632)
The optimization that uses a unique index to remove
GROUP BY did not ensure that the index was
actually used, thus violating the ORDER BY
that is implied by GROUP BY.
(Bug#30596)
SHOW STATUS LIKE 'Ssl_cipher_list' from a
MySQL client connected via SSL returned an empty string rather
than a list of available ciphers.
(Bug#30593)
Issuing a DELETE statement having
both an ORDER BY clause and a
LIMIT clause could cause
mysqld to crash.
(Bug#30385)
The Last_query_cost status
variable value can be computed accurately only for simple
“flat” queries, not complex queries such as those
with subqueries or UNION.
However, the value was not consistently being set to 0 for
complex queries.
(Bug#30377)
Queries that had a GROUP BY clause and
selected COUNT(DISTINCT
returned
incorrect results.
(Bug#30324)bit_column)
Using DISTINCT or GROUP BY
on a BIT column in a
SELECT statement caused the
column to be cast internally as an integer, with incorrect
results being returned from the query.
(Bug#30245)
Short-format mysql commands embedded within
/*! ... */ comments were parsed incorrectly
by mysql, which discarded the rest of the
comment including the terminating */
characters. The result was a malformed (unclosed) comment. Now
mysql does not discard the
*/ characters.
(Bug#30164)
When mysqldump wrote
DROP DATABASE statements within
version-specific comments, it included the terminating semicolon
in the wrong place, causing following statements to fail when
the dump file was reloaded.
(Bug#30126)
If a view used a function in its
SELECT statement, the columns
from the view were not inserted into the
INFORMATION_SCHEMA.COLUMNS table.
(Bug#29408)
Killing an SSL connection on platforms where MySQL is compiled
with -DSIGNAL_WITH_VIO_CLOSE (Windows, Mac OS
X, and some others) could crash the server.
(Bug#28812)
A SELECT in one connection could
be blocked by
INSERT ...
ON DUPLICATE KEY UPDATE in another connection even
when low_priority_updates is
set.
(Bug#28587)
mysql_upgrade could run binaries dynamically linked against incorrect versions of shared libraries. (Bug#28560)
SHOW COLUMNS returned
NULL instead of the empty string for the
Default value of columns that had no default
specified.
(Bug#27747)
With recent versions of DBD::mysql, mysqlhotcopy generated table names that were doubly qualified with the database name. (Bug#27694)
For InnoDB tables, CREATE TABLE a AS
SELECT * FROM A would fail.
(Bug#25164)
Under heavy load with a large query cache, invalidating part of the cache could cause the server to freeze (that is, to be unable to service other operations until the invalidation was complete). (Bug#21074)
Worked around an icc problem with an incorrect machine instruction being generated in the context of software pre-fetching after a subroutine got in-lined. (Upgrading to icc 10.0.026 makes the workaround unnecessary.) (Bug#20803)
Parameters of type DATETIME or
DATE in stored procedures were
silently converted to VARBINARY.
(Bug#13675)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This release was withdrawn from production and is no longer available.
This section documents all changes and bugfixes that have been applied since the last last MySQL Enterprise Server release (5.0.46). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
If a MyISAM table is created with no
DATA DIRECTORY option, the
.MYD file is created in the database
directory. By default, if MyISAM finds an
existing .MYD file in this case, it
overwrites it. The same applies to .MYI
files for tables created with no INDEX
DIRECTORY option. To suppress this behavior, start the
server with the new --keep_files_on_create
option, in which case MyISAM will not
overwrite existing files and returns an error instead.
(Bug#29325)
MySQL source distributions are now available in Zip format. (Bug#27742)
The EXAMPLE storage engine is now enabled by
default.
Bugs fixed:
Incompatible Change:
Failure to consider collation when comparing space characters
could result in incorrect index entry order, leading to
incorrect comparisons, inability to find some index values,
misordered index entries, misordered ORDER BY
results, or tables that CHECK
TABLE reports as having corrupt indexes.
As a result of this bug fix, indexes must be rebuilt for columns
that use any of these character sets:
eucjpms, euc_kr,
gb2312, latin7,
macce, ujis. See
Section 2.18.3, “Checking Whether Table Indexes Must Be Rebuilt”.
(Bug#29461)
MySQL Cluster:
Warnings and errors generated by ndb_config
--config-file=
were sent to filestdout, rather than to
stderr.
(Bug#25941)
MySQL Cluster:
When a cluster backup was terminated using the ABORT
BACKUP command in the management client, a misleading
error message Backup aborted by application:
Permanent error: Internal error was returned. The
error message returned in such cases now reads Backup
aborted by user request.
(Bug#21052)
MySQL Cluster: Large file support did not work in AIX server binaries. (Bug#10776)
Replication:
SHOW SLAVE STATUS failed when
slave I/O was about to terminate.
(Bug#34305)
Replication:
The thread ID was not reset properly after execution of
mysql_change_user(), which could
cause replication failure when replicating temporary tables.
(Bug#29734)
Replication: Operations that used the time zone replicated the time zone only for successful operations, but did not replicate the time zone for errors that need to know it. (Bug#29536)
Replication:
INSERT DELAYED statements on a
master server are replicated as non-DELAYED
inserts on slaves (which is normal, to preserve serialization),
but the inserts on the slave did not use concurrent inserts. Now
INSERT DELAYED on a slave is
converted to a concurrent insert when possible, and to a normal
insert otherwise.
(Bug#29152)
Replication:
An error that happened inside
INSERT,
UPDATE, or
DELETE statements performed from
within a stored function or trigger could cause inconsistency
between master and slave servers.
(Bug#27417)
Replication: Slave servers could incorrectly interpret an out-of-memory error from the master and reconnect using the wrong binary log position. (Bug#24192)
Memory corruption occurred for some queries with a top-level
OR operation in the WHERE
condition if they contained equality predicates and other
sargable predicates in disjunctive parts of the condition.
(Bug#30396)
The server created temporary tables for filesort operations in
the working directory, not in the directory specified by the
tmpdir system variable.
(Bug#30287)
The query cache does not support retrieval of statements for which column level access control applies, but the server was still caching such statements, thus wasting memory. (Bug#30269)
GROUP BY on
BIT columns produced incorrect
results.
(Bug#30219)
Using KILL QUERY
or KILL
CONNECTION to kill a
SELECT statement caused a server
crash if the query cache was enabled.
(Bug#30201)
Prepared statements containing
CONNECTION_ID() could be written
improperly to the binary log.
(Bug#30200)
When a thread executing a DROP
TABLE statement was killed, the table name locks that
had been acquired were not released.
(Bug#30193)
Use of local variables with non-ASCII names in stored procedures crashed the server. (Bug#30120)
On Windows, client libraries lacked symbols required for linking. (Bug#30118)
--myisam-recover='' (empty option value) did
not disable MyISAM recovery.
(Bug#30088)
The IS_UPDATABLE column in the
INFORMATION_SCHEMA.VIEWS table was
not always set correctly.
(Bug#30020)
Statements within stored procedures ignored the value of the
low_priority_updates system
variable.
(Bug#29963)
See also Bug#26162.
For MyISAM tables on Windows,
INSERT,
DELETE, or
UPDATE followed by
ALTER TABLE within
LOCK TABLES could cause table
corruption.
(Bug#29957)
With auto-reconnect enabled, row fetching for a prepared statement could crash after reconnect occurred because loss of the statement handler was not accounted for. (Bug#29948)
LOCK TABLES did not pre-lock
tables used in triggers of the locked tables. Unexpected locking
behavior and statement failures similar to failed:
1100: Table 'xx' was not locked with
LOCK TABLES could result.
(Bug#29929)
INSERT ... VALUES(CONNECTION_ID(), ...)
statements were written to the binary log in such a way that
they could not be properly restored.
(Bug#29928)
Adding DISTINCT could cause incorrect rows to
appear in a query result.
(Bug#29911)
Using the DATE() function in a
WHERE clause did not return any records after
encountering NULL. However, using
TRIM or CAST produced the
correct results.
(Bug#29898)
Very long prepared statements in stored procedures could cause a server crash. (Bug#29856)
If query execution involved a temporary table,
GROUP_CONCAT() could return a
result with an incorrect character set.
(Bug#29850)
If one thread was performing concurrent inserts, other threads reading from the same table using equality key searches could see the index values for new rows before the data values had been written, leading to reports of table corruption. (Bug#29838)
Repeatedly accessing a view in a stored procedure (for example, in a loop) caused a small amount of memory to be allocated per access. Although this memory is deallocated on disconnect, it could be a problem for a long running stored procedures that make repeated access of views. (Bug#29834)
mysqldump produced output that incorrectly
discarded the
NO_AUTO_VALUE_ON_ZERO value of
the sql_mode variable after
dumping triggers.
(Bug#29788)
An assertion failure occurred within yaSSL for very long keys. (Bug#29784)
For MEMORY tables, the
index_merge union access
method could return incorrect results.
(Bug#29740)
Comparison of TIME values using
the BETWEEN operator led to string
comparison, producing incorrect results in some cases. Now the
values are compared as integers.
(Bug#29739)
For a table with a DATE column
date_col such that selecting rows
with WHERE yielded
a nonempty result, adding date_col =
'date_val 00:00:00'GROUP BY
caused the result
to be empty.
(Bug#29729)date_col
In some cases, INSERT INTO ... SELECT ... GROUP
BY could insert rows even if the
SELECT by itself produced an
empty result.
(Bug#29717)
For the embedded server, the
mysql_stmt_store_result() C API
function caused a memory leak for empty result sets.
(Bug#29687)
EXPLAIN produced
Impossible where for statements of the form
SELECT ... FROM t WHERE c=0, where
c was an ENUM
column defined as a primary key.
(Bug#29661)
On Windows, ALTER TABLE hung if
records were locked in share mode by a long-running transaction.
(Bug#29644)
A left join between two views could produce incorrect results. (Bug#29604)
Certain statements with unions, subqueries, and joins could result in huge memory consumption. (Bug#29582)
Clients using SSL could hang the server. (Bug#29579)
A slave running with
--log-slave-updates would fail to
write INSERT DELAY IGNORE statements to its
binary log, resulting in different binary log contents on the
master and slave.
(Bug#29571)
An incorrect result was returned when comparing string values
that were converted to TIME
values with CAST().
(Bug#29555)
In the ascii character set, conversion of DEL
(0x7F) to Unicode incorrectly resulted in
QUESTION MARK (0x3F) rather than DEL.
(Bug#29499)
A field packet with NULL fields caused a
libmysqlclient crash.
(Bug#29494)
When using a combination of HANDLER... READ
and DELETE on a table, MySQL
continued to open new copies of the table every time, leading to
an exhaustion of file descriptors.
(Bug#29474)
This regression was introduced by Bug#21587.
On Windows, the mysql client died if the user entered a statement and Return after entering Control-C. (Bug#29469)
Killing an INSERT DELAYED thread
caused a server crash.
(Bug#29431)
The special “zero”
ENUM value was coerced to the
normal empty string ENUM value
during a column-to-column copy. This affected CREATE
... SELECT statements and
SELECT statements with aggregate
functions on ENUM columns in the
GROUP BY clause.
(Bug#29360)
Optimization of queries with DETERMINISTIC
stored functions in the WHERE clause was
ineffective: A sequential scan was always used.
(Bug#29338)
MyISAM corruption could occur with the
cp932_japanese_ci collation for the
cp932 character set due to incorrect
comparison for trailing space.
(Bug#29333)
The mysql_list_fields() C API
function incorrectly set
MYSQL_FIELD::decimals for some view columns.
(Bug#29306)
InnoDB refused to start on some versions of
FreeBSD with LinuxThreads. This is fixed by enabling file
locking on FreeBSD.
(Bug#29155)
A maximum of 4TB InnoDB free space was
reported by SHOW TABLE STATUS, which is
incorrect on systems with more than 4TB space.
(Bug#29097)
A byte-order issue in writing a spatial index to disk caused bad index files on some systems. (Bug#29070)
Creation of a legal stored procedure could fail if no default database had been selected. (Bug#29050)
Coercion of ASCII values to character sets that are a superset of ASCII sometimes was not done, resulting in illegal mix of collations errors. These cases now are resolved using repertoire, a new string expression attribute (see Section 9.1.7, “String Repertoire”). (Bug#28875)
Fast ALTER TABLE (that works
without rebuilding the table) acquired duplicate locks in the
storage engine. In MyISAM, if
ALTER TABLE was issued under
LOCK
TABLE, it caused all data inserted after
LOCK
TABLE to disappear.
(Bug#28838)
Tables using the InnoDB storage engine
incremented AUTO_INCREMENT values incorrectly
with ON DUPLICATE KEY UPDATE.
(Bug#28781)
Starting the server with an
innodb_force_recovery value of
4 did not work.
(Bug#28604)
For InnoDB tables, MySQL unnecessarily sorted
records in certain cases when the records were retrieved by
InnoDB in the proper order already.
(Bug#28591)
mysql_install_db could fail to find script files that it needs. (Bug#28585)
If a stored procedure was created and invoked prior to selecting
a default database with USE, a
No database selected error occurred.
(Bug#28551)
On Mac OS X, shared-library installation path names were incorrect. (Bug#28544)
Using the
--skip-add-drop-table
option with mysqldump generated
incorrect SQL if the database included any views. The recreation
of views requires the creation and removal of temporary tables.
This option suppressed the removal of those temporary tables.
The same applied to --compact
since this option also invokes
--skip-add-drop-table.
(Bug#28524)
A race condition in the interaction between
MyISAM and the query cache code caused the
query cache not to invalidate itself for concurrently inserted
data.
(Bug#28249)
Indexing column prefixes in InnoDB tables
could cause table corruption.
(Bug#28138)
Index creation could fail due to truncation of key values to the maximum key length rather than to a mulitiple of the maximum character length. (Bug#28125)
On Windows, symbols for yaSSL and taocrypt were missing from
mysqlclient.lib, resulting in unresolved
symbol errors for clients linked against that library.
(Bug#27861)
Some SHOW statements and
INFORMATION_SCHEMA queries could expose
information not allowed by the user's access privileges.
(Bug#27629)
Some character mappings in the ascii.xml
file were incorrect.
As a result of this bug fix, indexes must be rebuilt for columns
that use the ascii_general_ci collation for
columns that contain any of these characters:
'`', '[',
'\', ']',
'~'. See
Section 2.18.3, “Checking Whether Table Indexes Must Be Rebuilt”.
(Bug#27562)
A SELECT with more than 31 nested
dependent subqueries returned an incorrect result.
(Bug#27352)
INSERT INTO ...
SELECT caused a crash if
innodb_locks_unsafe_for_binlog
was enabled.
(Bug#27294)
FEDERATED tables had an artificially low
maximum of key length.
(Bug#26909)
After the first read of a TEMPORARY table,
CHECK TABLE could report the
table as being corrupt.
(Bug#26325)
If an operation had an InnoDB table, and two
triggers, AFTER UPDATE and AFTER
INSERT, competing for different resources (such as two
distinct MyISAM tables), the triggers were
unable to execute concurrently. In addition,
INSERT and
UPDATE statements for the
InnoDB table were unable to run concurrently.
(Bug#26141)
ALTER DATABASE did not require at
least one option.
(Bug#25859)
Using HANDLER to open a table
having a storage engine not supported by
HANDLER properly returned an
error, but also improperly prevented the table from being
dropped by other connections.
(Bug#25856)
When using a FEDERATED table, the value of
LAST_INSERT_ID() would not
correctly update the C API interface, which would affect the
autogenerated ID returned both through the C API and the MySQL
protocol, affecting Connectors that used the protocol and/or C
API.
(Bug#25714)
The server was blocked from opening other tables while the
FEDERATED engine was attempting to open a
remote table. Now the server does not check the correctness of a
FEDERATED table at
CREATE TABLE time, but waits
until the table actually is accessed.
(Bug#25679)
Several InnoDB assertion failures were
corrected.
(Bug#25645)
In a stored function or trigger, when InnoDB
detected deadlock, it attempted rollback and displayed an
incorrect error message (Explicit or implicit commit
is not allowed in stored function or trigger). Now
InnoDB returns an error under these
conditions and does not attempt rollback. Rollback is handled
outside of InnoDB above the function/trigger
level.
(Bug#24989)
Dropping a temporary InnoDB table that had
been locked with LOCK TABLES
caused a server crash.
(Bug#24918)
On Windows, executables did not include Vista manifests. (Bug#24732)
See also Bug#22563.
If MySQL/InnoDB crashed very quickly after
starting up, it would not force a checkpoint. In this case,
InnoDB would skip crash recovery at next
startup, and the database would become corrupt. Now, if the redo
log scan at InnoDB startup goes past the last
checkpoint, crash recovery is forced.
(Bug#23710)
SHOW INNODB STATUS caused an
assertion failure under high load.
(Bug#22819)
A statement of the form CREATE TABLE IF NOT EXISTS t1
SELECT f1() AS i failed with a deadlock error if the
stored function f1() referred to a table with
the same name as the to-be-created table. Now it correctly
produces a message that the table already exists.
(Bug#22427)
Read lock requests that were blocked by a pending write lock request were not allowed to proceed if the statement requesting the write lock was killed. (Bug#21281)
On Windows, the server used 10MB of memory for each connection thread, resulting in memory exhaustion. Now each thread uses 1MB. (Bug#20815)
InnoDB produced an unnecessary (and harmless)
warning: InnoDB: Error: trying to
declare trx to enter InnoDB, but
InnoDB: it already is declared
SQL_BIG_RESULT had no effect for
CREATE TABLE ... SELECT SQL_BIG_RESULT ...
statements.
(Bug#15130)
mysql_setpermission tried to grant global-only privileges at the database level. (Bug#14618)
For the general query log, logging of prepared statements
executed via the C API differed from logging of prepared
statements performed with PREPARE
and EXECUTE. Logging for the
latter was missing the Prepare and
Execute lines.
(Bug#13326)
Backup software can cause
ERROR_SHARING_VIOLATION or
ERROR_LOCK_VIOLATION conditions during file
operations. InnoDB now retries forever until
the condition goes away.
(Bug#9709)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bugfixes that have been applied since the last MySQL Enterprise Server release (5.0.44). If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise/advisors.html.
Functionality added or changed:
MySQL Cluster:
auto_increment_increment and
auto_increment_offset are now
supported for NDB tables.
(Bug#26342)
Replication:
The sql_mode,
foreign_key_checks,
unique_checks, character
set/collations, and
sql_auto_is_null session
variables are written to the binary log and honored during
replication. See Section 5.2.3, “The Binary Log”.
If a MERGE table cannot be opened or used
because of a problem with an underlying table,
CHECK TABLE now displays
information about which table caused the problem.
(Bug#26976)
Bugs fixed:
MySQL Cluster: When restarting a data node, queries could hang during that node's start phase 5, and continue only after the node had entered phase 6. (Bug#29364)
MySQL Cluster: Replica redo logs were inconsistently handled during a system restart. (Bug#29354)
MySQL Cluster:
The management client's response to START BACKUP
WAIT COMPLETED did not include the backup ID.
(Bug#27640)
Replication:
DROP USER statements that named
multiple users, only some of which could be dropped, were
replicated incorrectly.
(Bug#29030)
On the IBM i5 platform, the installation script in the
.savf binaries unconditionally executed the
mysql_install_db script.
(Bug#30084)
gcov coverage-testing information was not written if the server crashed. (Bug#29543)
Corrupt data resulted from use of SELECT ... INTO
OUTFILE ', where
file_name' FIELDS ENCLOSED
BY 'c'c is a digit or minus sign, followed
by LOAD DATA INFILE
'.
(Bug#29442)file_name' FIELDS ENCLOSED BY
'c'
Use of SHOW BINLOG EVENTS for a
nonexistent log file followed by PURGE
BINARY LOGS caused a server crash.
(Bug#29420)
Assertion failure could occur for grouping queries that employed
DECIMAL user variables with
assignments to them.
(Bug#29417)
For CAST(,
the limits of 65 and 30 on the precision
(expr AS
DECIMAL(M,D))M) and scale
(D) were not enforced.
(Bug#29415)
Results for a select query that aliases the column names against
a view could duplicate one column while omitting another. This
bug could occur for a query over a multiple-table view that
includes an ORDER BY clause in its
definition.
(Bug#29392)
mysqldump created a stray file when a given a too-long file name argument. (Bug#29361)
FULLTEXT indexes could be corrupted by
certain gbk characters.
(Bug#29299)
SELECT ... INTO
OUTFILE followed by LOAD
DATA could result in garbled characters when the
FIELDS ENCLOSED BY clause named a delimiter
of '0', 'b',
'n', 'r',
't', 'N', or
'Z' due to an interaction of character
encoding and doubling for data values containing the enclosed-by
character.
(Bug#29294)
Sort order of the collation wasn't used when comparing trailing
spaces. This could lead to incorrect comparison results,
incorrectly created indexes, or incorrect result set order for
queries that include an ORDER BY clause.
(Bug#29261)
If an ENUM column contained
'' as one of its members (represented with
numeric value greater than 0), and the column contained error
values (represented as 0 and displayed as
''), using ALTER
TABLE to modify the column definition caused the 0
values to be given the numeric value of the nonzero
'' member.
(Bug#29251)
Calling mysql_options() after
mysql_real_connect() could cause
clients to crash.
(Bug#29247)
CHECK TABLE for
ARCHIVE tables could falsely report table
corruption or cause a server crash.
(Bug#29207)
Mixing binary and utf8 columns in a union
caused field lengths to be calculated incorrectly, resulting in
truncation.
(Bug#29205)
AsText() could fail with a buffer overrun.
(Bug#29166)
LOCK TABLES was not atomic when
more than one InnoDB tables were locked.
(Bug#29154)
A network structure was initialized incorrectly, leading to embedded server crashes. (Bug#29117)
An assertion failure occurred if a query contained a conjunctive
predicate of the form
in
the view_column = constantWHERE clause and the GROUP
BY clause contained a reference to a different view
column. The fix also enables application of an optimization that
was being skipped if a query contained a conjunctive predicate
of the form view_column =
constantWHERE clause and
the GROUP BY clause contained a reference to
the same view column.
(Bug#29104)
If an INSERT INTO
... SELECT statement inserted into the same table that
the SELECT retrieved from, and
the SELECT included
ORDER BY and LIMIT
clauses, different data was inserted than the data produced by
the SELECT executed by itself.
(Bug#29095)
Queries that performed a lookup into a
BINARY index containing key
values ending with spaces caused an assertion failure for debug
builds and incorrect results for nondebug builds.
(Bug#29087)
The semantics of BIGINT depended
on platform-specific characteristics.
(Bug#29079)
If one of the queries in a UNION
used the SQL_CACHE option and another query
in the UNION contained a
nondeterministic function, the result was still cached. For
example, this query was incorrectly cached:
SELECT NOW() FROM t1 UNION SELECT SQL_CACHE 1 FROM t1;
REPLACE,
INSERT IGNORE,
and UPDATE IGNORE did not work for
FEDERATED tables.
(Bug#29019)
Inserting into InnoDB tables and executing
RESET MASTER in multiple threads
cause assertion failure in debug server binaries.
(Bug#28983)
For a ucs2 column,
GROUP_CONCAT() did not convert
separators to the result character set before inserting them,
producing a result containing a mixture of two different
character sets.
(Bug#28925)
Queries using UDFs or stored functions were cached. (Bug#28921)
For a join with GROUP BY and/or
ORDER BY and a view reference in the
FROM list, the query metadata erroneously
showed empty table aliases and database names for the view
columns.
(Bug#28898)
Non-utf8 characters could get mangled when
stored in CSV tables.
(Bug#28862)
ALTER VIEW is not supported as a
prepared statement but was not being rejected.
ALTER VIEW is now prohibited as a
prepared statement or when called within stored routines.
(Bug#28846)
In strict SQL mode, errors silently stopped the SQL thread even
for errors named using the --slave-skip-errors
option.
(Bug#28839)
Runtime changes to the
log_queries_not_using_indexes
system variable were ignored.
(Bug#28808)
Selecting a column not present in the selected-from table caused
an extra error to be produced by SHOW
ERRORS.
(Bug#28677)
For a statement of the form CREATE t1 SELECT
, the
server created the column using the
integer_constantDECIMAL data type for large
negative values that are within the range of
BIGINT.
(Bug#28625)
When one thread attempts to lock two (or more) tables and
another thread executes a statement that aborts these locks
(such as REPAIR TABLE,
OPTIMIZE TABLE, or
CHECK TABLE), the thread might
get a table object with an incorrect lock type in the table
cache. The result is table corruption or a server crash.
(Bug#28574)
mysqlbinlog --hexdump generated incorrect
output due to omission of the “ #
” comment character for some comment lines.
(Bug#28293)
The LOCATE() function returned
NULL if any of its arguments evaluated to
NULL. Likewise, the predicate,
LOCATE(, erroneously evaluated to
str,NULL)
IS NULLFALSE.
(Bug#27932)
The modification of a table by a partially completed multi-column update was not recorded in the binlog, rather than being marked by an event and a corresponding error code. (Bug#27716)
A stack overrun could occur when storing
DATETIME values using repeated
prepared statements.
(Bug#27592)
Dropping a user-defined function could cause a server crash if the function was still in use by another thread. (Bug#27564)
Unsafe aliasing in the source caused a client library crash when compiled with gcc 4 at high optimization levels. (Bug#27383)
Index-based range reads could fail for comparisons that involved
contraction characters (such as ch in Czech
or ll in Spanish).
(Bug#27345)
Aggregations in subqueries that refer to outer query columns were not always correctly referenced to the proper outer query. (Bug#27333)
Error returns from the time() system call
were ignored.
(Bug#27198)
Phantom reads could occur under InnoDB
SERIALIZABLE isolation level.
(Bug#27197)
The SUBSTRING() function returned
the entire string instead of an empty string when it was called
from a stored procedure and when the length parameter was
specified by a variable with the value “
0 ”.
(Bug#27130)
ALTER TABLE ... ENABLE KEYS could cause
mysqld to crash when executed on a table
containing on a MyISAM table containing
billions of rows.
(Bug#27029)
Binary content 0x00 in a
BLOB column sometimes became
0x5C 0x00 following a dump and reload, which
could cause problems with data using multi-byte character sets
such as GBK (Chinese). This was due to a
problem with SELECT INTO OUTFILE whereby
LOAD DATA later incorrectly
interpreted 0x5C as the second byte of a
multi-byte sequence rather than as the
SOLIDUS (“\”) character, used by
MySQL as the escape character.
(Bug#26711)
Index creation could corrupt the table definition in the
.frm file: 1) A table with the maximum
number of key segments and maximum length key name would have a
corrupted .frm file, due to incorrect
calculation of the total key length. 2)
MyISAM would reject a table with the maximum
number of keys and the maximum number of key segments in all
keys. (It would allow one less than this total maximum.) Now
MyISAM accepts a table defined with the
maximum.
(Bug#26642)
The index merge union access algorithm could produce incorrect
results with InnoDB tables. The problem could
also occur for queries that used DISTINCT.
(Bug#25798)
Under ActiveState Perl, mysql-test-run.pl
could kill itself when attempting to kill other processes.
(Bug#25657)
A query with DISTINCT in the select list to
which the loose-scan optimization for grouping queries was
applied returned an incorrect result set when the query was used
with the SQL_BIG_RESULT option.
(Bug#25602)
For a multiple-row insert into a FEDERATED
table that refers to a remote transactional table, if the insert
failed for a row due to constraint failure, the remote table
would contain a partial commit (the rows preceding the failed
one) instead of rolling back the statement completely. This
occurred because the rows were treated as individual inserts.
Now FEDERATED performs bulk-insert handling
such that multiple rows are sent to the remote table in a batch.
This provides a performance improvement and enables the remote
table to perform statement rollback properly should an error
occur. This capability has the following limitations:
The size of the insert cannot exceed the maximum packet size between servers. If the insert exceeds this size, it is broken into multiple packets and the rollback problem can occur.
Bulk-insert handling does not occur for
INSERT
... ON DUPLICATE KEY UPDATE.
The FEDERATED storage engine failed silently
for INSERT
... ON DUPLICATE KEY UPDATE if a duplicate key
violation occurred. FEDERATED does not
support ON DUPLICATE KEY UPDATE, so now it
correctly returns an ER_DUP_KEY
error if a duplicate key violation occurs.
(Bug#25511)
A too-long shared-memory-base-name value
could cause a buffer overflow and crash the server or clients.
(Bug#24924)
The server deducted some bytes from the
key_cache_block_size option
value and reduced it to the next lower 512 byte boundary. The
resulting block size was not a power of two. Setting the
key_cache_block_size system
variable to a value that is not a power of two resulted in
MyISAM table corruption.
(Bug#23068, Bug#28478, Bug#25853)
SHOW BINLOG EVENTS displayed
incorrect values of End_log_pos for events
associated with transactional storage engines.
(Bug#22540)
Under ActiveState Perl, mysql-test-run.pl
would not run.
(Bug#18415)
The server crashed when the size of an
ARCHIVE table grew larger than 2GB.
(Bug#15787)
On 64-bit Windows systems, the Config Wizard failed to complete
the setup because 64-bit Windows does not resolve dynamic
linking of the 64-bit libmysql.dll to a
32-bit application like the Config Wizard.
(Bug#14649)
The server returned data from SHOW CREATE
TABLE statement or a
SELECT statement on an
INFORMATION_SCHEMA table using the binary
character set.
(Bug#10491)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.41.
Functionality added or changed:
Incompatible Change:
Prior to this release, when DATE
values were compared with
DATETIME values, the time portion
of the DATETIME value was
ignored, or the comparison could be performed as a string
compare. Now a DATE value is
coerced to the DATETIME type by
adding the time portion as 00:00:00. To mimic
the old behavior, use the CAST()
function as shown in this example: SELECT
.
(Bug#28929)date_col = CAST(NOW() AS DATE) FROM
table;
Incompatible Change:
INSERT DELAYED is now downgraded
to a normal INSERT if the
statement uses functions that access tables or triggers, or that
is called from a function or a trigger.
This was done to resolve the following interrelated issues:
The server could abort or deadlock for
INSERT DELAYED statements for
which another insert was performed implicitly (for example,
via a stored function that inserted a row).
A trigger using an INSERT
DELAYED caused the error INSERT DELAYED
can't be used with table ... because it is locked with LOCK
TABLES although the target table was not
actually locked.
INSERT DELAYED into a table
with a BEFORE INSERT or AFTER
INSERT trigger gave an incorrect
NEW pseudocolumn value and caused the
server to deadlock or abort.
MySQL Cluster: The server source tree now includes scripts to simplify building MySQL with SCI support. For more information about SCI interconnects and these build scripts, see Section 17.9.1, “Configuring MySQL Cluster to use SCI Sockets”. (Bug#25470)
Binaries for the Linux x86 statically linked
tar.gz Community package were linked
dynamically, not statically. Static linking has been re-enabled.
(Bug#29617)
INSERT DELAYED statements on
BLACKHOLE tables are now rejected, due to the
fact that the BLACKHOLE storage engine does
not support them.
(Bug#27998)
A new status variable, Com_call_procedure,
indicates the number of calls to stored procedures.
(Bug#27994)
Potential memory leaks in SHOW
PROFILE were eliminated.
(Bug#24795)
Bugs fixed:
Security Fix: A malformed password packet in the connection protocol could cause the server to crash. Thanks for Dormando for reporting this bug, and for providing details and a proof of concept. (Bug#28984, CVE-2007-3780)
Security Fix: Use of a view could allow a user to gain update privileges for tables in other databases. (Bug#27878, CVE-2007-3782)
Security Fix:
The requirement of the DROP
privilege for RENAME TABLE was
not enforced.
(Bug#27515, CVE-2007-2691)
Security Fix:
If a stored routine was declared using SQL SECURITY
INVOKER, a user who invoked the routine could gain
privileges.
(Bug#27337, CVE-2007-2692)
Security Fix:
CREATE TABLE LIKE did not require any
privileges on the source table. Now it requires the
SELECT privilege.
In addition, CREATE TABLE LIKE was not
isolated from alteration by other connections, which resulted in
various errors and incorrect binary log order when trying to
execute concurrently a CREATE TABLE LIKE
statement and either DDL statements on the source table or DML
or DDL statements on the target table.
(Bug#23667, Bug#25578, CVE-2007-3781)
Incompatible Change:
When mysqldump was run with the
--delete-master-logs option,
binary log files were deleted before it was known that the dump
had succeeded, not after. (The method for removing log files
used RESET MASTER prior to the
dump. This also reset the binary log sequence numbering to
.000001.) Now mysqldump
flushes the logs (which creates a new binary log number with the
next sequence number), performs the dump, and then uses
PURGE BINARY LOGS to remove the
log files older than the new one. This also preserves log
numbering because the new log with the next number is generated
and only the preceding logs are removed. However, this may
affect applications if they rely on the log numbering sequence
being reset.
(Bug#24733)
Incompatible Change:
The use of an ORDER BY or
DISTINCT clause with a query containing a
call to the GROUP_CONCAT()
function caused results from previous queries to be redisplayed
in the current result. The fix for this includes replacing a
BLOB value used internally for
sorting with a VARCHAR. This
means that for long results (more than 65,535 bytes), it is
possible for truncation to occur; if so, an appropriate warning
is issued.
(Bug#23856, Bug#28273)
MySQL Cluster: A corrupt schema file could cause a File already open error. (Bug#28770)
MySQL Cluster:
Setting InitialNoOpenFiles equal to
MaxNoOfOpenFiles caused an error. This was
due to the fact that the actual value of
MaxNoOfOpenFiles as used by the cluster was
offset by 1 from the value set in
config.ini.
(Bug#28749)
MySQL Cluster:
UPDATE IGNORE statements involving the
primary keys of multiple tables could result in data corruption.
(Bug#28719)
MySQL Cluster:
A race condition could result when nonmaster nodes (in addition
to the master node) tried to update active status due to a local
checkpoint (that is, between NODE_FAILREP and
COPY_GCIREQ events). Now only the master
updates the active status.
(Bug#28717)
MySQL Cluster: A fast global checkpoint under high load with high usage of the redo buffer caused data nodes to fail. (Bug#28653)
MySQL Cluster:
When an API node sent more than 1024 signals in a single batch,
NDB would process only the first
1024 of these, and then hang.
(Bug#28443)
MySQL Cluster:
A delay in obtaining AUTO_INCREMENT IDs could
lead to excess temporary errors.
(Bug#28410)
MySQL Cluster: The cluster waited 30 seconds instead of 30 milliseconds before reading table statistics. (Bug#28093)
MySQL Cluster:
INSERT IGNORE
wrongly ignored NULL values in unique
indexes.
(Bug#27980)
MySQL Cluster: The name of the month “March” was given incorrectly in the cluster error log. (Bug#27926)
MySQL Cluster:
It was not possible to add a unique index to an
NDB table while in single user
mode.
(Bug#27710)
MySQL Cluster:
Repeated insertion of data generated by
mysqldump into
NDB tables could eventually lead to
failure of the cluster.
(Bug#27437)
MySQL Cluster:
ndb_connectstring did not appear in the
output of SHOW VARIABLES.
(Bug#26675)
MySQL Cluster: A failure to release internal resources following an error could lead to problems with single user mode. (Bug#25818)
Replication:
The result of executing of a prepared statement created with
PREPARE s FROM "SELECT 1 LIMIT ?" was not
replicated correctly.
(Bug#28464)
Replication: Recreating a view that already exists on the master would cause a replicating slave to terminate replication with a 'different error message on slave and master' error. (Bug#28244)
Replication: Binary logging of prepared statements could produce syntactically incorrect queries in the binary log, replacing some parameters with variable names rather than variable values. This could lead to incorrect results on replication slaves. (Bug#26842, Bug#12826)
Replication:
Connections from one mysqld server to another
failed on Mac OS X, affecting replication and
FEDERATED tables.
(Bug#26664)
See also Bug#29083.
Replication: Aborting a statement on the master that applied to a nontransactional statement broke replication. The statement was written to the binary log but not completely executed on the master. Slaves receiving the statement executed it completely, resulting in loss of data synchrony. Now an error code is written to the error log so that the slaves stop without executing the aborted statement. (That is, replication stops, but synchrony to the point of the stop is preserved and you can investigate the problem.) (Bug#26551)
Replication: Restoration of the default database after stored routine or trigger execution on a slave could cause replication to stop if the database no longer existed. (Bug#25082)
Replication: When using transactions and replication, shutting down the master in the middle of a transaction would cause all slaves to stop replicating. (Bug#22725)
Replication:
Using CREATE TABLE LIKE ... would raise an
assertion when replicated to a slave.
(Bug#18950)
Cluster API:
For BLOB reads on operations with
lock mode LM_CommittedRead, the lock mode was
not upgraded to LM_Read before the state of
the BLOB had already been
calculated. The NDB API methods
affected by this problem included the following:
NdbOperation::readTuple()
NdbScanOperation::readTuples()
NdbIndexScanOperation::readTuples()
On the IBM i5 platform, the installation script in the
.savf binaries unconditionally executed the
mysql_install_db script. This problem was
fixed in a repackaged distribution numbered 5.0.45b.
(Bug#30084)
Long path names for internal temporary tables could cause stack overflows. (Bug#29015)
Using an INTEGER column from a
table to ROUND() a number
produced different results than using a constant with the same
value as the INTEGER column.
(Bug#28980)
If a program binds a given number of parameters to a prepared
statement handle and then somehow changes
stmt->param_count to a different number,
mysql_stmt_execute() could crash
the client or server.
(Bug#28934)
INSERT .. ON DUPLICATE KEY UPDATE could under
some circumstances silently update rows when it should not have.
(Bug#28904)
Queries that used UUID() were
incorrectly allowed into the query cache. (This should not
happen because UUID() is
nondeterministic.)
(Bug#28897)
Using a VIEW created with a nonexisting
DEFINER could lead to incorrect results under
some circumstances.
(Bug#28895)
On Windows, USE_TLS was not defined for
mysqlclient.lib.
(Bug#28860)
A subquery with ORDER BY and LIMIT
1 could cause a server crash.
(Bug#28811)
Using BETWEEN with nonindexed date
columns and short formats of the date string could return
incorrect results.
(Bug#28778)
Selecting GEOMETRY columns in a
UNION caused a server crash.
(Bug#28763)
When constructing the path to the original
.frm file, ALTER ..
RENAME was unnecessarily (and incorrectly) lowercasing
the entire path when not on a case-insensitive file system,
causing the statement to fail.
(Bug#28754)
Searches on indexed and nonindexed
ENUM columns could return
different results for empty strings.
(Bug#28729)
Executing EXPLAIN EXTENDED on a query using a
derived table over a grouping subselect could lead to a server
crash. This occurred only when materialization of the derived
tables required creation of an auxiliary temporary table, an
example being when a grouping operation was carried out with
usage of a temporary table.
(Bug#28728)
The result of evaluation for a view's CHECK
OPTION option over an updated record and records of
merged tables was arbitrary and dependant on the order of
records in the merged tables during the execution of the
SELECT statement.
(Bug#28716)
The “manager thread” of the LinuxThreads implementation was unintentionally started before mysqld had dropped privileges (to run as an unprivileged user). This caused signaling between threads in mysqld to fail when the privileges were finally dropped. (Bug#28690)
For debug builds, ALTER TABLE
could trigger an assertion failure due to occurrence of a
deadlock when committing changes.
(Bug#28652)
After an upgrade, the names of stored routines referenced by
views were no longer displayed by SHOW
CREATE VIEW.
(Bug#28605)
This regression was introduced by Bug#23491.
Killing from one connection a long-running EXPLAIN
QUERY started from another connection caused
mysqld to crash.
(Bug#28598)
Outer join queries with ON conditions over
constant outer tables did not return
NULL-complemented rows when conditions were
evaluated to FALSE.
(Bug#28571)
An update on a multiple-table view with the CHECK OPTION clause and a subquery in the WHERE condition could cause an assertion failure. (Bug#28561)
PURGE MASTER LOGS BEFORE
( caused a server
crash. Subqueries are forbidden in the subquery)BEFORE
clause now.
(Bug#28553)
mysqldump calculated the required memory for a hex-blob string incorrectly causing a buffer overrun. This in turn caused mysqldump to crash silently and produce incomplete output. (Bug#28522)
Passing a DECIMAL value as a
parameter of a statement prepared with
PREPARE resulted in an error.
(Bug#28509)
mysql_affected_rows() could
return an incorrect result for
INSERT ...
ON DUPLICATE KEY UPDATE if the
CLIENT_FOUND_ROWS flag was set.
(Bug#28505)
A query that grouped by the result of an expression returned a different result when the expression was assigned to a user variable. (Bug#28494)
Subselects returning LONG values in MySQL
versions later than 5.0.24a returned LONGLONG
prior to this. The previous behavior was restored.
(Bug#28492)
This regression was introduced by Bug#19714.
Forcing the use of an index on a
SELECT query when the index had
been disabled would raise an error without running the query.
The query now executes, with a warning generated noting that the
use of a disabled index has been ignored.
(Bug#28476)
The query SELECT '2007-01-01' + INTERVAL
caused
mysqld to fail.
(Bug#28450)column_name DAY FROM
table_name
A server crash could happen under rare conditions such that a
temporary table outgrew heap memory reserved for it and the
remaining disk space was not big enough to store the table as a
MyISAM table.
(Bug#28449)
mysql_upgrade failed if certain SQL modes were set. Now it sets the mode itself to avoid this problem. (Bug#28401)
A query with a NOT IN subquery predicate
could cause a crash when the left operand of the predicate
evaluated to NULL.
(Bug#28375)
The test case for mysqldump failed with
bin-log disabled.
(Bug#28372)
Attempting to LOAD_FILE from an empty floppy
drive under Windows, caused the server to hang. For example, if
you opened a connection to the server and then issued the
command SELECT LOAD_FILE('a:test');, with no
floppy in the drive, the server was inaccessible until the modal
pop-up dialog box was dismissed.
(Bug#28366)
A buffer overflow could occur when using
DECIMAL columns on Windows
operating systems.
(Bug#28361)
libmysql.dll could not be dynamically loaded
on Windows.
(Bug#28358)
Grouping queries with correlated subqueries in
WHERE conditions could produce incorrect
results.
(Bug#28337)
mysqltest used a too-large stack size on PPC/Debian Linux, causing thread-creation failure for tests that use many threads. (Bug#28333)
EXPLAIN for a query on an empty
table immediately after its creation could result in a server
crash.
(Bug#28272)
The IS_UPDATABLE column in the
INFORMATION_SCHEMA.VIEWS table was
not always set correctly.
(Bug#28266)
Comparing a DATETIME column value
with a user variable yielded incorrect results.
(Bug#28261)
For CAST() of a
NULL value with type
DECIMAL, the return value was
incorrectly initialized, producing a runtime error for binaries
built using Visual C++ 2005.
(Bug#28250)
Portability problems caused by use of isinf()
were corrected.
(Bug#28240)
When dumping procedures, mysqldump
--compact generated
output that restored the session variable
sql_mode without first
capturing it. When dumping routines, mysqldump
--compact neither
set nor retrieved the value of
sql_mode.
(Bug#28223)
Comparison of the string value of a date showed as unequal to
CURTIME(). Similar behavior was
exhibited for DATETIME values.
(Bug#28208)
For InnoDB, in some rare cases the optimizer
preferred a more expensive
ref access to a less
expensive range access.
(Bug#28189)
A performance degradation was observed for outer join queries to which a not-exists optimization was applied. (Bug#28188)
SELECT * INTO OUTFILE ... FROM
INFORMATION_SCHEMA.SCHEMATA failed with an
Access denied error, even for a user who
had the FILE privilege.
(Bug#28181)
The Bytes_received and
Bytes_sent status variables
could hold only 32-bit values (not 64-bit values) on some
platforms.
(Bug#28149)
Comparisons of DATE or
DATETIME values for the
IN() function could yield
incorrect results.
(Bug#28133)
Storing a large number into a
FLOAT or
DOUBLE column with a fixed length
could result in incorrect truncation of the number if the
column's length was greater than 31.
(Bug#28121)
The server could hang for INSERT IGNORE ... ON
DUPLICATE KEY UPDATE if an update failed.
(Bug#28000)
DECIMAL values beginning with
nine 9 digits could be incorrectly rounded.
(Bug#27984)
For INSERT
... ON DUPLICATE KEY UPDATE statements that affected
many rows, updates could be applied to the wrong rows.
(Bug#27954)
Early NULL-filtering optimization did not
work for eq_ref table access.
(Bug#27939)
The second execution of a prepared statement from a
UNION query with ORDER
BY RAND() caused the server to crash. This problem
could also occur when invoking a stored procedure containing
such a query.
(Bug#27937)
For attempts to open a nonexistent table, the server should
report ER_NO_SUCH_TABLE but
sometimes reported
ER_TABLE_NOT_LOCKED.
(Bug#27907)
A stored program that uses a variable name containing multibyte characters could fail to execute. (Bug#27876)
Nongrouped columns were allowed by * in
ONLY_FULL_GROUP_BY SQL mode.
(Bug#27874)
ON conditions from JOIN
expressions were ignored when checking the CHECK
OPTION clause while updating a multiple-table view
that included such a clause.
(Bug#27827)
Debug builds on Windows generated false alarms about uninitialized variables with some Visual Studio runtime libraries. (Bug#27811)
Certain queries that used uncorrelated scalar subqueries caused
EXPLAIN to crash.
(Bug#27807)
Changes to some system variables should invalidate statements in the query cache, but invalidation did not happen. (Bug#27792)
Performing a UNION on two views
that had ORDER BY clauses resulted in an
Unknown column error.
(Bug#27786)
mysql_install_db is supposed to detect existing system tables and create only those that do not exist. Instead, it was exiting with an error if tables already existed. (Bug#27783)
On some systems, udf_example.c returned an
incorrect result length. Also on some systems,
mysql-test-run.pl could not find the shared
object built from udf_example.c.
(Bug#27741)
mysqld did not check the length of option values and could crash with a buffer overflow for long values. (Bug#27715)
Comparisons using row constructors could fail for rows
containing NULL values.
(Bug#27704)
LOAD DATA did not use
CURRENT_TIMESTAMP as the default value for a
TIMESTAMP column for which no
value was provided.
(Bug#27670)
On Linux, the server could not create temporary tables if
lower_case_table_names was set
to 1 and the value of tmpdir was a directory
name containing any uppercase letters.
(Bug#27653)
For InnoDB tables, a multiple-row
INSERT of the form
INSERT INTO t (id...) VALUES (NULL...) ON DUPLICATE KEY
UPDATE id=VALUES(id), where id is
an AUTO_INCREMENT column, could cause
ERROR 1062 (23000): Duplicate entry... errors
or lost rows.
(Bug#27650)
HASH indexes on
VARCHAR columns with binary
collations did not ignore trailing spaces from strings before
comparisons. This could result in duplicate records being
successfully inserted into a MEMORY table
with unique key constraints. A consequence was that internal
MEMORY tables used for GROUP
BY calculation contained duplicate rows that resulted
in duplicate-key errors when converting those temporary tables
to MyISAM, and that error was incorrectly
reported as a table is full error.
(Bug#27643)
The XML output representing an empty result was an empty string
rather than an empty <resultset/>
element.
(Bug#27608)
An error occurred trying to connect to mysqld-debug.exe. (Bug#27597)
Comparison of a DATE with a
DATETIME did not treat the
DATE as having a time part of
00:00:00.
(Bug#27590)
See also Bug#32198.
Selecting MIN() on an indexed
column that contained only NULL values caused
NULL to be returned for other result columns.
(Bug#27573)
If a stored function or trigger was killed, it aborted but no error was thrown, allowing the calling statement to continue without noticing the problem. This could lead to incorrect results. (Bug#27563)
The fix for Bug#17212 provided correct sort order for misordered output of certain queries, but caused significant overall query performance degradation. (Results were correct (good), but returned much more slowly (bad).) The fix also affected performance of queries for which results were correct. The performance degradation has been addressed. (Bug#27531)
The CRC32() function returns an
unsigned integer, but the metadata was signed, which could cause
certain queries to return incorrect results. (For example,
queries that selected a CRC32()
value and used that value in the GROUP BY
clause.)
(Bug#27530)
An interaction between SHOW TABLE
STATUS and other concurrent statements that modify the
table could result in a divide-by-zero error and a server crash.
(Bug#27516)
When ALTER TABLE was used to add
a new DATE column with no
explicit default value, '0000-00-00' was used
as the default even if the SQL mode included the
NO_ZERO_DATE mode to prohibit
that value. A similar problem occurred for
DATETIME columns.
(Bug#27507)
A race condition between DROP
TABLE and SHOW TABLE
STATUS could cause the latter to display incorrect
information.
(Bug#27499)
Using a TEXT local variable in a
stored routine in an expression such as SET
produced
an incorrect result.
(Bug#27415)var =
SUBSTRING(var, 3)
Nested aggregate functions could be improperly evaluated. (Bug#27363)
A stored function invocation in the WHERE
clause was treated as a constant.
(Bug#27354)
Failure to allocate memory associated with
transaction_prealloc_size could
cause a server crash.
(Bug#27322)
mysqldump crashed if it got no data from
SHOW CREATE PROCEDURE (for
example, when trying to dump a routine defined by a different
user and for which the current user had no privileges). Now it
prints a comment to indicate the problem. It also returns an
error, or continues if the --force option is
given.
(Bug#27293)
The error message for error number 137 did
not report which database/table combination reported the
problem.
(Bug#27173)
mysqlbinlog produced different output with
the -R option than without it.
(Bug#27171)
A large filesort could result in a division by zero error and a server crash. (Bug#27119)
Times displayed by SHOW PROFILE
were incorrectly associated with the profile entry one later
than the corrrect one.
(Bug#27060)
Flow control optimization in stored routines could cause exception handlers to never return or execute incorrect logic. (Bug#26977)
SHOW PROFILE hung if executed
before enabling the @@profiling session
variable.
(Bug#26938)
mysqldump would not dump a view for which the
DEFINER no longer exists.
(Bug#26817)
Creating a temporary table with InnoDB when
using the one-file-per-table setting, and when the host file
system for temporary tables was tmpfs, would
cause an assertion within mysqld. This was
due to the use of O_DIRECT when opening the
temporary table file.
(Bug#26662)
mysql_upgrade did not detect failure of external commands that it runs. (Bug#26639)
Some test suite files were missing from some MySQL-test packages. (Bug#26609)
Statements within triggers ignored the value of the
low_priority_updates system
variable.
(Bug#26162)
See also Bug#29963.
Index hints (USE INDEX, IGNORE
INDEX, FORCE INDEX) cannot be used
with FULLTEXT indexes, but were not being
ignored.
(Bug#25951)
If CREATE TABLE t1 LIKE t2 failed due to a
full disk, an empty t2.frm file could be
created but not removed. This file then caused subsequent
attempts to create a table named t2 to fail.
This is easily corrected at the file system level by removing
the t2.frm file manually, but now the
server removes the file if the create operation does not
complete successfully.
(Bug#25761)
Running CHECK TABLE concurrently
with a SELECT,
INSERT or other statement on
Windows could corrupt a MyISAM table.
(Bug#25712)
On Windows, connection handlers did not properly decrement the server's thread count when exiting. (Bug#25621)
mysql_upgrade did not pass a password to mysqlcheck if one was given. (Bug#25452)
On Windows, mysql_upgrade was sensitive to lettercase of the names of some required components. (Bug#25405)
For storage engines that allow the current auto-increment value
to be set, using ALTER TABLE ... ENGINE to
convert a table from one such storage engine to another caused
loss of the current value. (For storage engines that do not
support setting the value, it cannot be retained anyway when
changing the storage engine.)
(Bug#25262)
Due to a race condition, executing
FLUSH
PRIVILEGES in one thread could cause brief table
unavailability in other threads.
(Bug#24988)
Several math functions produced incorrect results for large
unsigned values. ROUND() produced
incorrect results or a crash for a large number-of-decimals
argument.
(Bug#24912)
The result set of a query that used WITH
ROLLUP and DISTINCT could lack some
rollup rows (rows with NULL values for
grouping attributes) if the GROUP BY list
contained constant expressions.
(Bug#24856)
For queries that used ORDER BY with
InnoDB tables, if the optimizer chose an
index for accessing the table but found a covering index that
enabled the ORDER BY to be skipped, no
results were returned.
(Bug#24778)
Concurrent execution of
CREATE TABLE ...
SELECT and other statements involving the target table
suffered from various race conditions, some of which might have
led to deadlocks.
(Bug#24738)
On some Linux distributions where LinuxThreads and NPTL
glibc versions both are available, statically
built binaries can crash because the linker defaults to
LinuxThreads when linking statically, but calls to external
libraries (such as libnss) are resolved to
NPTL versions. This cannot be worked around in the code, so
instead if a crash occurs on such a binary/OS combination, print
an error message that provides advice about how to fix the
problem.
(Bug#24611)
An attempt to execute
CREATE TABLE ...
SELECT when a temporary table with the same name
already existed led to the insertion of data into the temporary
table and creation of an empty nontemporary table.
(Bug#24508)
The MERGE storage engine could return
incorrect results when several index values that compare
equality were present in an index (for example,
'gross' and 'gross ',
which are considered equal but have different lengths).
(Bug#24342)
Some upgrade problems are detected and better error messages suggesting that mysql_upgrade be run are produced. (Bug#24248)
Some views could not be created even when the user had the requisite privileges. (Bug#24040)
Using CAST() to convert
DATETIME values to numeric values
did not work.
(Bug#23656)
The AUTO_INCREMENT value would not be
correctly reported for InnoDB tables when
using SHOW CREATE TABLE statement
or mysqldump command.
(Bug#23313)
Implicit conversion of 9912101 to
DATE did not match
CAST(9912101 AS DATE).
(Bug#23093)
Conversion errors could occur when constructing the condition
for an IN predicate. The predicate was
treated as if the affected column contains
NULL, but if the IN
predicate is inside NOT, incorrect results
could be returned.
(Bug#22855)
SELECT COUNT(*) from a table containing a
DATETIME NOT NULL column could produce
spurious warnings with the
NO_ZERO_DATE SQL mode enabled.
(Bug#22824)
Using SET
GLOBAL to change the
lc_time_names system variable
had no effect on new connections.
(Bug#22648)
A multiple-table UPDATE could
return an incorrect rows-matched value if, during insertion of
rows into a temporary table, the table had to be converted from
a MEMORY table to a MyISAM
table.
(Bug#22364)
yaSSL crashed on pre-Pentium Intel CPUs. (Bug#21765)
Linux binaries were unable to dump core after executing a
setuid() call.
(Bug#21723)
A slave that used
--master-ssl-cipher
could not connect to the master.
(Bug#21611)
Quoted labels in stored routines were mishandled, rendering the routines unusable. (Bug#21513)
Stack overflow caused server crashes. (Bug#21476)
CURDATE() is less than
NOW(), either when comparing
CURDATE() directly
(CURDATE() < NOW() is true) or when
casting CURDATE() to
DATE (CAST(CURDATE() AS
DATE) < NOW() is true). However, storing
CURDATE() in a
DATE column and comparing
incorrectly yielded false. This is fixed by
comparing a col_name <
NOW()DATE column as
DATETIME for comparisons to a
DATETIME constant.
(Bug#21103)
CREATE TABLE IF NOT EXISTS ... SELECT caused
a server crash if the target table already existed and had a
BEFORE INSERT trigger.
(Bug#20903)
Deadlock occurred for attempts to execute CREATE TABLE
IF NOT EXISTS ... SELECT when
LOCK TABLES had been used to
acquire a read lock on the target table.
(Bug#20662, Bug#15522)
Changing a utf8 column in an
InnoDB table to a shorter length did not
shorten the data values.
(Bug#20095)
For dates with 4-digit year parts less than 200, an incorrect
implicit conversion to add a century was applied for date
arithmetic performed with
DATE_ADD(),
DATE_SUB(), +
INTERVAL, and - INTERVAL. (For
example, DATE_ADD('0050-01-01 00:00:00',
INTERVAL 0 SECOND) became '2050-01-01
00:00:00'.)
(Bug#18997)
Granting access privileges to an individual table where the database or table name contained an underscore would fail. (Bug#18660)
The -lmtmalloc library was removed from the
output of mysql_config on Solaris, as it
caused problems when building DBD::mysql (and
possibly other applications) on that platform that tried to use
dlopen() to access the client library.
(Bug#18322)
The check-cpu script failed to detect AMD64 Turion processors correctly. (Bug#17707)
Trying to shut down the server following a failed
LOAD DATA
INFILE caused mysqld to crash.
(Bug#17233)
The omission of leading zeros in dates could lead to erroneous results when these were compared with the output of certain date and time functions. (Bug#16377)
INSERT...ON DUPLICATE KEY UPDATE could cause
Error 1032: Can't find record in ... for
inserts into an InnoDB table unique index
using key column prefixes with an underlying
utf8 string column.
(Bug#13191)
Having the EXECUTE privilege for
a routine in a database should make it possible to
USE that database, but the server
returned an error instead. This has been corrected. As a result
of the change, SHOW TABLES for a
database in which you have only the
EXECUTE privilege returns an
empty set rather than an error.
(Bug#9504)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.44).
Bugs fixed:
Using the DATE() function in a
WHERE clause did not return any records after
encountering NULL. However, using
TRIM or CAST produced the
correct results.
(Bug#29898)
For a table with a DATE column
date_col such that selecting rows
with WHERE yielded
a nonempty result, adding date_col =
'date_val 00:00:00'GROUP BY
caused the result
to be empty.
(Bug#29729)date_col
Optimization of queries with DETERMINISTIC
stored functions in the WHERE clause was
ineffective: A sequential scan was always used.
(Bug#29338)
Creation of a legal stored procedure could fail if no default database had been selected. (Bug#29050)
If a stored procedure was created and invoked prior to selecting
a default database with USE, a
No database selected error occurred.
(Bug#28551)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.42).
Functionality added or changed:
MySQL Cluster: The server source tree now includes scripts to simplify building MySQL with SCI support. For more information about SCI interconnects and these build scripts, see Section 17.9.1, “Configuring MySQL Cluster to use SCI Sockets”. (Bug#25470)
Enterprise builds did not include the CSV
storage engine. CSV is now included in
Enterprise builds for all platforms except Windows, QNX, and
NetWare.
(Bug#28844)
INSERT DELAYED statements on
BLACKHOLE tables are now rejected, due to the
fact that the BLACKHOLE storage engine does
not support them.
(Bug#27998)
A new status variable, Com_call_procedure,
indicates the number of calls to stored procedures.
(Bug#27994)
Bugs fixed:
Security Fix: A malformed password packet in the connection protocol could cause the server to crash. Thanks for Dormando for reporting this bug, and for providing details and a proof of concept. (Bug#28984, CVE-2007-3780)
Security Fix:
CREATE TABLE LIKE did not require any
privileges on the source table. Now it requires the
SELECT privilege.
In addition, CREATE TABLE LIKE was not
isolated from alteration by other connections, which resulted in
various errors and incorrect binary log order when trying to
execute concurrently a CREATE TABLE LIKE
statement and either DDL statements on the source table or DML
or DDL statements on the target table.
(Bug#23667, Bug#25578, CVE-2007-3781)
Incompatible Change:
When mysqldump was run with the
--delete-master-logs option,
binary log files were deleted before it was known that the dump
had succeeded, not after. (The method for removing log files
used RESET MASTER prior to the
dump. This also reset the binary log sequence numbering to
.000001.) Now mysqldump
flushes the logs (which creates a new binary log number with the
next sequence number), performs the dump, and then uses
PURGE BINARY LOGS to remove the
log files older than the new one. This also preserves log
numbering because the new log with the next number is generated
and only the preceding logs are removed. However, this may
affect applications if they rely on the log numbering sequence
being reset.
(Bug#24733)
Incompatible Change:
The use of an ORDER BY or
DISTINCT clause with a query containing a
call to the GROUP_CONCAT()
function caused results from previous queries to be redisplayed
in the current result. The fix for this includes replacing a
BLOB value used internally for
sorting with a VARCHAR. This
means that for long results (more than 65,535 bytes), it is
possible for truncation to occur; if so, an appropriate warning
is issued.
(Bug#23856, Bug#28273)
MySQL Cluster: A corrupt schema file could cause a File already open error. (Bug#28770)
MySQL Cluster:
Setting InitialNoOpenFiles equal to
MaxNoOfOpenFiles caused an error. This was
due to the fact that the actual value of
MaxNoOfOpenFiles as used by the cluster was
offset by 1 from the value set in
config.ini.
(Bug#28749)
MySQL Cluster:
UPDATE IGNORE statements involving the
primary keys of multiple tables could result in data corruption.
(Bug#28719)
MySQL Cluster:
A race condition could result when nonmaster nodes (in addition
to the master node) tried to update active status due to a local
checkpoint (that is, between NODE_FAILREP and
COPY_GCIREQ events). Now only the master
updates the active status.
(Bug#28717)
MySQL Cluster: A fast global checkpoint under high load with high usage of the redo buffer caused data nodes to fail. (Bug#28653)
MySQL Cluster:
When an API node sent more than 1024 signals in a single batch,
NDB would process only the first
1024 of these, and then hang.
(Bug#28443)
MySQL Cluster:
A delay in obtaining AUTO_INCREMENT IDs could
lead to excess temporary errors.
(Bug#28410)
MySQL Cluster: A failure to release internal resources following an error could lead to problems with single user mode. (Bug#25818)
Replication:
The result of executing of a prepared statement created with
PREPARE s FROM "SELECT 1 LIMIT ?" was not
replicated correctly.
(Bug#28464)
Replication: Recreating a view that already exists on the master would cause a replicating slave to terminate replication with a 'different error message on slave and master' error. (Bug#28244)
Replication: Binary logging of prepared statements could produce syntactically incorrect queries in the binary log, replacing some parameters with variable names rather than variable values. This could lead to incorrect results on replication slaves. (Bug#26842, Bug#12826)
Replication:
Connections from one mysqld server to another
failed on Mac OS X, affecting replication and
FEDERATED tables.
(Bug#26664)
See also Bug#29083.
Replication: When using transactions and replication, shutting down the master in the middle of a transaction would cause all slaves to stop replicating. (Bug#22725)
Replication:
Using CREATE TABLE LIKE ... would raise an
assertion when replicated to a slave.
(Bug#18950)
On the IBM i5 platform, the installation script in the
.savf binaries unconditionally executed the
mysql_install_db script. This problem was
fixed in a repackaged distribution numbered 5.0.44b.
(Bug#30084)
Long path names for internal temporary tables could cause stack overflows. (Bug#29015)
Using an INTEGER column from a
table to ROUND() a number
produced different results than using a constant with the same
value as the INTEGER column.
(Bug#28980)
If a program binds a given number of parameters to a prepared
statement handle and then somehow changes
stmt->param_count to a different number,
mysql_stmt_execute() could crash
the client or server.
(Bug#28934)
INSERT .. ON DUPLICATE KEY UPDATE could under
some circumstances silently update rows when it should not have.
(Bug#28904)
Queries that used UUID() were
incorrectly allowed into the query cache. (This should not
happen because UUID() is
nondeterministic.)
(Bug#28897)
Using a VIEW created with a nonexisting
DEFINER could lead to incorrect results under
some circumstances.
(Bug#28895)
For InnoDB tables that use the
utf8 character set, incorrect results could
occur for DML statements such as
DELETE or
UPDATE that use an index on
character-based columns.
(Bug#28878)
See also Bug#29449, Bug#30485, Bug#31395.
This regression was introduced by Bug#13195.
On Windows, USE_TLS was not defined for
mysqlclient.lib.
(Bug#28860)
A subquery with ORDER BY and LIMIT
1 could cause a server crash.
(Bug#28811)
Using BETWEEN with nonindexed date
columns and short formats of the date string could return
incorrect results.
(Bug#28778)
Selecting GEOMETRY columns in a
UNION caused a server crash.
(Bug#28763)
When constructing the path to the original
.frm file, ALTER ..
RENAME was unnecessarily (and incorrectly) lowercasing
the entire path when not on a case-insensitive file system,
causing the statement to fail.
(Bug#28754)
Searches on indexed and nonindexed
ENUM columns could return
different results for empty strings.
(Bug#28729)
Executing EXPLAIN EXTENDED on a query using a
derived table over a grouping subselect could lead to a server
crash. This occurred only when materialization of the derived
tables required creation of an auxiliary temporary table, an
example being when a grouping operation was carried out with
usage of a temporary table.
(Bug#28728)
The result of evaluation for a view's CHECK
OPTION option over an updated record and records of
merged tables was arbitrary and dependant on the order of
records in the merged tables during the execution of the
SELECT statement.
(Bug#28716)
The “manager thread” of the LinuxThreads implementation was unintentionally started before mysqld had dropped privileges (to run as an unprivileged user). This caused signaling between threads in mysqld to fail when the privileges were finally dropped. (Bug#28690)
For debug builds, ALTER TABLE
could trigger an assertion failure due to occurrence of a
deadlock when committing changes.
(Bug#28652)
After an upgrade, the names of stored routines referenced by
views were no longer displayed by SHOW
CREATE VIEW.
(Bug#28605)
This regression was introduced by Bug#23491.
Killing from one connection a long-running EXPLAIN
QUERY started from another connection caused
mysqld to crash.
(Bug#28598)
Outer join queries with ON conditions over
constant outer tables did not return
NULL-complemented rows when conditions were
evaluated to FALSE.
(Bug#28571)
An update on a multiple-table view with the CHECK OPTION clause and a subquery in the WHERE condition could cause an assertion failure. (Bug#28561)
PURGE MASTER LOGS BEFORE
( caused a server
crash. Subqueries are forbidden in the subquery)BEFORE
clause now.
(Bug#28553)
mysqldump calculated the required memory for a hex-blob string incorrectly causing a buffer overrun. This in turn caused mysqldump to crash silently and produce incomplete output. (Bug#28522)
Passing a DECIMAL value as a
parameter of a statement prepared with
PREPARE resulted in an error.
(Bug#28509)
mysql_affected_rows() could
return an incorrect result for
INSERT ...
ON DUPLICATE KEY UPDATE if the
CLIENT_FOUND_ROWS flag was set.
(Bug#28505)
A query that grouped by the result of an expression returned a different result when the expression was assigned to a user variable. (Bug#28494)
Subselects returning LONG values in MySQL
versions later than 5.0.24a returned LONGLONG
prior to this. The previous behavior was restored.
(Bug#28492)
This regression was introduced by Bug#19714.
Forcing the use of an index on a
SELECT query when the index had
been disabled would raise an error without running the query.
The query now executes, with a warning generated noting that the
use of a disabled index has been ignored.
(Bug#28476)
The query SELECT '2007-01-01' + INTERVAL
caused
mysqld to fail.
(Bug#28450)column_name DAY FROM
table_name
A server crash could happen under rare conditions such that a
temporary table outgrew heap memory reserved for it and the
remaining disk space was not big enough to store the table as a
MyISAM table.
(Bug#28449)
mysql_upgrade failed if certain SQL modes were set. Now it sets the mode itself to avoid this problem. (Bug#28401)
The test case for mysqldump failed with
bin-log disabled.
(Bug#28372)
Attempting to LOAD_FILE from an empty floppy
drive under Windows, caused the server to hang. For example, if
you opened a connection to the server and then issued the
command SELECT LOAD_FILE('a:test');, with no
floppy in the drive, the server was inaccessible until the modal
pop-up dialog box was dismissed.
(Bug#28366)
A buffer overflow could occur when using
DECIMAL columns on Windows
operating systems.
(Bug#28361)
libmysql.dll could not be dynamically loaded
on Windows.
(Bug#28358)
Grouping queries with correlated subqueries in
WHERE conditions could produce incorrect
results.
(Bug#28337)
mysqltest used a too-large stack size on PPC/Debian Linux, causing thread-creation failure for tests that use many threads. (Bug#28333)
EXPLAIN for a query on an empty
table immediately after its creation could result in a server
crash.
(Bug#28272)
The IS_UPDATABLE column in the
INFORMATION_SCHEMA.VIEWS table was
not always set correctly.
(Bug#28266)
Comparing a DATETIME column value
with a user variable yielded incorrect results.
(Bug#28261)
For CAST() of a
NULL value with type
DECIMAL, the return value was
incorrectly initialized, producing a runtime error for binaries
built using Visual C++ 2005.
(Bug#28250)
Portability problems caused by use of isinf()
were corrected.
(Bug#28240)
When dumping procedures, mysqldump
--compact generated
output that restored the session variable
sql_mode without first
capturing it. When dumping routines, mysqldump
--compact neither
set nor retrieved the value of
sql_mode.
(Bug#28223)
Comparison of the string value of a date showed as unequal to
CURTIME(). Similar behavior was
exhibited for DATETIME values.
(Bug#28208)
The Bytes_received and
Bytes_sent status variables
could hold only 32-bit values (not 64-bit values) on some
platforms.
(Bug#28149)
Storing a large number into a
FLOAT or
DOUBLE column with a fixed length
could result in incorrect truncation of the number if the
column's length was greater than 31.
(Bug#28121)
DECIMAL values beginning with
nine 9 digits could be incorrectly rounded.
(Bug#27984)
The second execution of a prepared statement from a
UNION query with ORDER
BY RAND() caused the server to crash. This problem
could also occur when invoking a stored procedure containing
such a query.
(Bug#27937)
For attempts to open a nonexistent table, the server should
report ER_NO_SUCH_TABLE but
sometimes reported
ER_TABLE_NOT_LOCKED.
(Bug#27907)
A stored program that uses a variable name containing multibyte characters could fail to execute. (Bug#27876)
ON conditions from JOIN
expressions were ignored when checking the CHECK
OPTION clause while updating a multiple-table view
that included such a clause.
(Bug#27827)
On some systems, udf_example.c returned an
incorrect result length. Also on some systems,
mysql-test-run.pl could not find the shared
object built from udf_example.c.
(Bug#27741)
HASH indexes on
VARCHAR columns with binary
collations did not ignore trailing spaces from strings before
comparisons. This could result in duplicate records being
successfully inserted into a MEMORY table
with unique key constraints. A consequence was that internal
MEMORY tables used for GROUP
BY calculation contained duplicate rows that resulted
in duplicate-key errors when converting those temporary tables
to MyISAM, and that error was incorrectly
reported as a table is full error.
(Bug#27643)
An error occurred trying to connect to mysqld-debug.exe. (Bug#27597)
Selecting MIN() on an indexed
column that contained only NULL values caused
NULL to be returned for other result columns.
(Bug#27573)
If a stored function or trigger was killed, it aborted but no error was thrown, allowing the calling statement to continue without noticing the problem. This could lead to incorrect results. (Bug#27563)
When ALTER TABLE was used to add
a new DATE column with no
explicit default value, '0000-00-00' was used
as the default even if the SQL mode included the
NO_ZERO_DATE mode to prohibit
that value. A similar problem occurred for
DATETIME columns.
(Bug#27507)
Using a TEXT local variable in a
stored routine in an expression such as SET
produced
an incorrect result.
(Bug#27415)var =
SUBSTRING(var, 3)
The error message for error number 137 did
not report which database/table combination reported the
problem.
(Bug#27173)
A large filesort could result in a division by zero error and a server crash. (Bug#27119)
Some test suite files were missing from some MySQL-test packages. (Bug#26609)
Statements within triggers ignored the value of the
low_priority_updates system
variable.
(Bug#26162)
See also Bug#29963.
Running CHECK TABLE concurrently
with a SELECT,
INSERT or other statement on
Windows could corrupt a MyISAM table.
(Bug#25712)
On Windows, connection handlers did not properly decrement the server's thread count when exiting. (Bug#25621)
Due to a race condition, executing
FLUSH
PRIVILEGES in one thread could cause brief table
unavailability in other threads.
(Bug#24988)
When mysqld was run as a Windows service, shared memory objects were not created in the global namespace and could not be used by clients to connect. (Bug#24731)
On some Linux distributions where LinuxThreads and NPTL
glibc versions both are available, statically
built binaries can crash because the linker defaults to
LinuxThreads when linking statically, but calls to external
libraries (such as libnss) are resolved to
NPTL versions. This cannot be worked around in the code, so
instead if a crash occurs on such a binary/OS combination, print
an error message that provides advice about how to fix the
problem.
(Bug#24611)
Implicit conversion of 9912101 to
DATE did not match
CAST(9912101 AS DATE).
(Bug#23093)
Conversion errors could occur when constructing the condition
for an IN predicate. The predicate was
treated as if the affected column contains
NULL, but if the IN
predicate is inside NOT, incorrect results
could be returned.
(Bug#22855)
Linux binaries were unable to dump core after executing a
setuid() call.
(Bug#21723)
Stack overflow caused server crashes. (Bug#21476)
CURDATE() is less than
NOW(), either when comparing
CURDATE() directly
(CURDATE() < NOW() is true) or when
casting CURDATE() to
DATE (CAST(CURDATE() AS
DATE) < NOW() is true). However, storing
CURDATE() in a
DATE column and comparing
incorrectly yielded false. This is fixed by
comparing a col_name <
NOW()DATE column as
DATETIME for comparisons to a
DATETIME constant.
(Bug#21103)
For dates with 4-digit year parts less than 200, an incorrect
implicit conversion to add a century was applied for date
arithmetic performed with
DATE_ADD(),
DATE_SUB(), +
INTERVAL, and - INTERVAL. (For
example, DATE_ADD('0050-01-01 00:00:00',
INTERVAL 0 SECOND) became '2050-01-01
00:00:00'.)
(Bug#18997)
Granting access privileges to an individual table where the database or table name contained an underscore would fail. (Bug#18660)
The -lmtmalloc library was removed from the
output of mysql_config on Solaris, as it
caused problems when building DBD::mysql (and
possibly other applications) on that platform that tried to use
dlopen() to access the client library.
(Bug#18322)
The check-cpu script failed to detect AMD64 Turion processors correctly. (Bug#17707)
Trying to shut down the server following a failed
LOAD DATA
INFILE caused mysqld to crash.
(Bug#17233)
Using up-arrow for command-line recall in mysql could cause a segmentation fault. (Bug#10218)
The result for CAST() when
casting a value to UNSIGNED was limited to
the maximum signed BIGINT value
(9223372036854775808), rather than the maximum unsigned value
(18446744073709551615).
(Bug#8663)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.40).
Functionality added or changed:
Incompatible Change:
Prior to this release, when DATE
values were compared with
DATETIME values, the time portion
of the DATETIME value was
ignored, or the comparison could be performed as a string
compare. Now a DATE value is
coerced to the DATETIME type by
adding the time portion as 00:00:00. To mimic
the old behavior, use the CAST()
function as shown in this example: SELECT
.
(Bug#28929)date_col = CAST(NOW() AS DATE) FROM
table;
Incompatible Change:
INSERT DELAYED is now downgraded
to a normal INSERT if the
statement uses functions that access tables or triggers, or that
is called from a function or a trigger.
This was done to resolve the following interrelated issues:
The server could abort or deadlock for
INSERT DELAYED statements for
which another insert was performed implicitly (for example,
via a stored function that inserted a row).
A trigger using an INSERT
DELAYED caused the error INSERT DELAYED
can't be used with table ... because it is locked with LOCK
TABLES although the target table was not
actually locked.
INSERT DELAYED into a table
with a BEFORE INSERT or AFTER
INSERT trigger gave an incorrect
NEW pseudocolumn value and caused the
server to deadlock or abort.
mysqld_multi now understands the
--no-defaults,
--defaults-file, and
--defaults-extra-file
options. The --config-file
option is deprecated; if given, it is treated like
--defaults-extra-file.
(Bug#27390)
Bugs fixed:
Security Fix: Use of a view could allow a user to gain update privileges for tables in other databases. (Bug#27878, CVE-2007-3782)
Security Fix:
The requirement of the DROP
privilege for RENAME TABLE was
not enforced.
(Bug#27515, CVE-2007-2691)
Security Fix:
If a stored routine was declared using SQL SECURITY
INVOKER, a user who invoked the routine could gain
privileges.
(Bug#27337, CVE-2007-2692)
MySQL Cluster: The cluster waited 30 seconds instead of 30 milliseconds before reading table statistics. (Bug#28093)
MySQL Cluster:
INSERT IGNORE
wrongly ignored NULL values in unique
indexes.
(Bug#27980)
MySQL Cluster: The name of the month “March” was given incorrectly in the cluster error log. (Bug#27926)
MySQL Cluster:
It was not possible to add a unique index to an
NDB table while in single user
mode.
(Bug#27710)
MySQL Cluster:
Repeated insertion of data generated by
mysqldump into
NDB tables could eventually lead to
failure of the cluster.
(Bug#27437)
MySQL Cluster:
ndb_connectstring did not appear in the
output of SHOW VARIABLES.
(Bug#26675)
Replication: Aborting a statement on the master that applied to a nontransactional statement broke replication. The statement was written to the binary log but not completely executed on the master. Slaves receiving the statement executed it completely, resulting in loss of data synchrony. Now an error code is written to the error log so that the slaves stop without executing the aborted statement. (That is, replication stops, but synchrony to the point of the stop is preserved and you can investigate the problem.) (Bug#26551)
Replication: Restoration of the default database after stored routine or trigger execution on a slave could cause replication to stop if the database no longer existed. (Bug#25082)
Cluster API:
For BLOB reads on operations with
lock mode LM_CommittedRead, the lock mode was
not upgraded to LM_Read before the state of
the BLOB had already been
calculated. The NDB API methods
affected by this problem included the following:
NdbOperation::readTuple()
NdbScanOperation::readTuples()
NdbIndexScanOperation::readTuples()
On the IBM i5 platform, the installation script in the
.savf binaries unconditionally executed the
mysql_install_db script. This problem was
fixed in a repackaged distribution numbered 5.0.42b.
(Bug#30084)
A query with a NOT IN subquery predicate
could cause a crash when the left operand of the predicate
evaluated to NULL.
(Bug#28375)
For InnoDB, in some rare cases the optimizer
preferred a more expensive
ref access to a less
expensive range access.
(Bug#28189)
A performance degradation was observed for outer join queries to which a not-exists optimization was applied. (Bug#28188)
SELECT * INTO OUTFILE ... FROM
INFORMATION_SCHEMA.SCHEMATA failed with an
Access denied error, even for a user who
had the FILE privilege.
(Bug#28181)
Comparisons of DATE or
DATETIME values for the
IN() function could yield
incorrect results.
(Bug#28133)
The server could hang for INSERT IGNORE ... ON
DUPLICATE KEY UPDATE if an update failed.
(Bug#28000)
For INSERT
... ON DUPLICATE KEY UPDATE statements that affected
many rows, updates could be applied to the wrong rows.
(Bug#27954)
Early NULL-filtering optimization did not
work for eq_ref table access.
(Bug#27939)
Nongrouped columns were allowed by * in
ONLY_FULL_GROUP_BY SQL mode.
(Bug#27874)
Debug builds on Windows generated false alarms about uninitialized variables with some Visual Studio runtime libraries. (Bug#27811)
Certain queries that used uncorrelated scalar subqueries caused
EXPLAIN to crash.
(Bug#27807)
Changes to some system variables should invalidate statements in the query cache, but invalidation did not happen. (Bug#27792)
Performing a UNION on two views
that had ORDER BY clauses resulted in an
Unknown column error.
(Bug#27786)
mysql_install_db is supposed to detect existing system tables and create only those that do not exist. Instead, it was exiting with an error if tables already existed. (Bug#27783)
mysqld did not check the length of option values and could crash with a buffer overflow for long values. (Bug#27715)
Comparisons using row constructors could fail for rows
containing NULL values.
(Bug#27704)
LOAD DATA did not use
CURRENT_TIMESTAMP as the default value for a
TIMESTAMP column for which no
value was provided.
(Bug#27670)
On Linux, the server could not create temporary tables if
lower_case_table_names was set
to 1 and the value of tmpdir was a directory
name containing any uppercase letters.
(Bug#27653)
For InnoDB tables, a multiple-row
INSERT of the form
INSERT INTO t (id...) VALUES (NULL...) ON DUPLICATE KEY
UPDATE id=VALUES(id), where id is
an AUTO_INCREMENT column, could cause
ERROR 1062 (23000): Duplicate entry... errors
or lost rows.
(Bug#27650)
The XML output representing an empty result was an empty string
rather than an empty <resultset/>
element.
(Bug#27608)
Comparison of a DATE with a
DATETIME did not treat the
DATE as having a time part of
00:00:00.
(Bug#27590)
See also Bug#32198.
The fix for Bug#17212 provided correct sort order for misordered output of certain queries, but caused significant overall query performance degradation. (Results were correct (good), but returned much more slowly (bad).) The fix also affected performance of queries for which results were correct. The performance degradation has been addressed. (Bug#27531)
The CRC32() function returns an
unsigned integer, but the metadata was signed, which could cause
certain queries to return incorrect results. (For example,
queries that selected a CRC32()
value and used that value in the GROUP BY
clause.)
(Bug#27530)
An interaction between SHOW TABLE
STATUS and other concurrent statements that modify the
table could result in a divide-by-zero error and a server crash.
(Bug#27516)
A race condition between DROP
TABLE and SHOW TABLE
STATUS could cause the latter to display incorrect
information.
(Bug#27499)
Nested aggregate functions could be improperly evaluated. (Bug#27363)
A stored function invocation in the WHERE
clause was treated as a constant.
(Bug#27354)
Failure to allocate memory associated with
transaction_prealloc_size could
cause a server crash.
(Bug#27322)
mysqldump crashed if it got no data from
SHOW CREATE PROCEDURE (for
example, when trying to dump a routine defined by a different
user and for which the current user had no privileges). Now it
prints a comment to indicate the problem. It also returns an
error, or continues if the --force option is
given.
(Bug#27293)
mysqlbinlog produced different output with
the -R option than without it.
(Bug#27171)
Flow control optimization in stored routines could cause exception handlers to never return or execute incorrect logic. (Bug#26977)
mysqldump would not dump a view for which the
DEFINER no longer exists.
(Bug#26817)
Creating a temporary table with InnoDB when
using the one-file-per-table setting, and when the host file
system for temporary tables was tmpfs, would
cause an assertion within mysqld. This was
due to the use of O_DIRECT when opening the
temporary table file.
(Bug#26662)
mysql_upgrade did not detect failure of external commands that it runs. (Bug#26639)
Index hints (USE INDEX, IGNORE
INDEX, FORCE INDEX) cannot be used
with FULLTEXT indexes, but were not being
ignored.
(Bug#25951)
If CREATE TABLE t1 LIKE t2 failed due to a
full disk, an empty t2.frm file could be
created but not removed. This file then caused subsequent
attempts to create a table named t2 to fail.
This is easily corrected at the file system level by removing
the t2.frm file manually, but now the
server removes the file if the create operation does not
complete successfully.
(Bug#25761)
mysql_upgrade did not pass a password to mysqlcheck if one was given. (Bug#25452)
On Windows, mysql_upgrade was sensitive to lettercase of the names of some required components. (Bug#25405)
For storage engines that allow the current auto-increment value
to be set, using ALTER TABLE ... ENGINE to
convert a table from one such storage engine to another caused
loss of the current value. (For storage engines that do not
support setting the value, it cannot be retained anyway when
changing the storage engine.)
(Bug#25262)
Several math functions produced incorrect results for large
unsigned values. ROUND() produced
incorrect results or a crash for a large number-of-decimals
argument.
(Bug#24912)
The result set of a query that used WITH
ROLLUP and DISTINCT could lack some
rollup rows (rows with NULL values for
grouping attributes) if the GROUP BY list
contained constant expressions.
(Bug#24856)
For queries that used ORDER BY with
InnoDB tables, if the optimizer chose an
index for accessing the table but found a covering index that
enabled the ORDER BY to be skipped, no
results were returned.
(Bug#24778)
Concurrent execution of
CREATE TABLE ...
SELECT and other statements involving the target table
suffered from various race conditions, some of which might have
led to deadlocks.
(Bug#24738)
An attempt to execute
CREATE TABLE ...
SELECT when a temporary table with the same name
already existed led to the insertion of data into the temporary
table and creation of an empty nontemporary table.
(Bug#24508)
The MERGE storage engine could return
incorrect results when several index values that compare
equality were present in an index (for example,
'gross' and 'gross ',
which are considered equal but have different lengths).
(Bug#24342)
Some upgrade problems are detected and better error messages suggesting that mysql_upgrade be run are produced. (Bug#24248)
Some views could not be created even when the user had the requisite privileges. (Bug#24040)
Using CAST() to convert
DATETIME values to numeric values
did not work.
(Bug#23656)
The AUTO_INCREMENT value would not be
correctly reported for InnoDB tables when
using SHOW CREATE TABLE statement
or mysqldump command.
(Bug#23313)
SELECT COUNT(*) from a table containing a
DATETIME NOT NULL column could produce
spurious warnings with the
NO_ZERO_DATE SQL mode enabled.
(Bug#22824)
Using SET
GLOBAL to change the
lc_time_names system variable
had no effect on new connections.
(Bug#22648)
A multiple-table UPDATE could
return an incorrect rows-matched value if, during insertion of
rows into a temporary table, the table had to be converted from
a MEMORY table to a MyISAM
table.
(Bug#22364)
yaSSL crashed on pre-Pentium Intel CPUs. (Bug#21765)
A slave that used
--master-ssl-cipher
could not connect to the master.
(Bug#21611)
Quoted labels in stored routines were mishandled, rendering the routines unusable. (Bug#21513)
CREATE TABLE IF NOT EXISTS ... SELECT caused
a server crash if the target table already existed and had a
BEFORE INSERT trigger.
(Bug#20903)
Deadlock occurred for attempts to execute CREATE TABLE
IF NOT EXISTS ... SELECT when
LOCK TABLES had been used to
acquire a read lock on the target table.
(Bug#20662, Bug#15522)
Changing a utf8 column in an
InnoDB table to a shorter length did not
shorten the data values.
(Bug#20095)
The omission of leading zeros in dates could lead to erroneous results when these were compared with the output of certain date and time functions. (Bug#16377)
INSERT...ON DUPLICATE KEY UPDATE could cause
Error 1032: Can't find record in ... for
inserts into an InnoDB table unique index
using key column prefixes with an underlying
utf8 string column.
(Bug#13191)
Having the EXECUTE privilege for
a routine in a database should make it possible to
USE that database, but the server
returned an error instead. This has been corrected. As a result
of the change, SHOW TABLES for a
database in which you have only the
EXECUTE privilege returns an
empty set rather than an error.
(Bug#9504)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.37.
Functionality added or changed:
If a set function S with an outer
reference
cannot be aggregated in the outer query against which the outer
reference has been resolved, MySQL interprets S(outer_ref)
the same way that it would interpret S(outer_ref)
.
However, standard SQL requires throwing an error in this
situation. An error now is thrown for such queries if the
S(const)ANSI SQL mode is enabled.
(Bug#27348)
Prefix lengths for columns in SPATIAL indexes
are no longer displayed in SHOW CREATE
TABLE output. mysqldump uses that
statement, so if a table with SPATIAL indexes
containing prefixed columns is dumped and reloaded, the index is
created with no prefixes. (The full column width of each column
is indexed.)
(Bug#26794)
The output of mysql
--xml and
mysqldump
--xml now includes a valid XML
namespace.
(Bug#25946)
If you use SSL for a client connection, you can tell the client
not to authenticate the server certificate by specifying neither
--ssl-ca nor
--ssl-capath. The server still
verifies the client according to any applicable requirements
established via GRANT statements
for the client, and it still uses any
--ssl-ca/--ssl-capath
values that were passed to server at startup time.
(Bug#25309)
The server now includes a timestamp in error messages that are
logged as a result of unhandled signals (such as mysqld
got signal 11 messages).
(Bug#24878)
The syntax for index hints has been extended to enable explicit specification that the hint applies only to join processing. See Section 12.2.8.2, “Index Hint Syntax”.
This is a new fix for this issue, and replaces the fix made in MySQL 5.0.25 and reverted in 5.0.26. (Bug#21174)
Added the --secure-file-priv
option for mysqld, which limits the effect of
the LOAD_FILE() function and the
LOAD DATA and
SELECT ... INTO
OUTFILE statements to work only with files in a given
directory.
(Bug#18628)
Binary distributions for some platforms did not include shared
libraries; now shared libraries are shipped for all platforms
except AIX 5.2 64-bit. Exception: The
library for the libmysqld embedded server is
not shared except on Windows.
(Bug#16520, Bug#26767, Bug#13450)
Added the read-only hostname
system variable, which the server sets at startup to the server
host name.
The mysql_create_system_tables script was removed because mysql_install_db no longer uses it in MySQL 5.0.
To satisfy different user requirements, we provide several servers. mysqld is an optimized server that is a smaller, faster binary. Each package now also includes mysqld-debug, which is compiled with debugging support but is otherwise configured identically to the nondebug server.
Bugs fixed:
Incompatible Change:
INSERT DELAYED statements are not
supported for MERGE tables, but the
MERGE storage engine was not rejecting such
statements, resulting in table corruption. Applications
previously using INSERT DELAYED
into MERGE table will break when upgrading to
versions with this fix. To avoid the problem, remove
DELAYED from such statements.
(Bug#26464)
MySQL Cluster:
NDB tables having
MEDIUMINT AUTO_INCREMENT columns were not
restored correctly by ndb_restore, causing
spurious duplicate key errors. This issue did not affect
TINYINT,
INT, or
BIGINT columns with
AUTO_INCREMENT.
(Bug#27775)
MySQL Cluster:
NDB tables with indexes whose names
contained space characters were not restored correctly by
ndb_restore (the index names were truncated).
(Bug#27758)
MySQL Cluster:
Under certain rare circumstances performing a
DROP TABLE or
TRUNCATE on an
NDB table could cause a node
failure or forced cluster shutdown.
(Bug#27581)
MySQL Cluster: Memory usage of a mysqld process grew even while idle. (Bug#27560)
MySQL Cluster:
It was not possible to set
LockPagesInMainMemory equal to
0.
(Bug#27291)
MySQL Cluster: A race condition could sometimes occur if the node acting as master failed while node IDs were still being allocated during startup. (Bug#27286)
MySQL Cluster: When a data node was taking over as the master node, a race condition could sometimes occur as the node was assuming responsibility for handling of global checkpoints. (Bug#27283)
MySQL Cluster: Error messages displayed when running in single user mode were inconsistent. (Bug#27021)
MySQL Cluster: The failure of a data node while restarting could cause other data nodes to hang or crash. (Bug#27003)
MySQL Cluster:
On Solaris, the value of an NDB
table column declared as BIT(33) was always
displayed as 0.
(Bug#26986)
MySQL Cluster: mysqld processes would sometimes crash under high load. (Bug#26825)
MySQL Cluster:
The output from ndb_restore
--print_data was incorrect for a
backup made of a database containing tables with
TINYINT or
SMALLINT columns.
(Bug#26740)
MySQL Cluster: An inadvertent use of unaligned data caused ndb_restore to fail on some 64-bit platforms, including Sparc and Itanium-2. (Bug#26739)
MySQL Cluster:
An invalid pointer was returned following a
FSCLOSECONF signal when accessing the REDO
logs during a node restart or system restart.
(Bug#26515)
MySQL Cluster:
The failure of a data node when restarting it with
--initial could lead to failures of subsequent
data node restarts.
(Bug#26481)
MySQL Cluster: Takeover for local checkpointing due to multiple failures of master nodes was sometimes incorrectly handled. (Bug#26457)
MySQL Cluster:
The LockPagesInMainMemory parameter was not
read until after distributed communication had already started
between cluster nodes. When the value of this parameter was
1, this could sometimes result in data node
failure due to missed heartbeats.
(Bug#26454)
MySQL Cluster: Under some circumstances, following the restart of a management node, all data nodes would connect to it normally, but some of them subsequently failed to log any events to the management node. (Bug#26293)
MySQL Cluster:
In some cases, AFTER UPDATE and
AFTER DELETE triggers on
NDB tables that referenced subject
table did not see the results of operation which caused
invocation of the trigger, but rather saw the row as it was
prior to the update or delete operation.
This was most noticeable when an update operation used a
subquery to obtain the rows to be updated. An example would be
UPDATE tbl1 SET col2 = val1 WHERE tbl1.col1 IN (SELECT
col3 FROM tbl2 WHERE c4 = val2) where there was an
AFTER UPDATE trigger on table
tbl1. In such cases, the trigger would fail
to execute.
The problem occurred because the actual update or delete
operations were deferred to be able to perform them later as one
batch. The fix for this bug solves the problem by disabling this
optimization for a given update or delete if the table has an
AFTER trigger defined for this operation.
(Bug#26242)
MySQL Cluster: Condition pushdown did not work with prepared statements. (Bug#26225)
MySQL Cluster:
Joins on multiple tables containing
BLOB columns could cause data
nodes run out of memory, and to crash with the error
NdbObjectIdMap::expand unable to expand.
(Bug#26176)
MySQL Cluster:
After entering single user mode it was not possible to alter
non-NDB tables on any SQL nodes
other than the one having sole access to the cluster.
(Bug#25275)
MySQL Cluster:
The management client command
displayed
the message node_id STATUSNode when node_id:
not connectednode_id
was not the node ID of a data node.
The ALL STATUS command in the cluster
management client still displays status information for data
nodes only. This is by design. See
Section 17.7.2, “Commands in the MySQL Cluster Management Client”, for more
information.
MySQL Cluster:
The message Error 0 in readAutoIncrementValue(): no
Error was written to the error log whenever
SHOW TABLE STATUS was performed
on a Cluster table that did not have an
AUTO_INCREMENT column.
(Bug#21033)
MySQL Cluster:
Some values of MaxNoOfTables caused the error
Job buffer congestion to occur.
(Bug#19378)
Replication: Out-of-memory errors were not reported. Now they are written to the error log. (Bug#26844)
Replication: Improved out-of-memory detection when sending logs from a master server to slaves, and log a message when allocation fails. (Bug#26837)
Replication: A multiple-row delayed insert with an auto-increment column could cause duplicate entries to be created on the slave in a replication environment. (Bug#26116, Bug#25507)
Replication:
When RAND() was called multiple
times inside a stored procedure, the server did not write the
correct random seed values to the binary log, resulting in
incorrect replication.
(Bug#25543)
Replication:
GRANT statements were not
replicated if the server was started with the
--replicate-ignore-table or
--replicate-wild-ignore-table
option.
(Bug#25482)
Replication: Duplicating the usage of a user variable in a stored procedure or trigger would not be replicated correctly to the slave. (Bug#25167)
Replication:
DROP TRIGGER statements would not
be filtered on the slave when using the
replication-wild-do-table option.
(Bug#24478)
Replication:
For INSERT
... ON DUPLICATE KEY UPDATE statements where some
AUTO_INCREMENT values were generated
automatically for inserts and some rows were updated, one
auto-generated value was lost per updated row, leading to faster
exhaustion of the range of the AUTO_INCREMENT
column.
Because the original problem can affect replication (different values on master and slave), it is recommended that the master and its slaves be upgraded to the current version. (Bug#24432)
Replication:
Replication between master and slave would infinitely retry
binary log transmission where the
max_allowed_packet on the master was larger
than that on the slave if the size of the transfer was between
these two values.
(Bug#23775)
Replication:
Loading data using
LOAD DATA
INFILE may not replicate correctly (due to character
set incompatibilities) if the
character_set_database variable
is set before the data is loaded.
(Bug#15126)
Replication: User defined variables used within stored procedures and triggers are not replicated correctly when operating in statement-based replication mode. (Bug#14914, Bug#20141)
Cluster Replication: Some queries that updated multiple tables were not backed up correctly. (Bug#27748)
Cluster API:
Using NdbBlob::writeData() to write data in
the middle of an existing blob value (that is, updating the
value) could overwrite some data past the end of the data to be
changed.
(Bug#27018)
Some equi-joins containing a WHERE clause
that included a NOT IN subquery caused a
server crash.
(Bug#27870)
SELECT
DISTINCT could return incorrect results if the select
list contained duplicated columns.
(Bug#27659)
With NO_AUTO_VALUE_ON_ZERO SQL
mode enabled, LOAD DATA
operations could assign incorrect
AUTO_INCREMENT values.
(Bug#27586)
Incorrect results could be returned for some queries that
contained a select list expression with IN or
BETWEEN together with an
ORDER BY or GROUP BY on
the same expression using NOT IN or
NOT BETWEEN.
(Bug#27532)
Evaluation of an IN() predicate containing a
decimal-valued argument caused a server crash.
(Bug#27513, Bug#27362, CVE-2007-2583)
Profiling overhead was incurred even with profiling disabled. (Bug#27501)
In out-of-memory conditions, the server might crash or otherwise not report an error to the Windows event log. (Bug#27490)
Passing nested row expressions with different structures to an
IN predicate caused a server crash.
(Bug#27484)
The decimal.h header file was incorrectly
omitted from binary distributions.
(Bug#27456)
With innodb_file_per_table
enabled, attempting to rename an InnoDB table
to a nonexistent database caused the server to exit.
(Bug#27381)
A subquery could get incorrect values for references to outer query columns when it contained aggregate functions that were aggregated in outer context. (Bug#27321)
In a view, a column that was defined using a
GEOMETRY function was treated as having the
LONGBLOB data type rather than
the GEOMETRY type.
(Bug#27300)
Queries containing subqueries with
COUNT(*) aggregated in an outer
context returned incorrect results. This happened only if the
subquery did not contain any references to outer columns.
(Bug#27257)
SELECT ... INTO
OUTFILE with a long FIELDS ENCLOSED
BY value could crash the server.
(Bug#27231)
Use of an aggregate function from an outer context as an
argument to GROUP_CONCAT() caused
a server crash.
(Bug#27229)
String truncation upon insertion into an integer or year column did not generate a warning (or an error in strict mode). (Bug#27176, Bug#26359)
Storing NULL values in spatial fields caused
excessive memory allocation and crashes on some systems.
(Bug#27164)
Row equalities in WHERE clauses could cause
memory corruption.
(Bug#27154)
GROUP BY on a ucs2 column
caused a server crash when there was at least one empty string
in the column.
(Bug#27079)
Duplicate members in SET or
ENUM definitions were not
detected. Now they result in a warning; if strict SQL mode is
enabled, an error occurs instead.
(Bug#27069)
For INSERT
... ON DUPLICATE KEY UPDATE statements on tables
containing AUTO_INCREMENT columns,
LAST_INSERT_ID() was reset to 0
if no rows were successfully inserted or changed. “Not
changed” includes the case where a row was updated to its
current values, but in that case,
LAST_INSERT_ID() should not be
reset to 0. Now LAST_INSERT_ID()
is reset to 0 only if no rows were successfully inserted or
touched, whether or not touched rows were changed.
(Bug#27033)
See also Bug#27210, Bug#27006.
This regression was introduced by Bug#19978.
mysql_install_db could terminate with an error after failing to determine that a system table already existed. (Bug#27022)
AFTER UPDATE triggers were not activated by
the update part of
INSERT ...
ON DUPLICATE KEY UPDATE statements.
(Bug#27006)
See also Bug#27033, Bug#27210.
This regression was introduced by Bug#19978.
In a MEMORY table, using a
BTREE index to scan for updatable rows could
lead to an infinite loop.
(Bug#26996)
Invalid optimization of pushdown conditions for queries where an outer join was guaranteed to read only one row from the outer table led to results with too few rows. (Bug#26963)
Windows binaries contained no debug symbol file. Now
.map and .pdb files are
included in 32-bit builds for mysqld-nt.exe,
mysqld-debug.exe, and
mysqlmanager.exe.
(Bug#26893)
For MERGE tables defined on underlying tables
that contained a short VARCHAR
column (shorter than four characters), using
ALTER TABLE on at least one but
not all of the underlying tables caused the table definitions to
be considered different from that of the
MERGE table, even if the
ALTER TABLE did not change the
definition.
(Bug#26881)
For InnoDB tables having a clustered index
that began with a CHAR or
VARCHAR column, deleting a record
and then inserting another before the deleted record was purged
could result in table corruption.
(Bug#26835)
Use of a subquery containing GROUP BY and
WITH ROLLUP caused a server crash.
(Bug#26830)
Duplicates were not properly identified among (potentially) long
strings used as arguments for
GROUP_CONCAT(DISTINCT).
(Bug#26815)
ALTER VIEW requires the
CREATE VIEW and
DROP privileges for the view.
However, if the view was created by another user, the server
erroneously required the SUPER
privilege.
(Bug#26813)
Added support for --debugger=dbx for
mysql-test-run.pl and added support for
--debugger=devenv,
--debugger=DevEnv, and
--debugger=.
(Bug#26792)/path/to/devenv
A result set column formed by concatention of string literals
was incomplete when the column was produced by a subquery in the
FROM clause.
(Bug#26738)
SSL connections failed on Windows. (Bug#26678)
When using the result of
SEC_TO_TIME() for time value
greater than 24 hours in an ORDER BY clause,
either directly or through a column alias, the rows were sorted
incorrectly as strings.
(Bug#26672)
Use of a subquery containing a
UNION with an invalid
ORDER BY clause caused a server crash.
(Bug#26661)
The range optimizer could cause the server to run out of memory. (Bug#26625)
The range optimizer could consume a combinatorial amount of
memory for certain classes of WHERE clauses.
(Bug#26624)
In some error messages, inconsistent format specifiers were used for the translations in different languages. comp_err (the error message compiler) now checks for mismatches. (Bug#26571)
Views that used a scalar correlated subquery returned incorrect results. (Bug#26560)
UNHEX() IS NULL comparisons failed when
UNHEX() returned
NULL.
(Bug#26537)
On 64-bit Windows, large timestamp values could be handled incorrectly. (Bug#26536)
mysqldump could crash or exhibit incorrect
behavior when some options were given very long values, such as
--fields-terminated-by=". The code has been cleaned up to
remove a number of fixed-sized buffers and to be more careful
about error conditions in memory allocation.
(Bug#26346)some very long
string"
If the server was started with
--skip-grant-tables, Selecting
from INFORMATION_SCHEMA tables causes a
server crash.
(Bug#26285)
For some values of the position argument, the
INSERT() function could insert a
NUL byte into the result.
(Bug#26281)
For an INSERT statement that
should fail due to a column with no default value not being
assigned a value, the statement succeeded with no error if the
column was assigned a value in an ON DUPLICATE KEY
UPDATE clause, even if that clause was not used.
(Bug#26261)
INSERT DELAYED statements
inserted incorrect values into
BIT columns.
(Bug#26238)
The temporary file-creation code was cleaned up on Windows to improve server stability. (Bug#26233)
For MyISAM tables,
COUNT(*) could return an
incorrect value if the WHERE clause compared
an indexed TEXT column to the
empty string (''). This happened if the
column contained empty strings and also strings starting with
control characters such as tab or newline.
(Bug#26231)
For INSERT INTO
... SELECT where index searches used column prefixes,
insert errors could occur when key value type conversion was
done.
(Bug#26207)
For DELETE FROM (with no
tbl_name ORDER BY
col_nameWHERE or LIMIT clause),
the server did not check whether
col_name was a valid column in the
table.
(Bug#26186)
REPAIR TABLE ... USE_FRM with an
ARCHIVE table deleted all records from the
table.
(Bug#26138)
BENCHMARK() did not work
correctly for expressions that produced a
DECIMAL result.
(Bug#26093)
LOAD DATA
INFILE sent an okay to the client before writing the
binary log and committing the changes to the table had finished,
thus violating ACID requirements.
(Bug#26050)
X() IS NULL and Y() IS
NULL comparisons failed when
X() and
Y() returned
NULL.
(Bug#26038)
mysqldump crashed for
MERGE tables if the
--complete-insert
(-c) option was given.
(Bug#25993)
Indexes on TEXT columns were
ignored when ref accesses
were evaluated.
(Bug#25971)
If a thread previously serviced a connection that was killed, excessive memory and CPU use by the thread occurred if it later serviced a connection that had to wait for a table lock. (Bug#25966)
Setting a column to NOT NULL with an
ON DELETE SET NULL clause foreign key crashes
the server.
(Bug#25927)
VIEW restrictions were applied to
SELECT statements after a
CREATE VIEW statement failed, as
though the CREATE had succeeded.
(Bug#25897)
Several deficiencies in resolution of column names for
INSERT ...
SELECT statements were corrected.
(Bug#25831)
Inserting utf8 data into a
TEXT column that used a
single-byte character set could result in spurious warnings
about truncated data.
(Bug#25815)
On Windows, debug builds of mysqld could fail with heap assertions. (Bug#25765)
In certain situations, MATCH ... AGAINST
returned false hits for NULL values produced
by LEFT JOIN when no full-text index was
available.
(Bug#25729)
In certain cases it could happen that deleting a row corrupted
an RTREE index. This affected indexes on
spatial columns.
(Bug#25673)
OPTIMIZE TABLE might fail on
Windows when it attempts to rename a temporary file to the
original name if the original file had been opened, resulting in
loss of the .MYD file.
(Bug#25521)
For SHOW ENGINE
INNODB STATUS, the LATEST DEADLOCK
INFORMATION was not always cleared properly.
(Bug#25494)
mysql_stmt_fetch() did an
invalid memory deallocation when used with the embedded server.
(Bug#25492)
Expressions involving SUM(), when
used in an ORDER BY clause, could lead to
out-of-order results.
(Bug#25376)
Use of a GROUP BY clause that referred to a
stored function result together with WITH
ROLLUP caused incorrect results.
(Bug#25373)
A stored procedure that made use of cursors failed when the procedure was invoked from a stored function. (Bug#25345)
Difficult repair or optimization operations could cause an assertion failure, resulting in a server crash. (Bug#25289)
On Windows, the server exhibited a file-handle leak after reaching the limit on the number of open file descriptors. (Bug#25222)
The REPEAT() function did not
allow a column name as the count
parameter.
(Bug#25197)
A reference to a nonexistent column in the ORDER
BY clause of an UPDATE ... ORDER BY
statement could cause a server crash.
(Bug#25126)
A view on a join is insertable for
INSERT statements that store
values into only one table of the join. However, inserts were
being rejected if the inserted-into table was used in a
self-join because MySQL incorrectly was considering the insert
to modify multiple tables of the view.
(Bug#25122)
MySQL would not compile when configured using
--without-query-cache.
(Bug#25075)
Duplicate entries were not assessed correctly in a
MEMORY table with a BTREE
primary key on a utf8
ENUM column.
(Bug#24985)
Selecting the result of AVG()
within a UNION could produce
incorrect values.
(Bug#24791)
MBROverlaps() returned incorrect values in
some cases.
(Bug#24563)
Increasing the width of a DECIMAL
column could cause column values to be changed.
(Bug#24558)
IF(expr,
unsigned_expr,
unsigned_expr) was evaluated to a
signed result, not unsigned. This has been corrected. The fix
also affects constructs of the form IS [NOT]
{TRUE|FALSE}, which were transformed internally into
IF() expressions that evaluated
to a signed result.
For existing views that were defined using IS [NOT]
{TRUE|FALSE} constructs, there is a related
implication. The definitions of such views were stored using the
IF() expression, not the original
construct. This is manifest in that SHOW
CREATE VIEW shows the transformed
IF() expression, not the original
one. Existing views will evaluate correctly after the fix, but
if you want SHOW CREATE VIEW to
display the original construct, you must drop the view and
re-create it using its original definition. New views will
retain the construct in their definition.
(Bug#24532)
A problem in handling of aggregate functions in subqueries caused predicates containing aggregate functions to be ignored during query execution. (Bug#24484)
The test for the
MYSQL_OPT_SSL_VERIFY_SERVER_CERT option for
mysql_options() was performed
incorrectly. Also changed as a result of this bug fix: The
arg option for the
mysql_options() C API function
was changed from char * to void
*.
(Bug#24121)
A user-defined variable could be assigned an incorrect value if a temporary table was employed in obtaining the result of the query used to determine its value. (Bug#24010)
Queries that used a temporary table for the outer query when evaluating a correlated subquery could return incorrect results. (Bug#23800)
On Windows, debug builds of mysqlbinlog could fail with a memory error. (Bug#23736)
When using certain server SQL modes, the
mysql.proc table was not created by
mysql_install_db.
(Bug#23669)
The values displayed for the
Innodb_row_lock_time,
Innodb_row_lock_time_avg, and
Innodb_row_lock_time_max
status variables were incorrect.
(Bug#23666)
DOUBLE values such as
20070202191048.000000 were being treated as
illegal arguments by WEEK().
(Bug#23616)
The server could crash if two or more threads initiated query cache resize operation at moments very close in time. (Bug#23527)
SHOW CREATE VIEW qualified
references to stored functions in the view definition with the
function's database name, even when the database was the default
database. This affected mysqldump (which uses
SHOW CREATE VIEW to dump views)
because the resulting dump file could not be used to reload the
database into a different database. SHOW
CREATE VIEW now suppresses the database name for
references to functions in the default database.
(Bug#23491)
An INTO OUTFILE clause is allowed only for
the final SELECT of a
UNION, but this restriction was
not being enforced correctly.
(Bug#23345)
NOW() returned the wrong value in
statements executed at server startup with the
--init-file option.
(Bug#23240)
With the NO_AUTO_VALUE_ON_ZERO
SQL mode enabled,
LAST_INSERT_ID() could return 0
after
INSERT ...
ON DUPLICATE KEY UPDATE. Additionally, the next rows
inserted (by the same INSERT, or
the following INSERT with or
without ON DUPLICATE KEY UPDATE), would
insert 0 for the auto-generated value if the value for the
AUTO_INCREMENT column was
NULL or missing.
(Bug#23233)
SOUNDEX() returned an invalid
string for international characters in multi-byte character
sets.
(Bug#22638)
When nesting stored procedures within a trigger on a table, a
false dependency error was thrown when one of the nested
procedures contained a DROP TABLE
statement.
(Bug#22580)
Instance Manager did not remove the angel PID file on a clean shutdown. (Bug#22511)
EXPLAIN EXTENDED did not show
WHERE conditions that were optimized away.
(Bug#22331)
COUNT(
sometimes generated a spurious truncation warning.
(Bug#21976)decimal_expr)
IN ((,
subquery))IN (((,
and so forth, are equivalent to subquery)))IN
(, which is always
interpreted as a table subquery (so that it is allowed to return
more than one row). MySQL was treating the
“over-parenthesized” subquery as a single-row
subquery and rejecting it if it returned more than one row. This
bug primarily affected automatically generated code (such as
queries generated by Hibernate), because humans rarely write the
over-parenthesized forms.
(Bug#21904)subquery)
An INSERT trigger invoking a
stored routine that inserted into a table other than the one on
which the trigger was defined would fail with a Table
'...' doesn't exist referring to the second table
when attempting to delete records from the first table.
(Bug#21825)
InnoDB: The first read statement, if served
from the query cache, was not consistent with the
READ COMMITTED isolation
level.
(Bug#21409)
CURDATE() is less than
NOW(), either when comparing
CURDATE() directly
(CURDATE() < NOW() is true) or when
casting CURDATE() to
DATE (CAST(CURDATE() AS
DATE) < NOW() is true). However, storing
CURDATE() in a
DATE column and comparing
incorrectly yielded false. This is fixed by
comparing a col_name <
NOW()DATE column as
DATETIME for comparisons to a
DATETIME constant.
(Bug#21103)
When a stored routine attempted to execute a statement accessing a nonexistent table, the error was not caught by the routine's exception handler. (Bug#20713, Bug#8407)
For a stored procedure containing a
SELECT statement that used a
complicated join with an ON expression, the
expression could be ignored during re-execution of the
procedure, yielding an incorrect result.
(Bug#20492)
The conditions checked by the optimizer to allow use of indexes
in IN predicate calculations were
unnecessarily tight and were relaxed.
(Bug#20420)
When a TIME_FORMAT() expression
was used as a column in a GROUP BY clause,
the expression result was truncated.
(Bug#20293)
The creation of MySQL system tables was not checked for by mysql-test-run.pl. (Bug#20166)
For index reads, the BLACKHOLE engine did not
return end-of-file (which it must because
BLACKHOLE tables contain no rows), causing
some queries to crash.
(Bug#19717)
In some cases, the optimizer preferred a range or full index scan access method over lookup access methods when the latter were much cheaper. (Bug#19372)
For expr
IN(value_list)BIGINT UNSIGNED values
were used for expr or in the value
list.
(Bug#19342)
When attempting to call a stored procedure creating a table from
a trigger on a table tbl in a database
db, the trigger failed with ERROR
1146 (42S02): Table 'db.tbl' doesn't exist. However,
the actual reason that such a trigger fails is due to the fact
that CREATE TABLE causes an
implicit COMMIT, and so a trigger
cannot invoke a stored routine containing this statement. A
trigger which does so now fails with ERROR 1422
(HY000): Explicit or implicit commit is not allowed in stored
function or trigger, which makes clear the reason
for the trigger's failure.
(Bug#18914)
The update columns for INSERT ... SELECT ... ON
DUPLICATE KEY UPDATE could be assigned incorrect
values if a temporary table was used to evaluate the
SELECT.
(Bug#16630)
Conversion of DATETIME values in
numeric contexts sometimes did not produce a double
(YYYYMMDDHHMMSS.uuuuuu) value.
(Bug#16546)
For SUBSTRING() evaluation using
a temporary table, when
SUBSTRING() was used on a
LONGTEXT column, the max_length metadata
value of the result was incorrectly calculated and set to 0.
Consequently, an empty string was returned instead of the
correct result.
(Bug#15757)
Local variables in stored routines or triggers, when declared as
the BIT type, were interpreted as
strings.
(Bug#12976)
CONNECTION is no longer treated as a reserved
word.
(Bug#12204)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.38).
Functionality added or changed:
MySQL Cluster: The behavior of the ndb_restore utility has been changed as follows:
It is now possible to restore selected databases or tables using ndb_restore.
Several options have been added for use with
ndb_restore
--print_data to facilitate the creation
of structured data dump files. These options can be used
to make dumps made using ndb_restore
more like those produced by mysqldump.
For details of these changes, see Section 17.6.15, “ndb_restore — Restore a MySQL Cluster Backup”. (Bug#26899, Bug#26900)
If a set function S with an outer
reference
cannot be aggregated in the outer query against which the outer
reference has been resolved, MySQL interprets S(outer_ref)
the same way that it would interpret S(outer_ref)
.
However, standard SQL requires throwing an error in this
situation. An error now is thrown for such queries if the
S(const)ANSI SQL mode is enabled.
(Bug#27348)
Added the
--service-startup-timeout
option for mysql.server to specify how long
to wait for the server to start. If the server does not start
within the timeout period, mysql.server exits
with an error.
(Bug#26952)
Prefix lengths for columns in SPATIAL indexes
are no longer displayed in SHOW CREATE
TABLE output. mysqldump uses that
statement, so if a table with SPATIAL indexes
containing prefixed columns is dumped and reloaded, the index is
created with no prefixes. (The full column width of each column
is indexed.)
(Bug#26794)
The output of mysql
--xml and
mysqldump
--xml now includes a valid XML
namespace.
(Bug#25946)
If you use SSL for a client connection, you can tell the client
not to authenticate the server certificate by specifying neither
--ssl-ca nor
--ssl-capath. The server still
verifies the client according to any applicable requirements
established via GRANT statements
for the client, and it still uses any
--ssl-ca/--ssl-capath
values that were passed to server at startup time.
(Bug#25309)
The syntax for index hints has been extended to enable explicit specification that the hint applies only to join processing. See Section 12.2.8.2, “Index Hint Syntax”.
This is a new fix for this issue, and replaces the fix made in MySQL 5.0.25 and reverted in 5.0.26. (Bug#21174)
The mysql_create_system_tables script was removed because mysql_install_db no longer uses it in MySQL 5.0.
Bugs fixed:
Important Note: The parser accepted invalid code in SQL condition handlers, leading to server crashes or unexpected execution behavior in stored programs. Specifically, the parser allowed a condition handler to refer to labels for blocks that enclose the handler declaration. This was incorrect because block label scope does not include the code for handlers declared within the labeled block.
The parser now rejects this invalid construct, but if you perform a binary upgrade (without dumping and reloading your databases), existing handlers that contain the construct are still invalid and should be rewritten even if they appear to function as you expect.
To find affected handlers, use mysqldump to dump all stored procedures and functions, triggers, and events. Then attempt to reload them into an upgraded server. Handlers that contain illegal label references will be rejected.
For more information about condition handlers and writing them
to avoid invalid jumps, see Section 12.8.4.2, “DECLARE for Handlers”.
(Bug#26503)
MySQL Cluster:
NDB tables having
MEDIUMINT AUTO_INCREMENT columns were not
restored correctly by ndb_restore, causing
spurious duplicate key errors. This issue did not affect
TINYINT,
INT, or
BIGINT columns with
AUTO_INCREMENT.
(Bug#27775)
MySQL Cluster:
NDB tables with indexes whose names
contained space characters were not restored correctly by
ndb_restore (the index names were truncated).
(Bug#27758)
MySQL Cluster:
Under certain rare circumstances performing a
DROP TABLE or
TRUNCATE on an
NDB table could cause a node
failure or forced cluster shutdown.
(Bug#27581)
MySQL Cluster: Memory usage of a mysqld process grew even while idle. (Bug#27560)
MySQL Cluster:
It was not possible to set
LockPagesInMainMemory equal to
0.
(Bug#27291)
MySQL Cluster: A race condition could sometimes occur if the node acting as master failed while node IDs were still being allocated during startup. (Bug#27286)
MySQL Cluster: When a data node was taking over as the master node, a race condition could sometimes occur as the node was assuming responsibility for handling of global checkpoints. (Bug#27283)
MySQL Cluster: Error messages displayed when running in single user mode were inconsistent. (Bug#27021)
MySQL Cluster: The failure of a data node while restarting could cause other data nodes to hang or crash. (Bug#27003)
MySQL Cluster:
On Solaris, the value of an NDB
table column declared as BIT(33) was always
displayed as 0.
(Bug#26986)
MySQL Cluster: mysqld processes would sometimes crash under high load. (Bug#26825)
MySQL Cluster:
The output from ndb_restore
--print_data was incorrect for a
backup made of a database containing tables with
TINYINT or
SMALLINT columns.
(Bug#26740)
MySQL Cluster:
In some cases, AFTER UPDATE and
AFTER DELETE triggers on
NDB tables that referenced subject
table did not see the results of operation which caused
invocation of the trigger, but rather saw the row as it was
prior to the update or delete operation.
This was most noticeable when an update operation used a
subquery to obtain the rows to be updated. An example would be
UPDATE tbl1 SET col2 = val1 WHERE tbl1.col1 IN (SELECT
col3 FROM tbl2 WHERE c4 = val2) where there was an
AFTER UPDATE trigger on table
tbl1. In such cases, the trigger would fail
to execute.
The problem occurred because the actual update or delete
operations were deferred to be able to perform them later as one
batch. The fix for this bug solves the problem by disabling this
optimization for a given update or delete if the table has an
AFTER trigger defined for this operation.
(Bug#26242)
MySQL Cluster: Condition pushdown did not work with prepared statements. (Bug#26225)
MySQL Cluster:
Joins on multiple tables containing
BLOB columns could cause data
nodes run out of memory, and to crash with the error
NdbObjectIdMap::expand unable to expand.
(Bug#26176)
MySQL Cluster:
After entering single user mode it was not possible to alter
non-NDB tables on any SQL nodes
other than the one having sole access to the cluster.
(Bug#25275)
MySQL Cluster: When a cluster data node suffered a “hard” failure (such as a power failure or loss of a network connection) TCP sockets to the missing node were maintained indefinitely. Now socket-based transporters check for a response and terminate the socket if there is no activity on the socket after 2 hours. (Bug#24793)
MySQL Cluster:
The management client command
displayed
the message node_id STATUSNode when node_id:
not connectednode_id
was not the node ID of a data node.
The ALL STATUS command in the cluster
management client still displays status information for data
nodes only. This is by design. See
Section 17.7.2, “Commands in the MySQL Cluster Management Client”, for more
information.
MySQL Cluster:
Some values of MaxNoOfTables caused the error
Job buffer congestion to occur.
(Bug#19378)
MySQL Cluster:
When trying to create tables on an SQL node not connected to the
cluster, a misleading error message Table
'tbl_name' already exists
was generated. The error now generated is Could not
connect to storage engine.
(Bug#11217, Bug#18676)
Replication: Out-of-memory errors were not reported. Now they are written to the error log. (Bug#26844)
Replication: Improved out-of-memory detection when sending logs from a master server to slaves, and log a message when allocation fails. (Bug#26837)
Replication:
When RAND() was called multiple
times inside a stored procedure, the server did not write the
correct random seed values to the binary log, resulting in
incorrect replication.
(Bug#25543)
Replication:
GRANT statements were not
replicated if the server was started with the
--replicate-ignore-table or
--replicate-wild-ignore-table
option.
(Bug#25482)
Replication:
Replication between master and slave would infinitely retry
binary log transmission where the
max_allowed_packet on the master was larger
than that on the slave if the size of the transfer was between
these two values.
(Bug#23775)
Cluster Replication: Some queries that updated multiple tables were not backed up correctly. (Bug#27748)
Cluster API:
Using NdbBlob::writeData() to write data in
the middle of an existing blob value (that is, updating the
value) could overwrite some data past the end of the data to be
changed.
(Bug#27018)
Cluster API:
After defining a delete operation (using
NdbOperation::deleteTuple()) on a nonexistent
primary key of a table having a
BLOB or
TEXT column, invoking
NdbTransaction::execute() caused the calling
application to enter an endless loop rather than raising an
error.
This issue also affected ndb_restore; when
restoring tables containing BLOB
or TEXT columns, this could cause
it to consume all available memory and then crash.
(Bug#24028)
Some equi-joins containing a WHERE clause
that included a NOT IN subquery caused a
server crash.
(Bug#27870)
SELECT
DISTINCT could return incorrect results if the select
list contained duplicated columns.
(Bug#27659)
With NO_AUTO_VALUE_ON_ZERO SQL
mode enabled, LOAD DATA
operations could assign incorrect
AUTO_INCREMENT values.
(Bug#27586)
Incorrect results could be returned for some queries that
contained a select list expression with IN or
BETWEEN together with an
ORDER BY or GROUP BY on
the same expression using NOT IN or
NOT BETWEEN.
(Bug#27532)
Evaluation of an IN() predicate containing a
decimal-valued argument caused a server crash.
(Bug#27513, Bug#27362, CVE-2007-2583)
In out-of-memory conditions, the server might crash or otherwise not report an error to the Windows event log. (Bug#27490)
Passing nested row expressions with different structures to an
IN predicate caused a server crash.
(Bug#27484)
The decimal.h header file was incorrectly
omitted from binary distributions.
(Bug#27456)
With innodb_file_per_table
enabled, attempting to rename an InnoDB table
to a nonexistent database caused the server to exit.
(Bug#27381)
A subquery could get incorrect values for references to outer query columns when it contained aggregate functions that were aggregated in outer context. (Bug#27321)
The server did not shut down cleanly. (Bug#27310)
In a view, a column that was defined using a
GEOMETRY function was treated as having the
LONGBLOB data type rather than
the GEOMETRY type.
(Bug#27300)
Queries containing subqueries with
COUNT(*) aggregated in an outer
context returned incorrect results. This happened only if the
subquery did not contain any references to outer columns.
(Bug#27257)
Use of an aggregate function from an outer context as an
argument to GROUP_CONCAT() caused
a server crash.
(Bug#27229)
String truncation upon insertion into an integer or year column did not generate a warning (or an error in strict mode). (Bug#27176, Bug#26359)
Storing NULL values in spatial fields caused
excessive memory allocation and crashes on some systems.
(Bug#27164)
Row equalities in WHERE clauses could cause
memory corruption.
(Bug#27154)
GROUP BY on a ucs2 column
caused a server crash when there was at least one empty string
in the column.
(Bug#27079)
Duplicate members in SET or
ENUM definitions were not
detected. Now they result in a warning; if strict SQL mode is
enabled, an error occurs instead.
(Bug#27069)
For INSERT
... ON DUPLICATE KEY UPDATE statements on tables
containing AUTO_INCREMENT columns,
LAST_INSERT_ID() was reset to 0
if no rows were successfully inserted or changed. “Not
changed” includes the case where a row was updated to its
current values, but in that case,
LAST_INSERT_ID() should not be
reset to 0. Now LAST_INSERT_ID()
is reset to 0 only if no rows were successfully inserted or
touched, whether or not touched rows were changed.
(Bug#27033)
See also Bug#27210, Bug#27006.
This regression was introduced by Bug#19978.
mysql_install_db could terminate with an error after failing to determine that a system table already existed. (Bug#27022)
In a MEMORY table, using a
BTREE index to scan for updatable rows could
lead to an infinite loop.
(Bug#26996)
Invalid optimization of pushdown conditions for queries where an outer join was guaranteed to read only one row from the outer table led to results with too few rows. (Bug#26963)
Windows binaries contained no debug symbol file. Now
.map and .pdb files are
included in 32-bit builds for mysqld-nt.exe,
mysqld-debug.exe, and
mysqlmanager.exe.
(Bug#26893)
For InnoDB tables having a clustered index
that began with a CHAR or
VARCHAR column, deleting a record
and then inserting another before the deleted record was purged
could result in table corruption.
(Bug#26835)
Duplicates were not properly identified among (potentially) long
strings used as arguments for
GROUP_CONCAT(DISTINCT).
(Bug#26815)
ALTER VIEW requires the
CREATE VIEW and
DROP privileges for the view.
However, if the view was created by another user, the server
erroneously required the SUPER
privilege.
(Bug#26813)
A result set column formed by concatention of string literals
was incomplete when the column was produced by a subquery in the
FROM clause.
(Bug#26738)
When using the result of
SEC_TO_TIME() for time value
greater than 24 hours in an ORDER BY clause,
either directly or through a column alias, the rows were sorted
incorrectly as strings.
(Bug#26672)
The range optimizer could cause the server to run out of memory. (Bug#26625)
The range optimizer could consume a combinatorial amount of
memory for certain classes of WHERE clauses.
(Bug#26624)
mysqldump could crash or exhibit incorrect
behavior when some options were given very long values, such as
--fields-terminated-by=". The code has been cleaned up to
remove a number of fixed-sized buffers and to be more careful
about error conditions in memory allocation.
(Bug#26346)some very long
string"
If the server was started with
--skip-grant-tables, Selecting
from INFORMATION_SCHEMA tables causes a
server crash.
(Bug#26285)
For an INSERT statement that
should fail due to a column with no default value not being
assigned a value, the statement succeeded with no error if the
column was assigned a value in an ON DUPLICATE KEY
UPDATE clause, even if that clause was not used.
(Bug#26261)
The temporary file-creation code was cleaned up on Windows to improve server stability. (Bug#26233)
For MyISAM tables,
COUNT(*) could return an
incorrect value if the WHERE clause compared
an indexed TEXT column to the
empty string (''). This happened if the
column contained empty strings and also strings starting with
control characters such as tab or newline.
(Bug#26231)
For INSERT INTO
... SELECT where index searches used column prefixes,
insert errors could occur when key value type conversion was
done.
(Bug#26207)
For DELETE FROM (with no
tbl_name ORDER BY
col_nameWHERE or LIMIT clause),
the server did not check whether
col_name was a valid column in the
table.
(Bug#26186)
REPAIR TABLE ... USE_FRM with an
ARCHIVE table deleted all records from the
table.
(Bug#26138)
mysqldump crashed for
MERGE tables if the
--complete-insert
(-c) option was given.
(Bug#25993)
Setting a column to NOT NULL with an
ON DELETE SET NULL clause foreign key crashes
the server.
(Bug#25927)
On Windows, debug builds of mysqld could fail with heap assertions. (Bug#25765)
In certain situations, MATCH ... AGAINST
returned false hits for NULL values produced
by LEFT JOIN when no full-text index was
available.
(Bug#25729)
OPTIMIZE TABLE might fail on
Windows when it attempts to rename a temporary file to the
original name if the original file had been opened, resulting in
loss of the .MYD file.
(Bug#25521)
For SHOW ENGINE
INNODB STATUS, the LATEST DEADLOCK
INFORMATION was not always cleared properly.
(Bug#25494)
mysql_stmt_fetch() did an
invalid memory deallocation when used with the embedded server.
(Bug#25492)
Difficult repair or optimization operations could cause an assertion failure, resulting in a server crash. (Bug#25289)
Duplicate entries were not assessed correctly in a
MEMORY table with a BTREE
primary key on a utf8
ENUM column.
(Bug#24985)
Selecting the result of AVG()
within a UNION could produce
incorrect values.
(Bug#24791)
MBROverlaps() returned incorrect values in
some cases.
(Bug#24563)
Increasing the width of a DECIMAL
column could cause column values to be changed.
(Bug#24558)
A problem in handling of aggregate functions in subqueries caused predicates containing aggregate functions to be ignored during query execution. (Bug#24484)
The test for the
MYSQL_OPT_SSL_VERIFY_SERVER_CERT option for
mysql_options() was performed
incorrectly. Also changed as a result of this bug fix: The
arg option for the
mysql_options() C API function
was changed from char * to void
*.
(Bug#24121)
On Windows, debug builds of mysqlbinlog could fail with a memory error. (Bug#23736)
The values displayed for the
Innodb_row_lock_time,
Innodb_row_lock_time_avg, and
Innodb_row_lock_time_max
status variables were incorrect.
(Bug#23666)
SHOW CREATE VIEW qualified
references to stored functions in the view definition with the
function's database name, even when the database was the default
database. This affected mysqldump (which uses
SHOW CREATE VIEW to dump views)
because the resulting dump file could not be used to reload the
database into a different database. SHOW
CREATE VIEW now suppresses the database name for
references to functions in the default database.
(Bug#23491)
An INTO OUTFILE clause is allowed only for
the final SELECT of a
UNION, but this restriction was
not being enforced correctly.
(Bug#23345)
With the NO_AUTO_VALUE_ON_ZERO
SQL mode enabled,
LAST_INSERT_ID() could return 0
after
INSERT ...
ON DUPLICATE KEY UPDATE. Additionally, the next rows
inserted (by the same INSERT, or
the following INSERT with or
without ON DUPLICATE KEY UPDATE), would
insert 0 for the auto-generated value if the value for the
AUTO_INCREMENT column was
NULL or missing.
(Bug#23233)
SOUNDEX() returned an invalid
string for international characters in multi-byte character
sets.
(Bug#22638)
COUNT(
sometimes generated a spurious truncation warning.
(Bug#21976)decimal_expr)
InnoDB: The first read statement, if served
from the query cache, was not consistent with the
READ COMMITTED isolation
level.
(Bug#21409)
For a stored procedure containing a
SELECT statement that used a
complicated join with an ON expression, the
expression could be ignored during re-execution of the
procedure, yielding an incorrect result.
(Bug#20492)
In some cases, the optimizer preferred a range or full index scan access method over lookup access methods when the latter were much cheaper. (Bug#19372)
Conversion of DATETIME values in
numeric contexts sometimes did not produce a double
(YYYYMMDDHHMMSS.uuuuuu) value.
(Bug#16546)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.36).
Functionality added or changed:
The server now includes a timestamp in error messages that are
logged as a result of unhandled signals (such as mysqld
got signal 11 messages).
(Bug#24878)
Added the --secure-file-priv
option for mysqld, which limits the effect of
the LOAD_FILE() function and the
LOAD DATA and
SELECT ... INTO
OUTFILE statements to work only with files in a given
directory.
(Bug#18628)
Added the read-only hostname
system variable, which the server sets at startup to the server
host name.
To satisfy different user requirements, we provide several servers. mysqld is an optimized server that is a smaller, faster binary. Each package now also includes mysqld-debug, which is compiled with debugging support but is otherwise configured identically to the nondebug server.
Bugs fixed:
Incompatible Change:
INSERT DELAYED statements are not
supported for MERGE tables, but the
MERGE storage engine was not rejecting such
statements, resulting in table corruption. Applications
previously using INSERT DELAYED
into MERGE table will break when upgrading to
versions with this fix. To avoid the problem, remove
DELAYED from such statements.
(Bug#26464)
MySQL Cluster: An inadvertent use of unaligned data caused ndb_restore to fail on some 64-bit platforms, including Sparc and Itanium-2. (Bug#26739)
MySQL Cluster: An infinite loop in an internal logging function could cause trace logs to fill up with Unknown Signal type error messages and thus grow to unreasonable sizes. (Bug#26720)
MySQL Cluster:
An invalid pointer was returned following a
FSCLOSECONF signal when accessing the REDO
logs during a node restart or system restart.
(Bug#26515)
MySQL Cluster:
The failure of a data node when restarting it with
--initial could lead to failures of subsequent
data node restarts.
(Bug#26481)
MySQL Cluster: Takeover for local checkpointing due to multiple failures of master nodes was sometimes incorrectly handled. (Bug#26457)
MySQL Cluster:
The LockPagesInMainMemory parameter was not
read until after distributed communication had already started
between cluster nodes. When the value of this parameter was
1, this could sometimes result in data node
failure due to missed heartbeats.
(Bug#26454)
MySQL Cluster: Under some circumstances, following the restart of a management node, all data nodes would connect to it normally, but some of them subsequently failed to log any events to the management node. (Bug#26293)
MySQL Cluster:
The message Error 0 in readAutoIncrementValue(): no
Error was written to the error log whenever
SHOW TABLE STATUS was performed
on a Cluster table that did not have an
AUTO_INCREMENT column.
(Bug#21033)
Replication: A multiple-row delayed insert with an auto-increment column could cause duplicate entries to be created on the slave in a replication environment. (Bug#26116, Bug#25507)
Replication: Duplicating the usage of a user variable in a stored procedure or trigger would not be replicated correctly to the slave. (Bug#25167)
Replication:
DROP TRIGGER statements would not
be filtered on the slave when using the
replication-wild-do-table option.
(Bug#24478)
Replication:
For INSERT
... ON DUPLICATE KEY UPDATE statements where some
AUTO_INCREMENT values were generated
automatically for inserts and some rows were updated, one
auto-generated value was lost per updated row, leading to faster
exhaustion of the range of the AUTO_INCREMENT
column.
Because the original problem can affect replication (different values on master and slave), it is recommended that the master and its slaves be upgraded to the current version. (Bug#24432)
Replication:
Loading data using
LOAD DATA
INFILE may not replicate correctly (due to character
set incompatibilities) if the
character_set_database variable
is set before the data is loaded.
(Bug#15126)
Replication: User defined variables used within stored procedures and triggers are not replicated correctly when operating in statement-based replication mode. (Bug#14914, Bug#20141)
SELECT ... INTO
OUTFILE with a long FIELDS ENCLOSED
BY value could crash the server.
(Bug#27231)
An INSERT
... ON DUPLICATE KEY UPDATE statement might modify
values in a table but not flush affected data from the query
cache, causing subsequent selects to return stale results. This
made the combination of query cache plus ON DUPLICATE
KEY UPDATE very unreliable.
(Bug#27210)
See also Bug#27006, Bug#27033.
This regression was introduced by Bug#19978.
For MERGE tables defined on underlying tables
that contained a short VARCHAR
column (shorter than four characters), using
ALTER TABLE on at least one but
not all of the underlying tables caused the table definitions to
be considered different from that of the
MERGE table, even if the
ALTER TABLE did not change the
definition.
(Bug#26881)
Use of a subquery containing GROUP BY and
WITH ROLLUP caused a server crash.
(Bug#26830)
Added support for --debugger=dbx for
mysql-test-run.pl and added support for
--debugger=devenv,
--debugger=DevEnv, and
--debugger=.
(Bug#26792)/path/to/devenv
SSL connections failed on Windows. (Bug#26678)
Use of a subquery containing a
UNION with an invalid
ORDER BY clause caused a server crash.
(Bug#26661)
In some error messages, inconsistent format specifiers were used for the translations in different languages. comp_err (the error message compiler) now checks for mismatches. (Bug#26571)
Views that used a scalar correlated subquery returned incorrect results. (Bug#26560)
UNHEX() IS NULL comparisons failed when
UNHEX() returned
NULL.
(Bug#26537)
On 64-bit Windows, large timestamp values could be handled incorrectly. (Bug#26536)
For some values of the position argument, the
INSERT() function could insert a
NUL byte into the result.
(Bug#26281)
INSERT DELAYED statements
inserted incorrect values into
BIT columns.
(Bug#26238)
BENCHMARK() did not work
correctly for expressions that produced a
DECIMAL result.
(Bug#26093)
LOAD DATA
INFILE sent an okay to the client before writing the
binary log and committing the changes to the table had finished,
thus violating ACID requirements.
(Bug#26050)
X() IS NULL and Y() IS
NULL comparisons failed when
X() and
Y() returned
NULL.
(Bug#26038)
Indexes on TEXT columns were
ignored when ref accesses
were evaluated.
(Bug#25971)
If a thread previously serviced a connection that was killed, excessive memory and CPU use by the thread occurred if it later serviced a connection that had to wait for a table lock. (Bug#25966)
VIEW restrictions were applied to
SELECT statements after a
CREATE VIEW statement failed, as
though the CREATE had succeeded.
(Bug#25897)
Several deficiencies in resolution of column names for
INSERT ...
SELECT statements were corrected.
(Bug#25831)
Inserting utf8 data into a
TEXT column that used a
single-byte character set could result in spurious warnings
about truncated data.
(Bug#25815)
In certain cases it could happen that deleting a row corrupted
an RTREE index. This affected indexes on
spatial columns.
(Bug#25673)
Expressions involving SUM(), when
used in an ORDER BY clause, could lead to
out-of-order results.
(Bug#25376)
Use of a GROUP BY clause that referred to a
stored function result together with WITH
ROLLUP caused incorrect results.
(Bug#25373)
A stored procedure that made use of cursors failed when the procedure was invoked from a stored function. (Bug#25345)
On Windows, the server exhibited a file-handle leak after reaching the limit on the number of open file descriptors. (Bug#25222)
The REPEAT() function did not
allow a column name as the count
parameter.
(Bug#25197)
A reference to a nonexistent column in the ORDER
BY clause of an UPDATE ... ORDER BY
statement could cause a server crash.
(Bug#25126)
A view on a join is insertable for
INSERT statements that store
values into only one table of the join. However, inserts were
being rejected if the inserted-into table was used in a
self-join because MySQL incorrectly was considering the insert
to modify multiple tables of the view.
(Bug#25122)
MySQL would not compile when configured using
--without-query-cache.
(Bug#25075)
IF(expr,
unsigned_expr,
unsigned_expr) was evaluated to a
signed result, not unsigned. This has been corrected. The fix
also affects constructs of the form IS [NOT]
{TRUE|FALSE}, which were transformed internally into
IF() expressions that evaluated
to a signed result.
For existing views that were defined using IS [NOT]
{TRUE|FALSE} constructs, there is a related
implication. The definitions of such views were stored using the
IF() expression, not the original
construct. This is manifest in that SHOW
CREATE VIEW shows the transformed
IF() expression, not the original
one. Existing views will evaluate correctly after the fix, but
if you want SHOW CREATE VIEW to
display the original construct, you must drop the view and
re-create it using its original definition. New views will
retain the construct in their definition.
(Bug#24532)
A user-defined variable could be assigned an incorrect value if a temporary table was employed in obtaining the result of the query used to determine its value. (Bug#24010)
Queries that used a temporary table for the outer query when evaluating a correlated subquery could return incorrect results. (Bug#23800)
When using certain server SQL modes, the
mysql.proc table was not created by
mysql_install_db.
(Bug#23669)
DOUBLE values such as
20070202191048.000000 were being treated as
illegal arguments by WEEK().
(Bug#23616)
The server could crash if two or more threads initiated query cache resize operation at moments very close in time. (Bug#23527)
NOW() returned the wrong value in
statements executed at server startup with the
--init-file option.
(Bug#23240)
When nesting stored procedures within a trigger on a table, a
false dependency error was thrown when one of the nested
procedures contained a DROP TABLE
statement.
(Bug#22580)
Instance Manager did not remove the angel PID file on a clean shutdown. (Bug#22511)
EXPLAIN EXTENDED did not show
WHERE conditions that were optimized away.
(Bug#22331)
IN ((,
subquery))IN (((,
and so forth, are equivalent to subquery)))IN
(, which is always
interpreted as a table subquery (so that it is allowed to return
more than one row). MySQL was treating the
“over-parenthesized” subquery as a single-row
subquery and rejecting it if it returned more than one row. This
bug primarily affected automatically generated code (such as
queries generated by Hibernate), because humans rarely write the
over-parenthesized forms.
(Bug#21904)subquery)
An INSERT trigger invoking a
stored routine that inserted into a table other than the one on
which the trigger was defined would fail with a Table
'...' doesn't exist referring to the second table
when attempting to delete records from the first table.
(Bug#21825)
When a stored routine attempted to execute a statement accessing a nonexistent table, the error was not caught by the routine's exception handler. (Bug#20713, Bug#8407)
The conditions checked by the optimizer to allow use of indexes
in IN predicate calculations were
unnecessarily tight and were relaxed.
(Bug#20420)
When a TIME_FORMAT() expression
was used as a column in a GROUP BY clause,
the expression result was truncated.
(Bug#20293)
The creation of MySQL system tables was not checked for by mysql-test-run.pl. (Bug#20166)
For index reads, the BLACKHOLE engine did not
return end-of-file (which it must because
BLACKHOLE tables contain no rows), causing
some queries to crash.
(Bug#19717)
For expr
IN(value_list)BIGINT UNSIGNED values
were used for expr or in the value
list.
(Bug#19342)
When attempting to call a stored procedure creating a table from
a trigger on a table tbl in a database
db, the trigger failed with ERROR
1146 (42S02): Table 'db.tbl' doesn't exist. However,
the actual reason that such a trigger fails is due to the fact
that CREATE TABLE causes an
implicit COMMIT, and so a trigger
cannot invoke a stored routine containing this statement. A
trigger which does so now fails with ERROR 1422
(HY000): Explicit or implicit commit is not allowed in stored
function or trigger, which makes clear the reason
for the trigger's failure.
(Bug#18914)
The update columns for INSERT ... SELECT ... ON
DUPLICATE KEY UPDATE could be assigned incorrect
values if a temporary table was used to evaluate the
SELECT.
(Bug#16630)
For SUBSTRING() evaluation using
a temporary table, when
SUBSTRING() was used on a
LONGTEXT column, the max_length metadata
value of the result was incorrectly calculated and set to 0.
Consequently, an empty string was returned instead of the
correct result.
(Bug#15757)
Local variables in stored routines or triggers, when declared as
the BIT type, were interpreted as
strings.
(Bug#12976)
CONNECTION is no longer treated as a reserved
word.
(Bug#12204)
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.33.
Functionality added or changed:
Incompatible Change: MySQL Cluster:
The LockPagesInMainMemory configuration
parameter has changed its type and possible values. For more
information, see
LockPagesInMainMemory
.
The values true and
false are no longer accepted for this
parameter. If you were using this parameter and had it set to
false in a previous release, you must
change it to 0. If you had this parameter
set to true, you should instead use
1 to obtain the same behavior as
previously, or 2 to take advantage of new
functionality introduced with this release, as described in
the section cited above.
Incompatible Change:
Previously, the DATE_FORMAT()
function returned a binary string. Now it returns a string with
a character set and collation given by
character_set_connection and
collation_connection so that it
can return month and weekday names containing non-ASCII
characters.
(Bug#22646)
Important Change:
When using MERGE tables the definition of the
MERGE table and the MyISAM
tables are checked each time the tables are opened for access
(including any SELECT or
INSERT statement. Each table is
compared for column order, types, sizes and associated. If there
is a difference in any one of the tables then the statement will
fail.
Added the
Uptime_since_flush_status
status variable, which indicates the number of seconds since the
most recent FLUSH STATUS statement.
(Community contribution by Jeremy Cole)
(Bug#24822)
Added the SHOW PROFILES and
SHOW PROFILE statements to
display statement profile data, and the accompanying
INFORMATION_SCHEMA.PROFILING table.
Profiling is controlled via the
profiling and
profiling_history_size session
variables. see Section 12.5.5.29, “SHOW PROFILES Syntax”, and
Section 19.17, “The INFORMATION_SCHEMA PROFILING Table”. (Community contribution by
Jeremy Cole)
The profiling feature is enabled via the
--enable-community-features
and --enable-profiling options
to configure. These options are enabled by
default; to disable them, use
--disable-community-features
and
--disable-profiling.
(Bug#24795)
The localhost anonymous user account created
during MySQL installation on Windows now has no global
privileges. Formerly this account had all global privileges. For
operations that require global privileges, the
root account can be used instead.
(Bug#24496)
The --skip-thread-priority option
now is enabled by default for binary Mac OS X distributions. Use
of thread priorities degrades performance on Mac OS X.
(Bug#18526)
This is the last version for which MySQL-Max RPM distributions are available. (This change was already made for non-RPM binary distributions in 5.0.27.)
The bundled yaSSL library was upgraded to version 1.5.8.
Added the
--disable-grant-options option
to configure. If configure
is run with this option, the
--bootstrap,
--skip-grant-tables, and
--init-file options for
mysqld are disabled and cannot be used. For
Windows, the configure.js script recognizes
the DISABLE_GRANT_OPTIONS flag, which has the
same effect.
Bugs fixed:
Security Fix:
Using an INFORMATION_SCHEMA table with
ORDER BY in a subquery could cause a server
crash.
We would like to thank Oren Isacson of Flowgate Security Consulting and Stefan Streichsbier of SEC Consult for informing us of this problem. (Bug#24630, Bug#26556, CVE-2007-1420)
Incompatible Change:
For ENUM columns that had
enumeration values containing commas, the commas were mapped to
0xff internally. However, this rendered the
commas indistinguishable from true 0xff
characters in the values. This no longer occurs. However, the
fix requires that you dump and reload any tables that have
ENUM columns containing any true
0xff values. Dump the tables using
mysqldump with the current server before
upgrading from a version of MySQL 5.0 older than 5.0.36 to
version 5.0.36 or newer.
(Bug#24660)
Partitioning: MySQL Cluster:
A query with an IN clause against an
NDB table employing explicit
user-defined partitioning did not always return all matching
rows.
(Bug#25821)
MySQL Cluster:
It was not possible to create an
NDB table with a key on two
VARCHAR columns where both
columns had a storage length in excess of 256.
(Bug#25746)
MySQL Cluster: Hosts in clusters with large numbers of nodes could experience excessive CPU usage while obtaining configuration data. (Bug#25711)
MySQL Cluster: In some circumstances, shutting down the cluster could cause connected mysqld processes to crash. (Bug#25668)
MySQL Cluster:
Memory allocations for TEXT
columns were calculated incorrectly, resulting in space being
wasted and other issues.
(Bug#25562)
MySQL Cluster: The failure of a master node during a node restart could lead to a resource leak, causing later node failures. (Bug#25554)
MySQL Cluster:
An UPDATE using an
IN clause on an
NDB table on which there was a
trigger caused mysqld to crash.
(Bug#25522)
MySQL Cluster: A node shutdown occurred if the master failed during a commit. (Bug#25364)
MySQL Cluster:
Creating a nonunique index with the USING
HASH clause silently created an ordered index instead
of issuing a warning.
(Bug#24820)
MySQL Cluster:
The ndb_size.tmpl file (necessary for using
the ndb_size.pl script) was missing from
binary distributions.
(Bug#24191)
MySQL Cluster:
When a data node was shut down using the management client
STOP command, a connection event
(NDB_LE_Connected) was logged instead of a
disconnection event (NDB_LE_Disconnected).
(Bug#22773)
MySQL Cluster: The management server did not handle logging of node shutdown events correctly in certain cases. (Bug#22013)
MySQL Cluster:
SELECT statements with a
BLOB or
TEXT column in the selected
column list and a WHERE condition including a
primary key lookup on a VARCHAR
primary key produced empty result sets.
(Bug#19956)
Replication:
When SET PASSWORD was written to
the binary log double quotes were included in the statement. If
the slave was running in with the server SQL mode set to
ANSI_QUOTES, then the event
failed, which halted the replication process.
(Bug#24158)
Replication: A stored procedure, executed from a connection using a binary character set, and which wrote multibyte data, would write incorrectly escaped entries to the binary log. This caused syntax errors, and caused replication to fail. (Bug#23619, Bug#24492)
Replication:
Changes to the lc_time_names
system variable were not replicated.
(Bug#22645)
Replication:
For SET,
SELECT, and
DO statements that invoked a
stored function from a database other than the default database,
the function invocation could fail to be replicated.
(Bug#19725)
Replication: If a slave server closed its relay log (for example, due to an error during log rotation), the I/O thread did not recognize this and still tried to write to the log, causing a server crash. (Bug#10798)
Cluster API:
Deletion of an Ndb_cluster_connection object
took a very long time.
(Bug#25487)
Cluster API:
Invoking the NdbTransaction::execute() method
using execution type Commit and abort option
AO_IgnoreError could lead to a crash of the
transaction coordinator (DBTC).
(Bug#25090)
Cluster API: A unique index lookup on a nonexistent tuple could lead to a data node timeout (error 4012). (Bug#25059)
Cluster API:
libndbclient.so was not versioned.
(Bug#13522)
Using ORDER BY or GROUP BY
could yield different results when selecting from a view and
selecting from the underlying table.
(Bug#26209)
DISTINCT queries that were executed using a
loose scan for an InnoDB table that had been
emptied caused a server crash.
(Bug#26159)
A WHERE clause that used
BETWEEN for
DATETIME values could be treated
differently for a SELECT and a
view defined as that SELECT.
(Bug#26124)
Collation for LEFT JOIN comparisons could be
evaluated incorrectly, leading to improper query results.
(Bug#26017)
The WITH CHECK OPTION clause for views was
ignored for updates of multiple-table views when the updates
could not be performed on fly and the rows to update had to be
put into temporary tables first.
(Bug#25931)
LOAD DATA
INFILE did not work with pipes.
(Bug#25807)
The SEC_TO_TIME() and
QUARTER() functions sometimes did
not handle NULL values correctly.
(Bug#25643)
The InnoDB parser sometimes did not account
for null bytes, causing spurious failure of some queries.
(Bug#25596)
View definitions that used the ! operator
were treated as containing the NOT operator,
which has a different precedence and can produce different
results. .
(Bug#25580)
An error in the name resolution of nested JOIN ...
USING constructs was corrected.
(Bug#25575)
GROUP BY and DISTINCT did
not group NULL values for columns that have a
UNIQUE index. .
(Bug#25551)
The --with-readline option for
configure did not work for commercial source
packages, but no error message was printed to that effect. Now a
message is printed.
(Bug#25530)
mysql_stmt_fetch() did an
invalid memory deallocation when used with the embedded server.
(Bug#25492)
Referencing an ambiguous column alias in an expression in the
ORDER BY clause of a query caused the server
to crash.
(Bug#25427)
A yaSSL program named test was installed, causing conflicts with the test system utility. It is no longer installed. (Bug#25417)
For a UNIQUE index containing many
NULL values, the optimizer would prefer the
index for col IS
NULL
An AFTER UPDATE trigger on an
InnoDB table with a composite primary key
caused the server to crash.
(Bug#25398)
Passing a NULL value to a user-defined
function from within a stored procedure crashes the server.
(Bug#25382)
perror crashed on some platforms due to
failure to handle a NULL pointer.
(Bug#25344)
mysql.server stop timed out too quickly (35 seconds) waiting for the server to exit. Now it waits up to 15 minutes, to ensure that the server exits. (Bug#25341)
A query that contained an EXIST subquery with
a UNION over correlated and
uncorrelated SELECT queries could
cause the server to crash.
(Bug#25219)
mysql_kill() caused a server
crash when used on an SSL connection.
(Bug#25203)
yaSSL was sensitive to the presence of whitespace at the ends of lines in PEM-encoded certificates, causing a server crash. (Bug#25189)
A query with ORDER BY and GROUP
BY clauses where the ORDER BY
clause had more elements than the GROUP BY
clause caused a memory overrun leading to a crash of the server.
(Bug#25172)
Use of ON DUPLICATE KEY UPDATE defeated the
usual restriction against inserting into a join-based view
unless only one of the underlying tables is used.
(Bug#25123)
Using a view in combination with a USING
clause caused column aliases to be ignored.
(Bug#25106)
A multiple-table DELETE QUICK could sometimes
cause one of the affected tables to become corrupted.
(Bug#25048)
ALTER TABLE ... ENABLE KEYS acquired a global
lock, preventing concurrent execution of other statements that
use tables. .
(Bug#25044)
An assertion failed incorrectly for prepared statements that
contained a single-row uncorrelated subquery that was used as an
argument of the IS NULL predicate.
(Bug#25027)
A return value of -1 from user-defined
handlers was not handled well and could result in conflicts with
server code.
(Bug#24987)
Accessing a fixed record format table with a crashed key definition results in server/myisamchk segmentation fault. (Bug#24855)
mysqld_multi and
mysqlaccess looked for option files in
/etc even if the
--sysconfdir option for
configure had been given to specify a
different directory.
(Bug#24780)
If there was insufficient memory available to mysqld, this could sometimes cause the server to hang during startup. (Bug#24751)
Optimizations that are legal only for subqueries without tables
and WHERE conditions were applied for any
subquery without tables.
(Bug#24670)
If an ORDER BY or GROUP BY
list included a constant expression being optimized away and, at
the same time, containing single-row subselects that returned
more that one row, no error was reported. If a query required
sorting by expressions containing single-row subselects that
returned more than one row, execution of the query could cause a
server crash.
(Bug#24653)
For ALTER TABLE, using
ORDER BY
could cause a
server crash. Now the expressionORDER BY clause allows
only column names to be specified as sort criteria (which was
the only documented syntax, anyway).
(Bug#24562)
A workaround was implemented to avoid a race condition in the
NPTL pthread_exit() implementation.
(Bug#24507)
mysqltest crashed with a stack overflow. (Bug#24498)
Within stored routines or prepared statements, inconsistent
results occurred with multiple use of INSERT ... SELECT
... ON DUPLICATE KEY UPDATE when the ON
DUPLICATE KEY UPDATE clause erroneously tried to
assign a value to a column mentioned only in its
SELECT part.
(Bug#24491)
Expressions of the form (a, b) IN (SELECT a, MIN(b)
FROM t GROUP BY a) could produce incorrect results
when column a of table t
contained NULL values while column
b did not.
(Bug#24420)
If a prepared statement accessed a view, access to the tables listed in the query after that view was checked in the security context of the view. (Bug#24404)
Attempts to access a MyISAM table with a
corrupt column definition caused a server crash.
(Bug#24401)
When opening a corrupted .frm file during a
query, the server crashes.
(Bug#24358)
Some joins in which one of the joined tables was a view could return erroneous results or crash the server. (Bug#24345)
A view was not handled correctly if the
SELECT part contained “
\Z ”.
(Bug#24293)
A query using WHERE
could
cause the server to crash.
(Bug#24261)unsigned_column NOT IN
('negative_value')
Expressions of the form (a, b) IN (SELECT c, d
...) could produce incorrect results if
a, b, or both were
NULL.
(Bug#24127)
A FETCH statement using a cursor
on a table which was not in the table cache could sometimes
cause the server to crash.
(Bug#24117)
Queries that evaluate NULL IN (SELECT ... UNION SELECT
...) could produce an incorrect result
(FALSE instead of NULL).
(Bug#24085)
Hebrew-to-Unicode conversion failed for some characters. Definitions for the following Hebrew characters (as specified by the ISO/IEC 8859-8:1999) were added: LEFT-TO-RIGHT MARK (LRM), RIGHT-TO-LEFT MARK (RLM) (Bug#24037)
Some UPDATE statements were
slower than in previous versions when the search key could not
be converted to a valid value for the type of the search column.
(Bug#24035)
ISNULL(DATE(NULL)) and
ISNULL(CAST(NULL AS DATE))
erroneously returned false.
(Bug#23938)
Within a stored routine, accessing a declared routine variable
with PROCEDURE ANALYSE() caused a server
crash.
(Bug#23782)
When reading from the standard input on Windows, mysqlbinlog opened the input in text mode rather than binary mode and consequently misinterpreted some characters such as Control-Z. (Bug#23735)
OPTIMIZE TABLE tried to sort
R-tree indexes such as spatial indexes, although this is not
possible (see Section 12.5.2.5, “OPTIMIZE TABLE Syntax”).
(Bug#23578)
For an InnoDB table with any ON
DELETE trigger,
TRUNCATE
TABLE mapped to DELETE
and activated triggers. Now a fast truncation occurs and
triggers are not activated. .
(Bug#23556)
The row count for MyISAM tables was not
updated properly, causing SHOW TABLE
STATUS to report incorrect values.
(Bug#23526)
User-defined variables could consume excess memory, leading to a
crash caused by the exhaustion of resources available to the
MEMORY storage engine, due to the fact that
this engine is used by MySQL for variable storage and
intermediate results of GROUP BY queries.
Where SET had been used, such a condition
could instead give rise to the misleading error message
You may only use constant expressions with
SET, rather than Out of memory (Needed
NNNNNN bytes).
(Bug#23443)
With ONLY_FULL_GROUP_BY
enables, the server was too strict: Some expressions involving
only aggregate values were rejected as nonaggregate (for
example, MAX(a) –
MIN(a)).
(Bug#23417)
The arguments to the ENCODE() and
the DECODE() functions were not
printed correctly, causing problems in the output of
EXPLAIN EXTENDED and in view definitions.
(Bug#23409)
A table created with the ROW_FORMAT = FIXED
table option lost the option if an index was added or dropped
with CREATE INDEX or
DROP INDEX.
(Bug#23404)
A deadlock could occur, with the server hanging on
Closing tables, with a sufficient number of
concurrent INSERT DELAYED,
FLUSH TABLES,
and ALTER TABLE operations.
(Bug#23312)
Some queries against INFORMATION_SCHEMA that
used subqueries failed. .
(Bug#23299)
readline detection did not work correctly on
NetBSD.
(Bug#23293)
If there was insufficient memory to store or update a blob
record in a MyISAM table then the table will
marked as crashed.
(Bug#23196)
LAST_INSERT_ID() was not reset to
0 if INSERT ...
SELECT inserted no rows.
(Bug#23170)
A compressed MyISAM table that became
corrupted could crash myisamchk and possibly
the MySQL Server.
(Bug#23139)
The number of setsockopt() calls performed
for reads and writes to the network socket was reduced to
decrease system call overhead.
(Bug#22943)
mysql_upgrade failed when called with a
--basedir path name containing spaces.
(Bug#22801)
SET lc_time_names = allowed only exact literal values, not expression
values.
(Bug#22647)value
The STDDEV() function returned a
positive value for data sets consisting of a single value.
(Bug#22555)
Storing values specified as hexadecimal values 64 or more bits
long in BIT(64),
BIGINT, or BIGINT
UNSIGNED columns did not raise any warning or error if
the value was out of range.
(Bug#22533)
SHOW COLUMNS reported some
NOT NULL columns as NULL.
(Bug#22377)
Type conversion errors during formation of index search conditions were not correctly checked, leading to incorrect query results. (Bug#22344)
Changing the value of MI_KEY_BLOCK_LENGTH in
myisam.h and recompiling MySQL resulted in
a myisamchk that saw existing
MyISAM tables as corrupt.
(Bug#22119)
A crash of the MySQL Server could occur when unpacking a
BLOB column from a row in a
corrupted MyISAM table. This could happen when trying to repair
a table using either REPAIR TABLE
or myisamchk; it could also happen when
trying to access such a “broken” row using
statements like SELECT if the
table was not marked as crashed.
(Bug#22053)
The code for generating USE
statements for binary logging of CREATE
PROCEDURE statements resulted in confusing output from
mysqlbinlog for DROP
PROCEDURE statements.
(Bug#22043)
For the IF() and
COALESCE() function and
CASE expressions, large unsigned
integer values could be mishandled and result in warnings.
(Bug#22026)
SSL connections could hang at connection shutdown. (Bug#21781, Bug#24148)
The FEDERATED storage engine did not support
the euckr character set.
(Bug#21556)
When updating a table that used a JOIN of the
table itself (for example, when building trees) and the table
was modified on one side of the expression, the table would
either be reported as crashed or the wrong rows in the table
would be updated.
(Bug#21310)
mysqld_error.h was not installed when only
the client libraries were built.
(Bug#21265)
InnoDB: During a restart of the MySQL Server
that followed the creation of a temporary table using the
InnoDB storage engine, MySQL failed to clean
up in such a way that InnoDB still attempted
to find the files associated with such tables.
(Bug#20867)
Inserting DEFAULT into a column with no
default value could result in garbage in the column. Now the
same result occurs as when inserting NULL
into a NOT NULL column.
(Bug#20691)
A stored routine containing semicolon in its body could not be reloaded from a dump of a binary log. (Bug#20396)
SELECT ... FOR
UPDATE, SELECT
... LOCK IN SHARE MODE,
DELETE, and
UPDATE statements executed using
a full table scan were not releasing locks on rows that did not
satisfy the WHERE condition.
(Bug#20390)
On Windows, if the server was installed as a service, it did not auto-detect the location of the data directory. (Bug#20376)
The BUILD/check-cpu script did not recognize Celeron processors. (Bug#20061)
If a duplicate key value was present in the table,
INSERT ...
ON DUPLICATE KEY UPDATE reported a row count
indicating that a record was updated, even when no record
actually changed due to the old and new values being the same.
Now it reports a row count of zero.
(Bug#19978)
ORDER BY values of the
DOUBLE or
DECIMAL types could change the
result returned by a query.
(Bug#19690)
The readline library wrote to uninitialized
memory, causing mysql to crash.
(Bug#19474)
mysqltest incorrectly tried to retrieve result sets for some queries where no result set was available. (Bug#19410)
Use of already freed memory caused SSL connections to hang forever. (Bug#19209)
Some
CASE
statements inside stored routines could lead to excessive
resource usage or a crash of the server.
(Bug#19194, Bug#24854)
Instance Manager could crash during shutdown. (Bug#19044)
The server might fail to use an appropriate index for
DELETE when ORDER
BY, LIMIT, and a nonrestricting
WHERE are present.
(Bug#17711)
No warning was issued for use of the DATA
DIRECTORY or INDEX DIRECTORY table
options on a platform that does not support them.
(Bug#17498)
The FEDERATED storage engine did not support
the utf8 character set.
(Bug#17044)
The optimizer removes expressions from GROUP
BY and DISTINCT clauses if they
happen to participate in
predicates of the
expression =
constantWHERE clause, the idea being that, if the
expression is equal to a constant, then it cannot take on
multiple values. However, for predicates where the expression
and the constant item are of different result types (for
example, when a string column is compared to 0), this is not
valid, and can lead to invalid results in such cases. The
optimizer now performs an additional check of the result types
of the expression and the constant; if their types differ, then
the expression is not removed from the GROUP
BY list.
(Bug#15881)
When a prepared statement failed during the prepare operation, the error code was not cleared when it was reused, even if the subsequent use was successful. (Bug#15518)
Dropping a user-defined function sometimes did not remove the
UDF entry from the mysql.proc table.
(Bug#15439)
Inserting a row into a table without specifying a value for a
BINARY(
column caused the column to be set to spaces, not zeroes.
(Bug#14171)N) NOT NULL
On Windows, the SLEEP() function
could sleep too long, especially after a change to the system
clock.
(Bug#14094, Bug#24686, Bug#17635)
mysqldump --order-by-primary failed if the primary key name was an identifier that required quoting. (Bug#13926)
To enable installation of MySQL RPMs on Linux systems running RHEL 4 (which includes SE-Linux) additional information was provided to specify some actions that are allowed to the MySQL binaries. (Bug#12676)
The presence of ORDER BY in a view definition
prevented the MERGE algorithm from being used
to resolve the view even if nothing else in the definition
required the TEMPTABLE algorithm.
(Bug#12122)
The internal functions for table preparation, creation, and
alteration were not re-execution friendly, causing problems in
code that: repeatedly altered a table; repeatedly created and
dropped a table; opened and closed a cursor on a table, altered
the table, and then reopened the cursor; used
ALTER TABLE to change a table's
current AUTO_INCREMENT value; created indexes
on utf8 columns.
Re-execution of CREATE DATABASE,
CREATE TABLE, and
ALTER TABLE statements in stored
routines or as prepared statements also caused incorrect results
or crashes.
(Bug#4968, Bug#6895, Bug#19182, Bug#19733, Bug#22060, Bug#24879)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.36).
Bugs fixed:
SELECT ... INTO
OUTFILE with a long FIELDS ENCLOSED
BY value could crash the server.
(Bug#27231)
For MERGE tables defined on underlying tables
that contained a short VARCHAR
column (shorter than four characters), using
ALTER TABLE on at least one but
not all of the underlying tables caused the table definitions to
be considered different from that of the
MERGE table, even if the
ALTER TABLE did not change the
definition.
(Bug#26881)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
After release, a trigger failure problem was found to have been introduced. (Bug#27006) Users affected by this issue should upgrade to MySQL 5.0.38, which corrects the problem.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.34).
Functionality added or changed:
Incompatible Change: MySQL Cluster:
The LockPagesInMainMemory configuration
parameter has changed its type and possible values. For more
information, see
LockPagesInMainMemory
.
The values true and
false are no longer accepted for this
parameter. If you were using this parameter and had it set to
false in a previous release, you must
change it to 0. If you had this parameter
set to true, you should instead use
1 to obtain the same behavior as
previously, or 2 to take advantage of new
functionality introduced with this release, as described in
the section cited above.
Incompatible Change:
Previously, the DATE_FORMAT()
function returned a binary string. Now it returns a string with
a character set and collation given by
character_set_connection and
collation_connection so that it
can return month and weekday names containing non-ASCII
characters.
(Bug#22646)
Important Change:
When using MERGE tables the definition of the
MERGE table and the MyISAM
tables are checked each time the tables are opened for access
(including any SELECT or
INSERT statement. Each table is
compared for column order, types, sizes and associated. If there
is a difference in any one of the tables then the statement will
fail.
The localhost anonymous user account created
during MySQL installation on Windows now has no global
privileges. Formerly this account had all global privileges. For
operations that require global privileges, the
root account can be used instead.
(Bug#24496)
The bundled yaSSL library was upgraded to version 1.5.8.
Bugs fixed:
Security Fix:
Using an INFORMATION_SCHEMA table with
ORDER BY in a subquery could cause a server
crash.
We would like to thank Oren Isacson of Flowgate Security Consulting and Stefan Streichsbier of SEC Consult for informing us of this problem. (Bug#24630, Bug#26556, CVE-2007-1420)
Incompatible Change:
For ENUM columns that had
enumeration values containing commas, the commas were mapped to
0xff internally. However, this rendered the
commas indistinguishable from true 0xff
characters in the values. This no longer occurs. However, the
fix requires that you dump and reload any tables that have
ENUM columns containing any true
0xff values. Dump the tables using
mysqldump with the current server before
upgrading from a version of MySQL 5.0 older than 5.0.36 to
version 5.0.36 or newer.
(Bug#24660)
Partitioning: MySQL Cluster:
A query with an IN clause against an
NDB table employing explicit
user-defined partitioning did not always return all matching
rows.
(Bug#25821)
MySQL Cluster:
It was not possible to create an
NDB table with a key on two
VARCHAR columns where both
columns had a storage length in excess of 256.
(Bug#25746)
MySQL Cluster: In some circumstances, shutting down the cluster could cause connected mysqld processes to crash. (Bug#25668)
MySQL Cluster:
Memory allocations for TEXT
columns were calculated incorrectly, resulting in space being
wasted and other issues.
(Bug#25562)
MySQL Cluster: The failure of a master node during a node restart could lead to a resource leak, causing later node failures. (Bug#25554)
MySQL Cluster:
An UPDATE using an
IN clause on an
NDB table on which there was a
trigger caused mysqld to crash.
(Bug#25522)
MySQL Cluster: A node shutdown occurred if the master failed during a commit. (Bug#25364)
MySQL Cluster:
Creating a nonunique index with the USING
HASH clause silently created an ordered index instead
of issuing a warning.
(Bug#24820)
MySQL Cluster:
The ndb_size.tmpl file (necessary for using
the ndb_size.pl script) was missing from
binary distributions.
(Bug#24191)
MySQL Cluster: The management server did not handle logging of node shutdown events correctly in certain cases. (Bug#22013)
MySQL Cluster:
SELECT statements with a
BLOB or
TEXT column in the selected
column list and a WHERE condition including a
primary key lookup on a VARCHAR
primary key produced empty result sets.
(Bug#19956)
MySQL Cluster: The loss of one or more data nodes could sometimes cause ndb_mgmd to use a high amount of CPU (15 percent or more, as opposed to 1 to 2 percent normally).
Replication:
When SET PASSWORD was written to
the binary log double quotes were included in the statement. If
the slave was running in with the server SQL mode set to
ANSI_QUOTES, then the event
failed, which halted the replication process.
(Bug#24158)
Replication: A stored procedure, executed from a connection using a binary character set, and which wrote multibyte data, would write incorrectly escaped entries to the binary log. This caused syntax errors, and caused replication to fail. (Bug#23619, Bug#24492)
Replication:
Changes to the lc_time_names
system variable were not replicated.
(Bug#22645)
Replication:
For SET,
SELECT, and
DO statements that invoked a
stored function from a database other than the default database,
the function invocation could fail to be replicated.
(Bug#19725)
Replication: If a slave server closed its relay log (for example, due to an error during log rotation), the I/O thread did not recognize this and still tried to write to the log, causing a server crash. (Bug#10798)
Cluster API:
Deletion of an Ndb_cluster_connection object
took a very long time.
(Bug#25487)
Cluster API:
libndbclient.so was not versioned.
(Bug#13522)
Using ORDER BY or GROUP BY
could yield different results when selecting from a view and
selecting from the underlying table.
(Bug#26209)
DISTINCT queries that were executed using a
loose scan for an InnoDB table that had been
emptied caused a server crash.
(Bug#26159)
A WHERE clause that used
BETWEEN for
DATETIME values could be treated
differently for a SELECT and a
view defined as that SELECT.
(Bug#26124)
Collation for LEFT JOIN comparisons could be
evaluated incorrectly, leading to improper query results.
(Bug#26017)
The WITH CHECK OPTION clause for views was
ignored for updates of multiple-table views when the updates
could not be performed on fly and the rows to update had to be
put into temporary tables first.
(Bug#25931)
LOAD DATA
INFILE did not work with pipes.
(Bug#25807)
The SEC_TO_TIME() and
QUARTER() functions sometimes did
not handle NULL values correctly.
(Bug#25643)
The InnoDB parser sometimes did not account
for null bytes, causing spurious failure of some queries.
(Bug#25596)
View definitions that used the ! operator
were treated as containing the NOT operator,
which has a different precedence and can produce different
results. .
(Bug#25580)
An error in the name resolution of nested JOIN ...
USING constructs was corrected.
(Bug#25575)
GROUP BY and DISTINCT did
not group NULL values for columns that have a
UNIQUE index. .
(Bug#25551)
The --with-readline option for
configure did not work for commercial source
packages, but no error message was printed to that effect. Now a
message is printed.
(Bug#25530)
A yaSSL program named test was installed, causing conflicts with the test system utility. It is no longer installed. (Bug#25417)
For a UNIQUE index containing many
NULL values, the optimizer would prefer the
index for col IS
NULL
An AFTER UPDATE trigger on an
InnoDB table with a composite primary key
caused the server to crash.
(Bug#25398)
Passing a NULL value to a user-defined
function from within a stored procedure crashes the server.
(Bug#25382)
perror crashed on some platforms due to
failure to handle a NULL pointer.
(Bug#25344)
mysql.server stop timed out too quickly (35 seconds) waiting for the server to exit. Now it waits up to 15 minutes, to ensure that the server exits. (Bug#25341)
A query that contained an EXIST subquery with
a UNION over correlated and
uncorrelated SELECT queries could
cause the server to crash.
(Bug#25219)
mysql_kill() caused a server
crash when used on an SSL connection.
(Bug#25203)
yaSSL was sensitive to the presence of whitespace at the ends of lines in PEM-encoded certificates, causing a server crash. (Bug#25189)
A query with ORDER BY and GROUP
BY clauses where the ORDER BY
clause had more elements than the GROUP BY
clause caused a memory overrun leading to a crash of the server.
(Bug#25172)
Use of ON DUPLICATE KEY UPDATE defeated the
usual restriction against inserting into a join-based view
unless only one of the underlying tables is used.
(Bug#25123)
ALTER TABLE ... ENABLE KEYS acquired a global
lock, preventing concurrent execution of other statements that
use tables. .
(Bug#25044)
A return value of -1 from user-defined
handlers was not handled well and could result in conflicts with
server code.
(Bug#24987)
Accessing a fixed record format table with a crashed key definition results in server/myisamchk segmentation fault. (Bug#24855)
mysqld_multi and
mysqlaccess looked for option files in
/etc even if the
--sysconfdir option for
configure had been given to specify a
different directory.
(Bug#24780)
If there was insufficient memory available to mysqld, this could sometimes cause the server to hang during startup. (Bug#24751)
If an ORDER BY or GROUP BY
list included a constant expression being optimized away and, at
the same time, containing single-row subselects that returned
more that one row, no error was reported. If a query required
sorting by expressions containing single-row subselects that
returned more than one row, execution of the query could cause a
server crash.
(Bug#24653)
For ALTER TABLE, using
ORDER BY
could cause a
server crash. Now the expressionORDER BY clause allows
only column names to be specified as sort criteria (which was
the only documented syntax, anyway).
(Bug#24562)
A workaround was implemented to avoid a race condition in the
NPTL pthread_exit() implementation.
(Bug#24507)
mysqltest crashed with a stack overflow. (Bug#24498)
Within stored routines or prepared statements, inconsistent
results occurred with multiple use of INSERT ... SELECT
... ON DUPLICATE KEY UPDATE when the ON
DUPLICATE KEY UPDATE clause erroneously tried to
assign a value to a column mentioned only in its
SELECT part.
(Bug#24491)
Expressions of the form (a, b) IN (SELECT a, MIN(b)
FROM t GROUP BY a) could produce incorrect results
when column a of table t
contained NULL values while column
b did not.
(Bug#24420)
If a prepared statement accessed a view, access to the tables listed in the query after that view was checked in the security context of the view. (Bug#24404)
Attempts to access a MyISAM table with a
corrupt column definition caused a server crash.
(Bug#24401)
When opening a corrupted .frm file during a
query, the server crashes.
(Bug#24358)
A query using WHERE
could
cause the server to crash.
(Bug#24261)unsigned_column NOT IN
('negative_value')
Expressions of the form (a, b) IN (SELECT c, d
...) could produce incorrect results if
a, b, or both were
NULL.
(Bug#24127)
A FETCH statement using a cursor
on a table which was not in the table cache could sometimes
cause the server to crash.
(Bug#24117)
Queries that evaluate NULL IN (SELECT ... UNION SELECT
...) could produce an incorrect result
(FALSE instead of NULL).
(Bug#24085)
Hebrew-to-Unicode conversion failed for some characters. Definitions for the following Hebrew characters (as specified by the ISO/IEC 8859-8:1999) were added: LEFT-TO-RIGHT MARK (LRM), RIGHT-TO-LEFT MARK (RLM) (Bug#24037)
Some UPDATE statements were
slower than in previous versions when the search key could not
be converted to a valid value for the type of the search column.
(Bug#24035)
ISNULL(DATE(NULL)) and
ISNULL(CAST(NULL AS DATE))
erroneously returned false.
(Bug#23938)
Within a stored routine, accessing a declared routine variable
with PROCEDURE ANALYSE() caused a server
crash.
(Bug#23782)
When reading from the standard input on Windows, mysqlbinlog opened the input in text mode rather than binary mode and consequently misinterpreted some characters such as Control-Z. (Bug#23735)
For an InnoDB table with any ON
DELETE trigger,
TRUNCATE
TABLE mapped to DELETE
and activated triggers. Now a fast truncation occurs and
triggers are not activated. .
(Bug#23556)
The row count for MyISAM tables was not
updated properly, causing SHOW TABLE
STATUS to report incorrect values.
(Bug#23526)
With ONLY_FULL_GROUP_BY
enables, the server was too strict: Some expressions involving
only aggregate values were rejected as nonaggregate (for
example, MAX(a) –
MIN(a)).
(Bug#23417)
The arguments to the ENCODE() and
the DECODE() functions were not
printed correctly, causing problems in the output of
EXPLAIN EXTENDED and in view definitions.
(Bug#23409)
Some queries against INFORMATION_SCHEMA that
used subqueries failed. .
(Bug#23299)
readline detection did not work correctly on
NetBSD.
(Bug#23293)
If there was insufficient memory to store or update a blob
record in a MyISAM table then the table will
marked as crashed.
(Bug#23196)
LAST_INSERT_ID() was not reset to
0 if INSERT ...
SELECT inserted no rows.
(Bug#23170)
The number of setsockopt() calls performed
for reads and writes to the network socket was reduced to
decrease system call overhead.
(Bug#22943)
mysql_upgrade failed when called with a
--basedir path name containing spaces.
(Bug#22801)
SET lc_time_names = allowed only exact literal values, not expression
values.
(Bug#22647)value
The STDDEV() function returned a
positive value for data sets consisting of a single value.
(Bug#22555)
Storing values specified as hexadecimal values 64 or more bits
long in BIT(64),
BIGINT, or BIGINT
UNSIGNED columns did not raise any warning or error if
the value was out of range.
(Bug#22533)
SHOW COLUMNS reported some
NOT NULL columns as NULL.
(Bug#22377)
Type conversion errors during formation of index search conditions were not correctly checked, leading to incorrect query results. (Bug#22344)
The code for generating USE
statements for binary logging of CREATE
PROCEDURE statements resulted in confusing output from
mysqlbinlog for DROP
PROCEDURE statements.
(Bug#22043)
For the IF() and
COALESCE() function and
CASE expressions, large unsigned
integer values could be mishandled and result in warnings.
(Bug#22026)
SSL connections could hang at connection shutdown. (Bug#21781, Bug#24148)
When updating a table that used a JOIN of the
table itself (for example, when building trees) and the table
was modified on one side of the expression, the table would
either be reported as crashed or the wrong rows in the table
would be updated.
(Bug#21310)
Inserting DEFAULT into a column with no
default value could result in garbage in the column. Now the
same result occurs as when inserting NULL
into a NOT NULL column.
(Bug#20691)
A stored routine containing semicolon in its body could not be reloaded from a dump of a binary log. (Bug#20396)
SELECT ... FOR
UPDATE, SELECT
... LOCK IN SHARE MODE,
DELETE, and
UPDATE statements executed using
a full table scan were not releasing locks on rows that did not
satisfy the WHERE condition.
(Bug#20390)
On Windows, if the server was installed as a service, it did not auto-detect the location of the data directory. (Bug#20376)
The BUILD/check-cpu script did not recognize Celeron processors. (Bug#20061)
If a duplicate key value was present in the table,
INSERT ...
ON DUPLICATE KEY UPDATE reported a row count
indicating that a record was updated, even when no record
actually changed due to the old and new values being the same.
Now it reports a row count of zero.
(Bug#19978)
ORDER BY values of the
DOUBLE or
DECIMAL types could change the
result returned by a query.
(Bug#19690)
The readline library wrote to uninitialized
memory, causing mysql to crash.
(Bug#19474)
mysqltest incorrectly tried to retrieve result sets for some queries where no result set was available. (Bug#19410)
Use of already freed memory caused SSL connections to hang forever. (Bug#19209)
The server might fail to use an appropriate index for
DELETE when ORDER
BY, LIMIT, and a nonrestricting
WHERE are present.
(Bug#17711)
No warning was issued for use of the DATA
DIRECTORY or INDEX DIRECTORY table
options on a platform that does not support them.
(Bug#17498)
When a prepared statement failed during the prepare operation, the error code was not cleared when it was reused, even if the subsequent use was successful. (Bug#15518)
On Windows, the SLEEP() function
could sleep too long, especially after a change to the system
clock.
(Bug#14094, Bug#24686, Bug#17635)
mysqldump --order-by-primary failed if the primary key name was an identifier that required quoting. (Bug#13926)
To enable installation of MySQL RPMs on Linux systems running RHEL 4 (which includes SE-Linux) additional information was provided to specify some actions that are allowed to the MySQL binaries. (Bug#12676)
The presence of ORDER BY in a view definition
prevented the MERGE algorithm from being used
to resolve the view even if nothing else in the definition
required the TEMPTABLE algorithm.
(Bug#12122)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.32).
Functionality added or changed:
The --skip-thread-priority option
now is enabled by default for binary Mac OS X distributions. Use
of thread priorities degrades performance on Mac OS X.
(Bug#18526)
Added the
--disable-grant-options option
to configure. If configure
is run with this option, the
--bootstrap,
--skip-grant-tables, and
--init-file options for
mysqld are disabled and cannot be used. For
Windows, the configure.js script recognizes
the DISABLE_GRANT_OPTIONS flag, which has the
same effect.
Bugs fixed:
MySQL Cluster: Hosts in clusters with large numbers of nodes could experience excessive CPU usage while obtaining configuration data. (Bug#25711)
MySQL Cluster:
When a data node was shut down using the management client
STOP command, a connection event
(NDB_LE_Connected) was logged instead of a
disconnection event (NDB_LE_Disconnected).
(Bug#22773)
Cluster API:
Invoking the NdbTransaction::execute() method
using execution type Commit and abort option
AO_IgnoreError could lead to a crash of the
transaction coordinator (DBTC).
(Bug#25090)
Cluster API: A unique index lookup on a nonexistent tuple could lead to a data node timeout (error 4012). (Bug#25059)
Referencing an ambiguous column alias in an expression in the
ORDER BY clause of a query caused the server
to crash.
(Bug#25427)
Using a view in combination with a USING
clause caused column aliases to be ignored.
(Bug#25106)
A multiple-table DELETE QUICK could sometimes
cause one of the affected tables to become corrupted.
(Bug#25048)
An assertion failed incorrectly for prepared statements that
contained a single-row uncorrelated subquery that was used as an
argument of the IS NULL predicate.
(Bug#25027)
Optimizations that are legal only for subqueries without tables
and WHERE conditions were applied for any
subquery without tables.
(Bug#24670)
Some joins in which one of the joined tables was a view could return erroneous results or crash the server. (Bug#24345)
A view was not handled correctly if the
SELECT part contained “
\Z ”.
(Bug#24293)
The server was built even when configure was
run with the --without-server
option.
(Bug#23973)
See also Bug#32898.
OPTIMIZE TABLE tried to sort
R-tree indexes such as spatial indexes, although this is not
possible (see Section 12.5.2.5, “OPTIMIZE TABLE Syntax”).
(Bug#23578)
User-defined variables could consume excess memory, leading to a
crash caused by the exhaustion of resources available to the
MEMORY storage engine, due to the fact that
this engine is used by MySQL for variable storage and
intermediate results of GROUP BY queries.
Where SET had been used, such a condition
could instead give rise to the misleading error message
You may only use constant expressions with
SET, rather than Out of memory (Needed
NNNNNN bytes).
(Bug#23443)
A table created with the ROW_FORMAT = FIXED
table option lost the option if an index was added or dropped
with CREATE INDEX or
DROP INDEX.
(Bug#23404)
A deadlock could occur, with the server hanging on
Closing tables, with a sufficient number of
concurrent INSERT DELAYED,
FLUSH TABLES,
and ALTER TABLE operations.
(Bug#23312)
A compressed MyISAM table that became
corrupted could crash myisamchk and possibly
the MySQL Server.
(Bug#23139)
Changing the value of MI_KEY_BLOCK_LENGTH in
myisam.h and recompiling MySQL resulted in
a myisamchk that saw existing
MyISAM tables as corrupt.
(Bug#22119)
A crash of the MySQL Server could occur when unpacking a
BLOB column from a row in a
corrupted MyISAM table. This could happen when trying to repair
a table using either REPAIR TABLE
or myisamchk; it could also happen when
trying to access such a “broken” row using
statements like SELECT if the
table was not marked as crashed.
(Bug#22053)
The FEDERATED storage engine did not support
the euckr character set.
(Bug#21556)
mysqld_error.h was not installed when only
the client libraries were built.
(Bug#21265)
InnoDB: During a restart of the MySQL Server
that followed the creation of a temporary table using the
InnoDB storage engine, MySQL failed to clean
up in such a way that InnoDB still attempted
to find the files associated with such tables.
(Bug#20867)
InnoDB: Optimizations removed in MySQL 5.0.28
were re-enabled except for files under the
innobase/mem directory. (This is a
fine-tuning of optimization disabling.)
(Bug#19424)
Some
CASE
statements inside stored routines could lead to excessive
resource usage or a crash of the server.
(Bug#19194, Bug#24854)
Instance Manager could crash during shutdown. (Bug#19044)
The FEDERATED storage engine did not support
the utf8 character set.
(Bug#17044)
The optimizer removes expressions from GROUP
BY and DISTINCT clauses if they
happen to participate in
predicates of the
expression =
constantWHERE clause, the idea being that, if the
expression is equal to a constant, then it cannot take on
multiple values. However, for predicates where the expression
and the constant item are of different result types (for
example, when a string column is compared to 0), this is not
valid, and can lead to invalid results in such cases. The
optimizer now performs an additional check of the result types
of the expression and the constant; if their types differ, then
the expression is not removed from the GROUP
BY list.
(Bug#15881)
Dropping a user-defined function sometimes did not remove the
UDF entry from the mysql.proc table.
(Bug#15439)
Inserting a row into a table without specifying a value for a
BINARY(
column caused the column to be set to spaces, not zeroes.
(Bug#14171)N) NOT NULL
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.27.
This version of MySQL Community Server has been released as a source tarball only; there are no binaries built by MySQL.
Functionality added or changed:
Incompatible Change:
InnoDB rolls back only the last statement on
a transaction timeout. A new option,
--innodb_rollback_on_timeout,
causes InnoDB to abort and roll back the
entire transaction if a transaction timeout occurs (the same
behavior as in MySQL 5.0.13 and earlier).
(Bug#24200)
Incompatible Change:
The prepared_stmt_count system
variable has been converted to the
Prepared_stmt_count global
status variable (viewable with the
SHOW GLOBAL
STATUS statement).
(Bug#23159)
MySQL Cluster:
Setting the configuration parameter
LockPagesInMainMemory had no effect.
(Bug#24461)
MySQL Cluster:
The ndb_config utility now accepts
-c as a short form of the
--ndb-connectstring option.
(Bug#22295)
MySQL Cluster:
Added the --bind-address option for
ndbd. This allows a data node process to be
bound to a specific network interface.
(Bug#22195)
MySQL Cluster:
It is now possible to create a unique hashed index on a column
that is not defined as NOT NULL.
This change applies only to tables using the
NDB storage engine.
Unique indexes on columns in NDB
tables do not store null values because they are mapped to
primary keys in an internal index table (and primary keys cannot
contain nulls).
Normally, an additional ordered index is created when one
creates unique indexes on NDB table
columns; this can be used to search for NULL
values. However, if USING HASH is specified
when such an index is created, no ordered index is created.
The reason for permitting unique hash indexes with null values
is that, in some cases, the user wants to save space if a large
number of records are pre-allocated but not fully initialized.
This also assumes that the user will not
try to search for null values. Since MySQL does not support
indexes that are not allowed to be searched in some cases, the
NDB storage engine uses a full
table scan with pushed conditions for the referenced index
columns to return the correct result.
A warning is returned if one creates a unique nullable hash
index, since the query optimizer should be provided a hint not
to use it with NULL values if this can be
avoided.
(Bug#21507)
MySQL Cluster:
The Ndb_number_of_storage_nodes system
variable was renamed to
Ndb_number_of_data_nodes.
(Bug#20848)
MySQL Cluster:
The HELP command in the Cluster
management client now provides command-specific help. For
example, HELP RESTART in
ndb_mgm provides detailed information about
the RESTART command.
(Bug#19620)
DROP TRIGGER now supports an
IF EXISTS clause.
(Bug#23703)
The Com_create_user status variable was added
(for counting CREATE USER
statements).
(Bug#22958)
The --memlock option relies on
system calls that are unreliable on some operating systems. If a
crash occurs, the server now checks whether
--memlock was specified and if so
issues some information about possible workarounds.
(Bug#22860)
If the user specified the server options
--max-connections= or
N
--table-cache=, a warning would be given in some cases that some
values were recalculated, with the result that
M
--table-cache could be assigned
greater value.
In such cases, both the warning and the increase in the
--table-cache value were
completely harmless. Note also that it is not possible for the
MySQL Server to predict or to control limitations on the maximum
number of open files, since this is determined by the operating
system.
The value of --table-cache is no
longer increased automatically, and a warning is now given only
if some values had to be decreased due to operating system
limits.
(Bug#21915)
For the CALL statement, stored
procedures that take no arguments now can be invoked without
parentheses. That is, CALL p() and
CALL p are equivalent.
(Bug#21462)
mysql_upgrade now passes all the parameters
specified on the command line to both
mysqlcheck and mysql using
the upgrade_defaults file.
(Bug#20100)
SHOW STATUS is no longer logged
to the slow query log.
(Bug#19764)
mysqldump --single-transaction now uses
START TRANSACTION /*!40100 WITH CONSISTENT SNAPSHOT
*/ rather than
BEGIN to start
a transaction, so that a consistent snapshot will be used on
those servers that support it.
(Bug#19660)
The bundled yaSSL library was upgraded to version 1.5.0.
Bugs fixed:
MySQL Cluster: The failure of a data node failure during a schema operation could lead to additional node failures. (Bug#24752)
MySQL Cluster: A committed read could be attempted before a data node had time to connect, causing a timeout error. (Bug#24717)
MySQL Cluster: Sudden disconnection of an SQL or data node could lead to shutdown of data nodes with the error failed ndbrequire. (Bug#24447)
MySQL Cluster: ndb_config failed when trying to use 2 management servers and node IDs. (Bug#23887)
MySQL Cluster: Backup of a cluster failed if there were any tables with 128 or more columns. (Bug#23502)
MySQL Cluster: Cluster backups failed when there were more than 2048 schema objects in the cluster. (Bug#23499)
MySQL Cluster:
The management client command ALL DUMP 1000
would cause the cluster to crash if data nodes were connected to
the cluster but not yet fully started.
(Bug#23203)
MySQL Cluster:
INSERT ...
ON DUPLICATE KEY UPDATE on an
NDB table could lead to deadlocks
and memory leaks.
(Bug#23200)
MySQL Cluster: (NDB API): Inacivity timeouts for scans were not correctly handled. (Bug#23107)
MySQL Cluster: If a node restart could not be performed from the REDO log, no node takeover took place. This could cause partitions to be left empty during a system restart. (Bug#22893)
MySQL Cluster: Multiple node restarts in rapid succession could cause a system restart to fail , or induce a race condition. (Bug#22892, Bug#23210)
MySQL Cluster:
(NDB API): Attempting to read a nonexistent tuple using
Commit mode for
NdbTransaction::execute() caused node
failures.
(Bug#22672)
MySQL Cluster:
The --help output from
NDB binaries did not include
file-related options.
(Bug#21994)
MySQL Cluster: (NDB API): Scans closed before being executed were still placed in the send queue. (Bug#21941)
MySQL Cluster: A scan timeout returned Error 4028 (Node failure caused abort of transaction) instead of Error 4008 (Node failure caused abort of transaction...). (Bug#21799)
MySQL Cluster:
The node recovery algorithm was missing a version check for
tables in the ALTER_TABLE_COMMITTED state (as
opposed to the TABLE_ADD_COMMITTED state,
which has the version check). This could cause inconsistent
schemas across nodes following node recovery.
(Bug#21756)
MySQL Cluster:
The output for the --help option used with
NDB executable programs (such as
ndbd, ndb_mgm,
ndb_restore, ndb_config,
and others mentioned in
Section 17.6, “MySQL Cluster Programs”) referred to the
Ndb.cfg file, instead of to
my.cnf.
(Bug#21585)
MySQL Cluster: Partition distribution keys were updated only for the primary and starting replicas during node recovery. This could lead to node failure recovery for clusters having an odd number of replicas.
For best results, use values for
NumberOfReplicas that are even powers of 2.
MySQL Cluster: The ndb_mgm management client did not set the exit status on errors, always returning 0 instead. (Bug#21530)
MySQL Cluster: Cluster logs were not rotated following the first rotation cycle. (Bug#21345)
MySQL Cluster:
When inserting a row into an NDB
table with a duplicate value for a nonprimary unique key, the
error issued would reference the wrong key.
(Bug#21072)
MySQL Cluster:
Condition pushdown did not work correctly with
DATETIME columns.
(Bug#21056)
MySQL Cluster: Under some circumstances, local checkpointing would hang, keeping any unstarted nodes from being started. (Bug#20895)
MySQL Cluster:
Using an invalid node ID with the management client
STOP command could cause
ndb_mgm to hang.
(Bug#20575)
MySQL Cluster: Data nodes added while the cluster was running in single user mode were all assigned node ID 0, which could later cause multiple node failures. Adding nodes while in single user mode is no longer possible. (Bug#20395)
MySQL Cluster:
In some cases where SELECT COUNT(*) from an
NDB table should have yielded an
error, MAX_INT was returned instead.
(Bug#19914)
MySQL Cluster: Following the restart of a management node, the Cluster management client did not automatically reconnect. (Bug#19873)
MySQL Cluster:
Error messages given when trying to make online changes to
parameters such as NoOfReplicas that can only
be changed via a complete shutdown and restart of the cluster
did not indicate the true nature of the problem.
(Bug#19787)
MySQL Cluster: ndb_restore did not always make clear that it had recovered successfully from temporary errors while restoring a cluster backup. (Bug#19651)
MySQL Cluster:
In rare situations with resource shortages, a crash could result
from insufficient IndexScanOperations.
(Bug#19198)
MySQL Cluster: ndb_mgm -e show | head would hang after displaying the first 10 lines of output. (Bug#19047)
MySQL Cluster: The error returned by the cluster when too many nodes were defined did not make clear the nature of the problem. (Bug#19045)
MySQL Cluster:
A unique constraint violation was not ignored by an
UPDATE IGNORE statement when the constraint
violation occurred on a nonprimary key.
(Bug#18487, Bug#24303)
MySQL Cluster:
The ndb_config utility did not perform host
lookups correctly when using the --host option
(Bug#17582)
MySQL Cluster: A problem with takeover during a system restart caused ordered indexes to be rebuilt incorrectly. (Bug#15303)
Replication: Changes to character set variables prior to an action on a replication-ignored table were forgotten by slave servers. (Bug#22877)
Replication: On slave servers, transactions that exceeded the lock wait timeout failed to roll back properly. (Bug#20697)
Replication: Column names were not quoted properly for replicated views. (Bug#19736)
Replication:
SQL statements close to the size of
max_allowed_packet could
produce binary log events larger than
max_allowed_packet that could
not be read by slave servers.
(Bug#19402)
Replication:
Slave servers would retry the execution of an SQL statement an
infinite number of times, ignoring the value
SLAVE_TRANSACTION_RETRIES when using the NDB
engine.
(Bug#16228)
Replication:
Transient errors in replication from master to slave may trigger
multiple Got fatal error 1236: 'binlog truncated in the
middle of event' errors on the slave.
(Bug#4053)
Cluster API:
Using BIT values with any of the
comparison methods of the NdbScanFilter class
caused data nodes to fail.
(Bug#24503)
Cluster API: Some MGM API function calls could yield incorrect return values in certain cases where the cluster was operating under a very high load, or experienced timeouts in inter-node communications. (Bug#24011)
Cluster API:
The NdbOperation::getBlobHandle() method,
when called with the name of a nonexistent column, caused a
segmentation fault.
(Bug#21036)
Cluster API: When multiple processes or threads in parallel performed the same ordered scan with exclusive lock and updated the retrieved records, the scan could skip some records, which as a result were not updated. (Bug#20446)
The REPEAT() function could
return NULL when passed a column for the
count argument.
(Bug#24947)
mysql_upgrade failed if the
--password (or -p) option
was given.
(Bug#24896)
With innodb_file_per_table
enabled, InnoDB displayed incorrect file
times in the output from SHOW TABLE
STATUS.
(Bug#24712)
ALTER ENABLE KEYS or ALTER TABLE
DISABLE KEYS combined with another
ALTER TABLE option other than
RENAME TO did nothing. In addition, if ALTER
TABLE was used on a table having disabled keys, the keys of the
resulting table were enabled.
(Bug#24395)
The InnoDB mutex structure was simplified to
reduce memory load.
(Bug#24386)
The --extern option for
mysql-test-run.pl did not function correctly.
(Bug#24354)
Foreign key identifiers for InnoDB tables
could not contain certain characters.
(Bug#24299)
The mysql.server script used the source command, which is less portable than the . command; it now uses . instead. (Bug#24294)
ALTER TABLE statements that
performed both RENAME TO and
{ENABLE|DISABLE} KEYS operations caused a
server crash.
(Bug#24219)
The loose index scan optimization for GROUP
BY with MIN or
MAX was not applied within other queries,
such as CREATE
TABLE ... SELECT ..., INSERT ... SELECT
..., or in the FROM clauses of
subqueries.
(Bug#24156)
There was a race condition in the InnoDB
fil_flush_file_spaces() function.
(Bug#24089)
This regression was introduced by Bug#15653.
Subqueries for which a pushed-down condition did not produce exactly one key field could cause a server crash. (Bug#24056)
The size of MEMORY tables and internal
temporary tables was limited to 4GB on 64-bit Windows systems.
(Bug#24052)
yaSSL-related memory leaks were detected by Valgrind. (Bug#23981)
The internal SQL interpreter of InnoDB placed
an unnecessary lock on the supremum record with
innodb_locks_unsafe_for_binlog
enabled. This caused an assertion failure when
InnoDB was built with debugging enabled.
(Bug#23769)
ROW_COUNT() did not work properly
as an argument to a stored procedure.
(Bug#23760)
LAST_DAY('0000-00-00') could
cause a server crash.
(Bug#23653)
A trigger that invoked a stored function could cause a server crash when activated by different client connections. (Bug#23651)
The stack size for NetWare binaries was increased to 128KB to prevent problems caused by insufficient stack size. (Bug#23504)
If elements in a nontop-level IN subquery
were accessed by an index and the subquery result set included a
NULL value, the quantified predicate that
contained the subquery was evaluated to NULL
when it should return a non-NULL value.
(Bug#23478)
When applying the
group_concat_max_len limit,
GROUP_CONCAT() could truncate
multi-byte characters in the middle.
(Bug#23451)
MySQL 5.0.26 introduced an ABI incompatibility, which this release reverts. Programs compiled against 5.0.26 are not compatible with any other version and must be recompiled. (Bug#23427)
M % 0NULL, but (
evaluated to
false.
(Bug#23411)M % 0) IS NULL
mysql_affected_rows() could
return values different from
mysql_stmt_affected_rows() for
the same sequence of statements.
(Bug#23383)
For not-yet-authenticated connections, the
Time column in SHOW
PROCESSLIST was a random value rather than
NULL.
(Bug#23379)
Accuracy was improved for comparisons between
DECIMAL columns and numbers
represented as strings.
(Bug#23260)
MySQL failed to build on Linux/Alpha. (Bug#23256)
This regression was introduced by Bug#21250.
If COMPRESS() returned
NULL, subsequent invocations of
COMPRESS() within a result set or
within a trigger also returned NULL.
(Bug#23254)
Calculation of COUNT(DISTINCT),
AVG(DISTINCT), or
SUM(DISTINCT) when they are
referenced more than once in a single query with GROUP
BY could cause a server crash.
(Bug#23184)
Insufficient memory
(myisam_sort_buffer_size) could
cause a server crash for several operations on
MyISAM tables: repair table, create index by
sort, repair by sort, parallel repair, bulk insert.
(Bug#23175)
The column default value in the output from
SHOW COLUMNS or SELECT
FROM INFORMATION_SCHEMA.COLUMNS was truncated to 64
characters.
(Bug#23037)
mysql did not check for errors when fetching data during result set printing. (Bug#22913)
InnoDB exhibited thread thrashing with more
than 50 concurrent connections under an update-intensive
workload.
(Bug#22868)
The return value from my_seek() was ignored.
(Bug#22828)
The optimizer failed to use equality propagation for
BETWEEN and IN
predicates with string arguments.
(Bug#22753)
The Handler_rollback status
variable sometimes was incremented when no rollback had taken
place.
(Bug#22728)
The Host column in SHOW
PROCESSLIST output was blank when the server was
started with the
--skip-grant-tables option.
(Bug#22723)
If a table contains an AUTO_INCREMENT column,
inserting into an insertable view on the table that does not
include the AUTO_INCREMENT column should not
change the value of
LAST_INSERT_ID(), because the
side effects of inserting default values into columns not part
of the view should not be visible. MySQL was incorrectly setting
LAST_INSERT_ID() to zero.
(Bug#22584)
Queries using a column alias in an expression as part of an
ORDER BY clause failed, an example of such a
query being SELECT mycol + 1 AS mynum FROM mytable
ORDER BY 30 - mynum.
(Bug#22457)
Using EXPLAIN caused a server
crash for queries that selected from
INFORMATION_SCHEMA in a subquery in the
FROM clause.
(Bug#22413)
Instance Manager had a race condition involving mysqld PID file removal. (Bug#22379)
A server crash occurred when using LOAD
DATA to load a table containing a NOT
NULL spatial column, when the statement did not load
the spatial column. Now a NULL supplied to NOT NULL
column error occurs.
(Bug#22372)
The optimizer used the ref
join type rather than eq_ref
for a simple join on strings.
(Bug#22367)
Some queries that used MAX() and
GROUP BY could incorrectly return an empty
result.
(Bug#22342)
DATE_ADD() requires complete
dates with no “zero” parts, but sometimes did not
return NULL when given such a date.
(Bug#22229)
If an init_connect SQL
statement produced an error, the connection was silently
terminated with no error message. Now the server writes a
warning to the error log.
(Bug#22158)
Some small double precision numbers (such as
1.00000001e-300) that should have been
accepted were truncated to zero.
(Bug#22129)
For a nonexistent table, DROP TEMPORARY TABLE
failed with an incorrect error message if
read_only was enabled.
(Bug#22077)
Trailing spaces were not removed from Unicode
CHAR column values when used in
indexes. This resulted in excessive usage of storage space, and
could affect the results of some ORDER BY
queries that made use of such indexes.
When upgrading, it is necessary to re-create any existing
indexes on Unicode CHAR columns
in order to take advantage of the fix. This can be done by
using a REPAIR TABLE statement
on each affected table.
The code for generating USE
statements for binary logging of CREATE
PROCEDURE statements resulted in confusing output from
mysqlbinlog for DROP
PROCEDURE statements.
(Bug#22043)
STR_TO_DATE() returned
NULL if the format string contained a space
following a nonformat character.
(Bug#22029)
Use of a DES-encrypted SSL certificate file caused a server crash. (Bug#21868)
Use of PREPARE with a
CREATE PROCEDURE statement that
contained a syntax error caused a server crash.
(Bug#21856)
Adding a day, month, or year interval to a
DATE value produced a
DATE, but adding a week interval
produced a DATETIME value. Now
all produce a DATE value.
(Bug#21811)
In some cases, the parser failed to distinguish a user-defined function from a stored function. (Bug#21809)
Use of a subquery that invoked a function in the column list of the outer query resulted in a memory leak. (Bug#21798)
Inserting a default or invalid value into a spatial column could
fail with Unknown error rather than a more
appropriate error.
(Bug#21790)
It was possible to use DATETIME
values whose year, month, and day parts were all zeroes but
whose hour, minute, and second parts contained nonzero values,
an example of such an illegal
DATETIME being
'0000-00-00 11:23:45'.
This fix was reverted in MySQL 5.0.40.
See also Bug#25301.
yaSSL crashed on pre-Pentium Intel CPUs. (Bug#21765)
Evaluation of subqueries that require the filesort algorithm
were allocating and freeing the
sort_buffer_size buffer many
times, resulting in slow performance. Now the buffer is
allocated once and reused.
(Bug#21727)
Through the C API, the member strings in
MYSQL_FIELD for a query that contains
expressions may return incorrect results.
(Bug#21635)
Selecting from a MERGE table could result in
a server crash if the underlying tables had fewer indexes than
the MERGE table itself.
(Bug#21617, Bug#22937)
View columns were always handled as having implicit derivation,
leading to illegal mix of collation errors
for some views in UNION
operations. Now view column derivation comes from the original
expression given in the view definition.
(Bug#21505)
InnoDB crashed while performing XA recovery
of prepared transactions.
(Bug#21468)
INET_ATON() returned a signed
BIGINT value, not an unsigned
value.
(Bug#21466)
After FLUSH TABLES WITH
READ LOCK followed by
UNLOCK
TABLES, attempts to drop or alter a stored routine
failed with an error that the routine did not exist, and
attempts to execute the routine failed with a lock conflict
error.
(Bug#21414)
It was possible to set the backslash character (“
\ ”) as the delimiter character using
DELIMITER, but not actually possible to use
it as the delimiter.
(Bug#21412)
For multiple-table UPDATE
statements, storage engines were not notified of duplicate-key
errors.
(Bug#21381)
Within a prepared statement, SELECT (COUNT(*) =
1) (or similar use of other aggregate functions) did
not return the correct result for statement re-execution.
(Bug#21354)
It was possible for a stored routine with a
non-latin1 name to cause a stack overrun.
(Bug#21311)
Certain malformed INSERT
statements could crash the mysql client.
(Bug#21142)
Creating a TEMPORARY table with the same name
as an existing table that was locked by another client could
result in a lock conflict for DROP TEMPORARY
TABLE because the server unnecessarily tried to
acquire a name lock.
(Bug#21096)
Incorrect results could be obtained from re-execution of a
parametrized prepared statement or a stored routine with a
SELECT that uses LEFT
JOIN with a second table having only one row.
(Bug#21081)
Within a stored routine, a view definition cannot refer to routine parameters or local variables. However, an error did not occur until the routine was called. Now it occurs during parsing of the routine creation statement.
A side effect of this fix is that if you have already created
such routines, and error will occur if you execute
SHOW CREATE PROCEDURE or
SHOW CREATE FUNCTION. You
should drop these routines because they are erroneous.
In mysql, invoking connect
or \r with very long
db_name or
host_name parameters caused buffer
overflow.
(Bug#20894)
SHOW VARIABLES truncated the
Value field to 256 characters.
(Bug#20862)
Selecting into variables sometimes returned incorrect wrong results. (Bug#20836)
WITH ROLLUP could group unequal values.
(Bug#20825)
Range searches on columns with an index prefix could miss records. (Bug#20732)
Inserting DEFAULT into a column with no
default value could result in garbage in the column. Now the
same result occurs as when inserting NULL
into a NOT NULL column.
(Bug#20691)
An UPDATE that referred to a key
column in the WHERE clause and activated a
trigger that modified the column resulted in a loop.
(Bug#20670)
CONCURRENT did not work correctly for
LOAD DATA
INFILE.
(Bug#20637)
mysql_fix_privilege_tables.sql altered the
table_privs.table_priv column to contain too
few privileges, causing loss of the CREATE
VIEW and SHOW VIEW
privileges.
(Bug#20589)
LIKE searches failed for indexed
utf8 character columns.
(Bug#20471)
With lower_case_table_names set
to 1, SHOW CREATE TABLE printed
incorrect output for table names containing Turkish I (LATIN
CAPITAL LETTER I WITH DOT ABOVE).
(Bug#20404)
A query with a subquery that references columns of a view from
the outer SELECT could return an
incorrect result if used from a prepared statement.
(Bug#20327)
For queries that select from a view, the server was returning
MYSQL_FIELD metadata inconsistently for view
names and table names. For view columns, the server now returns
the view name in the table field and, if the
column selects from an underlying table, the table name in the
org_table field.
(Bug#20191)
Invalidating the query cache caused a server crash for
INSERT INTO ...
SELECT statements that selected from a view.
(Bug#20045)
With sql_mode = TRADITIONAL, MySQL
incorrectly aborted on warnings within stored routines and
triggers.
(Bug#20028)
Unsigned BIGINT values treated as
signed values by the MOD()
function.
(Bug#19955)
Compiling PHP 5.1 with the MySQL static libraries failed on some versions of Linux. (Bug#19817)
The DELIMITER statement did not work
correctly when used in an SQL file run using the
SOURCE statement.
(Bug#19799)
mysqldump --xml produced invalid XML for
BLOB data.
(Bug#19745)
For a cast of a DATETIME value
containing microseconds to
DECIMAL, the microseconds part
was truncated without generating a warning. Now the microseconds
part is preserved.
(Bug#19491)
InnoDB: Reduced optimization level for
Windows 64 builds to handle possible memory overrun.
(Bug#19424)
VARBINARY column values inserted
on a MySQL 4.1 server had trailing zeroes following upgrade to
MySQL 5.0 or later.
(Bug#19371)
FLUSH INSTANCES in Instance Manager triggered
an assertion failure.
(Bug#19368)
For a debug server, a reference to an undefined user variable in
a prepared statement executed with
EXECUTE caused an assertion
failure.
(Bug#19356)
The server could send incorrect column count information to the client for queries that produce a larger number of columns than can fit in a two-byte number. (Bug#19216)
Within a trigger for a base table, selecting from a view on that base table failed. (Bug#19111)
The value of the warning_count
system variable was not being calculated correctly (also
affecting SHOW COUNT(*) WARNINGS).
(Bug#19024)
For some problems relating to character set conversion or
incorrect string values for
INSERT or
UPDATE, the server was reporting
truncation or length errors instead.
(Bug#18908)
DELETE IGNORE could hang for foreign key
parent deletes.
(Bug#18819)
Constant expressions and some numeric constants used as input parameters to user-defined functions were not treated as constants. (Bug#18761)
InnoDB used table locks (not row locks)
within stored functions.
(Bug#18077)
myisampack wrote to unallocated memory, causing a crash. (Bug#17951)
FLUSH LOGS or
mysqladmin flush-logs caused a server crash
if the binary log was not open.
(Bug#17733)
mysql_fix_privilege_tables did not accept a password containing embedded space or apostrophe characters. (Bug#17700)
mysql would lose its connection to the server if its standard output was not writable. (Bug#17583)
Attempting to use a view containing DEFINER
information for a nonexistent user resulted in an error message
that revealed the definer account. Now the definer is revealed
only to superusers. Other users receive only an access
denied message.
(Bug#17254)
mysql-test-run did not work correctly for RPM-based installations. (Bug#17194)
IN() and
CHAR() can return
NULL, but did not signal that to the query
processor, causing incorrect results for
IS NULL
operations.
(Bug#17047)
A client library crash was caused by executing a statement such
as SELECT * FROM t1 PROCEDURE ANALYSE() using
a server side cursor on a table t1 that does
not have the same number of columns as the output from
PROCEDURE ANALYSE().
(Bug#17039)
The WITH CHECK OPTION for a view failed to
prevent storing invalid column values for
UPDATE statements.
(Bug#16813)
InnoDB showed substandard performance with
multiple queries running concurrently.
(Bug#15815)
ALTER TABLE was not able to
rename a view.
(Bug#14959)
Statements such as DROP PROCEDURE
and DROP VIEW were written to the
binary log too late due to a race condition.
(Bug#14262)
A literal string in a GROUP BY clause could
be interpreted as a column name.
(Bug#14019)
Instance Manager didn't close the client socket file when starting a new mysqld instance. mysqld inherited the socket, causing clients connected to Instance Manager to hang. (Bug#12751)
Entries in the slow query log could have an incorrect
Rows_examined value.
(Bug#12240)
Warnings were generated when explicitly casting a character to a
number (for example, CAST('x' AS
SIGNED)), but not for implicit conversions in simple
arithmetic operations (such as 'x' + 0). Now
warnings are generated in all cases.
(Bug#11927)
Lack of validation for input and output
TIME values resulted in several
problems: SEC_TO_TIME() in some
cases did not clip large values to the
TIME range appropriately;
SEC_TO_TIME() treated
BIGINT UNSIGNED values as signed; only
truncation warnings were produced when both truncation and
out-of-range TIME values
occurred.
(Bug#11655, Bug#20927)
Metadata for columns calculated from scalar subqueries was limited to integer, double, or string, even if the actual type of the column was different. (Bug#11032)
Several string functions could return incorrect results when given very large length arguments. (Bug#10963)
FROM_UNIXTIME() did not accept
arguments up to POWER(2,31)-1,
which it had previously.
(Bug#9191)
Subqueries of the form NULL IN (SELECT ...)
returned invalid results.
(Bug#8804, Bug#23485)
OPTIMIZE TABLE with
myisam_repair_threads > 1
could result in MyISAM table corruption.
(Bug#8283)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.30).
Functionality added or changed:
Incompatible Change:
InnoDB rolls back only the last statement on
a transaction timeout. A new option,
--innodb_rollback_on_timeout,
causes InnoDB to abort and roll back the
entire transaction if a transaction timeout occurs (the same
behavior as in MySQL 5.0.13 and earlier).
(Bug#24200)
Incompatible Change:
The prepared_stmt_count system
variable has been converted to the
Prepared_stmt_count global
status variable (viewable with the
SHOW GLOBAL
STATUS statement).
(Bug#23159)
MySQL Cluster:
Setting the configuration parameter
LockPagesInMainMemory had no effect.
(Bug#24461)
MySQL Cluster:
It is now possible to create a unique hashed index on a column
that is not defined as NOT NULL.
This change applies only to tables using the
NDB storage engine.
Unique indexes on columns in NDB
tables do not store null values because they are mapped to
primary keys in an internal index table (and primary keys cannot
contain nulls).
Normally, an additional ordered index is created when one
creates unique indexes on NDB table
columns; this can be used to search for NULL
values. However, if USING HASH is specified
when such an index is created, no ordered index is created.
The reason for permitting unique hash indexes with null values
is that, in some cases, the user wants to save space if a large
number of records are pre-allocated but not fully initialized.
This also assumes that the user will not
try to search for null values. Since MySQL does not support
indexes that are not allowed to be searched in some cases, the
NDB storage engine uses a full
table scan with pushed conditions for the referenced index
columns to return the correct result.
A warning is returned if one creates a unique nullable hash
index, since the query optimizer should be provided a hint not
to use it with NULL values if this can be
avoided.
(Bug#21507)
DROP TRIGGER now supports an
IF EXISTS clause.
(Bug#23703)
The Com_create_user status variable was added
(for counting CREATE USER
statements).
(Bug#22958)
The --memlock option relies on
system calls that are unreliable on some operating systems. If a
crash occurs, the server now checks whether
--memlock was specified and if so
issues some information about possible workarounds.
(Bug#22860)
mysqldump now accepts the
--debug-info option, which displays debugging
information and memory and CPU usage statistics at program exit.
The bundled yaSSL library was upgraded to version 1.5.0.
Bugs fixed:
MySQL Cluster: The failure of a data node failure during a schema operation could lead to additional node failures. (Bug#24752)
MySQL Cluster: A committed read could be attempted before a data node had time to connect, causing a timeout error. (Bug#24717)
MySQL Cluster: Sudden disconnection of an SQL or data node could lead to shutdown of data nodes with the error failed ndbrequire. (Bug#24447)
MySQL Cluster: ndb_config failed when trying to use 2 management servers and node IDs. (Bug#23887)
MySQL Cluster:
If the value set for MaxNoOfAttributes is
excessive, a suitable error message is now returned.
(Bug#19352)
MySQL Cluster:
A unique constraint violation was not ignored by an
UPDATE IGNORE statement when the constraint
violation occurred on a nonprimary key.
(Bug#18487, Bug#24303)
Replication: Changes to character set variables prior to an action on a replication-ignored table were forgotten by slave servers. (Bug#22877)
Replication: On slave servers, transactions that exceeded the lock wait timeout failed to roll back properly. (Bug#20697)
Replication:
SQL statements close to the size of
max_allowed_packet could
produce binary log events larger than
max_allowed_packet that could
not be read by slave servers.
(Bug#19402)
Replication:
Slave servers would retry the execution of an SQL statement an
infinite number of times, ignoring the value
SLAVE_TRANSACTION_RETRIES when using the NDB
engine.
(Bug#16228)
Cluster API:
Using BIT values with any of the
comparison methods of the NdbScanFilter class
caused data nodes to fail.
(Bug#24503)
Cluster API: Some MGM API function calls could yield incorrect return values in certain cases where the cluster was operating under a very high load, or experienced timeouts in inter-node communications. (Bug#24011)
The REPEAT() function could
return NULL when passed a column for the
count argument.
(Bug#24947)
mysql_upgrade failed if the
--password (or -p) option
was given.
(Bug#24896)
With innodb_file_per_table
enabled, InnoDB displayed incorrect file
times in the output from SHOW TABLE
STATUS.
(Bug#24712)
ALTER ENABLE KEYS or ALTER TABLE
DISABLE KEYS combined with another
ALTER TABLE option other than
RENAME TO did nothing. In addition, if ALTER
TABLE was used on a table having disabled keys, the keys of the
resulting table were enabled.
(Bug#24395)
The InnoDB mutex structure was simplified to
reduce memory load.
(Bug#24386)
The --extern option for
mysql-test-run.pl did not function correctly.
(Bug#24354)
Foreign key identifiers for InnoDB tables
could not contain certain characters.
(Bug#24299)
The mysql.server script used the source command, which is less portable than the . command; it now uses . instead. (Bug#24294)
ALTER TABLE statements that
performed both RENAME TO and
{ENABLE|DISABLE} KEYS operations caused a
server crash.
(Bug#24219)
The loose index scan optimization for GROUP
BY with MIN or
MAX was not applied within other queries,
such as CREATE
TABLE ... SELECT ..., INSERT ... SELECT
..., or in the FROM clauses of
subqueries.
(Bug#24156)
Subqueries for which a pushed-down condition did not produce exactly one key field could cause a server crash. (Bug#24056)
The size of MEMORY tables and internal
temporary tables was limited to 4GB on 64-bit Windows systems.
(Bug#24052)
ROW_COUNT() did not work properly
as an argument to a stored procedure.
(Bug#23760)
LAST_DAY('0000-00-00') could
cause a server crash.
(Bug#23653)
A trigger that invoked a stored function could cause a server crash when activated by different client connections. (Bug#23651)
The stack size for NetWare binaries was increased to 128KB to prevent problems caused by insufficient stack size. (Bug#23504)
If elements in a nontop-level IN subquery
were accessed by an index and the subquery result set included a
NULL value, the quantified predicate that
contained the subquery was evaluated to NULL
when it should return a non-NULL value.
(Bug#23478)
When applying the
group_concat_max_len limit,
GROUP_CONCAT() could truncate
multi-byte characters in the middle.
(Bug#23451)
mysql_affected_rows() could
return values different from
mysql_stmt_affected_rows() for
the same sequence of statements.
(Bug#23383)
Accuracy was improved for comparisons between
DECIMAL columns and numbers
represented as strings.
(Bug#23260)
Calculation of COUNT(DISTINCT),
AVG(DISTINCT), or
SUM(DISTINCT) when they are
referenced more than once in a single query with GROUP
BY could cause a server crash.
(Bug#23184)
Queries using a column alias in an expression as part of an
ORDER BY clause failed, an example of such a
query being SELECT mycol + 1 AS mynum FROM mytable
ORDER BY 30 - mynum.
(Bug#22457)
Using EXPLAIN caused a server
crash for queries that selected from
INFORMATION_SCHEMA in a subquery in the
FROM clause.
(Bug#22413)
A server crash occurred when using LOAD
DATA to load a table containing a NOT
NULL spatial column, when the statement did not load
the spatial column. Now a NULL supplied to NOT NULL
column error occurs.
(Bug#22372)
DATE_ADD() requires complete
dates with no “zero” parts, but sometimes did not
return NULL when given such a date.
(Bug#22229)
Some small double precision numbers (such as
1.00000001e-300) that should have been
accepted were truncated to zero.
(Bug#22129)
For a nonexistent table, DROP TEMPORARY TABLE
failed with an incorrect error message if
read_only was enabled.
(Bug#22077)
Trailing spaces were not removed from Unicode
CHAR column values when used in
indexes. This resulted in excessive usage of storage space, and
could affect the results of some ORDER BY
queries that made use of such indexes.
When upgrading, it is necessary to re-create any existing
indexes on Unicode CHAR columns
in order to take advantage of the fix. This can be done by
using a REPAIR TABLE statement
on each affected table.
STR_TO_DATE() returned
NULL if the format string contained a space
following a nonformat character.
(Bug#22029)
In some cases, the parser failed to distinguish a user-defined function from a stored function. (Bug#21809)
Inserting a default or invalid value into a spatial column could
fail with Unknown error rather than a more
appropriate error.
(Bug#21790)
It was possible to use DATETIME
values whose year, month, and day parts were all zeroes but
whose hour, minute, and second parts contained nonzero values,
an example of such an illegal
DATETIME being
'0000-00-00 11:23:45'.
This fix was reverted in MySQL 5.0.40.
See also Bug#25301.
yaSSL crashed on pre-Pentium Intel CPUs. (Bug#21765)
Evaluation of subqueries that require the filesort algorithm
were allocating and freeing the
sort_buffer_size buffer many
times, resulting in slow performance. Now the buffer is
allocated once and reused.
(Bug#21727)
Through the C API, the member strings in
MYSQL_FIELD for a query that contains
expressions may return incorrect results.
(Bug#21635)
Using FLUSH
TABLES in one connection while another connection is
using HANDLER statements caused a
server crash.
This fix was reverted in MySQL 5.0.48
See also Bug#29474.
View columns were always handled as having implicit derivation,
leading to illegal mix of collation errors
for some views in UNION
operations. Now view column derivation comes from the original
expression given in the view definition.
(Bug#21505)
InnoDB crashed while performing XA recovery
of prepared transactions.
(Bug#21468)
INET_ATON() returned a signed
BIGINT value, not an unsigned
value.
(Bug#21466)
It was possible to set the backslash character (“
\ ”) as the delimiter character using
DELIMITER, but not actually possible to use
it as the delimiter.
(Bug#21412)
Selecting into variables sometimes returned incorrect wrong results. (Bug#20836)
CONCURRENT did not work correctly for
LOAD DATA
INFILE.
(Bug#20637)
mysql_fix_privilege_tables.sql altered the
table_privs.table_priv column to contain too
few privileges, causing loss of the CREATE
VIEW and SHOW VIEW
privileges.
(Bug#20589)
With lower_case_table_names set
to 1, SHOW CREATE TABLE printed
incorrect output for table names containing Turkish I (LATIN
CAPITAL LETTER I WITH DOT ABOVE).
(Bug#20404)
A query with a subquery that references columns of a view from
the outer SELECT could return an
incorrect result if used from a prepared statement.
(Bug#20327)
For queries that select from a view, the server was returning
MYSQL_FIELD metadata inconsistently for view
names and table names. For view columns, the server now returns
the view name in the table field and, if the
column selects from an underlying table, the table name in the
org_table field.
(Bug#20191)
Invalidating the query cache caused a server crash for
INSERT INTO ...
SELECT statements that selected from a view.
(Bug#20045)
Unsigned BIGINT values treated as
signed values by the MOD()
function.
(Bug#19955)
Compiling PHP 5.1 with the MySQL static libraries failed on some versions of Linux. (Bug#19817)
The DELIMITER statement did not work
correctly when used in an SQL file run using the
SOURCE statement.
(Bug#19799)
For a cast of a DATETIME value
containing microseconds to
DECIMAL, the microseconds part
was truncated without generating a warning. Now the microseconds
part is preserved.
(Bug#19491)
VARBINARY column values inserted
on a MySQL 4.1 server had trailing zeroes following upgrade to
MySQL 5.0 or later.
(Bug#19371)
The server could send incorrect column count information to the client for queries that produce a larger number of columns than can fit in a two-byte number. (Bug#19216)
For some problems relating to character set conversion or
incorrect string values for
INSERT or
UPDATE, the server was reporting
truncation or length errors instead.
(Bug#18908)
Constant expressions and some numeric constants used as input parameters to user-defined functions were not treated as constants. (Bug#18761)
myisampack wrote to unallocated memory, causing a crash. (Bug#17951)
FLUSH LOGS or
mysqladmin flush-logs caused a server crash
if the binary log was not open.
(Bug#17733)
mysql_fix_privilege_tables did not accept a password containing embedded space or apostrophe characters. (Bug#17700)
Attempting to use a view containing DEFINER
information for a nonexistent user resulted in an error message
that revealed the definer account. Now the definer is revealed
only to superusers. Other users receive only an access
denied message.
(Bug#17254)
IN() and
CHAR() can return
NULL, but did not signal that to the query
processor, causing incorrect results for
IS NULL
operations.
(Bug#17047)
Warnings were generated when explicitly casting a character to a
number (for example, CAST('x' AS
SIGNED)), but not for implicit conversions in simple
arithmetic operations (such as 'x' + 0). Now
warnings are generated in all cases.
(Bug#11927)
Metadata for columns calculated from scalar subqueries was limited to integer, double, or string, even if the actual type of the column was different. (Bug#11032)
Subqueries of the form NULL IN (SELECT ...)
returned invalid results.
(Bug#8804, Bug#23485)
This is a Service Pack release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.30).
Functionality added or changed:
Incompatible Change:
InnoDB rolls back only the last statement on
a transaction timeout. A new option,
--innodb_rollback_on_timeout,
causes InnoDB to abort and roll back the
entire transaction if a transaction timeout occurs (the same
behavior as in MySQL 5.0.13 and earlier).
(Bug#24200)
Bugs fixed:
Replication: A stored procedure, executed from a connection using a binary character set, and which wrote multibyte data, would write incorrectly escaped entries to the binary log. This caused syntax errors, and caused replication to fail. (Bug#23619, Bug#24492)
The loose index scan optimization for GROUP
BY with MIN or
MAX was not applied within other queries,
such as CREATE
TABLE ... SELECT ..., INSERT ... SELECT
..., or in the FROM clauses of
subqueries.
(Bug#24156)
The size of MEMORY tables and internal
temporary tables was limited to 4GB on 64-bit Windows systems.
(Bug#24052)
Accuracy was improved for comparisons between
DECIMAL columns and numbers
represented as strings.
(Bug#23260)
Calculation of COUNT(DISTINCT),
AVG(DISTINCT), or
SUM(DISTINCT) when they are
referenced more than once in a single query with GROUP
BY could cause a server crash.
(Bug#23184)
Evaluation of subqueries that require the filesort algorithm
were allocating and freeing the
sort_buffer_size buffer many
times, resulting in slow performance. Now the buffer is
allocated once and reused.
(Bug#21727)
InnoDB crashed while performing XA recovery
of prepared transactions.
(Bug#21468)
Certain malformed INSERT
statements could crash the mysql client.
(Bug#21142)
CONCURRENT did not work correctly for
LOAD DATA
INFILE.
(Bug#20637)
Several string functions could return incorrect results when given very large length arguments. (Bug#10963)
This is a Monthly Rapid Update release of the MySQL Enterprise Server 5.0.
This section documents all changes and bug fixes that have been applied since the last MySQL Enterprise Server release (5.0.28).
Functionality added or changed:
MySQL Cluster:
The ndb_config utility now accepts
-c as a short form of the
--ndb-connectstring option.
(Bug#22295)
MySQL Cluster:
Added the --bind-address option for
ndbd. This allows a data node process to be
bound to a specific network interface.
(Bug#22195)
MySQL Cluster:
The NDB storage engine could leak
memory during file operations.
(Bug#21858)
MySQL Cluster:
The Ndb_number_of_storage_nodes system
variable was renamed to
Ndb_number_of_data_nodes.
(Bug#20848)
MySQL Cluster:
The HELP command in the Cluster
management client now provides command-specific help. For
example, HELP RESTART in
ndb_mgm provides detailed information about
the RESTART command.
(Bug#19620)
If the user specified the server options
--max-connections= or
N
--table-cache=, a warning would be given in some cases that some
values were recalculated, with the result that
M
--table-cache could be assigned
greater value.
In such cases, both the warning and the increase in the
--table-cache value were
completely harmless. Note also that it is not possible for the
MySQL Server to predict or to control limitations on the maximum
number of open files, since this is determined by the operating
system.
The value of --table-cache is no
longer increased automatically, and a warning is now given only
if some values had to be decreased due to operating system
limits.
(Bug#21915)
For the CALL statement, stored
procedures that take no arguments now can be invoked without
parentheses. That is, CALL p() and
CALL p are equivalent.
(Bug#21462)
mysql_upgrade now passes all the parameters
specified on the command line to both
mysqlcheck and mysql using
the upgrade_defaults file.
(Bug#20100)
SHOW STATUS is no longer logged
to the slow query log.
(Bug#19764)
mysqldump --single-transaction now uses
START TRANSACTION /*!40100 WITH CONSISTENT SNAPSHOT
*/ rather than
BEGIN to start
a transaction, so that a consistent snapshot will be used on
those servers that support it.
(Bug#19660)
Bugs fixed:
MySQL Cluster: Backup of a cluster failed if there were any tables with 128 or more columns. (Bug#23502)
MySQL Cluster: Cluster backups failed when there were more than 2048 schema objects in the cluster. (Bug#23499)
MySQL Cluster:
The management client command ALL DUMP 1000
would cause the cluster to crash if data nodes were connected to
the cluster but not yet fully started.
(Bug#23203)
MySQL Cluster:
INSERT ...
ON DUPLICATE KEY UPDATE on an
NDB table could lead to deadlocks
and memory leaks.
(Bug#23200)
MySQL Cluster: (NDB API): Inacivity timeouts for scans were not correctly handled. (Bug#23107)
MySQL Cluster: If a node restart could not be performed from the REDO log, no node takeover took place. This could cause partitions to be left empty during a system restart. (Bug#22893)
MySQL Cluster: Multiple node restarts in rapid succession could cause a system restart to fail , or induce a race condition. (Bug#22892, Bug#23210)
MySQL Cluster:
(NDB API): Attempting to read a nonexistent tuple using
Commit mode for
NdbTransaction::execute() caused node
failures.
(Bug#22672)
MySQL Cluster:
The --help output from
NDB binaries did not include
file-related options.
(Bug#21994)
MySQL Cluster: (NDB API): Scans closed before being executed were still placed in the send queue. (Bug#21941)
MySQL Cluster: A scan timeout returned Error 4028 (Node failure caused abort of transaction) instead of Error 4008 (Node failure caused abort of transaction...). (Bug#21799)
MySQL Cluster:
The node recovery algorithm was missing a version check for
tables in the ALTER_TABLE_COMMITTED state (as
opposed to the TABLE_ADD_COMMITTED state,
which has the version check). This could cause inconsistent
schemas across nodes following node recovery.
(Bug#21756)
MySQL Cluster: Partition distribution keys were updated only for the primary and starting replicas during node recovery. This could lead to node failure recovery for clusters having an odd number of replicas.
For best results, use values for
NumberOfReplicas that are even powers of 2.
MySQL Cluster: The ndb_mgm management client did not set the exit status on errors, always returning 0 instead. (Bug#21530)
MySQL Cluster:
Attempting to create an NDB table
on a MySQL with an existing non-Cluster table with the same name
in the same database could result in data loss or corruption.
MySQL now issues a warning when a SHOW
TABLES or other statement causing table discovery
finds such a table.
(Bug#21378)
MySQL Cluster: Cluster logs were not rotated following the first rotation cycle. (Bug#21345)
MySQL Cluster:
When inserting a row into an NDB
table with a duplicate value for a nonprimary unique key, the
error issued would reference the wrong key.
(Bug#21072)
MySQL Cluster:
Condition pushdown did not work correctly with
DATETIME columns.
(Bug#21056)
MySQL Cluster: Under some circumstances, local checkpointing would hang, keeping any unstarted nodes from being started. (Bug#20895)
MySQL Cluster:
Using an invalid node ID with the management client
STOP command could cause
ndb_mgm to hang.
(Bug#20575)
MySQL Cluster: Data nodes added while the cluster was running in single user mode were all assigned node ID 0, which could later cause multiple node failures. Adding nodes while in single user mode is no longer possible. (Bug#20395)
MySQL Cluster:
In some cases where SELECT COUNT(*) from an
NDB table should have yielded an
error, MAX_INT was returned instead.
(Bug#19914)
MySQL Cluster: Following the restart of a management node, the Cluster management client did not automatically reconnect. (Bug#19873)
MySQL Cluster:
Error messages given when trying to make online changes to
parameters such as NoOfReplicas that can only
be changed via a complete shutdown and restart of the cluster
did not indicate the true nature of the problem.
(Bug#19787)
MySQL Cluster: ndb_restore did not always make clear that it had recovered successfully from temporary errors while restoring a cluster backup. (Bug#19651)
MySQL Cluster:
In rare situations with resource shortages, a crash could result
from insufficient IndexScanOperations.
(Bug#19198)
MySQL Cluster: ndb_mgm -e show | head would hang after displaying the first 10 lines of output. (Bug#19047)
MySQL Cluster: The error returned by the cluster when too many nodes were defined did not make clear the nature of the problem. (Bug#19045)
MySQL Cluster:
The ndb_config utility did not perform host
lookups correctly when using the --host option
(Bug#17582)
MySQL Cluster: A problem with takeover during a system restart caused ordered indexes to be rebuilt incorrectly. (Bug#15303)
Replication: Column names were not quoted properly for replicated views. (Bug#19736)
Replication:
Transient errors in replication from master to slave may trigger
multiple Got fatal error 1236: 'binlog truncated in the
middle of event' errors on the slave.
(Bug#4053)
Cluster API:
The NdbOperation::getBlobHandle() method,
when called with the name of a nonexistent column, caused a
segmentation fault.
(Bug#21036)
Cluster API: When multiple processes or threads in parallel performed the same ordered scan with exclusive lock and updated the retrieved records, the scan could skip some records, which as a result were not updated. (Bug#20446)
There was a race condition in the InnoDB
fil_flush_file_spaces() function.
(Bug#24089)
This regression was introduced by Bug#15653.
yaSSL-related memory leaks were detected by Valgrind. (Bug#23981)
The internal SQL interpreter of InnoDB placed
an unnecessary lock on the supremum record with
innodb_locks_unsafe_for_binlog
enabled. This caused an assertion failure when
InnoDB was built with debugging enabled.
(Bug#23769)
M % 0NULL, but (
evaluated to
false.
(Bug#23411)M % 0) IS NULL
For not-yet-authenticated connections, the
Time column in SHOW
PROCESSLIST was a random value rather than
NULL.
(Bug#23379)
MySQL failed to build on Linux/Alpha. (Bug#23256)
This regression was introduced by Bug#21250.
If COMPRESS() returned
NULL, subsequent invocations of
COMPRESS() within a result set or
within a trigger also returned NULL.
(Bug#23254)
Insufficient memory
(myisam_sort_buffer_size) could
cause a server crash for several operations on
MyISAM tables: repair table, create index by
sort, repair by sort, parallel repair, bulk insert.
(Bug#23175)
The column default value in the output from
SHOW COLUMNS or SELECT
FROM INFORMATION_SCHEMA.COLUMNS was truncated to 64
characters.
(Bug#23037)
mysql did not check for errors when fetching data during result set printing. (Bug#22913)
InnoDB exhibited thread thrashing with more
than 50 concurrent connections under an update-intensive
workload.
(Bug#22868)
The return value from my_seek() was ignored.
(Bug#22828)
The optimizer failed to use equality propagation for
BETWEEN and IN
predicates with string arguments.
(Bug#22753)
The Handler_rollback status
variable sometimes was incremented when no rollback had taken
place.
(Bug#22728)
The Host column in SHOW
PROCESSLIST output was blank when the server was
started with the
--skip-grant-tables option.
(Bug#22723)
If a table contains an AUTO_INCREMENT column,
inserting into an insertable view on the table that does not
include the AUTO_INCREMENT column should not
change the value of
LAST_INSERT_ID(), because the
side effects of inserting default values into columns not part
of the view should not be visible. MySQL was incorrectly setting
LAST_INSERT_ID() to zero.
(Bug#22584)
Instance Manager had a race condition involving mysqld PID file removal. (Bug#22379)
The optimizer used the ref
join type rather than eq_ref
for a simple join on strings.
(Bug#22367)
Some queries that used MAX() and
GROUP BY could incorrectly return an empty
result.
(Bug#22342)
If an init_connect SQL
statement produced an error, the connection was silently
terminated with no error message. Now the server writes a
warning to the error log.
(Bug#22158)
Use of a DES-encrypted SSL certificate file caused a server crash. (Bug#21868)
Use of PREPARE with a
CREATE PROCEDURE statement that
contained a syntax error caused a server crash.
(Bug#21856)
Adding a day, month, or year interval to a
DATE value produced a
DATE, but adding a week interval
produced a DATETIME value. Now
all produce a DATE value.
(Bug#21811)
Use of a subquery that invoked a function in the column list of the outer query resulted in a memory leak. (Bug#21798)
Selecting from a MERGE table could result in
a server crash if the underlying tables had fewer indexes than
the MERGE table itself.
(Bug#21617, Bug#22937)
After FLUSH TABLES WITH
READ LOCK followed by
UNLOCK
TABLES, attempts to drop or alter a stored routine
failed with an error that the routine did not exist, and
attempts to execute the routine failed with a lock conflict
error.
(Bug#21414)
For multiple-table UPDATE
statements, storage engines were not notified of duplicate-key
errors.
(Bug#21381)
Within a prepared statement, SELECT (COUNT(*) =
1) (or similar use of other aggregate functions) did
not return the correct result for statement re-execution.
(Bug#21354)
It was possible for a stored routine with a
non-latin1 name to cause a stack overrun.
(Bug#21311)
Creating a TEMPORARY table with the same name
as an existing table that was locked by another client could
result in a lock conflict for DROP TEMPORARY
TABLE because the server unnecessarily tried to
acquire a name lock.
(Bug#21096)
Incorrect results could be obtained from re-execution of a
parametrized prepared statement or a stored routine with a
SELECT that uses LEFT
JOIN with a second table having only one row.
(Bug#21081)
Within a stored routine, a view definition cannot refer to routine parameters or local variables. However, an error did not occur until the routine was called. Now it occurs during parsing of the routine creation statement.
A side effect of this fix is that if you have already created
such routines, and error will occur if you execute
SHOW CREATE PROCEDURE or
SHOW CREATE FUNCTION. You
should drop these routines because they are erroneous.
In mysql, invoking connect
or \r with very long
db_name or
host_name parameters caused buffer
overflow.
(Bug#20894)
SHOW VARIABLES truncated the
Value field to 256 characters.
(Bug#20862)
WITH ROLLUP could group unequal values.
(Bug#20825)
Range searches on columns with an index prefix could miss records. (Bug#20732)
An UPDATE that referred to a key
column in the WHERE clause and activated a
trigger that modified the column resulted in a loop.
(Bug#20670)
LIKE searches failed for indexed
utf8 character columns.
(Bug#20471)
With sql_mode = TRADITIONAL, MySQL
incorrectly aborted on warnings within stored routines and
triggers.
(Bug#20028)
mysqldump --xml produced invalid XML for
BLOB data.
(Bug#19745)
FLUSH INSTANCES in Instance Manager triggered
an assertion failure.
(Bug#19368)
For a debug server, a reference to an undefined user variable in
a prepared statement executed with
EXECUTE caused an assertion
failure.
(Bug#19356)
Within a trigger for a base table, selecting from a view on that base table failed. (Bug#19111)
The value of the warning_count
system variable was not being calculated correctly (also
affecting SHOW COUNT(*) WARNINGS).
(Bug#19024)
DELETE IGNORE could hang for foreign key
parent deletes.
(Bug#18819)
InnoDB used table locks (not row locks)
within stored functions.
(Bug#18077)
mysql would lose its connection to the server if its standard output was not writable. (Bug#17583)
mysql-test-run did not work correctly for RPM-based installations. (Bug#17194)
A client library crash was caused by executing a statement such
as SELECT * FROM t1 PROCEDURE ANALYSE() using
a server side cursor on a table t1 that does
not have the same number of columns as the output from
PROCEDURE ANALYSE().
(Bug#17039)
The WITH CHECK OPTION for a view failed to
prevent storing invalid column values for
UPDATE statements.
(Bug#16813)
InnoDB showed substandard performance with
multiple queries running concurrently.
(Bug#15815)
ALTER TABLE was not able to
rename a view.
(Bug#14959)
Statements such as DROP PROCEDURE
and DROP VIEW were written to the
binary log too late due to a race condition.
(Bug#14262)
A literal string in a GROUP BY clause could
be interpreted as a column name.
(Bug#14019)
Instance Manager didn't close the client socket file when starting a new mysqld instance. mysqld inherited the socket, causing clients connected to Instance Manager to hang. (Bug#12751)
Entries in the slow query log could have an incorrect
Rows_examined value.
(Bug#12240)
Lack of validation for input and output
TIME values resulted in several
problems: SEC_TO_TIME() in some
cases did not clip large values to the
TIME range appropriately;
SEC_TO_TIME() treated
BIGINT UNSIGNED values as signed; only
truncation warnings were produced when both truncation and
out-of-range TIME values
occurred.
(Bug#11655, Bug#20927)
A locking safety check in InnoDB reported a
spurious error stored_select_lock_type is 0 inside
::start_stmt() for
INSERT ...
SELECT statements in
innodb_locks_unsafe_for_binlog
mode. The safety check was removed.
(Bug#10746)
FROM_UNIXTIME() did not accept
arguments up to POWER(2,31)-1,
which it had previously.
(Bug#9191)
OPTIMIZE TABLE with
myisam_repair_threads > 1
could result in MyISAM table corruption.
(Bug#8283)
This is the first MySQL Enterprise Server release, following the last Community Server release (5.0.27).
Functionality added or changed:
Binary MySQL distributions no longer include a mysqld-max server, except for RPM distributions. Instead, distributions contain a mysqld binary that includes the features previously included in the mysqld-max binary.
Bugs fixed:
This is a bugfix release for the current MySQL Community Server production release family. It replaces MySQL 5.0.26.
Functionality added or changed:
This is the last version for which binary MySQL-Max distributions are available, except for RPM distributions. (For RPM distributions, the last version is 5.0.37.)
Bugs fixed:
MySQL 5.0.26 introduced an ABI incompatibility, which this release reverts. Programs compiled against 5.0.26 are not compatible with any other version and must be recompiled. (Bug#23427)
This is a bugfix release for the current production release family.
This section documents all changes and bug fixes that have been applied since the last official MySQL release. If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise.
Functionality added or changed:
Important Change: MySQL Cluster:
LOAD DATA
INFILE no longer causes an implicit commit for all
storage engines. It now causes an implicit commit only for
tables using the NDB storage
engine.
(Bug#11151)
The number of InnoDB threads is no longer
limited to 1,000 on Windows.
(Bug#22268)
mysqldump now has a
--flush-privileges option. It
causes mysqldump to emit a
FLUSH
PRIVILEGES statement after dumping the
mysql database. This option should be used
any time the dump contains the mysql database
and any other database that depends on the data in the
mysql database for proper restoration.
(Bug#21424)
For mysqldump, the output generated by the
server when using the --xml
option has changed with regard to null values. It now matches
the output from mysqldump
--xml. That is, a column
containing a NULL value is now reported as
<field name="column_name" xsi:nil="true" />
whereas a column containing the string value
'NULL' is reported as
<field name="column_name">NULL</field>
and a column containing an empty string is reported as
<field name="column_name"></field>
The source distribution has been updated so that the UDF example can be compiled under Windows with CMake. See Section 21.2.2.5, “Compiling and Installing User-Defined Functions”. (Bug#19121)
The LOAD DATA FROM MASTER and LOAD
TABLE FROM MASTER statements are deprecated. See
Section 12.6.2.2, “LOAD DATA FROM MASTER Syntax”, for recommended
alternatives.
(Bug#9125, Bug#20596, Bug#14399, Bug#12187, Bug#15025, Bug#18822)
Bugs fixed:
Replication: Column names supplied for a view created on a master server could be lost on a slave server. (Bug#19419)
Deleting entries from a large MyISAM index
could cause index corruption when it needed to shrink. Deletes
from an index can happen when a record is deleted, when a key
changes and must be moved, and when a key must be un-inserted
because of a duplicate key. This can also happen in
REPAIR TABLE when a duplicate key
is found and in myisamchk when sorting the
records by an index.
(Bug#22384)
yaSSL had a conflicting definition for
socklen_t on hurd-i386 systems.
(Bug#22326)
Conversion of values inserted into a
BIT column could affect adjacent
columns.
(Bug#22271)
mysql_com.h unnecessarily referred to the
ulong type.
(Bug#22227)
The source distribution would not build on Windows due to a
spurious dependency on ib_config.h.
(Bug#22224)
Execution of a prepared statement that uses an
IN subquery with aggregate functions in the
HAVING clause could cause a server crash.
(Bug#22085)
Using GROUP_CONCAT() on the
result of a subquery in the FROM clause that
itself used GROUP_CONCAT() could
cause a server crash.
(Bug#22015)
A query that used GROUP BY and an
ALL or ANY quantified
subquery in a HAVING clause could trigger an
assertion failure.
(Bug#21853)
UPGRADE was treated as a reserved word,
although it is not.
(Bug#21772)
The value of LAST_INSERT_ID() was
not always updated correctly within stored routines.
(Bug#21726)
A function result in a comparison was replaced with a constant by the optimizer under some circumstances when this optimization was invalid. (Bug#21698)
If mysqld was linked against a
system-installed zlib library compiled
without large-file support, it would likely exit with a
SIGXFSZ (file size exceeded) signal if an
ARCHIVE table reached 2GB. The server now
checks for space before writing.
(Bug#21675)
The presence of a subquery in the ON clause
of a join in a view definition prevented the
MERGE algorithm from being used for the view
in cases where it should be allowed.
(Bug#21646)
When records are merged from the insert buffer and the page
needs to be reorganized, InnoDB used
incorrect column length information when interpreting the
records of the page. This caused a server crash due to apparent
corruption of secondary indexes in
ROW_FORMAT=COMPACT that contain prefix
indexes of fixed-length columns. Data files should not be
corrupted, but the crash was likely to repeat every time the
server was restarted.
(Bug#21638)
For character sets having a mbmaxlen value of
2, any ALTER TABLE statement
changed TEXT columns to
MEDIUMTEXT.
(Bug#21620)
mysql displayed an empty string for
NULL values.
(Bug#21618)
For INSERT
... ON DUPLICATE KEY UPDATE, use of
VALUES(
within the col_name)UPDATE clause
sometimes was handled incorrectly.
(Bug#21555)
Subqueries with aggregate functions but no
FROM clause could return incorrect results.
(Bug#21540)
The server could crash for the second execution of a function
containing a SELECT statement
that uses an aggregating IN subquery.
(Bug#21493)
myisam_ftdump produced bad counts for common words. (Bug#21459)
The URL into the online manual that is printed in the stack trace message by the server was out of date. (Bug#21449)
Table aliases in multiple-table
DELETE statements sometimes were
not resolved.
(Bug#21392)
mysql_config --libmysqld-libs did not produce
any SSL options necessary for linking
libmysqld with SSL support enabled.
(Bug#21239)
In the package of pre-built time zone tables that is available
for download at
http://dev.mysql.com/downloads/timezones.html,
the tables now explicitly use the utf8
character set so that they work the same way regardless of the
system character set value.
(Bug#21208)
A subquery that uses an index for both the
WHERE and ORDER BY clauses
produced an empty result.
(Bug#21180)
mysql_upgrade produced a malformed
upgrade_defaults file by overwriting the
[client] group header with a
password option. This prevented
mysqlcheck from running successfully when
invoked by mysql_upgrade.
(Bug#21011)
On Windows, inserting into a MERGE table
after renaming an underlying MyISAM table
caused a server crash.
(Bug#20789)
Within stored routines, some error messages were printed incorrectly. A nonnull-terminated string was passed to a message-printing routine that expected a null-terminated string. (Bug#20778)
INSERT DELAYED did not honor
SET INSERT_ID or the
auto_increment_* system variables.
(Bug#20627, Bug#20830)
If the auto_increment_offset
setting causes MySQL to generate a value larger than the
column's maximum possible value, the
INSERT statement is accepted in
strict SQL mode, whereas but should fail with an error.
(Bug#20573)
User names have a maximum length of 16 characters (even if they contain multi-byte characters), but were being truncated to 16 bytes. (Bug#20393)
PROCEDURE ANALYSE() returned incorrect values
of M
FLOAT( and
M,
D)DOUBLE(.
(Bug#20305)M,
D)
For a MyISAM table locked with LOCK
TABLES ...WRITE, queries optimized using the
index_merge method did not
show rows inserted with the lock in place.
(Bug#20256)
SUBSTRING() results sometimes
were stored improperly into a temporary table when multi-byte
character sets were used.
(Bug#20204)
For an ENUM column that used the
ucs2 character set, using
ALTER TABLE to modify the column
definition caused the default value to be lost.
(Bug#20108)
Join conditions using index prefixes on utf8
columns of InnoDB tables incorrectly ignored
rows where the length of the actual value was greater than the
length of the index prefix.
(Bug#19960)
make install tried to build files that should already have been built by make all, causing a failure if installation was performed using a different account than the one used for the initial build. (Bug#19738)
For a MyISAM table with a
FULLTEXT index, compression with
myisampack or a check with
myisamchk after compression resulted in table
corruption.
(Bug#19702)
The build process incorrectly tried to overwrite
sql/lex_hash.h. This caused the build to
fail when using a shadow link tree pointing to original sources
that were owned by another account.
(Bug#18888)
Linking the pthreads library to
single-threaded MySQL libraries caused
dlopen() to fail at runtime on HP-UX.
(Bug#18267)
The source distribution failed to compile when configured with
the --with-libwrap option.
(Bug#18246)
Queries containing a subquery that used aggregate functions could return incorrect results. (Bug#16792)
Row equalities (such as WHERE (a,b) = (c,d)
were not taken into account by the optimizer, resulting in slow
query execution. Now they are treated as conjunctions of
equalities between row elements.
(Bug#16081)
BIN(),
OCT(), and
CONV() did not work with BIT
values.
(Bug#15583)
The parser rejected queries that selected from a table twice
using a UNION within a subquery.
The parser now supports arbitrary subquery, join, and
parenthesis operations within EXISTS
subqueries. A limitation still exists for scalar subqueries: If
the subquery contains UNION, the
first SELECT of the
UNION cannot be within
parentheses. For example, SELECT (SELECT a FROM t1
UNION SELECT b FROM t2) will work, but SELECT
((SELECT a FROM t1) UNION (SELECT b FROM t2)) will
not.
(Bug#14654)
On Mac OS X, zero-byte read() or
write() calls to an SMB-mounted file system
could return a nonstandard return value, leading to data
corruption. Now such calls are avoided.
(Bug#12620)
The server returns a more informative error message when it
attempts to open a MERGE table that has been
defined to use non-MyISAM tables.
(Bug#10974)
With TRADITIONAL SQL mode,
assignment of out-of-bound values and rounding of assigned
values was done correctly, but assignment of the same numbers
represented as strings sometimes was handled differently.
(Bug#6147)
On an INSERT into an updatable
but noninsertable view, an error message was issued stating that
the view was not updatable. Now the message says the view is not
insertable-into.
(Bug#5505)
EXPLAIN sometimes returned an
incorrect select_type for a
SELECT from a view, compared to
the select_type for the equivalent
SELECT from the base table.
(Bug#5500)
Incorporated portability fixes into the definition of
__attribute__ in
my_global.h.
(Bug#2717)
End of Product LifecycleActive development and support for MySQL database server versions 3.23, 4.0, and 4.1 has ended. However, for MySQL 4.0 and 4.1, there is still extended support available. For details, see http://www.mysql.com/company/legal/lifecycle/#calendar. According to the MySQL Lifecycle Policy (see http://www.mysql.com/company/legal/lifecycle/#policy), only Security and Severity Level 1 issues will still be fixed for MySQL 4.0 and 4.1. Please consider upgrading to a recent version (MySQL 5.0 or 5.1).
This is a bugfix release for the current production release family. This version was released as MySQL Classic 5.0.25 to commercial customers only.
This section documents all changes and bug fixes that have been applied since the last official MySQL release. If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise.
Functionality added or changed:
MySQL did not properly do stack dumps on
x86_64 and i386/NPTL
systems. (Note that the initial fix for this problem was
discovered not to be correct. Further work on the problem was
undertaken only for MySQL 5.1 and up. See Bug#31891.)
(Bug#21250)
The mysqld and mysqlmanager man pages have been reclassified from volume 1 to volume 8. (Bug#21220)
InnoDB now honors IGNORE
INDEX. Perviously using IGNORE
INDEX in cases where an index sort would be slower
than a filesort had no effect when used with
InnoDB tables.
This fix was reverted in MySQL 5.0.26, and a new fix made in MySQL 5.0.40.
TIMESTAMP columns that are
NOT NULL now are reported that way by
SHOW COLUMNS and
INFORMATION_SCHEMA.
(Bug#20910)
The MySQL distribution now compiles on UnixWare 7.13. (Bug#20190)
configure now defines the symbol
DBUG_ON in config.h to
indicate whether the source tree is configured to be compiled
with debugging support.
(Bug#19517)
mysql_upgrade no longer reads the
[client] option file group because it is not
a client and did not understand client options such as
host. Now it reads only the
[mysql_upgrade] group.
(Bug#19452)
For mysqlshow, if a database name argument
contains wildcard characters (such as “
_ ”) but matches a single database
name exactly, treat the name as a literal name. This allows a
command such as mysqlshow information_schema
work without having to escape the wildcard character.
(Bug#19147)
On Windows, typing Control-C while a query was running caused the mysql client to crash. Now it causes mysql to attempt to kill the current statement. If this cannot be done, or Control-C is typed again before the statement is killed, mysql exits. (In other words, mysql's behavior with regard to Control-C is now the same as it is on Unix platforms.) (Bug#17926)
See also Bug#1989.
The VIEW_DEFINITION column of the
INFORMATION_SCHEMA
VIEWS table now contains
information about the view algorithm.
(Bug#16832)
The bundled yaSSL library licensing has added a FLOSS exception
similar to MySQL to resolve licensing incompatibilities with
MySQL. (See the
extra/yassl/FLOSS-EXCEPTIONS file in a
MySQL source distribution for details.)
(Bug#16755)
Table comments longer than 60 characters and column comments longer than 255 characters were truncated silently. Now a warning is issued, or an error in strict mode. (Bug#13934)
The mysql client used the default character
set if it automatically reconnected to the server, which is
incorrect if the character set had been changed. To enable the
character set to remain synchronized on the client and server,
the mysql command charset
(or \C) that changes the default character
set and now also issues a SET NAMES
statement. The changed character set is used for reconnects.
(Bug#11972)
If a DROP VIEW statement named
multiple views, it stopped with an error if a nonexistent view
was named and did not drop the remaining views. Now it continues
on and reports an error at the end, similar to
DROP TABLE.
(Bug#11551)
The server now issues a warning if it removes leading spaces from an alias. (Bug#10977)
For a successful dump, mysqldump now writes a SQL comment to the end of the dump file in the following format:
-- Dump completed on YYYY-MM-DD hh:mm:ss
For spatial data types, the server formerly returned these as
VARSTRING values with a binary collation. Now
the server returns spatial values as
BLOB values.
(Bug#10166)
A new system variable,
lc_time_names, specifies the
locale that controls the language used to display day and month
names and abbreviations. This variable affects the output from
the DATE_FORMAT(),
DAYNAME() and
MONTHNAME() functions. See
Section 9.8, “MySQL Server Locale Support”.
Using --with-debug to
configure MySQL with debugging support enables you to use the
--debug="d,parser_debug" option
when you start the server. This causes the Bison parser that is
used to process SQL statements to dump a parser trace to the
server's standard error output. Typically, this output is
written to the error log.
The bundled yaSSL library was upgraded to version 1.3.7.
Bugs fixed:
Security Fix:
A stored routine created by one user and then made accessible to
a different user using
GRANT EXECUTE
could be executed by that user with the privileges of the
routine's definer.
(Bug#18630, CVE-2006-4227)
Security Fix: On Linux, and possibly other platforms using case-sensitive file systems, it was possible for a user granted rights on a database to create or access a database whose name differed only from that of the first by the case of one or more letters. (Bug#17647, CVE-2006-4226)
MySQL Cluster: Packaging:
The ndb_mgm program was included in both the
MySQL-ndb-tools and
MySQL-ndb-management RPM packages, resulting
in a conflict if both were installed. Now
ndb_mgm is included only in
MySQL-ndb-tools.
(Bug#21058)
MySQL Cluster:
Setting TransactionDeadlockDetectionTimeout
to a value greater than 12000 would cause scans to deadlock,
time out, fail to release scan records, until the cluster ran
out of scan records and stopped processing.
(Bug#21800)
MySQL Cluster: A memory leak occurred when running ndb_mgm -e "SHOW". (Bug#21670)
MySQL Cluster: The server provided a nondescriptive error message when encountering a fatally corrupted REDO log. (Bug#21615)
MySQL Cluster: A partial rollback could lead to node restart failures. (Bug#21536)
MySQL Cluster: The failure of a unique index read due to an invalid schema version could be handled incorrectly in some cases, leading to unpredictable results. (Bug#21384)
MySQL Cluster: In a cluster with more than 2 replicas, a manual restart of one of the data nodes could fail and cause the other nodes in the same node group to shut down. (Bug#21213)
MySQL Cluster:
Some queries involving joins on very large
NDB tables could crash the MySQL
server.
(Bug#21059)
MySQL Cluster: Restarting a data node while DDL operations were in progress on the cluster could cause other data nodes to fail. This could also lead to mysqld hanging or crashing under some circumstances. (Bug#21017, Bug#21050)
MySQL Cluster: In some situations with a high disk-load, writing of the redo log could hang, causing a crash with the error message GCP STOP detected. (Bug#20904)
MySQL Cluster:
When the redo buffer ran out of space, a Pointer too
large error was raised and the cluster could become
unusable until restarted with --initial.
(Bug#20892)
MySQL Cluster: A vague error message was returned when reading both schema files during a restart of the cluster. (Bug#20860)
MySQL Cluster:
Incorrect values were inserted into
AUTO_INCREMENT columns of tables restored
from a cluster backup.
(Bug#20820)
MySQL Cluster: When attempting to restart the cluster following a data import, the cluster failed during Phase 4 of the restart with Error 2334: Job buffer congestion. (Bug#20774)
MySQL Cluster:
REPLACE statements did not work
correctly on an NDB table having
both a primary key and a unique key. In such cases, proper
values were not set for columns which were not explicitly
referenced in the statement.
(Bug#20728)
MySQL Cluster:
The server did not honor the value set for
ndb_cache_check_time in the
my.cnf file.
(Bug#20708)
MySQL Cluster: ndb_size.pl and ndb_error_reporter were missing from RPM packages. (Bug#20426)
MySQL Cluster:
Running ndbd
--nowait-nodes=
where id id was the node ID of a node
that was already running would fail with an invalid error
message.
(Bug#20419)
MySQL Cluster:
(Direct APIs): NdbScanOperation::readTuples()
and NdbIndexScanOperation::readTuples()
ignored the batch parameter.
(Bug#20252)
MySQL Cluster: A node failure during a scan could sometime cause the node to crash when restarting too quickly following the failure. (Bug#20197)
MySQL Cluster:
It was possible to use port numbers greater than 65535 for
ServerPort in the
config.ini file.
(Bug#19164)
MySQL Cluster: Under certain circumstances, a node that was shut down then restarted could hang during the restart. (Bug#18863)
MySQL Cluster: Trying to create or drop a table while a node was restarting caused the node to crash. This is now handled by raising an error. (Bug#18781)
MySQL Cluster: The server failed with a nondescriptive error message when out of data memory. (Bug#18475)
MySQL Cluster:
For NDB and possibly
InnoDB tables, a BEFORE
UPDATE trigger could insert incorrect values.
(Bug#18437)
MySQL Cluster:
SELECT ... FOR
UPDATE failed to lock the selected rows.
(Bug#18184)
MySQL Cluster:
perror did not properly report
NDB error codes.
(Bug#16561)
MySQL Cluster:
A Cluster whose storage nodes were installed from the
MySQL-ndb-storage- RPMs could not perform *
CREATE or
ALTER operations that made use of nondefault
character sets or collations.
(Bug#14918)
MySQL Cluster:
The management client ALL STATUS command
could sometimes report the status of some data nodes
incorrectly.
(Bug#13985)
MySQL Cluster: An issue that arose from a patch for Bug#19852 made in MySQL 5.0.23 was corrected. (See Section C.1.59, “Changes in MySQL 5.0.23 (Not released)”.)
Cluster Replication: Replication:
In some cases, a large number of MySQL servers sending requests
to the cluster simultaneously could cause the cluster to crash.
This could also be triggered by many
NDB API clients making simultaneous
event subscriptions or unsubscriptions.
(Bug#20683)
Replication:
CREATE PROCEDURE, CREATE
FUNTION, CREATE
TRIGGER, and CREATE
VIEW statements containing multi-line comments
(/* ... */) could not be replicated.
(Bug#20438)
Cluster API:
Invoking the MGM API function
ndb_mgm_listen_event() caused a memory leak.
(Bug#21671)
Cluster API:
The MGM API function ndb_logevent_get_fd()
was not implemented.
(Bug#21129)
Some Linux-x86_64-icc packages (of previous releases) mistakenly contained 32-bit binaries. Only ICC builds are affected, not gcc builds. Solaris and FreeBSD x86_64 builds are not affected. (Bug#22238)
Running SHOW
MASTER LOGS at the same time as binary log files were
being switched would cause mysqld to hang.
(Bug#21965)
libmysqlclient defined a symbol
BN_bin2bn which belongs to OpenSSL. This
could break applications that also linked against OpenSSL's
libcrypto library. The fix required
correcting an error in a build script that was failing to add
rename macros for some functions.
(Bug#21930)
character_set_results can be
NULL to signify “no conversion,”
but some code did not check for NULL,
resulting in a server crash.
(Bug#21913)
A NUL byte within a prepared statement string
caused the rest of the string not to be written to the query
log, allowing logging to be bypassed.
(Bug#21813)
COUNT(*) queries with
ORDER BY and LIMIT could
return the wrong result.
This problem was introduced by the fix for Bug#9676, which
limited the rows stored in a temporary table to the
LIMIT clause. This optimization is not
applicable to nongroup queries with aggregate functions. The
current fix disables the optimization in such cases.
INSERT ...
SELECT sometimes generated a spurious Column
count doesn't match value count error.
(Bug#21774)
EXPORT_SET() did not accept
arguments with coercible character sets.
(Bug#21531)
mysqldump incorrectly tried to use
LOCK TABLES for tables in the
INFORMATION_SCHEMA database.
(Bug#21527)
Memory overruns could occur for certain kinds of subqueries. (Bug#21477)
A DATE can be represented as an
integer (such as 20060101) or as a string
(such as '2006.01.01'). When a
DATE (or
TIME) column is compared in one
SELECT against both
representations, constant propagation by the optimizer led to
comparison of DATE as a string
against DATE as an integer. This
could result in integer comparisons such as
2006 against 20060101,
erroneously producing a false result.
(Bug#21475)
Adding ORDER BY to a SELECT
DISTINCT( query could
produce incorrect results.
(Bug#21456)expr)
Database and table names have a maximum length of 64 characters (even if they contain multi-byte characters), but were truncated to 64 bytes.
This fix was reverted in MySQL 5.0.26.
With max_sp_recursion set to 0, a stored
procedure that executed a SHOW CREATE
PROCEDURE statement for itself triggered a recursion
limit exceeded error, though the statement involves no
recursion.
(Bug#21416)
On 64-bit Windows, a missing table generated error 1017, not the correct value of 1146. (Bug#21396)
The optimizer sometimes produced an incorrect row-count estimate
after elimination of const
tables. This resulted in choosing extremely inefficient
execution plans in same cases when distribution of data in joins
were skewed.
(Bug#21390)
A query result could be sorted improperly when using
ORDER BY for the second table in a join.
(Bug#21302)
Query results could be incorrect if the WHERE
clause contained t., where
key_part
NOT IN (val_list)val_list is a list of more than 1000
constants.
(Bug#21282)
For user-defined functions created with
CREATE FUNCTION, the
DEFINER clause is not legal, but no error was
generated.
(Bug#21269)
The SELECT privilege was required
for an insert on a view, instead of the
INSERT privilege.
(Bug#21261)
This regression was introduced by Bug#20989.
Subqueries on INFORMATION_SCHEMA tables could
erroneously return an empty result.
(Bug#21231)
mysql_upgrade created temporary files in a possibly insecure way. (Bug#21224)
When DROP DATABASE or
SHOW OPEN TABLES was issued while
concurrently in another connection issuing
DROP TABLE,
RENAME TABLE, CREATE
TABLE LIKE or any other statement that required a name
lock, the server crashed.
(Bug#21216, Bug#19403)
The --master-data option for
mysqldump requires certain privileges, but
mysqldump generated a truncated dump file
without producing an appropriate error message or exit status if
the invoking user did not have those privileges.
(Bug#21215)
Some prepared statements caused a server crash when executed a second time. (Bug#21166)
The optimizer assumed that if (a=x AND b=x)
is true, (a=x AND b=x) AND a=b is also true.
But that is not always so if a and
b have different data types.
(Bug#21159)
SHOW INNODB STATUS contained some
duplicate output.
(Bug#21113)
InnoDB was slow with more than 100,000
.idb files.
(Bug#21112)
Performing an INSERT on a view
that was defined using a SELECT
that specified a collation and a column alias caused the server
to crash .
(Bug#21086)
ALTER VIEW did not retain
existing values of attributes that had been originally specified
but were not changed in the ALTER
VIEW statement.
(Bug#21080)
For InnoDB tables, the server could crash
when executing NOT IN(...) subqueries.
(Bug#21077)
The myisam_stats_method
variable was mishandled when set from an option file or on the
command line.
(Bug#21054)
With query_cache_type set to 0,
RESET QUERY CACHE was very slow and other
threads were blocked during the operation. Now a cache reset is
faster and nonblocking.
(Bug#21051)
mysql crashed for very long arguments to the
connect command.
(Bug#21042)
A query using WHERE did not
return consistent results on successive invocations. The
column =
constant OR
column IS NULLcolumn in each part of the
WHERE clause could be either the same column,
or two different columns, for the effect to be observed.
(Bug#21019)
Performance during an import on a table with a trigger that called a stored procedure was severely degraded. This issue first arose in MySQL 5.0.18. (Bug#21013)
A query of the form shown here caused the server to crash:
SELECT * FROM t1 NATURAL JOIN (
t2 JOIN (
t3 NATURAL JOIN t4,
t5 NATURAL JOIN t6
)
ON (t3.id3 = t2.id3 AND t5.id5 = t2.id5)
);
STR_TO_DATE() sometimes would
return NULL if the %D
format specifier was not the last specifier in the format
string.
(Bug#20987)
A query using WHERE NOT
( yielded a
different result from the same query using the same
column < ANY
(subquery))column and
subquery with WHERE
(.
(Bug#20975)column > ANY
(subquery))
In debugging mode, mysqld printed
server_init rather than
network_init during network initialization.
(Bug#20968)
Under certain circumstances,
AVG(
returned a value but
key_val)MAX(
returned an empty set due to incorrect application of
key_val)MIN()/MAX() optimization.
(Bug#20954)
On Windows, mysql_upgrade.exe could not find mysqlcheck.exe. (Bug#20950)
Use of zero-length variable names caused a server crash. (Bug#20908)
The server crashed when using the range access method to execut
a subquery with a ORDER BY DESC clause.
(Bug#20869)
For certain queries, the server incorrectly resolved a reference to an aggregate function and crashed. (Bug#20868)
Using aggregate functions in subqueries yielded incorrect
results under certain circumstances due to incorrect application
of
MIN()/MAX()
optimization.
(Bug#20792)
If a column definition contained a character set declaration,
but a DEFAULT value began with an introducer,
the introducer character set was used as the column character
set.
(Bug#20695)
Multiplication of DECIMAL values
could produce incorrect fractional part and trailing garbage
caused by signed overflow.
(Bug#20569)
Users who had the SHOW VIEW
privilege for a view and privileges on one of the view's
base tables could not see records in
INFORMATION_SCHEMA tables relating to the
base table.
(Bug#20543)
The MD5(),
SHA1(), and
ENCRYPT() functions should return
a binary string, but the result sometimes was converted to the
character set of the argument.
MAKE_SET() and
EXPORT_SET() now use the correct
character set for their default separators, resulting in
consistent result strings which can be coerced according to
normal character set rules.
(Bug#20536)
A subquery that contained LIMIT
could return more than
one row.
(Bug#20519)N,1
Creation of a view as a join of views or tables could fail if the views or tables are in different databases. (Bug#20482)
SELECT statements using
GROUP BY against a view could have missing
columns in the output when there was a trigger defined on one of
the base tables for the view.
(Bug#20466)
For connections that required a SUBJECT
value, a check was performed to verify that the value was
correct, but the connection was not refused if not.
(Bug#20411)
Some user-level errors were being written to the server's error log, which is for server errors. (Bug#20402)
perror crashed on Solaris due to
NULL return value of
strerror() system call.
(Bug#20145)
For mysql, escaping with backslash sometimes did not work. (Bug#20103)
Use of MIN() or
MAX() with GROUP
BY on a ucs2 column could cause a
server crash.
(Bug#20076)
mysqld --flush failed to flush
MyISAM table changes to disk following an
UPDATE statement for which no
updated column had an index.
(Bug#20060)
A user-defined function that is called on each row of a returned
result set, could receive an in_null state
that is set, if it was set previously. Now, the
is_null state is reset to false before each
invocation of a UDF.
(Bug#19904)
The query command for
mysqltest did not work.
(Bug#19890)
When executing a SELECT with
ORDER BY on a view that is constructed from a
SELECT statement containing a
stored function, the stored function was evaluated too many
times.
(Bug#19862)
The first time a user who had been granted the
CREATE ROUTINE privilege used
that privilege to create a stored function or procedure, the
Password column in that user's row in the
mysql.user table was set to
NULL.
(Bug#19857)
For TIME_FORMAT(), the
%H and %k format
specifiers can return values larger than two digits (if the hour
is greater than 99), but for some query results that contained
three-character hours, column values were truncated.
(Bug#19844)
Using SELECT on a corrupt
MyISAM table using the dynamic record format
could cause a server crash.
(Bug#19835)
Using cursors with READ
COMMITTED isolation level could cause
InnoDB to crash.
(Bug#19834)
The yaSSL library bundled with libmysqlclient
had some conflicts with OpenSSL. Now macros are used to rename
the conflicting symbols to have a prefix of
ya.
(Bug#19810)
On 64-bit systems, use of the cp1250
character set with a primary key column in a
LIKE clause caused a server crash for
patterns having letters in the range 128..255.
(Bug#19741)
DESCRIBE returned the type
BIGINT for a column of a view if
the column was specified by an expression over values of the
type INT.
(Bug#19714)
An issue with yaSSL prevented Connector/J clients from connecting to the server using a certificate. (Bug#19705)
A cast problem caused incorrect results for prepared statements that returned float values when MySQL was compiled with gcc 4.0. (Bug#19694)
The mysql_list_fields() C API
function returned the incorrect table name for views.
(Bug#19671)
If a query had a condition of the form
, which participated in equality propagation and also
was used for tableX.key
=
tableY.key
ref access, then
early ref-access
NULL filtering was not performed for the
condition. This could make query execution slower.
(Bug#19649)
Repeated DROP TABLE statements in
a stored procedure could sometimes cause the server to crash.
(Bug#19399)
When not running in strict mode, the server failed to convert
the invalid years portion of a
DATE or
DATETIME value to
'0000' when inserting it into a table.
This fix was reverted in MySQL 5.0.40.
See also Bug#25301.
The final parenthesis of a CREATE
INDEX statement occurring in a stored procedure was
omitted from the binary log when the stored procedure was
called.
(Bug#19207)
A SELECT with a subquery that was
bound to the outer query over multiple columns returned
different results when a constant was used instead of one of the
dependant columns.
(Bug#18925)
Setting myisam_repair_threads
caused any repair operation on a MyISAM table
to fail to update the cardinality of indexes, instead making
them always equal to 1.
(Bug#18874)
FEDERATED tables raised invalid duplicate key
errors when attempting on one server to insert rows having the
same primary key values as rows that had been deleted from the
linked table on the other server.
(Bug#18764)
The implementation for
UNCOMPRESS() did not indicate
that it could return NULL, causing the
optimizer to do the wrong thing.
(Bug#18539)
Using > ALL with subqueries that return no
rows yielded incorrect results under certain circumstances due
to incorrect application of
MIN()/MAX()
optimization.
(Bug#18503)
Referring to a stored function qualified with the name of one database and tables in another database caused a “table doesn't exist” error. (Bug#18444)
Triggers on tables in the mysql database
caused a server crash. Triggers for tables in this database now
are disallowed.
(Bug#18361, Bug#18005)
The length of the pattern string prefix for
LIKE operations was calculated incorrectly
for multi-byte character sets. As a result, the scanned range
was wider than necessary if the prefix contained any multi-byte
characters, and rows could be missing from the result set.
(Bug#18359, Bug#16674)
Multiple invocations of the
REVERSE() function could return
different results.
(Bug#18243)
The optimizer did not take advantage of indexes on columns used
for the second or third arguments of
BETWEEN.
(Bug#18165)
For table-format output, mysql did not always calculate columns widths correctly for columns containing multi-byte characters in the column name or contents. (Bug#17939)
The character set was not being properly initialized for
CAST() with a type like
CHAR(2) BINARY, which resulted in incorrect
results or even a server crash.
(Bug#17903)
Checking a MyISAM table (using
CHECK TABLE) having a spatial
index and only one row would wrongly indicate that the table was
corrupted.
(Bug#17877)
A stored procedure that created and invoked a prepared statement was not executed when called in a mysqld init-file. (Bug#17843)
It is possible to create MERGE tables into
which data cannot be inserted (by not specifying a
UNION clause. However, when an
insert was attempted, the error message was confusing. Now an
error occurs indicating that the table is read-only.
(Bug#17766)
Attempting to insert a string of greater than 4096 bytes into a
FEDERATED table resulted in the error
ERROR 1296 (HY000) at line 2: Got error 10000 'Error
on remote system: 1054: Unknown column
'string-value' from
FEDERATED. This error was raised regardless of the
type of column involved (VARCHAR,
TEXT, and so on.)
(Bug#17608)
Views could not be updated within a stored function or trigger. (Bug#17591)
Use of the --prompt option or
prompt command caused
mysql to be unable to connect to the Instance
Manager.
(Bug#17485)
N'xxx' and _utf8'xxx' were
not treated as equivalent because N'xxx'
failed to unescape backslashes (\) and
doubled apostrophe/single quote characters
('').
(Bug#17313)
Use of the join cache in favor of an index for ORDER
BY operations could cause incorrect result sorting.
(Bug#17212)
The PASSWORD() function returned
invalid results when used in some
UNION queries.
(Bug#16881)
ORDER BY RAND() LIMIT 1 always set a user
variable to the last possible value from the table.
(Bug#16861)
When performing a GROUP_CONCAT(),
the server transformed BLOB
columns VARCHAR columns, which
could cause erroneous results when using Connector/J and
possibly other MySQL APIs.
(Bug#16712)
Stored procedures did not use the character set defined for the database in which they were created. (Bug#16676)
Some server errors were not reported to the client, causing both to try to read from the connection until a hang or crash resulted. (Bug#16581)
On Windows, a definition for
mysql_set_server_option() was
missing from the C client library.
(Bug#16513)
Updating a column of a FEDERATED table to
NULL sometimes failed.
(Bug#16494)
For SELECT ... FOR
UPDATE statements that used
DISTINCT or GROUP BY over
all key parts of a unique index (or primary key), the optimizer
unnecessarily created a temporary table, thus losing the linkage
to the underlying unique index values. This caused a
Result set not updatable error. (The
temporary table is unnecessary because under these circumstances
the distinct or grouped columns must also be unique.)
(Bug#16458)
Using ANY with “nontable”
subqueries such as SELECT 1 yielded incorrect
results under certain circumstances due to incorrect application
of
MIN()/MAX()
optimization.
(Bug#16302)
A subquery in the WHERE clause of the outer
query and using IN and GROUP
BY returned an incorrect result.
(Bug#16255)
A query could produce different results with and without and
index, if the WHERE clause contained a range
condition that used an invalid
DATETIME constant.
(Bug#16249)
TIMESTAMPDIFF() examined only the
date and ignored the time when the requested difference unit was
months or quarters.
(Bug#16226)
Using tables from MySQL 4.x in MySQL 5.x, in particular those
with VARCHAR fields and using
INSERT DELAYED to update data in
the table would result in either data corruption or a server
crash.
(Bug#16218, Bug#17294, Bug#16611)
The value returned by a stored function returning a string value was not of the declared character set. (Bug#16211)
The
index_merge/Intersection
optimizer could experience a memory overrun when the number of
table columns covered by an index was sufficiently large,
possibly resulting in a server crash.
(Bug#16201)
DECIMAL columns were handled
incorrectly in two respects :
When the precision of the column was too small for the value. In this case, the original value was returned instead of an error.
When the scale of the column was set to 0. In this case, the value. In this case, the value was treated as though the scale had been defined as 2.
Certain queries having a WHERE clause that
included conditions on multi-part keys with more than 2 key
parts could produce incorrect results and send [Note]
Use_count: Wrong count for key at... messages to
STDERR.
(Bug#16168)
When a row was inserted through a view but did not specify a value for a column that had no default value in the base table, no warning or error occurred. Now a warning occurs, or an error in strict SQL mode. (Bug#16110)
When NOW() was used in a
BETWEEN clause of the definition
for a view, it was replaced with a constant in the view.
(Bug#15950)
The C API failed to return a status message when invoking a stored procedure. (Bug#15752)
mysqlimport sends a set
@@character_set_database=binary statement to the
server, but this is not understood by pre-4.1 servers. Now
mysqlimport encloses the statement within a
/*!40101 ... */ comment so that old servers
will ignore it.
(Bug#15690)
For the CSV storage engine, memory-mapped
pages of the data file were not invalidated when new data was
appended to the file via traditional (file descriptor-based) I/O
primitives.
(Bug#15669)
SHOW GRANTS FOR CURRENT_USER did not return
definer grants when executed in DEFINER
context (such as within a stored prodedure defined with
SQL SECURITY DEFINER), it returned the
invoker grants.
(Bug#15298)
The --collation-server server
option was being ignored. With the fix, if you choose a
nondefault character set with
--character-set-server, you
should also use
--collation-server to specify the
collation.
(Bug#15276)
The server crashed if it tried to access a
CSV table for which the data file had been
removed.
(Bug#15205)
Tables created with the FEDERATED storage
engine did not permit indexes using NULL
columns.
(Bug#15133)
When using tables containing
VARCHAR columns created under
MySQL 4.1 with a 5.0 or later server, for some queries the
metadata sent to the client could have an empty column name.
(Bug#14897)
CREATE TABLE ...
SELECT statements that selected
GEOMETRY values resulted in a table that
contained BLOB columns, not
GEOMETRY columns.
(Bug#14807)
When setting a column to its implicit default value as the
result of inserting a NULL into a
NOT NULL column as part of a multi-row insert
or LOAD DATA operation, the
server returned a misleading warning message.
(Bug#14770)
The use of WHERE in col_name IS
NULLSELECT
statements reset the value of
LAST_INSERT_ID() to zero.
(Bug#14553)
Inserts into BIT columns of
FEDERATED tables did not work.
(Bug#14532)
Using SELECT and a table join
while running a concurrent INSERT
operation would join incorrect rows.
(Bug#14400)
Prepared statements caused general log and server memory corruption. (Bug#14346)
libmysqld produced some warnings to
stderr which could not be silenced. These
warnings now are suppressed.
(Bug#13717)
The Instance Manager allowed STOP INSTANCE to
be used on a server instance that was not running.
(Bug#12673)
For very complex SELECT
statements could create temporary tables that were too large,
and for which the temporary files were not removed, causing
subsequent queries to fail.
(Bug#11824)
USE did not refresh database
privileges when employed to re-select the current database.
(Bug#10979)
The type of the value returned by the
VARIANCE() function varied
according to the type of the input value. The function should
always return a DOUBLE value.
(Bug#10966)
The same trigger error message was produced under two conditions: The trigger duplicated an existing trigger name, or the trigger duplicated an existing combination of action and event. Now different messages are produced for the two conditions so as to be more informative. (Bug#10946)
CREATE USER did not respect the
16-character user name limit.
(Bug#10668)
A server or network failure with an open client connection would cause the client to hang even though the server was no longer available.
As a result of this change, the
MYSQL_OPT_READ_TIMEOUT and
MYSQL_OPT_WRITE_TIMEOUT options for
mysql_options() now apply to
TCP/IP connections on all platforms. Previously, they applied
only to Windows.
(Bug#9678)
INSERT INTO ... SELECT ... LIMIT 1 could be
slow because the LIMIT was ignored when
selecting candidate rows.
(Bug#9676)
The optimizer could produce an incorrect result after
AND with collations such as
latin1_german2_ci,
utf8_czech_ci, and
utf8_lithianian_ci.
(Bug#9509)
A stored procedure with a CONTINUE handler
that encountered an error continued to execute a statement that
caused an error, rather with the next statement following the
one that caused the error.
(Bug#8153)
For ODBC compatibility, MySQL supports use of WHERE
for
col_name IS NULLDATE or
DATETIME columns that are
NOT NULL, to allow column values of
'0000-00-00' or '0000-00-00
00:00:00' to be selected. However, this was not
working for WHERE clauses in
DELETE statements.
(Bug#8143)
A user variable set to a value selected from an unsigned column was stored as a signed value. (Bug#7498)
The --with-collation option
was not honored for client connections.
(Bug#7192)
This is a bugfix release for the current production release family. It replaces MySQL 5.0.24.
Bugs fixed:
The shared compatibility RPM files were missing some files. (Bug#22251)
mysqld could crash when closing temporary tables. (Bug#21582)
MySQL 5.0.24 introduced an ABI incompatibility, which this release reverts. Programs compiled against 5.0.24 are not compatible with any other version and must be recompiled. (Bug#21543)
Path name separator and device characters were not correctly parameterized for NetWare, causing mysqld startup errors. (Bug#21537)
Closing of temporary tables failed if binary logging was not enabled. (Bug#20919)
For statements that have a DEFINER clause
such as CREATE TRIGGER or
CREATE VIEW, long user names or
host names could cause a buffer overflow.
(Bug#16899)
This is a bugfix release for the current production release family.
This section documents all changes and bug fixes that have been applied since the last official MySQL release. If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise.
Functionality added or changed:
In the INFORMATION_SCHEMA.ROUTINES
table the ROUTINE_DEFINITION column now is
defined as NULL rather than NOT
NULL. Also, NULL rather than the
empty string is returned as the column value if the user does
not have sufficient privileges to see the routine definition.
(Bug#20230)
The LEFT() and
RIGHT() functions return
NULL if any argument is
NULL.
(Bug#11728)
The innodb_log_arch_dir system
variable (which has been unused since MySQL 4.0.6) is now
deprecated and should no longer be used. It will be removed in
MySQL 5.1.
Program Database files (extension .pdb) are
now included by default in Windows distributions. These can be
used to help diagnose problems with mysqld
and other tools. See Section 21.4.1, “Debugging a MySQL Server”.
Bugs fixed:
Security Fix:
If a user has access to MyISAM table
t, that user can create a
MERGE table m that
accesses t. However, if the user's
privileges on t are subsequently
revoked, the user can continue to access
t by doing so through
m. If this behavior is undesirable,
you can start the server with the new
--skip-merge option to disable
the MERGE storage engine.
(Bug#15195, CVE-2006-4031)
MySQL Cluster:
The ndb_size.pl script did not account for
TEXT and
BLOB column values correctly.
(Bug#21204)
MySQL Cluster:
The repeated creating and dropping of a table would eventually
lead to NDB Error 826,
Too many tables and attributes ... Insufficient
space.
(Bug#20847)
Replication: A race condition during slave server shutdown caused an assert failure. (Bug#20850)
Replication:
With the
auto_increment_increment system
variable set larger than 1, if the next generated
AUTO_INCREMENT value would be larger than the
column's maximum value, the value would be clipped down to that
maximum value and inserted, even if the resulting value would
not be in the generated sequence. This could cause problems for
master-master replication. Now the server clips the value down
to the previous value in the sequence, which correctly produces
a duplicate-key error if that value already exists in the
column.
(Bug#20524)
Replication:
If a table on a slave server had a higher
AUTO_INCREMENT counter than the corresponding
master table (even though all rows of the two tables were
identical), in some cases REPLACE
or INSERT
... ON DUPLICATE KEY UPDATE would not replicate
properly using statement-based logging. (Different values would
be inserted on the master and slave.)
(Bug#20188)
Under heavy load (executing more than 1024 simultaneous complex queries), a problem in the code that handles internal temporary tables could lead to writing beyond allocated space and memory corruption.
Use of more than 1024 simultaneous cursors server wide also could lead to memory corruption. This applies to both stored procedure cursors and C API cursors. (Bug#21206)
Failure to account for a NULL table pointer
on big-endian machines could cause a server crash during type
conversion.
(Bug#21135)
mysqldump sometimes did not select the correct database before trying to dump views from it, resulting in an empty result set that caused mysqldump to die with a segmentation fault. (Bug#21014)
A SELECT that used a subquery in
the FROM clause that did not select from a
table failed when the subquery was used in a join.
(Bug#21002)
REPLACE ...
SELECT for a view required the
INSERT privilege for tables other
than the table being modified.
(Bug#20989)
Issuing a SHOW CREATE FUNCTION or
SHOW CREATE PROCEDURE statement
without sufficient privileges could crash the
mysql client.
(Bug#20664)
In a view defined with SQL SECURITY DEFINER,
the CURRENT_USER() function
returned the invoker, not the definer.
(Bug#20570)
SELECT @@INSERT_ID displayed a value
unrelated to a preceding SET INSERT_ID. (It
was returning LAST_INSERT_ID instead.)
(Bug#20392)
The mysql client did not understand
help commands that had spaces at the end.
(Bug#20328)
mysqldump produced a malformed dump file when dumping multiple databases that contained views. (Bug#20221)
For a DATE parameter sent via a
MYSQL_TIME data structure,
mysql_stmt_execute() zeroed the
hour, minute, and second members of the structure rather than
treating them as read-only.
(Bug#20152)
Performing INSERT ... SELECT ... JOIN ...
USING without qualifying the column names caused
ERROR 1052 "column 'x' in field list is
ambiguous" even in cases where the column references
were unambiguous.
(Bug#18080)
Using the extended syntax for
TRIM() — that is,
TRIM(... FROM ...) — in a
SELECT statement defining a view
caused an invalid syntax error when selecting from the view.
(Bug#17526)
Assignments of values to variables of type
TEXT were handled incorrectly in
stored routines.
(Bug#17225)
DATE_ADD() and
DATE_SUB() returned
NULL when the result date was on the day
'9999-12-31'.
(Bug#12356)
The DATA DIRECTORY table option did not work
for TEMPORARY tables.
(Bug#8706)
Bug#10952 may cause inadvertent data loss. A fix for this bug was included in MySQL 5.0.23, but the approach used caused a loss of intended functionality. Because of this, that fix has been reverted in MySQL 5.0.24. As a consequence, the risk of inadvertent data loss still exists (see Bug#10952).
MySQL 5.0.23 was never officially released.
This section documents all changes and bug fixes that have been applied since the last official MySQL release. If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise.
Functionality added or changed:
Incompatible Change:
For GRANT and REVOKE,
ON * previously granted and revoked
privileges for the default database if there was a default
database and global privileges if there was none. Now
ON * requires a default database and produces
an error if there is none.
Important Change: MySQL Cluster:
The status variables Ndb_connected_host and
Ndb_connected_port were renamed to
Ndb_config_from_host and
Ndb_config_from_port,
respectively.
MySQL Cluster:
The limit of 2048 ordered indexes per cluster has been lifted.
There is now no upper limit on the number of ordered indexes
(including AUTO_INCREMENT columns) that may
be used.
(Bug#14509)
The mysqldumpslow script has been moved from client RPM packages to server RPM packages. This corrects a problem where mysqldumpslow could not be used with a client-only RPM install, because it depends on my_print_defaults which is in the server RPM. (Bug#20216)
Added the
log_queries_not_using_indexes
system variable.
(Bug#19616)
Added the ssl_ca,
ssl_capath,
ssl_cert,
ssl_cipher, and
ssl_key system variables, which
display the values given via the corresponding command options.
See Section 5.5.7.3, “SSL Command Options”.
(Bug#19606)
SQL syntax for prepared statements now supports
ANALYZE TABLE,
OPTIMIZE TABLE, and
REPAIR TABLE.
(Bug#19308)
For a table with an AUTO_INCREMENT column,
SHOW CREATE TABLE now shows the
next AUTO_INCREMENT value to be generated.
(Bug#19025)
The ONLY_FULL_GROUP_BY SQL
mode now also applies to the HAVING clause.
That is, columns not named in the GROUP BY
clause cannot be used in the HAVING clause if
not used in an aggregate function.
(Bug#18739)
Added the --set-charset option
to mysqlbinlog to allow the character set to
be specified for processing binary log files.
(Bug#18351)
The bundled yaSSL library was upgraded to version 1.3.5. This improves handling of certain problems with SSL-related command options. (Bug#17737)
Added the
--ssl-verify-server-cert option
to MySQL client programs. This option causes the server's Common
Name value in its certificate to be verified against the host
name used when connecting to the server, and the connection is
rejected if there is a mismatch. Added
MYSQL_OPT_SSL_VERIFY_SERVER_CERT option for
the mysql_options() C API
function to enable this verification. This feature can be used
to prevent man-in-the-middle attacks. Verification is disabled
by default.
(Bug#17208)
It is now possible to use
NEW.
values within triggers as var_nameINOUT parameters to
stored procedures.
(Bug#14635)
Added the --angel-pid-file option to
mysqlmanager for specifying the file in which
the angel process records its process ID when
mysqlmanager runs in daemon mode.
(Bug#14106)
The mysql_get_ssl_cipher() C API
function was added.
The mysql_upgrade command has been converted from a shell script to a C program, so it is available on non-Unix systems such as Windows. This program should be run for each MySQL upgrade. See Section 4.4.9, “mysql_upgrade — Check Tables for MySQL Upgrade”.
Binary distributions that include SSL support now are built using yaSSL when possible.
Bugs fixed:
Security Fix:
A NUL byte within a comment in a statement
string caused the rest of the string not to be written to the
query log, allowing logging to be bypassed.
(Bug#17667, CVE-2006-0903)
MySQL Cluster:
The ndb_mgm client command ALL
CLUSTERLOG STATISTICS=15 had no effect.
(Bug#20336)
MySQL Cluster:
The failure of a data node when preparing to commit a
transaction (that is, while the node's status was
CS_PREPARE_TO_COMMIT) could cause the failure
of other cluster data nodes.
(Bug#20185)
MySQL Cluster: An internal formatting error caused some management client error messages to be unreadable. (Bug#20016)
MySQL Cluster: Renaming a table in such a way as to move it to a different database failed to move the table's indexes. (Bug#19967)
MySQL Cluster: Running management client commands while mgmd was in the process of disconnecting could cause the management server to fail. (Bug#19932)
MySQL Cluster:
Running ALL START in the
NDB management client or restarting
multiple nodes simultaneously could under some circumstances
cause the cluster to crash.
(Bug#19930)
MySQL Cluster:
TEXT columns in Cluster tables
having both an explicit primary key and a unique key were not
correctly updated by REPLACE
statements.
(Bug#19906)
MySQL Cluster:
The cluster's data nodes failed while trying to load data when
NoOfFrangmentLogFiles was set equal to 1.
(Bug#19894)
MySQL Cluster:
Restoring a backup with ndb_restore failed
when the backup had been taken from a cluster whose
DataMemory had been completely used up.
(Bug#19852)
MySQL Cluster:
Resources for unique indexes on Cluster table columns were
incorrectly allocated, so that only one-fourth as many unique
indexes as indicated by the value of
UniqueHashIndexes could be created.
(Bug#19623)
MySQL Cluster:
(NDBAPI): On big-endian platforms,
NdbOperation::write_attr() did not update
32-bit fields correctly.
(Bug#19537)
MySQL Cluster:
LOAD DATA
LOCAL failed to ignore duplicate keys in Cluster
tables.
(Bug#19496)
MySQL Cluster: For ndb_mgmd, Valgrind revealed problems with a memory leak and a dependency on an uninitialized variable. (Bug#19318, Bug#20333)
MySQL Cluster:
A problem with error handling when
ndb_use_exact_count was enabled
could lead to incorrect values returned from queries using
COUNT(). A warning is now
returned in such cases.
(Bug#19202)
MySQL Cluster:
TRUNCATE failed on tables having
BLOB or
TEXT columns with the error
Lock wait timeout exceeded.
(Bug#19201)
MySQL Cluster:
mysql-test-run.pl started
NDB even for test cases that did
not need it.
(Bug#19083)
MySQL Cluster: Stopping multiple nodes could cause node failure handling not to be completed. (Bug#19039)
MySQL Cluster:
The management client ALL STOP command shut
down mgmd processes (as well as
ndbd processes).
(Bug#18966)
MySQL Cluster:
TRUNCATE
TABLE failed to reset the
AUTO_INCREMENT counter.
(Bug#18864)
MySQL Cluster:
Repeated CREATE -
INSERT - DROP
operations on tables could in some circumstances cause the MySQL
table definition cache to become corrupt, so that some
mysqld processes could access table
information but others could not.
(Bug#18595)
MySQL Cluster:
Repeated use of the SHOW and
ALL STATUS commands in the
ndb_mgm client could cause the
mgmd process to crash.
(Bug#18591)
MySQL Cluster: ndbd sometimes failed to start with the error Node failure handling not completed following a graceful restart. (Bug#18550)
MySQL Cluster:
Backups could fail for large clusters with many tables, where
the number of tables approached
MaxNoOfTables.
(Bug#17607)
MySQL Cluster:
An issue with ndb_mgmd prevented more than 27
mysqld processes from connecting to a single
cluster at one time.
(Bug#17150)
MySQL Cluster:
Using “stale” mysqld
.FRM files could cause a newly-restored
cluster to fail. This situation could arise when restarting a
MySQL Cluster using the --initial option while
leaving connected mysqld processes running.
(Bug#16875)
MySQL Cluster: Data node failures could cause excessive CPU usage by ndb_mgmd. (Bug#13987)
MySQL Cluster: Cluster system status variables were not updated properly. (Bug#11459)
MySQL Cluster:
Some queries having a WHERE clause of the
form c1=val1 OR c2 LIKE 'val2' were not
evaluated correctly. (Bug # 17421)
MySQL Cluster: (NDBAPI): Update operations on blobs were not checked for illegal operations.
Read locks with blob update operations are now upgraded from read committed to read shared.
Replication:
Valgrind revealed several issues with mysqld
that were corrected: A dangling stack pointer being overwritten;
possible uninitialized data in a string comparison; memory
corruption in replication slaves when switching databases;
syscall() write parameter pointing to an
uninitialized byte.
(Bug#19022, Bug#20579, Bug#20769, Bug#20783, Bug#20791)
Replication:
The binary log would create an incorrect DROP
query when creating temporary tables during replication.
(Bug#17263)
Replication:
An invalid GRANT statement for
which Ok was returned on a replication master
caused an error on the slave and replication to fail.
(Bug#6774)
A buffer overwrite error in Instance Manager caused a crash. (Bug#20622)
On Windows, temporary tables containing “
: ” in the name could not be created.
(Bug#20616)
The fill_help_tables.sql file did not
contain a SET NAMES 'utf8' statement to
indicate its encoding. This caused problems for some settings of
the MySQL character set such as big5.
(Bug#20551)
The fill_help_tables.sql file did not load
properly if the ANSI_QUOTES
SQL mode was enabled.
(Bug#20542)
mysql_upgrade was missing from binary MySQL distributions. (Bug#20403, Bug#18516, Bug#20556)
Several aspects of view privileges were being checked incorrectly. (Bug#20363, Bug#18681)
Queries using an indexed column as the argument for the
MIN() and
MAX() functions following an
ALTER TABLE .. DISABLE KEYS statement
returned Got error 124 from storage
engine until ALTER TABLE ... ENABLE
KEYS was run on the table.
(Bug#20357)
The thread for INSERT DELAYED
rows was maintaining a separate
AUTO_INCREMENT counter, resulting in
incorrect values being assigned if DELAYED
and non-DELAYED inserts were mixed.
(Bug#20195)
On Linux, libmysqlclient when compiled with
yaSSL using the icc compiler had a spurious
dependency on C++ libraries.
(Bug#20119)
A number of dependency issues in the RPM
bench and test packages
caused installation of these packages to fail.
(Bug#20078)
A compatibility issue with NPTL (Native POSIX Thread Library) on
Linux could result in a deadlock with
FLUSH TABLES WITH READ
LOCK under some conditions.
(Bug#20048)
Some outer joins were incorrectly converted to inner joins. (Bug#19816)
This regression was introduced by Bug#17146.
CREATE DATABASE, RENAME
DATABASE, and DROP
DATABASE could deadlock in cases where there was a
global read lock.
(Bug#19815)
The WITH CHECK OPTION was not enforced when a
REPLACE statement was executed
against a view.
(Bug#19789)
Multiple-table updates with FEDERATED tables
could cause a server crash.
(Bug#19773)
InnoDB unlocked its data directory before
committing a transaction, potentially resulting in
nonrecoverable tables if a server crash occurred before the
commit.
(Bug#19727)
Subqueries that produced a BIGINT UNSIGNED
value were being treated as returning a signed value.
(Bug#19700)
GROUP BY on an expression that contained a
cast to DECIMAL produced an
incorrect result.
(Bug#19667)
MERGE tables did not work reliably with
BIT columns.
(Bug#19648)
Re-execution of a prepared multiple-table
DELETE statement that involves a
trigger or stored function can result in a server crash.
(Bug#19634)
The range operator failed and caused a server crash for clauses
of the form
.
(Bug#19618)tbl_name.unsigned_keypart
NOT IN (negative_const,
...)
CHECK TABLE on a
MyISAM table briefly cleared its
AUTO_INCREMENT value, while holding only a
read lock. Concurrent inserts to that table could use the wrong
AUTO_INCREMENT value.
CHECK TABLE no longer modifies
the AUTO_INCREMENT value.
(Bug#19604)
Using
CONCAT(@, where
user_var,
col_name)col_name is a column in an
INFORMATION_SCHEMA table, could cause
erroneous duplication of data in the query result.
(Bug#19599)
Some yaSSL public function names conflicted with those from
OpenSSL, causing conflicts for applications that linked against
both OpenSSL and a version of libmysqlclient
that was built with yaSSL support. The yaSSL public functions
now are renamed to avoid this conflict.
(Bug#19575)
A view definition that referred to an alias in the
HAVING clause could be saved in the
.frm file with the alias replaced by the
expression that it referred to, causing failure of subsequent
SELECT * FROM statements.
(Bug#19573)view_name
mysql displayed NULL for
strings that are empty or contain only spaces.
(Bug#19564)
InnoDB failed to increment the
handler_read_prev counter.
(Bug#19542)
Selecting from a view that used GROUP BY on a
nonconstant temporal interval (such as
DATE(
could cause a server crash.
(Bug#19490)col) + INTERVAL
TIME_TO_SEC(col) SECOND
mysqldump did not dump the table name
correctly for some table identifiers that contained unusual
characters such as “ : ”.
(Bug#19479)
On 64-bit Windows systems, REGEXP for regular
expressions with exactly 31 characters did not work.
(Bug#19407)
An outer join of two views that was written using { OJ
... } syntax could cause a server crash.
(Bug#19396)
Race conditions on certain platforms could cause the Instance Manager to fail to initialize. (Bug#19391)
Use of the --no-pager option
caused mysql to crash.
(Bug#19363)
In the INFORMATION_SCHEMA.COLUMNS
table, the values for the
CHARACTER_MAXIMUM_LENGTH and
CHARACTER_OCTET_LENGTH columns were incorrect
for multi-byte character sets.
(Bug#19236)
Multiple-table DELETE statements
containing a subquery that selected from one of the tables being
modified caused a server crash.
(Bug#19225)
On Windows, removal of binary log files would fail if the files were already open. (Bug#19208)
Flushing the compression buffer (via
FLUSH TABLE) no
longer increases the size of an unmodified
ARCHIVE table.
(Bug#19204)
An ALTER TABLE operation that
does not need to copy data, when executed on a table created
prior to MySQL 4.0.25, could result in a server crash for
subsequent accesses to the table.
(Bug#19192)
SSL connections using yaSSL on OpenBSD could fail. (Bug#19191)
Attempting to set the default value of an
ENUM or SET
column to NULL caused a server crash.
(Bug#19145)
Use of uninitialized user variables in a subquery in the
FROM clause resulted in invalid entries in
the binary log.
(Bug#19136)
A CREATE TABLE statement that
created a table from a materialized view did not inherit default
values from the underlying table.
(Bug#19089)
Index prefixes for utf8
VARCHAR columns did not work for
UPDATE statements.
(Bug#19080)
Premature optimization of nested subqueries in the
FROM clause that refer to aggregate functions
could lead to incorrect results.
(Bug#19077)
The parser leaked memory when its stack needed to be extended. (Bug#18930)
BIT columns in a table could
cause joins that use the table to fail.
(Bug#18895)
The MySQL server startup script /etc/init.d/mysql (created from mysql.server) is now marked to ensure that the system services ypbind, nscd, ldap, and NTP are started first (if these are configured on the machine). (Bug#18810)
The COM_STATISTICS command was changed in
5.0.3 to display session status variable values rather than
global values. This causes mysqladmin status
information not to be useful for the Slow
queries and Opens values. Now
COM_STATISTICS displays the global values for
Slow queries and Opens.
(Bug#18669)
LOAD DATA FROM MASTER would fail when trying
to load the INFORMATION_SCHEMA database from
the master, because the INFORMATION_SCHEMA
system database would already exist on the slave.
(Bug#18607)
BLOB or
TEXT arguments to or values
returned from stored functions were not copied properly if too
long and could become garbled.
(Bug#18587)
The IN-to-EXISTS
transformation was making a reference to a parse tree fragment
that was left out of the parse tree. This caused problems with
prepared statements.
(Bug#18492)
mysqldump produced garbled output for view definitions. (Bug#18462)
The configuration information for building the embedded server on Windows was missing a file. (Bug#18455)
In mysqltest, --sleep=0 had
no effect. Now it correctly causes sleep
commands in test case files to sleep for 0 seconds.
(Bug#18312)
INFORMATION_SCHEMA.TABLES provided
inconsistent info about invalid views. This could cause server
crashes or result in incorrect data being returned for queries
that attempt to obtain information from
INFORMATION_SCHEMA tables about views using
stored functions.
(Bug#18282)
On Windows, corrected a crash stemming from differences in Visual C runtime library routines from POSIX behavior regarding invalid file descriptors. (Bug#18275)
On Windows, terminating mysqld with Control-C could result in a crash during shutdown. (Bug#18235)
Selecting data from a MEMORY table with a
VARCHAR column and a
HASH index over it returned only the first
row matched.
(Bug#18233)
The use of MIN() and
MAX() on columns with an index
prefix produced incorrect results in some queries.
(Bug#18206)
An entry in the mysql.proc table with an
empty routine name caused access to the
INFORMATION_SCHEMA.ROUTINES table
to crash the server.
(Bug#18177)
A UNION over more than 128
SELECT statements that use an
aggregate function failed.
(Bug#18175)
Updates to a MEMORY table caused the size of
BTREE indexes for the table to increase.
(Bug#18160)
SELECT
DISTINCT queries sometimes returned only the last row.
(Bug#18068)
Returning the value of a system variable from a stored function caused a server crash. (Bug#18037)
An update that used a join of a table to itself and modified the table on both sides of the join reported the table as crashed. (Bug#18036)
Race conditions on certain platforms could cause the Instance Manager to try to restart the same instance multiple times. (Bug#18023)
For a reference to a nonexistent index in FORCE
INDEX, the error message referred to a column, not an
index.
(Bug#17873)
The sql_big_selects system
variable was not displayed by SHOW
VARIABLES.
(Bug#17849)
REPAIR TABLE did not restore the
length for packed keys in tables created under MySQL 4.x, which
caused them to appear corrupt to CHECK
TABLE but not to REPAIR
TABLE.
(Bug#17810)
Results from
INFORMATION_SCHEMA.SCHEMATA could
contain uppercase information when
lower_case_table_names was not
0.
(Bug#17661)
CREATE TABLE ...
SELECT did not always produce the proper column
default value in TRADITIONAL
SQL mode.
(Bug#17626)
A range access optimizer heuristic was invalid, causing some queries to be much slower in MySQL 5.0 than in 4.0. (Bug#17379, Bug#18940)
mysqldump would not dump views that had
become invalid because a table named in the view definition had
been dropped. Instead, it quit with an error message. Now you
can specify the --force option to cause
mysqldump to keep going and write an SQL
comment containing the view definition to the dump output.
(Bug#17371)
The --core-file-size option
for mysqld_safe was effective only for
root.
(Bug#17353)
On Windows, multiple clients simultaneously attempting to
perform ALTER TABLE operations on
an InnoDB table could deadlock.
(Bug#17264)
Revised memory allocation for local objects within stored functions and triggers to avoid memory leak for repeated function or trigger invocation. (Bug#17260)
Multiple calls to a stored procedure that selects from
INFORMATION_SCHEMA could cause a server
crash.
(Bug#17204)
Views created from prepared statements inside of stored
procedures were created with a definition that included both
SQL_CACHE and
SQL_NO_CACHE.
(Bug#17203)
mysqldump wrote an extra pair of
DROP DATABASE and
CREATE DATABASE statements if run
with the --add-drop-database
option and the database contained views.
(Bug#17201)
A Table ... doesn't exist error could occur for statements that called a function defined in another database. (Bug#17199)
For certain CREATE
TABLE ... SELECT statements, the selected values were
truncated when inserted into the new table.
(Bug#17048)
ALTER TABLE on a table created
prior to 5.0.3 would cause table corruption if the
ALTER TABLE did one of the
following:
Change the default value of a column.
Change the table comment.
Change the table password.
MyISAM table deadlock was possible if one
thread issued a LOCK TABLES
request for write locks and then an administrative statement
such as OPTIMIZE TABLE, if
between the two statements another client meanwhile issued a
multiple-table SELECT for some of
the locked tables.
(Bug#16986)
Symlinking .mysql_history to
/dev/null to suppress statement history
saving by mysql did not work.
(mysql deleted the symlink and recreated
.mysql_history as a regular file, and then
wrote history to it.)
(Bug#16803)
Concatenating the results of multiple constant subselects produced incorrect results. (Bug#16716)
Privilege checking on the contents of the
INFORMATION_SCHEMA.VIEWS table was
insufficiently restrictive.
(Bug#16681)
mysqlcheck tried to check views instead of ignoring them. (Bug#16502)
IS_USED_LOCK() could return an
incorrect connection identifier.
(Bug#16501)
Concurrent reading and writing of privilege structures could crash the server. (Bug#16372)
Grant table modifications sometimes did not refresh the
in-memory tables if the host name was '' or
not specified.
(Bug#16297)
The sql_notes and
sql_warnings system variables
were not always displayed correctly by SHOW
VARIABLES (for example, they were displayed as
ON after being set to
OFF).
(Bug#16195)
The max_length metadata value for columns
created from CONCAT() could be
incorrect when the collation of an argument differed from the
collation of the CONCAT() itself.
In some contexts such as UNION,
this could lead to truncation of the column contents.
(Bug#15962)
The server no longer uses a signal handler for signal 0 because it could cause a crash on some platforms. (Bug#15869)
InnoDB does not support
SPATIAL indexes, but did not prevent creation
of such an index.
(Bug#15860)
Long multiple-row INSERT
statements could take a very long time for some multi-byte
character sets.
(Bug#15811)
The system_time_zone and
version_* system variables could not be
accessed via SELECT
@@ syntax.
(Bug#15684, Bug#12792)var_name
EXPLAIN ... SELECT INTO caused the client to
hang.
(Bug#15463)
Nested natural joins worked executed correctly when executed as
a nonprepared statement could fail with an Unknown
column ' error when executed as a prepared statement, due
to a name resolution problem.
(Bug#15355)col_name' in 'field
list'
The MD5() and
SHA() functions
treat their arguments as case-sensitive strings. But when they
are compared, their arguments were compared as case-insensitive
strings, which leads to two function calls with different
arguments (and thus different results) compared as being
identical. This can lead to a wrong decision made in the range
optimizer and thus to an incorrect result set.
(Bug#15351)
Invalid escape sequences in option files caused MySQL programs that read them to abort. (Bug#15328)
Re-executing a stored procedure with a complex stored procedure cursor query could lead to a server crash. (Bug#15217)
CREATE TABLE ...
SELECT ... statements that used a stored function
explicitly or implicitly (through a view) resulted in a
Table not locked error.
(Bug#15137, Bug#12472)
An invalid comparison between keys with index prefixes over
multi-byte character fields could lead to incorrect result sets
if the selected query execution plan used a range scan by an
index prefix over a UTF8 character field.
This also caused incorrect results under similar circumstances
with many other character sets.
(Bug#14896)
A view with a nonexistent account in the
DEFINER clause caused
SHOW CREATE VIEW to fail. Now
SHOW CREATE VIEW issues a warning
instead.
(Bug#14875)
For BOOLEAN mode full-text
searches on nonindexed columns, NULL rows
generated by a LEFT JOIN caused incorrect
query results.
(Bug#14708, Bug#25637)
SHOW CREATE TABLE did not display
the AUTO_INCREMENT column attribute if the
SQL mode was MYSQL323 or
MYSQL40. This also affected
mysqldump, which uses
SHOW CREATE TABLE to get table
definitions.
(Bug#14515)
Some queries were slower in 5.0 than in 4.1 because some 4.1 cost-evaluation code had not been merged into 5.0. (Bug#14292)
The binary log lacked character set information for table names when dropping temporary tables. (Bug#14157)
The result from CONV() is a
string, but was not always treated the same way as a string when
converted to a real value for an arithmetic operation.
(Bug#13975)
RPM packages had spurious dependencies on Perl modules and other programs. (Bug#13634)
REPLACE statements caused
activation of UPDATE triggers,
not DELETE and
INSERT triggers.
(Bug#13479)
With settings of
read_buffer_size >= 2G and
read_rnd_buffer_size >=2G,
LOAD DATA
INFILE failed with no error message or caused a server
crash for files larger than 2GB.
(Bug#12982)
A B-TREE index on a MEMORY
table erroneously reported duplicate entry error for multiple
NULL values.
(Bug#12873)
Use of CONVERT_TZ() in a stored
function or trigger (or in a stored procedure called from a
stored function or trigger) caused an error.
(Bug#11081)
LOAD_FILE() returned an error if
the file did not exist, rather than NULL as
it should according to the manual.
(Bug#10418)
When myisamchk needed to rebuild a table,
AUTO_INCREMENT information was lost.
(Bug#10405)
For certain CREATE VIEW
statements, the server did not detect invalid subqueries within
the SELECT part.
(Bug#7549)
Within a trigger, SET used the SQL mode of
the invoking statement, not the mode in effect at trigger
creation time.
(Bug#6951)
Some queries that used ORDER BY and
LIMIT performed quickly in MySQL 3.23, but
slowly in MySQL 4.x/5.x due to an optimizer problem.
(Bug#4981)
The basedir and
tmpdir system variables could not be accessed
via @@
syntax.
(Bug#1039)var_name
This is a security fix release for the previous production release family.
This section documents all changes and bug fixes that have been applied since the last official MySQL release. If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise.
Bugs fixed:
Security Fix:
An SQL-injection security hole has been found in multi-byte
encoding processing. The bug was in the server, incorrectly
parsing the string escaped with the
mysql_real_escape_string() C API
function.
This vulnerability was discovered and reported by Josh Berkus
<[email protected]> and Tom Lane
<[email protected]> as part of the inter-project
security collaboration of the OSDB consortium. For more
information about SQL injection, please see the following text.
Discussion.
An SQL injection security hole has been found in multi-byte
encoding processing. An SQL injection security hole can
include a situation whereby when a user supplied data to be
inserted into a database, the user might inject SQL statements
into the data that the server will execute. With regards to
this vulnerability, when character set-unaware escaping is
used (for example, addslashes() in PHP), it
is possible to bypass the escaping in some multi-byte
character sets (for example, SJIS, BIG5 and GBK). As a result,
a function such as addslashes() is not able
to prevent SQL-injection attacks. It is impossible to fix this
on the server side. The best solution is for applications to
use character set-aware escaping offered by a function such
mysql_real_escape_string().
However, a bug was detected in how the MySQL server parses the
output of
mysql_real_escape_string(). As a
result, even when the character set-aware function
mysql_real_escape_string() was
used, SQL injection was possible. This bug has been fixed.
Workarounds.
If you are unable to upgrade MySQL to a version that includes
the fix for the bug in
mysql_real_escape_string()
parsing, but run MySQL 5.0.1 or higher, you can use the
NO_BACKSLASH_ESCAPES SQL
mode as a workaround. (This mode was introduced in MySQL
5.0.1.) NO_BACKSLASH_ESCAPES
enables an SQL standard compatibility mode, where backslash is
not considered a special character. The result will be that
queries will fail.
To set this mode for the current connection, enter the following SQL statement:
SET sql_mode='NO_BACKSLASH_ESCAPES';
You can also set the mode globally for all clients:
SET GLOBAL sql_mode='NO_BACKSLASH_ESCAPES';
This SQL mode also can be enabled automatically when the server
starts by using the command-line option
--sql-mode=NO_BACKSLASH_ESCAPES
or by setting sql-mode=NO_BACKSLASH_ESCAPES
in the server option file (for example,
my.cnf or my.ini,
depending on your system).
(Bug#8378, CVE-2006-2753)
See also Bug#8303.
Replication:
The dropping of a temporary table whose name contained a
backtick ('`') character was not correctly
written to the binary log, which also caused it not to be
replicated correctly.
(Bug#19188)
The client libraries were not compiled for position-independent code on Solaris-SPARC and AMD x86_64 platforms. (Bug#18091, Bug#13159, Bug#14202)
Running myisampack followed by
myisamchk with the
--unpack option would corrupt
the auto_increment key.
(Bug#12633)
The patch for Bug#8303 broke the fix for Bug#8378 and was reverted.
In string literals with an escape character
(\) followed by a multi-byte character that
had (\) as its second byte, the literal was
not interpreted correctly. Now only next byte now is escaped,
and not the entire multi-byte character. This means it is a
strict reverse of the
mysql_real_escape_string()
function.
This MySQL 5.0.21 release includes the patches for recently
reported security vulnerabilites in the MySQL client-server
protocol. We would like to thank Stefano Di Paola
<[email protected]> for finding and
reporting these to us.
This is a bugfix release for the current production release family.
This section documents all changes and bug fixes that have been applied since the last official MySQL release. If you would like to receive more fine-grained and personalized update alerts about fixes that are relevant to the version and features you use, please consider subscribing to MySQL Enterprise (a commercial MySQL offering). For more details please see http://www.mysql.com/products/enterprise.
Functionality added or changed:
Security Enhancement:
Added the global
max_prepared_stmt_count system
variable to limit the total number of prepared statements in the
server. This limits the potential for denial-of-service attacks
based on running the server out of memory by preparing huge
numbers of statements. The current number of prepared statements
is available through the
prepared_stmt_count system
variable.
(Bug#16365)
MySQL Cluster:
It is now possible to perform a partial start of a cluster. That
is, it is now possible to bring up the cluster without first
running ndbd --initial on
all configured data nodes.
(Bug#18606)
MySQL Cluster:
Added the --nowait-nodes startup option for
ndbd, making it possible to skip specified
nodes without waiting for them to start when starting the
cluster. See
Section 17.6.22.2, “Program Options for ndbd”.
MySQL Cluster:
It is now possible to install MySQL with Cluster support to a
nondefault location and change the search path for font
description files using either the
--basedir or
--character-sets-dir options.
(Previously in MySQL 5.0, ndbd searched only
the default path for character sets.)
Packaging:
The
MySQL-shared-compat-5.0.
shared compatibility RPMs no longer contain libraries for MySQL
5.1. This avoids a conflict because the 5.0 and 5.1 libraries
share the same soname number. They now contain libraries for
MySQL 3.23, 4.0, 4.1, and 5.0 only.
(Bug#19288)X-.i386.rpm
The default for the
innodb_thread_concurrency
system variable was changed to 8.
(Bug#15868)
Server and clients ignored the --sysconfdir
option that was passed to configure. The
directory specified by this option, if set, now is used as one
of the standard locations in which to look for option files.
(Bug#15069)
In result set metadata, the
MYSQL_FIELD.length value for
BIT columns now is reported in
number of bits. For example, the value for a
BIT(9) column is 9. (Formerly, the value was
related to number of bytes.)
(Bug#13601)
Bugs fixed:
Security Fix:
Invalid arguments to
DATE_FORMAT() caused a server
crash. Thanks to Jean-David Maillefer for discovering and
reporting this problem to the Debian project and to Christian
Hammers from the Debian Team for notifying us of it.
(Bug#20729, CVE-2006-3469)
Security Fix:
A malicious client, using specially crafted invalid
COM_TABLE_DUMP packets was able to trigger an
exploitable buffer overflow on the server. Thanks to Stefano Di
Paola <[email protected]> for finding and
reporting this bug.
(CVE-2006-1518)
Security Fix:
A malicious client, using specially crafted invalid login or
COM_TABLE_DUMP packets was able to read
uninitialized memory, which potentially, though unlikely in
MySQL, could have led to an information disclosure. (, ) Thanks
to Stefano Di Paola <[email protected]> for
finding and reporting this bug.
(CVE-2006-1516, CVE-2006-1517)
MySQL Cluster:
A simultaneous DROP TABLE and
table update operation utilising a table scan could trigger a
node failure.
(Bug#18597)
MySQL Cluster: When multiple node restarts were attempted without allowing each restart to complete, the error message returned was Array index out of bounds rather than Too many crashed replicas. (Bug#18349)
MySQL Cluster:
In a 2-node cluster with a node failure, restarting the node
with a low value for StartPartialTimeout
could cause the cluster to come up partitioned
(“split-brain” issue).
A similar issue could occur when the cluster was first started with a sufficiently low value for this parameter. (Bug#16447, Bug#18612)
MySQL Cluster: On systems with multiple network interfaces, data nodes would get “stuck” in startup phase 2 if the interface connecting them to the management server was working on node startup while the interface interconnecting the data nodes experienced a temporary outage. (Bug#15695)
MySQL Cluster:
On slow networks or CPUs, the management client
SHOW command could sometimes
erroneously show all data nodes as being master nodes belonging
to nodegroup 0.
(Bug#15530)
MySQL Cluster:
TRUNCATE did not reset the
AUTO_INCREMENT counter for
MyISAM tables when issued inside a stored
procedure.
This bug did not affect InnoDB tables.
In addition, TRUNCATE does not
reset the AUTO_INCREMENT counter for
NDB tables regardless of when it
is called.
See also Bug#18864.
MySQL Cluster:
Unused open handlers for tables in which the metadata had
changed were not properly closed. This could result in stale
results from NDB tables following
an ALTER TABLE statement.
(Bug#13228)
MySQL Cluster: Uninitialized internal variables could lead to unexpected results. (Bug#11033, Bug#11034)
MySQL Cluster:
When attempting to create an index on a
BIT or
BLOB column, Error
743: Unsupported character set in table or index was
returned instead of Error 906: Unsupported attribute
type in index.
Replication:
CREATE VIEW statements would not
be replicated to the slave if the
--replicate-wild-ignore-table
rule was enabled.
(Bug#18715)
Replication:
Updating a field value when also requesting a lock with
GET_LOCK() would cause slave
servers in a replication environment to terminate.
(Bug#17284)
InnoDB could read a delete mark from its
system tables incorrectly.
(Bug#19217)
Corrected a syntax error in mysql-test-run.sh. (Bug#19190)
Index corruption could occur in cases when
key_cache_block_size was not a
multiple of the myisam-block-size
value (for example, with
--key_cache_block_size=1536 and
--myisam-block-size=1024).
(Bug#19079)
The optimizer could cause a server crash or use a nonoptimal
subset of indexes when evaluating whether to use Index
Merge/Intersection variant of
index_merge optimization.
(Bug#19021)
A missing DBUG_RETURN() caused the server
to emit a spurious error message: missing DBUG_RETURN
or DBUG_VOID_RETURN macro in function "open_table".
(Bug#18964)
Creating a table in an InnoDB database with a
column name that matched the name of an internal
InnoDB column (including
DB_ROW_ID, DB_TRX_ID,
DB_ROLL_PTR and DB_MIX_ID)
would cause a crash. MySQL now returns Error 1005
Cannot create table with
errno set to -1.
(Bug#18934)
MySQL would not compile on Linux distributions that use the
tinfo library.
(Bug#18912)
mysql_reconnect() sent a SET
NAMES statement to the server, even for pre-4.1
servers that do not understand the statement.
(Bug#18830)
For a reference to a nonexistent stored function in a stored
routine that had a CONTINUE handler, the
server continued as though a useful result had been returned,
possibly resulting in a server crash.
(Bug#18787)
For single-SELECT union
constructs of the form (SELECT ... ORDER BY
order_list1 [LIMIT
n]) ORDER BY
order_list2, the ORDER
BY lists were concatenated and the
LIMIT clause was ignored.
(Bug#18767)
Conversion of a number to a CHAR UNICODE
string returned an invalid result.
(Bug#18691)
UNCOMPRESS(NULL) could cause
subsequent UNCOMPRESS() calls to
return NULL for legal
non-NULL arguments.
(Bug#18643)
If the second or third argument to
BETWEEN was a constant expression
such as '2005-09-01 - INTERVAL 6 MONTH and
the other two arguments were columns,
BETWEEN was evaluated incorrectly.
(Bug#18618)
A LOCK TABLES statement that
failed could cause MyISAM not to update table
statistics properly, causing a subsequent
CHECK TABLE to report table
corruption.
(Bug#18544)
The yaSSL library returned a cipher list in a manner incompatible with OpenSSL. (Bug#18399)
InnoDB did not use a consistent read for
CREATE ... SELECT when
innodb_locks_unsafe_for_binlog
was set.
(Bug#18350)
DROP DATABASE did not drop stored
routines associated with the database if the database name was
longer than 21 characters.
(Bug#18344)
The euro sign (€) was not stored
correctly in columns using the
latin1_german1_ci or
latin1_general_ci collation.
(Bug#18321)
A recent change caused the mysql client not
to display NULL values correctly and to
display numeric columns left-justified rather than
right-justified. The problems have been corrected.
(Bug#18265)
COUNT(*) on a
MyISAM table could return different results
for the base table and a view on the base table.
(Bug#18237)
EXTRACT(QUARTER FROM
returned unexpected
results.
(Bug#18100)date)
Executing SELECT on a large table
that had been compressed within myisampack
could cause a crash.
(Bug#17917)
Casting a string to DECIMAL
worked, but casting a trimmed string (using
LTRIM() or
RTRIM()) resulted in loss of
decimal digits.
(Bug#17043)
mysql-test-run could not be run as
root.
(Bug#17002)
Queries of the form SELECT DISTINCT
did not return
all matching rows.
(Bug#16710)timestamp_column WHERE
date_function(timestamp_col)
= constant
IA-64 RPM packages for Red Hat and SuSE Linux that were built with the icc compiler incorrectly depended on icc runtime libraries. (Bug#16662)
MySQL-shared-compat-5.0.13-0.i386.rpm,
MySQL-shared-compat-5.0.15-0.i386.rpm,
MySQL-shared-compat-5.0.18-0.i386.rpm,
MySQL-shared-compat-5.0.19-0.i386.rpm,
MySQL-shared-compat-5.0.20-0.i386.rpm, and
MySQL-shared-compat-5.0.20a-0.i386.rpm
incorrectly depended on glibc 2.3 and could
not be installed on a glibc 2.2 system.
(Bug#16539)
The presence of multiple equalities in a condition after reading a constant table could cause the optimizer not to use an index. This resulted in certain queries being much slower than in MySQL 4.1. (Bug#16504)
Within a trigger, CONNECTION_ID()
did not return the connection ID of the thread that caused the
trigger to be activated.
(Bug#16461)
For tables created in a MySQL 4.1 installation upgraded to MySQL 5.0 and up, multiple-table updates could update only the first matching row. (Bug#16281)
A query using WHERE (column_1,
column_2) IN
((value_1,
value_2)[, (..., ...), ...]) would
return incorrect results.
(Bug#16248)
For mysql.server, if the
basedir option was specified after
datadir in an option file, the setting for
datadir was ignored and assumed to be located
under basedir.
(Bug#16240)
If the first argument to BETWEEN
was a DATE or
TIME column of a view and the
other arguments were constants,
BETWEEN did not perform conversion
of the constants to the appropriate temporary type, resulting in
incorrect evaluation.
(Bug#16069)
After calling FLUSH STATUS, the
max_used_connections variable did not
increment for existing connections and connections which use the
thread cache.
(Bug#15933)
Lettercase in database name qualifiers was not consistently
handled properly in queries when
lower_case_table_names was set
to 1.
(Bug#15917)
DELETE and
UPDATE statements that used large
NOT IN
( clauses could
use large amounts of memory.
(Bug#15872)value_list)
InnoDB failure to release an adaptive hash
index latch could cause a server crash if the query cache was
enabled.
(Bug#15758)
LAST_INSERT_ID() in a stored
function or trigger returned zero. .
(Bug#15728)
DELETE with LEFT
JOIN for InnoDB tables could crash
the server if
innodb_locks_unsafe_for_binlog
was enabled.
(Bug#15650)
When running a query that contained a
GROUP_CONCAT(SELECT GROUP_CONCAT(...)
), the result was NULL except in
the ROLLUP part of the result, if there was
one.
(Bug#15560)
Use of CONVERT_TZ() in a view
definition could result in spurious syntax or access errors.
(Bug#15153)
CAST( for large
double AS
SIGNED INT)double values outside the signed
integer range truncated the result to be within range, but the
result sometimes had the wrong sign, and no warning was
generated.
(Bug#15098)
For InnoDB tables, an expression of the form
col_name BETWEEN
col_name2 - INTERVAL
x DAY AND
col_name2 + INTERVAL
x DAY
Prevent recursive views caused by using
RENAME TABLE on a view after
creating it.
(Bug#14308)
INSERT DELAYED into a view caused
an infinite loop.
(Bug#13683)
Avoid trying to include
<asm/atomic.h> when it doesn't work
in C++ code.
(Bug#13621)
Within stored routines, user names were parsed incorrectly if they were enclosed within quotes. (Bug#13310)
The server was always built as though
--with-extra-charsets=complex
had been specified.
(Bug#12076)
This is a bugfix release for the current production release family. It replaces MySQL 5.0.20.
Additional information about SSL support
Please note that the original 5.0.20 announcement included
inexact wording: SSL support is “included” in both
server and client, but by default not “enabled”.
SSL can be enabled by passing the SSL-related options
(--ssl,
--ssl-key=...,
--ssl-cert=...,
--ssl-ca=...) when starting the
server and the client or by specifying these options in an
option file. For more information, see
Section 5.5.7, “Using SSL for Secure Connections”.
With version 5.0.20a, SSL support is contained in all binaries
for all Unix (including Linux) and Windows platforms except AIX,
HP-UX, OpenServer 6, and the RPMs specific for RHAS3/RHAS4/SLES9
on Itanium CPUs (ia64); It is also not
contained in those for Novell Netware. We are trying to add
these platforms in future versions.
Bugs fixed:
The fix for “Command line options are ignored for mysql client” has been revoked because it introduced an incompatible change in the way the mysql command-line client selects the server to connect to. In the worst case, this might have led to a client issuing commands to a server for which they were not intended, and this must not happen. To help all users in understanding this subject, Section 4.2.1, “Invoking MySQL Programs” now includes additional explanation of how command options function with regard to host selection. (Bug#16855)
The code of the yaSSL library has been
improved to avoid a dependency on a C++ runtime library, so a
link with pure C applications is now possible on additional (but
not yet all) platforms. We are working on fixing the remaining
issues.
Functionality added or changed:
MySQL Cluster:
The NDBCLUSTER storage engine now
supports INSERT
IGNORE and REPLACE
statements. Previously, these statements failed with an error.
(Bug#17431)
Replication:
Triggers from older servers that included no
DEFINER clause in the trigger definition now
execute with the privileges of the invoker (which on the slave
is the slave SQL thread). Previously, replication slaves could
not replicate such triggers.
(Bug#16266)
Builds for Windows, Linux, and Unix (except AIX) platforms now have SSL support enabled, in the server as well as in the client libraries. Because part of the SSL code is written in C++, this does introduce dependencies on the system's C++ runtime libraries in several cases, depending on compiler specifics. (Bug#18195)
Large file support added to build for QNX
platform.
(Bug#17336)
InnoDB: The InnoDB storage
engine now provides a descriptive error message if
ibdata file information is omitted from
my.cnf.
(Bug#16827)
Added the --sysdate-is-now option
to mysqld to enable
SYSDATE() to be treated as an
alias for NOW(). See
Section 11.6, “Date and Time Functions”.
(Bug#15101)
Large file support was re-enabled for the MySQL server binary for the AIX 5.2 platform. (Bug#13571)
The syntax for CREATE PROCEDURE
and CREATE FUNCTION statements
now includes a DEFINER clause. The
DEFINER value specifies the security context
to be used when checking access privileges at routine invocation
time if the routine has the SQL SECURITY
DEFINER characteristic. See
Section 12.1.9, “CREATE PROCEDURE and
CREATE FUNCTION Syntax”, for more information.
When mysqldump is invoked with the
--routines option, it now
dumps the DEFINER value for stored routines.
Bugs fixed:
MySQL Cluster:
A timeout in the handling of an ABORT
condition with more that 32 operations could yield a node
failure.
(Bug#18414)
MySQL Cluster:
A node restart immediately following a
CREATE TABLE would fail.
This fix supports 2-node Clusters only.
MySQL Cluster: In event of a node failure during a rollback, a “false” lock could be established on the backup for that node, which lock could not be removed without restarting the node. (Bug#18352)
MySQL Cluster: The cluster created a crashed replica of a table having an ordered index — or when logging was not enabled, of a table having a table or unique index — leading to a crash of the cluster following 8 successive restarts. (Bug#18298)
MySQL Cluster: When replacing a failed master node, the replacement node could cause the cluster to crash from a buffer overflow if it had an excessively large amount of data to write to the cluster log. (Bug#18118)
MySQL Cluster:
Certain queries using ORDER BY ... ASC in the
WHERE clause could return incorrect results.
(Bug#17729)
MySQL Cluster: If a mysql or other client could not parse the result set returned from a mysqld process acting as an SQL node in a cluster, the client would crash instead of returning the appropriate error. For example, this could happen when the client attempted to use a character set was not available to the mysqld. (Bug#17380)
MySQL Cluster: Some query cache statistics were not always correctly reported for Cluster tables. (Bug#16795)
MySQL Cluster: Restarting nodes were allowed to start and join the cluster too early. (Bug#16772)
MySQL Cluster:
Inserting and deleting BLOB
column values while a backup was in process could cause data
nodes to shut down.
(Bug#14028)
MySQL Cluster:
The server would not compile with
NDB support on AIX 5.2.
(Bug#10776)
Replication:
Use of TRUNCATE
TABLE for a TEMPORARY table on a
master server was propagated to slaves properly, but slaves did
not decrement the
Slave_open_temp_tables counter
properly.
(Bug#17137)
Replication:
The DEFINER value for stored routines was not
replicated.
(Bug#15963)
A SELECT ... ORDER BY ... from a view defined
using a function could crash the server. An example of such a
view is CREATE VIEW v1 AS SELECT SQRT(c1) FROM
t1.
(Bug#18386)
InnoDB had a memory leak for duplicate-key
errors with tables having 90 columns or more.
(Bug#18384)
A DELETE using a subquery could
crash the server.
(Bug#18306)
If a row was inserted inside a stored procedure using the parameters passed to the procedure in the INSERT statement, the resulting binary log entry was not escaped properly. (Bug#18293)
If InnoDB encountered a
HA_ERR_LOCK_TABLE_FULL error and rolled back
a transaction, the transaction was still written to the binary
log.
(Bug#18283)
When using ORDER BY with a nonstring column
inside GROUP_CONCAT() the
result's character set was converted to binary.
(Bug#18281)
See also Bug#14169.
Complex queries with nested joins could cause a server crash. (Bug#18279)
For InnoDB tables created in MySQL 4.1 or
earlier, or created in 5.0 or later with compact format,
updating a row so that a long column is updated or the length of
some column changes, InnoDB later would fail
to reclaim the BLOB storage space
if the row was deleted.
(Bug#18252)
If InnoDB ran out of buffer space for row
locks and adaptive hashes, the server would crash. Now
InnoDB rolls back the transaction.
(Bug#18238)
Views that incorporated tables from the
INFORMATION_SCHEMA database resulted in a
server crash when queried.
(Bug#18224)
REPAIR TABLE,
OPTIMIZE TABLE, and
ALTER TABLE operations on
transactional tables (or on tables of any type on Windows) could
corrupt triggers associated with those tables.
(Bug#18153)
The server could deadlock under heavy load while writing to the binary log. (Bug#18116)
A SELECT * query on an
INFORMATION_SCHEMA table by a user with
limited privileges resulted in a server crash.
(Bug#18113)
Connecting to a server with a UCS2 default character set with a client using a non-UCS2 character set crashed the server. (Bug#18004)
MyISAM: Performing a bulk insert on a table
referenced by a trigger would crash the table.
(Bug#17764)
Updating a view that filters certain rows to set a filtered out
row to be included in the table caused infinite loop. For
example, if the view has a WHERE clause of salary >
100 then issuing an UPDATE statement of SET
salary = 200 WHERE id = 10, caused an infinite loop.
(Bug#17726)
MyISAM: Keys for which the first part of the
key was a CHAR or
VARCHAR column using the UTF-8
character set and longer than 254 bytes could become corrupted.
(Bug#17705)
Updating the value of a Unicode
VARCHAR column with the result
returned by a stored function would cause the insertion of ASCII
characters into the column instead of Unicode, even where the
function's return type was also declared as Unicode.
(Bug#17615)
For FEDERATED tables, a
SELECT statement with an
ORDER BY clause did not return rows in the
proper order.
(Bug#17377)
SELECT ... WHERE , when column LIKE
'A%'column had a key
and used the latin2_czech_cs collation,
caused the wrong number of rows to be returned.
(Bug#17374)
A LEFT JOIN with a
UNION that selects literal values
could crash the server.
(Bug#17366)
Checks for permissions on database operations could be performed
in a case-insensitive manner (a user with permissions on
database MYDATABASE could by accident get
permissions on database myDataBase), if the
privilege data were still cached from a previous check.
(Bug#17279)
Stored procedures that call UDFs and pass local string variables caused server crashes. (Bug#17261)
If the WHERE condition of a query contained
an OR-ed FALSE term, the
set of tables whose rows cannot serve for null-complements in
outer joins was determined incorrectly. This resulted in
blocking possible conversions of outer joins into joins by the
optimizer for such queries.
(Bug#17164)
InnoDB tables with an adaptive hash blocked
other queries during CHECK TABLE
statements while the entire hash was checked. This could be a
long time for a large hash.
(Bug#17126)
Stored routine names longer than 64 characters were silently truncated. Now the limit is properly enforced and an error occurs. (Bug#17015)
InnoDB: The LATEST FOREIGN KEY
ERROR section in the output of
SHOW INNODB STATUS was sometimes
formatted incorrectly, causing problems with scripts that parsed
the output of this statement.
(Bug#16814)
If the server was started with the
--skip-grant-tables option, it
was impossible to create a trigger or a view without explicitly
specifying a DEFINER clause.
(Bug#16777)
The FORMAT() function returned an
incorrect result when the client's
character_set_connection value
was utf8.
(Bug#16678)
Using ORDER BY within a stored procedure (where
intvar
intvar is an integer variable or
expression) would crash the server.
The use of an integer i in an
ORDER BY
clause for sorting the result by the
ii th
column is deprecated (and nonstandard). It should
not be used in new applications. See
Section 12.2.8, “SELECT Syntax”.
Character set conversion of string constants for
UNION of constant and table
column was not done when it was safe to do so.
(Bug#15949)
Triggers created in MySQL 5.0.16 and earlier could not be dropped after upgrading the server to 5.0.17 or later. (Bug#15921)
The mysql_close() C API function
leaked handles for shared-memory connections on Windows.
(Bug#15846)
COUNT(DISTINCT
and
col1,
col2)COUNT(DISTINCT
CONCAT( operations produced
different results if one of the columns was an indexed
col1,
col2))DECIMAL column.
(Bug#15745)
A SELECT using a function against
a nested view would crash the server.
(Bug#15683)
The server displayed garbage in the error message warning about
bad assignments to DECIMAL
columns or routine variables.
(Bug#15480)
During conversion from one character set to
ucs2, multi-byte characters with no
ucs2 equivalent were converted to multiple
characters, rather than to 0x003F QUESTION
MARK.
(Bug#15375)
Certain combinations of joins with mixed ON
and USING claused caused unknown
column errors.
(Bug#15229)
SELECT COUNT(*) for a
MyISAM table could return different results
depending on whether an index was used.
(Bug#14980)
Attempting to access an InnoDB table after
starting the server with
--skip-innodb caused a server
crash.
(Bug#14575)
Use of stored functions with DISTINCT or
GROUP BY can produce incorrect results when
ORDER BY is also used.
(Bug#13575)
The server would execute stored routines that had a nonexistent definer. (Bug#13198)
mysql_config returned incorrect libraries on
x86_64 systems.
(Bug#13158)
Loading of UDFs in a statically linked MySQL caused a server crash. UDF loading is now blocked if the MySQL server is statically linked. (Bug#11835)
Functionality added or changed:
Incompatible Change:
The InnoDB storage engine no longer ignores
trailing spaces when comparing
BINARY or
VARBINARY column values. This
means that (for example) the binary values
'a' and 'a ' are now
regarded as unequal any time they are
compared, as they are in MyISAM tables.
See Section 10.4.2, “The BINARY and
VARBINARY Types” for more information
about the BINARY and
VARBINARY types.
(Bug#14189)
MySQL Cluster:
More descriptive warnings are now issued when inappropriate
logging parameters are set in config.ini.
(Formerly, the warning issued was simply Could not
add logfile destination.)
(Bug#11331)
MySQL Cluster:
The ndb_mgm client commands
node_id STARTnode_id STOPALL for node_id
continues to affect all data nodes only.
mysql no longer terminates data value display when it encounters a NUL byte. Instead, it displays NUL bytes as spaces. (Bug#16859)
New charset command added to
mysql command-line client. By typing
charset or
name\C (such as
name\C UTF8), the client character set can be
changed without reconnecting.
(Bug#16217)
Added the --wait-timeout option to
mysqlmanager to allow configuration of the
timeout for dropping an inactive connection, and increased the
default timeout from 30 seconds to 28,800 seconds (8 hours).
(Bug#15980, Bug#12674)
The INFORMATION_SCHEMA now skips data
contained in unlistable/unreadable directories rather than
returning an error.
(Bug#15851)
InnoDB now caches a list of unflushed files
instead of scanning for unflushed files during a table flush
operation. This improves performance when
--innodb_file_per_table is set on
a system with a large number of InnoDB
tables.
(Bug#15653)
Added the --port-open-timeout
option to mysqld to control how many seconds
the server should wait for the TCP/IP port to become free if it
cannot be opened.
(Bug#15591)
Wording of error 1329 changed to No data - zero rows fetched, selected, or processed. (Bug#15206)
The message for error 1109 changed from Unknown table ... in order clause to Unknown table ... in field list. (Bug#15091)
A number of performance issues were resolved that had previously
been encountered when using statements that repeatedly invoked
stored functions. For example, calling
BENCHMARK() using a stored
function executed much more slowly than when invoking it with
inline code that accomplished the same task. In most cases the
two should now execute with approximately the same speed.
(Bug#14946, Bug#15014)
mysqldump now surrounds the
DEFINER, SQL SECURITY
DEFINER and WITH CHECK OPTION
clauses of a CREATE VIEW
statement with "not in version" comments to prevent errors in
earlier versions of MySQL.
(Bug#14871)
When using the GROUP_CONCAT()
function where the
group_concat_max_len system
variable was greater than 512, the type of the result was
BLOB only if the query included
an ORDER BY clause; otherwise the result was
a VARCHAR.
The result type of the
GROUP_CONCAT() function is now
VARCHAR only if the value of the
group_concat_max_len system
variable is less than or equal to 512. Otherwise, this function
returns a BLOB.
(Bug#14169)
The mysql_ping() function will
now retry if the reconnect flag is set and
error CR_SERVER_LOST is
encountered during the first attempt to ping the server.
(Bug#14057)
The mysqltest utility now converts all
CR/LF combinations to LF
to allow test cases intended for Windows to work properly on
UNIX-like systems.
(Bug#13809)
libmysqlclient now uses versioned symbols
with GNU ld.
(Bug#3074)
The client API now attempts to reconnect using TCP/IP if the
reconnect flag is set, as is the case with
sockets.
(Bug#2845)
Two new Hungarian collations are included:
utf8_hungarian_ci and
ucs2_hungarian_ci. These support the correct
sort order for Hungarian vowels. However, they do not support
the correct order for sorting Hungarian consonant contractions;
we expect to fix this issue in a future release.
Several changes were made to make upgrades easier:
Added the mysql_upgrade program that checks all tables for incompatibilities with the current version of MySQL Server and repairs them if necessary. This program should be run for each MySQL upgrade (rather than mysql_fix_privilege_tables). See Section 4.4.9, “mysql_upgrade — Check Tables for MySQL Upgrade”.
Added the FOR UPGRADE option for the
CHECK TABLE statement. This
option checks whether tables are incompatible with the
current version of MySQL Server.
Added the --check-upgrade
to mysqlcheck that invokes
CHECK TABLE with the
FOR UPGRADE option.
Bugs fixed:
MySQL Cluster:
Cluster log file paths were truncated to 128 characters. They
may now be as long as MAX_PATH (the maximum
path length permitted by the operating system).
(Bug#17411)
MySQL Cluster:
Following multiple forced shutdowns and restarts of data nodes,
DROP DATABASE could fail.
(Bug#17325)
MySQL Cluster:
The REDO log would become corrupted (and thus
unreadable) in some circumstances, due to a failure in the query
handler.
(Bug#17295)
MySQL Cluster:
An UPDATE with an inner join
failed to match any records if both tables in the join did not
have a primary key.
(Bug#17257)
MySQL Cluster:
A DELETE with a join in the
WHERE clause failed to retrieve any records
if both tables in the join did not have a primary key.
(Bug#17249)
MySQL Cluster:
The error message returned by perror
--ndb was prefixed with
OS error code instead of NDB
error code.
(Bug#17235)
MySQL Cluster:
In some cases, LOAD
DATA INFILE did not load all data into
NDB tables.
(Bug#17081)
MySQL Cluster:
ndb_delete_all ran out of memory when
processing tables containing BLOB
columns.
(Bug#16693)
MySQL Cluster:
A BIT column whose offset and
length totaled 32 caused the cluster to crash.
(Bug#16125)
MySQL Cluster:
UNIQUE keys in Cluster tables were limited to
225 bytes in length.
(Bug#15918)
MySQL Cluster:
The ndb_autodiscover test failed sporadically
due to a node not being permitted to connect to the cluster.
(Bug#15619)
MySQL Cluster:
NDB returned an incorrect
Can't find file error for OS error 24;
this has been changed to Too many open
files.
(Bug#15020)
MySQL Cluster:
No error message was generated for setting
NoOfFragmentLogFiles too low.
(Bug#13966)
MySQL Cluster:
No error message was generated for setting
MaxNoOfAttributes too low.
(Bug#13965)
MySQL Cluster: When running more than one management process in a cluster:
ndb_mgm -c
host:port
-e "node_id STOP"
stopped a management process running only on the same
system where the command was issued.
ndb_mgm -e "SHUTDOWN" failed to shut down any management processes at all.
MySQL Cluster:
ndb_mgm -c
host:port
-e "node_id STOP" would
stop a management process running only on the same system on
which the command was issued.
MySQL Cluster: ndb_mgm -e "SHUTDOWN" failed to shut down any management processes at all.
Replication:
For a transaction that used MyISAM and
InnoDB tables, interruption of the
transaction due to a dropped connection on a master server
caused slaves to lose synchrony.
(Bug#16559)
Replication:
The --replicate-do and
--replicate-ignore options were not being
enforced on multiple-table statements.
(Bug#16487, Bug#15699)
Replication:
Previously, a stored function invocation was written to the
binary log as DO
if the
invocation changes data and occurs within a nonlogged statement,
or if the function invokes a stored procedure that produces an
error. These invocations now are logged as func_name()SELECT
instead for
better control over error code checking (slave servers could
stop due to detecting a different error than occurred on the
master).
(Bug#14769)func_name()
Replication:
BIT fields were not properly
handled when using row-based replication.
(Bug#13418)
Cluster API: Upon the completion of a scan where a key request remained outstanding on the primary replica and a starting node died, the scan did not terminate. This caused incomplete error handling for the failed node. (Bug#15908)
type_decimal failed with the prepared
statement protocol.
(Bug#17826)
The MySQL server could crash with out of memory errors when
performing aggregate functions on a
DECIMAL column.
(Bug#17602)
Using DROP FUNCTION IF EXISTS
to drop a
user-defined function caused a server crash if the server was
running with the
func_name--skip-grant-tables option.
(Bug#17595)
Data truncations on non-UNIQUE indexes could
crash InnoDB when using multi-byte character
sets.
(Bug#17530)
A natural join between INFORMATION_SCHEMA
tables failed.
(Bug#17523)
A stored procedure failed to return data the first time it was called per connection. (Bug#17476)
For certain MERGE tables, the optimizer
wrongly assumed that using
index_merge/intersection was too expensive.
(Bug#17314)
The parser allowed CREATE AGGREGATE FUNCTION
for creating stored functions, even though
AGGREGATE does not apply. (It is used only
for CREATE FUNCTION only when
creating user-defined functions.)
(Bug#16896)
Cursors in stored routines could cause a server crash. (Bug#16887)
Triggers created without BEGIN and
END clauses resulted in “You have an
error in your SQL syntax” errors when dumping and
replaying a binary log.
(Bug#16878)
Using ALTER TABLE to increase the
length of a
BINARY( column
caused column values to be padded with spaces rather than
M)0x00 bytes.
(Bug#16857)
A RETURN statement within a
trigger caused a server crash.
RETURN now is disallowed within
triggers. To exit immediately, use
LEAVE.
(Bug#16829)
For a MySQL 5.0 server, using MySQL 4.1 tables in queries with a
GROUP BY clause could result in buffer
overrun or a server crash.
(Bug#16752)
An INSERT statement in a stored
procedure corrupted the binary log.
(Bug#16621)
If the query optimizer transformed a GROUP BY
clause in a subquery, it did not also transform the
HAVING clause if there was one, producing
incorrect results.
(Bug#16603)
In a highly concurrent environment, a server crash or deadlock could result from execution of a statement that used stored functions or activated triggers coincident with alteration of the tables used by these functions or triggers. (Bug#16593)
A race condition could occur when dropping the adaptive hash
index for a B-tree page in InnoDB.
(Bug#16582)
When evaluation of the test in a
CASE
failed in a stored procedure that contained a
CONTINUE handler, execution resumed at the
beginning of the CASE statement instead of at the end.
(Bug#16568)
Clients compiled from source with the
--without-readline did not save command history
from session to session.
(Bug#16557)
The DECIMAL data type was not
being handled correctly with prepared statements.
(Bug#16511)
Instance Manager searched wrong location for password file on some platforms. (Bug#16499)
UPDATE statement crashed
multi-byte character set FULLTEXT index if
update value was almost identical to initial value only
differing in some spaces being changed to .
(Bug#16489)
Certain nested LEFT JOIN operations were not
properly optimized.
(Bug#16393)
Dropping InnoDB constraints named
tbl_name_ibfk_0
SELECT with GROUP
BY on a view could cause a server crash.
(Bug#16382)
An invalid stored routine could not be dropped. (Bug#16303)
InnoDB: After upgrading an
InnoDB table having a VARCHAR
BINARY column created in MySQL 4.0 to MySQL 5.0,
update operations on the table would cause the server to crash.
(Bug#16298)
Parallel builds occasionally failed on Solaris. (Bug#16282)
A call to the IF() function using
decimal arguments could return incorrect results.
(Bug#16272)
MySQL server dropped client connection for certain
SELECT statements against views
defined that used MERGE algorithm.
(Bug#16260)
InnoDB used full explicit table locks in
trigger processing.
(Bug#16229)
Using GROUP BY on column used in
WHERE clause could cause empty set to be
returned.
(Bug#16203)
A memory leak caused warnings on slaves for certain statements that executed without warning on the master. (Bug#16175)
The FORCE INDEX keyword in a query would
prevent an index merge from being used where an index merge
would normally be chosen by the optimizer.
(Bug#16166)
Setting InnoDB path settings to an empty
string caused InnoDB storage engine to crash
upon server startup.
(Bug#16157)
The mysql_stmt_sqlstate() C API
function incorrectly returned an empty string rather than
'00000' when no error occurred.
(Bug#16143)
MIN() and
MAX() operations were not
optimized for views.
(Bug#16016)
Performing a RENAME TABLE on an
InnoDB table when the server was started with
the --innodb_file_per_table
option and the data directory was a symlink caused a server
crash.
(Bug#15991)
Executing a SHOW CREATE VIEW
query of an invalid view caused the
mysql_next_result() function of
libMySQL.dll to hang.
(Bug#15943)
Test suite sp test left behind tables when
the test failed that could cause future tests to fail.
(Bug#15866)
STR_TO_DATE(1,NULL) caused a
server crash.
(Bug#15828, CVE-2006-3081)
CAST(... AS TIME) operations
returned different results when using versus not using
prepared-statement protocol.
(Bug#15805)
Issuing a DROP USER command could
cause some users to encounter a
error.
(Bug#15775)hostname is not allowed to connect to
this MySQL server
The contents of fill_help_tables.sql could
not be loaded in strict SQL mode.
(Bug#15760)
fill_help_tables.sql was not included in
binary distributions for several platforms.
(Bug#15759)
Certain
LEAVE
statements in stored procedures were not properly optimized.
(Bug#15737)
The mysql_real_connect() C API
function incorrectly reset the
MYSQL_OPT_RECONNECT option to its default
value.
(Bug#15719)
Created a user function with an empty string (that is,
CREATE FUNCTION ''()), was accepted by the
server. Following this, calling SHOW
FUNCTION STATUS would cause the server to crash.
(Bug#15658)
Trying to compile the server on Windows generated a stack
overflow warning due to a recursive definition of the internal
Field_date::store() method.
(Bug#15634)
In some cases the query optimizer did not properly perform multiple joins where inner joins followed left joins, resulting in corrupted result sets. (Bug#15633)
Certain permission management statements could create a
NULL host name for a user, resulting in a
server crash.
(Bug#15598)
Improper memory handling for stored routine variables could cause memory overruns and binary log corruption. (Bug#15588)
The COALESCE() function truncated
data in a TINYTEXT column.
(Bug#15581)
Binary distributions for Solaris contained files with group
ownership set to the nonexisting wheel group.
Now the bin group is used.
(Bug#15562)
The absence of a table in the left part of a left or right join was not checked prior to name resolution, which resulted in a server crash. (Bug#15538)
A SELECT of a stored function
that references the INFORMATION_SCHEMA could
crash the server.
(Bug#15533)
Characters in the gb2312 and
euckr character sets which did not have
Unicode mappings were truncated.
(Bug#15377)
Certain subqueries where the inner query was the result of a aggregate function would return different results with MySQL 5.0 than with MySQL 4.1.
Subselects could also return wrong results when the query cache and grouping were involved. (Bug#15347)
Performing an ORDER BY on an indexed
ENUM column returned error.
(Bug#15308)
A SELECT query which contained a
GROUP_CONCAT() and an
ORDER BY clause against the
INFORMATION_SCHEMA resulted in an empty
result set.
(Bug#15307)
The NOT FOUND condition handler for stored
procedures did not distinguish between a NOT
FOUND condition and an exception or warning.
(Bug#15231)
The SELECT privilege was required
for triggers that performed no selects.
(Bug#15196)
An attempt to open a table that requires a disabled storage engine could cause a server crash. (Bug#15185)
The UPDATE privilege was required
for triggers that performed no updates.
(Bug#15166)
Tarball install package was missing a proper
fill_help_tables.sql file.
(Bug#15151)
Setting innodb_log_file_size to
a value greater than 4G crashed the server.
(Bug#15108)
When multiple handlers are created for the same MySQL error number within nested blocks, the outermost handler took precedence. (Bug#15011)
A statement containing GROUP BY and
HAVING clauses could return incorrect results
when the HAVING clause contained logic that
returned FALSE for every row.
(Bug#14927)
Stored routines that contained only a single statement were not
written properly to the dumpfile when using
mysqldump.
(Bug#14857)
Killing a long-running query containing a subquery could cause a server crash. (Bug#14851)
GRANT statements specifying
schema names that included underscore characters (that is,
my_schema) did not match if the underscore
was escaped in the GRANT
statement (that is, GRANT ALL ON `my\_schema`
...).
(Bug#14834)
Generating an AUTO_INCREMENT value through a
FEDERATED table did not set the value
returned by LAST_INSERT_ID().
(Bug#14768)
SUBSTRING_INDEX() could yield
inconsistent results when applied with the same arguments to
consecutive rows in a query.
(Bug#14676)
Running out of diskspace in the location specified by the
tmpdir option resulted in incorrect error
message.
(Bug#14634)
InnoDB: Comparison of indexed
VARCHAR CHARACTER SET ucs2 COLLATE ucs2_bin
columns using LIKE could fail.
(Bug#14583)
A stored procedure with an undefined variable and an exception handler would hang the client when called. (Bug#14498)
A FULLTEXT query in a prepared statement
could result in unexpected behavior.
(Bug#14496)
Using an aggregate function as the argument for a
HAVING clause resulted in the aggregate
function always returning FALSE.
(Bug#14274)
The use of LOAD INDEX within a
stored routine was permitted and caused the server to crash.
LOAD INDEX statements within
stored routines are not supported, and
now yield an error if attempted. This behavior is intended.
A COMMIT statement followed by a
ALTER TABLE statement on a BDB
table caused server crash.
(Bug#14212)
The mysql_stmt_store_result() C
API function could not be used for a prepared statement if a
cursor had been opened for the statement.
(Bug#14013)
SET sql_mode = ,
where NN > 31, did not work
properly.
(Bug#13897)
Attempts to create FULLTEXT indexes on
VARCHAR columns larger than 1000
bytes resulted in error.
(Bug#13835)
The RENAME TABLE statement did
not move triggers to the new table.
(Bug#13525)
The length of a VARCHAR() column that used
the utf8 character set would increase each
time the table was re-created in a stored procedure or prepared
statement, eventually causing the CREATE
TABLE statement to fail.
(Bug#13134)
Instance Manager erroneously accepted a list of instance
identifiers for the START INSTANCE and
STOP INSTANCE commands (should accept only a
single identifier).
(Bug#12813)
A prepared statement created from a SELECT ...
LIKE query (such as PREPARE stmt1 FROM
'SELECT col_1 FROM tedd_test WHERE col_1 LIKE ?';)
would begin to produce erratic results after being executed
repeatedly numerous (thousands) of times.
(Bug#12734)
Multi-byte path names for LOAD
DATA and
SELECT ... INTO
OUTFILE caused errors. Added the
character_set_filesystem system
variable, which controls the interpretation of string literals
that refer to file names.
(Bug#12448)
Temporary table aliasing did not work inside stored functions. (Bug#12198)
The embedded server did not allow binding of columns to the
MYSQL_TYPE_VAR_STRING data type in prepared
statements.
(Bug#12070)
When MyODBC or any other client called
my_init()/my_end()
several times, it caused corruption of charset data stored in
once_mem_pool.
(Bug#11892)
Setting the
myisam_repair_threads system
variable to a value larger than 1 could cause corruption of
large MyISAM tables.
(Bug#11527)
The mysqlbinlog utility did not output
DELIMITER statements, causing syntax errors
for stored routine creation statements.
(Bug#11312)
The embedded server failed various tests in the automated test suite. (Bug#10801, Bug#10925, Bug#15433, Bug#9633, Bug#10926, Bug#9631, Bug#10930, Bug#10911, Bug#9630, Bug#10924)
A large BIGINT value specified in
a WHERE clause could be treated differently
depending on whether it is specified as a quoted string. (For
example, WHERE bigint_col =
17666000000000000000 versus WHERE bigint_col
= '17666000000000000000').
(Bug#9088)
CHECKSUM TABLE returned different
values for MyISAM tables depending on whether
the QUICK or EXTENDED
option was used.
(Bug#8841)
Using the TRUNCATE() function
with a negative number for the second argument on a
BIGINT column returned incorrect
results.
(Bug#8461)
Issuing GRANT
EXECUTE on a procedure would display any warnings
related to the creation of the procedure.
(Bug#7787)
Repeated invocation of my_init() and
my_end() caused corruption of character set
data and connection failure.
(Bug#6536)
An INSERT ...
SELECT statement between tables in a
MERGE set can return errors when statement
involves insert into child table from merge table or vice-versa.
(Bug#5390)
Functionality added or changed:
The server treats stored routine parameters and local variables
(and stored function return values) according to standard SQL.
Previously, parameters, variables, and return values were
treated as items in expressions and were subject to automatic
(silent) conversion and truncation. Now the data type is
observed. Data type conversion and overflow problems that occur
in assignments result in warnings, or errors in strict mode. The
CHARACTER SET clause for character data type
declarations is used. Parameters, variables, and return values
must be scalars; it is no longer possible to assign a row value.
Also, stored functions execute using the
sql_mode value in force at
function creation time rather than ignoring it. For more
information, see Section 12.1.9, “CREATE PROCEDURE and
CREATE FUNCTION Syntax”.
(Bug#13808, Bug#12903, Bug#9078, Bug#14161, Bug#13705, Bug#13909, Bug#15148, Bug#8769, Bug#8702, Bug#9572, Bug#8768)
It is now possible to build the server such that
MyISAM tables can support up to 128 keys
rather than the standard 64. This can be done by configuring the
build using the option
--with-max-indexes=, where N
N≤128 is the
maximum number of indexes to permit per table.
(Bug#10932)
Bugs fixed:
MySQL Cluster: If an abort by the Transaction Coordinator timed out, the abort condition was incorrectly handled, causing the transaction record to be released prematurely. (Bug#15685)
MySQL Cluster:
The ndb_read_multi_range.test script failed
to drop a table, causing the test to fail.
(Bug#15675)
See also Bug#15402.
MySQL Cluster: Under some circumstances, it was possible for a restarting node to undergo a forced shutdown. (Bug#15632)
MySQL Cluster: A node which failed during cluster startup was sometimes not removed from the internal list of active nodes. (Bug#15587)
Replication:
A replication slave server could sometimes crash on a
BEFORE UPDATE trigger if the
UPDATE query was not executed in
the same database as the table with the trigger.
(Bug#14614)
When a connection using yaSSL was aborted, the server would
continue to try to read the closed socket, and the thread
continued to appear in the output of SHOW
PROCESSLIST. Note that this issue did not affect
secure connection attempts using OpenSSL.
(Bug#15772)
API function
mysql_stmt_prepare() returned
wrong field length for TEXT columns.
(Bug#15613)
InnoDB: Having two tables in a parent-child
relationship enforced by a foreign key where one table used
ROW_FORMAT=COMPACT and the other used
ROW_FORMAT=REDUNDANT could result in a MySQL
server crash. Note that this problem did not exist prior to
MySQL 5.0.3, when the compact row format for
InnoDB was introduced.
(Bug#15550)
BDB: A DELETE,
INSERT, or
UPDATE of a
BDB table could cause the server to crash
where the query contained a subquery using an index read.
(Bug#15536)
Resolution of the argument to the
VALUES() function to a variable
inside a stored routine caused a server crash. The argument must
be a table column.
(Bug#15441)
A left join on a column that having a NULL
value could cause the server to crash.
(Bug#15268)
The output of mysqldump --triggers did not
contain the DEFINER clause in dumped trigger
definitions.
(Bug#15110)
Reversing the order of operands in a WHERE
clause testing a simple equality (such as WHERE t1.col1
= t2.col2) would produce different output from
EXPLAIN.
(Bug#15106)
The output of SHOW TRIGGERS
contained extraneous whitespace.
(Bug#15103)
Creating a trigger caused a server crash if the table or trigger database was not known because no default database had been selected. (Bug#14863)
Column aliases were displayed incorrectly in a
SELECT from a view following an
update to a base table of the view.
(Bug#14861)
SHOW [FULL] COLUMNS and
SHOW INDEX did not function with
temporary tables.
(Bug#14387, Bug#15224)
The INFORMATION_SCHEMA.COLUMNS table did not report the size of BINARY or VARBINARY columns. (Bug#14271)
InnoDB: If
foreign_key_checks was 0,
InnoDB allowed inconsistent foreign keys to
be created.
(Bug#13778)
The server would not compile under Cygwin. (Bug#13640)
DESCRIBE did not function with
temporary tables.
(Bug#12770)
Set functions could not be aggregated in outer subqueries. (Bug#12762)
A race condition when creating temporary files caused a deadlock
on Windows with threads in Opening tables or
Waiting for table states.
(Bug#12071)
Functionality added or changed:
Replication:
The syntax for CREATE TRIGGER now
includes a DEFINER clause for specifying
which access privileges to check at trigger invocation time. See
Section 12.1.11, “CREATE TRIGGER Syntax”, for more information.
Known issue.
If you attempt to replicate from a master server older than
MySQL 5.0.17 to a slave running MySQL 5.0.17 through 5.0.19,
replication of CREATE TRIGGER
statements fails on the slave with a Definer not
fully qualified error. A workaround is to create
triggers on the master using a version-specific comment
embedded in each CREATE TRIGGER
statement:
CREATE /*!50017 DEFINER = 'root'@'localhost' */ TRIGGER ... ;
CREATE TRIGGER statements
written this way will replicate to newer slaves, which pick up
the DEFINER clause from the comment and
execute successfully.
Support files for compiling with Visual Studio 6 have been removed. (Bug#15094)
In the latin5_turkish_ci collation, the order
of the characters A WITH CIRCUMFLEX,
I WITH CIRCUMLEX, and U WITH
CIRCUMFLEX was changed. If you have used these
characters in any indexed columns, you should rebuild those
indexes.
(Bug#13421)
Recursion is allowed in stored procedures. Recursive stored functions and triggers still are disallowed. (Bug#10100)
Added a DEFINER column to the
INFORMATION_SCHEMA.TRIGGERS table.
Invoking a stored function or trigger creates a new savepoint level. When the function or trigger finishes, the previous savepoint level is restored.
See also Bug#13825.
The maximum key length for InnoDB indexes was
increased from 1024 bytes to 3072 bytes for all builds. (In
MySQL 5.0.15, the length was increased but only for 64-bit
builds.)
Added the SHOW FUNCTION CODE and
SHOW PROCEDURE CODE statements
(available only for servers that have been built with debugging
support). See Section 12.5.5.25, “SHOW PROCEDURE CODE Syntax”.
Bugs fixed:
MySQL Cluster:
A forced cluster shutdown occurred when the management daemon
was restarted with a changed config.ini
file that added an API or SQL node.
(Bug#15512)
MySQL Cluster: There was a small window for a node failure to occur during a backup without an error being reported. (Bug#15425)
MySQL Cluster:
Using ORDER BY
when
selecting from a table having the primary key on a
primary_key_columnVARCHAR column caused a forced
shutdown of the cluster.
(Bug#15240, Bug#15682, Bug#14828, Bug#15517)
MySQL Cluster: Under certain circumstances, when mysqld connected to a cluster management server, the connection would fail before a node ID could be allocated. (Bug#15215)
MySQL Cluster:
Creating a table with packed keys failed silently.
NDB now supports the
PACK_KEYS option to
CREATE TABLE correctly.
(Bug#14514)
MySQL Cluster:
REPLACE failed when attempting to
update a primary key value in a Cluster table.
(Bug#14007)
Replication: Stored functions making use of cursors were not replicated. (Bug#14077)
Replication: On Windows, the server could crash during shutdown if both replication threads and normal client connection threads were active. (Bug#11796)
Replication:
InnoDB: During replication, There was a
failure to record events in the binary log that still occurred
even in the event of a
ROLLBACK. For
example, this sequence of commands:
BEGIN; CREATE TEMPORARY TABLE t1 (a INT) ENGINE=INNODB; ROLLBACK; INSERT INTO t1 VALUES (1);
would succeed on the replication master as expected. However,
the INSERT would fail on the
slave because the
ROLLBACK would
(erroneously) cause the
CREATE TEMPORARY
TABLE statement not to be written to the binlog.
(Bug#7947)
Corrected an error-handling problem within stored routines on 64-bit platforms. (Bug#15630)
Slave SQL thread cleanup was not handled properly on Mac OS X when a statement was killed, resulting in a slave crash. (Bug#15623, Bug#15668)
The CREATE test case in
mysql-test-run.pl failed on AIX and SCO.
(Bug#15607)
A bug in mysql-test/t/mysqltest.test caused
that test to fail.
(Bug#15605)
A statement that produced a warning, when fetched via
mysql_stmt_fetch(), did not
produce a warning count according to
mysql_warning_count().
(Bug#15510)
The database-changing code for stored routine handling caused an error-handling problem resulting in a server crash. (Bug#15392)
The original Linux RPM packages (5.0.17-0) had an issue with a
zlib dependency that would result in an error
during an install or upgrade. They were replaced by new
binaries, 5.0.17-1. Here is a list of the new RPM binaries:
MySQL-{Max,client,devel,server,shared,ndb*}-5.0.17-1.i386.rpm
MySQL-*-standard-5.0.17-1.rhel3.i386.rpm, MySQL-*-standard-5.0.17-1.rhel3.ia64.rpm, MySQL-*-standard-5.0.17-1.rhel3.x86_64.rpm
MySQL-*-pro-5.0.17-1.rhel3.i386.rpm, MySQL-*-pro-5.0.17-1.rhel3.ia64.rpm, MySQL-*-pro-5.0.17-1.rhel3.x86_64.rpm
MySQL-*-pro-gpl-5.0.17-1.rhel3.i386.rpm, MySQL-*-pro-gpl-5.0.17-1.rhel3.ia64.rpm, MySQL-*-pro-gpl-5.0.17-1.rhel3.x86_64.rpm
mysqld would not start on Windows 9X operating systems including Windows Me. (Bug#15209)
Queries that select records based on comparisons to a set of column could crash the server if there was one index covering the columns, and a set of other noncovering indexes that taken together cover the columns. (Bug#15204)
Selecting from a view processed with the temptable algorithm caused a server crash if the query cache was enabled. (Bug#15119)
mysql --help was missing a newline after the
version string when the bundled readline
library was not used.
(Bug#15097)
Creating a view that referenced a stored function that selected from a view caused a crash upon selection from the view. (Bug#15096)
The server crashed if compiled without any transactional storage engines. (Bug#15047)
Multiple-table update operations were counting updates and not updated rows. As a result, if a row had several updates it was counted several times for the “rows matched” value but updated only once. (Bug#15028)
Symbolic links did not function properly on Windows platforms. (Bug#14960, Bug#14310)
ROW_COUNT() returned an incorrect
result after EXECUTE of a
prepared statement.
(Bug#14956)
When using an aggregate function to select from a table that has
a multiple-column primary key, adding ORDER
BY to the query could produce an incorrect result.
(Bug#14920)
ANALYZE TABLE did not properly
update table statistics for a MyISAM table
with a FULLTEXT index containing stopwords,
so a subsequent ANALYZE TABLE
would not recognize the table as having already been analyzed.
(Bug#14902)
Creating a view within a stored procedure could result in an out of memory error or a server crash. (Bug#14885)
GROUP BY on a view column did not correctly
account for the possibility that the column could contain
NULL values.
(Bug#14850)
The mysql_stmt_fetch() C API
function could return MYSQL_NO_DATA for a
SELECT COUNT(*) FROM
statement, which should return 1 row.
(Bug#14845)tbl_name WHERE 1 = 0
Selecting from a view used filesort retrieval
when faster retrieval was possible.
(Bug#14816)
InnoDB: A race condition allowed two threads
to drop a hash index simultaneously.
(Bug#14747)
SHOW CREATE TABLE for a view
could fail if the client had locked the view.
(Bug#14726)
The grammar for supporting the DEFINER =
CURRENT_USER clause in CREATE
VIEW and ALTER VIEW was
incorrect.
(Bug#14719)
ALTER TABLE ... SET DEFAULT had no effect.
(Bug#14693)
Using ORDER BY on a column from a view, when
also selecting the column normally, and via an alias, caused a
mistaken Column 'x' in order clause is
ambiguous error.
(Bug#14662)
SELECT queries that began with an
opening parenthesis were not being placed in the query cache.
(Bug#14652)
In a stored procedure, continuing (via a condition handler) after a failed variable initialization caused a server crash. (Bug#14643)
A LIMIT-related optimization failed to take
into account that MyISAM table indexes can be
disabled, causing Error 124 when it tried to use such an index.
(Bug#14616)
mysqlhotcopy tried to copy
INFORMATION_SCHEMA tables.
(Bug#14610)
A server crash resulted from the following sequence of events:
1) With no default database selected, create a stored procedure
with the procedure name explicitly qualified with a database
name (CREATE PROCEDURE
). 2) Create another stored procedure with no
database name qualifier. 3) Execute db_name.proc_name
...SHOW
PROCEDURE STATUS.
(Bug#14569)
mysqldump --triggers did not account for the
SQL mode and could dump trigger definitions with missing
whitespace if the IGNORE_SPACE
mode was enabled.
(Bug#14554)
CREATE TABLE could crash the server and write invalid
data into the tbl_name (...)
SELECT ....frm file if the
CREATE TABLE and
SELECT both contained a column
with the same name. Also, if a default value is specified in the
column definition, it is now actually used.
(Bug#14480)
The value of
INFORMATION_SCHEMA.TABLES.TABLE_TYPE
sometimes was reported as empty.
(Bug#14476)
mysql_fix_privilege_tables.sql contained an
erroneous comment that resulted in an error when the file
contents were processed.
(Bug#14469)
Queries on ARCHIVE tables that used the
filesort sorting method could result in a
server crash.
(Bug#14433)
Creating a table containing an
ENUM or SET
column from within a stored procedure or prepared statement
caused a server crash later when executing the procedure or
statement.
(Bug#14410)
For a table that had been opened with HANDLER
OPEN, issuing OPTIMIZE
TABLE, ALTER TABLE, or
REPAIR TABLE caused a server
crash.
(Bug#14397)
Declaring a stored routine variable to have a
DEFAULT value that referred to a variable of
the same name caused a server crash. (For example:
DECLARE x INT DEFAULT x) Now the
DEFAULT variable is interpreted as referring
to a variable in an outer scope, if there is one.
(Bug#14376)
Complex subqueries could cause improper internal query execution environment initialization and crash the server. (Bug#14342)
Within a stored procedure, inserting with
INSERT ...
SELECT into a table with an
AUTO_INCREMENT column did not generate the
correct sequence number.
(Bug#14304)
Space truncation was being ignored when inserting into
BINARY or
VARBINARY columns. Now space
truncation results in a warning, or an error in strict mode.
(Bug#14299)
Casting a FLOAT or
DOUBLE whose value was less than
1.0E-06 to
DECIMAL would yield an
inappropriate value.
(Bug#14268)
CAST() did not pad with
0x00 to a length of expr AS
BINARY(N)N bytes.
(Bug#14255)
Manual manipulation of the mysql.proc table
could cause a server crash. This should not happen, but it is
also not supported that the server will notice such changes.
(Bug#14233)
A UNION of
DECIMAL columns could produce
incorrect results.
(Bug#14216)
The maximum value of MAX_ROWS was handled
incorrectly on 64-bit systems.
(Bug#14155)
CHAR(... USING ...) and
CONVERT(CHAR(...) USING ...),
though logically equivalent, could produce different results.
(Bug#14146)
The server could misinterpret old trigger definition files created before MySQL 5.0.17. Now they are interpreted correctly, but this takes more time and the server issues a warning that the trigger should be re-created. (Bug#14090)
For a invalid view definition, selecting from the
INFORMATION_SCHEMA.VIEWS table or
using SHOW CREATE VIEW failed,
making it difficult to determine what part of the definition was
invalid. Now the server returns the definition and issues a
warning.
(Bug#13818)
InnoDB: Activity on an
InnoDB table caused execution time for
SHOW CREATE TABLE for the table
to increase.
(Bug#13762)
Within a stored procedure, exception handling for
UPDATE statements that caused a
duplicate-key error caused a Packets out of
order error for the following statement.
(Bug#13729)
Statements that implicitly commit a transaction are prohibited in stored functions and triggers. An attempt to create a function or trigger containing such a statement produces an error. (The originally reported symptom was that a trigger that dropped another trigger could cause a server crash. That problem was fixed by the patch for Bug#13343.) (Bug#13627)
A newline character in a column alias in a view definition caused an error when selecting from the view later. (Bug#13622)
Invoking a stored procedure within another stored procedure caused the server to crash. (Bug#13549)
Warnings from a previous command were not being reset when fetching from a cursor. (Bug#13524)
In some cases, a left outer join could yield an invalid result
or cause the server to crash, due to a
MYSQL_DATA_TRUNCATED error.
(Bug#13488)
DELETE from
CSV tables reported an incorrect
rows-affected value.
(Bug#13406)
A server crash could occur if a prepared statement updated a table for which a trigger existed when the statement was prepared but had been dropped prior to statement execution. (Bug#13399)
RESET MASTER failed to delete log
files on Windows. One consequence of this change is that server
opens the general query and slow log files in shared mode, so
now they can be renamed while the server has them open
(something not true in previous versions).
(Bug#13377)
For binary string data types, mysqldump
--hex-blob produced an illegal output value of
0x rather than ''.
(Bug#13318)
REPAIR TABLES, BACKUP
TABLES, RESTORE TABLES within a
stored procedure caused a server crash.
(Bug#13012)
Implicit versus explicit conversion of float to integer (such as
inserting a float value into an integer column versus using
CAST(... AS UNSIGNED) before
inserting the value) could produce different results. Implicit
and explicit typecasts now are done the same way, with a value
equal to the nearest integer according to the prevailing
rounding mode.
(Bug#12956)
Some comparisons for the IN()
operator were inconsistent with equivalent comparisons for the
= operator.
(Bug#12612)
A server crash could occur if a prepared statement invoked a stored procedure that existed when the statement was prepared but had been dropped and re-created prior to statement execution. (Bug#12329)
make failed when attempting to build MySQL in different directory other than that containing the source. (Bug#11827)
Revised table locking to allow proper assessment of view security. (Bug#11555)
Perform character set conversion of constant values whenever possible without data loss. (Bug#10446)
Within a trigger definition the
CURRENT_USER() function evaluated
to the user whose actions caused the trigger to be activated.
Now that triggers have a DEFINER value,
CURRENT_USER() evaluates to the
trigger definer.
(Bug#5861)
mysql ignored the
MYSQL_TCP_PORT environment variable.
(Bug#5792)
Functionality added or changed:
MySQL Cluster:
The InnoDB, NDB,
BDB, and ARCHIVE storage
engines now support spatial columns. See
Section 11.12, “Spatial Extensions”.
Replication:
Added a --hexdump option to
mysqlbinlog that displays a hex dump of the
log in comments. This output can be helpful for replication
debugging.
When a date column is set NOT NULL and
contains 0000-00-00, it will be updated for
UPDATE statements that contains
in the
WHERE clause.
(Bug#14186)columnname IS NULL
When trying to run the server with yaSSL enabled, MySQL now
tries to open /dev/random automatically if
/dev/urandom is not available.
(Bug#13164)
MySQL 5.0 now supports character set conversion for
seven additional cp950 characters into the
big5 character set:
0xF9D6, 0xF9D7,
0xF9D8, 0xF9D9,
0xF9DA, 0xF9DB, and
0xF9DC.
If you move data containing these additional characters to an older MySQL installation which does not support them, you may encounter errors.
You must now declare a prefix for an index on any column of any
Geometry class, the only exception being when
the column is a POINT.
(Bug#12267)
The read_only system variable
no longer applies to TEMPORARY tables.
(Bug#4544)
Due to changes in binary logging, the restrictions on which
stored routine creators can be trusted not to create unsafe
routines have been lifted for stored procedures (but not stored
functions). Consequently, the
log_bin_trust_routine_creators system
variable and the corresponding
--log-bin-trust-routine-creators server option
were renamed to
log_bin_trust_function_creators
and --log-bin-trust-function-creators. For
backward compatibility, the old names are recognized but result
in a warning. See Section 18.5, “Binary Logging of Stored Programs”.
The CHECK TABLE statement now
works for ARCHIVE tables.
In MySQL 5.0.13, syntax for DEFINER and
SQL SECURITY clauses was added to the
CREATE VIEW and
ALTER VIEW statements, but the
clauses had no effect. They now are enabled. They specify the
security context to be used when checking access privileges at
view invocation time. See Section 12.1.12, “CREATE VIEW Syntax”, for
more information.
Added the Compression status
variable, which indicates whether the client connection uses
compression in the client/server protocol.
Bugs fixed:
MySQL Cluster:
Repeated transactions using unique index lookups could cause a
memory leak leading to error 288, Out of index
operations in transaction coordinator.
(Bug#14199)
MySQL Cluster: A memory leak occurred when performing ordered index scans using indexes on columns larger than 32 bytes. This would eventually lead to the forced shutdown of all mysqld server processes used with the cluster. (Bug#13078)
For some stored functions dumped by mysqldump --routines, the function definition could not be reloaded later due to a parsing error. (Bug#14723)
Deletes from a CSV table could cause table
corruption.
(Bug#14672)
Executing REPAIR TABLE,
ANALYZE TABLE, or
OPTIMIZE TABLE on a view for
which an underlying table had been dropped caused a server
crash.
(Bug#14540)
mysqlmanager did not start up correctly on Windows 2003. (Bug#14537)
Selecting from a table in both an outer query and a subquery could cause a server crash. (Bug#14482)
ORDER BY DESC within the
GROUP_CONCAT() function was not
honored when used in a view.
(Bug#14466)
The input polling loop for Instance Manager did not sleep properly. Instance Manager used up too much CPU as a result. (Bug#14388)
Indexes for BDB tables were being limited
incorrectly to 255 bytes.
(Bug#14381)
The mysql parser did not properly strip the
delimiter from input lines less than nine characters long. For
example, this could cause USE abc; to result
in an Unknown database: abc; error.
(Bug#14358)
The displayed value for the
CHARACTER_MAXIMUM_LENGTH column in the
INFORMATION_SCHEMA.COLUMNS table
was not adjusted for multi-byte character sets.
(Bug#14290)
The parser did not correctly recognize wildcards in the host
part of the DEFINER user in
CREATE VIEW statements.
(Bug#14256)
Memory corruption and a server crash could be caused by
statements that used a cursor and generated a result set larger
than max_heap_table_size.
(Bug#14210)
A bug fix in MySQL 5.0.15 caused the displayed values for the
CHARACTER_MAXIMUM_LENGTH and
CHARACTER_OCTET_LENGTH columns in the
INFORMATION_SCHEMA.COLUMNS table to
be reversed.
(Bug#14207)
Statements of the form
CREATE TABLE ...
SELECT ... that created a column with a multi-byte
character set could incorrectly calculate the maximum length of
the column, resulting in a Specified key was too
long error.
(Bug#14139)
Use of WITH ROLLUP PROCEDURE ANALYSE() could
hang the server.
(Bug#14138)
On Windows, the value of
character_sets_dir in
SHOW VARIABLES output was
displayed inconsistently (using both “
/ ” and “ \
” as path name component separators).
(Bug#14137)
A comparison with an invalid date (such as WHERE
)
caused any index on col_name > '2005-09-31'col_name not to
be used and a string comparison for each row, resulting in slow
performance.
(Bug#14093)
Subqueries in the FROM clause failed if the
current database was INFORMATION_SCHEMA.
(Bug#14089)
For InnoDB tables, using a column prefix for
a utf8 column in a primary key caused
Cannot find record errors when attempting to
locate records.
(Bug#14056)
Some updatable views could not be updated. (Bug#14027)
A prepared statement that selected from a view processed using the merge algorithm could crash on the second execution. (Bug#14026)
When the DATE_FORMAT() function
appeared in both the SELECT and
ORDER BY clauses of a query but with
arguments that differ by case (for example,
%m and %M), incorrect
sorting may have occurred.
(Bug#14016)
TIMEDIFF(),
ADDTIME(), and
STR_TO_DATE() were not reporting
that they could return NULL, so functions
that invoked them might misinterpret their results.
(Bug#14009)
Within stored routines, REPLACE()
could return an empty string (rather than the original string)
when no replacement was done, and
IFNULL() could return garbage
results.
(Bug#13941)
Inserting a new row into an InnoDB table
could cause DATETIME values
already stored in the table to change.
(Bug#13900)
An update of a CSV table could cause a server
crash.
(Bug#13894)
Corrected a parser precedence problem that resulted in an
Unknown column ... in 'on clause' error for
some joins.
(Bug#13832)
Trying to take the logarithm of a negative value is now handled
in the same fashion as division by zero. That is, it produces a
warning when
ERROR_FOR_DIVISION_BY_ZERO is
set, and an error in strict mode.
(Bug#13820)
The example configuration files supplied with MySQL
distributions listed the
thread_cache_size variable as
thread_cache.
(Bug#13811)
mysqld_safe did not correctly start the
-max version of the server (if it was
present) if the --ledir option was given.
(Bug#13774)
SHOW CREATE TABLE did not display
the CONNECTION string for
FEDERATED tables.
(Bug#13724)
For a MyISAM table originally created in
MySQL 4.1, INSERT DELAYED could
cause a server crash.
(Bug#13707)
The server incorrectly accepted column definitions of the form
DECIMAL(0, for
D)D less than 11.
(Bug#13667)
Trying to create a stored routine with no database selected would crash the server. (Bug#13587, Bug#13514)
Inserts of too-large DECIMAL
values were handled inconsistently (sometimes set to the maximum
DECIMAL value, sometimes set to
0).
(Bug#13573)
TIMESTAMPDIFF() returned an
incorrect result if one argument but not the other was a leap
year and a date was from March or later.
(Bug#13534)
Specifying --default-character-set=cp-932 for
mysqld would cause SQL scripts containing
comments written using that character set to fail with a syntax
error.
(Bug#13487)
Use of col_name =
VALUES(col_name)ON DUPLICATE KEY UPDATE clause of an
INSERT statement failed with an
Column ' error.
(Bug#13392)col_name' in field
list is ambiguous
The default value of
query_prealloc_size was set to
8192, lower than its minimum of 16384. The minimum has been
lowered to 8192.
(Bug#13334)
InnoDB: When dropping and adding a
PRIMARY KEY, if a loose index scan using only
the second part of multiple-part index was chosen, incorrect
keys were created and an endless loop resulted.
(Bug#13293)
mysqladmin and mysqldump would hang on SCO OpenServer. (Bug#13238)
SELECT DISTINCT
CHAR( returned
incorrect results after col_name)SET NAMES utf8.
(Bug#13233)
For queries with nested outer joins, the optimizer could choose join orders that query execution could not handle. The fix is that now the optimizer avoids choosing such join orders. (Bug#13126)
The server did not take character set into account in checking
the width of the mysql.user.Password column.
As a result, it could incorrectly generate long password hashes
even if the column was not long enough to hold them.
(Bug#13064)
The source distribution failed to compile when configured with
the --without-geometry option.
(Bug#12991)
Use of the deprecated --sql-bin-update-same
option caused a server crash.
(Bug#12974)
Maximum values were handled incorrectly for command-line options
of type GET_LL.
(Bug#12925)
mysqldump could not dump views if the
-x option was given.
(Bug#12838)
Two threads that were creating triggers on an
InnoDB table at the same time could deadlock.
(Bug#12739)
InnoDB: Large
innobase_buffer_pool_size and
innobase_log_file_size values were displayed
incorrectly on 64-bit systems.
(Bug#12701)
For LIKE ... ESCAPE, an escape sequence
longer than one character was accepted as valid. Now the
sequence must be empty or one character long. If the
NO_BACKSLASH_ESCAPES SQL mode
is enabled, the sequence must be one character long.
(Bug#12595)
Inserting cp932 strings into a
VARCHAR column caused a server
crash rather than string truncation if the string was longer
than the column definition.
(Bug#12547)
A prepared statement failed with Illegal mix of
collations if the client character set was
utf8 and the statement used a table that had
a character set of latin1.
(Bug#12371)
Using ALTER TABLE to add an index
could fail if the operation ran out of temporary file space. Now
it automatically makes a second attempt that uses a slower
method but no temporary file. In this case, problems that
occurred during the first attempt can be displayed with
SHOW WARNINGS.
(Bug#12166)
mysqlimport now issues a SET
@@character_set_database = binary statement before
loading data so that a file containing mixed character sets
(columns with different character sets) can be loaded properly.
(Bug#12123)
Running OPTIMIZE TABLE and other
data-updating statements concurrently on an
InnoDB table could cause a crash or the
following warnings in the error log: Warning: Found
locks from different threads in write: enter
write_lock, Warning: Found locks from
different threads in write: start of release lock.
(Bug#11704)
LOAD DATA
INFILE would not accept the same character for both
the ESCAPED BY and the ENCLOSED
BY clauses.
(Bug#11203)
The value of Last_query_cost
was not updated for queries served from the query cache.
(Bug#10303)
Starting mysqld with the
--skip-innodb and
--default-storage-engine=innodb
(or --default-table-type=innodb
caused a server crash.
(Bug#9815)
The --exit-info=65536 option conflicted with
--temp-pool and caused problems
with the server's use of temporary files. Now
--temp-pool is ignored if
--exit-info=65536 is specified.
(Bug#9551)
For a user that has the SELECT
privilege on a view, the server erroneously was also requiring
the user to have the EXECUTE
privilege at view execution time for stored functions used in
the view definition.
(Bug#9505)
Where one stored procedure called another stored procedure: If
the second stored procedure generated an exception, the
exception was not caught by the calling stored procedure. For
example, if stored procedure A used an
EXIT statement to handle an exception,
subsequent statements in A would be executed
regardless when A was called by another
stored procedure B, even if an exception that
should have been handled by the EXIT was
generated in A.
(Bug#7049)
On Windows, the server was not ignoring hidden or system directories that Windows may have created in the data directory, and would treat them as available databases. (Bug#4375)
Functionality added or changed:
Incompatible Change:
For BINARY columns, the pad value
and how it is handled has changed. The pad value for inserts now
is 0x00 rather than space, and there is no
stripping of the pad value for selects. For details, see
Section 10.4.2, “The BINARY and
VARBINARY Types”.
Incompatible Change:
The CHAR() function now returns a
binary string rather than a string in the connection character
set. An optional USING
clause may be used
to produce a result in a specific character set instead. Also,
arguments larger than 256 produce multiple characters. They are
no longer interpreted modulo 256 to produce a single character
each. These changes may cause some incompatibilities, as noted
in Section 2.18.1.2, “Upgrading from MySQL 4.1 to 5.0”.
charset
MySQL Cluster: The ndb_mgm client now reports node startup phases automatically. (Bug#16197)
MySQL Cluster:
A new “smart” node allocation algorithm means that
it is no longer necessary to use sequential IDs for cluster
nodes, and that nodes not explicitly assigned IDs should now
have IDs allocated automatically in most cases. In practical
terms, this means that it is now possible to assign a set of
node IDs such as 1, 2,
4, 5 without an error
being generated due to the missing 3.
(Bug#13009)
MySQL Cluster: A number of new or improved error messages have been implemented in this release in order to provide better and more accurate diagnostic information regarding cluster configuration issues and problems. (Bug#12786, Bug#11749, Bug#13197, Bug#11739, Bug#12044)
The following statements now cause an implicit
COMMIT:
Added the --tz-utc option to
mysqldump. This option adds SET
TIME_ZONE='+00:00' to the dump file so that
TIMESTAMP columns can be dumped
and reloaded between servers in different time zones and
protected from changes due to daylight saving time.
(Bug#13052)
When executing single-table
UPDATE or
DELETE queries containing an
ORDER BY ... LIMIT
clause, but not having
any NWHERE clause, MySQL can now take
advantage of an index to read the first
N rows in the ordering specified in
the query. If an index is used, only the first
N records will be read, as opposed to
scanning the entire table.
(Bug#12915)
The MySQL-server RPM now explicitly assigns
the mysql system user to the
mysql user group during the postinstallation
process. This corrects an issue with upgrading the server on
some Linux distributions whereby a previously existing
mysql user was not changed to the
mysql group, resulting in wrong groups for
files created following the installation.
(Bug#12823)
The maximum key length for InnoDB indexes was
increased from 1024 bytes to 3072 bytes for 64-bit builds.
When declaring a local variable (or parameter) named
password or name, and
setting it with SET (for example,
SET password = ''), the new error message
ERROR 42000: Variable 'nnn' must be quoted with `...`,
or renamed is returned (where 'nnn' is 'password' or
'names'). This means there is a syntax conflict with special
sentences like SET PASSWORD = PASSWORD(...)
(for setting a user's password) and set names
default (for setting charset and collation).
This must be resolved either by quoting the variable name:
SET `password` = ..., which will set the
local variable `password`, or by renaming the
variable to something else (if setting the user's password is
the desired effect).
Bugs fixed:
MySQL Cluster:
The perror utility included with the
MySQL-Server RPM did not provide support for
the --ndb option. It now supports
this option, and so can be used to obtain error message text for
MySQL Cluster error codes.
(Bug#13740)
MySQL Cluster:
Placing multiple [tcp default] sections in
the cluster's config.ini file crashed
ndb_mgmd. (The process now exits gracefully
in such cases, with an appropriate error message.)
(Bug#13611)
MySQL Cluster: ndb_mgmd allowed a node to be stopped or restarted while another node was still starting up, which could crash the cluster. It should now not be possible to issue a node stop or restart while a different node is still restarting, and the cluster management client should issue an error when such an attempt is made. (Bug#13461)
MySQL Cluster:
Trying to run ndbd as system
root when connecting to a
mysqld process running as the
mysql system user via SHM caused the
ndbd process to crash.
(ndbd should now exit gracefully with an
appropriate error message instead.)
(Bug#9249)
Replication:
An UPDATE query using a join
would be executed incorrectly on a replication slave.
(Bug#12618)
Tests containing SHOW TABLE
STATUS or INFORMATION_SCHEMA failed
on opnsrv6c.
(Bug#14064, Bug#14065)
mysqlcheck
--all-databases
--analyze
--optimize failed because it
also tried to analyze and optimize the
INFORMATION_SCHEMA tables which it can't.
(Bug#13783)
Character set conversion was not being done for
FIND_IN_SET().
(Bug#13751)
On BSD systems, the system crypt() call could
return an error for some salt values. The error was not handled,
resulting in a server crash.
(Bug#13619)
When calling a stored procedure with the syntax CALL
and no default schema selected, schema.procedurename
ERROR
1046 was displayed after the procedure returned.
(Bug#13616)
A column in the ON condition of a join that
referenced a table in a nested join could not be resolved if the
nested join was a right join.
(Bug#13597)
The server could over-allocate memory when performing a
FULLTEXT search for stopwords only.
(Bug#13582)
CREATE DEFINER=... VIEW ... caused the server
to crash when run with
--skip-grant-tables.
(Bug#13504)
InnoDB: Queries that were executed using an
index_merge union or
intersection could produce incorrect results if the underlying
table used the InnoDB storage engine and had
a primary key containing VARCHAR
members.
(Bug#13484)
A qualified reference to a view column in the
HAVING clause could not be resolved.
(Bug#13410)
CAST(1E+300 TO SIGNED INT)
produced an incorrect result on little-endian machines.
(Bug#13344)
Queries that use indexes in normal
SELECT statements may cause range
scans in VIEWs.
(Bug#13327)
SELECT * INTO OUTFILE ... FROM
INFORMATION_SCHEMA.schemata failed with an
Access denied error.
(Bug#13202)
mysqldump --triggers did not quote
identifiers properly if the
--compatible option was given,
so the dump output could not be reloaded.
(Bug#13146)
A table or view named Ç (C-cedilla) couldn't be dropped. (Bug#13145)
For XA transaction IDs (
), uniqueness is supposed to be assessed based on
gtrid.bqual.formatID
gtrid and
bqual. MySQL was also including
formatID in the uniqueness check.
(Bug#13143)
Trying to create a view dynamically using a prepared statement within a stored procedure failed with error 1295. (Bug#13095)
comp_err did not detect when multiple error messages for a language were given for an error symbol. (Bug#13071)
If special characters such as '_' ,
'%', or the escape character were included
within the prefix of a column index, LIKE
pattern matching on the indexed column did not return the
correct result.
(Bug#13046, Bug#13919)
Using an undefined variable in an
IF or
SET clause inside a stored routine produced
an incorrect unknown column ... in 'order
clause' error message.
(Bug#13037)
With --log-slave-updates
Exec_master_log_pos of SQL thread lagged IO
(Bug#13023)
SHOW CREATE TABLE did not display
any FOREIGN KEY clauses if a temporary file
could not be created. Now SHOW CREATE
TABLE displays an error message in an SQL comment if
this occurs.
(Bug#13002)
Local (non-XA) and XA transactions are supposed to be mutually exclusive within a given client connection, but this prohibition was not always enforced. (Bug#12935)
Server crashed during a SELECT
statement, writing a message like this to the error log:
InnoDB: Error: MySQL is trying to perform a SELECTInnoDB: but it has not locked any tables in ::external_lock()!
An expression in an ORDER BY clause failed
with Unknown column
' if the expression referred to a column alias.
(Bug#11694)col_name' in 'order
clause'
Issuing STOP SLAVE after having
acquired a global read lock with
FLUSH TABLES WITH READ
LOCK caused a deadlock. Now STOP
SLAVE is generates an error in such circumstances.
(Bug#10942)
Corrected a memory-copying problem for big5
values when using icc compiler on Linux IA-64
systems.
(Bug#10836)
The --interactive-timeout and
--slave-net-timeout options for
mysqld were not being obeyed on Mac OS X and
other BSD-based platforms.
(Bug#8731)
Queries of the form (SELECT ...) ORDER BY ...
were being treated as a UNION.
This improperly resulted in only distinct values being returned
(because UNION by default
eliminates duplicate results). Also, references to column
aliases in ORDER BY clauses following
parenthesized SELECT statements
were not resolved properly.
(Bug#7672)
Character set file parsing during
mysql_real_connect() read past
the end of a memory buffer.
(Bug#6413)
Functionality added or changed:
Replication:
Multiple-table UPDATE and
DELETE statements that do not
affect any rows are now written to the binary log and will
replicate.
(Bug#13348, Bug#12844)
Range scans can now be performed for queries on VIEWs such as
column IN (<constants>) and
column BETWEEN ConstantA AND ConstantB.
(Bug#13317)
The limit of 255 characters on the input buffer for mysql on Windows has been lifted. The exact limit depends on what the system allows, but can be up to 64K characters. A typical limit is 16K characters. (Bug#12929)
Added the myisam_stats_method,
which controls whether NULL values in indexes
are considered the same or different when collecting statistics
for MyISAM tables. This influences the query
optimizer as described in
Section 7.4.6, “MyISAM Index Statistics Collection”.
(Bug#12232)
The CHAR() function now takes
into account the character set and collation given by the
character_set_connection and
collation_connection system
variables. For an argument n to
CHAR(), the result is
n mod 256 for single-byte character
sets. For multi-byte character sets,
n must be a valid code point in the
character set. Also, the result string from
CHAR() is checked for
well-formedness. For invalid arguments, or a result that is not
well-formed, MySQL generates a warning (or, in strict SQL mode,
an error).
(Bug#10504)
Re-enabled the
--delayed-insert option for
mysqldump, which now checks for each table
dumped whether its storage engine supports
DELAYED inserts.
(Bug#7815)
RENAME TABLE now works for views
as well, as long as you do not try to rename a view into a
different database.
(Bug#5508)
Configure-time checking for the availability of multi-byte
macros and functions in the bundled readline
library. This improves handling of multi-byte character sets in
the mysql client.
(Bug#3982)
When an InnoDB foreign key constraint is
violated, the error message now indicates which table, column,
and constraint names are involved.
(Bug#3443)
Bugs fixed:
MySQL Cluster:
A trigger updating the value of an
AUTO_INCREMENT column in an
NDB table would insert an error
code rather than the expected value into the column.
(Bug#13961)
MySQL Cluster: If ndb_restore could not find a free mysqld process, it crashed. (Bug#13512)
MySQL Cluster: Adding an index to a table with a large number of columns (more then 100) crashed the storage node. (Bug#13316)
MySQL Cluster:
BIT columns and following columns
in NDB tables were corrupt when
dumped by mysqldump.
(Bug#13152)
MySQL Cluster:
Queries on NDB tables that were
executed using index_merge
could produce incorrect results.
(Bug#13081)
MySQL Cluster:
Receipt of several ENTER SINGLE USER MODE
commands by multiple ndb_mgmd processes
within a short period of time resulted in cluster shutdown.
(Bug#13053)
MySQL Cluster: Multiple ndb_mgmd processes in a cluster did not know each other's IP addresses. (Bug#12037)
MySQL Cluster:
With two mgmd processes in a cluster,
ndb_mgm output for
SHOW would display the same IP
address for both processes, even when they were on different
hosts.
(Bug#11595)
MySQL Cluster:
LOAD DATA
INFILE with a large data file failed.
(Bug#10694)
MySQL Cluster:
When deleting a great many (tens of thousands of) rows at once
from an NDB table, an improperly
dereferenced pointer could cause the mysqld
process to crash.
(Bug#9282)
Replication:
The --replicate-rewrite-db and
--replicate-do-table options did
not work for statements in which tables were aliased to names
other than those listed by the options.
(Bug#11139)
Certain joins using Range checked for each
record in the query execution plan could cause the
server to crash.
(Bug#24776)
Joins nested under NATURAL or
USING joins were sometimes not initialized
properly, causing a server crash.
(Bug#13545)
After running configure with the
--with-embedded-privilege-control option, the
embedded server failed to build.
(Bug#13501)
The optimizer chose a less efficient execution plan for
col_name BETWEEN
const AND
const
, even though the two
expressions are logically equivalent. Now the optimizer can use
the col_name =
constref access method for
both expressions.
(Bug#13455)
Locking a view with the query cache enabled and
query_cache_wlock_invalidate
enabled could cause a server crash.
(Bug#13424)
A HAVING clause that references an
unqualified view column name could crash the server.
(Bug#13411)
The --skip-innodb-doublewrite option disables
use of the InnoDB doublewrite buffer.
However, having this option in effect when creating a new MySQL
installation prevented the buffer from even being created,
resulting in a server crash later.
(Bug#13367)
Calling the FORMAT() function
with a DECIMAL column value
caused a server crash when the value was
NULL.
(Bug#13361)
Comparisons involving row constructors containing constants could cause a server crash. (Bug#13356)
Aggregate functions sometimes incorrectly were allowed in the
WHERE clause of
UPDATE and
DELETE statements.
(Bug#13180)
NATURAL joins and joins with
USING against a view could return
NULL rather than the correct value.
(Bug#13127)
For queries with DISTINCT and WITH
ROLLUP, the DISTINCT should be
applied after the rollup operation, but was not always.
(Bug#12887)
It was possible to create a view that executed a stored function
for which you did not have the
EXECUTE privilege.
(Bug#12812)
Shared-memory connections were not working on Windows. (Bug#12723)
The server was not rejecting
FLOAT(
or
M,D)DOUBLE(
columns specifications when M,D)M was
less than D.
(Bug#12694)
CHECKSUM TABLE locked
InnoDB tables and did not use a consistent
read.
(Bug#12669)
Incorrect creation of DECIMAL
local variables in a stored procedure could cause a server
crash.
(Bug#12589)
For queries for which the optimizer determined a join type of
“Range checked for each record” (as shown by
EXPLAIN, the query sometimes
could cause a server crash, depending on the data distribution.
(Bug#12291)
After running configure with the
--without-server option, the
distribution failed to build.
(Bug#11680, Bug#13550)
Use of a user-defined function within the
HAVING clause of a query resulted in an
Unknown column error.
(Bug#11553)
The server crashed when processing a view that invoked the
CONVERT_TZ() function.
(Bug#11416)
When SELECT ... FOR
UPDATE or
SELECT ... LOCK IN SHARE
MODE for an InnoDB table were
executed from within a stored function or a trigger, they were
converted to a nonlocking consistent read.
(Bug#11238)
Queries against a MERGE table that has a
composite index could produce incorrect results.
(Bug#9112)
MySQL programs in binary distributions for Solaris 8/9/10 x86 systems would not run on Pentium III machines. (Bug#6772)
Nested handlers within stored procedures didn't work. (Bug#6127)
Functionality added or changed:
Replication:
Better detection of connection timeout for replication servers
on Windows allows elimination of extraneous Lost
connection errors in the error log.
(Bug#5588)
OPTIMIZE TABLE and
HANDLER now are prohibited in
stored procedures and functions and in triggers.
(Bug#12953, Bug#12995)
The LEAST() and
GREATEST() functions used to
return NULL only if all arguments were
NULL. Now they return NULL
if any argument is NULL, the same as Oracle.
(Bug#12791)
InnoDB: The
TRUNCATE
TABLE statement for InnoDB tables
always resets the counter for an
AUTO_INCREMENT column now, regardless of
whether there is a foreign key constraint on the table.
(Beginning with 5.0.3,
TRUNCATE
TABLE reset the counter, but only if there was no such
constraint.)
(Bug#11946)
Reorder network startup to come after all other initialization, particularly storage engine startup which can take a long time. This also prevents MySQL from being run on a privileged port (any port under 1024) unless run as the root user. (Bug#11707)
The restriction on the use of
PREPARE,
EXECUTE, and
DEALLOCATE PREPARE within stored
procedures was lifted. The restriction still applies to stored
functions and triggers.
(Bug#10975, Bug#10605)
See also Bug#7115.
A new command line argument was added to mysqld to ignore client character set information sent during handshake, and use server side settings instead, to reproduce 4.0 behavior :
mysqld --skip-character-set-client-handshake
(Bug#9948)
Added a --routines option for
mysqldump that enables dumping of stored
routines.
(Bug#9056)
RAND() no longer allows
nonconstant initializers. (Prior to MySQL 5.0.13, the effect of
nonconstant initializers is undefined.)
(Bug#6172)
The syntax for CREATE VIEW and
ALTER VIEW statements now
includes DEFINER and SQL
SECURITY clauses for specifying the security context
to be used when checking access privileges at view invocation
time. (The syntax is present in 5.0.13, but these clauses have
no effect until 5.0.16.) See Section 12.1.12, “CREATE VIEW Syntax”, for
more information.
The --hex-dump option for
mysqldump now also applies to
BIT columns.
Two new collations have been added for Esperanto:
utf8_esperanto_ci and
ucs2_esperanto_ci.
The Windows binary packages are now compiled with the Microsoft Visual Studio 2003 compiler instead of Microsoft Visual C++ 6.0.
The connection string for FEDERATED tables
now is specified using a CONNECTION table
option rather than a COMMENT table option.
The binaries compiled with the Intel icc compiler are now built using icc 9.0 instead of icc 8.1. You will have to install new versions of the Intel icc runtime libraries, which are available from here: http://dev.mysql.com/downloads/os-linux.html
Bugs fixed:
Incompatible Change:
A lock wait timeout caused InnoDB to roll
back the entire current transaction. Now it rolls back only the
most recent SQL statement.
(Bug#12308)
MySQL Cluster:
The cluster management client START BACKUP
command could be interrupted by a
SHOW command.
(Bug#13054)
MySQL Cluster: A cluster shutdown following the crash of a data node failed to terminate any remaining node processes, even though ndb_mgm showed the shutdown request as having been completed. (Bug#9996, Bug#10938, Bug#11623)
MySQL Cluster:
The average row size for Cluster tables was calculated
incorrectly. This affected the values shown for the
Data_length and
Avg_row_length columns in the output
generated by SHOW TABLE STATUS as
well as the values for the data_length and
data_length/table_rows columns shown in the
TABLES table of the
INFORMATION_SCHEMA database with respect to
Cluster tables.
Tables using storage engines other than
NDB were not affected by this bug.
(Bug#9896)
Replication:
Within a transaction, the following statements now cause an
implicit commit: CREATE FUNCTION,
DROP FUNCTION,
DROP PROCEDURE (for stored
functions, not UDFs), ALTER
FUNCTION, ALTER
PROCEDURE, CREATE
PROCEDURE. This corrects a problem where these
statements followed by
ROLLBACK might
not be replicated properly.
(Bug#12870)
Replication:
Replication of LOAD
DATA INFILE failed between systems using different
path name syntax (such as delimiter characters).
(Bug#11815)
Local variables in stored routines were not always initialized correctly. (Bug#13133)
The FEDERATED storage engine does not support
ALTER TABLE, but no appropriate
error message was issued.
(Bug#13108)
Columns named in the USING() clause of
JOIN ... USING() were incorrectly resolved in
case-sensitive fashion.
(Bug#13067)
For a server compiled with yaSSL, clients that used MySQL Connector/J were not able to establish SSH connections. (Bug#13029)
When used in view definitions,
DAYNAME(,
expr)DAYOFWEEK(,
expr)WEEKDAY(
were incorrectly treated as though the expression was
expr)TO_DAYS(
or
expr)TO_DAYS(TO_DAYS(.
(Bug#13000)expr))
Using AS to rename a column selected from a
view in a subquery made it not possible to refer to that column
in the outer query.
(Bug#12993)
Using an INOUT parameter with a
DECIMAL data type in a stored
procedure caused a server crash.
(Bug#12979)
SELECT ... JOIN ... ON ... JOIN ... USING
caused a server crash.
(Bug#12977)
A bug introduced in MySQL 5.0.12 caused
SHOW TABLE STATUS to display an
Auto_increment value of 0 for
InnoDB tables.
(Bug#12973)
On HP-UX 11.x (PA-RISC), the -L option caused
mysqlimport to crash.
(Bug#12958)
InnoDB: A consistent read could return
inconsistent results due to a bug introduced in MySQL 5.0.5.
(Bug#12947)
Incorrect implicit nesting of joins caused the parser to fail on
queries of the form SELECT ... FROM t1 JOIN t2 JOIN t3
ON t1.t1col = t3.t3col with an Unknown column
't1.t1col' in 'on clause' error.
(Bug#12943)
Incorrect results could be returned from a view processed using a temporary table. (Bug#12941)
Multiplying a DECIMAL value
within a loop in a stored routine could incorrectly result in a
value of NULL.
(Bug#12938)
Using GROUP BY when selecting from a view in
some cases could cause incorrect results to be returned.
(Bug#12922)
The counters for the
Key_read_requests,
Key_reads,
Key_write_requests, and
Key_writes status variables
were changed from unsigned long to
unsigned longlong to accommodate larger
values before the variables roll over and restart from 0.
(Bug#12920)
mysql and mysqldump were
ignoring the
--defaults-extra-file option.
(Bug#12917)
SHOW FIELDS FROM
caused error 1046 when no default schema was set.
(Bug#12905)schemaname.viewname
UNION [DISTINCT] was not removing all
duplicates for multi-byte character values.
(Bug#12891)
A column that can be NULL was not handled
properly for WITH ROLLUP in a subquery or
view.
(Bug#12885)
GROUP_CONCAT() ignored an empty
string if it was the first value to occur in the result.
(Bug#12863)
If a client has opened an InnoDB table for
which the .ibd file is missing,
InnoDB would not honor a
DROP TABLE statement for the
table.
(Bug#12852)
Within a stored procedure, a server crash was caused by
assigning to a VARCHAR INOUT parameter the
value of an expression that included the variable itself. (For
example, SET c = c.)
(Bug#12849)
The server crashed when one thread resized the query cache while another thread was using it. (Bug#12848)
A concurrency problem for CREATE ... SELECT
could cause a server crash.
(Bug#12845)
DO IFNULL(NULL, NULL) and SELECT
CAST(IFNULL(NULL, NULL) AS DECIMAL) caused a server
crash.
(Bug#12841)
After changing the character set with SET CHARACTER
SET, the result of the
GROUP_CONCAT() function was not
converted to the proper character set.
(Bug#12829)
The Windows installer made a change to one of the
mysql.proc table files, causing stored
routine functionality to be compromised. The Windows installer
now never overwrites files in the MySQL data directory. During
an upgrade from one version to another, a file in the data
directory will not be overwritten even if it has not been
modified since it was put there by an older installer.
If you have already lost access to stored routines because of this problem, you can get them back using the following procedure:
Stop the server.
In the mysql\data directory under your
MySQL installation directory, and replace the
proc.frm file with corresponding file
from the version of MySQL that you were using before you
upgraded.
Start the server.
Start the mysql command-line client (use
the root account or another account that
has full database privileges) and execute the
mysql_fix_privilege_tables.sql script
that upgrades the grant tables to the current structure.
Instructions for doing this are given in
Section 4.4.5, “mysql_fix_privilege_tables — Upgrade MySQL System Tables”.
After this, all stored routine functionality should work. (Bug#12820)
Queries with subqueries, where the inner subquery uses the
range or
index_merge access method,
could return incorrect results.
(Bug#12720)
The server failed to disallow
SET autocommit
in stored functions and triggers. It is allowed to change the
value of autocommit in stored
procedures, but a runtime error might occur if the procedure is
invoked from a stored function or trigger.
(Bug#12712)
Simultaneous execution of DML statements and
CREATE TRIGGER or
DROP TRIGGER statements on the
same table could cause server crashes or errors.
(Bug#12704)
Performing an IS NULL check on the
MIN() or
MAX() of an indexed column in a
complex query could produce incorrect results.
(Bug#12695)
Use of PREPARE and
EXECUTE with a statement that
selected from a view in a subquery could cause a server crash.
(Bug#12651)
If the binary log is enabled, execution of a stored procedure that modifies table data and uses user variables could cause a server crash or incorrect information to be written to the binary log. (Bug#12637)
The LIKE ... ESCAPE syntax produced invalid
results when escape character was larger than one byte.
(Bug#12611)
InnoDB: Limit recursion depth to 200 in
deadlock detection to avoid running out of stack space.
(Bug#12588)
The mysql.server script contained an
incorrect path for the libexec directory.
(Bug#12550)
A UNION of long
utf8 VARCHAR
columns was sometimes returned as a column with a
LONGTEXT data type rather than
VARCHAR. This could prevent such
queries from working at all if selected into a
MEMORY table because the
MEMORY storage engine does not support the
TEXT data types.
(Bug#12537)
A client connection thread cleanup problem caused the server to crash when closing the connection if the binary log was enabled. (Bug#12517)
Use of the mysql client
HELP command from within a stored
routine caused a “packets out of order” error and a
lost connection. Now HELP is
detected and disallowed within stored routines.
(Bug#12490)
The SYSDATE() function now
returns the time at which it was invoked. In particular, within
a stored routine or trigger,
SYSDATE() returns the time at
which it executes, not the time at which the stored routine or
triggering statement began to execute.
(Bug#12480)
CREATE VIEW inside a stored
procedure caused a server crash if the table underlying the view
had been deleted.
(Bug#12468)
Deadlock occurred when several account management statements
were run (particularly between
FLUSH
PRIVILEGES/SET PASSWORD
and
GRANT/REVOKE
statements).
(Bug#12423)
InnoDB was too permissive with LOCK
TABLE ... READ LOCAL and allowed new inserts into the
table. Now READ LOCAL is equivalent to
READ for InnoDB. This will
cause slightly more locking in mysqldump, but
makes InnoDB table dumps consistent with
MyISAM table dumps.
(Bug#12410)
If a stored function invoked from a
SELECT failed with an error, it
could cause the client connection to be dropped. Now such errors
generate warnings instead so as not to interrupt the
SELECT.
(Bug#12379)
The value of
character_set_results could be
set to NULL, but returned the string
"NULL" when retrieved.
(Bug#12363)
On Windows, the server was preventing tables from being created
if the table name was a prefix of a forbidden name. For example,
nul is a forbidden name because it is the
same as a Windows device name, but a table with the name of
n or nu was being
forbidden as well.
(Bug#12325)
ALTER TABLE ... DISCARD TABLESPACE for
non-InnoDB table caused the client to lose
the connection. (The server was not returning the error
properly.)
(Bug#12207)
Outer join elimination was erroneously applied for some queries
that used a NOT BETWEEN condition, an
IN(
condition, or an value_list)IF() condition.
(Bug#12102, Bug#12101)
Foreign keys were not properly enforced in
TEMPORARY tables. Foreign keys now are
disallowed in TEMPORARY tables.
(Bug#12084)
When using a cursor, a SELECT
statement that uses a GROUP BY clause could
return incorrect results.
(Bug#11904)
The character_set_system system
variable could not be selected with SELECT
@@character_set_system.
(Bug#11775)
A memory leak resulting from repeated SELECT ...
INTO statements inside a stored procedure could cause
the server to crash.
(Bug#11333)
Use of yaSSL for a secure client connection caused
LOAD DATA LOCAL
INFILE to fail.
(Bug#11286)
mysqld_multi now quotes arguments on command lines that it constructs to avoid problems with arguments that contain shell metacharacters. (Bug#11280)
The server allowed privileges to be granted explicitly for the
INFORMATION_SCHEMA database. Such privileges
are always implicit and should not be grantable.
(Bug#10734)
SHOW CREATE PROCEDURE and
SHOW CREATE FUNCTION no longer
qualify the routine name with the database name, for consistency
with the behavior of SHOW CREATE
TABLE.
(Bug#10362)
The server incorrectly generated an Unknown
table error message when for attempts to drop tables
in the INFORMATION_SCHEMA database. Now it
issues an Access denied message.
(Bug#9846)
Within a stored procedure, fetching a large number of rows in a
loop using a cursor could result in a server crash or an out of
memory error. Also, values inserted within a stored procedure
using a cursor were interpreted as latin1
even if character set variables had been set to a different
character set.
(Bug#9819, Bug#6513)
The server allowed TEMPORARY tables and
stored procedures to be created in the
INFORMATION_SCHEMA database.
(Bug#9683, Bug#10708)
SHOW FIELDS truncated the
TYPE column to 40 characters.
(Bug#7142)
See also Bug#12817.
A view-creation statement of the form CREATE VIEW
failed with a
name AS SELECT ... FROM
tbl_name AS
nameNot unique table/alias:
' error.
(Bug#6808)name'
myisampack did not properly pack
BLOB values larger than
224 bytes.
(Bug#4214)
Functionality added or changed:
Incompatible Change:
Beginning with MySQL 5.0.12, natural joins and joins with
USING, including outer join variants, are
processed according to the SQL:2003 standard. The changes
include elimination of redundant output columns for
NATURAL joins and joins specified with a
USING clause and proper ordering of output
columns. The precedence of the comma operator also now is lower
compared to JOIN.
In addition, a Duplicate column name
error no longer occurs when selecting from a view
defined as SELECT * from a join that uses a
USING clause on tables that have a common
column name.
These changes make MySQL more compliant with standard SQL.
However, they can result in different output columns for some
joins. Also, some queries that appeared to work correctly prior
to 5.0.12 must be rewritten to comply with the standard. For
details about the scope of the changes and examples that show
what query rewrites are necessary, see Section 12.2.8.1, “JOIN Syntax”.
(Bug#6495, Bug#6136, Bug#10972, Bug#9978, Bug#10428, Bug#10646, Bug#6276, Bug#6489, Bug#6558, Bug#9067, Bug#4789, Bug#12065, Bug#13551)
MySQL Cluster:
The parsing of the CLUSTERLOG command by
ndb_mgm was corrected to allow multiple
items.
(Bug#12833)
Replication: Interleaved execution of stored procedures and functions could be written to the binary log incorrectly, causing replication slaves to get out of sync. (Bug#12335)
Replication: Calls to stored procedures were written to the binary log even within transactions that were rolled back, causing them to be executed on replication slaves. (Bug#12334)
A query of the form SHOW TABLE STATUS FROM
would crash
the server.
(Bug#12636)db_name WHERE name IN
(select_query)
Using DESCRIBE on a view after
renaming a column in one of the view's base tables caused the
server to crash.
(Bug#12533)
If a thread (connection) has tables locked, the query cache is switched off for that thread. This prevents invalid results where the locking thread inserts values between a second thread connecting and selecting from the table. (Bug#12385)
SHOW TABLE STATUS FROM INFORMATION_SCHEMA now
sorts output by table name the same as it does for other
databases.
(Bug#12315)
It is no longer possible to issue
FLUSH commands from within stored
functions or triggers. See
Section D.1, “Restrictions on Stored Routines and Triggers”, for details.
(Bug#12280, Bug#12307)
SHOW OPEN TABLES now supports
FROM and LIKE clauses.
(Bug#12183)
Recursive triggers are detected and disallowed. Also, within a stored function or trigger, it is not allowable to modify a table that is already being used (for reading or writing) by the statement that invoked the function or trigger. (Bug#11896, Bug#12644)
INFORMATION_SCHEMA objects are now reported
as a SYSTEM VIEW table type.
(Bug#11711)
The stability of cursors when used with
InnoDB tables was greatly improved.
(Bug#11309, Bug#11832, Bug#12243)
Trying to drop the default keycache by setting
@@global.key_buffer_size to zero now returns
a warning that the default keycache cannot be dropped.
(Bug#10473)
SHOW ENGINE INNODB
STATUS now can display longer query strings.
(Bug#7819)
Added the SLEEP() function, which
pauses for the number of seconds given by its argument.
(Bug#6760)
SHOW TABLE STATUS for a view now
shows VIEW in uppercase, consistent with
SHOW TABLES and
INFORMATION_SCHEMA.
(Bug#5501)
Bugs fixed:
MySQL Cluster: When it could not copy a fragment, ndbd exited without printing a message about the condition to the error log. Now the message is written. (Bug#12900)
MySQL Cluster: When a Disk is full condition occurred, ndbd exited without reporting this condition in the error log. (Bug#12716)
MySQL Cluster: Cluster failed to take character set data into account when recomputing hashes (and thus could not locate records for updating or deletion) following a configuration change and node restart. (Bug#12220)
MySQL Cluster:
An ALTER TABLE command caused
loss of data stored prior to the issuing of the command.
(Bug#12118)
MySQL Cluster:
Invalid values in config.ini caused
ndb_mgmd to crash.
(Bug#12043)
MySQL Cluster: When a schema was detected to be corrupt, ndb neglected to close it, resulting in a file already open error if the schema was opened again later. written. (Bug#12027)
MySQL Cluster: Improved error messages related to file system issues. (Bug#11218)
MySQL Cluster: The wrong error message was displayed when the cluster management server port was closed while a mysqld process was trying to connect. (Bug#10950)
Replication: Some statements executed on a master server caused the SQL thread on a slave to run out of memory. (Bug#12532)
Replication: Trigger and stored procedure execution could break replication. (Bug#12482)
Replication:
NOW(),
CURRENT_TIME() and values
generated by timestamp columns are now constant for the duration
of a stored function or trigger. This prevents the breaking of
statement-based replication.
(Bug#12481)
Replication:
Slave I/O threads were considered to be in the running state
when launched (rather than after successfully connecting to the
master server), resulting in incorrect SHOW
SLAVE STATUS output.
(Bug#10780)
Replication:
If a DROP DATABASE fails on a
master server due to the presence of a nondatabase file in the
database directory, the master have the database tables deleted,
but not the slaves. To deal with failed database drops, we now
write DROP TABLE statements to
the binary log for the tables so that they are dropped on
slaves.
(Bug#4680)
An optimizer estimate of zero rows for a nonempty
InnoDB table used in a left or right join
could cause incomplete rollback for the table.
(Bug#12779)
mysql_fix_privilege_tables.sql was missing
a comma, causing a syntax error when executed.
(Bug#12705)
Invocations of the SLEEP()
function incorrectly could get optimized away for statements in
which it occurs. Statements containing
SLEEP() incorrectly could be
stored in the query cache.
(Bug#12689)
Improper use of loose index scan in InnoDB
sometimes caused incorrect query results.
(Bug#12672)
A SELECT
DISTINCT query with a constant value for one of the
columns would return only a single row.
(Bug#12625)
SHOW TABLES FROM returned wrong error message
if the schema specified did not exist.
(Bug#12591)
A server crash could result from an update of a view defined as a join, even though the update updated only a single table. (Bug#12569)
DELETE or
UPDATE for an indexed
MyISAM table could fail. This was due to a
change in end-space comparison behavior from 4.0 to 4.1.
(Bug#12565)
The COLUMN_DEFAULT column of the
INFORMATION_SCHEMA.COLUMNS table
should be returned as NULL if a column has no
default value. An empty string was being returned if the column
was defined as NOT NULL.
(Bug#12518)
The ROW() contructor returned an incorrect
result when comparison involved NULL values.
(Bug#12509)
Selecting from a view defined as a join over many tables could
result in a server crash due to miscalculation of the number of
conditions in the WHERE clause.
(Bug#12470)
MEMORY tables using B-Tree
index on 64-bit platforms could produce false table is full
errors.
(Bug#12460)
The CREATE_OPTIONS column of
INFORMATION_SCHEMA.TABLES showed
incorrect options for tables in
INFORMATION_SCHEMA.
(Bug#12397)
Mishandling of comparison for rows containing
NULL values against rows produced by an
IN subquery could cause a server crash.
(Bug#12392)
Selecting from a view after
INSERT statements for the view's
underlying table yielded different results than subsequent
selects.
(Bug#12382)
Concatenating USER() or
DATABASE() with a column produced
invalid results.
(Bug#12351)
Comparison of InnoDB multi-part primary keys
that include VARCHAR columns can
result in incorrect results.
(Bug#12340)
Renamed the rest() macro in
my_list.h to list_rest()
to avoid name clashes with user code.
(Bug#12327)
When restoring INFORMATION_SCHEMA as the
default database after failing to execute a stored procedure in
an inaccessible database, the server returned a spurious
ERROR 42000: Unknown database
'information_schema' message.
(Bug#12318)
Users created using an IP address or other alias rather than a
host name listed in /etc/hosts could not
set their own passwords.
(Bug#12302)
The NUMERIC_SCALE column of the
INFORMATION_SCHEMA.COLUMNS table
should be returned as 0 for integer columns.
It was being returned as NULL.
(Bug#12301)
Creating a view that included the
TIMESTAMPDIFF() function resulted
in a invalid view.
(Bug#12298)
CHECKSUM TABLE command returned
incorrect results for tables with deleted rows. After upgrading,
users who used stored checksum information to detect table
changes should rebuild their checksum data.
(Bug#12296)
Inserting NULL into a
GEOMETRY column for a table that has a
trigger could result in a server crash if the table was
subsequently dropped.
(Bug#12281)
myisampack failed to delete
.TMD temporary files when run with
-T option.
(Bug#12235)
A race condition between server threads could cause a crash if one thread deleted a stored routine while another thread was executing a stored routine. (Bug#12228)
Duplicate instructions in stored procedures resulted in incorrect execution when the optimizer optimized the duplicate code away. (Bug#12168)
XA allowed two active transactions to be
started with the same XID.
(Bug#12162)
NULL column definitions read incorrectly for
inner tables of nested outer joins.
(Bug#12154)
GROUP_CONCAT ignores the
DISTINCT modifier when used in a query
joining multiple tables where one of the tables has a single
row.
(Bug#12095)
A failure to obtain a lock for a LOCK IN SHARE
MODE query could result in a server crash.
(Bug#12082)
SELECT ... INTO within a trigger could cause a server crash.
(Bug#11973)var_name
Using cursors and nested queries for the same table, corrupted results were returned for the outer query. (Bug#11909)
A query using a LEFT JOIN, an
IN subquery on the outer table, and an
ORDER BY clause, caused the server to crash
when cursors were enabled.
(Bug#11901)
UNION query with
FULLTEXT could cause server crash.
(Bug#11869)
Some subqueries of the form SELECT ... WHERE ROW(...)
IN ( were being
handled incorrectly.
(Bug#11867)subquery)
Column names in subqueries must be unique, but were not being checked for uniqueness. (Bug#11864)
TRUNCATE
TABLE did not work with TEMPORARY
InnoDB tables.
(Bug#11816)
The mysql_info() C API function
could return incorrect data when executed as part of a
multi-statement that included a mix of statements that do and do
not return information.
(Bug#11688)
A trigger that included a SELECT
statement could cause a server crash.
(Bug#11587)
Built-in commands for the mysql client, such
as delimiter and \d are
now always parsed within files that are read using the
\. and source commands.
(Bug#11523)
Added portability check for Intel compiler to address a problem
compiling InnoDB code.
(Bug#11510)
ALTER TABLE did not move the table to
default database unless the new name was qualified with the
database name.
(Bug#11493)db_name.t RENAME
t
Joins on VARCHAR columns of
different lengths could produce incorrect results.
(Bug#11398)
For PKG installs on Mac OS X, the preinstallation and postinstallation scripts were being run only for new installations and not for upgrade installations, resulting in an incomplete installation process. (Bug#11380)
Prepared statement parameters could cause errors in the binary
log if the character set was cp932.
(Bug#11338)
Columns defined as TINYINT(1) were redefined
as TINYINT(4) when incorporated into a
VIEW.
(Bug#11335)
Stored procedures with particularly long loops could crash server due to memory leak. (Bug#11247, Bug#12297)
SET GLOBAL
TRANSACTION ISOLATION LEVEL was not working.
(Bug#11207)
A view was allowed to depend on a function that referred to a temporary table. (Bug#10970)
Issuing FLUSH INSTANCES followed by
STOP INSTANCE caused instance manager to
crash.
(Bug#10957)
User variables were not automatically cast for comparisons, causing queries to fail if the column and connection character sets differed. Now when mixing strings with different character sets but the same coercibility, allow conversion if one character set is a superset of the other. (Bug#10892)
An incorrect conversion from double to
ulonglong caused indexes not to be used for
BDB tables on HP-UX.
(Bug#10802)
DATE_ADD() and
DATE_SUB() were converting
invalid dates to NULL in
TRADITIONAL SQL mode rather
than rejecting them with an error.
(Bug#10627)
Views with multiple UNION and
UNION ALL
produced incorrect results.
(Bug#10624)
It was not possible to create a stored function with a spatial return value data type. (Bug#10499)
INSERT ... SELECT ... ON DUPLICATE KEY UPDATE
could fail with an erroneous “Column
'col_name' specified twice”
error.
(Bug#10109)
The only valid values for the PACK_KEYS table
option are 0 and 1, but other values were being accepted.
(Bug#10056)
Using a stored procedure that referenced tables in the
INFORMATION_SCHEMA database would return an
empty result set.
(Bug#10055, Bug#12278)
FLUSH TABLES WITH READ
LOCK combined with LOCK TABLE ..
WRITE caused deadlock.
(Bug#9459)
A data type of CHAR BINARY was not recognized
as valid for stored routine parameters.
(Bug#9048)
ISO-8601 formatted dates were not being
parsed correctly.
(Bug#7308)
On Windows when the
--innodb_buffer_pool_awe_mem_mb
option has been given, the server detects whether AWE support is
available and has been compiled into the server, and displays an
appropriate error message if not.
(Bug#6581)
Pathame values for options such as
--basedir or
--datadir didn't work on Japanese
Windows machines for directory names containing multi-byte
characters having a second byte of 0x5C
(“ \ ”).
(Bug#5439)
SHOW TABLE STATUS sometimes
reported a Row_format value of
Dynamic for MEMORY tables,
though such tables always have a format of
Fixed.
(Bug#3094)
Functionality added or changed:
MySQL Cluster:
Improved handling of the configuration variables
NoOfPagesToDiskDuringRestartACC,
NoOfPagesToDiskAfterRestartACC,
NoOfPagesToDiskDuringRestartTUP, and
NoOfPagesToDiskAfterRestartTUP should result
in noticeably faster startup times for MySQL Cluster.
(Bug#12149)
Added an optimization that avoids key access with
NULL keys for the
ref method when used in outer
joins.
(Bug#12144)
Added support of where clause for queries with FROM
DUAL.
(Bug#11745)
Maximum size of stored procedures increased from 64k to 4Gb. (Bug#11602)
SHOW CHARACTER SET and
INFORMATION_SCHEMA now properly report the
Latin1 character set as
cp1252.
(Bug#11216)
Added new
ER_STACK_OVERRUN_NEED_MORE error
message to indicate that, while the stack is not completely
full, more stack space is required.
(Bug#11213)
mysqldump now dumps triggers for each dumped
table. This can be suppressed with the
--skip-triggers
option.
(Bug#10431)
Added error message for users who attempt CREATE TABLE
... LIKE and specify a nontable in the
LIKE clause.
(Bug#6859)
Security improvement: Applied a patch that addresses a potential
zlib data vulnerability that could result in
an application crash. This only affects the binaries for
platforms that are linked statically against the bundled zlib
(most notably Microsoft Windows and HP-UX).
(CVE-2005-1849)
Bugs fixed:
MySQL Cluster: The MySQL Cluster backup log was invalid where the number of Cluster nodes was not equal to a power of 2. (Bug#11675)
Creation of the mysql group account failed
during the RPM installation.
(Bug#12348)
Updated dependency list for RPM builds to include missing
dependencies such as useradd and
groupadd.
(Bug#12233)
A delayed insert that would duplicate an existing record crashed the server instead. (Bug#12226)
When DROP DATABASE was called
concurrently with a DROP TABLE of
any table, the MySQL Server crashed.
(Bug#12212)
InnoDB: True
VARCHAR: Return
NULL columns in the format expected by MySQL.
(Bug#12186)
Information about a trigger was not displayed in the output of
SELECT ... FROM INFORMATION_SCHEMA.TRIGGERS
when the selected database was
INFORMATION_SCHEMA, prior to the trigger's
first invocation.
(Bug#12127)
InnoDB: Do not flush after each write, not
even before setting up the doublewrite buffer. Flushing can be
extremely slow on some systems.
(Bug#12125)
Two threads could potentially initialize different characters sets and overwrite each other. (Bug#12109)
big5 strings were not being stored in
FULLTEXT index.
(Bug#12075)
Character data truncated when GBK characters
0xA3A0 and 0xA1 are
present.
(Bug#11987)
ALTER TABLE when
sql_mode = 'TRADITIONAL' gave rise to an
invalid error message.
(Bug#11964)
Issuing successive FLUSH
TABLES WITH READ LOCK would cause the
mysql client to hang.
(Bug#11934)
mysql_install_db used static
localhost value in
GRANT tables even when server
host name is not localhost, such as
localhost.localdomain. This change is applied
to version 5.0.10b on Windows.
(Bug#11822)
Comparisons like SELECT "A\\" LIKE "A\\";
fail when using SET NAMES utf8;.
(Bug#11754)
Attempting to repair a table having a fulltext index on a column
containing words whose length exceeded 21 characters and where
myisam_repair_threads was
greater than 1 would crash the server.
(Bug#11684)
When used in a SELECT query
against a view, the
GROUP_CONCAT() function returned
only a single row.
(Bug#11412)
Multiplying ABS() output by a
negative number would return incorrect results.
(Bug#11402)
The LPAD() and
RPAD() functions returned the
wrong length to
mysql_fetch_fields().
(Bug#11311)
A UNIQUE VARCHAR column would be
mis-identified as MUL in table descriptions.
(Bug#11227)
DDL statements are now allowed in stored procedures if the
procedure is not invoked from a stored function or a trigger.
This fix also resolves a problem where a
TEMPORARY statement created by one stored
routine was inaccessible to another routine invoked during the
same connection.
(Bug#11126)
Calling the C API function
mysql_stmt_fetch() after all
rows of a result set were exhausted would return an error
instead of MYSQL_NO_DATA.
(Bug#11037)
SELECT @@local... returned
@@session... in the column header.
(Bug#10724)
Incorrect error message displayed if user attempted to create a
table in a nonexisting database using CREATE
syntax.
(Bug#10407)database_name.table_name
Unsigned LONG system variables may return
incorrect value when retrieved with a
SELECT for certain values.
(Bug#10351)
GROUP_CONCAT() sometimes returned
a result with a different collation from that of its arguments.
(Bug#10201)
Prepared statements were not being written to the Slow Query log. (Bug#9968)
The value of max_connections_per_hour was
capped by the unrelated
max_user_connections setting.
(Bug#9947)
In stored procedures, a cursor that fetched an empty string into
a variable would set the variable to NULL
instead.
(Bug#8692)
Added checks to prevent error when allocating memory when there was insufficient memory available. (Bug#7003)
Multiple SELECT SQL_CACHE queries in a stored
procedure causes error and client hang.
(Bug#6897)
A trigger dependent on a feature of one
sql_mode setting would cause an
error when invoked after the
sql_mode was changed.
(Bug#5891)
Functionality added or changed:
Incompatible Change:
The namespace for triggers has changed. Previously, trigger
names had to be unique per table. Now they must be unique within
the schema (database). An implication of this change is that
DROP TRIGGER syntax now uses a
schema name instead of a table name (schema name is optional
and, if omitted, the current schema will be used).
When upgrading from a version of MySQL 5 older than 5.0.10 to
MySQL 5.0.10 or newer, you must drop all triggers and
re-create them or DROP TRIGGER
will not work after the upgrade. A suggested procedure for
doing this is given in
Section 2.18.1.2, “Upgrading from MySQL 4.1 to 5.0”.
(Bug#5892)
MySQL Cluster:
A new -P option is available for use with the
ndb_mgmd client. When called with this
option, ndb_mgmd prints all configuration
data to stdout, then exits.
On Windows, the search path used by MySQL applications for
my.ini now includes
..\my.ini (that is, the application's
parent directory, and hence, the installation directory).
(Bug#10419)
The viewing of triggers and trigger metadata has been enhanced as follows:
An extension to the SHOW
command has been added: SHOW
TRIGGERS can be used to view a listing of
triggers. See Section 12.5.5.35, “SHOW TRIGGERS Syntax”, for details.
The INFORMATION_SCHEMA database now
includes a TRIGGERS table. See
Section 19.16, “The INFORMATION_SCHEMA TRIGGERS Table”, for details.
(Bug#9586)
It is no longer necessary to issue an explicit
LOCK TABLES for any tables
accessed by a trigger prior to executing any statements that
might invoke the trigger.
Previously, executing a statement that invoked a trigger would
cause problems unless a LOCK
TABLES was first issued for any tables accessed by the
trigger. The exact nature of the problem depended upon the MySQL
5.0 release being used: prior to 5.0.3, this resulted in a
crash; from 5.0.3 to 5.0.7, MySQL would issue a warning; in
5.0.9, the server would issue an error.
The same issue caused LOCK TABLES
to fail following
UNLOCK
TABLES if triggers were involved.
(Bug#8406, Bug#9581)
The MySQL server now starts correctly with all combinations of
--basedir and
--datadir, resolving an issue
introduced by the original fix for this bug in MySQL 4.1.9.
(Bug#7249)
See also Bug#7518.
Added
mysql_get_character_set_info() C
API function for obtaining information about the default
character set of the current connection.
An extension to the SHOW command
has been added: SHOW TRIGGERS can
be used to view a listing of triggers. See
Section 12.5.5.35, “SHOW TRIGGERS Syntax”, for details.
Add the --defaults-group-suffix
option. See Section 4.2.3.3, “Using Option Files”.
The bundled version of the readline library
was upgraded to version 5.0.
Triggers can now reference tables by name. See
Section 12.1.11, “CREATE TRIGGER Syntax”, for more information.
Add table_lock_wait_timeout
global system variable.
Bugs fixed:
Security Fix:
A vulnerability in zlib could result in a
buffer overflow and arbitrary code execution.
(Bug#11844, CVE-2005-2096, CVE-2005-1849)
MySQL Cluster:
The temporary tables created by an ALTER
TABLE on an NDB table
were visible to all SQL nodes in the cluster.
(Bug#12055)
MySQL Cluster:
NDB ignored the
Hostname option in the [ndbd
default] section of the cluster configuration file.
(Bug#12028)
MySQL Cluster:
The output of perror --help
did not display any information about the
--ndb option.
(Bug#11999)
MySQL Cluster: Attempting to create or drop tables during a backup would cause the cluster to shut down. (Bug#11942)
MySQL Cluster: ndb_mgmd leaked file descriptors. (Bug#11898)
MySQL Cluster: The MySQL Server left core files following shutdown if data nodes had failed. (Bug#11516)
MySQL Cluster:
When attempting to drop a table with a broken unique index,
NDB failed to drop the table and
erroneously report that the table was unknown.
(Bug#11355)
MySQL Cluster:
Trying to use a greater number of tables than specified by the
value of MaxNoOfTables caused table
corruption such that data nodes could not be restarted.
(Bug#9994)
The server did not compile correctly when using gcc4 on AMD64 platforms. (Bug#12040)
SHOW BINARY LOGS displayed a file
size of 0 for all log files but the current one if the files
were not located in the data directory.
(Bug#12004)
Increased the version number of the
libmysqlclient shared library from 14 to 15
because it is binary incompatible with the MySQL 4.1 client
library.
(Bug#11893)
The server crashed when dropping a trigger that invoked a stored procedure, if the procedure was not yet in the connection-specific stored routine cache. (Bug#11889)
SELECT ... NOT IN() gave unexpected results
when only static value present between the
().
(Bug#11885)
A recent optimizer change caused DELETE ... WHERE ...
NOT LIKE and DELETE ... WHERE ... NOT
BETWEEN to not properly identify the rows to be
deleted.
(Bug#11853)
Execution of a prepared statement that invoked a nonexistent or dropped stored routine would crash the server. (Bug#11834)
Selecting the result of an aggregate function for an
ENUM or SET
column within a subquery could result in a server crash.
(Bug#11821)
Creating a table with a SET or
ENUM column with the
DEFAULT 0 clause caused a server crash if the
table's character set was utf8.
(Bug#11819)
Incorrect column values could be retrieved from views defined
using statements of the form SELECT * FROM
.
(Bug#11771)tbl_name
When invoked within a view,
SUBTIME() returned incorrect
values.
(Bug#11760)
For several character sets, MySQL incorrectly converted the
character code for the division sign to the
eucjpms character set.
(Bug#11717)
Performing an ORDER BY on a
SELECT from a
VIEW produced unexpected results when
VIEW and underlying table had the same column
name on different columns.
(Bug#11709)
Execution of SHOW TABLES failed
to increment the Com_show_tables status
variable.
(Bug#11685)
LIKE pattern matching using prefix index didn't return correct result. (Bug#11650)
Invoking the DES_ENCRYPT()
function could cause a server crash if the server was started
without the --des-key-file
option.
(Bug#11643)
IP addresses not shown in ndb_mgm SHOW
command on second ndb_mgmd (or on ndb_mgmd restart).
(Bug#11596)
SHOW PROCEDURE/FUNCTION STATUS didn't work
for users with limited access.
(Bug#11577)
mysqlbinlog was failing the test suite on
Windows due to BOOL being
incorrectly cast to INT.
(Bug#11567)
The server crashed upon execution of a statement that used a
stored function indirectly (via a view) if the function was not
yet in the connection-specific stored routine cache and the
statement would update a
Handler_ status
variable. This fix allows the use of stored routines under
xxxLOCK TABLES without explicitly
locking the mysql.proc table. However, you
cannot use mysql.proc in statements that will
combine locking of it with modifications for other tables.
(Bug#11554)
Aliasing the column names in a VIEW did not
work when executing a SELECT
query on the VIEW.
(Bug#11399)
The mysql.proc table was not being created
properly with the proper utf8 character set
and collation, causing server crashes for stored procedure
operations if the server was using a multi-byte character set.
To take advantage of the bug fix,
mysql_fix_privilege_tables should be run to
correct the structure of the mysql.proc
table.
Note that it is necessary to run
mysql_fix_privileges_tables when upgrading
from a previous installation that contains the
mysql.proc table (that is, from a previous
5.0 installation). Otherwise, creating stored procedures might
not work.
(Bug#11365)
For prepared statements, the SQL parser did not disallow “
? ” parameter markers immediately
adjacent to other tokens, which could result in malformed
statements in the binary log. (For example, SELECT *
FROM t WHERE? = 1 could become SELECT * FROM
t WHERE0 = 1.)
(Bug#11299)
The C API function
mysql_stmt_reset() did not clear
error information.
(Bug#11183)
INFORMATION_SCHEMA.COLUMNS had some
inaccurate values for some data types.
(Bug#11057)
MySQL server would crash is a fetch was performed after a
ROLLBACK when
cursors were involved.
(Bug#10760)
When two threads competed for the same table, a deadlock could
occur if one thread also had a lock on another table through
LOCK TABLES and the thread was
attempting to remove the table in some manner while the other
thread tried to place locks on both tables.
(Bug#10600)
When used within a subquery,
SUBSTRING() returned an empty
string.
(Bug#10269)
Multiple-table UPDATE queries
using CONVERT_TZ() would fail
with an error.
(Bug#9979)
With strict SQL mode enabled, ALTER
TABLE reported spurious “Invalid default
value” messages for columns that had no
DEFAULT clause.
(Bug#9881)
mysql_fetch_fields() returned
incorrect length information for MEDIUM and
LONG TEXT and
BLOB columns.
(Bug#9735)
Within a stored procedure, selecting from a table through a view caused subsequent updates to the table to fail with a message that the table was read-locked. (Bug#9597)
Within a stored procedure that selects from a table, invoking another procedure that requires a write lock for the table caused that procedure to fail with a message that the table was read-locked. (Bug#9565)
Server-side prepared statements failed for columns with a
character set of ucs2.
(Bug#9442)
In SQL prepared statements, comparisons could fail for values
not equally space-padded. For example, SELECT 'a' =
'a '; returns 1, but PREPARE s FROM
'SELECT ?=?'; SET @a = 'a', @b = 'a '; PREPARE s FROM
'SELECT ?=?'; EXECUTE s USING @a, @b; incorrectly
returned 0.
(Bug#9379)
References to system variables in an SQL statement prepared with
PREPARE were evaluated during
EXECUTE to their values at
prepare time, not to their values at execution time.
(Bug#9359)
The server did not accept some fully-qualified trigger names. (Bug#8758)
Creating a trigger in one database that references a table in another database was being allowed without generating errors. (Bug#8751)
For server shutdown on Windows, error messages of the form
Forcing close of thread were being
written to the error log. Now connections are closed more
gracefully without generating error messages.
(Bug#7403)n
user: 'name'
For a stored procedure defined with SQL SECURITY
DEFINER characteristic,
CURRENT_USER() incorrectly
reported the use invoking the procedure, not the user who
defined it.
(Bug#7291)
Labels in stored routines did not work if the character set was
not latin1.
(Bug#7088)
Duplicate trigger names were allowed within a single schema. (Bug#6182)
For execution of a stored procedure that refers to a view, changes to the view definition were not seen. The procedure continued to see the old contents of the view. (Bug#6120)
The traditional SQL mode accepted invalid
dates if the date value provided was the result of an implicit
type conversion.
(Bug#5906)
In a shared Windows environment, MySQL could not find its
configuration file unless the file was in the
C:\ directory.
(Bug#5354)
Functions that evaluate to constants (such as
NOW() and
CURRENT_USER() were being
evaluated in the definition of a VIEW rather
than included verbatim.
(Bug#4663)
Functionality added or changed:
The handling of BIT columns has
been improved, and should now be much more reliable in a number
of cases.
(Bug#11572, Bug#11091, Bug#10617)
Recursion in stored routines is now disabled because it was crashing the server. We plan to modify stored routines to allow this to operate safely in a future release. (Bug#11394)
An attempt to create a TIMESTAMP
column with a display width (for example,
TIMESTAMP(6)) now results in a warning.
Display widths have not been supported for
TIMESTAMP since MySQL 4.1.
(Bug#10466)
mysql_real_escape_string() API
function now respects
NO_BACKSLASH_ESCAPES SQL mode.
(Bug#10214)
InnoDB: Made CHECK
TABLE killable.
(Bug#9730)
InnoDB: Make
innodb_thread_concurrency = 20 by default.
Bypass the concurrency checking if the setting is greater than
or equal to 20.
InnoDB: Various optimizations. Removed
unreachable debug code from nondebug builds. Added hints for the
branch predictor in gcc. Made assertions
occupy less space.
InnoDB: When creating or extending an
InnoDB data file, at most one megabyte at a
time is allocated for initializing the file. Previously,
InnoDB allocated and initialized 1 or 8
megabytes of memory, even if only a few 16-kilobyte pages were
to be written. This improves the performance of
CREATE TABLE in
innodb_file_per_table mode.
Bugs fixed:
MySQL Cluster: When trying to open a table that could not be discovered or unpacked, the cluster returned error codes which the MySQL server falsely interpreted as operating system errors. (Bug#10365)
The --master-data option for
mysqldump resulted in no error if the binary
log was not enabled. Now an error occurs unless the
--force option is given.
(Bug#11678)
When a table had a primary key containing a
BLOB column, creation of another
index failed with the error BLOB/TEXT column used in
key specification without keylength, even when the new
index did not contain a BLOB
column.
(Bug#11657)
Incorrect results when using GROUP BY ... WITH
ROLLUP on a VIEW.
(Bug#11639)
MySQL would not compile correctly on QNX due to missing
rint() function.
(Bug#11544)
A SELECT DISTINCT would work correctly with a col_name
MyISAM
table only when there was an index on
col_name.
(Bug#11484)
Using CONCAT_WS() on a column set
NOT NULL caused incorrect results when used
in a LEFT JOIN.
(Bug#11469)
Temporary tables were created in the data directory instead of
tmpdir.
(Bug#11440)
Running a CHECK TABLES on multiple views
crashed the server.
(Bug#11337)
Manually inserting a row with host='' into
mysql.tables_priv and performing a
FLUSH
PRIVILEGES would cause the server to crash.
(Bug#11330)
Wrong comparison method used in VIEW when
relaxed date syntax used (for example,
2005.06.10).
(Bug#11325)
Signed BIGINT would not accept
-9223372036854775808 as a
DEFAULT value.
(Bug#11215)
Optimizer performed range check when comparing unsigned integers to negative constants, could cause errors. (Bug#11185)
A cursor using a query with a filter on a
DATE or
DATETIME column would cause the
server to crash server after the data was fetched.
(Bug#11172)
The mysql_config script did not handle
symbolic linking properly.
(Bug#10986)
mysqldump failed when reloading a view if the view was defined in terms of a different view that had not yet been reloaded. mysqldump now creates a dummy table to handle this case. (Bug#10927)
If a prepared statement cursor is opened but not completely fetched, attempting to open a cursor for a second prepared statement will fail. (Bug#10794)
Combining cursors and subqueries could cause server crash or memory leaks. (Bug#10736)
Instances of the VAR_SAMP()
function in view definitions were converted to
VARIANCE(). This is incorrect
because VARIANCE() is the same as
VAR_POP(), not
VAR_SAMP().
(Bug#10651)
DES_ENCRYPT() and
DES_DECRYPT() require SSL support
to be enabled, but were not checking for it. Checking for
incorrect arguments or resource exhaustion was also improved for
these functions.
(Bug#10589)
For MEMORY tables, it was possible for
updates to be performed using outdated key statistics when the
updates involved only very small changes in a very few rows.
This resulted in the random failures of queries such as
UPDATE t SET col = col + 1 WHERE col_key = 2;
where the same query with no WHERE clause
would succeed.
(Bug#10178)
When used in joins, SUBSTRING()
failed to truncate to zero those string values that could not be
converted to numbers.
(Bug#10124)
Views did not use indexes on all appropriate queries. (Bug#10031)
Closing a cursor that was already closed would cause MySQL to hang. (Bug#9814)
The server would lose table-level CREATE
VIEW and SHOW VIEW
privileges following a
FLUSH
PRIVILEGES or server restart.
(Bug#9795)
mysqldump --xml did not format
NULL column values correctly.
(Bug#9657)
The --no-data option for
mysqldump was being ignored if table names
were given after the database name.
(Bug#9558)
Clients would hang following some errors with stored procedures. (Bug#9503)
mysqldump could crash for illegal or nonexistent table names. (Bug#9358)
A compression algorithm issue caused
myisampackto fail for very large data sets
(where the total size of all records in a single column was on
the order of 3 GB or more) on 64-bit platforms. (A fix for other
platforms was made in MySQL 5.0.6.)
(Bug#8321)
The ENCRYPT() and
SUBSTRING_INDEX() functions would
cause errors when used with a VIEW.
(Bug#7024)
SHOW CREATE VIEW did not take the
ANSI MODE into account when quoting
identifiers.
(Bug#6903)
In strict mode, an INSERT into a
view that did not include a value for a NOT
NULL column but that did include a
WHERE test on the same column would succeed,
This happened even though the
INSERT should have been prevented
due to the failure to supply a value for the NOT
NULL column.
(Bug#6443)
Starting with version 5.0.8, changes for MySQL Cluster can be found in the combined Change History.
Functionality added or changed:
Incompatible Change:
Previously, conversion of
DATETIME values to numeric form
by adding zero produced a result in
YYYYMMDDHHMMSS format. The result of
DATETIME+0 is now in
YYYYMMDDHHMMSS.000000 format.
(Bug#12268)
Replication:
Some data definition statements (CREATE
TABLE where the table was not a temporary table,
TRUNCATE
TABLE, DROP DATABASE,
and CREATE DATABASE) were not
being written to the binary log after a
ROLLBACK. This
also caused problems with replication.
As a result of this fix, the folowing statements now cause an implicit commit:
(Bug#6883)
Where a GROUP BY query uses a grouping column
from the query's SELECT clause,
MySQL now issues a warning. This is done because the SQL
standard states that any grouping column must unambiguously
reference a column of the table resulting from the query's
FROM clause, and allowing columns from the
SELECT clause to be used as
grouping columns is a MySQL extension to the standard.
By way of example, consider the following table:
CREATE TABLE users ( userid INT NOT NULL PRIMARY KEY, username VARCHAR(25), usergroupid INT NOT NULL );
MySQL allows you to use the alias in this query:
SELECT usergroupid AS id, COUNT(userid) AS number_of_users FROM users GROUP BY id;
However, the SQL standard requires that the column name be used, as shown here:
SELECT usergroupid AS id, COUNT(userid) AS number_of_users FROM users GROUP BY usergroupid;
Queries such as the first of the two shown above will continue
to be supported in MySQL; however, beginning with MySQL 5.0.8,
using a column alias in this fashion will generate a warning.
Note that in the event of a collision between column names
and/or aliases used in joins, MySQL attempts to resolve the
conflict by giving preference to columns arising from tables
named in the query's FROM clause.
(Bug#11211)
Using prepared statements within a stored routine
(PREPARE,
EXECUTE,
DEALLOCATE PREPARE) could cause
the client connection to be dropped after the routine returned.
In addition, executing a statement which called a function
deallocating the same statement caused the server to crash. This
is prevented by disabling dynamic SQL within stored routines.
This restriction was lifted in 5.0.13 for stored procedures, but not stored functions or triggers.
See also Bug#7115.
Added support for B'10' syntax for bit
literal.
(Bug#10650)
MEMORY tables now support indexes of up to
500 bytes. See Section 13.4, “The MEMORY (HEAP) Storage Engine”.
(Bug#10566)
Expanded on information provided in general log and slow query log for prepared statements. (Bug#8367, Bug#9334)
New sql_mode mode
NO_ENGINE_SUBSTITUTION
prevents automatic substitution of storage engine when the
requested storage engine is disabled or not compiled in.
(Bug#6877)
Bugs fixed:
Security Fix:
On Windows systems, a user with any of the following privileges
on *.* could crash mysqld
by issuing a USE LPT1; or USE
PRN; command:
In addition, any of the commands USE NUL;,
USE CON;, USE COM1;, or
USE AUX; would report success even though the
database was not in fact changed.
Although this bug was thought to be fixed previously, it was later discovered to be present in the MySQL 5.0.7-beta release for Windows.
(Bug#9148)
MySQL Cluster:
Setting TransactionInactiveTimeout = 0 did
not result in an infinite timeout.
(Bug#11290)
MySQL Cluster: mysqld processes did not reconnect to the cluster following a restart of ndb_mgmd. (Bug#11221)
MySQL Cluster: Insert records were incorrectly applied by ndb_restore, thus making restoring from backup inconsistent if the binary log contained inserts. (Bug#11166)
MySQL Cluster:
A DELETE performed as part of a
transaction caused an erroneous result.
(Bug#11133)
MySQL Cluster: Connections between data nodes and management nodes were not closed following shutdown of ndb_mgmd. (Bug#11132)
MySQL Cluster:
The ndb_mgm client's
SHOW command displayed incorrect
output after master data node failure.
(Bug#11050)
MySQL Cluster: When using dynamically allocated ports on Linux, the cluster would hang on initial startup. (Bug#10893)
MySQL Cluster:
Not allowing sufficient parallelism in the cluster's
configuration (for example, by setting
NoOfTransactions too small) caused
ndb_restore to fail without providing any
error messages.
(Bug#10294)
MySQL Cluster: Running ndb_select_count crashed the cluster when running on Red Hat Enterprise 4/64-bit/Opteron. (Bug#10058)
MySQL Cluster: Data nodes failed to restart on 64-bit Solaris. (Bug#9025)
MySQL Cluster: On 64-bit Solaris 9, the cluster timed out and crashed after the first query was made. (Bug#8918)
Replication: An invalid comaprison caused warnings for packet length in replication on 64-bit compilers. (Bug#11064)
Multiple range accesses in a subquery cause server crash. (Bug#11487)
Server crashed when using GROUP BY on the
result of a DIV operation on a
DATETIME value.
(Bug#11385)
INSERT INTO SELECT FROM produced incorrect result when using view
ORDER
BY.
(Bug#11298)
Possible NULL values in
BLOB columns could crash the
server when a BLOB was used in a
GROUP BY query.
(Bug#11295)
An outer join with an ON condition that
evaluated to false could return an incorrect result.
(Bug#11285)
An outer join with an empty derived table (a result from a subquery) returned no result. (Bug#11284)
CAST( ... AS DECIMAL) didn't work
for strings.
(Bug#11283)
Corrected a problem with IFNULL()
returning an incorrect result on 64-bit systems.
(Bug#11235)
The SHOW INSTANCE OPTIONS command in MySQL
Instance Manager displayed option values incorrectly for options
for which no value had been given.
(Bug#11200)
The default host name for MySQL server was always
mysql.
(Bug#11174)
Some internal functions did not take into account that, for
multi-byte character sets, CHAR
columns could exceed 255 bytes and
VARCHAR columns could exceed
65,535 bytes, which could cause the server to crash.
(Bug#11167)
There were locking problems with multiple-statement
DELETE statements performed
within a stored routine, such as incorrectly locking the table
to be read with a read lock rather than a write lock.
(Bug#11158)
Testing for crypt() support caused
compilation problems when using OpenSSL/yaSSL on HP-UX and Mac
OS X.
(Bug#11150, Bug#10675)
The NULLIF() function could
produce incorrect results if the first argument was
NULL.
(Bug#11142)
mysqld_safe would sometimes fail to remove
the pid file for the old mysql process after
a crash. As a result, the server would fail to start due to a
false A mysqld process already exists...
error.
(Bug#11122)
Calling a stored procedure that made use of an INSERT
... SELECT ... UNION SELECT ... query caused a server
crash.
(Bug#11060)
sql_data_access column of
routines table of
INFORMATION_SCHEMA was empty.
(Bug#11055)
SELECT
DISTINCT queries or GROUP BY
queries without MIN() or
MAX() could return inconsistent
results for indexed columns.
(Bug#11044)
A CREATE TABLE
statement would crash the server when no
database was selected.
(Bug#11028)db_name.tbl_name
LIKE ...
On Windows, mysqlshow did not interpret
wildcard characters properly if they were given in the table
name argument.
(Bug#10947)
The host name cache was not working. (Bug#10931)
A three byte buffer overflow in the client functions caused improper exiting of the client when reading a command from the user. (Bug#10841)
The mysql client would output a prompt twice following input of very long strings, because it incorrectly assumed that a call to the _cgets() function would clear the input buffer. (Bug#10840)
Setting @@sql_mode = NULL caused an erroneous
error message.
(Bug#10732)
When using a cursor with a prepared statement, the first execution returned the correct result but was not cleaned up properly, causing subsequent executions to return incorrect results. (Bug#10729)
Converting a VARCHAR column
having an index to a different type (such as
TINYTEXT) gave rise to an
incorrect error message.
Note that this bug fix induces a slight change in the behavior
of indexes: If an index is defined to be the same length as a
field (or is left to default to that field's length), and the
length of the field is later changed, then the index will adopt
the new length of the field. Previously, the size of the index
did not change for some field types (such as
VARCHAR) when the field type was
changed.
(Bug#10543)
InnoDB: Pad UTF-8
VARCHAR columns with
0x20. Pad UCS2
CHAR columns with
0x0020.
(Bug#10511)
InnoDB: Enforce maximum
CHAR_LENGTH() of UTF-8 data in
ON UPDATE CASCADE.
(Bug#10409)
SELECT * FROM returned incorrect results when called from a stored
procedure, where table
table had a primary
key.
(Bug#10136)
The granting and revocation of privileges on a stored routine
was performed when running the server with
--skip-grant-tables even after
the statement SET @@GLOBAL.automatic_sp_privileges =
1; was executed.
(Bug#9993)
A stored procedure run while the query cache was enabled could cause the server to crash. (Bug#9715)
Table names were not handled correctly when
lower_case_table_names = 2 if the table name
lettercase differed in the FROM and
WHERE clauses.
(Bug#9500)
SHOW CREATE DATABASE INFORMATION_SCHEMA
returned an “unknown database” error.
(Bug#9434)
SELECT DISTINCT ... GROUP BY
returned multiple
rows (it should return a single row).
(Bug#8614)constant
An issue with index merging could cause suboptimal index merge
plans to be chosen when searching by indexes created on
DATE columns. The same issue
caused the InnoDB storage engine to issue the
warning using a partial-field key prefix in
search.
(Bug#8441)
The mysqlhotcopy script was not parsing the
output of SHOW SLAVE STATUS
correctly when called with the --record_log_pos
option.
(Bug#7967)
A Boolean full-text search where a query contained more query terms than one-third of the query length caused the server to hang or crash. (Bug#7858)
When used in defining a view, the
TIME_FORMAT() function failed
with calculated values, for example, when passed the value
returned by SEC_TO_TIME().
(Bug#7521)
Views could be created with duplicate column names. (Bug#7448)
An ORDER BY clause sometimes had no effect on
the ordering of a result when selecting specific columns (as
opposed to using SELECT *) from a view.
(Bug#7422)
Using PREPARE to prepare a
statement that invoked a stored routine that executed the
prepared statement caused a Packets out of order
error the second time the routine was invoked. This
is prevented by disabling dynamic SQL within stored routines.
This restriction was lifted in 5.0.13 for stored procedures, but not for stored functions or triggers.
(Bug#7115)
Selecting from a view defined using SELECT SUM(DISTINCT
...) caused an error; attempting to execute a
SELECT * FROM INFORMATION_SCHEMA.TABLES query
after defining such a view crashed the server.
(Bug#7015)
Functionality added or changed:
Security Fix: A UDF library-loading vulnerability could result in a buffer overflow and code execution. (CVE-2005-2558)
Improved the optimizer to be able to use indexes for expressions
of the form indexed_col NOT
IN (val1,
val2, ...)
.
(Bug#10561)indexed_col NOT BETWEEN
val1 AND
val2
The table, type, and
rows columns of
EXPLAIN output can now be
NULL. This is required for using
EXPLAIN on
SELECT queries that use no
tables, such as EXPLAIN SELECT 1).
(Bug#9899)
All characters occurring on the same line following the
DELIMITER keyword will be set as delimiter.
For example, DELIMITER :; will set
:; as the delimiter. This behavior is now
consistent between MySQL 5.1 and MySQL 5.0.
(Bug#9879)
Added mysql_set_character_set()
C API function for setting the default character set of the
current connection. This allows clients to affect the character
set used by
mysql_real_escape_string().
(Bug#8317)
The --delayed-insert option
for mysqldump was disabled to avoid causing
problems with storage engines that do not support
INSERT DELAYED.
(Bug#7815)
Placeholders now can be used for LIMIT in
prepared statements.
(Bug#7306)
InnoDB: In stored procedures and functions,
InnoDB no longer takes full explicit table
locks for every involved table. Only “intention”
locks are taken, similar to those in the execution of an
ordinary SQL statement. This greatly reduces the number of
deadlocks.
SHOW BINARY LOGS now displays a
File_size column that indicates the size of
each file.
Removed WinMySQLAdmin from the source
distribution and from the “No Installer” Windows
distribution (it had already been removed from the “With
Installer” distribution before).
The behavior of the
Last_query_cost system
variable has been changed. The default value is now 0 (rather
than -1) and it now has session-level scope (rather than being
global). See Section 5.1.6, “Server Status Variables”, for
additional information.
Removed mysqlshutdown.exe and
mysqlwatch.exe from the Windows “No
Installer” distribution (they had already been removed
from the “With Installer” distribution before).
Removed those programs from the source distribution.
Bugs fixed:
MySQL would pass an incorrect key length to storage engines for
MIN(). This could cause spurious
warnings such as InnoDB: Warning: using a
partial-field key prefix in search to appear in the
.err log.
(Bug#13218, Bug#11039)
Build failures occurred when compiling the server on Windows using Visual Studio 6. (Bug#11153)
Corrected a problem where an incorrect data type was returned in
the result set metadata when using a prepared
SELECT
DISTINCT statement to select from a view.
(Bug#11111)
The server could crash due to an attempt to allocate too much
memory when GROUP BY
and
blob_colCOUNT(DISTINCT) were used.
(Bug#11088)
Multiple-row REPLACE could fail
on a duplicate-key error when having one
AUTO_INCREMENT key and one unique key.
(Bug#11080)
InnoDB: A duplicate key error occurred with
REPLACE in a table having an
AUTO_INCREMENT column.
(Bug#11005)
WITH ROLLUP did not sum values properly.
(Bug#10982)
Security update: A user with
limited privileges could obtain information about the privileges
of other users by querying objects in the
INFORMATION_SCHEMA database for which that
user did not have the requisite privileges.
(Bug#10964)
The value returned by the FIELD()
function was incorrect when its parameter list contained one or
more instances of NULL.
(Bug#10944)
Failure of a BEFORE trigger did not prevent
the triggering statement from performing its operation on the
row for which the trigger error occurred. Now the triggering
statement fails as described in Section 18.3, “Using Triggers”.
(Bug#10902)
The FEDERATED storage engine properly handled
outer joins, but not inner joins.
(Bug#10848)
Executing LOAD INDEX
INTO CACHE for a table while other threads where
selecting from the table caused a deadlock.
(Bug#10602)
The TIME_FORMAT() function
returned incorrect results with some format specifiers. See
Section 11.6, “Date and Time Functions”.
(Bug#10590)
The LAST_DAY() failed to return
NULL when supplied with an invalid argument.
See Section 11.6, “Date and Time Functions”.
(Bug#10568)
A problem with the my_global.h file caused
compilation of MySQL to fail on single-processor Linux systems
running 2.6 kernels.
(Bug#10364)
Corrected inappropriate error messages that were displayed when
attempting to set the read-only
warning_count and
error_count system variables.
(Bug#10339)
The MySQL Instance manager caused the version to be displayed as
unknown by SHOW INSTANCE
STATUS.
(Bug#10229)
A simultaneous CREATE TABLE ... SELECT FROM
and tableALTER
TABLE on the same
table caused the server to crash.
(Bug#10224)table
Under certain rare circumstances, inserting into the
mysql.host table could cause the server to
crash.
(Bug#10181)
Consistently report INFORMATION_SCHEMA table
names in uppercase in SHOW TABLE
STATUS output.
(Bug#10059)
Accessing InnoDB tables within stored
functions caused the MySQL server to crash. Now, statements that
perform an implicit or explicit commit or rollback are
prohibited within stored functions or triggers.
(Bug#10015)
Issuing a write lock for a table from one client prevented other
clients from accessing the table's metadata. For example, if one
client issued a LOCK TABLES
, then a second client attempting to execute a
mydb.mytable
WRITEUSE would
hang.
(Bug#9998)mydb;
Dropping stored routines when the MySQL server had been started
with --skip-grant-tables
generated extraneous warnings.
(Bug#9993)
The mysql_next_result() function
could hang if you were executing many statements in a
mysql_real_query() call and one
of those statements raised an error.
(Bug#9992)
The functions COALESCE(),
IF(), and
IFNULL() performed incorrect
conversions of their arguments.
(Bug#9939)
An incorrect result was returned from a view that selected a
COALESCE() expression from the
result of an outer join.
(Bug#9938)
InnoDB: Do very fast shutdown only if
innodb_fast_shutdown = 2, but wait for
threads to exit and release allocated memory if
innodb_fast_shutdown = 1. Starting with
MySQL/InnoDB 5.0.5, InnoDB would do brutal
shutdown also when innodb_fast_shutdown = 1.
(Bug#9673)
Using ORDER BY to sort the results of an
IF() that contained a
FROM_UNIXTIME() expression
returned incorrect results due to integer overflow.
(Bug#9669)
On Windows, with
lower_case_table_names set to
2, using ALTER TABLE to alter a
MEMORY or InnoDB table
that had a mixed-case name also improperly changed the name to
lowercase.
(Bug#9660)
The combination of COUNT(),
DISTINCT, and
CONCAT() sometimes triggered a
memory deallocation bug on Windows resulting in a server crash.
(Bug#9593)
INSERT BEFORE triggers were not being
activated for implicit inserts (LOAD
DATA).
(Bug#8755)
The ucs2_turkish_ci collation failed with upper('i'). UPPER/LOWER now can return a string with different length. (Bug#8610)
The server timed out SSL connections too quickly on Windows. (Bug#8572)
If a stored function contained a
FLUSH statement, the function
crashed when invoked. FLUSH now
is disallowed within stored functions.
(Bug#8409)
OPTIMIZE run on an InnoDB
table did not return a Table is full
error if there was insufficient room in the tablespace.
(Bug#8135)
An incorrect result was obtained for columns that included an
aggregate function as part of an expression, and when
WITH ROLLUP was used with GROUP
BY.
(Bug#7914)
Queries with ROLLUP returned wrong results
for expressions containingGROUP BY columns.
(Bug#7894)
The second invocation of a stored procedure that selected from a
view defined as a join using ON in the join
condition could cause the server to crash.
(Bug#6866)
INSERT BEFORE triggers were not being
activated for
INSERT ...
SELECT statements.
(Bug#6812)
INSERT or
UPDATE when the
WHERE clause contained a correlated subquery
that referred to a column of the table being modified caused the
server to crash.
(Bug#6384)
MySQL was adding a DEFAULT clause to
ENUM columns that included no
explicit DEFAULT and were defined as
NOT NULL. (This is supposed to happen only
for columns that are NULL.)
(Bug#6267)
Using ALTER TABLE for a table
that had a trigger caused a crash when executing a statement
that activated the trigger, and also a crash later with
USE for
the database containing the table.
(Bug#5894)db_name
Triggers with dropped functions caused crashes. (Bug#5893)
Triggers were not being activated for multiple-table
UPDATE or
DELETE statements.
(Bug#5860)
The incorrect sequence of statements HANDLER
without a
preceding tbl_name READ
index_name NEXTHANDLER for an
tbl_name
READ index_name =
(value_list)InnoDB table resulted in a server crash
rather than an error.
(Bug#5373)
Multiple-table DELETE always
deleted on the fly from the first table that was to be deleted
from. In some cases, when using many tables and when necessary
to access the same row twice in the first table, some rows to be
deleted from other tables could be missed.
Functionality added or changed:
Incompatible Change:
The behavior of LOAD
DATA INFILE and
SELECT ... INTO
OUTFILE has changed when the FIELDS
TERMINATED BY and FIELDS ENCLOSED
BY values both are empty. Formerly, a column was read
or written the display width of the column. For example,
INT(4) was read or written using a field with
a width of 4. Now columns are read and written using a field
width wide enough to hold all values in the field. However, data
files written before this change was made might not be reloaded
correctly with LOAD
DATA INFILE for MySQL 4.1.12 and up. This change also
affects data files read by mysqlimport and
written by mysqldump --tab, which use
LOAD DATA
INFILE and
SELECT ... INTO
OUTFILE. For more information, see
Section 12.2.6, “LOAD DATA INFILE
Syntax”.
(Bug#12564)
Incompatible Change:
MyISAM and InnoDB tables
created with DECIMAL columns in
MySQL 5.0.3 to 5.0.5 will appear corrupt after an upgrade to
MySQL 5.0.6. Dump such tables with mysqldump
before upgrading, and then reload them after upgrading. (The
same incompatibility will occur for these tables created in
MySQL 5.0.6 after a downgrade to MySQL 5.0.3 to 5.0.5.)
(Bug#10465, Bug#10625)
When the server cannot read a table because it cannot read the
.frm file, print a message that the table
was created with a different version of MySQL. (This can happen
if you create tables that use new features and then downgrade to
an older version of MySQL.)
(Bug#10435)
The GRANT and
REVOKE statements now support an
object_type clause to be used for
disambiguating whether the grant object is a table, a stored
function, or a stored procedure. Use of this clause requires
that you upgrade your grant tables. See
Section 4.4.5, “mysql_fix_privilege_tables — Upgrade MySQL System Tables”.
(Bug#10246)
Added REFERENCED_TABLE_SCHEMA,
REFERENCED_TABLE_NAME, and
REFERENCED_COLUMN_NAME columns to the
KEY_COLUMN_USAGE table of
INFORMATION_SCHEMA.
(Bug#9587)
The use of SESSION or
GLOBAL for user variables, or for local
variables in stored routines, is now disallowed.
(Bug#9286)
New /*> prompt for
mysql. This prompt indicates that a
/* ... */ comment was begun on an earlier
line and the closing */ sequence has not yet
been seen.
(Bug#9186)
The INFORMATION_SCHEMA.SCHEMATA
table now has a DEFAULT_COLLATION_NAME
column.
(Bug#8998)
Added a --show-warnings option to
mysql to cause warnings to be shown after
each statement if there are any. This option applies to
interactive and batch mode. In interactive mode,
\w and \W may be used to
enable and disable warning display.
(Bug#8684)
If strict SQL mode is enabled,
VARCHAR and
VARBINARY columns with a length
greater than 65,535 no longer are silently converted to
TEXT or
BLOB columns. Instead, an error
occurs.
(Bug#8295, Bug#8296)
Removed a limitation that prevented use of FIFOs as logging targets (such as for the general query log). This modification does not apply to the binary log and the relay log. (Bug#8271)
SHOW VARIABLES now shows the
slave_compressed_protocol,
slave_load_tmpdir and
slave_skip_errors system
variables.
(Bug#7800)
InnoDB: When the maximum length of
SHOW INNODB STATUS output would
be exceeded, truncate the beginning of the list of active
transactions, instead of truncating the end of the output.
(Bug#5436)
Updated version of libedit to 2.9.
(Bug#2596)
InnoDB: If
innodb_locks_unsafe_for_binlog
is enabled and the isolation level of the transaction is not set
to SERIALIZABLE,
InnoDB uses a consistent read for select in
clauses such as
INSERT INTO ...
SELECT and UPDATE ... (SELECT) that
do not specify FOR UPDATE or LOCK IN
SHARE MODE. Thus, no locks are set to rows read from
selected table.
Added the
div_precision_increment system
variable, which indicates the number of digits by which to
increase the scale of the result of division operations
performed with the / operator.
Removed mysqlshutdown.exe and
mysqlwatch.exe from the Windows “With
Installer” distribution.
The precision of the DECIMAL data
type has been increased from 64 to 65 decimal digits.
Added the --log-bin-trust-routine-creators
server option for setting the
log_bin_trust_routine_creators system
variable from the command line.
Implemented the STMT_ATTR_PREFETCH_ROWS
option for the
mysql_stmt_attr_set() C API
function. This sets how many rows to fetch at a time when using
cursors with prepared statements.
Added the log_bin_trust_routine_creators
system variable, which applies when binary logging is enabled.
It controls whether stored routine creators can be trusted not
to create stored routines that will cause unsafe events to be
written to the binary log.
Removed unused system variable
myisam_max_extra_sort_file_size.
Changed default value of
myisam_data_pointer_size from 4
to 6. This allows us to avoid table is full
errors for most cases.
Added a --debug option
to my_print_defaults.
The variable concurrent_insert
now takes 3 values. Setting this to 2 changes
MyISAM to do concurrent inserts to end of
table if table is in use by another thread.
Bugs fixed:
Security Fix:
mysql_install_db created the
mysql_install_db.X file with a predictable
file name and insecure permissions, which allowed local users to
execute arbitrary SQL statements by modifying the file's
contents.
(CVE-2005-1636)
Replication: Statements that create and drop triggers were not being written to the binary log, which affects replication and data recovery options. Trigger-related statements now are logged, subject to the issues and limitations discussed in Section 18.5, “Binary Logging of Stored Programs”. (Bug#10417)
Replication: Statements that create and use stored routines were not being written to the binary log, which affects replication and data recovery options. Stored routine-related statements now are logged, subject to the issues and limitations discussed in Section 18.5, “Binary Logging of Stored Programs”. (Bug#2610)
MERGE tables could fail on Windows due to
incorrect interpretation of path name separator characters for
file names in the .MRG file.
(Bug#10687)
Repeated calls to ABS() when the
argument evaluated to NULL crashed the
server.
(Bug#10599)
SELECT 0/0 returned 0
rather than NULL.
(Bug#10404)
INSERT ...
ON DUPLICATE KEY UPDATE with MERGE
tables, which do not have unique indexes, caused the server to
crash.
(Bug#10400)
AUTO_INCREMENT in InnoDB
tables could assign the same value for several rows.
(Bug#10359)
mysqldump crashed using the
--complete-insert option while
dumping tables with a large number of long column names.
(Bug#10286)
Incomplete results were returned from
INFORMATION_SCHEMA.COLUMNS for
INFORMATION_SCHEMA tables for
non-root users.
(Bug#10261)
mysql.cc did not compile correctly using
VC++ on Windows.
(Bug#10245)
Using #pragma interface or #pragma
implementation in source files caused portability
issues for cygwin.
(Bug#10241)
Corrected a problem where DEFAULT values were
not assigned properly to BIT(1) or
CHAR(1) columns if certain other columns
preceded them in the table definition.
(Bug#10179)
The BLACKHOLE storage engine failed in
testing, causing the server to crash.
(Bug#10175)
The optimizer was choosing suboptimal execution plans for
certain outer joins where the right table of a left join (or
left table of a right join) had both ON and
WHERE conditions.
(Bug#10162)
Corrected a problem resolving outer column references in correlated subqueries when using the prepared statements. (Bug#10041)
INFORMATION_SCHEMA tables were inaccessible
depending on the lettercase used to refer to them.
(Bug#10018)
awk script portability problems were found in
cmd-line-utils/libedit/makelist.sh
.
(Bug#9954)
The error message for exceeding
MAX_CONNECTIONS_PER_HOUR mistakenly referred
to max_connections.
(Bug#9947)
RENAME TABLE for an
ARCHIVE table failed if the
.arn file was not present.
(Bug#9911)
A CHECK TABLE statement whose
arguments were a view name followed by a table name caused the
server to crash.
(Bug#9897)
my_print_defaults was ignoring the
--defaults-extra-file option or
crashing when the option was given.
(Bug#9851, Bug#9136)
Within a stored procedure, attempting to update a view defined
as an inner join failed with a Table
' error.
(Bug#9841)tbl_name' was locked with a READ
lock and can't be updated
The INFORMATION_SCHEMA.COLUMNS
table was missing columns of views for which the user has
access.
(Bug#9838)
Use of a CHAR or
VARCHAR column with
MIN() or
MAX() and GROUP BY ...
WITH ROLLUP caused the server to crash.
(Bug#9820)
Usi DISTINCT AVG() with GROUP BY ...
WITH ROLLUP caused the server to crash.
(Bug#9800)
Using AVG(DISTINCT) with
GROUP BY ... WITH ROLLUP caused the server to
crash.
(Bug#9799)
Using GROUP BY ... WITH ROLLUP on an indexed
column in an InnoDB table could cause the
server to crash.
(Bug#9798)
Corrected some failures of prepared statements for SQL
(PREPARE plus
EXECUTE) to return all rows for
some SELECT statements.
(Bug#9777, Bug#9096)
CREATE TABLE ... LIKE did not work correctly
when lower_case_table_names was
set on a case-sensitive file system and the source table name
was not given in lowercase.
(Bug#9761)
Corrected an inability to select from a view within a stored procedure. (Bug#9758)
net_read_timeout and
net_write_timeout were not
being respected on Windows.
(Bug#9721)
libsupc++ was longer required for building on
FreeBSD 5.3.
(Bug#9714)
The mysql_stmt_attr_set() C API
function now returns an error for option values that are defined
in mysql.h but not yet implemented, such as
CURSOR_TYPE_SCROLLABLE.
(Bug#9643)
Memory block allocation did not function correctly for the query cache in the embedded server. (Bug#9549)
CREATE TABLE t AS SELECT UUID() created a
VARCHAR(12) column, which is too small to
hold the 36-character result from
UUID().
(Bug#9535)
SELECT
DISTINCT with a prepared statement that used a cursor
could cause the server to crash.
(Bug#9520)
NULL key parts in hash indexes on
VARCHAR columns were not handled
correctly, resulting in incorrect query results.
(Bug#9489, Bug#10176)
The mysql_stmt_execute() and
mysql_stmt_reset() C API
functions now close any cursor that is open for the statement,
which prevents a server crash.
(Bug#9478)
SELECT from
INFORMATION_SCHEMA tables failed if the
statement has a GROUP BY clause and an
aggregate function in the select list.
(Bug#9404)
MAX() for an INT
UNSIGNED (unsigned 4-byte integer) column could return
negative values if the column contained values larger than
231.
(Bug#9298)
Disabled binary logging within stored routines to avoid writing
spurious extra statements to the binary log. For example, if a
routine p() executes an
INSERT statement, then for
CALL p(), the
CALL statement appears in the
binary log, but not the INSERT
statement.
(Bug#9100)
FORMAT() now performs better
rounding for double values (for example,
FORMAT(4.55,1) returns
4.6, not 4.5).
(Bug#9060)
SHOW CREATE VIEW got confused and
could not find the view if there was a temporary table with the
same name as the view.
(Bug#8921)
Selecting from a single-table view defined on multiple-table views caused a server crash. (Bug#8528)
Remove extra slashes in --tmpdir value (for
example, convert /var//tmp to
/var/tmp, because they caused various
errors.
(Bug#8497)
Invoking a stored function that executed a
SHOW statement resulted in a
server crash.
(Bug#8408)
An error in the implementation of the MyISAM
compression algorithm caused myisampack to
fail with very large sets of data (total size of all the records
in a single column needed to be at least 3 GB in order to
trigger this issue).
(Bug#8321)
Added Create_routine_priv,
Alter_routine_priv, and
Execute_priv privileges to the
mysql.host privilege table. (They had been
added to mysql.db in MySQL 5.0.3 but not to
the host table.)
(Bug#8166)
A deadlock resulted from using
FLUSH TABLES WITH READ
LOCK while an INSERT
DELAYED statement was in progress.
(Bug#7823)
In strict SQL mode, some assignments to numeric columns that
should have been rejected were not (such as the result of an
arithmetic expression or an explicit
CAST() operation).
(Bug#6961)
For MERGE tables, avoid writing absolute path
names in the .MRG file for the names of the
constituent MyISAM tables so that if the data
directory is moved, MERGE tables will not
break. For mysqld, write just the
MyISAM table name if it is in the same
database as the MERGE table, and a path
relative to the data directory otherwise. For the embedded
servers, absolute path names may still be used.
(Bug#5964)
Multiple calls to a stored procedure that assigned the result of
a subquery to a variable or compared it to a value with
IN could cause the server to crash.
(Bug#5963)
If the file named by a
--defaults-extra-file option
does not exist or is otherwise inaccessible, an error now
occurs.
(Bug#5056)
configure did not properly recognize whether NPTL was available on Linux. (Bug#2173)
No public release of MySQL 5.0.5 was made. The changes described in this section are available in MySQL 5.0.6.
Functionality added or changed:
MySQL Cluster:
More informative error messages are provided when a query is
issued against an NDB table that
has been modified by another mysqld server.
(Bug#6762)
InnoDB: When
foreign_key_checks = 0,
ALTER TABLE and
RENAME TABLE will ignore any type
incompatibilities between referencing and referenced columns.
Thus, it will be possible to convert the character sets of
columns that participate in a foreign key. Be sure to convert
all tables before modifying any data!
(Bug#9802)
SHOW VARIABLES no longer displays
the deprecated log_update system variable.
(Bug#9738)
Added support for the BIT data
type to the MEMORY,
InnoDB, and BDB storage
engines.
The behavior controlled by the
--innodb-fast-shutdown option now
can be changed at runtime by setting the value of the global
innodb_fast_shutdown system
variable. It now accepts values 0, 1 and 2 (except on Netware
where 2 is disabled). If set to 2, then when the MySQL server
shuts down, InnoDB will just flush its logs
and shut down brutally (and quickly) as if a MySQL crash had
occurred; no committed transaction will be lost, but a crash
recovery will be done at next startup.
Bugs fixed:
Security Fix:
Starting mysqld with
--user=
caused it to run using the privileges of the account from which
it was invoked, including the non_existent_userroot account.
(Bug#9833)
A memory leak occurred when selecting from a view that contained a subquery. (Bug#10107)
Setting the storage_engine
system variable to MEMORY succeeded, but
retrieving the variable resulted in a value of
HEAP (the old name for the
MEMORY storage engine) rather than
MEMORY.
(Bug#10039)
Queries containing CURRENT_USER()
incorrectly were registered in the query cache.
(Bug#9796)
Invoking a stored function that returned a value having an
ENUM or SET
data type caused the server to crash.
(Bug#9775)
A string length comparison problem caused
mysql to fail when loading dump files
containing certain escape sequences containing a backslash
character (\).
(Bug#9756)
After an internal temporary table became too large in memory and had to be converted to an on-disk table, the error indicator was not cleared and the query failed with error 1023 Can't find record in ''. (Bug#9703)
Use of a subquery that used WITH ROLLUP in
the FROM clause of the main query sometimes
resulted in a Column cannot be null error.
(Bug#9681)
InnoDB: Assertion failures of types
ut_a(cursor->old_stored ==
BTR_PCUR_OLD_STORED) and
prebuilt->template_type == 0 could occur
when performing multi-table updates. This bug was introduced in
4.1.10 and 4.0.24.
(Bug#9670)
A problem with readlinecaused the
mysql client to crash when the user pressed
Control-R..
(Bug#9568)
Executing LOCK TABLES and then
calling a stored procedure caused an error and resulting in the
server thinking that no stored procedures exist.
(Bug#9566)
The server died with signal 11 if a nonexistent location was specified for the location of the binary log. Now the server exits after printing an appropriate error message. (Bug#9542)
Incorrect results were returned for queries of the form
SELECT ... LEFT JOIN ... WHERE EXISTS
(, where the
subquery selected rows based on an subquery)IS
NULL condition.
(Bug#9516)
A segmentation fault in mysqlcheck occurred
when the last table checked in
--auto-repair mode returned
an error (such as the table being a MERGE
table).
(Bug#9492)
Within a stored procedure, attempting to execute a
multiple-table UPDATE failed with
a Table ' error.
(Bug#9486)tbl_name' was
locked with a READ lock and can't be updated
mysqlshow displayed an incorrect row count for tables. (Bug#9391)
InnoDB: Next-key locking did not allow
inserts which did not produce a “phantom”. If the
range is of type 'a' <= uniquecolumn,
InnoDB lock only the RECORD, if the record
with the column value 'a' exists in a
CLUSTERED index. This allows inserts before a range.
(Bug#9354)
The optimizer did not compute the union of two ranges for the
OR operator correctly.
(Bug#9348)
Corrected a failure to resolve a column reference correctly for
a LEFT JOIN that compared a join column to an
IN subquery.
(Bug#9338)
OPTIMIZE TABLE was written twice
to the binary log when used on InnoDB tables.
(Bug#9149)
Multiple-table updates could produce spurious data-truncation warnings if they used a join across columns that are indexed using a column prefix. (Bug#9103)
Invocation of a stored function that returned a value having a
BLOB data type caused the server
to crash.
(Bug#9102)
For stored functions that should return a
YEAR value, corrected a failure
of the value to be in YEAR
format.
(Bug#8861)
Selecting from a view containing a subquery caused the server to hang. (Bug#8490)
TIMEDIFF() with a negative time
first argument and positive time second argument produced
incorrect results.
(Bug#8068)
Invocation of a stored function that returned a value having a
BIT data type caused the server
to crash.
(Bug#7648)
SET @var= CAST(NULL AS [INTEGER|CHAR]) now
sets the result type of the variable to
INTEGER/CHAR.
(Bug#6598)
The client/server protocol allowed the server to close the connection before sending the final error message. The problem could show up as a Lost connection to MySQL server error during a query when attempting to connect and access a nonexistent database. (Bug#6387, Bug#9455)
Column references were not properly resolved when an outer join involving a view contained a subquery and the column was used both in the subquery and the outer query. (Bug#6107, Bug#6106)
InnoDB: Prevent ALTER
TABLE from changing the storage engine if there are
foreign key constraints on the table.
(Bug#5574, Bug#5670)
Functionality added or changed:
Replication: The way the time zone information is stored in the binary log was changed, so that it is now possible to have a replication master and slave running with different global time zones. A disadvantage is that replication from 5.0.4 masters to pre-5.0.4 slaves is impossible.
Added ENGINE=MyISAM table option when
creating mysql.proc table in
mysql_create_system_tables script to make
sure the table is created as a MyISAM table
even if the default storage engine has been changed.
(Bug#9496)
SHOW CREATE TABLE for an
INFORMATION_SCHEMA table no longer prints a
MAX_ROWS value because the value has no
meaning.
(Bug#8941)
Invalid DEFAULT values for
CREATE TABLE now generate errors.
(Bug#5903)
Added --show-table-type option
to mysqlshow, to display a column indicating
the table type, as in SHOW FULL TABLES.
(Bug#5036)
New configuration directives !include and
!includedir implemented for including option
files and searching directories for option files. See
Section 4.2.3.3, “Using Option Files”, for usage.
Added --with-big-tables
compilation option to configure. (Previously
it was necessary to pass -DBIG_TABLES to the
compiler manually in order to enable large table support.) See
Section 2.16.2, “Typical configure Options”, for details.
Bugs fixed:
Security Fix:
Information in INFORMATION_SCHEMA could be
exposed to a user with insufficient privileges.
(Bug#7214)
MySQL Cluster:
The commit count cache for NDB was
not properly invalidated when deleting a record using a cursor.
(Bug#8585)
Replication:
If, on a replication master, a
LOAD DATA
INFILE operation was interrupted (by, for example, an
integrity constraint violation or killed connection), the slave
skipped the LOAD DATA
INFILE entirely, thus missing changes if this command
permanently inserted or updated table records before being
interrupted.
(Bug#3247)
mysql.server no longer uses nonportable alias command or LSB functions. (Bug#9852)
A server installed as a Windows service and started with
--shared-memory could not be
stopped.
(Bug#9665)
Selecting a BIT column failed if
the binary client/server protocol was used.
(Bug#9608)
Creating a PRIMARY KEY on a table having a
BIT column caused the server to
crash.
(Bug#9571)
ENUM and SET
columns in InnoDB tables were treated
incorrectly as character strings. This bug did not manifest
itself with latin1 collations, but it caused
malfunction with utf8. Old tables will
continue to work. In new tables,
ENUM and SET
will be stored internally as unsigned integers.
(Bug#9526)
An error in division of floating point numbers could cause nine
zeros (000000000) to be inserted in the
middle of the quotient.
(Bug#9501)
Fixed option-parsing code for the embedded server to understand
K, M, and
G suffixes for the
net_buffer_length and
max_allowed_packet options.
(Bug#9472)
Some user variables were not being handled with “implicit” coercibility. (Bug#9425)
Using CREATE TABLE
... SELECT or
INSERT INTO ...
SELECT to select from multiple-table view caused the
server to crash.
(Bug#9398, Bug#8703)
Multiple executions of a prepared statement involving a join of
an INFORMATION_SCHEMA table with another
table could lead to a crash of the server.
(Bug#9383)
An InnoDB test suite failure was caused by a
locking conflict between two server instances at server shutdown
or startup. This conflict on advisory locks appears to be the
result of a bug in the operating system; these locks should be
released when the files are closed, but somehow that does not
always happen immediately in Linux.
(Bug#9381)
Allow extra HKSCS and cp950 characters (big5
extension characters) to be accepted in big5
columns.
(Bug#9357)
The value of the CHARACTER_MAXIMUM_LENGTH and
CHARACTER_OCTET_LENGTH columns of the
INFORMATION_SCHEMA.COLUMNS table
must be NULL for numeric columns, but were
not.
(Bug#9344)
INFORMATION_SCHEMA tables had an implicit
upper limit for the number of rows. As a result, not all data
could be returned for some queries.
(Bug#9317)
InnoDB: True
VARCHAR:
InnoDB stored the 'position' of a row wrong
in a column prefix primary key index; this could cause MySQL to
complain ERROR 1032: Can't find record … in
an update of the primary key, and also some ORDER
BY or DISTINCT queries.
(Bug#9314)
ORDER BY sometimes caused incorrect sorting
of UTF8 data.
(Bug#9309)
The utf8_spanish2_ci and
ucs2_spanish2_ci collations no longer
consider r equal to rr .
If you upgrade to this version from an earlier version, you
should rebuild the indexes of any affected tables.
(Bug#9269)
CREATE OR REPLACE VIEW and
ALTER VIEW now require the
CREATE VIEW and
priv privileges, not
CREATE VIEW and
DELETE.
(DELETE is a row-level privilege,
not a table-level privilege.)
(Bug#9260)
Using GROUP BY on a decimal expression caused
the server to crash.
(Bug#9210)
mysqldump dumped core when invoked with
--tmp and
--single-transaction options
and a nonexistent table name.
(Bug#9175)
Calling mysql_stmt_close() for a
single-row result set could cause the server to crash.
(Bug#9159)
Setting the max_error_count
system variable to 0 resulted in a setting of 1.
(Bug#9072)
The use of XOR together with NOT
ISNULL() erroneously resulted in some outer joins
being converted to inner joins by the optimizer.
(Bug#9017)
Two prepared statements for single-row result sets being open simultaneously caused a Commands out of sync error error. (Bug#8880)
Extraneous comparisons between NULL values in
indexed columns were performed by the optimzer for operators
such as = that are never true for
NULL.
(Bug#8877)
In the client/server protocol for prepared statements, reconnection failed when the connection was killed with reconnection enabled. (Bug#8866)
In prepared statements, subqueries containing parameters were
erroneously treated as const
tables during preparation, resulting in a server crash.
(Bug#8807)
Do not try to space-pad BLOB
columns containing ucs2 characters.
(Bug#8771)
This regression was introduced by Bug#7350.
The warning message from
GROUP_CONCAT() did not always
indicate the correct number of lines.
(Bug#8681)
InnoDB: SQL statements were not rolled back
on error.
(Bug#8650)
Too many rows were returned from queries that combined
ROLLUP and LIMIT if
SQL_CALC_FOUND_ROWS was given.
(Bug#8617)
Incorrect results were returned from queries that combined
SELECT
DISTINCT, GROUP BY , and
ROLLUP.
(Bug#8616)
Queries that combined
SELECT
DISTINCT, SUM(), and
ROLLUP could cause the MySQL server to crash.
(Bug#8615)
The tee command could sometimes cause the
mysql client to crash.
(Bug#8499)
DROP TABLE did not drop triggers
that were defined for the table. DROP
DATABASE did not drop triggers in the database.
(Bug#6559, Bug#5859)
Added linking with libsupc++ on Fedora Core 3
to get language support functions.
(Bug#6554)
Unions between binary and nonbinary columns failed due to a a collation coercibility problem. (Bug#6519)
Using CONVERT('0000-00-00',DATE)
or CAST('0000-00-00' as DATE)
with the NO_ZERO_DATE SQL mode
enabled now produces a warning.
(Bug#6145)
TRADITIONAL SQL mode should
prevent inserts where a column with no default value is omitted
or set to a value of DEFAULT; however, in
some cases, this restriction was not enforced.
(Bug#5986)
Inserting a zero date in a DATE,
DATETIME or
TIMESTAMP column during
TRADITIONAL mode now produces
an error.
(Bug#5933)
CAST() now produces warnings when
casting incorrect INTEGER and
CHAR values. This also applies to
implicit string to number
casts.
(Bug#5912)
An error now occurs if you try to insert an invalid value via a
stored procedure in STRICT mode.
(Bug#5907)
STR_TO_DATE() now produces errors
in strict mode (and warnings otherwise) when given an illegal
argument.
(Bug#5902)
Inserting a zero date into a
DATETIME column in
TRADITIONAL mode now produces
an error.
ALTER TABLE now fails in
STRICT mode if the alteration generates
warnings.
This Beta release, as any other pre-production release, should not be installed on “production” level systems or systems with critical data. It is good practice to back up your data before installing any new version of software. Although MySQL worked very hard to ensure a high level of quality, protect your data by making a backup as you would for any software beta release.
Functionality added or changed:
Incompatible Change:
The C API ER_WARN_DATA_TRUNCATED warning
symbol was renamed to
WARN_DATA_TRUNCATED.
Incompatible Change:
The DECIMAL and
NUMERIC data types now are
handled with a fixed-point library that allows for precision
math handling that results in more accurate results. See
Section 11.13, “Precision Math”.
A consequence of the change in handling of the
DECIMAL and
NUMERIC fixed-point data types is
that the server is more strict to follow standard SQL. For
example, a data type of DECIMAL(3,1) stores a
maximum value of 99.9. Previously, the server allowed larger
numbers to be stored. That is, it stored a value such as 100.0
as 100.0. Now the server clips 100.0 to the maximum allowable
value of 99.9. If you have tables that were created before MySQL
5.0.3 and that contain floating-point data not strictly legal
for the data type, you should alter the data types of those
columns. For example:
ALTER TABLEtbl_nameMODIFYcol_nameDECIMAL(4,1);
For user-defined functions, exact-value decimal arguments such
as 1.3 or
DECIMAL column values were passed
as REAL_RESULT values prior to MySQL 5.0.3.
As of 5.0.3, they are passed as strings with a type of
DECIMAL_RESULT. If you upgrade to 5.0.3 and
find that your UDF now receives string values, use the
initialization function to coerce the arguments to numbers as
described in Section 21.2.2.3, “UDF Argument Processing”.
For the FLOOR() and
CEILING() functions, the return
type is no longer always BIGINT.
For exact-value numeric arguments, the return value has an
exact-value numeric type. For string or floating-point
arguments, the return value has a floating-point type.
Replication: MySQL Cluster:
Added a new global system variable
slave_transaction_retries: If
the replication slave SQL thread fails to execute a transaction
because of an InnoDB deadlock or exceeded
InnoDB's
innodb_lock_wait_timeout or
NDBCLUSTER's
TransactionDeadlockDetectionTimeout or
TransactionInactiveTimeout, it automatically
retries
slave_transaction_retries times
before stopping with an error. The default is 10.
(Bug#8325)
MySQL Cluster:
When using this storage engine, the output of
SHOW TABLE STATUS now displays
properly-calculated values in the
Avg_row_length and
Data_length columns. (Note that
BLOB columns are not yet taken
into account.) In addition, the number of replicas is now shown
in the Comment column (as
number_of_replicas).
Replication:
The LOAD DATA statement was
extended to support user variables in the target column list,
and an optional SET clause. Now one can
perform some transformations on data after they have been read
and before they are inserted into the table. For example:
LOAD DATA INFILE 'file.txt' INTO TABLE t1 (column1, @var1) SET column2 = @var1/100;
Also, replication of LOAD DATA
was changed, so you can't replicate such statements from a 5.0.3
master to pre-5.0.3 slaves.
Replication: The way the character set information is stored into the binary log was changed, so that it is now possible to have a replication master and slave running with different global character sets. A disadvantage is that replication from 5.0.3 masters to pre-5.0.3 slaves is impossible.
Replication:
If the MySQL server is started without an argument to
--log-bin and without
--log-bin-index, thus not
providing a name for the binary log index file, a warning is
issued because MySQL falls back to using the host name for that
name, and this is prone to replication issues if the server's
host name gets changed later. See Section B.1.8.1, “Open Issues in MySQL”.
Nonoptimal index_merge query
execution plans were chosen on IRIX.
(Bug#8578)
mysqld_safe will create the directory where the UNIX socket file is to be located if the directory does not exist. This applies only to the last component of the directory path name. (Bug#8513)
ONLY_FULL_GROUP_BY no longer
is included in the ANSI
composite SQL mode.
(Bug#8510)
The server now includes a timestamp in the Ready for
connections message that is written to the error log
at startup.
(Bug#8444)
CHECKSUM TABLE returns a warning
for nonexisting tables. The checksum value remains
NULL as before.
(Bug#8256)
Setting the connection collation to a value different from the
server collation followed by a CREATE
TABLE statement that included a quoted default value
resulted in a server crash.
(Bug#8235)
When a client releases a user-level lock, DO
RELEASE_LOCK() will not be written to the binary log
anymore (this makes the binary log smaller); as a counterpart,
the slave does not actually take the lock when it executes
GET_LOCK(). This is mainly an
optimization and should not affect existing setups.
(Bug#7998)
InnoDB: Corrected a bug in the crash recovery
of ROW_FORMAT=COMPACT tables that caused
corruption. There may still be bugs in the crash recovery,
especially in COMPACT tables.
(Bug#7973)
Allowed the service-installation command for Windows servers to
specify a single option other than
--defaults-file following the
service name. This is for compatibility with MySQL 4.1.
(Bug#7856)
Changed XML format for mysql from
<
to col_name>col_value</col_name><field
name="
to allow for proper encoding of column names that are not legal
as element names.
(Bug#7811)col_name">col_value</field>
SHOW CREATE TABLE now uses
USING
rather than index_typeTYPE
to specify an
index type.
(Bug#7233)index_type
InnoDB: Implemented fast
TRUNCATE
TABLE. The old approach (deleting rows one by one) may
be used if the table is being referenced by foreign keys.
(Bug#7150)
Out-of-order packets were sent (ERROR after
OK or EOF) following a
KILL QUERY
statement.
(Bug#6804)
Added sql_notes session
variable to cause Note-level warnings not to
be recorded.
(Bug#6662)
Added mysql_library_init() and
mysql_library_end() as synonyms
for the mysql_server_init() and
mysql_server_end() C API
functions. mysql_library_init()
and mysql_library_end() are
#define symbols, but the names more clearly
indicate that they should be called when beginning and ending
use of a MySQL C API library no matter whether the application
uses libmysqlclient or
libmysqld.
(Bug#6149)
Added VAR_POP() and
STDDEV_POP() as standard SQL
aliases for the VARIANCE() and
STDDEV() functions that compute
population variance and standard deviation. Added new
VAR_SAMP() and
STDDEV_SAMP() functions to
compute sample variance and standard deviation.
(Bug#3190)
InnoDB: A commit is now performed after every
10,000 copied rows when executing ALTER
TABLE, CREATE INDEX,
DROP INDEX or
OPTIMIZE TABLE. This makes
recovery from an aborted operations of these types much faster
than previous to this change.
Added support for AVG(DISTINCT).
A new CREATE USER privilege was
added.
Support for RAID options in
MyISAM tables has been removed. If you have
tables that use these options, you should convert them before
upgrading. See Section 2.18.1.2, “Upgrading from MySQL 4.1 to 5.0”.
InnoDB: A shared record lock
(LOCK_REC_NOT_GAP) is now taken for a
matching record in the foreign key check because inserts can be
allowed into gaps.
The MySQL server now aborts when started with the option
--log-bin-index and without
--log-bin, and when started with
--log-slave-updates and without
--log-bin.
API change: the reconnect flag in the
MYSQL structure is now set to 0 by
mysql_real_connect(). Only those
client programs which didn't explicitly set this flag to 0 or 1
after mysql_real_connect()
experience a change. Having automatic reconnection enabled by
default was considered too dangerous (after reconnection, table
locks, temporary tables, user and session variables are lost).
Bit-field values can be written using
b' notation.
value'value is a binary value written using
0s and 1s.
InnoDB: Relaxed locking in
INSERT ...
SELECT, single table UPDATE ...
(SELECT) and single table DELETE ...
(SELECT) clauses when
innodb_locks_unsafe_for_binlog
is used and isolation level of the transaction is not
SERIALIZABLE.
InnoDB uses consistent read in these cases
for a selected table.
InnoDB now supports a fast
TRUNCATE
TABLE. One visible change from this is that
auto-increment values for this table are reset on
TRUNCATE.
InnoDB: Introduced a compact record format
that does not store the number of columns or the lengths of
fixed-size columns. The old format can be requested by
specifying ROW_FORMAT=REDUNDANT. The new
format (ROW_FORMAT=COMPACT) is the default.
The new format typically saves 20% of disk space and memory.
The presence of the new compact row format decreases row storage space by about 20% at the cost of increasing CPU use for some operations. If your workload is a typical one that is limited by cache hit rates and disk speed it is likely to be faster. If it is a rare case that is limited by CPU speed, it might be slower.
From the Windows distribution, predefined accounts without passwords for remote users ("root@%", "@%") were removed (other distributions never had them).
Added the FEDERATED storage engine. See
Section 13.7, “The FEDERATED Storage Engine”.
InnoDB: Setting the initial
AUTO_INCREMENT value for an
InnoDB table using CREATE TABLE ...
AUTO_INCREMENT = now
works, and nALTER TABLE ... AUTO_INCREMENT =
resets the current value.
n
User variable coercibility has been changed from “coercible” to “implicit.” That is, user variables have the same coercibility as column values.
Security improvement: User-defined functions should have at
least one symbol defined in addition to the
xxx symbol that corresponds to the main
xxx() function. These auxiliary symbols
correspond to the xxx_init(),
xxx_deinit(), xxx_reset(),
xxx_clear(), and xxx_add()
functions. mysqld by default no longer loads
UDFs unless they have at least one auxiliary symbol defined in
addition to the main symbol. The
--allow-suspicious-udfs option
controls whether UDFs that have only an xxx
symbol can be loaded. By default, the option is off.
mysqld also checks UDF file names when it
reads them from the mysql.func table and
rejects those that contain directory path name separator
characters. (It already checked names as given in
CREATE FUNCTION statements.) See
Section 21.2.2.1, “UDF Calling Sequences for Simple Functions”,
Section 21.2.2.2, “UDF Calling Sequences for Aggregate Functions”, and
Section 21.2.2.6, “User-Defined Function Security Precautions”. Thanks to Stefano Di Paola
<[email protected]> for finding and
informing us about this issue.
(CVE-2005-0709, CVE-2005-0710)
Added the --large-pages option
for mysqld. Large page support can be used on
Linux systems. See Section 7.5.9, “Enabling Large Page Support”.
Added an error member to the
MYSQL_BIND data structure that is used in the
C API for prepared statements. This member is used for reporting
data truncation errors. Truncation reporting is enabled via the
new MYSQL_REPORT_DATA_TRUNCATION option for
the mysql_options() C API
function.
Added the multi_range_count system variable.
The coercibility for the return value of functions such as
USER() or
VERSION() now is “system
constant” rather than “implicit.” This makes
these functions more coercible than column values so that
comparisons of the two do not result in Illegal mix of
collations errors.
COERCIBILITY() was modified to
accommodate this new coercibility value. See
Section 11.10.3, “Information Functions”.
InnoDB: Upgrading from
4.1: The sorting order for end-space in
TEXT columns for
InnoDB tables has changed. Starting from
5.0.3, InnoDB compares
TEXT columns as space-padded at
the end. If you have a nonunique index on a
TEXT column, you should run
CHECK TABLE on it, and run
OPTIMIZE TABLE if the check
reports errors. If you have a UNIQUE INDEX on
a TEXT column, you should rebuild
the table with OPTIMIZE TABLE.
Boolean full-text phrase searching now requires only that matches contain exactly the same words as the phrase and in the same order. Nonword characters no longer need match exactly.
my.cnf in the compile-time datadir (usually
/usr/local/mysql/data/ in the binary
tarball distributions) is not being read anymore. The value of
the environment variable MYSQL_HOME is used
instead of the hard-coded path.
Additional control over transaction completion was implemented.
The COMMIT and
ROLLBACK
statements support AND [NO] CHAIN and
RELEASE clauses. There is a new
RELEASE
SAVEPOINT statement. The
completion_type system variable
was added for setting the global and session default completion
type.
Security improvement: The server creates
.frm, .MYD,
.MYI, .MRG,
.ISD, and .ISM table
files only if a file with the same name does not already exist.
Thanks to Stefano Di Paola
<[email protected]> for finding and
informing us about this issue.
(CVE-2005-0711)
Added the
engine_condition_pushdown
system variable. For NDB, setting this variable to 1 allows
processing of some WHERE clause conditions to
be processed in NDB nodes before rows are sent to the MySQL
server, rather than having rows sent to the server for
evaluation.
Support for the ISAM storage engine has been
removed. If you have ISAM tables, you should
convert them before upgrading. See
Section 2.18.1.2, “Upgrading from MySQL 4.1 to 5.0”.
Added the CREATE ROUTINE and
ALTER ROUTINE privileges, and
made the EXECUTE privilege
operational.
BIT in column definitions now is
a distinct data type; it no longer is treated as a synonym for
TINYINT(1).
Added cp932 (SJIS for Windows Japanese) and
eucjpms (UJIS for Windows Japanese) character
sets.
MEMORY (HEAP) can have
VARCHAR fields.
SHOW DATABASES,
SHOW TABLES,
SHOW COLUMNS, and so forth,
display information about the
INFORMATION_SCHEMA database. Also, several
SHOW statements now accept a
WHERE clause specifying which output rows to
display. See Chapter 19, INFORMATION_SCHEMA Tables.
SHOW COLUMNS now displays
NO rather than blank in the
Null output column if the corresponding table
column cannot be NULL.
When the MyISAM storage engine detects
corruption of a MyISAM table, a message
describing the problem now is written to the error log.
A VARCHAR column can now contain
up to 65535 bytes. In addition,
VARCHAR columns now remember
trailing spaces. For more details, see
Section C.1, “Changes in Release 5.0.x (Production)”.
Added --innodb-checksums and
--innodb-doublewrite options for
mysqld.
Added several InnoDB status variables. See
Section 5.1.6, “Server Status Variables”.
Added account-specific MAX_USER_CONNECTIONS
limit, which allows you to specify the maximum number of
concurrent connections for the account. Also, all limited
resources now are counted per account (instead of being counted
per user + host pair as it was before). Use the
--old-style-user-limits option to
get the old behavior.
Implemented support for XA transactions. See
Section 12.4.7, “XA Transactions”. The implementation makes the
--innodb_safe_binlog option obsolete, so it has
been removed.
mysqlbinlog now prints a
ROLLBACK
statement at the end of its output, in case the server crashed
while it was in the process of writing the final entry into the
last binary log named on the command line. This causes any
half-written transaction to be rolled back when the output is
executed. The
ROLLBACK is
harmless if the binary log file was written and closed normally.
Seconds_Behind_Master is
NULL (which means “unknown”) if
the slave SQL thread is not running, or if the slave I/O thread
is not running or not connected to master. It is zero if the SQL
thread has caught up to the I/O thread. It no longer grows
indefinitely if the master is idle.
FLUSH TABLES WITH READ
LOCK is now killable while it is waiting for running
COMMIT statements to finish.
The MySQL server aborts immediately instead of simply issuing a
warning if it is started with the
--log-bin option but cannot
initialize the binary log at startup (that is, an error occurs
when writing to the binary log file or binary log index file).
The binary log file and binary log index file now are handled
the same way as MyISAM tables when there is a
“disk full” or “quota exceeded” error.
See Section B.1.4.3, “How MySQL Handles a Full Disk”.
InnoDB: When MySQL/InnoDB is compiled on Mac
OS X 10.2 or earlier, detect the operating system version at run
time and use the fcntl() file flush method
on Mac OS X versions 10.3 and later. In Mac OS X,
fsync() does not flush the write cache in
the disk drive, but the special fcntl()
does; however, the flush request is ignored by some external
devices. Failure to flush the buffers may cause severe database
corruption at power outages.
Bugs fixed:
Replication: If multiple semicolon-separated statements were received in a single packet, they were written to the binary log as a single event rather than as separate per-statement events. For a server serving as a replication master, this caused replication to fail when the event was sent to slave servers. (Bug#8436)
Replication:
A replication master stamped a generated statement (such as a
SET statement) with an error code intended
only for another statement. This could happen, for example, when
a statement generated a duplicate key error on the master but
still had be to replicated to the slave.
(Bug#8412)
Replication:
If the slave was running with
--replicate-*-table options which excluded one
temporary table and included another, and the two tables were
used in a single DROP TEMPORARY TABLE IF
EXISTS statement, as the ones the master automatically
writes to its binary log upon client's disconnection when client
has not explicitly dropped these, the slave could forget to
delete the included replicated temporary table. Only the slave
needs to be upgraded.
(Bug#8055)
Replication:
Multiple-table updates did not replicate properly to slave
servers where --replicate-*-table options had
been specified.
(Bug#7011)
Replication:
A replication slave could crash after replicating many
ANALYZE TABLE,
OPTIMIZE TABLE, or
REPAIR TABLE statements from the
master.
(Bug#6461, Bug#7658)
Replication:
Changed semantics of CREATE/ALTER/DROP
DATABASE statements so that replication of
CREATE DATABASE is possible when
using --binlog-do-db and
--binlog-ignore-db.
(Bug#6391)
Replication: DDL statements for views were not being written to the binary log (and thus not subject to replication). (Bug#4838)
mysqldump misinterpreted “
_ ” and “ %
” characters in the names of tables to be dumped as
wildcard characters.
(Bug#9123)
In strict or traditional SQL mode, too-long string values
assigned to string columns (CHAR,
VARCHAR,
BINARY,
VARBINARY,
TEXT, or
BLOB) were correctly truncated,
but the server returned an SQLSTATE value of
01000 (should be 22001).
(Bug#9029, Bug#6999)
The definition of the enumeration-valued
sql_mode column of the
mysql.proc table was missing some of the
current allowable SQL modes, so stored routines would not
necessarily execute with the SQL mode in effect at the time of
routine definition.
(Bug#8902)
TRUNCATE did not work within
stored procedures. Now, within stored procedures,
TRUNCATE is executed in the same
way as DELETE. This change was
necessary because TRUNCATE
implicitly locks tables.
(Bug#8850)
A rare race condition could cause
FLUSH TABLES WITH READ
LOCK to hang.
(Bug#8682)
AES_DECRYPT(
could fail to return col_name,key)NULL for invalid values
in col_name, if
col_name was declared as NOT
NULL.
(Bug#8669)
If SELECT
DISTINCT named an index column multiple times in the
select list, the server tried to access different key fields for
each instance of the column, which could result in a crash.
(Bug#8532)
MATCH ... AGAINST in natural language mode
could cause a server crash if the FULLTEXT
index was not used in a join (that is,
EXPLAIN did not show
fulltext join mode) and the
search query matched no rows in the table.
(Bug#8522)
REPAIR TABLE did not invalidate
query results in the query cache that were generated from the
table.
(Bug#8480)
LOAD INDEX statement now loads
the index into memory.
(Bug#8452)
For a stored function that refers to a given table, invoking the function while selecting from the same table resulted in a server crash. (Bug#8405)
Comparison of a DECIMAL column
containing NULL to a subquery that produced
DECIMAL values resulted in a
server crash.
(Bug#8397)
DELETE FROM when the tbl_name ...
WHERE ... ORDER BY
tbl_name.col_name
ORDER BY column was
qualified with the table name caused the server to crash.
(Bug#8392)
Stored functions that used cursors could return incorrect results. (Bug#8386)
The Cyrillic letters I
(И) and SHORT I
(Й) were treated as being the same
character by the utf8_general_ci collation.
(Bug#8385)
When performing boolean full-text searches on
utf8 columns, a double-quote character in the
search string caused the server to crash.
(Bug#8351)
The --set-character-set option for
myisamchk was changed to
--set-collation. The value
needed for specifying how to sort indexes is a collation name,
not a character set name.
(Bug#8349)
Corruption of MyISAM table indexes could
occur with TRUNCATE
TABLE if the table had already been opened. For
example, this was possible if the table had been opened
implicitly by selecting from a MERGE table
that mapped to the MyISAM table. The server
now issues an error message for
TRUNCATE
TABLE under these conditions.
(Bug#8306)
For a query with both GROUP BY and
COUNT(DISTINCT) clauses and a
FROM clause with a subquery,
NULL was returned for any
VARCHAR column selected by the
subquery.
(Bug#8218)
Selecting from an INFORMATION_SCHEMA table
combined with a subquery on an
INFORMATION_SCHEMA table caused an error with
the message Table .
(Bug#8164)tbl_name
is corrupted
Matching of table names by mysqlhotcopy now
accommodates DBD::mysql versions 2.9003 and
up, which implement identifier quoting.
(Bug#8136)
Re-execution of prepared statements containing subqueries caused the server to crash. (Bug#8125)
A problem with equality propagation optimization for prepared statements and stored procedures caused a server crash upon re-execution of the prepared statement or stored procedure. (Bug#8115, Bug#8849)
Selecting from a view defined as a join caused a server crash if the query cache was enabled. (Bug#8054)
Results in the query cache generated from a view were not
properly invalidated after ALTER
VIEW or DROP VIEW on
that view.
(Bug#8050)
Creating a table using a name containing a character that is
illegal in character_set_client
resulted in the character being stripped from the name and no
error. The character now is considered an error.
(Bug#8041)
Certain correlated subqueries with forward references (referring to an alias defined later in the outer query) could crash the server. (Bug#8025)
Corrected a problem with references to DUAL
where statements such as SELECT 1 AS a FROM
DUAL would succeed but statements such as
SELECT 1 AS a FROM DUAL LIMIT 1 would fail.
(Bug#8023)
Comparing a nested row expression (such as
ROW(1,(2,3))) with a subquery caused the
server to crash.
(Bug#8022)
The number of columns in a row comparison against a subquery was calculated incorrectly. (Bug#8020)
mysqldump now avoids writing SET
NAMES to the dump output if the server is older than
version 4.1 and would not understand that statement.
(Bug#7997)
A deadlock could occur on an update followed by a
SELECT on an
InnoDB table without any explicit locks being
taken. InnoDB now takes an exclusive lock
when INSERT ON DUPLICATE KEY UPDATE is
checking duplicate keys.
(Bug#7975)
A slave running MySQL 3.23.51 or newer hung while trying to
connect to a master running MySQL 3.23.50 or older. (The reason
for this was a bug in the old masters — SELECT
@@ caused the
server to hang — which was fixed in MySQL 3.23.50.)
(Bug#7965)unknown_var
Erroneous output resulted from
SELECT
DISTINCT combined with a subquery and GROUP
BY.
(Bug#7946)
FOUND_ROWS() returned an
incorrect value after a SELECT SQL_CALC_FOUND_ROWS
DISTINCT statement that selected constants and
included GROUP BY and
LIMIT clauses.
(Bug#7945)
mysqld_safe now understands the
--help option. Previously, it ignored the
option and attempted to start the server anyway.
(Bug#7931)
Creating a user with grants failed when specifying a password but worked without one. (Bug#7905)
Comparing the result of a subquery to a nonexistent column caused the server to crash. This issue affected MySQL on Windows platforms only. (Bug#7885)
ALTER TABLE improperly accepted
an index on a TIMESTAMP column
that CREATE TABLE would reject.
(Bug#7884)
MySQL allowed concurrent updates (including inserts and deletes) to a table if binary logging was enabled. Now, all updates are executed in a serialized fashion, because they are executed serialized when the binary log is replayed. (Bug#7879)
Ensured that mysqldump --single-transaction
sets its transaction isolation level to
REPEATABLE READ before
proceeding (otherwise if the MySQL server was configured to run
with a default isolation level lower than
REPEATABLE READ it could give
an inconsistent dump).
(Bug#7850)
mysqlbinlog forgot to add backquotes around
the collation of user variables (causing later parsing problems
as BINARY is a reserved word).
(Bug#7793)
A Table is full error occurred when the
table was still smaller than
max_heap_table_size.
(Bug#7791)
Use of GROUP_CONCAT() with
HAVING caused the server to crash.
(Bug#7769)
The CONV() function returned an
unsigned BIGINT number, which
does not fit in 32 bits.
(Bug#7751)
The IN() operator did not return
correct results if all values in the list were constants and
some of them used substring functions such as
LEFT(),
RIGHT(), or
MID().
(Bug#7716)
When encountering a disk full or
quota exceeded write error,
MyISAM sometimes failed to sleep and retry
the write, resulting in a corrupted table.
(Bug#7714)
The CONVERT_TZ() function, when
its second or third argument was from a
const table, caused the
server to crash. (See Section 12.3.2, “EXPLAIN Syntax”.)
(Bug#7705)
The output of the STATUS
(\s) command in mysql had
the values for the server and client character sets reversed.
(Bug#7571)
A LEFT OUTER JOIN between an empty base table
and a view on an empty base table caused a server crash.
(Bug#7433)
Ordering by an unsigned expression (more complex than a column
reference) was treating the value as signed, producing
incorrectly sorted results. HAVING was also
treating unsigned columns as signed.
(Bug#7425)
The server crashed when an error occurred during the filling of a temporary table created for handling a view or derived table. (Bug#7413)
Made the MySQL server accept executing SHOW
CREATE DATABASE even if the connection has an open
transaction or locked tables. Refusing it made
mysqldump --single-transaction sometimes fail
to print a complete CREATE
DATABASE statement for some dumped databases.
(Bug#7358)
Corrected the handling of trailing spaces in the
ucs2 character set.
(Bug#7350)
--expire-logs-days was not
honored if using only transactions.
(Bug#7236)
Some INFORMATION_SCHEMA columns that
contained timestamp values were of type
VARBINARY. These were changed to
TIMESTAMP.
(Bug#7217)
Some INFORMATION_SCHEMA columns that
contained catalog identifiers were of type
LONGTEXT. These were changed to
VARCHAR(, where
NN is the appropriate maximum
identifier length.
(Bug#7215)
Use of GROUP_CONCAT() in the
select list when selecting from a view caused a server crash.
(Bug#7116)
An expression that tested a case-insensitive character column
against string constants that differed in lettercase could fail
because the constants were treated as having a binary collation.
(For example, WHERE city='London' AND
city='london' could fail.)
(Bug#7098, Bug#8690)
Setting the initial AUTO_INCREMENT value for
an InnoDB table using CREATE TABLE
... AUTO_INCREMENT = did
not work, and nALTER TABLE ... AUTO_INCREMENT =
did not reset the current
value.
(Bug#7061)n
When setting integer system variables to a negative value with
SET VARIABLES, the value was treated as a
positive value modulo 232.
(Bug#6958)
Use of a view in a correlated subquery that contains
HAVING but no GROUP BY
caused a server crash.
(Bug#6894)
Praparing a query using the
CONVERT_TZ() function with
constant arguments caused the server to crash.
(Bug#6849)
Handling by mysql_list_fields()
of references to stored functions within views was incorrect and
could result in a server crash.
(Bug#6814)
A sequence of
BEGIN (or
SET autocommit = 0),
FLUSH TABLES WITH READ
LOCK, transactional update,
COMMIT,
FLUSH TABLES WITH READ
LOCK could hang the connection forever and possibly
the MySQL server itself. This happened for example when running
the innobackup script several times.
(Bug#6732)
Prevent adding CREATE TABLE .. SELECT query
to the binary log when the insertion of new records partially
failed.
(Bug#6682)
mysqlbinlog did not print SET
PSEUDO_THREAD_ID statements in front of
LOAD DATA
INFILE statements inserting into temporary tables,
thus causing potential problems when rolling forward these
statements after restoring a backup.
(Bug#6671)
If a MyISAM table on Windows had
INDEX DIRECTORY or DATA
DIRECTORY table options, mysqldump
dumped the directory path names with single-backslash path name
separators. This would cause syntax errors when importing the
dump file. mysqldump now changes “
\ ” to “ /
” in the path names on Windows.
(Bug#6660)
SHOW CREATE TABLE now reports
ENGINE=MEMORY rather than
ENGINE=HEAP for a MEMORY
table (unless the MYSQL323 SQL
mode is enabled).
(Bug#6659)
Incorrectly ordered results were returned from a query using a
FULLTEXT index to retrieve rows and there was
another index that was usable for ORDER BY.
For such a query, EXPLAIN showed
the fulltext join type, but
showed the other (not FULLTEXT) index in the
Key column.
(Bug#6635)
CREATE TABLE ... LIKE failed on Windows when
the source or destination table was located in a symlinked
database directory.
(Bug#6607)
Retrieving from a view defined as a
SELECT that mixed
UNION ALL and
UNION DISTINCT
resulted in a different result than retrieving from the original
SELECT.
(Bug#6565)
Selecting from a view that had an EXISTS or
NOT EXISTS subquery did not always work
properly, and selecting columns by name could cause a server
crash. With SELECT *, crashes did not occur,
but columns in the outer query were not resolved properly.
(Bug#6394)
Fixed a problem in
NO_BACKSLASH_ESCAPES SQL mode
for strings that contained both the string quoting character and
backslash.
(Bug#6368)
The CHAR() function was not
ignoring NULL arguments, contrary to the
documentation.
(Bug#6317)
Starting and stopping the slave thread (only) could in some circumstance cause the server to crash. (Bug#6148)
InnoDB: Honor the
--tmpdir startup option when
creating temporary files. Previously, InnoDB
temporary files were always created in the temporary directory
of the operating system. On Netware, InnoDB
will continue to ignore --tmpdir.
(Bug#5822)
A HAVING clause that referred to
RAND() or a user-defined function
in the SELECT part of a query
through an alias could cause MySQL to crash or to return an
incorrect value.
(Bug#5185)
Platform and architecture information in version information
produced for --version option on Windows was
always Win95/Win98 (i32). More accurately
determine platform as Win32 or
Win64 for 32-bit or 64-bit Windows, and
architecture as ia32 for x86,
ia64 for Itanium, and axp
for Alpha.
(Bug#4445)
When using the RPAD() function
(or any function adding spaces to the right) in a query that had
to be resolved by using a temporary table, all resulting strings
had rightmost spaces removed (that is,
RPAD() did not work)
(Bug#4048)
Host name matching didn't work if a netmask was specified for table-specific privileges. (Bug#3309)
mysql_fix_privilege_tables now makes it
possible for mysql privilege tables created
in MySQL 5.0 to be used with MySQL 4.1. This makes it possible
to downgrade from 5.0 to 4.1, or to run MySQL 4.1 and 5.0 using
the same privilege table files for testing purposes.
Giving mysqld a SIGHUP
caused it to crash.
Prepared statements using
SUM(DISTINCT...) did not perform
correctly.
InnoDB: Use native
tmpfile() function on Netware. All
InnoDB temporary files are created under
sys:\tmp. Previously,
InnoDB temporary files were never deleted on
Netware.
A symlink vulnerability in the mysqlaccess script was reported by Javier Fernandez-Sanguino Pena and Debian Security Audit Team. (CVE-2005-0004)
A number of portability issues relating to overflow in floating point values were corrected.
Prepared statements now gives warnings on prepare.
The combination of -not and
trunc* operators in a full-text search did
not work correctly. Using more than one truncated negative
search term caused the result to be empty.
Prepared statements did not work correctly with OUTER
JOIN.
Functionality added or changed:
Incompatible Change:
The precedence of NOT operator has changed so
that expressions such as NOT a BETWEEN b AND
c are parsed correctly as NOT (a BETWEEN b
AND c) rather than as (NOT a) BETWEEN b AND
c. The pre-5.0 higher-precedence behavior can be
obtained by enabling the new
HIGH_NOT_PRECEDENCE SQL mode.
Incompatible Change:
SHOW STATUS now shows the session
(thread-specific) status variables and
SHOW GLOBAL
STATUS shows the status variables for the whole
server.
Before MySQL 5.0.2, SHOW STATUS
returned global status values. Because the default as of 5.0.2
is to return session values, this is incompatible with previous
versions. To issue a SHOW STATUS
statement that will retrieve global status values for all
versions of MySQL, write it like this:
SHOW /*!50002 GLOBAL */ STATUS;
Replication:
mysqldump --single-transaction
--master-data is now able to take an online
(nonblocking) dump of InnoDB and report the
corresponding binary log coordinates, which makes a backup
suitable for point-in-time recovery, roll-forward or replication
slave creation. See Section 4.5.4, “mysqldump — A Database Backup Program”.
Replication:
Two new system variables were introduced.
auto_increment_increment and
auto_increment_offset can be
set locally or globally, and are intended for use in controlling
the behavior of AUTO_INCREMENT columns in
master-to-master replication. Note that these variables are not
intended to take the place of sequences. See
Section 5.1.3, “Server System Variables”.
If a write to a MyISAM table fails because of
a full disk or an exceeded disk quota, it now prints a message
to the error log every 10 minutes, and waits until disk space
becomes available.
(Bug#3248)
Made the MySQL server ignore SIGHUP and
SIGQUIT on Mac OS X 10.3. This is needed
because under this OS, the MySQL server receives lots of these
signals.
(Bug#2030)
If the server finds that the user table has
not been upgraded to include the view-related privilege columns,
it treats each account as having view privileges that are the
same as its CREATE privilege.
A connection doing a rollback now displays "Rolling back" in the
State column of SHOW
PROCESSLIST.
Renamed the sql_updatable_view_key system
variable to
updatable_views_with_limit.
This variable now can have only two values:
1 or YES: Don't issue
an error message (warning only) if a VIEW without presence
of a key in the underlying table is used in queries with a
LIMIT clause for updating. (This is the
default value.)
0 or NO: Prohibit
update of a VIEW, which does not contain a key in the
underlying table and the query uses a
LIMIT clause (usually get from GUI
tools).
Reverted output format of SHOW
TABLES to old pre-5.0.1 format that did not include a
table type column. To get the additional column that lists the
table type, use SHOW FULL TABLES now.
CHECK TABLE now works for views.
Modify DROP USER so that it drops
the account, including all its privileges. Formerly, it removed
the account record only for an account that had all privileges
revoked.
0 or NO: Prohibit update
of a VIEW, which does not contain a key in the underlying table
and the query uses a LIMIT clause (usually
get from GUI tools).
We now detect too-large floating point numbers during statement parsing and generate an error messages for them.
New auto_increment_increment
and auto_increment_offset
system variables. These enable you to set up a server to
generate auto-increment values that don't conflict with another
server.
Added the CREATE USER and
RENAME USER statements.
MySQL now by default checks dates and in strict mode allows only
fully correct dates. If you want MySQL to behave as before, you
should enable the new
ALLOW_INVALID_DATES SQL mode.
Added NO_AUTO_CREATE_USER SQL
mode to prevent GRANT from
automatically creating new users if it would otherwise do so,
unless a password also is specified.
MySQL now remembers which columns were declared to have default
values. In
STRICT_TRANS_TABLES/STRICT_ALL_TABLES
mode, you now get an error if you do an
INSERT without specifying all
columns that don't have a default value. A side effect of this
is that when you do SHOW CREATE for a new
table, you no longer see a DEFAULT value for
a column for which you didn't specify a default value.
InnoDB: If you specify the
innodb_locks_unsafe_for_binlog
option in my.cnf, for an
UPDATE or a
DELETE, InnoDB
locks only the rows that it updates or deletes. This greatly
reduces the probability of deadlocks.
A HAVING clause in a
SELECT statement now can refer to
columns in the GROUP BY clause, as required
by standard SQL.
The SCHEMA and SCHEMAS
keywords are now accepted as synonyms for
DATABASE and DATABASES.
Added several InnoDB status variables. See
Section 5.1.6, “Server Status Variables”.
Added STRICT_TRANS_TABLES,
STRICT_ALL_TABLES,
NO_ZERO_IN_DATE,
NO_ZERO_DATE,
ERROR_FOR_DIVISION_BY_ZERO,
and TRADITIONAL SQL modes. The
TRADITIONAL mode is shorthand
for all the preceding modes. When using mode
TRADITIONAL, MySQL generates
an error if you try to insert a wrong value in a column. It does
not adjust the value to the closest possible legal value.
The mysql_fix_privilege_tables script now
initializes the global CREATE
VIEW and SHOW VIEW
privileges in the user table to the value of
the CREATE privilege in that
table.
1 or YES: Don't issue an
error message (warning only) if a VIEW without presence of a key
in the underlying table is used in queries with a
LIMIT clause for updating. (This is the
default value.)
The compilation flag DONT_USE_DEFAULT_FIELDS
was removed because you can get the same behavior by setting the
sql_mode system variable to
STRICT_TRANS_TABLES.
mysqlbinlog now prints an informative
commented line (thread id, timestamp, server id, and so forth)
before each LOAD DATA
INFILE, like it does for other queries; unless
--short-form is used.
Added IS [NOT] syntax, where
boolean_value
boolean_value is
TRUE, FALSE, or
UNKNOWN.
Added --start-datetime,
--stop-datetime,
--start-position, and
--stop-position options to
mysqlbinlog. These make point-in-time
recovery easier.
Added initial support for rudimentary triggers (the
CREATE TRIGGER and
DROP TRIGGER statements).
Added basic support for read-only server side cursors.
Implemented the WITH CHECK OPTION clause for
CREATE VIEW.
Added support for the INFORMATION_SCHEMA
“information database” that provides database
metadata. See Chapter 19, INFORMATION_SCHEMA Tables.
Bugs fixed:
Replication:
A problem introduced in MySQL 4.0.21 caused replication slaves
to stop (error 1223) where a connection started a transaction,
performed updates, then issued a
FLUSH TABLES WITH READ
LOCK followed by a
COMMIT. This issue occurred when
using the InnoDB
innobackup script.
(Bug#5949)
Replication:
SET COLLATION_SERVER... statements replicated
by the slave SQL thread no longer advance its position. This is
so that, if the thread is interrupted before the update is
completed, it later performs the SET again.
(Bug#5705)
Replication:
OPTIMIZE TABLE,
REPAIR TABLE, and
ANALYZE TABLE are now replicated
without any error code in the binary log.
(Bug#5551)
Replication:
A CREATE TABLE ... TYPE=HEAP ... AS SELECT...
statement caused the replication slave to stop.
(Bug#4971)
mysqlbinlog was unable to read from
stdin, for example, when piping the output
from zcat to mysqlbinlog.
(Bug#7853)
If a connection was interrupted by a network error and did a
rollback, the network error code got stored into the
BEGIN and
ROLLBACK
binary log events; that caused superfluous slave stops.
(Bug#6522)
If a connection had an open transaction but had done no updates
to transactional tables (for example if had just done a
SELECT FOR UPDATE then executed a
nontransactional update, that update automatically committed the
transaction (thus releasing InnoDB's
row-level locks etc).
(Bug#5714)
If the slave SQL thread finds a syntax error in a query (which should be rare, as the master parsed it successfully), it now stops immediately. (Bug#5711)
disable-local-infile option had no effect if
the client read it from a configuration file using
mysql_options(...,MYSQL_READ_DEFAULT,...).
(Bug#5073)
SET GLOBAL SYNC_BINLOG did not work on some
platforms (Mac OS X).
(Bug#5064)
mysql-test-run failed the
rpl_trunc_binlog test when running the test
from the installation directory.
(Bug#5050)
mysql_options(...,MYSQL_OPT_LOCAL_INFILE,...)
failed to disable
LOAD DATA LOCAL
INFILE.
(Bug#5038)
The counter for an AUTO_INCREMENT column was
not reset by TRUNCATE
TABLE if the table was a temporary one.
(Bug#5033)
FLUSH TABLES WITH READ
LOCK now blocks COMMIT
statements if the server is running with binary logging enabled;
this ensures that the binary log position is trustable when
doing a full backup of tables and the binary log.
(Bug#4953)
KILLing a connection while it was
performing START SLAVE caused the
server to crash.
(Bug#4827)
A deadlock could happen under certain rare circumstances when
using KILL.
(Bug#4810)
mysql-test-run failed the
grant_cache test when run as Unix root user.
(Bug#4678)
mysqlbinlog --read-from-remote-server sometimes could not accept 2 binary logs in a single invocation. (Bug#4507)
mysqlbinlog --position
--read-from-remote-server had incorrect output for
# at .
(Bug#4506)log_pos
This build passes our test suite and fixes a lot of reported bugs found in the previous 5.0.0 release. However, please be aware that this is not a “standard MySQL build” in the sense that there are still some open critical bugs in our bugs database at http://bugs.mysql.com/ that affect this release as well. We are actively fixing these and will make a new release where these are fixed as soon as possible. However, this binary should be a good candidate for testing new MySQL 5.0 features for future products.
Functionality added or changed:
Incompatible Change:
C API change: mysql_shutdown()
now requires a second argument. This is a source-level
incompatibility that affects how you compile client programs; it
does not affect the ability of compiled clients to communicate
with older servers. See Section 20.9.3.65, “mysql_shutdown()”.
Replication:
For replication of MEMORY
(HEAP) tables: Made the master automatically
write a DELETE FROM statement to its binary
log when a MEMORY table is opened for the
first time since the master's startup. This is for the case
where the slave has replicated a nonempty
MEMORY table, and then the master is shut
down and restarted: the table is now empty on the master; the
DELETE FROM empties it on the slave as well.
Even with this fix, between the master's restart and the first
use of the table on master, the slave still has out-of-date data
in the table. However, if you use the
--init-file option to populate
the MEMORY table on the master at startup, it
ensures that the failing time interval is zero.
(Bug#2477)
Replication:
Added the
--replicate-same-server-id server
option.
Replication:
DROP DATABASE IF EXISTS, DROP TABLE
IF EXISTS, single-table
DELETE, and single-table
UPDATE now are written to the
binary log even if they changed nothing on the master (for
example, even if a DELETE matched
no rows). The old behavior sometimes caused bad surprises in
replication setups.
Replication: Replication and mysqlbinlog now have better support for the case that the session character set and collation variables are changed within a given session. See Section 16.3.1, “Replication Features and Issues”.
The Type column name and values in the output
from SHOW TABLES are now shown
according to standard. The column name has changed from
Type to
table_type; permitted values
are BASE TABLE, VIEW, and
ERROR.
(Bug#4603)
Added Last_query_cost status
variable that reports optimizer cost for last compiled query.
Changed that when the MySQL server has binary logging disabled
(that is, no --log-bin option was
used), then no transaction binary log cache is allocated for
connections. This should save
binlog_cache_size bytes of
memory (32KB by default) for every connection.
Implemented a new “greedy search” optimizer that
can significantly reduce the time spent on query optimization
for some many-table joins. (You are affected if not only some
particular SELECT is slow, but
even using EXPLAIN for it takes a
noticeable amount of time.) Two new system variables,
optimizer_search_depth and
optimizer_prune_level, can be
used to fine-tune optimizer behavior.
OPTIMIZE TABLE for
InnoDB tables is now mapped to
ALTER TABLE instead of
ANALYZE TABLE. This rebuilds the
table, which updates index statistics and frees space in the
clustered index.
When a session having open temporary tables terminates, the
statement automatically written to the binary log is now
DROP TEMPORARY TABLE IF EXISTS instead of
DROP TEMPORARY TABLE, for more robustness.
Added support for read-only and updatable views based on a single table or other updatable views. View use requires that you upgrade your grant tables to add the view-related privileges. See Section 4.4.5, “mysql_fix_privilege_tables — Upgrade MySQL System Tables”.
Added the sql_updatable_view_key system
variable.
The MySQL server now returns an error if SET
sql_log_bin is issued by a user without the
SUPER privilege (in previous
versions it just silently ignored the statement in this case).
When a database is dropped, all routines belonging to that database are also dropped.
Added the --to-last-log option to
mysqlbinlog, for use in conjunction with
--read-from-remote-server.
sync_frm is now a settable
global variable (not only a startup option).
Added --innodb_safe_binlog server
option, which adds consistency guarantees between the content of
InnoDB tables and the binary log. See
Section 5.2.3, “The Binary Log”.
Explicit USE statements no longer are allowed in a stored
procedure.
db_name
Added the sync_binlog=N global variable and
startup option, which makes the MySQL server synchronize its
binary log to disk (fdatasync()) after every
Nth write to the binary log.
Killing a CHECK TABLE statement
does not result in the table being marked as
“corrupted” any more; the table remains as if
CHECK TABLE had not even started.
See Section 12.5.6.3, “KILL Syntax”.
Changed the slave SQL thread to print less useless error
messages (no more message duplication; no more messages when an
error is skipped because of
slave-skip-errors).
When executed from another database, an implicit USE
is in effect.
db_name
When installing a MySQL server as a Windows service, the
installation command can include a
--local-service option following the service
name to cause the server to run using the
LocalService Windows account that has limited
privileges. This is in addition to the
--defaults-file option that also
can be given following the service name.
Procedure names may be qualified, for example,
db.p()
A stored procedure is no longer “global.” That is, it now belongs to a specific database:
When a database is dropped, all routines belonging to that database are also dropped.
Procedure names may be qualified, for example,
db.p()
When executed from another database, an implicit
USE is
in effect.
db_name
Explicit USE statements no longer are allowed in a stored
procedure.
db_name
See Section 18.2, “Using Stored Routines (Procedures and Functions)”.
Bugs fixed:
Replication:
When a multiple-table DROP TABLE
failed to drop a table on the master server, the error was not
written to the binary log.
(Bug#4553)
Replication:
When the slave SQL thread was replicating a
LOAD DATA
INFILE statement, it didn't show the statement in the
output of SHOW PROCESSLIST.
(Bug#4326)
Replication:
Complex expressions using AND,
OR, or both could result in a crash if the
query containing the expression query was ignored, either by a
replication server due to --replicate-*-table
rules, or by any MySQL server due to a syntax error.
(Bug#3969, Bug#4494)
Replication:
The slave SQL thread refused to replicate
INSERT ...
SELECT if it examined more than 4 billion rows.
(Bug#3871)
Replication:
If server-id was not set using startup
options but with SET
GLOBAL, the replication slave still complained that it
was not set.
(Bug#3829)
Replication:
A MySQL slave server built using
--with-debug, and replicating
itself, crashed.
(Bug#3568)
Replication:
Multiple-table DELETE statements
were always replicated by the slave if there were some
--replicate-*-ignore-table options and no
--replicate-*-do-table options.
(Bug#3461)
Replication:
Memory could be corrupted by replicating a
LOAD DATA
INFILE from a MySQL 3.23 master. Some less critical
issues remain; see Section 16.3.1, “Replication Features and Issues”.
(Bug#3422)
Replication: In some replication error messages, a very long query caused the rest of the message to be invisible (truncated), by putting the query last in the message. (Bug#3357)
Replication:
Changed that when a thread handling INSERT
DELAYED (also known as a
delayed_insert thread) is killed, its
statements are recorded with an error code of value zero
(killing such a thread does not endanger replication, so we thus
avoid a superfluous error on the slave).
(Bug#3081)
Replication:
Corrected the master's binary log position that
InnoDB reports when it is doing a crash
recovery on a slave server.
(Bug#3015)
Replication:
--replicate-wild-*-table rules now apply to
ALTER DATABASE when the table
pattern is %, as is the case for
CREATE DATABASE and
DROP DATABASE.
(Bug#3000)
Replication:
Statements did not raise errors on the slave, if the slave was
excluded given the --replicate-* options in use
at the time. The effect of this problem was: when a statement
was killed on the master, the slave stopped.
(Bug#2983)
Replication:
Multiple-table DELETE statements
were never replicated by the slave if there were any
--replicate-*-table options.
(Bug#2527)
Replication:
Replication: If a client connects to a slave server and issues
an administrative statement for a table (for example,
OPTIMIZE TABLE or
REPAIR TABLE), this could
sometimes stop the slave SQL thread. This does not lead to any
corruption, but you must use START
SLAVE to get replication going again.
(Bug#1858)
If CREATE TEMPORARY TABLE t SELECT failed
while loading the data, the temporary table was not dropped.
(Bug#4551)
mysql_fix_privilege_tables did not handle the
--password=
option correctly.
(Bug#4240, Bug#4543)password_val
Made DROP DATABASE honor the
value of
lower_case_table_names.
(Bug#4066)
During the installation process of the server RPM on Linux, if
mysqld was run as the root
system user and with --log-bin
pointing to a directory outside of
/var/lib/mysql, it created binary log files
owned by root in this directory, which
remained owned by root after the
installation. Now mysqld is started as the
mysql system user instead.
(Bug#4038)
A potential memory overrun could occur in
mysql_real_connect() (which
required a compromised DNS server and certain operating
systems).
(Bug#4017)
mysqlbinlog didn't escape the string content of user variables, and did not deal well when these variables were in non-ASCII character sets; this is now fixed by always printing the string content of user variables in hexadecimal. The character set and collation of the string is now also printed. (Bug#3875)
mysqlbinlog failed to print a
USE statement under those rare
circumstances where the binary log contained a
LOAD DATA
INFILE statement.
(Bug#3415)
A rare error condition caused the slave SQL thread spuriously to print the message Binlog has bad magic number and stop when it was not necessary to do so. (Bug#3401)
mysqlbinlog --read-from-remote-server now
print the exact positions of events in lines beginning with
at # in the log.
(Bug#3214)
mysqlbinlog --read-from-remote-server read
all binary logs following the one that was requested. It now
stops at the end of the requested file, the same as it does when
reading a local binary log. There is an option
--to-last-log to get the old behavior.
(Bug#3204)
Strange results with index (x, y) ... WHERE
x=
(Bug#3155)val_1 AND
y>=val_2 ORDER BY
pk;
Adding ORDER BY to a query that uses a
subquery can cause incorrect results.
(Bug#3118)
Changed that when a DROP TEMPORARY TABLE
statement is automatically written to the binary log when a
session ends, the statement is recorded with an error code of
value zero (this ensures that killing a
SELECT on the master does not
result in a superfluous error on the slave).
(Bug#3063)
When a Rotate event was found by the slave
SQL thread in the middle of a transaction, the value of
Relay_Log_Pos in SHOW
SLAVE STATUS was incorrectly altered.
(Bug#3017)
Running LOAD DATA FROM MASTER after
RESET SLAVE caused a segmentation
fault.
(Bug#2922)
A deadlock ocurred when two START
SLAVE commands were run at the same time.
(Bug#2921)
Changed the column Seconds_Behind_Master in
SHOW SLAVE STATUS to never show a
value of -1.
(Bug#2826)
Made clearer the error message that one gets when an update is
refused because of the
--read-only option.
(Bug#2757)
The MySQL server did not report any error if a statement
(submitted through
mysql_real_query() or
mysql_stmt_prepare()) was
terminated by garbage characters. This can happen if you pass a
wrong length parameter to these functions.
The result was that the garbage characters were written into the
binary log.
(Bug#2703)
SLAVE START (which is a deprecated syntax,
START SLAVE should be used
instead) could crash the slave.
(Bug#2516)
ALTER DATABASE caused the client
to hang if the database did not exist.
(Bug#2333)
The --local-load option of
mysqlbinlog now requires an argument.
Functionality added or changed:
Important Change:
If you upgrade to MySQL 4.1.1 or higher, it is difficult to
downgrade back to 4.0 or 4.1.0. That is because, for earlier
versions, InnoDB is not aware of multiple
tablespaces.
Replication: Easier replication upgrade (5.0.0 masters can read older binary logs and 5.0.0 slaves can read older relay logs). See Section 16.3.2, “Replication Compatibility Between MySQL Versions”, for more details). The format of the binary log and relay log is changed compared to that of MySQL 4.1 and older.
Replication:
New binary log format that enables replication of these session
variables: sql_mode,
sql_auto_is_null,
foreign_key_checks (which was
replicated since 4.0.14, but here it is done more efficiently
and takes less space in the binary logs),
unique_checks. Other variables
(like character sets,
sql_select_limit, ...) will be
replicated in upcoming 5.0.x releases.
Added TIMESTAMPADD() and
TIMESTAMPDIFF() functions.
The KILL statement now takes
CONNECTION and QUERY
modifiers. The first is the same as
KILL with no modifier (it kills a
given connection thread). The second kills only the statement
currently being executed by the connection.
Added support for SUM(DISTINCT),
MIN(DISTINCT), and
MAX(DISTINCT).
Basic support for stored procedures and functions (SQL:2003 style). See Section 18.2, “Using Stored Routines (Procedures and Functions)”.
The output of the SHOW BINLOG
EVENTS statement has been modified. The
Orig_log_pos column has been renamed to
End_log_pos and now represents the offset of
the last byte of the event, plus one.
Implemented Index Merge optimization for OR
clauses. See Section 7.2.6, “Index Merge Optimization”.
For user-defined functions (UDFs), the
UDF_ARGS structure now has
attributes and
attribute_lengths members that provide
information about the argument names.
Section 21.2.2.3, “UDF Argument Processing”.
Added WEEK and QUARTER
values as INTERVAL arguments for the
DATE_ADD() and
DATE_SUB() functions.
The precedence of the XOR operator now lies
between OR and AND.
Previously, XOR had the same precedence as
OR.
Added SELECT INTO
, which can be
of mixed (that is, global and local) types. See
Section 12.8.3.3, “list_of_varsSELECT ... INTO Statement”.
LOAD DATA
INFILE causes an implicit commit.
The behavior of
LOAD DATA
INFILE in this regard was changed again in MySQL
5.0.26. See Section C.1.55, “Changes in MySQL 5.0.26 (03 October 2006)”.
Starting from 4.1.13 and 5.0.7, all Cluster changes are included in the MySQL Change History, and this manual section is no longer separately maintained.
Starting with version 5.0.8, changes for MySQL Cluster can be found in the combined MySQL Change History.
Functionality added or changed:
Bugs fixed:
(Bug#11019) mgmapi start backup in some cases returns wrong backupid
(Bug#10190) Backup from cluster wih NoOfReplica=1 is corrupt
(Bug#9246) Condition pushdown and left join, wrong result
(Bug#10956) More than 7 node restarts with
--initial caused cluster to fail.
(Bug#9945) ALTER TABLE caused
server crash. (Linux/390)
(Bug#9826) (Bug#10948) Schema change
(DROP TABLE,
ALTER TABLE) crashed HPUX and
PPC32.
(Bug#10711) (Bug#9363) (Bug#8918) (Bug#10058) (Bug#9025)
Cluster would time out and crash after first query; setting
DataMemory to more than 2GB prevented cluster from starting;
calling ndb_select_count() crashed the
cluster. (64-bit Unix OSes)
Functionality added or changed:
Limit on number of metadata objects (number of tables, indexes and BLOBs) now increased to 20,320
Bugs fixed:
The server would hang on successive calls to an
INSERT
... ON DUPLICATE KEY UPDATE query. (Bug#9725)
(Bug#10193) Invalid DataDir in config causes ndbd segmentation fault
(Bug#10813) Build with SCI Transporter fails
(Bug#10831) ndb mgmd LogDestination maxfiles does not rotate logs properly
Functionality added or changed:
Decreased IndexMemory Usage
Parallel key lookup (read-multi-range) for queries like
SELECT * FROM t1 WHERE primary_key IN
(1,2,3,4,5,6,7,8,9,10);
Bugs fixed:
Patches merged from versions 4.1.11 and 4.1.12
(Bug#8315) NdbScanFilter cmp method only works for strings of exact word boundary length
(Bug#8103) Configuration handling error
(Bug#8035) mysqld signal 10 when ndbd is shutdown
(Bug#7631) NDB$EVENT contains unreadable event and table names
(Bug#7628) Filtered event types are ignored
(Bug#7627) Drop Event operation fails
(Bug#7424) create index on datetime fails
Functionality added or changed:
Condition pushdown to storage engine now works for update and delete as well
Bugs fixed:
(Bug#9675) Auto-increment not working with INSERT..SELECT and NDB storage
(Bug#9517) Condition pushdown to storage engine does not work for update/delete
(Bug#9282) API Node Crashes/Reloads on 'DELETE FROM'
(Bug#9280) Memory leak in cluster when dependent sub-queries are used
(Bug#8585) ndb_cache2 fails on aix52
Functionality added or changed:
Condition pushdown to storage engine
Query cache enabled for cluster
Bugs fixed:
Patches merged from version 4.1.10
Functionality added or changed:
This was the first MySQL Cluster release in the 5.0 series. As nearly all attention was still focused on getting 4.1 stable, it is not recommended to use MySQL 5.0.1 for MySQL Cluster.
Bugs fixed:
N/A
Functionality added or changed:
Bugs fixed:
(Bug#11132) Connections between data nodes and management
nodes were not being closed following shutdown of
ndb_mgmd.
(Bug#11050) ndb_mgm> show printed
incorrectly after master data node failure.
(Bug#10956) More than 7 node restarts with
--initial caused cluster to fail.
(Bug#9826) (Bug#10948) Schema change
(DROP TABLE,
ALTER TABLE) crashed HPUX and
PPC32.
(Bug#9025) Data nodes failed to restart on 64-bit Solaris.
(Bug#11166) Insert records were incorrectly applied by
ndb_restore, thus making restoration from
backup inconsistent if the binlog contained inserts.
(Bug#8918) (Bug#9363) (Bug#10711) (Bug#10058) (Bug#9025)
Cluster would time out and crash after first query; setting
DataMemory to more than 2GB prevented cluster from starting;
calling ndb_select_count() crashed the
cluster. (64-bit Unix OSes)
(Bug#10190) When making a backup of a cluster where
NumberOfReplicas was equal to 1, the
backup's metadata was corrupted. (Linux)
(Bug#9945) ALTER TABLE caused
server crash. (Linux/390)
(Bug#11133) A delete operation performed as part of a transaction caused an erroneous result.
(Bug#10294) Not allowing sufficient parallelism in cluster
configuration (for example,
NoOfTransactions too small) caused
ndb_restore to fail without generating any
error messages.
(Bug#11290) Setting TransactionInactiveTimeout= 0 did not result in an infinite timeout.
Functionality added or changed:
Bugs fixed:
(Bug#10471) Backup can become inconsistent with certain combinations of multiple-row updates
(Bug#10287) ndb_select_all "delimiter" option non functional
(Bug#10142) Unhandled resource shortage in UNIQUE index code
(Bug#10029) crash in ordered index scan after db full
(Bug#10001) 2 NDB nodes get signal 6 (abort) in DBTC
(Bug#9969) 4012 - has misleading error message
(Bug#9960) START BACKUP reports failure albeit succeeding
(Bug#9924) ABORT BACKUP 1 crashes 4 node cluster
(Bug#9892) Index activation file during node recovery
(Bug#9891) Crash in DBACC (line 7004) during commit
(Bug#9865) SELECT does not function properly
(Bug#9839) Column with AUTOINC contains -1 Value on node stop
(Bug#9757) Uncompleted node failure after gracefully stopping node
(Bug#9749) Transactions causes deadlock in ACC
(Bug#9724) Node fails to start: Message: File has already been opened
(Bug#9691) UPDATE fails on attempt to update primary key
(Bug#9675) Auto-increment not working with INSERT..SELECT and NDB storage
(Bug#9318) drop database does not drop ndb tables
(Bug#9280) Memory leak in cluster when dependent sub-queries are used
(Bug#8928) create table with keys will shutdown the cluster
Creating a table did not work for a cluster with 6 nodes. (Bug#8928) Databases with 1, 2, 4, 8, ...
(2n
nodes) did not have the problem. After a rolling upgrade,
restart each node manually by restarting it with the
--initial option. Otherwise, use dump and
restore after an upgrade.
Functionality added or changed:
Bugs fixed:
(Bug#9916) DbaccMain.cpp / DBACC (Line: 4876) / Pointer too large
(Bug#9435) TIMESTAMP columns do not update
(Bug#9052) Uninitialized data during unique index build, potential cluster crash
(Bug#8876) Timeout when committing aborted transaction after node failure
(Bug#8786) ndb_autodiscover, drop index can fail, wait 2 minutes timeout
(Bug#8853) Transaction aborted after long time during node failure (4012)
(Bug#8753) Invalid schema object version after dropping index (crash fixed, currently retry required)
(Bug#8645) Assertion failure with multiple management servers
(Bug#8557) ndbd does not get same nodeid on restart
(Bug#8556) corrupt ndb_mgm show printout for certain configurations
(Bug#8167) cluster shared memory and mysqld signal usage clash
Bugs fixed:
(Bug#8284) Out of fragment memory in DBACC
(Bug#8262) Node crash due to bug in DBLQH
(Bug#8208) node restart fails on Aix 5.2
(Bug#8167) cluster shared memory and mysqld signal usage clash
(Bug#8101) unique index and error 4209 while selecting
(Bug#8070) (Bug#7937) (Bug#6716) various ndb_restore core dumps on HP-UX
(Bug#8010) 4006 forces MySQL Node Restart
(Bug#7928) out of connection objects
(Bug#7898) mysqld crash with ndb (solaris)
(Bug#7864) Not possible to have more than 4.5G data memory
Functionality added or changed:
New implementation of shared memory transporter.
Cluster automatically configures shared memory transporter if possible.
Cluster prioritizes usage of transporters with shared memory and localhost TCP
Added switches to control the above functions,
ndb-shm and
ndb-optimized-node-selection.
Bugs fixed:
(Bug#7805) config.ini parsing error
(Bug#7798) Running range scan after alter table in different thread causes node failure
(Bug#7761) Alter table does not autocommit
(Bug#7725) Indexed DATETIME Columns Return Random Results
(Bug#7660) START BACKUP does not increment BACKUP-ID (Big Endian machines)
(Bug#7593) Cannot Create A Large NDB Data Warehouse
(Bug#7480) Mysqld crash in ha_ndbcluster using Query Browser
(Bug#7470) shared memory transporter does not connect
(Bug#7396) Primary Key not working in NDB Mysql Clustered table (solaris)
(Bug#7379) ndb restore fails to handle blobs and multiple databases
(Bug#7346) ndb_restore enters infinite loop
(Bug#7340) Problem for inserting data into the Text field on utf8
(Bug#7124) ndb_mgmd is aborted on startup when using SHM connection
Functionality added or changed:
Default port for ndb_mgmd was changed to 1186 (from 2200) as this port number was officially assigned to MySQL Cluster by IANA.
New command in ndb_mgm, PURGE STALE SESSIONS, as a workaround for cases where nodes fail to allocate a node id even if it is free to use.
New command in ndb_mgm, CONNECT.
The ndb executables have been changed to make use of the regular MySQL command-line option parsing features. See Section 17.6.21, “Options Common to MySQL Cluster Programs”, for notes on changes.
As bonus of the above you can now specify all command line
options in my.cnf using the executable
names as sections, that is, [ndbd],
[ndb_mgmd], [ndb_mgm],
[ndb_restore], and so forth.
[ndbd]
ndb-connectstring=myhost.domain.com:1234
[ndb_mgm]
ndb-connectstring=myhost.domain.com:1234
Added use of section [mysql_cluster] in
my.cnf. All cluster executables,
including mysqld, parse this section. For example, this is a
convenient place to put ndb-connectstring
so that it need be specified only once.
Added cluster log info events on allocation and deallocation of nodeid's.
Added cluster log info events on connection refuse as a result of version mismatch.
Extended connectstring syntax to allow for leaving the port
number out. For example,
ndb-connectstring|connect-string=myhost1,myhost2,myhost3
is a valid connectstring and connect occurs on default port
1186.
Clear text ndb error messages provided also for error codes
that are mapped to corresponding mysql error codes, by
executing SHOW WARNINGS after
an error has occurred which relates to the ndb storage engine.
Significant performance improvements done for read performance, especially for blobs.
Added some variables for performance tuning,
ndb_force_send and
ndb_use_exact_count. Do
show variables like 'ndb%'; in mysql client
for listing. Use set command to alter
variables.
Added variables to set some options,
ndb_use_transactions and
ndb_autoincrement_prefetch_sz.
Bugs fixed:
(Bug#7303) ndb_mgm: Trying to set CLUSTERLOG for a specific node id core dumps
(Bug#7193) start backup gives false error printout
(Bug#7153) Cluster nodes do not report error on endianness mismatch
(Bug#7152) ndb_mgmd segmentation fault on incorrect HostName in configuration
(Bug#7104) clusterlog filtering and level setting broken
(Bug#6995) ndb_recover on varchar fields results in changing case of data
(Bug#6919) all status only shows 2 nodes on a 8-node cluster
(Bug#6871) DBD execute failed: Got error 897 'Unknown error code' from ndbcluster
(Bug#6794) Wrong outcome of update operation of ndb table
(Bug#6791) Segmentation fault when config.ini is not correctly set
(Bug#6775) failure in acc when running many mysql clients
(Bug#6696) ndb_mgm command-line options inconsistent with behavior
(Bug#6684) ndb_restore doesn't give error messages if improper command given
(Bug#6677) ndb_mgm can crash on "ALL CLUSTERLOG"
(Bug#6538) Error code returned when select max() on empty table with index
(Bug#6451) failing create table givers "ghost" tables which are impossible to remove
(Bug#6435) strange behavior of left join
(Bug#6426) update with long pk fails
(Bug#6398) update of primary key fails
(Bug#6354) mysql does not complain about --ndbcluster option when NDB is not compiled in
(Bug#6331) INSERT IGNORE .. SELECT breaks subsequent inserts
(Bug#6288) cluster nodes crash on data import
(Bug#6031) To drop database you have to execute DROP DATABASE command twice
(Bug#6020) LOCK TABLE + delete returns error 208
(Bug#6018) REPLACE does not work for BLOBs + NDB
(Bug#6016) Strange crash with blobs + different DATABASES
(Bug#5973) ndb table belonging to different database shows up in show tables
(Bug#5872) ALTER TABLE with blob from ndb table to myisam fails
(Bug#5844) Failing mysql-test-run leaves stray NDB processes behind
(Bug#5824) HELP text messed up in ndb_mgm
(Bug#5786) Duplicate key error after restore
(Bug#5785) lock timeout during concurrent update
(Bug#5782) Unknown error when using LIMIT with ndb table
(Bug#5756) RESTART node from ndb_mgm fails
A few more not reported bugs fixed
Functionality added or changed:
Optimization 1: Improved performance on index scans. Measured 30% performance increase on query which do large amounts of index scans.
Optimization 2: Improved performance on primary key lookups. Around double performance for autocommitted primary key lookups.
Optimization 3: Improved performance when using blobs by avoiding usage of exclusive locks for blobs.
Bugs fixed:
A few bugs fixed.
Functionality added or changed:
Limited character set support for storage engine NDBCLUSTER:
| Char set | Collation |
| big5 | big5_chinese_ci |
| big5_bin | |
| binary | binary |
| euckr | euckr_korean_ci |
| euckr_bin | |
| gb2312 | gb2312_chinese_ci |
| gb2312_bin | |
| gbk | gbk_chinese_ci |
| gbk_bin | |
| latin1 | latin1_swedish_ci |
| latin1_bin | |
| sjis | sjis_japanese_ci |
| sjis_bin | |
| tis620 | tis620_bin |
| ucs2 | ucs2_general_ci |
| ucs2_bin | |
| ujis | ujis_japanese_ci |
| ujis_bin | |
| utf8 | utf8_general_ci |
| utf8_bin |
The SCI Transporter has been brought up-to-date with all changes and now works and has been documented as well.
Optimizations when several clients to a MySQL Server access ndb tables.
Added more checks and warnings for erroneous and inappropriate cluster configurations.
SHOW TABLES now directly shows
ndb tables created on a different MySQL server, that is,
without a prior table access.
Enhanced support for starting MySQL Server independently of ndbd and ndb_mgmd.
Clear text ndb error messages provided by executing
SHOW WARNINGS after an error
has occurred which relates to the ndb storage engine.
Bugs fixed:
Quite a few bugs fixed.
Functionality added or changed:
Many queries in MySQL Cluster are executed as range scans or full table scans. All queries that do not use a unique hash index or the primary hash index use this access method. In a distributed system it is crucial that batching is properly performed.
In previous versions, the batch size was fixed to 16 per data node. In this version it is configurable per MySQL Server. So for queries using lots of large scans it is appropriate to set this parameter rather large and for queries using many small scans only fetching a small amount of records it is appropriate to set it low.
The performance of queries can easily change as much as 40% based on how this variable is set.
In future versions more logic will be implemented for
assessing the batch size on a per-query basis. Thus, the
semantics of the new configuration variable
ScanBatchSize are likely to change.
The fixed size overhead of the ndbd process has been greatly decreased. This is also true for the overhead per operation record as well as overhead per table and index.
A number of new configuration variables have been introduced to enable configuration of system buffers. Configuration variables for specifying the numbers of tables, unique hash indexes, and ordered indexes have also been introduced.
New configuration variables:
MaxNoOfOrderedIndexes,
MaxNoOfUniqueHashIndexes
Configuration variables no longer used:
MaxNoOfIndexes (split into the two above).
In previous versions ALTER
TABLE,
TRUNCATE
TABLE, and LOAD DATA
were performed as one big transaction. In this version, all of
these statements are automatically separated into several
distinct transactions.
This removes the limitation that one could not change very
large tables due to the
MaxNoOfConcurrentOperations parameter.
MySQL CLuster's online backup feature now backs up indexes so that both data and indexes are restored.
In previous versions it was not possible to use
NULL in indexes. This is now possible for
all supported index types.
Much work has been put onto making
AUTO_INCREMENT features work as for other
table handlers. Autoincrements as a partial key is still only
supported by MyISAM.
In earlier versions, mysqld would crash if
the cluster wasn't started with the
--ndbcluster option. Now
mysqld handles cluster crashes and starts
without crashing.
The -i option for initial startup of
ndbd has been removed. Initial startup
still can be specified by using the --initial
option. The reason for this is to ensure that it is clear what
takes place when using --initial: this option
completely removes all data from the disk and should only be
used at initial start, in certain software upgrade cases, and
in some cases as a workaround when nodes cannot be restarted
successfully.
The management client (ndb_mgm) now has
additional commands and more information is printed for some
commands such as show.
In previous versions, the files were called
ndb_0.. when it wasn't possible to
allocate a node ID when starting the node. To ensure that
files are not so easily overwritten, these files are now named
ndb_pid.., where pid is the process ID
assigned by the OS.
The default parameters have changed for
ndb_mgmd and ndbd. In
particular, they are now started as daemons by default. The
-n option has been removed since it could
cause confusion as to its meaning (nostart or nodaemon).
In the configuration file, you can now use
[NDBD] as an alias for
[DB], [MYSQLD] as an
alias for [API], and
[NDB_MGMD] as an alias for
[MGM].
In fact, [NDBD],
[MYSQLD], and
[NDB_MGMD] are now the preferred
designations, although the older ones will continue to be
supported for some time to come in order to maintain
backward compatibility.
Many more checks for consistency in configuration have been introduced to in order to provide quicker feedback on configuration errors.
In the connect string, it is now possible to use both
“;” and
“,” as the separator between
entries. Thus, "nodeid=2,host=localhost:2200" is equivalent to
"nodeid=2;host=localhost:2200".
In the configuration file, it is also possible to use
“:” or
“=” for assignment values. For
example, MaxNoOfOrderedIndexes : 128 and
MaxNoOfOrderedIndexes = 128 are equivalent
expressions.
The configuration variable names are now case insensitive, so
MaxNoOfOrderedIndexes: 128 is equivalent to
MAXNOOFORDEREDINDEXES = 128.
It is possible now to set the backup directory separately from
the FileSystemPath by using the
BackupDir configuration variable.
Log files and trace files can now be placed in any directory
by setting the DataDir configuration
variable.
FileSystemPath is no longer mandatory and
defaults to DataDir.
Queries involving tables from different databases are now supported.
It is now possible to update the primary key.
The performance of ordered indexes has been greatly improved, particularly the maintenance of indexes on updates, inserts and deletes.
Bugs fixed:
Quite a few bugs fixed.
Functionality added or changed:
The names of the log files and trace files created by the ndbd and ndb_mgmd processes have changed.
Support for the many BLOB data
types was introduced in this version.
Bugs fixed:
Quite a few bugs were fixed in the 4.1.4 release.
This appendix lists the changes to the MySQL Enterprise Monitor, beginning with the most recent release. Each release section covers added or changed functionality, bug fixes, and known issues, if applicable. All bug fixes are referenced by bug number and include a link to the bug database. Bugs are listed in order of resolution. To find a bug quickly, search by bug number.
This section documents all changes and bug fixes that have been applied since the release of MySQL Enterprise Monitor, version 2.0.5.
Bugs fixed:
An error was generated when an attempt was made to rename a Replication Group to one that had previously been deleted. (Bug#43846)
A new topology was not discovered after the previous replication group was renamed. (Bug#43815)
On Unix systems, executing the command:
./mysqlmonitorctl.sh stop
did not make sure that mysqld was shutdown
before finishing.
This resulted in a situation such as the following:
# /opt/mysql/enterprise/monitor-2.0.0.7092/mysqlmonitorctl.sh stop Using CATALINA_BASE: /opt/mysql/enterprise/monitor/apache-tomcat Using CATALINA_HOME: /opt/mysql/enterprise/monitor/apache-tomcat Using CATALINA_TMPDIR: /opt/mysql/enterprise/monitor/apache-tomcat/temp Using JRE_HOME: /opt/mysql/enterprise/monitor-2.0.0.7092/java Stopping tomcat service ... [ OK ] /opt/mysql/enterprise/monitor-2.0.0.7092/mysqlmonitorctl.sh : mysql stopped
However, running the following command a few minutes later showed that the MySQL server was still running:
# /opt/mysql/enterprise/monitor-2.0.0.7092/mysqlmonitorctl.sh status MySQL Network MySQL is running MySQL Network Tomcat is not running
The MySQL Enterprise Monitor upgrade installer incorrectly
replaced the AdvisorScript.jar in
<instDir>/apache-tomcat/webapps/ROOT/WEB-INF/lib/
with the default Advisor JAR.
(Bug#43773)
The agent created the mysql.inventory table
with an engine type of InnoDB, instead of MyISAM, when InnoDB
was specified as the default engine type in
my.cnf. This happened because the agent did
not explicitly specify the table engine type to be of MyISAM.
(Bug#43551)
If a host was not a slave during the initial discovery phase, then it would not be displayed in the Replication tab if it subsequently became a slave.
This was because after the initial discovery phase, if a host
did not have slavestatus present, no
subsequent checks were made to check for the host being a slave.
It was therefore missed for the purposes of replication
discovery and never showed in the Replication tab.
(Bug#42997)
This section documents all changes and bug fixes that have been applied since the release of MySQL Enterprise Monitor, version 2.0.4.
Bugs fixed:
The FreeBSD 7 Agent was inadvertently a Linux binary.
shell> file mysqlmonitoragent-2.0.5.7153-freebsd7-x86-32bit-installer.bin mysqlmonitoragent-2.0.5.7153-freebsd7-x86-32bit-installer.bin: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.2.5, dynamically linked (uses shared libs), stripped
Calling the Agent with the option
--agent-run-os-tests resulted in a crash. This
happened on Linux x86-64 systems. The resultant stack trace was:
(qa-merlin) 2009-03-04 16:39:42: (critical) chassis.c:1097: could not raise RLIMIT_NOFILE to 8192, Invalid argument (22). Current limit still 1024. sigar-test-all.c.124 (test_sigar_pid_get): pid = 5188 sigar-test-all.c.106 (test_sigar_mem_get): ...cut... sigar-test-all.c.427 (test_sigar_file_system_list_get): (items = 13) [0] fs.dirname = / fs.devname = /dev/mapper/vg00-root fs.typename = local fs.sys-type-name = ext3 fs.type = 2 fsusage.total = 15481840 fsusage.free = 14116140 fsusage.used = 1365700 fsusage.avail = 13329708 fsusage.files = 1966080 fsusage.use_percent = 0.100000 [0] diskusage.reads = 315302 diskusage.writes = 6318240 diskusage.write_bytes = 25879511040 diskusage.read_bytes = 6561092608 diskusage.queue = 47457530080206 Segmentation fault
On some systems no output was shown, other than the message “Segmentation fault”. (Bug#43381)
Following a change in the replication configuration, MySQL Enterprise Monitor did not display the new topology correctly. (Bug#43240)
When a data collection became invalid, the agent sent NULLs for those collection values. However, the timestamps that it sent with the values were the timestamps from the last valid value that was collected.
Due to key constraints on the Service Manager side, MySQL Enterprise Monitor disregarded anything sent with duplicate timestamps, thus the NULLs received from the agent were never processed. Some mechanisms, such as replication discovery, depend on the NULLs to identify a situation where data has become invalid. (Bug#43239)
MySQL Enterprise Monitor did not add a log entry each time data was purged. The log entry should have noted how many rows of each type of data were purged (historical data, logs, quan data). (Bug#43159)
MySQL Enterprise Monitor generated a stack overflow. This was as a result of a bug in Hibernate, which caused Hibernate to enter infinite recursion while trying to load a relationship. (Bug#43107)
The “Table Cache Set Too Low For Startup” and
“Table Cache Not Optimal” rules were not supported
on MySQL 5.1 because the table_cache system
variable was deprecated and replaced with
table_open_cache.
(Bug#42663)
Migrated server was not completely deleted.
In a Monitor that had been updated from 1.3.2 to 2.0.4, with 2 database servers queued for migration, if a server being migrated was deleted, or a migrated server was deleted, this would not be reflected in the user interface or in the license count, until Tomcat was restarted. (Bug#42604)
The installer used to upgrade from version 1.3 corrupted passwords containing the “?” character. (Bug#42452)
Sun multi-core processors caused all cores to be reported on the meta information page.
The larger T-series SPARC processors have 32+ cores. This caused the meta information page in the Dashboard to scroll as it reported each one. (Bug#42355)
If an attempt was made to disable a rule using the link next to the rule, the following error message was generated:
U0002 You must log in to access the requested resource. Go to login page.
However, clicking on the link did not prompt the user to login again. (Bug#42313)
Changing ssh-agent from OpenSSH or specifying
a malevolent value of agent-host-id, could
inject data into the monitored MySQL Server.
For example, setting agent-host-id to the
value “I'm a test” would result in the
following message in the error log:
2009-01-23 15:45:11: ((error)) agent_mysqld.c:281: mysql_real_query('INSERT INTO
mysql.inventory (name, value) VALUES ( 'hostid', 'I'm a test' )') on 'mysql' failed: You
have an error in your SQL syntax; check the manual that corresponds to your MySQL server
version for the right syntax to use near 'm a test' )' at line 1 (mysql-errno = 1064)
When SHOW GLOBAL STATUS returned a value
greater than 214748364, it was sent to the Service Manager as
214748364.
(Bug#42245)
The Agent failed to identify local sockets as local on Mac OS X 10.4.
If the Agent was configured to use a Unix domain socket on Mac OS X 10.4, it did not treat the connection as local and failed to provide CPU and memory information to MySQL Enterprise Monitor. This is shown in the log file:
2009-01-20 18:15:02: (message) network-socket.c:752: is-local family 0 != 1 2009-01-20 18:15:02: (message) agent_mysqld.c:322: [mysql] mysqld is not local or not directly connected
However, this problem did not happen on Mac OS X 10.5. (Bug#42220)
Some graphs on the Graph tab were not updated after the page was refreshed, or was clicked.
The only way to get an updated graph was to change the graph size (in pixels) and then click . (Bug#42162)
The my.cnf file for the Enterprise Monitor
internal database had the following configuration item:
innodb_autoextend_increment = 50M
This generated the error:
16:36:23 [Warning] option 'innodb_autoextend_increment': unsigned value 52428800 adjusted to 1000
This variable is interpreted as being specified in MB, so 50M would be 50 TB. Such a high value results in the variable being adjusted to 1000 MB.
The value in the configuration file should be:
innodb_autoextend_increment = 50
A number of Advisor rules had advice text that had not been translated into Japanese. The Advisors that contained untranslated rules included Performance, Schema and Security. (Bug#42067)
The Service Manager did not handle the case where the agent
failed to supply a valid master_ip. The
Service Manager would then not restart. The logs contained the
following:
2009-01-09 14:39:50,472 ERROR [main:org.springframework.web.context.ContextLoader] Context initialization failed org.springframework.beans.factory.UnsatisfiedDependencyException: Error creating bean with name 'serverConnectionMonitor' defined in ServletContext resource [/WEB-INF/applicationContext.xml]: Unsatisfied dependency expressed through constructor argument with index 2 of type [com.mysql.etools.monitor.bo.Manager]: Error creating bean with name 'manager' defined in ServletContext resource [/WEB-INF/applicationContext.xml]: Invocation of init method failed; nested exception is com.mysql.etools.monitor.pom.UnsupportedAttributeException: 101c6b5b-15eb-49aa-916c-843c51b28d38: mysql.slavestatus.Master_ip
Having too many users with strong privileges generated Service Manager errors and events failed to be triggered.
If there were approximately 1000 users with full privileges and
the value of group_concat_max_len was set to
100001, the size of the data that the agent sent to the Service
Manager was too large and caused Service Manager errors. Also,
the Security event “Account has Strong MySQL
privileges” did not trigger.
(Bug#41987)
Query Analyzer's Query Type filter for SELECT
ignored statements starting with parenthesis.
If you sent a statement through Query Analyzer starting with parenthesis, such as:
(SELECT 1 FROM dual);
and then attempted to filter for SELECT
queries only, queries starting with parenthesis were not
displayed.
(Bug#41968)
The agent angel process was spawning too soon, which caused a duplicate UUID error.
g_error() logged a fatal error, and called
abort(). This terminated the program. Then
the angel respawned with the same UUID, but with a -1 session
resync request, which the MEM server denied because the old
session was still active. This resulted in a permissions denied
error (1142) on the mysql.inventory table.
2008-12-17 18:58:58: (message) agent_mysqld.c:313: [mysql] mysqld is local and directly connected
2008-12-17 18:58:58: ((error)) agent_mysqld.c:444: [mysql] mysql_real_query("SELECT value
FROM mysql.inventory WHERE name = 'uuid'") failed: SELECT command denied to user
'ent_agent'@'127.0.0.1' for table 'inventory' (errno=1142)
2008-12-17 18:58:58: (debug) chassis.c:282: 15134 returned: 15134
2008-12-17 18:58:58: (message) chassis.c:304: [angel] PID=15134 died on signal=6 (it used
0 kBytes max) ... waiting 3min before restart
2008-12-17 18:59:00: (debug) chassis.c:244: we are the child: 15149
2008-12-17 18:59:00: (message) chassis.c:259: [angel] we try to keep PID=15149 alive
2008-12-17 18:59:00: (debug) chassis.c:277: waiting for 15149
2008-12-17 18:59:00: (message) mysql-proxy 0.7.0 started
2008-12-17 18:59:00: (message) MySQL Monitor Agent 2.0.0.7111 started.
master_uuid discovery did not work with MySQL
Server versions prior to 5.1. master_uuid did
not show in any Agent to Monitor communications, and no log or
error messages were generated.
However, now the bug has been fixed, an Agent monitoring a 5.0 MySQL Server would register the following in its logs:
... <classname>slavestatus</classname> <instance>12515cdc-8c00-4223-9d2a-2666a403512c</instance> <attribute>Master_uuid</attribute> </target> <utc>2009-03-03T19:58:05.700Z</utc> <value>b2fd9f86-6e42-49f2-b930-e8fb3e728179</value>
Note the presence of master_uuid.
(Bug#41525)
The master_uuid was not used for replication
topology discovery.
The agent collected master_uuid by reading
the master.info file, and then running a
SELECT directly against its master. This was
to try and read the mysql.inventory table on
the master to obtain the instance
master_uuid.
However, this was not being used correctly by the replication topology discovery within the server. (Bug#41519)
Queries such as SELECT against the master
mysql.inventory was not logged on slave-based
agents. This made diagnosing topology discovery issues
difficult.
(Bug#41518)
After an error was generated due to an incorrect password while trying to create a new user, the following error was obtained when subsequently attempting to create a valid new user:
U0002 You must log in to access the requested resource
Agent startup failed on Solaris 10. The error generated was:
# ./mysql-monitor-agent start Bad string ERROR! /opt/mysql/enterprise/agent/etc/mysql-monitor-agent.ini has to have a [mysql-proxy].pid-file setting
This was caused by unexpected behavior of the tr command. (Bug#41260)
On the Query Analysis page, if a query was clicked the minimum displayed was greater than the average. (Bug#41259)
In some circumstances the agent/proxy ran out of file descriptors, causing secondary failures. It could not recover from that state. The relevant part of the log file is shown here:
2008-11-27 11:11:00: (critical) last message repeated 2 times
2008-11-27 11:11:00: (critical) job_collect_os.c:411: sigar_cpu_info_list_get() failed
2008-11-27 11:11:00: (critical) job_collect_os.c:445: sigar_cpu_list_get() failed
2008-11-27 11:11:00: (critical) job_collect_os.c:411: sigar_cpu_info_list_get() failed
2008-11-27 11:11:00: (critical) job_collect_os.c:445: sigar_cpu_list_get() failed
2008-11-27 11:11:00: (critical) job_collect_os.c:411: sigar_cpu_info_list_get() failed
2008-11-27 11:11:00: (critical) job_collect_os.c:445: sigar_cpu_list_get() failed
2008-11-27 11:11:00: (critical) job_collect_os.c:411: sigar_cpu_info_list_get() failed
2008-11-27 11:11:00: (critical) job_collect_os.c:445: sigar_cpu_list_get() failed
2008-11-27 11:11:30: (critical) network-socket.c.292: socket(127.0.0.1:3306) failed: Too many open files (24)
2008-11-27 11:11:30: (critical) proxy-plugin.c.1532: Cannot connect, all backends are down.
2008-11-27 11:20:22: (critical) last message repeated 4 times
2008-11-27 11:20:22: (critical) network-io.c:215:
curl_easy_perform('https://user:password@merlin-dashboard:443/heartbeat') failed:
SSL connection timeout (curl-error = 'Timeout was reached' (28), os-error = 'Connection refused' (111))
If an installation of Service Manager 2.0.0.7102 included a
backup directory, due to a previous
upgrade, and was upgraded using at least Service Manager
2.0.0.7103, then the installer displayed an error message and
exited.
The error message displayed was:
There has been an error. Error renaming /Applications/mysql/enterprise/monitor/apache-tomcat to /Applications/mysql/enterprise/monitor/backup/apache-tomcat The application will exit now
The Agent started without problems and connected to the master.
But it was unable to perform a SELECT from
the table mysql.inventory in order to obtain
server information.
(Bug#40933)
Canonical Query Text for Select -1 was
displayed as SELECT -? instead of
SELECT ? on the Query Analyzer tab.
(Bug#40435)
The agent installer for Solaris 8 x86 32-bit was missing. (Bug#40248)
The startup scripts supplied with MySQL Network Monitoring and
Advisory tool did not supply all of the LSB
init.d script options required. A list of
the required options can be found at the following website
http://refspecs.freestandards.org/LSB_3.1.0/LSB-Core-generic/LSB-Core-generic/iniscrptact.html
The required options missing include status
and force-reload. The option
status is used by monitoring tools and
cluster software such as Red Hat Cluster, to ensure that the
service is still running. The force-reload
option is an alias for restart.
(Bug#29848)
Multiple errors showed in the agent log after issuing a
SHOW INNODB STATUS statement. The
InnoDB Buffer Poolgraph also went blank.
(Bug#27372)
This section documents all changes and bug fixes that have been applied since the release of MySQL Enterprise Monitor, version 2.0.3.
Bugs fixed:
The Service Agent did not log connections to, and disconnects from, the monitored database.
This meant it was not possible to check if the Agent was connected to the database by simply looking at the log file. (Bug#42403)
The Service Agent failed to create the
mysql.inventory table. The logs displayed the
following error message:
(critical) (share/mysql-proxy/quan.lua:711) [proxy] please add SELECT permissions for this user on mysql.inventory to enable the QUAN feature, got Table 'mysql.inventory' doesn't exist
This happened even though the Service Agent used the
root account.
(Bug#42389)
After installing Service Agent 2.0.4.7138 and then starting it
with --version, the incorrect version was
displayed. Although 2.0.4.7138 was installed, the Service Agent
displayed the version number as 2.0.2.7138.
(Bug#42263)
An invalid path was shown in the error message if the upgrade installer failed to find the previous install location.
The error message is shown below, note that the error message displays a different path to that provided by the user:
Please specify the directory that contains the previous installation of the MySQL Enterprise Monitor Agent Installation directory [/home/mysql/mysql/enterprise/agent]: /var/lib/mysql/agent Warning: The directory /home/mysql/mysql/enterprise/agent does not contain a a previous installation. Please select the right installation directory. Press [Enter] to continue : ---------------------------------------------------------------------------- Please specify the directory that contains the previous installation of the MySQL Enterprise Monitor Agent
The agent password error in the Service Manager did not log the originating host, making it impossible to determine the agent that failed to log in:
ErrorJan 5, 2009 4:56:08 PM<?xml version='1.0'?><exceptions><error><![CDATA[E1702: IncorrectPasswordException: [agent]]]></error></exceptions>
The Service Manager's JDBC connections did not have
sql_mode set to include
NO_ENGINE_SUBSTITUTION. This resulted in the
failure of Data Definition Language (DDL) if the Service Manager
was inadvertently pointed to an incorrect
mysqld instance.
(Bug#42137)
A number of Advisor rules had advice text that had not been translated into Japanese. The Advisors that contained untranslated rules included Performance, Schema and Security. (Bug#42067)
Service Agent configuration files had global read privileges on the filesystem. (Bug#41794)
When the log files rotated to the maximum allowed by the
log4j configuration, the metadata contained
in the FileAppender became out of
synchronization due to a logic bug. This caused an assertion
error on any subsequent request for the logs tab or data.
(Bug#41593)
SNMP trap messages were sending 127.0.0.1 as the IP address, and there was no feature to allow the user to configure the IP address contained in the SNMP message, which would have been useful for troubleshooting. (Bug#41361)
On all available test systems, including SUSE Linux Enterprise Server (SLES) 9 and 10 x86-32, x86-64, IA64, and Ubuntu 6.10 x86-64, the agent memory grew incrementally within ~3 hours to ~25M.
The agents then switched to the 'red' condition on the dashboard and no more data was reported. The agent processes were still alive.
This was also present in MySQL Enterprise Monitor version 2.0.0.7111 on Suse 7.3, with the glibc2.2 x86 agent.
There was no load on the monitored servers, and the problem also occurred when the “self-quan” was not configured. (Bug#41244)
Although the Monitor tab loaded initially, after a 64-bit MySQL server running a 32-bit MySQL Monitor Agent was clicked, a Null Pointer Exception was generated. (Bug#41164)
The Service Agent startup configuration did not seem to work and did not generate a log file. (Bug#40583)
An error generated by a rule failed to clear.
When a rule was created with the
os:disk:fs_used data item, if the instance
was not a valid mount point then the Service Agent reported the
error:
2008-08-11 17:57:00: (critical) disk-get failed for all: 2
Note the above instance was for “all”, and similar results occurred for instance “disk”.
The issue was that upon editing the rule, or even deleting the rule, the agent still showed the above error type. Restarting the agent and the monitoring service failed to remove the error. (Bug#38709)
If the “On Save send test trap” checkbox was checked when the button was clicked and the locale was set to Japanese, an error occurred. The orange error banner was displayed at the top of the page with the error message in Japanese. (Bug#32069)
This section documents all changes and bug fixes that have been applied since the release of MySQL Enterprise Monitor, version 2.0.2.
Bugs fixed:
The Service Agent failed to connect to the server, and no error message was displayed in the log file.
However, when the log verbosity level was set to
message, the following message was recorded:
2009-01-12 13:34:43: (warning) agent_mysqld.c:600: agent connecting to mysql-server failed: mysql_real_connect(host = '127.0.0.1', port = 3306, socket = ''): Lost connection to MySQL server at 'reading initial communication packet', system error: 0 (mysql-errno = 2013)
There was a mismatch between the contents of an SNMP Trap from
MySQL Enterprise Monitor and the definition given in the file
MONITOR.MIB.
(Bug#41912)
If a copy was made of a standard rule, the resulting Wiki markup was incorrect, resulting in the display of user-interface text containing HTML markup. (Bug#41375)
Allowing the heat chart rules to be set to unscheduled caused the user interface to appear broken. (Bug#41312)
When a custom rule requiring disk information was created, for example:
[Expression] %disk_reads% > THRESHOLD [Thresholds] Critical Alert: 9000 Warning Alert: 3000 Info Alert: 1000 [Variable Assignment] variables: %disk_reads% Data Item: os:disk:disk_reads Instance: /
The following error was written to the file
mysql-service-agent.log:
2007-09-05 17:11:00: (critical) disk-get failed for c0d0p1: 2
It therefore appeared that the Service Agent was not able to obtain the required information. (Bug#30820)
This section documents all changes and bug fixes that have been applied since the release of MySQL Enterprise Monitor, version 2.0.1.
Bugs fixed:
The button was missing from the Setup page you were taken to after the current subscription expired. (Bug#41685)
If the host-id of the monitored instance and
the host-id of the current agent did not
match, the agent generated the following error message:
Please TRUNCATE TABLE mysql.inventory on this mysql-instance and restart the agent.
However, it did not suggest using sql_log_bin =
0. This is used for all other actions against this
table so that they are not replicated to slaves, each of which
has their own copy of this table.
(Bug#41673)
The Agent did not start up when the monitored server had many
databases and tables, and was under heavy load. This was because
the trigger_schema query was taking too long
on agent start up.
(Bug#41555)
The Service Agent failed to install on Solaris 10 x86. The following error was generated:
Installing 0% ______________ 50% ______________ 100% ######################################## Warning: Problem running post-install step. Installation may not complete correctly Error running /usr/local/mysqlagent/bin/mysql-monitor-agent --defaults-file=/usr/local/mysqlagent/etc/mysql-monitor-agent.ini --plugins=agent --agent-generate-uuid=true : 2008-12-12 13:06:02: (critical) Conversion from character set '646' to 'UTF-8' is not supported 2008-12-12 13:06:02: (message) shutting down normally
The console/stdout appender remained in the log4j configuration,
which meant that all the MySQL Enterprise Monitor server logs
were duplicated to Catalina's stdout, and thus
catalina.out, which was wasteful,
especially as that file was not rotated or managed.
(Bug#41439)
When creating new multiple user accounts, the first attempt worked fine. However, following attempts to create new users did not show the Query Analyzer Options in the Create User popup until the role field was changed. (Bug#41430)
When login privileges were required the Service Manager did not redirect the user to the login page. This resulted in error messages being displayed rather than simply redirecting the user to the login page. This problem typically occurred if it was necessary to restart Tomcat. (Bug#41320)
The monitor 2.0.0.7105 and 2.0.0.7122 Solaris Intel update installer quits unexpectedly. The installer exits from the GUI in the Backup of Previous Installation screen, when OpenSolaris is running on top of Sun xVM.
The console output for both installer versions is given below:
shell> ./mysqlmonitor-2.0.0.7105-solaris-intel-update-installer.bin X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 73 (X_GetImage) Serial number of failed request: 21161 Current serial number in output stream: 21161
shell> ./mysqlmonitor-2.0.0.7122-solaris-intel-update-installer.bin X Error of failed request: BadMatch (invalid parameter attributes) Major opcode of failed request: 73 (X_GetImage) Serial number of failed request: 21148 Current serial number in output stream: 21148
The Rename Group function failed if the new group name was different to the current one only in the case used; for example, if “Merlin” was changed to “MERLIN”.
The error generated was:
U0105 This group name is already in use. Enter a different name.
The Agent returned an inventory list of all databases and
tables. This information was not used by MySQL Enterprise
Monitor, other than to populate the
inv_databases and
inv_tables tables. For large-scale
deployments, where there were many databases and tables, this
resulted in redundant XML messages being sent from the Agent to
the Service Manager.
(Bug#33150)
This section documents all changes and bug fixes that have been applied since the release of MySQL Enterprise Monitor, version 2.0.0.
Bugs fixed:
This section documents all changes and bug fixes that have been applied since the release of MySQL Enterprise Monitor, version 1.3.3.
Functionality added or changed:
Important Change:
The server-name configuration parameter is
deprecated. For compatibility, during an upgrade, the
information will be migrated to a displayname
configuration parameter within the individual instance
configuration files. This configuration parameter is provided
only for compatibility, as display name information is now
stored within the main repository. Support for
displayname is also deprecated and will be
removed in a future release.
Important Note: The rules “32-Bit Binary Running on 64-Bit AMD Or Intel System” and “Key Buffer Size Greater Than 4 GB” occasionally do not evaluate correctly due to timing issues. This causes them to be displayed with the Severity level of “Unknown”. This is a known issue and will be resolved in future versions of MySQL Enterprise Monitor.
Important Note: When you start the Merlin 2.0 agent from the command line on Windows, you get the following error dialog:
"mysql-proxy.exe - Entry Point Not Found" "The procedure entry point libiconv_set_relocation_prefix could not be located in the dynamic link library iconv.dll"
If you click the agent works fine after that.
This only occurs when starting the agent from the command line,
and only when there is another version of one of the DLLs that
the agent uses somewhere on the current path. This error can be
avoided by opening a command prompt, typing SET
PATH= to clear the path, and then starting the agent.
Important Note: If you are monitoring one instance of MySQL server (mysqld) and then upgrade that MySQL server, the correct version of the MySQL server is not displayed in the Dashboard. This is a known issue that will be fixed in future versions of MySQL Enterprise Monitor.
The following has been added to the Tomcat
config.properties properties file:
# max connections in the pool for the repository
default.maxActive=70
The dashboard could be used to change the agent password to one
containing the @ character, or other special
characters, which subsequently caused errors. To fix this
problem, special characters in passwords are now prevented by
the dashboard. The list of disallowed special characters can be
found at the following location:
http://en.wikipedia.org/wiki/Percent-encoding#Types_of_URI_characters
(Bug#37172)
The Query Analysis page was missing the Refresh dropdown control that the Monitor, Events, and Graphs pages had at the top. (Bug#36831)
The User Interface only returned error strings, without any associated error codes. This meant that if the error string was in a language that the user did not understand, it would be very difficult to determine which error actually occurred.
The User Interface now supports error codes, as well as error strings. This change allows easier testing of multiple locales. (Bug#32131)
The wording was changed on the fading popup subscription alert. The text “account” was changed to “subscription”. (Bug#31492)
The alert thresholds for the Query Cache Advisor were changed:
| Information | Warning | Critical | |
|---|---|---|---|
| From | 95 | 85 | 75 |
| To | 60 | 50 | 40 |
The agent log file name has changed from
mysql-service-agent.log to
mysql-monitor-agent.log. The old log file
will be retained during a upgrade install.
Bugs fixed:
Starting a new agent against an instance that contained many databases broke up the initial discovery packet, causing collections such as CPU usage and their graphs to fail. See also Bug#33150 and Bug#41314. (Bug#41933)
When altering an existing rule you had to add an empty string,
"", to any threshold level that was empty.
Otherwise, the rule failed to run and the resulting exceptions
caused the Events page to be unusable due
to duplicate key exceptions.
(Bug#41310)
After installing the 2.0.0.7119 Dashboard the following error was generated in the logs:
com.opensymphony.xwork2.util.OgnlValueStack Warning Dec 5, 2008 10:41:26 AM Caught an Ognl exception while getting property titleAddition
The dc_ng_long_now table became very large
partly due to an unused column begin_time.
(Bug#41093)
Although there was a unique constraint on the user name, it was
not enforced during first-time setup. This resulted in a stack
trace being produced, rather than a more user-friendly error
message, if the same name was used for the
admin and agent accounts.
(Bug#40870)
MySQL Enterprise Monitor server had stopped sending up/down SNMP traps. (Bug#40861)
The Query Analyzer's Explain Query tab did
not have pop up text or a link to the documentation regarding
SELECT queries.
(Bug#40841)
When you tried to import a Trial-level advisor JAR using a Trial user account, one of the following error messages was generated:
U0009 The uploaded Advisor jar was invalid
U0161 Please import a Platinum level Advisor .jar to use with this Platinum level product key
Meta Info on the Dashboard did not display information for the
meta data os_description.
(Bug#40830)
Attempting to create an alias of statements such as
COMMIT, ROLLBACK,
BEGIN, resulted in the error:
Can't find template commitTnx.st; group hierarchy is [HTMLFormatting] org.antlr.stringtemplate.StringTemplateGroup.lookupTemplate(StringTemplateGroup.java:507)
Trying to upgrade from 2.0.0.7088 to 2.0.0.7092 failed as there
was a missing file. When the update program
mysqlmonitor-2.0.0.7092-solaris-intel-update-installer.bin
was run, the file
/tmp/com/mysql/merlin/server/version.props
could not be found.
(Bug#40692)
On OS X with Java 1.5, Tomcat crashed on launch with the error:
Invalid memory access of location 00000007 eip=013df179
When the agent was installed using the command line installer
and --enableproxy 0 was specified, the
installer should have removed quan.lua from
the agent-item-files option in the INI file.
(Bug#40551)
The Agent could not connect to a database with the
hostname set to localhost.
Doing so resulted in the error:
(critical) the MySQL server could not be reached at socket '(null)', we will check in 10 seconds
The update installer for the 2.0 Monitor did not have an
Is SSL support required? checkbox.
Therefore, the appropriate SSL connector definition was
commented out in the conf/server.xml file.
(Bug#40414)
The Service Manager installer did not uninstall or wipe out the previous installation if you answered “Yes” to the question:
“The directory you selected already contains a MySQL installation. If you continue installation all data will be lost. Do you wish to continue?” (Bug#40410)
If you unchecked the Enable Proxy checkbox on the Query Analysis Configuration screen, the agent's INI file still contained proxy configuration data and was not commented out. (Bug#40272)
As different queries were sent through the agent it used increasing amounts of memory. (Bug#40260)
The Service Manager installer set the Java Virtual Machine
option HeapDumpPath to
<install_root>\temp on Windows and
/tmp on other platforms. For consistency
the HeapDumpPath directory should have been
set to <install_root>\tmp on Windows.
(Bug#40215)
When using the command line agent setup program, a socket file was not accepted for the monitored instance. (Bug#40085)
The language option when installing the MySQL Enterprise Monitor in Japanese
using the command-line instllaer has been changed from
jp to ja.
(Bug#40082)
When monitoring MySQL 5.1.24 and above, the user used by the
agent to connect to the MySQL server for monitoring must have
the PROCESS privilege.
(Bug#40050)
The check box option string “Is SSL support required?”, on the Tomcat Server Option dialog of the Monitor installation, was not correctly translated into Japanese. (Bug#39814)
The base application directory for the MySQL Enterprise Dashboard has been
updated from http://localhost:18080/merlin to
http://localhost:18080/.
(Bug#39403)
If the Service Manager or the MySQL Server running the Repository crashed, they did not restart automatically. (Bug#39377)
If the agent crashed, there was no watchdog, angel or keep-alive mechanism to restart the agent and keep it running. (Bug#39374)
If the MySQL client libraries were located in a nonstandard location, the agent 2.0.0.7042 installer failed with a “library not found” error. (Bug#39317)
For the rule “Server-Enforced Data Integrity Checking Not
Strict”, the Recommended Action did not display
correctly. It displayed as SET
sql_mode=modes, rather than SET
[GLOBAL|SESSION] sql_mode=modes.
(Bug#39261)
A query that was issued through the proxy, and that had an
auto-explain performed on that query, did not give the correct
response to a subsequent query of SELECT
FOUND_ROWS().
(Bug#39223)
It was not possible to establish a connection with the Dashboard using the SSL protocol (https://). (Bug#39198)
When a 1.3 MySQL Enterprise Monitor installation with many rules scheduled was upgraded to 2.0, the upgraded installation was then found to have only the heat chart rules scheduled. (Bug#39043)
The agent installer did not allow a second agent to be installed on Windows. (Bug#38976)
The Dashboard incorrectly diplayed that insufficient licenses were available, even though sufficient licenses had been purchased. (Bug#38514)
After running the MySQL Service Agent uninstall program, the
file /etc/init.d/mysql-service-agent
remained present on the server.
(Bug#38490)
The notice fader continued to display English text after you changed the locale to Japanese. (Bug#38460)
The Subscription Expired pop-up window referred to the “Global Preferences” page, instead of the “Global Settings” page. (Bug#38358)
An inappropriate time zone was used to parse user-entered date and time strings. (Bug#38323)
A sigar network stats error was generated on the Solaris platform:
# /opt/mysql/enterprise/agent/bin/mysql-service-agent --version MySQL Service Agent - 1.2.0.7879, (glib lib=2.8.5, headers=2.8.5) SunOS mysqlprd01 5.10 Generic_127127-11 sun4v sparc SUNW,T5240 2008-07-21 10:07:24: (critical) sigar_net_interface_config_primary_get() failed: 6 2008-07-21 10:08:00: (critical) sigar_net_interface_config_primary_get() failed: 6 # /opt/mysql/enterprise/agent/bin/sigar-test-all >/tmp/test.txt sigar_net_interface_stat_get(e1000g0:2) failed#
After deleting a server from the Settings, Manage Servers tab, at the very bottom of the page the Monitoring x instances on x host values did not reflect the deletion. (Bug#38225)
The MySQL Enterprise Monitor alert “INFO Alert - Users Can View All Databases On MySQL Server (v 1.5 *)” from the Security advisor was incorrect. This is because the default server behavior allows users to see databases for which they have privileges, not “all databases on server” as suggested by the alert. (Bug#38052)
The “Maximum Connection Limit Nearing Or Reached” advisor did not generate a new Critical Alert event when there was an open info success event. (Bug#37816)
The Linux IA64 installer appeared to crash. The installer appeared to crash on RH4_IA64 if called with option "--version":
------------------------------------------------------------------------------- <INSTALLER> --version mysqlmonitorage(30704): unaligned access to 0x6000000000a8413c, ip=0x2000000003ddd5f0 mysqlmonitorage(30704): unaligned access to 0x6000000000a84144, ip=0x2000000003ddd5f1 mysqlmonitorage(30704): unaligned access to 0x6000000000a8414c, ip=0x2000000003ddd600 mysqlmonitorage(30704): unaligned access to 0x6000000000a84154, ip=0x2000000003ddd601 MySQL Enterprise Monitor Agent 0.7.0.1737 --- Built on 2008-06-25 19:31:53 -------------------------------------------------------------------------------
However, this warning is harmless and will not impact the operation of the agent. (Bug#37496)
If the log-level option in the agent configuration file was set
to an unknown level by mistake, the init.d
script appeared to enter an infinite loop.
(Bug#37108)
Malformatted server meta information appeared on the Dashboard; RAM and Disk Space appeared under the CPU category. (Bug#36740)
A “rename” link incorrectly appeared next to the Upgrade category on the Manage Rules page. (Bug#36584)
In the case where exceptions were passed through to the User Interface, the substituted arguments in the message contained developer-only information. (Bug#36580)
Agent Version, Last MySQL Contact, OS Info, CPU Info, and IP Addresses were all blank on the dashboard when the agent for the selected server was not functioning. (Bug#36301)
mysql-monitor-agent became confused by the
.DS_Store files that are created on Mac OS
X.
(Bug#36216)
The rule “Key Buffer Size Greater Than 4 GB” incorrectly triggered the following alert:
CRITICAL Alert - Key Buffer Size Greater Than 4 GB
However, on non-Windows systems, a key buffer size greater than 4 GB is supported. (Bug#36143)
Since the repository database for MySQL Enterprise Monitor uses InnoDB there was no way to reduce the size of the data files after an old log/event data purge operation. Further, the purge data operation executed once per day, and had no option to trigger the purge operation manually. (Bug#35971)
On the Graphs page, if all graphs were expanded, then the Time Display interval updated, the page was refreshed with the button displayed, even though all the graphs were already expanded. (Bug#35917, Bug#35133)
The Meta Info area of the Monitor page incorrectly reported the operating system version number for the MySQL version. (Bug#35836)
The rule “XA Distributed Transaction Support Enabled For InnoDB” incorrectly sent a warning when the binary log was on. (Bug#35786)
On the Monitor page, the time displayed for Last MySQL
Contact lagged behind that for Last Agent
Contact by a large amount.
(Bug#35774)
MySQL Enterprise Monitor did not update replication settings correctly. After a slave became the master, the Adviser still referred to it as a slave. (Bug#35771)
The Adviser email suggested using the
--log-queries-not-using-indexes option.
However, this option is not available in MySQL Server versions
prior to 4.1.
(Bug#35770)
Thumbnail graphs did not update properly after a time zone change. (Bug#35756)
If a system had a global wait_timeout lower
than the general activity of the agent, the agent was
disconnected. The monitored server then logged an error and
incremented Aborted_clients.
(Bug#35648)
Alerts fired after a blackout period based on data collections that happened during the blackout. (Bug#35617)
The translation of the button on the Rule Definition window was incorrect. (Bug#35495)
An uninstallation message asked about removing Apache files, even though Apache is no longer used. (Bug#35154)
After updating from a previous version to the latest 1.3 version, the Query Cache Has Sub-Optimal Hit Rate was still displayed in English after setting the locale to Japanese. Note, the rule was translated correctly if the full installer was used. (Bug#35134)
The MySQL Enterprise Monitor uninstall dialog box had missing text when using the Japanese locale. (Bug#34982)
Running the installer with the --help option
caused an incorrect message to be displayed.
(Bug#34200)
When using the unattended unInstall script
on Linux together with the option --env
deleteUserData=yes the correct warning text was
displayed. However, this text should not be displayed in
unattended mode. Further, the option --env
deleteUserData=yes was not displayed by the
--help output.
(Bug#34071)
Platinum Unlimited customers sometimes received a warning stating incorrectly that their subscription supported zero hosts. (Bug#34010)
Clicking on the resolution notes link for a closed event on the events tab showed incorrect behaviour. The popup initially showed the resolution notes, but when the resolution tab was clicked the notes disappeared and were replaced by an edit box. (Bug#33935)
The status on the product information page was not translated when the user locale was set to Japanese. (Bug#32785)
When the locale was set to Japanese, the date picker still had English month titles. (Bug#32741)
Flashing display of a pop-up used while saving outgoing email settings was caused by problematic initial placement calculations. (Bug#32579)
AIX 5.2 Agent did not work on AIX 5.3. (Bug#32414)
When the First Time Setup program was run it did not prompt the user to allow importing an Advisor bundle. (Bug#32199)
Agent on MacOSX did not read IP addresses for network interfaces correctly, so the monitor displayed empty host IP addresses. (Bug#32188)
HTML code in queries was not escaped when reporting replication errors, causing the code to be rendered into the page. (Bug#32186)
The First Run pop-up defaulted to English rather than to the locale set in the browser. (Bug#32129)
The error dialog box flashed in the upper left corner before being positioned in the center of the screen. This error dialog box now opens in the center of the screen. (Bug#32068)
The Events list did not take into account Daylight Time and Standard Time when listing events that happened during 1:00am-1:59am. An event that occurred at 1:10am Standard Time was listed before an event that occurred 50 minutes before it at 1:20am Daylight Time. (Bug#32016)
The pop-up for editing log levels failed to load due to bad instantiation data. (Bug#32013)
During the repeated hour of Daylight Savings Time (when 2am turns back into 1am), the graphs were not drawing data. Instead, there was a straight line from the point at 1:00 to the second 1:00, which is what happens if there is no data. The repository did, however, have data for this hour. (Bug#31997)
Only US English was supported for a locale setting. Other
English variants are now available for the
locale setting on the General
Settings or the User Preferences
pages.
(Bug#31801)
If the user locale was changed the graph cache would continue to display the graph in the last locale until it timed out. (Bug#31680)
No init script was installed for the MySQL Network Monitoring Service Manager, and so it did not restart automatically on reboot. (Bug#31676)
The graph's displayed time was not the local time of the Dashboard corresponding to the requested time on the monitored server. (Bug#31656)
Saving a rule with a name that already existed resulted in a stack trace in the window, instead of a more user-friendly error message. (Bug#30925)
The network.mysql.com error messages were
remapped thereby causing confusion. For example, the following
error message:
E9000: MySQL Enterprise Customer Center is having difficulties fetching your contract information. Please contact [email protected] for assistance.
Was remapped to:
Unable to connect to verify credentials (Bug#30873)
A newly added server showed as “down” in the user interface, and could potentially have sent a false alarm notification. (Bug#30735)
The information on the , page did not accurately reflect how many rules and graphs were actually in the database and available to the user. (Bug#29623)
The agent did not process SIGHUP.
(Bug#29380)
Monitor did not have a facility to stop or downgrade an agent collection frequency. (Bug#28589)
After an agent installation was updated from 1.0.1.4391 to
1.1.0.4899, the version in the menu was incorrectly displayed as 1.0.1.4391,
even though the update was successful and the file version of
agent.exe was correctly displayed as
1.1.0.4899.
(Bug#27447)
When viewing the Results of an Event in the Events tab of the Dashboard, the Notifications section did not reflect the Notifications settings at the time the Event was triggered, but rather the Notifications settings at the time the Event Results were viewed. (Bug#26349)
When the Service Agent was remotely monitoring a MySQL server it incorrectly reported that it could collect operating system information. (Bug#22497)
The Account Without Password advisor did not report all users who were without a password, it only reported one. (Bug#15165)
Bugs fixed:
Connector/ODBC failed to build with MySQL 5.1.30 due to
incorrect use of the data type bool.
(Bug#42120)
Calling SQLDescribeCol() with a NULL buffer
and nonzero buffer length caused a crash.
(Bug#41942)
MySQL Connector/ODBC updated some fields with random values,
rather than with NULL.
(Bug#41256)
Calling SQLDriverConnect() with a
NULL pointer for the output buffer caused a
crash if SQL_DRIVER_NOPROMPT was also
specified:
SQLDriverConnect(dbc, NULL, "DSN=myodbc5", SQL_NTS, NULL, 0, NULL, SQL_DRIVER_NOPROMPT)
Setting the ADO Recordset decimal field value
to 44.56 resulted in an incorrect value of 445600.0000 being
stored when the record set was updated with the
Update method.
(Bug#39961)
The SQLTablesW API gave incorrect results.
For example, table name and table type were returned as
NULL rather than as the correct values.
(Bug#39957)
MyODBC would crash when a character set was being used on the server that was not supported in the client, for example cp1251:
[MySQL][ODBC 5.1 Driver][mysqld-5.0.27-community-nt]Restricted data type attribute violation
The fix causes MyODBC to return an error message instead of crashing. (Bug#39831)
When the SQLTables method was called
with NULL passed as the
tablename parameter, only one row in the
resultset, with table name of
NULL was returned, instead of all tables for
the given database.
(Bug#39561)
The SQLGetInfo() function returned 0 for
SQL_CATALOG_USAGE information.
(Bug#39560)
MyODBC Driver 5.1.5 was not able to connect if the connection
string parameters contained spaces or tab symbols. For example,
if the SERVER parameter was specified as
“SERVER= localhost” instead of
“SERVER=localhost” the following error message will
be displayed:
[MySQL][ODBC 5.1 Driver] Unknown MySQL server host ' localhost' (11001).
The pointer passed to the
SQLDriverConnect method to retrieve the
output connection string length was one greater than it should
have been due to the inclusion of the NULL terminator.
(Bug#38949)
Data-at-execution parameters were not supported during
positioned update. This meant updating a long text field with a
cursor update would erroneously set the value to null. This
would lead to the error Column 'column_name' cannot be
null while updating the database, even when
column_name had been assigned a valid nonnull
string.
(Bug#37649)
The SQLDriverConnect method truncated
the OutputConnectionString parameter to 52
characters.
(Bug#37278)
The connection string option Enable
Auto-reconnect did not work. When the connection
failed, it could not be restored, and the errors generated were
the same as if the option had not been selected.
(Bug#37179)
Insertion of data into a LONGTEXT table field
did not work. If such an attempt was made the corresponding
field would be found to be empty on examination, or contain
random characters.
(Bug#36071)
No result record was returned for
SQLGetTypeInfo for the
TIMESTAMP data type. An application would
receive the result return code 100
(SQL_NO_DATA_FOUND).
(Bug#30626)
It was not possible to use Connector/ODBC to connect to a server using SSL. The following error was generated:
Runtime error '-2147467259 (80004005)': [MySQL][ODBC 3.51 Driver]SSL connection error.
When the recordSet.Update function was called
to update an adLongVarChar field, the field
was updated but the recordset was immediately lost. This
happened with driver cursors, whether the cursor was opened in
optimistic or pessimistic mode.
When the next update was called the test code would exit with the following error:
-2147467259 : Query-based update failed because the row to update could not be found.
Bugs fixed:
ODBC TIMESTAMP string format is
not handled properly by the MyODBC driver. When passing a
TIMESTAMP or
DATE to MyODBC, in the ODBC
format: {d <date>} or {ts <timestamp>}, the string
that represents this is copied once into the SQL statement, and
then added again, as an escaped string.
(Bug#37342)
The connector failed to prompt for additional information required to create a DSN-less connection from an application such as Microsoft Excel. (Bug#37254)
SQLDriverConnect does not return
SQL_NO_DATA on cancel. The ODBC documentation
specifies that this method should return
SQL_NO_DATA when the user cancels the dialog
to connect. The connector, however, returns
SQL_ERROR.
(Bug#36293)
Assigning a string longer than 67 characters to the
TableType parameter resulted in a buffer
overrun when the SQLTables() function was
called.
(Bug#36275)
The ODBC connector randomly uses logon information stored in
odbc-profile, or prompts the user for
connection information and ignores any settings stored in
odbc-profile.
(Bug#36203)
After having successfully established a connection, a crash
occurs when calling SQLProcedures()
followed by SQLFreeStmt(), using the ODBC C
API.
(Bug#36069)
Bugs fixed:
Wrong result obtained when using sum() on a
decimal(8,2) field type.
(Bug#35920)
The driver installer could not create a new DSN if many other drivers were already installed. (Bug#35776)
The SQLColAttribute() function returned
SQL_TRUE when querying the
SQL_DESC_FIXED_PREC_SCALE (SQL_COLUMN_MONEY)
attribute of a DECIMAL column.
Previously, the correct value of SQL_FALSE
was returned; this is now again the case.
(Bug#35581)
On Linux, SQLGetDiagRec() returned
SQL_SUCCESS in cases when it should have
returned SQL_NO_DATA.
(Bug#33910)
The driver crashes ODBC Administrator on attempting to add a new DSN. (Bug#32057)
Platform specific notes:
Important Change: You must uninstall previous 5.1.x editions of Connector/ODBC before installing the new version.
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
The installer for 64-bit Windows installs both the 32-bit and 64-bit driver. Please note that Microsoft does not yet supply a 64-bit bridge from ADO to ODBC.
Bugs fixed:
Important Change:
In previous versions, the SSL certificate would automatically be
verified when used as part of the Connector/ODBC connection. The
default mode is now to ignore the verificate of certificates. To
enforce verification of the SSL certificate during connection,
use the SSLVERIFY DSN parameter, setting the
value to 1.
(Bug#29955, Bug#34648)
Inserting characters to a UTF8 table using surrogate pairs would fail and insert invalid data. (Bug#34672)
Installation of Connector/ODBC would fail because it was unable to uninstall a previous installed version. The file being requested would match an older release version than any installed version of the connector. (Bug#34522)
Using SqlGetData in combination with
SQL_C_WCHAR would return overlapping data.
(Bug#34429)
Descriptor records were not cleared correctly when calling
SQLFreeStmt(SQL_UNBIND).
(Bug#34271)
The dropdown selection for databases on a server when creating a DSN was too small. The list size now automatically adjusts up to a maximum size of 20 potential databases. (Bug#33918)
Microsoft Access would be unable to use
DBEngine.RegisterDatabase to create a DSN
using the Connector/ODBC driver.
(Bug#33825)
Connector/ODBC erroneously reported that it supported the
CAST() and CONVERT() ODBC
functions for parsing values in SQL statements, which could lead
to bad SQL generation during a query.
(Bug#33808)
Using a linked table in Access 2003 where the table has a
BIGINT column as the first column
in the table, and is configured as the primary key, shows
#DELETED for all rows of the table.
(Bug#24535)
Updating a RecordSet when the query involves
a BLOB field would fail.
(Bug#19065)
MySQL Connector/ODBC 5.1.2-beta, a new version of the ODBC driver for the MySQL database management system, has been released. This release is the second beta (feature-complete) release of the new 5.1 series and is suitable for use with any MySQL server version since MySQL 4.1, including MySQL 5.0, 5.1, and 6.0. (It will not work with 4.0 or earlier releases.)
Keep in mind that this is a beta release, and as with any other pre-production release, caution should be taken when installing on production level systems or systems with critical data.
Platform specific notes:
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
The installer for 64-bit Windows installs both the 32-bit and 64-bit driver. Please note that Microsoft does not yet supply a 64-bit bridge from ADO to ODBC.
Due to differences with the installation process used on Windows and potential registry corruption, it is recommended that uninstall any existing versions of Connector/ODBC 5.1.x before upgrading.
See also Bug#34571.
Functionality added or changed:
Explicit descriptors are implemented. (Bug#32064)
A full implementation of SQLForeignKeys based on the information available from INFORMATION_SCHEMA in 5.0 and later versions of the server has been implemented.
Changed SQL_ATTR_PARAMSET_SIZE to return an
error until support for it is implemented.
Disabled MYSQL_OPT_SSL_VERIFY_SERVER_CERT
when using an SSL connection.
SQLForeignKeys uses
INFORMATION_SCHEMA when it is available on
the server, which allows more complete information to be
returned.
Bugs fixed:
The SSLCIPHER option would be incorrectly
recorded within the SSL configuration on Windows.
(Bug#33897)
Within the GUI interface, when connecting to a MySQL server on a nonstandard port, the connection test within the GUI would fail. The issue was related to incorrect parsing of numeric values within the DSN when the option was not configured as the last parameter within the DSN. (Bug#33822)
Specifying a nonexistent database name within the GUI dialog would result in an empty list, not an error. (Bug#33615)
When deleting rows from a static cursor, the cursor position would be incorrectly reported. (Bug#33388)
SQLGetInfo() reported characters for
SQL_SPECIAL_CHARACTERS that were not encoded
correctly.
(Bug#33130)
Retrieving data from a BLOB
column would fail within SQLGetDatawhen the
target data type was SQL_C_WCHAR due to
incorrect handling of the character buffer.
(Bug#32684)
Renaming an existing DSN entry would create a new entry with the new name without deleting the old entry. (Bug#31165)
Reading a TEXT column that had
been used to store UTF8 data would result in the wrong
information being returned during a query.
(Bug#28617)
SQLForeignKeys would return an empty string
for the schema columns instead of NULL.
(Bug#19923)
When accessing column data,
FLAG_COLUMN_SIZE_S32 did not limit the octet
length or display size reported for fields, causing problems
with Microsoft Visual FoxPro.
The list of ODBC functions that could have caused failures in
Microsoft software when retrieving the length of
LONGBLOB or
LONGTEXT columns includes:
SQLColumns
SQLColAttribute
SQLColAttributes
SQLDescribeCol
SQLSpecialColumns (theoretically can
have the same problem)
Dynamic cursors on statements with parameters were not supported. (Bug#11846)
Evaluating a simple numeric expression when using the OLEDB for ODBC provider and ADO would return an error, instead of the result. (Bug#10128)
Adding or updating a row using SQLSetPos()
on a result set with aliased columns would fail.
(Bug#6157)
MySQL Connector/ODBC 5.1.1-beta, a new version of the ODBC driver for the MySQL database management system, has been released. This release is the first beta (feature-complete) release of the new 5.1 series and is suitable for use with any MySQL server version since MySQL 4.1, including MySQL 5.0, 5.1, and 6.0. (It will not work with 4.0 or earlier releases.)
Keep in mind that this is a beta release, and as with any other pre-production release, caution should be taken when installing on production level systems or systems with critical data.
Includes changes from Connector/ODBC 3.51.21 and 3.51.22.
Built using MySQL 5.0.52.
Platform specific notes:
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
The installer for 64-bit Windows installs both the 32-bit and 64-bit driver. Please note that Microsoft does not yet supply a 64-bit bridge from ADO to ODBC.
Due to differences with the installation process used on Windows and potential registry corruption, it is recommended that uninstall any existing versions of Connector/ODBC 5.1.x before upgrading.
See also Bug#34571.
Functionality added or changed:
Incompatible Change: Replaced myodbc3i (now myodbc-installer) with Connector/ODBC 5.0 version.
Incompatible Change: Removed monitor (myodbc3m) and dsn-editor (myodbc3c).
Incompatible Change:
Disallow SET NAMES in initial statement and
in executed statements.
A wrapper for the
SQLGetPrivateProfileStringW() function,
which is required for Unicode support, has been created. This
function is missing from the unixODBC driver manager.
(Bug#32685)
Added MSI installer for Windows 64-bit. (Bug#31510)
Implemented support for SQLCancel().
(Bug#15601)
Removed nonthreadsafe configuration of the driver. The driver is now always built against the threadsafe version of libmysql.
Implemented native Windows setup library
Replaced the internal library which handles creation and loading of DSN information. The new library, which was originally a part of Connector/ODBC 5.0, supports Unicode option values.
The Windows installer now places files in a subdirectory of the
Program Files directory instead of the
Windows system directory.
Bugs fixed:
The SET NAMES statement has been disabled
because it causes problems in the ODBC driver when determining
the current client character set.
(Bug#32596)
SQLDescribeColW returned UTF-8 column as
SQL_VARCHAR instead of
SQL_WVARCHAR.
(Bug#32161)
ADO was unable to open record set using dynamic cursor. (Bug#32014)
ADO applications would not open a RecordSet
that contained a DECIMAL field.
(Bug#31720)
Memory usage would increase considerably. (Bug#31115)
SQLSetPos with SQL_DELETE
advances dynamic cursor incorrectly.
(Bug#29765)
Using an ODBC prepared statement with bound columns would produce an empty result set when called immediately after inserting a row into a table. (Bug#29239)
ADO Not possible to update a client side cursor. (Bug#27961)
Recordset Update() fails when using
adUseClient cursor.
(Bug#26985)
Connector/ODBC would fail to connect to the server if the password contained certain characters, including the semicolon and other punctuation marks. (Bug#16178)
Fixed SQL_ATTR_PARAM_BIND_OFFSET, and fixed
row offsets to work with updatable cursors.
SQLSetConnectAttr() did not clear previous
errors, possibly confusing SQLError().
SQLError() incorrectly cleared the error
information, making it unavailable from subsequent calls to
SQLGetDiagRec().
NULL pointers passed to SQLGetInfo() could
result in a crash.
SQL_ODBC_SQL_CONFORMANCE was not handled by
SQLGetInfo().
SQLCopyDesc() did not correctly copy all
records.
Diagnostics were not correctly cleared on connection and environment handles.
This release is the first of the new 5.1 series and is suitable for use with any MySQL server version since MySQL 4.1, including MySQL 5.0, 5.1, and 6.0. (It will not work with 4.0 or earlier releases.)
Keep in mind that this is a alpha release, and as with any other pre-production release, caution should be taken when installing on production level systems or systems with critical data. Not all of the features planned for the final Connector/ODBC 5.1 release are implemented.
Functionality is based on Connector/ODBC 3.51.20.
Platform specific notes:
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
There are no installer packages for Microsoft Windows x64 Edition.
Due to differences with the installation process used on Windows and potential registry corruption, it is recommended that uninstall any existing versions of Connector/ODBC 5.1.x before upgrading.
See also Bug#34571.
Functionality added or changed:
Added support for Unicode functions
(SQLConnectW, etc).
Added descriptor support (SQLGetDescField,
SQLGetDescRec, etc).
Added support for SQL_C_WCHAR.
Development on Connector/ODBC 5.0.x has ceased. New features and functionality will be incorporated into Connector/ODBC 5.1. See Section 20.1.2.1, “Connector/ODBC Roadmap”.
Bugs fixed:
Functionality added or changed:
Added support for ODBC v2 statement options using attributes.
Driver now builds and is partially tested under Linux with the iODBC driver manager.
Bugs fixed:
Connection string parsing for DSN-less connections could fail to identify some parameters. (Bug#25316)
Updates of MEMO or
TEXT columns from within
Microsoft Access would fail.
(Bug#25263)
Transaction support has been added and tested. (Bug#25045)
Internal function, my_setpos_delete_ignore()
could cause a crash.
(Bug#22796)
Fixed occasional mis-handling of the
SQL_NUMERIC_C type.
Fixed the binding of certain integer types.
Connector/ODBC 5.0.10 is the sixth BETA release.
Functionality added or changed:
Significant performance improvement when retrieving large text
fields in pieces using SQLGetData() with a
buffer smaller than the whole data. Mainly used in Access when
fetching very large text fields.
(Bug#24876)
Added initial unicode support in data and metadata. (Bug#24837)
Added initial support for removing braces when calling stored procedures and retrieving result sets from procedure calls. (Bug#24485)
Added loose handling of retrieving some diagnostic data. (Bug#15782)
Added wide-string type info for
SQLGetTypeInfo().
Bugs fixed:
Connector/ODBC 5.0.9 is the fifth BETA release.
This is an implementation and testing release, and is not designed for use within a production environment.
Functionality added or changed:
Added support for column binding as SQL_NUMBERIC_STRUCT.
Added recognition of SQL_C_SHORT and
SQL_C_TINYINT as C types.
Bugs fixed:
Fixed wildcard handling of and listing of catalogs and tables in
SQLTables.
Added limit of display size when requested via
SQLColAttribute/SQL_DESC_DISPLAY_SIZE.
Fixed buffer length return for SQLDriverConnect.
ODBC v2 behaviour in driver now supports ODBC v3 date/time types (since DriverManager maps them).
Catch use of SQL_ATTR_PARAMSET_SIZE and
report error until we fully support.
Fixed statistics to fail if it couldn't be completed.
Corrected retrieval multiple field types bit and blob/text.
Fixed SQLGetData to clear the NULL indicator correctly during multiple calls.
Connector/ODBC 5.0.8 is the fourth BETA release.
This is an implementation and testing release, and is not designed for use within a production environment.
Functionality added or changed:
Also made SQL_DESC_NAME only fill in the name
if there was a data pointer given, otherwise just the length.
Fixed display size to be length if max length isn’t available.
Wildcards now support escaped chars and underscore matching (needed to link tables with underscores in access).
Bugs fixed:
Fixed binding using SQL_C_LONG.
Fixed using wrong pointer for
SQL_MAX_DRIVER_CONNECTIONS in
SQLGetInfo.
Set default return to SQL_SUCCESS if nothing
is done for SQLSpecialColumns.
Fixed MDiagnostic to use correct v2/v3 error codes.
Allow SQLDescribeCol to be called to retrieve the length of the column name, but not the name itself.
Length now used when handling bind parameter (needed in
particular for SQL_WCHAR) - this enables
updating char data in MS Access.
Updated retrieval of descriptor fields to use the right pointer types.
Fixed hanlding of numeric pointers in SQLColAttribute.
Fixed type returned for MYSQL_TYPE_LONG to
SQL_INTEGER instead of
SQL_TINYINT.
Fix size return from SQLDescribeCol.
Fixed string length to chars, not bytes, returned by SQLGetDiagRec.
Connector/ODBC 5.0.7 is the third BETA release.
This is an implementation and testing release, and is not designed for use within a production environment.
Functionality added or changed:
Added support for SQLStatistics to
MYODBCShell.
Improved trace/log.
Bugs fixed:
SQLBindParameter now handles SQL_C_DEFAULT.
Corrected incorrect column index within
SQLStatistics. Many more tables can now be
linked into MS Access.
Fixed SQLDescribeCol returning column name
length in bytes rather than chars.
Connector/ODBC 5.0.6 is the second BETA release.
This is an implementation and testing release, and is not designed for use within a production environment.
Features, limitations and notes on this release
Connector/ODBC supports both User and
System DSNs.
Installation is provided in the form of a standard Microsoft System Installer (MSI).
You no longer have to have Connector/ODBC 3.51 installed before installing this version.
Bugs fixed:
You no longer have to have Connector/ODBC 3.51 installed before installing this version.
Connector/ODBC supports both User and
System DSNs.
Installation is provided in the form of a standard Microsoft System Installer (MSI).
Connector/ODBC 5.0.5 is the first BETA release.
This is an implementation and testing release, and is not designed for use within a production environment.
You no longer have to have Connector/ODBC 3.51 installed before installing this version.
Bugs fixed:
You no longer have to have Connector/ODBC 3.51 installed before installing this version.
This is an implementation and testing release, and is not designed for use within a production environment.
Features, limitations and notes on this release:
The following ODBC API functions have been added in this release:
SQLBindParameter
SQLBindCol
Connector/ODBC 5.0.2 was an internal implementation and testing release.
Features, limitations and notes on this release:
Connector/ODBC 5.0 is Unicode aware.
Connector/ODBC is currently limited to basic applications. ADO applications and Microsoft Office are not supported.
Connector/ODBC must be used with a Driver Manager.
The following ODBC API functions are implemented:
SQLAllocHandle
SQLCloseCursor
SQLColAttribute
SQLColumns
SQLConnect
SQLCopyDesc
SQLDisconnect
SQLExecDirect
SQLExecute
SQLFetch
SQLFreeHandle
SQLFreeStmt
SQLGetConnectAttr
SQLGetData
SQLGetDescField
SQLGetDescRec
SQLGetDiagField
SQLGetDiagRec
SQLGetEnvAttr
SQLGetFunctions
SQLGetStmtAttr
SQLGetTypeInfo
SQLNumResultCols
SQLPrepare
SQLRowcount
SQLTables
The following ODBC API function are implemented, but not yet support all the available attributes/options:
SQLSetConnectAttr
SQLSetDescField
SQLSetDescRec
SQLSetEnvAttr
SQLSetStmtAttr
Bugs fixed:
The client program hung when the network connection to the server was interrupted. (Bug#40407)
The connection string option Enable
Auto-reconnect did not work. When the connection
failed, it could not be restored, and the errors generated were
the same as if the option had not been selected.
(Bug#37179)
It was not possible to use Connector/ODBC to connect to a server using SSL. The following error was generated:
Runtime error '-2147467259 (80004005)': [MySQL][ODBC 3.51 Driver]SSL connection error.
Functionality added or changed:
There is a new connection option,
FLAG_NO_BINARY_RESULT. When set this option
disables charset 63 for columns with an empty
org_table.
(Bug#29402)
Bugs fixed:
When an ADOConnection is created and
attempts to open a schema with
ADOConnection.OpenSchema an access
violation occurs in myodbc3.dll.
(Bug#30770)
When SHOW CREATE TABLE was
invoked and then the field values read, the result was truncated
and unusable if the table had many rows and indexes.
(Bug#24131)
Bugs fixed:
The SQLColAttribute() function returned
SQL_TRUE when querying the
SQL_DESC_FIXED_PREC_SCALE (SQL_COLUMN_MONEY)
attribute of a DECIMAL column.
Previously, the correct value of SQL_FALSE
was returned; this is now again the case.
(Bug#35581)
The driver crashes ODBC Administrator on attempting to add a new DSN. (Bug#32057)
When accessing column data,
FLAG_COLUMN_SIZE_S32 did not limit the octet
length or display size reported for fields, causing problems
with Microsoft Visual FoxPro.
The list of ODBC functions that could have caused failures in
Microsoft software when retrieving the length of
LONGBLOB or
LONGTEXT columns includes:
SQLColumns
SQLColAttribute
SQLColAttributes
SQLDescribeCol
SQLSpecialColumns (theoretically can
have the same problem)
Bugs fixed:
Security Enhancement:
Accessing a parameer with the type of
SQL_C_CHAR, but with a numeric type and a
length of zero, the parameter marker would get stropped from the
query. In addition, an SQL injection was possible if the
parameter value had a nonzero length and was not numeric, the
text would be inserted verbatim.
(Bug#34575)
Important Change:
In previous versions, the SSL certificate would automatically be
verified when used as part of the Connector/ODBC connection. The
default mode is now to ignore the verificate of certificates. To
enforce verification of the SSL certificate during connection,
use the SSLVERIFY DSN parameter, setting the
value to 1.
(Bug#29955, Bug#34648)
When using ADO, the count of parameters in a query would always return zero. (Bug#33298)
Using tables with a single quote or other nonstandard characters in the table or column names through ODBC would fail. (Bug#32989)
When using Crystal Reports, table and column names would be truncated to 21 characters, and truncated columns in tables where the truncated name was the duplicated would lead to only a single column being displayed. (Bug#32864)
SQLExtendedFetch() and
SQLFetchScroll() ignored the rowset size if
the Don't cache result DSN option was set.
(Bug#32420)
When using the ODBC SQL_TXN_READ_COMMITTED
option, 'dirty' records would be read from tables as if the
option had not been applied.
(Bug#31959)
When creating a System DSN using the ODBC Administrator on Mac OS X, a User DSN would be created instead. The root cause is a problem with the iODBC driver manager used on Mac OS X. The fix works around this issue.
ODBC Administrator may still be unable to register a System
DSN unless the /Library/ODBC/odbc.ini
file has the correct permissions. You should ensure that the
file is writable by the admin group.
Calling SQLFetch or
SQLFetchScroll would return negative data
lengths when using SQL_C_WCHAR.
(Bug#31220)
SQLSetParam() caused memory allocation errors
due to driver manager's mapping of deprecated functions (buffer
length -1).
(Bug#29871)
Static cursor was unable to be used through ADO when dynamic cursors were enabled. (Bug#27351)
Using connection.Execute to create a record
set based on a table without declaring the cmd option as
adCmdTable will fail when communicating with
versions of MySQL 5.0.37 and higher. The issue is related to the
way that SQLSTATE is returned when ADO tries
to confirm the existence of the target object.
(Bug#27158)
Updating a RecordSet when the query involves
a BLOB field would fail.
(Bug#19065)
With some connections to MySQL databases using Connector/ODBC, the connection would mistakenly report 'user cancelled' for accesses to the database information. (Bug#16653)
Platform specific notes:
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
There are no installer packages for Microsoft Windows x64 Edition.
Bugs fixed:
Connector/ODBC would incorrectly return
SQL_SUCCESS when checking for distributed
transaction support.
(Bug#32727)
When using unixODBC or directly linked applications where the
thread level is set to less than 3 (within
odbcinst.ini), a thread synchronization
issue would lead to an application crash. This was because
SQLAllocStmt() and
SQLFreeStmt() did not synchronize access to
the list of statements associated with a connection.
(Bug#32587)
Cleaning up environment handles in multithread environments could result in a five (or more) second delay. (Bug#32366)
Renaming an existing DSN entry would create a new entry with the new name without deleting the old entry. (Bug#31165)
Setting the default database using the
DefaultDatabase property of an ADO
Connection object would fail with the error
Provider does not support this property. The
SQLGetInfo() returned the wrong value for
SQL_DATABASE_NAME when no database was
selected.
(Bug#3780)
Functionality added or changed:
The workaround for this bug was removed due to the fixes in MySQL Server 5.0.48 and 5.1.21.
This regression was introduced by Bug#10491.
Bugs fixed:
The English locale would be used when
formatting floating point values. The C
locale is now used for these values.
(Bug#32294)
When accessing information about supported operations, the
driver would return incorrect information about the support for
UNION.
(Bug#32253)
Unsigned integer values greater than the maximum value of a signed integer would be handled incorrectly. (Bug#32171)
The wrong result was returned by SQLGetData()
when the data was an empty string and a zero-sized buffer was
specified.
(Bug#30958)
Added the FLAG_COLUMN_SIZE_S32 option to
limit the reported column size to a signed 32-bit integer. This
option is automatically enabled for ADO applications to provide
a work around for a bug in ADO.
(Bug#13776)
Bugs fixed:
When using a rowset/cursor and add a new row with a number of
fields, subsequent rows with fewer fields will include the
original fields from the previous row in the final
INSERT statement.
(Bug#31246)
Uninitiated memory could be used when C/ODBC internally calls
SQLGetFunctions().
(Bug#31055)
The wrong SQL_DESC_LITERAL_PREFIX would be
returned for date/time types.
(Bug#31009)
The wrong COLUMN_SIZE would be returned by
SQLGetTypeInfo for the TIME columns
(SQL_TYPE_TIME).
(Bug#30939)
Clicking outside the character set selection box when configuring a new DSN could cause the wrong character set to be selected. (Bug#30568)
Not specifying a user in the DSN dialog would raise a warning even though the parameter is optional. (Bug#30499)
SQLSetParam() caused memory allocation errors
due to driver manager's mapping of deprecated functions (buffer
length -1).
(Bug#29871)
When using ADO, a column marked as
AUTO_INCREMENT could incorrectly report that
the column allowed NULL values. This was dur
to an issue with NULLABLE and
IS_NULLABLE return values from the call to
SQLColumns().
(Bug#26108)
Connector/ODBC would return the wrong the error code when the
server disconnects the active connection because the configured
wait_timeout has expired.
Previously it would return HY000.
Connector/ODBC now correctly returns an
SQLSTATE of 08S01.
(Bug#3456)
Bugs fixed:
Using FLAG_NO_PROMPT doesn't suppress the
dialogs normally handled by SQLDriverConnect.
(Bug#30840)
The specified length of the user name and authentication
parameters to SQLConnect() were not being
honored.
(Bug#30774)
The wrong column size was returned for binary data. (Bug#30547)
SQLGetData() will now always return
SQL_NO_DATA_FOUND on second call when no data
left, even if requested size is 0.
(Bug#30520)
SQLGetConnectAttr() did not reflect the
connection state correctly.
(Bug#14639)
Removed checkbox in setup dialog for
FLAG_FIELD_LENGTH (identified as
Don't Optimize Column Width within the GUI
dialog), which was removed from the driver in 3.51.18.
Connector/ODBC 3.51.19 fixes a specific issue with the 3.51.18 release. For a list of changes in the 3.51.18 release, see Section C.4.28, “Changes in MySQL Connector/ODBC 3.51.18 (08 August 2007)”.
Functionality added or changed:
Because of Bug#10491 in the server, character string results
were sometimes incorrectly identified as
SQL_VARBINARY. Until this server bug is
corrected, the driver will identify all variable-length strings
as SQL_VARCHAR.
Platform specific notes:
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
Binary packages for Sun Solaris are now available as
PKG packages.
Binary packages as disk images with installers are now available for Mac OS X.
A binary package without an installer is available for Microsoft Windows x64 Edition. There are no installer packages for Microsoft Windows x64 Edition.
Functionality added or changed:
Incompatible Change:
The FLAG_DEBUG option was removed.
When connecting to a specific database when using a DSN, the
system tables from the mysql database are no
longer also available. Previously, tables from the mysql
database (catalog) were listed as SYSTEM
TABLES by SQLTables() even when a
different catalog was being queried.
(Bug#28662)
Installed for Mac OS X has been re-instated. The installer registers the driver at a system (not user) level and makes it possible to create both user and system DSNs using the Connector/ODBC driver. The installer also fixes the situation where the necessary drivers would bge installed local to the user, not globally. (Bug#15326, Bug#10444)
Connector/ODBC now supports batched statements. In order to
enable cached statement support you must switch enable the
batched statement option
(FLAG_MULTI_STATEMENTS, 67108864, or
Allow multiple statements within a GUI
configuration). Be aware that batched statements create an
increased chance of SQL injection attacks and you must ensure
that your application protects against this scenario.
(Bug#7445)
The SQL_ATTR_ROW_BIND_OFFSET_PTR is now
supported for row bind offsets.
(Bug#6741)
The TRACE and TRACEFILE
DSN options have been removed. Use the ODBC driver manager trace
options instead.
Bugs fixed:
When using a table with multiple
TIMESTAMP columns, the final
TIMESTAMP column within the table
definition would not be updateable. Note that there is still a
limitation in MySQL server regarding multiple
TIMESTAMP columns . (Bug#9927)
(Bug#30081)
Fixed an issue where the myodbc3i would
update the user ODBC configuration file
(~/Library/ODBC/odbcinst.ini) instead of
the system /Library/ODBC/odbcinst.ini. This
was caused because myodbc3i was not honoring
the s and u modifiers for
the -d command-line option.
(Bug#29964)
Getting table metadata (through the
SQLColumns() would fail, returning a bad
table definition to calling applications.
(Bug#29888)
DATETIME column types would
return FALSE in place of
SQL_SUCCESS when requesting the column type
information.
(Bug#28657)
The SQL_COLUMN_TYPE,
SQL_COLUMN_DISPLAY and
SQL_COLUMN_PRECISION values would be returned
incorrectly by SQLColumns(),
SQLDescribeCol() and
SQLColAttribute() when accessing character
columns, especially those generated through
concat(). The lengths returned should now
conform to the ODBC specification. The
FLAG_FIELD_LENGTH option no longer has any
affect on the results returned.
(Bug#27862)
Obtaining the length of a column when using a character set for
the connection of utf8 would result in the
length being returned incorrectly.
(Bug#19345)
The SQLColumns() function could return
incorrect information about
TIMESTAMP columns, indicating
that the field was not nullable.
(Bug#14414)
The SQLColumns() function could return
incorrect information about AUTO_INCREMENT
columns, indicating that the field was not nullable.
(Bug#14407)
A binary package without an installer is available for Microsoft Windows x64 Edition. There are no installer packages for Microsoft Windows x64 Edition.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
BIT(n) columns are now treated as
SQL_BIT data where n = 1
and binary data where n > 1.
The wrong value from SQL_DESC_LITERAL_SUFFIX
was returned for binary fields.
The SQL_DATETIME_SUB column in SQLColumns()
was not correctly set for date and time types.
The value for SQL_DESC_FIXED_PREC_SCALE was
not returned correctly for values in MySQL 5.0 and later.
The wrong value for SQL_DESC_TYPE was
returned for date and time types.
SQLConnect() and
SQLDriverConnect() were rewritten to
eliminate duplicate code and ensure all options were supported
using both connection methods.
SQLDriverConnect() now only requires the
setup library to be present when the call requires it.
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
Binary packages as disk images with installers are now available for Mac OS X.
Binary packages for Sun Solaris are now available as
PKG packages.
The wrong value for DECIMAL_DIGITS in
SQLColumns() was reported for
FLOAT and
DOUBLE fields, as well as the
wrong value for the scale parameter to
SQLDescribeCol(), and the
SQL_DESC_SCALE attribute from
SQLColAttribute().
The SQL_DATA_TYPE column in
SQLColumns() results did not report the
correct value for date and time types.
Platform specific notes:
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
Binary packages for Sun Solaris are now available as
PKG packages.
Binary packages as disk images with installers are now available for Mac OS X.
A binary package without an installer is available for Microsoft Windows x64 Edition. There are no installer packages for Microsoft Windows x64 Edition.
Functionality added or changed:
It is now possible to specify a different character set as part
of the DSN or connection string. This must be used instead of
the SET NAMES statement. You can also
configure the character set value from the GUI configuration.
(Bug#9498, Bug#6667)
Fixed calling convention ptr and wrong free in myodbc3i, and fixed the null terminating (was only one, not two) when writing DSN to string.
Dis-allow NULL ptr for null indicator when calling SQLGetData() if value is null. Now returns SQL_ERROR w/state 22002.
The setup library has been split into its own RPM package, to allow installing the driver itself with no GUI dependencies.
Bugs fixed:
myodbc3i did not correctly format driver
info, which could cause the installation to fail.
(Bug#29709)
Connector/ODBC crashed with Crystal Reports due to a rproblem
with SQLProcedures().
(Bug#28316)
Fixed a problem where the GUI would crash when configuring or removing a System or User DSN. (Bug#27315)
Fixed error handling of out-of-memory and bad connections in catalog functions. This might raise errors in code paths that had ignored them in the past. (Bug#26934)
For a stored procedure that returns multiple result sets, Connector/ODBC returned only the first result set. (Bug#16817)
Calling SQLGetDiagField with
RecNumber 0, DiagIdentifier NOT 0 returned
SQL_ERROR, preventing access to diagnostic
header fields.
(Bug#16224)
Added a new DSN option
(FLAG_ZERO_DATE_TO_MIN) to retrieve
XXXX-00-00 dates as the minimum allowed ODBC
date (XXXX-01-01). Added another option
(FLAG_MIN_DATE_TO_ZERO) to mirror this but
for bound parameters. FLAG_MIN_DATE_TO_ZERO
only changes 0000-01-01 to
0000-00-00.
(Bug#13766)
If there was more than one unique key on a table, the correct
fields were not used in handling SQLSetPos().
(Bug#10563)
When inserting a large BLOB
field, Connector/ODBC would crash due to a memory allocation
error.
(Bug#10562)
The driver was using
mysql_odbc_escape_string(), which does not
handle the
NO_BACKSLASH_ESCAPES SQL mode.
Now it uses
mysql_real_escape_string(),
which does.
(Bug#9498)
SQLColumns() did not handle many of its
parameters correctly, which could lead to incorrect results. The
table name argument was not handled as a pattern value, and most
arguments were not escaped correctly when they contained
nonalphanumeric characters.
(Bug#8860)
There are no binary packages for Microsoft Windows x64 Edition.
There is no binary package for Mac OS X on 64-bit PowerPC because Apple does not currently provide a 64-bit PowerPC version of iODBC.
Correctly return error if SQLBindCol is
called with an invalid column.
Fixed possible crash if SQLBindCol() was not
called before SQLSetPos().
The Mac OS X binary packages are only provided as tarballs, there is no installer.
The binary packages for Sun Solaris are only provided as tarballs, not the PKG format.
The HP-UX 11.23 IA64 binary package does not include the GUI bits because of problems building Qt on that platform.
Functionality added or changed:
Connector/ODBC now supports using SSL for communication. This is not yet exposed in the setup GUI, but must be enabled through configuration files or the DSN. (Bug#12918)
Bugs fixed:
Calls to SQLNativeSql() could cause stack corruption due to an incorrect pointer cast. (Bug#28758)
Using curors on results sets with multi-column keys could select the wrong value. (Bug#28255)
SQLForeignKeys does not escape
_ and % in the table name
arguments.
(Bug#27723)
When using stored procedures, making a
SELECT or second stored procedure
call after an initial stored procedure call, the second
statement will fail.
(Bug#27544)
SQLTables() did not distinguish tables from views. (Bug#23031)
Data in TEXT columns would fail
to be read correctly.
(Bug#16917)
Specifying strings as parameters using the
adBSTR or adVarWChar
types, (SQL_WVARCHAR and
SQL_WLONGVARCHAR) would be incorrectly
quoted.
(Bug#16235)
SQL_WVARCHAR and SQL_WLONGVARCHAR parameters were not properly quoted and escaped. (Bug#16235)
Using BETWEEN with date values, the wrong
results could be returned.
(Bug#15773)
When using the Don't Cache Results (option
value 1048576) with Microsoft Access, the
connection will fail using DAO/VisualBasic.
(Bug#4657)
Return values from SQLTables() may be
truncated. (Bugs #22797)
Bugs fixed:
Connector/ODBC would incorrectly claim to support
SQLProcedureColumns (by returning true when
queried about SQLPROCEDURECOLUMNS with
SQLGetFunctions), but this functionality is
not supported.
(Bug#27591)
An incorrect transaction isolation level may not be returned when accessing the connection attributes. (Bug#27589)
Adding a new DSN with the myodbc3i utility
under AIX would fail.
(Bug#27220)
When inserting data using bulk statements (through
SQLBulkOperations), the indicators for all
rows within the insert would not updated correctly.
(Bug#24306)
Using SQLProcedures does not return the
database name within the returned resultset.
(Bug#23033)
The SQLTransact() function did not support an
empty connection handle.
(Bug#21588)
Using SQLDriverConnect instead of
SQLConnect could cause later operations to
fail.
(Bug#7912)
When using blobs and parameter replacement in a statement with
WHERE CURSOR OF, the SQL is truncated.
(Bug#5853)
Connector/ODBC would return too many foreign key results when accessing tables with similar names. (Bug#4518)
Functionality added or changed:
Use of SQL_ATTR_CONNECTION_TIMEOUT on the
server has now been disabled. If you attempt to set this
attribute on your connection the
SQL_SUCCESS_WITH_INFO will be returned, with
an error number/string of HYC00: Optional feature not
supported.
(Bug#19823)
Added auto is null option to Connector/ODBC option parameters. (Bug#10910)
Added auto-reconnect option to Connector/ODBC option parameters.
Added support for the HENV handlers in
SQLEndTran().
Bugs fixed:
On 64-bit systems, some types would be incorrectly returned. (Bug#26024)
When retrieving TIME columns,
C/ODBC would incorrectly interpret the type of the string and
could interpret it as a DATE type
instead.
(Bug#25846)
Connector/ODBC may insert the wrong parameter values when using prepared statements under 64-bit Linux. (Bug#22446)
Using Connector/ODBC, with SQLBindCol and
binding the length to the return value from
SQL_LEN_DATA_AT_EXEC fails with a memory
allocation error.
(Bug#20547)
Using DataAdapter, Connector/ODBC may
continually consume memory when reading the same records within
a loop (Windows Server 2003 SP1/SP2 only).
(Bug#20459)
When retrieving data from columns that have been compressed
using COMPRESS(), the retrieved data would be
truncated to 8KB.
(Bug#20208)
The ODBC driver name and version number were incorrectly reported by the driver. (Bug#19740)
A string format exception would be raised when using iODBC, Connector/ODBC and the embedded MySQL server. (Bug#16535)
The SQLDriverConnect() ODBC method did not
work with recent Connector/ODBC releases.
(Bug#12393)
Connector/ODBC 3.51.13 was an internal implementation and testing release.
Functionality added or changed:
N/A
Bugs fixed:
Bugs fixed:
mysql_list_dbcolumns() and
insert_fields() were retrieving all rows from
a table. Fixed the queries generated by these functions to
return no rows.
(Bug#8198)
SQLGetTypoInfo() returned
tinyblob for SQL_VARBINARY
and nothing for SQL_BINARY. Fixed to return
varbinary for
SQL_VARBINARY, binary for
SQL_BINARY, and longblob
for SQL_LONGVARBINARY.
(Bug#8138)
This is a new release, fixing recently discovered bugs.
Bugs fixed:
Adding the Allow Batch=False option to the
connection string caused MySQL Connector/NET to generate the error:
You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'SET character_set_results=NULL' at line 1
A MySQL Connector/NET test program that connected to MySQL Server using the
connection string option compress=true
crashed, but only when running on Mono. The program worked as
expected when running on Microsoft Windows.
This was due to a bug in Mono. MySQL Connector/NET was modified to avoid
using WeakReferences in the
Compressed stream class, which was causing
the crash.
(Bug#45463)
After a Reference to "C:\Program Files\MySQL\MySQL Connector Net 5.2.4\Compact Framework\MySql.Data.CF.dll" was added to a Windows Mobile 5.0 project, the project then failed to build, generating a Microsoft Visual C# compiler error.
The error generated was:
Error 2 The type 'System.Runtime.CompilerServices.CompilerGeneratedAttribute' has no constructors defined MysqlTest Error 3 Internal Compiler Error (0xc0000005 at address 5A7E3714): likely culprit is 'COMPILE'.
This is the first post-GA release, fixing recently discovered bugs.
Bugs fixed:
If a certain socket exception occurred when trying to establish a MySQL database connection, MySQL Connector/NET displayed an exception message that appeared to be unrelated to the underlying problem. This masked the problem and made diagnosing problems more difficult.
For example, if, when establishing a database connection via TCP/IP, Windows on the local machine allocated an ephemeral port that conflicted with a socket address still in use, then Windows/.NET would throw a socket exception with the following error text:
Only one usage of each socket address (protocol/network address/port) is normally
permitted IP ADDRESS/PORT.However, MySQL Connector/NET masked this socket exception and displayed an exception with the following text:
Unable to connect to any of the specified MySQL hosts.
A SQL query string containing an escaped backslash caused an exception to be generated:
Index and length must refer to a location within the string. Parameter name: length at System.String.InternalSubStringWithChecks(Int32 startIndex, Int32 length, Boolean fAlwaysCopy) at MySql.Data.MySqlClient.MySqlTokenizer.NextParameter() at MySql.Data.MySqlClient.Statement.InternalBindParameters(String sql, MySqlParameterCollection parameters, MySqlPacket packet) at MySql.Data.MySqlClient.Statement.BindParameters() at MySql.Data.MySqlClient.PreparableStatement.Execute() at MySql.Data.MySqlClient.MySqlCommand.ExecuteReader(CommandBehavior behavior) at MySql.Data.MySqlClient.MySqlCommand.ExecuteNonQuery()
The Microsoft Visual Studio solution file
MySQL-VS2005.sln was invalid. Several
projects could not be loaded and thus it was not possible to
build MySQL Connector/NET from source.
(Bug#44822)
The Data Set editor generated an error when attempts were made to modify insert, update or delete commands:
Error in WHERE clause near '@'. Unable to parse query text.
The DataReader in MySQL Connector/NET 6.0.3 considered a BINARY(16) field as a GUID with a length of 16. (Bug#44507)
When creating a new DataSet the following error was generated:
Failed to open a connection to database. Cannot load type with name 'MySQL.Data.VisualStudio.StoredProcedureColumnEnumerator'
The MySQL Connector/NET MySQLRoleProvider reported that there were no roles, even when roles existed. (Bug#44414)
MySQL Connector/NET was missing the capability to validate the server's certificate when using encryption. This made it possible to conduct a man-in-the-middle attack against the connection, which defeated the security provided by SSL. (Bug#38700)
First GA release.
Functionality added or changed:
The MySqlTokenizer failed to split fieldnames
from values if they were not separated by a space. This also
happened if the string contained certain characters. As a result
MySqlCommand.ExecuteNonQuery raised an index
out of range exception.
The resulting errors are illustrated by the following examples.
Note, the example statements do not have delimiting spaces
around the = operator.
INSERT INTO anytable SET Text='test--test';
The tokenizer incorrectly interpreted the value as containing a comment.
UPDATE anytable SET Project='123-456',Text='Can you explain this ?',Duration=15 WHERE ID=4711;'
A MySqlException was generated, as the
? in the value was interpreted by the
tokenizer as a parameter sign. The error message generated was:
Fatal error encountered during command execution. EXCEPTION: MySqlException - Parameter '?'' must be defined.
Bugs fixed:
MySQL.Data was not displayed as a Reference
inside Microsoft Visual Studio 2008 Professional.
When a new C# project was created in Microsoft Visual Studio
2008 Professional, MySQL.Data was not
displayed when , was selected.
(Bug#44141)
Column types for SchemaProvider and
ISSchemaProvider did not match.
When the source code in SchemaProvider.cs
and ISSchemaProvider.cs were compared it
was apparent that they were not using the same column types. The
base provider used SQL such as SHOW CREATE
TABLE, while ISSchemaProvider used
the schema information tables. Column types used by the base
class were INT64 and the column types used by
ISSchemaProvider were
UNSIGNED.
(Bug#44123)
This is a new development release, fixing recently discovered bugs.
Bugs fixed:
MySQL Connector/NET 6.0.1 did not load in Microsoft Visual Studio 2008 and Visual Studio 2005 Pro.
The following error message was generated:
.NET Framework Data Provider for MySQL: The data provider object factory service was not found.
This is a new Beta development release, fixing recently discovered bugs.
Bugs fixed:
An insert and update error was generated by the decimal data type in the Entity Framework, when a German collation was used. (Bug#43574)
Generating an Entity Data Model (EDM) schema with a table
containing columns with data types MEDIUMTEXT
and LONGTEXT generated a runtime error
message “Max value too long or too short for
Int32”.
(Bug#43480)
This is a new Alpha development release.
Bugs fixed:
A null reference exception was generated when
MySqlConnection.ClearPool(connection) was
called.
(Bug#42801)
Bugs fixed:
The Web Provider did not work at all on a remote host, and did
not create a database when using
autogenerateschema="true".
(Bug#39072)
The MySQL Connector/NET installer program ended prematurely without reporting the specific error. (Bug#39019)
When called with an incorrect password the
MembershipProvider.GetPassword() method
threw a
MySQLException
instead of a
MembershipPasswordException
.
(Bug#38939)
Possible overflow in
MySqlPacket.ReadLong().
(Bug#36997)
The TokenizeSql method was adding query
overhead and causing high CPU utilization for larger queries.
(Bug#36836)
Bugs fixed:
A MySQL Connector/NET test program that connected to MySQL Server using the
connection string option compress=true
crashed, but only when running on Mono. The program worked as
expected when running on Microsoft Windows.
This was due to a bug in Mono. MySQL Connector/NET was modified to avoid
using WeakReferences in the
Compressed stream class, which was causing
the crash.
(Bug#45463)
If a certain socket exception occurred when trying to establish a MySQL database connection, MySQL Connector/NET displayed an exception message that appeared to be unrelated to the underlying problem. This masked the problem and made diagnosing problems more difficult.
For example, if, when establishing a database connection via TCP/IP, Windows on the local machine allocated an ephemeral port that conflicted with a socket address still in use, then Windows/.NET would throw a socket exception with the following error text:
Only one usage of each socket address (protocol/network address/port) is normally
permitted IP ADDRESS/PORT.However, MySQL Connector/NET masked this socket exception and displayed an exception with the following text:
Unable to connect to any of the specified MySQL hosts.
The Microsoft Visual Studio solution file
MySQL-VS2005.sln was invalid. Several
projects could not be loaded and thus it was not possible to
build MySQL Connector/NET from source.
(Bug#44822)
The MySQL Connector/NET MySQLRoleProvider reported that there were no roles, even when roles existed. (Bug#44414)
After a Reference to "C:\Program Files\MySQL\MySQL Connector Net 5.2.4\Compact Framework\MySql.Data.CF.dll" was added to a Windows Mobile 5.0 project, the project then failed to build, generating a Microsoft Visual C# compiler error.
The error generated was:
Error 2 The type 'System.Runtime.CompilerServices.CompilerGeneratedAttribute' has no constructors defined MysqlTest Error 3 Internal Compiler Error (0xc0000005 at address 5A7E3714): likely culprit is 'COMPILE'.
When a TableAdapter was created on a DataSet, it was not possible to use a stored procedure with variables. The following error was generated:
The method or operation is not implemented
Functionality added or changed:
A new connection string option has been added: use
affected rows. When true the
connection will report changed rows instead of found rows.
(Bug#44194)
Bugs fixed:
Calling GetSchema() on
Indexes or IndexColumns
failed where index or column names were restricted.
In SchemaProvider.cs, methods
GetIndexes() and
GetIndexColumns() passed their restrictions
directly to GetTables(). This only worked if
the restrictions were no more specific than
schemaName and tableName.
If IndexName was given, this was passed to
GetTables() where it was treated as
TableType. As a result no tables were
returned, unless the index name happened to be BASE
TABLE or VIEW. This meant that both
methods failed to return any rows.
(Bug#43991)
GetSchema("MetaDataCollections") should have
returned a table with a column named
“NumberOfRestrictions” not
“NumberOfRestriction”.
This can be confirmed by referencing the Microsoft Documentation. (Bug#43990)
Requests sent to the MySQL Connector/NET role provider to remove a user from
a role failed. The query log showed the query was correctly
executed within a transaction which was immediately rolled back.
The rollback was caused by a missing call to the
Complete method of the transaction.
(Bug#43553)
When using MySqlBulkLoader.Load(), the text
file is opened by
NativeDriver.SendFileToServer. If it
encountered a problem opening the file as a stream, an exception
was generated and caught. An attempt to clean up resources was
then made in the finally{} clause by calling
fs.Close(), but since the stream was never
successfully opened, this was an attempt to execute a method of
a null reference.
(Bug#43332)
A null reference exception was generated when
MySqlConnection.ClearPool(connection) was
called.
(Bug#42801)
MySQLMembershipProvider.ValidateUser only
used the userId to validate. However, it
should also use the applicationId to perform
the validation correctly.
The generated query was, for example:
SELECT Password, PasswordKey, PasswordFormat, IsApproved, Islockedout FROM my_aspnet_Membership WHERE userId=13
Note that applicationId is not used.
(Bug#42574)
There was an error in the ProfileProvider
class in the private ProfileInfoCollection
GetProfiles() function. The column of the final table
was named “lastUpdatdDate” ('e' is
missing) instead of the correct “lastUpdatedDate”.
(Bug#41654)
The GetGuid() method of
MySqlDataReader did not treat
BINARY(16) column data as a GUID. When
operating on such a column a FormatException
exception was generated.
(Bug#41452)
When ASP.NET membership was configured to not require password
question and answer using
requiresQuestionAndAnswer="false", a
SqlNullValueException was generated when
using MembershipUser.ResetPassword() to reset
the user password.
(Bug#41408)
If a Stored Procedure contained spaces in its
parameter list, and was then called from MySQL Connector/NET, an exception
was generated. However, the same Stored
Procedure called from the MySQL Query Analyzer or the
MySQL Client worked correctly.
The exception generated was:
Parameter '0' not found in the collection.
The DATETIME format contained an erroneous
space.
(Bug#41021)
When MySql.Web.Profile.MySQLProfileProvider
was configured, it was not possible to assign a name other than
the default name MySQLProfileProvider.
If the name SCC_MySQLProfileProvider was
assigned, an exception was generated when attempting to use
Page.Context.Profile['custom prop'].
The exception generated was:
The profile default provider was not found.
Note that the exception stated: 'the profile default provider...', even though a different name was explicitly requested. (Bug#40871)
When ExecuteNonQuery was called with a
command type of Stored Procedure it worked
for one user but resulted in a hang for another user with the
same database permissions.
However, if CALL was used in the command text
and ExecuteNonQuery was used with a command
type of Text, the call worked for both users.
(Bug#40139)
Bugs fixed:
Visual Studio 2008 displayed the following error three times on start-up:
"Package Load Failure
Package 'MySql.Data.VisualStudio.MySqlDataProviderPackage, MySql.VisualStudio,
Version=5.2.4, Culture=neutral, PublicKeyTopen=null' has failed to load properly (GUID =
{79A115C9-B133-4891-9E7B-242509DAD272}). Please contact the package vendor for
assistance. Application restart is recommended, due to possible environment corruption.
Would you like to disable loading the package in the future? You may use
'devenve/resetskippkgs' to re-enable package loading."
Bugs fixed:
MySqlDataReader did not feature a
GetSByte method.
(Bug#40571)
When working with stored procedures MySQL Connector/NET generated an
exception Unknown "table parameters" in
information_schema.
(Bug#40382)
GetDefaultCollation and
GetMaxLength were not thread safe. These
functions called the database to get a set of parameters and
cached them in two static dictionaries in the function
InitCollections. However, if many threads
called them they would try to insert the same keys in the
collections resulting in duplicate key exceptions.
(Bug#40231)
If connection pooling was not set explicitly in the connection
string, MySQL Connector/NET added “;Pooling=False” to the end of
the connection string when
MySqlCommand.ExecuteReader() was called.
If connection pooling was explicitly set in the connection
string, when MySqlConnection.Open() was
called it converted “Pooling=True” to
“pooling=True”.
If MySqlCommand.ExecuteReader() was
subsequently called, it concatenated
“;Pooling=False” to the end of the connection
string. The resulting connection string was thus terminated with
“pooling=True;Pooling=False”. This disabled
connection pooling completely.
(Bug#40091)
The connection string option Functions Return
String did not set the correct encoding for the result
string. Even though the connection string option
Functions Return String=true; is set, the
result of SELECT DES_DECRYPT() contained
“??” instead of the correct national character
symbols.
(Bug#40076)
If, when using the MySqlTransaction
transaction object, an exception was thrown, the transaction
object was not disposed of and the transaction was not rolled
back.
(Bug#39817)
After the ConnectionString property was
initialized via the public setter of
DbConnectionStringBuilder, the
GetConnectionString method of
MySqlConnectionStringBuilder incorrectly
returned null when true
was assigned to the includePass parameter.
(Bug#39728)
When using ProfileProvider, attempting to
update a previously saved property failed.
(Bug#39330)
Reading a negative time value greater than -01:00:00 returned the absolute value of the original time value. (Bug#39294)
Inserting a negative time value (negative
TimeSpan) into a Time
column through the use of MySqlParameter
caused
MySqlException
to be thrown.
(Bug#39275)
When a data connection was created in the server explorer of Visual Studio 2008 Team, an error was generated when trying to expand stored procedures that had parameters.
Also, if TableAdapter was right-clicked and then , , selected, if you then attempted to select a stored procedure, the window would close and no error message would be displayed. (Bug#39252)
The Web Provider did not work at all on a remote host, and did
not create a database when using
autogenerateschema="true".
(Bug#39072)
MySQL Connector/NET called hashed password methods not supported in Mono 2.0 Preview 2. (Bug#38895)
Functionality added or changed:
Error string was returned after a 28000 second
wait_timeout. This has been
changed to generate a ConnectionState.Closed
event.
(Bug#38119)
Changed how the procedure schema collection is retrieved. If the
connection string contains “use procedure
bodies=true” then a
SELECT is performed on the
mysql.proc table directly, as this is up to
50 times faster than the current Information Schema
implementation. If the connection string contains
“use procedure bodies=false”,
then the Information Schema collection is queried.
(Bug#36694)
Changed how the procedure schema collection is retrieved. If
use procedure bodies=true then the
mysql.proc table is selected directly as this
is up to 50 times faster than the current
information_schema implementation. If
use procedure bodies=false, then the
information_schema collection is queried.
(Bug#36694)
String escaping functionality has been moved from the
MySqlString class to the
MySqlHelper class, where it can be
accessed by the EscapeString method.
(Bug#36205)
Bugs fixed:
The GetOrdinal() method failed to
return the ordinal if the column name string contained an
accent.
(Bug#38721)
MySQL Connector/NET uninstaller did not clean up all installed files. (Bug#38534)
There was a short circuit evaluation error in the
MySqlCommand.CheckState() method. When
the statement connection == null was true a
NullReferenceException was thrown and not
the expected InvalidOperationException.
(Bug#38276)
The provider did not silently create the user if the user did not exist. (Bug#38243)
Executing a command that resulted in a fatal exception did not close the connection. (Bug#37991)
When a prepared insert query is run that contains an
UNSIGNED TINYINT in the parameter list, the
complete query and data that should be inserted is corrupted and
no error is thrown.
(Bug#37968)
In a .NET application MySQL Connector/NET modifies the connection string so that it contains several occurrences of the same option with different values. This is illustrated by the example that follows.
The original connection string:
host=localhost;database=test;uid=*****;pwd=*****; connect timeout=25; auto enlist=false;pooling=false;
The connection string after after closing
MySqlDataReader:
host=localhost;database=test;uid=*****;pwd=*****; connect timeout=25;auto enlist=false;pooling=false; Allow User Variables=True;Allow User Variables=False; Allow User Variables=True;Allow User Variables=False;
Unnecessary network traffic was generated for the normal case where the web provider schema was up to date. (Bug#37469)
MySqlReader.GetOrdinal() performance
enhancements break existing functionality.
(Bug#37239)
The autogenerateschema option produced tables
with incorrect collations.
(Bug#36444)
GetSchema did not work correctly when
querying for a collection, if using a non-English locale.
(Bug#35459)
When reading back a stored double or single value using the .NET provider, the value had less precision than the one stored. (Bug#33322)
Using the MySQL Visual Studio plugin and a MySQL 4.1 server,
certain field types (ENUM) would
not be identified correctly. Also, when looking for tables, the
plugin would list all tables matching a wildcard pattern of the
database name supplied in the connection string, instead of only
tables within the specified database.
(Bug#30603)
Bugs fixed:
Product documentation incorrectly stated '?' is the preferred parameter marker. (Bug#37349)
An incorrect value for a bit field would returned in a multi-row
query if a preceding value for the field returned
NULL.
(Bug#36313)
Tables with GEOMETRY field types would return
an unknown datatype exception.
(Bug#36081)
When using the MySQLProfileProvider, setting
profile details and then reading back saved data would result in
the default values being returned instead of the updated values.
(Bug#36000)
When creating a connection, setting the
ConnectionString property of
MySqlConnection to NULL
would throw an exception.
(Bug#35619)
The DbCommandBuilder.QuoteIdentifer
method was not implemented.
(Bug#35492)
When using encrypted passwords, the
GetPassword() function would return the wrong
string.
(Bug#35336)
An error would be raised when calling
GetPassword() with a NULL
value.
(Bug#35332)
When retreiving data where a field has been identified as
containing a GUID value, the incorrect value would be returned
when a previous row contained a NULL value
for that field.
(Bug#35041)
Using the TableAdapter Wizard would fail when
generating commands that used stored procedures due to the
change in supported parameter characters.
(Bug#34941)
When creating a new stored procedured, the new parameter code
which allows the use of the @ symbol would
interfere with the specification of a
DEFINER.
(Bug#34940)
When using SqlDataSource to open a
connection, the connection would not automatically be closed
when access had completed.
(Bug#34460)
There was a high level of contention in the connection pooling code that could lead to delays when opening connections and submitting queries. The connection pooling code has been modified to try and limit the effects of the contention issue. (Bug#34001)
Using the TableAdaptor wizard in combination
with a suitable SELECT statement,
only the associated INSERT
statement would also be created, rather than the required
DELETE and
UPDATE statements.
(Bug#31338)
Fixed problem in datagrid code related to creating a new table. This problem may have been introduced with .NET 2.0 SP1.
Fixed profile provider that would throw an exception if you were updating a profile that already existed.
Bugs fixed:
When using the provider to generate or update users and passwords, the password checking algorithm would not validate the password strength or requirements correctly. (Bug#34792)
When executing statements that used stored procedures and functions, the new parameter code could fail to identify the correct parameter format. (Bug#34699)
The installer would fail to the DDEX provider binary if the Visual Studio 2005 component was not selected. The result would lead to MySQL Connector/NET not loading properly when using the interface to a MySQL server within Visual Studio. (Bug#34674)
A number issues were identified in the case, connection and
scema areas of the code for
MembershipProvider,
RoleProvider,
ProfileProvider.
(Bug#34495)
When using web providers, the MySQL Connector/NET would check the schema and cache the application id, even when the connection string had been set. The effect would be to break the memvership provider list. (Bug#34451)
Attempting to use an isolation level other than the default with a transaction scope would use the default isolation level. (Bug#34448)
When altering a stored procedure within Visual Studio, the parameters to the procedure could be lost. (Bug#34359)
A race condition could occur within the procedure cache resulting the cache contents overflowing beyond the configured cache size. (Bug#34338)
Fixed problem with Visual Studio 2008 integration that caused pop-up menus on server explorer nodes to not function
The provider code has been updated to fix a number of outstanding issues.
Functionality added or changed:
Performing GetValue() on a field
TINYINT(1) returned a
BOOLEAN. While not a bug, this
caused problems in software that expected an
INT to be returned. A new
connection string option Treat Tiny As
Boolean has been added with a default value of
true. If set to false the
provider will treat TINYINT(1) as
INT.
(Bug#34052)
Added support for DbDataAdapter
UpdateBatchSize. Batching is fully supported
including collapsing inserts down into the multi-value form if
possible.
DDEX provider now works under Visual Studio 2008 beta 2.
Added ClearPool and ClearAllPools features.
Bugs fixed:
Some speed improvements have been implemented in the
TokenizeSql process used to identify elements
of SQL statements.
(Bug#34220)
When accessing tables from different databases within the same
TransactionScope, the same user/password
combination would be used for each database connection. MySQL Connector/NET
does not handle multiple connections within the same transaction
scope. An error is now returned if you attempt this process,
instead of using the incorrect authorization information.
(Bug#34204)
The status of connections reported through the state change handler was not being updated correctly. (Bug#34082)
Incorporated some connection string cache optimizations sent to us by Maxim Mass. (Bug#34000)
In an open connection where the server had disconnected unexpectedly, the status information of the connection would not be updated properly. (Bug#33909)
Data cached from the connection string could return invalid information because the internal routines were not using case-sensitive semantics. This lead to updated connection string options not being recognized if they were of a different case than the existing cached values. (Bug#31433)
Column name metadata was not using the character set as deifned within the connection string being used. (Bug#31185)
Memory usage could increase and decrease significantly when updating or inserting a large number of rows. (Bug#31090)
Commands executed from within the state change handeler would
fail with a NULL exception.
(Bug#30964)
When running a stored procedure multiple times on the same connection, the memory usage could increase indefinitely. (Bug#30116)
Using compression in the MySQL connection with MySQL Connector/NET would be slower than using native (uncompressed) communication. (Bug#27865)
The MySqlDbType.Datetime has been replaced
with MySqlDbType.DateTime. The old format has
been obsoleted.
(Bug#26344)
Bugs fixed:
Calling GetSchema() on
Indexes or IndexColumns
failed where index or column names were restricted.
In SchemaProvider.cs, methods
GetIndexes() and
GetIndexColumns() passed their restrictions
directly to GetTables(). This only worked if
the restrictions were no more specific than
schemaName and tableName.
If IndexName was given, this was passed to
GetTables() where it was treated as
TableType. As a result no tables were
returned, unless the index name happened to be BASE
TABLE or VIEW. This meant that both
methods failed to return any rows.
(Bug#43991)
The DATETIME format contained an erroneous
space.
(Bug#41021)
If connection pooling was not set explicitly in the connection
string, MySQL Connector/NET added “;Pooling=False” to the end of
the connection string when
MySqlCommand.ExecuteReader() was called.
If connection pooling was explicitly set in the connection
string, when MySqlConnection.Open() was
called it converted “Pooling=True” to
“pooling=True”.
If MySqlCommand.ExecuteReader() was
subsequently called, it concatenated
“;Pooling=False” to the end of the connection
string. The resulting connection string was thus terminated with
“pooling=True;Pooling=False”. This disabled
connection pooling completely.
(Bug#40091)
If, when using the MySqlTransaction
transaction object, an exception was thrown, the transaction
object was not disposed of and the transaction was not rolled
back.
(Bug#39817)
When a prepared insert query is run that contains an
UNSIGNED TINYINT in the parameter list, the
complete query and data that should be inserted is corrupted and
no error is thrown.
(Bug#37968)
Bugs fixed:
There was a short circuit evaluation error in the
MySqlCommand.CheckState() method. When
the statement connection == null was true a
NullReferenceException was thrown and not
the expected InvalidOperationException.
(Bug#38276)
Executing a command that resulted in a fatal exception did not close the connection. (Bug#37991)
In a .NET application MySQL Connector/NET modifies the connection string so that it contains several occurrences of the same option with different values. This is illustrated by the example that follows.
The original connection string:
host=localhost;database=test;uid=*****;pwd=*****; connect timeout=25; auto enlist=false;pooling=false;
The connection string after after closing
MySqlDataReader:
host=localhost;database=test;uid=*****;pwd=*****; connect timeout=25;auto enlist=false;pooling=false; Allow User Variables=True;Allow User Variables=False; Allow User Variables=True;Allow User Variables=False;
As MySqlDbType.DateTime is not available
in VB.Net the warning The datetime
enum value is obsolete was always shown during
compilation.
(Bug#37406)
An unknown MySqlErrorCode was encountered
when opening a connection with an incorrect password.
(Bug#37398)
Documentation incorrectly stated that “the DataColumn class in .NET 1.0 and 1.1 does not allow columns with type of UInt16, UInt32, or UInt64 to be autoincrement columns”. (Bug#37350)
SemaphoreFullException is generated when
application is closed.
(Bug#36688)
GetSchema did not work correctly when
querying for a collection, if using a non-English locale.
(Bug#35459)
When reading back a stored double or single value using the .NET provider, the value had less precision than the one stored. (Bug#33322)
Using the MySQL Visual Studio plugin and a MySQL 4.1 server,
certain field types (ENUM) would
not be identified correctly. Also, when looking for tables, the
plugin would list all tables matching a wildcard pattern of the
database name supplied in the connection string, instead of only
tables within the specified database.
(Bug#30603)
Bugs fixed:
When creating a connection pool, specifying an invalid IP address will cause the entire application to crash, instead of providing an exception. (Bug#36432)
An incorrect value for a bit field would returned in a multi-row
query if a preceding value for the field returned
NULL.
(Bug#36313)
The MembershipProvider will raise an
exception when the connection string is configured with
enablePasswordRetrival = true and
RequireQuestionAndAnswer = false.
(Bug#36159)
When calling GetNumberOfUsersOnline an
exception is raised on the submitted query due to a missing
parameter.
(Bug#36157)
Tables with GEOMETRY field types would return
an unknown datatype exception.
(Bug#36081)
When creating a connection, setting the
ConnectionString property of
MySqlConnection to NULL
would throw an exception.
(Bug#35619)
The DbCommandBuilder.QuoteIdentifer
method was not implemented.
(Bug#35492)
When using SqlDataSource to open a
connection, the connection would not automatically be closed
when access had completed.
(Bug#34460)
Attempting to use an isolation level other than the default with a transaction scope would use the default isolation level. (Bug#34448)
When altering a stored procedure within Visual Studio, the parameters to the procedure could be lost. (Bug#34359)
A race condition could occur within the procedure cache resulting the cache contents overflowing beyond the configured cache size. (Bug#34338)
Using the TableAdaptor wizard in combination
with a suitable SELECT statement,
only the associated INSERT
statement would also be created, rather than the required
DELETE and
UPDATE statements.
(Bug#31338)
Functionality added or changed:
Performing GetValue() on a field
TINYINT(1) returned a
BOOLEAN. While not a bug, this
caused problems in software that expected an
INT to be returned. A new
connection string option Treat Tiny As
Boolean has been added with a default value of
true. If set to false the
provider will treat TINYINT(1) as
INT.
(Bug#34052)
Bugs fixed:
Some speed improvements have been implemented in the
TokenizeSql process used to identify elements
of SQL statements.
(Bug#34220)
When accessing tables from different databases within the same
TransactionScope, the same user/password
combination would be used for each database connection. MySQL Connector/NET
does not handle multiple connections within the same transaction
scope. An error is now returned if you attempt this process,
instead of using the incorrect authorization information.
(Bug#34204)
The status of connections reported through the state change handler was not being updated correctly. (Bug#34082)
Incorporated some connection string cache optimizations sent to us by Maxim Mass. (Bug#34000)
In an open connection where the server had disconnected unexpectedly, the status information of the connection would not be updated properly. (Bug#33909)
MySQL Connector/NET would fail to compile properly with nant. (Bug#33508)
Problem with membership provider would mean that
FindUserByEmail would fail with a
MySqlException because it was trying to add a
second parameter with the same name as the first.
(Bug#33347)
Using compression in the MySQL connection with MySQL Connector/NET would be slower than using native (uncompressed) communication. (Bug#27865)
Bugs fixed:
Setting the size of a string parameter after the value could cause an exception. (Bug#32094)
Creation of parameter objects with noninput direction using a constructor would fail. This was cause by some old legacy code preventing their use. (Bug#32093)
A date string could be returned incorrectly by
MySqlDataTime.ToString() when the date
returned by MySQL was 0000-00-00 00:00:00.
(Bug#32010)
A syntax error in a set of batch statements could leave the data adapter in a state that appears hung. (Bug#31930)
Installing over a failed uninstall of a previous version could
result in multiple clients being registered in the
machine.config. This would prevent certain
aspects of the MySQL connection within Visual Studio to work
properly.
(Bug#31731)
MySQL Connector/NET would incorrectly report success when enlisting in a distributed transaction, although distributed transactions are not supported. (Bug#31703)
Data cached from the connection string could return invalid information because the internal routines were not using case-sensitive semantics. This lead to updated connection string options not being recognized if they were of a different case than the existing cached values. (Bug#31433)
Trying to use a connection that was not open could return an ambiguous and misleading error message. (Bug#31262)
Column name metadata was not using the character set as deifned within the connection string being used. (Bug#31185)
Memory usage could increase and decrease significantly when updating or inserting a large number of rows. (Bug#31090)
Commands executed from within the state change handeler would
fail with a NULL exception.
(Bug#30964)
Extracting data through XML functions within a query returns the
data as System.Byte[]. This was due to MySQL Connector/NET
incorrectly identifying BLOB
fields as binary, rather than text.
(Bug#30233)
When running a stored procedure multiple times on the same connection, the memory usage could increase indefinitely. (Bug#30116)
Column types with only 1-bit (such as
BOOLEAN and
TINYINT(1) were not returned as boolean
fields.
(Bug#27959)
When accessing certain statements, the command would timeout
before the command completed. Because this cannot always be
controlled through the individual command timeout options, a
default command timeout has been added to the
connection string options.
(Bug#27958)
The server error code was not updated in the
Data[] hash, which prevented
DbProviderFactory users from accessing the
server error code.
(Bug#27436)
The MySqlDbType.Datetime has been replaced
with MySqlDbType.DateTime. The old format has
been obsoleted.
(Bug#26344)
Changing the connection string of a connection to one that changes the parameter marker after the connection had been assigned to a command but before the connection is opened could cause parameters to not be found. (Bug#13991)
This is a new Beta development release, fixing recently discovered bugs.
Bugs fixed:
An incorrect ConstraintException could be
raised on an INSERT when adding
rows to a table with a multiple-column unique key index.
(Bug#30204)
A DATE field would be updated
with a date/time value, causing a
MySqlDataAdapter.Update() exception.
(Bug#30077)
The Saudi Hijri calendar was not supported. (Bug#29931)
Calling SHOW CREATE PROCEDURE for
routines with a hyphen in the catalog name produced a syntax
error.
(Bug#29526)
Connecting to a MySQL server earlier than version 4.1 would
raise a NullException.
(Bug#29476)
The availability of a MySQL server would not be reset when using
pooled connections (pooling=true). This would
lead to the server being reported as unavailable, even if the
server become available while the application was still running.
(Bug#29409)
A FormatException error would be raised if a
parameter had not been found, instead of
Resources.ParameterMustBeDefined.
(Bug#29312)
An exception would be thrown when using the Manage Role functionality within the web administrator to assign a role to a user. (Bug#29236)
Using the membership/role providers when
validationKey or
decryptionKey parameters are set to
AutoGenerate, an exception would be raised
when accessing the corresponding values.
(Bug#29235)
Certain operations would not check the
UsageAdvisor setting, causing log messages
from the Usage Advisor even when it was disabled.
(Bug#29124)
Using the same connection string multiple times would result in
Database=
appearing multiple times in the resulting string.
(Bug#29123)dbname
Visual Studio Plugin: Adding a new query
based on a stored procedure that uses the
SELECT statement would terminate
the query/TableAdapter wizard.
(Bug#29098)
Using TransactionScope would cause an
InvalidOperationException.
(Bug#28709)
This is a new Beta development release, fixing recently discovered bugs.
Bugs fixed:
Log messages would be truncated to 300 bytes. (Bug#28706)
Creating a user would fail due to the application name being set incorrectly. (Bug#28648)
Visual Studio Plugin: Adding a new query
based on a stored procedure that used a
UPDATE,
INSERT or
DELETE statement would terminate
the query/TableAdapter wizard.
(Bug#28536)
Visual Studio Plugin: Query Builder would
fail to show TINYTEXT columns,
and any columns listed after a
TINYTEXT column correctly.
(Bug#28437)
Accessing the results from a large query when using data compression in the connection would fail to return all the data. (Bug#28204)
Visual Studio Plugin: Update commands would not be generated correctly when using the TableAdapter wizard. (Bug#26347)
Bugs fixed:
Running the statement SHOW
PROCESSLIST would return columns as byte arrays
instead of native columns.
(Bug#28448)
Installation of the MySQL Connector/NET on Windows would fail if VisualStudio had not already been installed. (Bug#28260)
MySQL Connector/NET would look for the wrong table when executing
User.IsRole().
(Bug#28251)
Building a connection string within a tight loop would show slow performance. (Bug#28167)
The UNSIGNED flag for parameters in a stored
procedure would be ignored when using
MySqlCommandBuilder to obtain the parameter
information.
(Bug#27679)
Using MySQLDataAdapter.FillSchema() on a
stored procedure would raise an exception: Invalid
attempt to access a field before calling Read().
(Bug#27668)
DATETIME fields from versions of
MySQL bgefore 4.1 would be incorrectly parsed, resulting in a
exception.
(Bug#23342)
Fixed password property on
MySqlConnectionStringBuilder to use
PasswordPropertyText attribute. This causes
dots to show instead of actual password text.
Functionality added or changed:
Now compiles for .NET CF 2.0.
Rewrote stored procedure parsing code using a new SQL tokenizer. Really nasty procedures including nested comments are now supported.
GetSchema will now report objects relative to the currently selected database. What this means is that passing in null as a database restriction will report objects on the currently selected database only.
Added Membership and Role provider contributed by Sean Wright (thanks!).
Bugs fixed:
If, when using the MySqlTransaction
transaction object, an exception was thrown, the transaction
object was not disposed of and the transaction was not rolled
back.
(Bug#39817)
Executing a command that resulted in a fatal exception did not close the connection. (Bug#37991)
When a prepared insert query is run that contains an
UNSIGNED TINYINT in the parameter list, the
complete query and data that should be inserted is corrupted and
no error is thrown.
(Bug#37968)
In a .NET application MySQL Connector/NET modifies the connection string so that it contains several occurrences of the same option with different values. This is illustrated by the example that follows.
The original connection string:
host=localhost;database=test;uid=*****;pwd=*****; connect timeout=25; auto enlist=false;pooling=false;
The connection string after after closing
MySqlDataReader:
host=localhost;database=test;uid=*****;pwd=*****; connect timeout=25;auto enlist=false;pooling=false; Allow User Variables=True;Allow User Variables=False; Allow User Variables=True;Allow User Variables=False;
When creating a connection pool, specifying an invalid IP address will cause the entire application to crash, instead of providing an exception. (Bug#36432)
GetSchema did not work correctly when
querying for a collection, if using a non-English locale.
(Bug#35459)
When reading back a stored double or single value using the .NET provider, the value had less precision than the one stored. (Bug#33322)
Bugs fixed:
The DbCommandBuilder.QuoteIdentifer
method was not implemented.
(Bug#35492)
Setting the size of a string parameter after the value could cause an exception. (Bug#32094)
Creation of parameter objects with noninput direction using a constructor would fail. This was cause by some old legacy code preventing their use. (Bug#32093)
A date string could be returned incorrectly by
MySqlDataTime.ToString() when the date
returned by MySQL was 0000-00-00 00:00:00.
(Bug#32010)
A syntax error in a set of batch statements could leave the data adapter in a state that appears hung. (Bug#31930)
Installing over a failed uninstall of a previous version could
result in multiple clients being registered in the
machine.config. This would prevent certain
aspects of the MySQL connection within Visual Studio to work
properly.
(Bug#31731)
Data cached from the connection string could return invalid information because the internal routines were not using case-sensitive semantics. This lead to updated connection string options not being recognized if they were of a different case than the existing cached values. (Bug#31433)
Column name metadata was not using the character set as deifned within the connection string being used. (Bug#31185)
Memory usage could increase and decrease significantly when updating or inserting a large number of rows. (Bug#31090)
Commands executed from within the state change handeler would
fail with a NULL exception.
(Bug#30964)
When running a stored procedure multiple times on the same connection, the memory usage could increase indefinitely. (Bug#30116)
The server error code was not updated in the
Data[] hash, which prevented
DbProviderFactory users from accessing the
server error code.
(Bug#27436)
Changing the connection string of a connection to one that changes the parameter marker after the connection had been assigned to a command but before the connection is opened could cause parameters to not be found. (Bug#13991)
This version introduces a new installer technology.
Bugs fixed:
Extracting data through XML functions within a query returns the
data as System.Byte[]. This was due to MySQL Connector/NET
incorrectly identifying BLOB
fields as binary, rather than text.
(Bug#30233)
An incorrect ConstraintException could be
raised on an INSERT when adding
rows to a table with a multiple-column unique key index.
(Bug#30204)
A DATE field would be updated
with a date/time value, causing a
MySqlDataAdapter.Update() exception.
(Bug#30077)
Fixed bug where MySQL Connector/NET was hand building some date time patterns rather than using the patterns provided under CultureInfo. This caused problems with some calendars that do not support the same ranges as Gregorian.. (Bug#29931)
Calling SHOW CREATE PROCEDURE for
routines with a hyphen in the catalog name produced a syntax
error.
(Bug#29526)
The availability of a MySQL server would not be reset when using
pooled connections (pooling=true). This would
lead to the server being reported as unavailable, even if the
server become available while the application was still running.
(Bug#29409)
A FormatException error would be raised if a
parameter had not been found, instead of
Resources.ParameterMustBeDefined.
(Bug#29312)
Certain operations would not check the
UsageAdvisor setting, causing log messages
from the Usage Advisor even when it was disabled.
(Bug#29124)
Using the same connection string multiple times would result in
Database=
appearing multiple times in the resulting string.
(Bug#29123)dbname
Log messages would be truncated to 300 bytes. (Bug#28706)
Accessing the results from a large query when using data compression in the connection will fail to return all the data. (Bug#28204)
Fixed problem where
MySqlConnection.BeginTransaction checked the
drivers status var before checking if the connection was open.
The result was that the driver could report an invalid condition
on a previously opened connection.
Fixed problem where we were not closing prepared statement handles when commands are disposed. This could lead to using up all prepared statement handles on the server.
Fixed the database schema collection so that it works on servers
that are not properly respecting the
lower_case_table_names setting.
Fixed problem where any attempt to not read all the records returned from a select where each row of the select is greater than 1024 bytes would hang the driver.
Fixed problem where a command timing out just after it actually finished would cause an exception to be thrown on the command timeout thread which would then be seen as an unhandled exception.
Fixed some serious issues with command timeout and cancel that could present as exceptions about thread ownership. The issue was that not all queries cancel the same. Some produce resultsets while others don't. ExecuteReader had to be changed to check for this.
Bugs fixed:
Running the statement SHOW
PROCESSLIST would return columns as byte arrays
instead of native columns.
(Bug#28448)
Building a connection string within a tight loop would show slow performance. (Bug#28167)
Using logging (with the logging=true
parameter to the connection string) would not generate a log
file.
(Bug#27765)
The UNSIGNED flag for parameters in a stored
procedure would be ignored when using
MySqlCommandBuilder to obtain the parameter
information.
(Bug#27679)
Using MySQLDataAdapter.FillSchema() on a
stored procedure would raise an exception: Invalid
attempt to access a field before calling Read().
(Bug#27668)
If you close an open connection with an active transaction, the transaction is not automatically rolled back. (Bug#27289)
When cloning an open
MySqlClient.MySqlConnection with the
Persist Security Info=False option set, the
cloned connection is not usable because the security information
has not been cloned.
(Bug#27269)
Enlisting a null transaction would affect the current connection object, such that further enlistment operations to the transaction are not possible. (Bug#26754)
Attempting to change the Connection Protocol
property within a PropertyGrid control would
raise an exception.
(Bug#26472)
The characterset property would not be
identified during a connection (also affected Visual Studion
Plugin).
(Bug#26147, Bug#27240)
The CreateFormat column of the
DataTypes collection did not contain a format
specification for creating a new column type.
(Bug#25947)
DATETIME fields from versions of
MySQL bgefore 4.1 would be incorrectly parsed, resulting in a
exception.
(Bug#23342)
Bugs fixed:
Publisher listed in "Add/Remove Programs" is not consistent with other MySQL products. (Bug#27253)
DESCRIBE .... SQL statement returns byte
arrays rather than data on MySQL versions older than 4.1.15.
(Bug#27221)
cmd.Parameters.RemoveAt("Id") will cause an
error if the last item is requested.
(Bug#27187)
MySqlParameterCollection and parameters added
with Insert method can not be retrieved later
using ParameterName.
(Bug#27135)
Exception thrown when using large values in
UInt64 parameters.
(Bug#27093)
MySQL Visual Studio Plugin 1.1.2 does not work with MySQL Connector/NET 5.0.5. (Bug#26960)
Functionality added or changed:
Reverted behavior that required parameter names to start with
the parameter marker. We apologize for this back and forth but
we mistakenly changed the behavior to not match what
SqlClient supports. We now support using
either syntax for adding parameters however we also respond
exactly like SqlClient in that if you ask for
the index of a parameter using a syntax different from when you
added the parameter, the result will be -1.
Assembly now properly appears in the Visual Studio 2005 Add/Remove Reference dialog.
Fixed problem that prevented use of
SchemaOnly or SingleRow
command behaviors with stored procedures or prepared statements.
Added MySqlParameterCollection.AddWithValue
and marked the Add(name, value) method as
obsolete.
Return parameters created with DeriveParameters now have the
name RETURN_VALUE.
Fixed problem with parameter name hashing where the hashes were not getting updated when parameters were removed from the collection.
Fixed problem with calling stored functions when a return parameter was not given.
Added Use Procedure Bodies connection string
option to allow calling procedures without using procedure
metadata.
Bugs fixed:
MySqlConnection.GetSchema fails with
NullReferenceException for Foreign Keys.
(Bug#26660)
MySQL Connector/NET would fail to install under Windows Vista. (Bug#26430)
Opening a connection would be slow due to host name lookup. (Bug#26152)
Incorrect values/formats would be applied when the
OldSyntax connection string option was used.
(Bug#25950)
Registry would be incorrectly populated with installation locations. (Bug#25928)
Times with negative values would be returned incorrectly. (Bug#25912)
Returned data types of a DataTypes collection
do not contain the right correctl CLR Datatype.
(Bug#25907)
GetSchema and DataTypes
would throw an exception due to an incorrect table name.
(Bug#25906)
MySqlConnection throws an exception when
connecting to MySQL v4.1.7.
(Bug#25726)
SELECT did not work correctly
when using a WHERE clause containing a UTF-8
string.
(Bug#25651)
When closing and then re-opening a connection to a database, the character set specification is lost. (Bug#25614)
Filling a table schema through a stored procedure triggers a runtime error. (Bug#25609)
BINARY and
VARBINARY columns would be
returned as a string, not binary, datatype.
(Bug#25605)
A critical ConnectionPool error would result
in repeated System.NullReferenceException.
(Bug#25603)
The UpdateRowSource.FirstReturnedRecord
method does not work.
(Bug#25569)
When connecting to a MySQL Server earlier than version 4.1, the connection would hang when reading data. (Bug#25458)
Using ExecuteScalar() with more than one
query, where one query fails, will hang the connection.
(Bug#25443)
When a MySqlConversionException is raised on
a remote object, the client application would receive a
SerializationException instead.
(Bug#24957)
When connecting to a server, the return code from the connection could be zero, even though the host name was incorrect. (Bug#24802)
High CPU utilization would be experienced when there is no idle
connection waiting when using pooled connections through
MySqlPool.GetConnection.
(Bug#24373)
MySQL Connector/NET would not compile properly when used with Mono 1.2. (Bug#24263)
Applications would crash when calling with
CommandType set to
StoredProcedure.
This is a new Beta development release, fixing recently discovered bugs.
Functionality added or changed:
Usage Advisor has been implemented. The Usage Advisor checks your queries and will report if you are using the connection inefficiently.
PerfMon hooks have been added to monitor the stored procedure cache hits and misses.
The MySqlCommand object now supports
asynchronous query methods. This is implemented useg the
BeginExecuteNonQuery and
EndExecuteNonQuery methods.
Metadata from storaed procedures and stored function execution are cached.
The CommandBuilder.DeriveParameters function
has been updated to the procedure cache.
The ViewColumns GetSchema
collection has been updated.
Improved speed and performance by re-architecting certain sections of the code.
Support for the embedded server and client library have been removed from this release. Support will be added back to a later release.
The ShapZipLib library has been replaced with the deflate support provided within .NET 2.0.
SSL support has been updated.
Bugs fixed:
Additional text added to error message (Bug#25178)
An exception would be raised, or the process would hang, if
SELECT privileges on a database
were not granted and a stored procedure was used.
(Bug#25033)
When adding parameter objects to a command object, if the
parameter direction is set to ReturnValue
before the parameter is added to the command object then when
the command is executed it throws an error.
(Bug#25013)
Using Driver.IsTooOld() would return the
wrong value.
(Bug#24661)
When using a DbNull.Value as the value for a
parameter value, and then later setting a specific value type,
the command would fail with an exception because the wrong type
was implied from the DbNull.Value.
(Bug#24565)
Stored procedure executions are not thread safe. (Bug#23905)
Deleting a connection to a disconnected server when using the Visual Studio Plugin would cause an assertion failure. (Bug#23687)
Nested transactions (which are unsupported)do not raise an error or warning. (Bug#22400)
Functionality added or changed:
An Ignore Prepare option has been added to
the connection string options. If enabled, prepared statements
will be disabled application-wide. The default for this option
is true.
Implemented a stored procedure cache. By default, the connector
caches the metadata for the last 25 procedures that are seen.
You can change the numbver of procedures that are cacheds by
using the procedure cache connection string.
Important change: Due to a number of issues with the use of server-side prepared statements, MySQL Connector/NET 5.0.2 has disabled their use by default. The disabling of server-side prepared statements does not affect the operation of the connector in any way.
To enable server-side prepared statements you must add the following configuration property to your connector string properties:
ignore prepare=false
The default value of this property is true.
Bugs fixed:
One system where IPv6 was enabled, MySQL Connector/NET would incorrectly resolve host names. (Bug#23758)
Column names with accented characters were not parsed properly causing malformed column names in result sets. (Bug#23657)
An exception would be thrown when calling
GetSchemaTable and fields
was null.
(Bug#23538)
A System.FormatException exception would be
raised when invoking a stored procedure with an
ENUM input parameter.
(Bug#23268)
During installation, an antivirus error message would be raised (indicating a malicious script problem). (Bug#23245)
Creating a connection through the Server Explorer when using the Visual Studio Plugin would fail. The installer for the Visual Studio Plugin has been updated to ensure that MySQL Connector/NET 5.0.2 must be installed. (Bug#23071)
Using Windows Vista (RC2) as a nonprivileged user would raise a
Registry key 'Global' access denied.
(Bug#22882)
Within Mono, using the PreparedStatement
interface could result in an error due to a
BitArray copying error.
(Bug#18186)
MySQL Connector/NET did not work as a data source for the
SqlDataSource object used by ASP.NET 2.0.
(Bug#16126)
Bugs fixed:
MySQL Connector/NET on a Tukish operating system, may fail to execute certain SQL statements correctly. (Bug#22452)
Starting a transaction on a connection created by
MySql.Data.MySqlClient.MySqlClientFactory,
using BeginTransaction without specifying an
isolation level, causes the SQL statement to fail with a syntax
error.
(Bug#22042)
The MySqlexception class is now derived from
the DbException class.
(Bug#21874)
The # would not be accepted within
column/table names, even though it was valid.
(Bug#21521)
You can now install the MySQL Connector/NET MSI package from the command line
using the /passive,
/quiet, /q options.
(Bug#19994)
Submitting an empty string to a command object through
prepare raises an
System.IndexOutOfRangeException, rather than
a MySQL Connector/NET exception.
(Bug#18391)
Using ExecuteScalar with a datetime field,
where the value of the field is "0000-00-00 00:00:00", a
MySqlConversionException exception would be
raised.
(Bug#11991)
An MySql.Data.Types.MySqlConversionException
would be raised when trying to update a row that contained a
date field, where the date field contained a zero value
(0000-00-00 00:00:00).
(Bug#9619)
Executing multiple queries as part of a transaction returns
There is already an openDataReader associated with this
Connection which must be closed first.
(Bug#7248)
Incorrect field/data lengths could be returned for
VARCHAR UTF8 columns. Bug
(#14592)
Functionality added or changed:
Replaced use of ICSharpCode with .NET 2.0 internal deflate support.
Refactored test suite to test all protocols in a single pass.
Added usage advisor warnings for requesting column values by the wrong type.
Reimplemented PacketReader/PacketWriter support into
MySqlStream class.
Reworked connection string classes to be simpler and faster.
Added procedure metadata caching.
Added internal implemention of SHA1 so we don't have to distribute the OpenNetCF on mobile devices.
Implemented MySqlClientFactory class.
Added perfmon hooks for stored procedure cache hits and misses.
Implemented classes and interfaces for ADO.Net 2.0 support.
Added Async query methods.
Implemented Usage Advisor.
Completely refactored how column values are handled to avoid boxing in some cases.
Implemented MySqlConnectionBuilder class.
Bugs fixed:
CommandText: Question mark in comment line is being parsed as a parameter. (Bug#6214)
Bugs fixed:
Attempting to utilize MySQL Connector .Net version 1.0.10 throws a fatal exception under Mono when pooling is enabled. (Bug#33682)
Setting the size of a string parameter after the value could cause an exception. (Bug#32094)
Creation of parameter objects with noninput direction using a constructor would fail. This was cause by some old legacy code preventing their use. (Bug#32093)
Memory usage could increase and decrease significantly when updating or inserting a large number of rows. (Bug#31090)
Commands executed from within the state change handeler would
fail with a NULL exception.
(Bug#30964)
Extracting data through XML functions within a query returns the
data as System.Byte[]. This was due to MySQL Connector/NET
incorrectly identifying BLOB
fields as binary, rather than text.
(Bug#30233)
Using compression in the MySQL connection with MySQL Connector/NET would be slower than using native (uncompressed) communication. (Bug#27865)
Changing the connection string of a connection to one that changes the parameter marker after the connection had been assigned to a command but before the connection is opened could cause parameters to not be found. (Bug#13991)
Bugs fixed:
An incorrect ConstraintException could be
raised on an INSERT when adding
rows to a table with a multiple-column unique key index.
(Bug#30204)
The availability of a MySQL server would not be reset when using
pooled connections (pooling=true). This would
lead to the server being reported as unavailable, even if the
server become available while the application was still running.
(Bug#29409)
Publisher listed in "Add/Remove Programs" is not consistent with other MySQL products. (Bug#27253)
MySqlParameterCollection and parameters added
with Insert method can not be retrieved later
using ParameterName.
(Bug#27135)
BINARY and
VARBINARY columns would be
returned as a string, not binary, datatype.
(Bug#25605)
A critical ConnectionPool error would result
in repeated System.NullReferenceException.
(Bug#25603)
When a MySqlConversionException is raised on
a remote object, the client application would receive a
SerializationException instead.
(Bug#24957)
High CPU utilization would be experienced when there is no idle
connection waiting when using pooled connections through
MySqlPool.GetConnection.
(Bug#24373)
Functionality added or changed:
The ICSharpCode ZipLib is no longer used by the Connector, and is no longer distributed with it.
Important change: Binaries for .NET 1.0 are no longer supplied with this release. If you need support for .NET 1.0, you must build from source.
Improved CommandBuilder.DeriveParameters to
first try and use the procedure cache before querying for the
stored procedure metadata. Return parameters created with
DeriveParameters now have the name
RETURN_VALUE.
An Ignore Prepare option has been added to
the connection string options. If enabled, prepared statements
will be disabled application-wide. The default for this option
is true.
Implemented a stored procedure cache. By default, the connector
caches the metadata for the last 25 procedures that are seen.
You can change the numbver of procedures that are cacheds by
using the procedure cache connection string.
Important change: Due to a number of issues with the use of server-side prepared statements, MySQL Connector/NET 5.0.2 has disabled their use by default. The disabling of server-side prepared statements does not affect the operation of the connector in any way.
To enable server-side prepared statements you must add the following configuration property to your connector string properties:
ignore prepare=false
The default value of this property is true.
Bugs fixed:
Times with negative values would be returned incorrectly. (Bug#25912)
MySqlConnection throws a
NullReferenceException and
ArgumentNullException when connecting to
MySQL v4.1.7.
(Bug#25726)
SELECT did not work correctly
when using a WHERE clause containing a UTF-8
string.
(Bug#25651)
When closing and then re-opening a connection to a database, the character set specification is lost. (Bug#25614)
Trying to fill a table schema through a stored procedure triggers a runtime error. (Bug#25609)
Using ExecuteScalar() with more than one
query, where one query fails, will hang the connection.
(Bug#25443)
Additional text added to error message. (Bug#25178)
When adding parameter objects to a command object, if the
parameter direction is set to ReturnValue
before the parameter is added to the command object then when
the command is executed it throws an error.
(Bug#25013)
When connecting to a server, the return code from the connection could be zero, even though the host name was incorrect. (Bug#24802)
Using Driver.IsTooOld() would return the
wrong value.
(Bug#24661)
When using a DbNull.Value as the value for a
parameter value, and then later setting a specific value type,
the command would fail with an exception because the wrong type
was implied from the DbNull.Value.
(Bug#24565)
Stored procedure executions are not thread safe. (Bug#23905)
The CommandBuilder would mistakenly add
insert parameters for a table column with auto incrementation
enabled.
(Bug#23862)
One system where IPv6 was enabled, MySQL Connector/NET would incorrectly resolve host names. (Bug#23758)
Nested transactions do not raise an error or warning. (Bug#22400)
An System.OverflowException would be raised
when accessing a varchar field over 255 bytes. Bug (#23749)
Within Mono, using the PreparedStatement
interface could result in an error due to a
BitArray copying error. (Bug 18186)
Functionality added or changed:
Stored procedures are now cached.
The method for retrieving stored procedured metadata has been
changed so that users without
SELECT privileges on the
mysql.proc table can use a stored procedure.
Bugs fixed:
MySQL Connector/NET on a Tukish operating system, may fail to execute certain SQL statements correctly. (Bug#22452)
The # would not be accepted within
column/table names, even though it was valid.
(Bug#21521)
Calling Close on a connection after
calling a stored procedure would trigger a
NullReferenceException.
(Bug#20581)
You can now install the MySQL Connector/NET MSI package from the command line
using the /passive,
/quiet, /q options.
(Bug#19994)
The DiscoverParameters function would fail when a stored
procedure used a NUMERIC
parameter type.
(Bug#19515)
When running a query that included a date comparison, a DateReader error would be raised. (Bug#19481)
IDataRecord.GetString would raise
NullPointerException for null values in
returned rows. Method now throws
SqlNullValueException.
(Bug#19294)
Parameter substitution in queries where the order of parameters and table fields did not match would substitute incorrect values. (Bug#19261)
Submitting an empty string to a command object through
prepare raises an
System.IndexOutOfRangeException, rather than
a MySQL Connector/NET exception.
(Bug#18391)
An exception would be raised when using an output parameter to a
System.String value.
(Bug#17814)
CHAR type added to MySqlDbType. (Bug#17749)
A SELECT query on a table with a
date with a value of '0000-00-00' would hang
the application.
(Bug#17736)
The CommandBuilder ignored Unsigned flag at Parameter creation. (Bug#17375)
When working with multiple threads, character set initialization would generate errors. (Bug#17106)
When using an unsigned 64-bit integer in a stored procedure, the unsigned bit would be lost stored. (Bug#16934)
DataReader would show the value of the
previous row (or last row with nonnull data) if the current row
contained a datetime field with a null value.
(Bug#16884)
Unsigned data types were not properly supported. (Bug#16788)
The connection string parser did not allow single or double quotes in the password. (Bug#16659)
The MySqlDateTime class did not contain
constructors.
(Bug#15112)
Called MySqlCommandBuilder.DeriveParameters
for a stored procedure that has no paramers would cause an
application crash.
(Bug#15077)
Using ExecuteScalar with a datetime field,
where the value of the field is "0000-00-00 00:00:00", a
MySqlConversionException exception would be
raised.
(Bug#11991)
An MySql.Data.Types.MySqlConversionException
would be raised when trying to update a row that contained a
date field, where the date field contained a zero value
(0000-00-00 00:00:00).
(Bug#9619)
When using MySqlDataAdapter, connections to a
MySQL server may remain open and active, even though the use of
the connection has been completed and the data received.
(Bug#8131)
Executing multiple queries as part of a transaction returns
There is already an openDataReader associated with this
Connection which must be closed first.
(Bug#7248)
Incorrect field/data lengths could be returned for
VARCHAR UTF8 columns. Bug
(#14592)
Bugs fixed:
Unsigned tinyint (NET byte) would lead to and
incorrectly determined parameter type from the parameter value.
(Bug#18570)
A #42000Query was empty exception occurred
when executing a query built with
MySqlCommandBuilder, if the query string
ended with a semicolon.
(Bug#14631)
The parameter collection object's Add()
method added parameters to the list without first checking to
see whether they already existed. Now it updates the value of
the existing parameter object if it exists.
(Bug#13927)
Added support for the cp932 character set.
(Bug#13806)
Calling a stored procedure where a parameter contained special
characters (such as '@') would produce an
exception. Note that
ANSI_QUOTES had to be enabled
to make this possible.
(Bug#13753)
The Ping() method did not update the
State property of the
Connection object.
(Bug#13658)
Implemented the
MySqlCommandBuilder.DeriveParameters method
that is used to discover the parameters for a stored procedure.
(Bug#13632)
A statement that contained multiple references to the same parameter could not be prepared. (Bug#13541)
Bugs fixed:
MySQL Connector/NET 1.0.5 could not connect on Mono. (Bug#13345)
Serializing a parameter failed if the first value passed in was
NULL.
(Bug#13276)
Field names that contained the following characters caused
errors: ()%<>/
(Bug#13036)
The nant build sequence had problems.
(Bug#12978)
The MySQL Connector/NET 1.0.5 installer would not install alongside MySQL Connector/NET 1.0.4. (Bug#12835)
Bugs fixed:
MySQL Connector/NET could not connect to MySQL 4.1.14. (Bug#12771)
With multiple hosts in the connection string, MySQL Connector/NET would not connect to the last host in the list. (Bug#12628)
The ConnectionString property could not be
set when a MySqlConnection object was added
with the designer.
(Bug#12551, Bug#8724)
The cp1250 character set was not supported.
(Bug#11621)
A call to a stored procedure caused an exception if the stored procedure had no parameters. (Bug#11542)
Certain malformed queries would trigger a Connection
must be valid and open error message.
(Bug#11490)
Trying to use a stored procedure when
Connection.Database was not populated
generated an exception.
(Bug#11450)
MySQL Connector/NET interpreted the new decimal data type as a byte array. (Bug#11294)
Added support to call a stored function from MySQL Connector/NET. (Bug#10644)
Connection could fail when .NET thread pool had no available worker threads. (Bug#10637)
Calling MySqlConnection.clone when a
connection string had not yet been set on the original
connection would generate an error.
(Bug#10281)
Decimal parameters caused syntax errors. (Bug#10152, Bug#11550, Bug#10486)
Parameters were not recognized when they were separated by linefeeds. (Bug#9722)
The MySqlCommandBuilder class could not
handle queries that referenced tables in a database other than
the default database.
(Bug#8382)
Trying to read a TIMESTAMP column
generated an exception.
(Bug#7951)
MySQL Connector/NET could not work properly with certain regional settings. (WL#8228)
Bugs fixed:
MySqlReader.GetInt32 throws exception if
column is unsigned.
(Bug#7755)
Quote character \222 not quoted in
EscapeString.
(Bug#7724)
GetBytes is working no more. (Bug#7704)
MySqlDataReader.GetString(index) returns
non-Null value when field is Null.
(Bug#7612)
Clone method bug in MySqlCommand.
(Bug#7478)
Problem with Multiple resultsets. (Bug#7436)
MySqlAdapter.Fill method throws error message
Non-negative number required.
(Bug#7345)
MySqlCommand.Connection returns an
IDbConnection.
(Bug#7258)
Calling prepare causing exception. (Bug#7243)
Fixed problem with shared memory connections.
Added or filled out several more topics in the API reference documentation.
Fixed another small problem with prepared statements.
Fixed problem that causes named pipes to not work with some blob functionality.
Bugs fixed:
Invalid query string when using inout parameters (Bug#7133)
Inserting DateTime causes
System.InvalidCastException to be thrown.
(Bug#7132)
MySqlDateTime in Datatables sorting by Text,
not Date.
(Bug#7032)
Exception stack trace lost when re-throwing exceptions. (Bug#6983)
Errors in parsing stored procedure parameters. (Bug#6902)
InvalidCast when using DATE_ADD-function.
(Bug#6879)
Int64 Support in MySqlCommand Parameters.
(Bug#6863)
Test suite fails with MySQL 4.0 because of case sensitivity of table names. (Bug#6831)
MySqlDataReader.GetChar(int i) throws
IndexOutOfRange exception.
(Bug#6770)
Integer "out" parameter from stored procedure returned as string. (Bug#6668)
An Open Connection has been Closed by the Host System. (Bug#6634)
Fixed Invalid character set index: 200. (Bug#6547)
Connections now do not have to give a database on the connection string.
Installer now includes options to install into GAC and create items.
Fixed major problem with detecting null values when using prepared statements.
Fixed problem where multiple resultsets having different numbers of columns would cause a problem.
Added ServerThread property to
MySqlConnection to expose server thread id.
Added Ping method to MySqlConnection.
Changed the name of the test suite to
MySql.Data.Tests.dll.
Now SHOW COLLATION is used upon
connection to retrieve the full list of charset ids.
Made MySQL the default named pipe name.
Bugs fixed:
Fixed Objects not being disposed (Bug#6649)
Fixed Charset-map for UCS-2 (Bug#6541)
Fixed Zero date "0000-00-00" is returned wrong when filling Dataset (Bug#6429)
Fixed double type handling in MySqlParameter(string parameterName, object value) (Bug#6428)
Fixed Installation directory ignored using custom installation (Bug#6329)
Fixed #HY000 Illegal mix of collations (latin1_swedish_ci,IMPLICIT) and (utf8_general_ (Bug#6322)
Added the TableEditor CS and VB sample
Added charset connection string option
Fixed problem with MySqlBinary where string values could not be used to update extended text columns
Provider is now using character set specified by server as default
Updated the installer to include the new samples
Fixed problem where setting command text leaves the command in a prepared state
Fixed Long inserts take very long time (Bu #5453)
Fixed problem where calling stored procedures might cause an "Illegal mix of collations" problem.
Bugs fixed:
Fixed IndexOutOfBounds when reading BLOB with DataReader with GetString(index) (Bug#6230)
Fixed GetBoolean returns wrong values (Bug#6227)
Fixed Method TokenizeSql() uses only a limited set of valid characters for parameters (Bug#6217)
Fixed NET Connector source missing resx files (Bug#6216)
Fixed System.OverflowException when using YEAR datatype (Bug#6036)
Fixed MySqlDateTime sets IsZero property on all subseq.records after first zero found (Bug#6006)
Fixed serializing of floating point parameters (double, numeric, single, decimal) (Bug#5900)
Fixed missing Reference in DbType setter (Bug#5897)
Fixed Parsing the ';' char (Bug#5876)
Fixed DBNull Values causing problems with retrieving/updating queries. (Bug#5798)
IsNullable error (Bug#5796)
Fixed problem where MySqlParameterCollection.Add() would throw unclear exception when given a null value (Bug#5621)
Fixed construtor initialize problems in MySqlCommand() (Bug#5613)
Fixed Yet Another "object reference not set to an instance of an object" (Bug#5496)
Fixed Can't display Chinese correctly (Bug#5288)
Fixed MySqlDataReader and 'show tables from ...' behavior (Bug#5256)
Fixed problem in PacketReader where it could try to allocate the wrong buffer size in EnsureCapacity
Fixed problem where using old syntax while using the interfaces caused problems
Fixed Bug#5458 Calling GetChars on a longtext column throws an exception
Added test case for resetting the command text on a prepared command
Fixed Bug#5388 DataReader reports all rows as NULL if one row is NULL
Fixed problem where connection lifetime on the connect string was not being respected
Fixed Bug#5602 Possible bug in MySqlParameter(string, object) constructor
Field buffers being reused to decrease memory allocations and increase speed
Fixed Bug#5392 MySqlCommand sees "?" as parameters in string literals
Added Aggregate function test (wasn't really a bug)
Using PacketWriter instead of Packet for writing to streams
Implemented SequentialAccess
Fixed problem with ConnectionInternal where a key might be added more than once
Fixed Russian character support as well
Fixed Bug#5474 cannot run a stored procedure populating mysqlcommand.parameters
Fixed problem where connector was not issuing a CMD_QUIT before closing the socket
Fixed problem where Min Pool Size was not being respected
Refactored compression code into CompressedStream to clean up NativeDriver
CP1252 is now used for Latin1 only when the server is 4.1.2 and later
Fixed Bug#5469 Setting DbType throws NullReferenceException
Virtualized driver subsystem so future releases could easily support client or embedded server support
Bugs fixed:
Thai encoding not correctly supported. (Bug#3889)
Bumped version number to 1.0.0 for beta 1 release.
Removed all of the XML comment warnings.
Added COPYING.rtf file for use in
installer.
Updated many of the test cases.
Fixed problem with using compression.
Removed some last references to ByteFX.
Added test fixture for prepared statements.
All type classes now implement a
SerializeBinary method for sending their
data to a PacketWriter.
Added PacketWriter class that will enable
future low-memory large object handling.
Fixed many small bugs in running prepared statements and stored procedures.
Changed command so that an exception will not be thrown in executing a stored procedure with parameters in old syntax mode.
SingleRow behavior now working right even
with limit.
GetBytes now only works on binary columns.
Logger now truncates long sql commands so blob columns do not blow out our log.
Host and database now have a default value of "" unless otherwise set.
Connection Timeout seems to be ignored. (Bug#5214)
Added test case for bug# 5051: GetSchema not working correctly.
Fixed problem where GetSchema would return
false for IsUnique when the column is key.
MySqlDataReader GetXXX methods now using
the field level MySqlValue object and not
performing conversions.
DataReader returning
NULL for time column. (Bug#5097)
Added test case for
LOAD DATA LOCAL
INFILE.
Added replacetext custom nant task.
Added CommandBuilderTest fixture.
Added Last One Wins feature to
CommandBuilder.
Fixed persist security info case problem.
Fixed GetBool so that 1, true, "true", and
"yes" all count as true.
Make parameter mark configurable.
Added the "old syntax" connection string parameter to allow use of @ parameter marker.
MySqlCommandBuilder. (Bug#4658)
ByteFX.MySqlClient caches passwords if
Persist Security Info is false. (Bug#4864)
Updated license banner in all source files to include FLOSS exception.
Added new .Types namespace and implementations for most current MySql types.
Added MySqlField41 as a subclass of
MySqlField.
Changed many classes to now use the new .Types types.
Changed type enum int to
Int32, short to
Int16, and bigint to
Int64.
Added dummy types UInt16,
UInt32, and UInt64 to
allow an unsigned parameter to be made.
Connections are now reset when they are pulled from the connection pool.
Refactored auth code in driver so it can be used for both auth and reset.
Added UserReset test in
PoolingTests.cs.
Connections are now reset using
COM_CHANGE_USER when pulled from the pool.
Implemented SingleResultSet behavior.
Implemented support of unicode.
Added char set mappings for utf-8 and ucs-2.
Time fields overflow using bytefx .net mysql driver (Bug#4520)
Modified time test in data type test fixture to check for time spans where hours > 24.
Wrong string with backslash escaping in
ByteFx.Data.MySqlClient.MySqlParameter.
(Bug#4505)
Added code to Parameter test case TestQuoting to test for backslashes.
MySqlCommandBuilder fails with multi-word
column names. (Bug#4486)
Fixed bug in TokenizeSql where underscore
would terminate character capture in parameter name.
Added test case for spaces in column names.
MySqlDataReader.GetBytes do not work
correctly. (Bug#4324)
Added GetBytes() test case to
DataReader test fixture.
Now reading all server variables in
InternalConnection.Configure into
Hashtable.
Now using string[] for index map in
CharSetMap.
Added CRInSQL test case for carriage returns in SQL.
Setting maxPacketSize to default value in
Driver.ctor.
Setting MySqlDbType on a parameter doesn't
set generic type. (Bug#4442)
Removed obsolete data types Long and
LongLong.
Overflow exception thrown when using "use pipe" on connection string. (Bug#4071)
Changed "use pipe" keyword to "pipe name" or just "pipe".
Allow reading multiple resultsets from a single query.
Added flags attribute to ServerStatusFlags
enum.
Changed name of ServerStatus enum to
ServerStatusFlags.
Inserted data row doesn't update properly.
Error processing show create table. (Bug#4074)
Change Packet.ReadLenInteger to
ReadPackedLong and added
packet.ReadPackedInteger that always reads
integers packed with 2,3,4.
Added syntax.cs test fixture to test
various SQL syntax bugs.
Improper handling of time values. Now time value of 00:00:00 is not treated as null. (Bug#4149)
Moved all test suite files into TestSuite
folder.
Fixed bug where null column would move the result packet pointer backward.
Added new nant build script.
Clear tablename so it will be regen'ed properly during the
next GenerateSchema. (Bug#3917)
GetValues was always returning zero and was
also always trying to copy all fields rather than respecting
the size of the array passed in. (Bug#3915)
Implemented shared memory access protocol.
Implemented prepared statements for MySQL 4.1.
Implemented stored procedures for MySQL 5.0.
Renamed MySqlInternalConnection to
InternalConnection.
SQL is now parsed as chars, fixes problems with other languages.
Added logging and allow batch connection string options.
RowUpdating event not set when setting the
DataAdapter property. (Bug#3888)
Fixed bug in char set mapping.
Implemented 4.1 authentication.
Improved open/auth code in driver.
Improved how connection bits are set during connection.
Database name is now passed to server during initial handshake.
Changed namespace for client to
MySql.Data.MySqlClient.
Changed assembly name of client to
MySql.Data.dll.
Changed license text in all source files to GPL.
Added the MySqlClient.build Nant file.
Removed the mono batch files.
Moved some of the unused files into notused folder so nant build file can use wildcards.
Implemented shared memory access.
Major revamp in code structure.
Prepared statements now working for MySql 4.1.1 and later.
Finished implementing auth for 4.0, 4.1.0, and 4.1.1.
Changed namespace from
MySQL.Data.MySQLClient back to
MySql.Data.MySqlClient.
Fixed bug in CharSetMapping where it was
trying to use text names as ints.
Changed namespace to
MySQL.Data.MySQLClient.
Integrated auth changes from UC2004.
Fixed bug where calling any of the GetXXX methods on a datareader before or after reading data would not throw the appropriate exception (thanks Luca Morelli).
Added TimeSpan code in parameter.cs to
properly serialize a timespan object to mysql time format
(thanks Gianluca Colombo).
Added TimeStamp to parameter serialization
code. Prevented DataAdatper updates from
working right (thanks Michael King).
Fixed a misspelling in MySqlHelper.cs
(thanks Patrick Kristiansen).
Driver now using charset number given in handshake to create encoding.
Changed command editor to point to
MySqlClient.Design.
Fixed bug in Version.isAtLeast.
Changed DBConnectionString to support
changes done to MySqlConnectionString.
Removed SqlCommandEditor and
DataAdapterPreviewDialog.
Using new long return values in many places.
Integrated new CompressedStream class.
Changed ConnectionString and added
attributes to allow it to be used in
MySqlClient.Design.
Changed packet.cs to support newer
lengths in ReadLenInteger.
Changed other classes to use new properties and fields of
MySqlConnectionString.
ConnectionInternal is now using PING to see
whether the server is alive.
Moved toolbox bitmaps into resource folder.
Changed field.cs to allow values to come
directly from row buffer.
Changed to use the new driver.Send syntax.
Using a new packet queueing system.
Started work handling the "broken" compression packet handling.
Fixed bug in StreamCreator where failure to
connect to a host would continue to loop infinitly (thanks
Kevin Casella).
Improved connectstring handling.
Moved designers into Pro product.
Removed some old commented out code from
command.cs.
Fixed a problem with compression.
Fixed connection object where an exception throw prior to the connection opening would not leave the connection in the connecting state (thanks Chris Cline).
Added GUID support.
Fixed sequence out of order bug (thanks Mark Reay).
Enum values now supported as parameter values (thanks Philipp Sumi).
Year datatype now supported.
Fixed compression.
Fixed bug where a parameter with a TimeSpan
as the value would not serialize properly.
Fixed bug where default constructor would not set default connection string values.
Added some XML comments to some members.
Work to fix/improve compression handling.
Improved ConnectionString handling so that
it better matches the standard set by
SqlClient.
A MySqlException is now thrown if a user
name is not included in the connection string.
Localhost is now used as the default if not specified on the connection string.
An exception is now thrown if an attempt is made to set the connection string while the connection is open.
Small changes to ConnectionString docs.
Removed MultiHostStream and
MySqlStream. Replaced it with
Common/StreamCreator.
Added support for Use Pipe connection string value.
Added Platform class for easier access to platform utility functions.
Fixed small pooling bug where new connection was not getting
created after IsAlive fails.
Added Platform.cs and
StreamCreator.cs.
Fixed Field.cs to properly handle 4.1
style timestamps.
Changed Common.Version to
Common.DBVersion to avoid name conflict.
Fixed field.cs so that text columns
return the right field type.
Added MySqlError class to provide some
reference for error codes (thanks Geert Veenstra).
Added Unix socket support (thanks Mohammad DAMT).
Only calling Thread.Sleep when no data is
available.
Improved escaping of quote characters in parameter data.
Removed misleading comments from
parameter.cs.
Fixed pooling bug.
Fixed ConnectionString editor dialog
(thanks marco p (pomarc)).
UserId now supported in connection strings
(thanks Jeff Neeley).
Attempting to create a parameter that is not input throws an exception (thanks Ryan Gregg).
Added much documentation.
Checked in new MultiHostStream capability.
Big thanks to Dan Guisinger for this. he originally submitted
the code and idea of supporting multiple machines on the
connect string.
Added a lot of documentation.
Fixed speed issue with 0.73.
Changed to Thread.Sleep(0) in MySqlDataStream to help optimize the case where it doesn't need to wait (thanks Todd German).
Prepopulating the idlepools to MinPoolSize.
Fixed MySqlPool deadlock condition as well
as stupid bug where CreateNewPooledConnection was not ever
adding new connections to the pool. Also fixed
MySqlStream.ReadBytes and
ReadByte to not use
TicksPerSecond which does not appear to
always be right. (thanks Matthew J. Peddlesden)
Fix for precision and scale (thanks Matthew J. Peddlesden).
Added Thread.Sleep(1) to stream reading
methods to be more cpu friendly (thanks Sean McGinnis).
Fixed problem where ExecuteReader would
sometime return null (thanks Lloyd Dupont).
Fixed major bug with null field handling (thanks Naucki).
Enclosed queries for
max_allowed_packet and
characterset inside try catch (and set
defaults).
Fixed problem where socket was not getting closed properly (thanks Steve!).
Fixed problem where ExecuteNonQuery was not
always returning the right value.
Fixed InternalConnection to not use
@@session.max_allowed_packet but use
@@max_allowed_packet. (Thanks Miguel)
Added many new XML doc lines.
Fixed sql parsing to not send empty queries (thanks Rory).
Fixed problem where the reader was not unpeeking the packet on close.
Fixed problem where user variables were not being handled (thanks Sami Vaaraniemi).
Fixed loop checking in the MySqlPool (thanks Steve M. Brown)
Fixed ParameterCollection.Add method to
match SqlClient (thanks Joshua Mouch).
Fixed ConnectionString parsing to handle no
and yes for boolean and not lowercase values (thanks Naucki).
Added InternalConnection class, changes to
pooling.
Implemented Persist Security Info.
Added security.cs and
version.cs to project
Fixed DateTime handling in
Parameter.cs (thanks Burkhard
Perkens-Golomb).
Fixed parameter serialization where some types would throw a cast exception.
Fixed DataReader to convert all returned
values to prevent casting errors (thanks Keith Murray).
Added code to Command.ExecuteReader to
return null if the initial SQL statement throws an exception
(thanks Burkhard Perkens-Golomb).
Fixed ExecuteScalar bug introduced with
restructure.
Restructure to allow for LOCAL DATA INFILE
and better sequencing of packets.
Fixed several bugs related to restructure.
Early work done to support more secure passwords in Mysql 4.1. Old passwords in 4.1 not supported yet.
Parameters appearing after system parameters are now handled correctly (Adam M. (adammil)).
Strings can now be assigned directly to blob fields (Adam M.).
Fixed float parameters (thanks Pent).
Improved Parameter constructor and
ParameterCollection.Add methods to better
match SqlClient (thanks Joshua Mouch).
Corrected Connection.CreateCommand to
return a MySqlCommand type.
Fixed connection string designer dialog box problem (thanks Abraham Guyt).
Fixed problem with sending commands not always reading the response packet (thanks Joshua Mouch).
Fixed parameter serialization where some blobs types were not being handled (thanks Sean McGinnis).
Removed spurious MessageBox.show from
DataReader code (thanks Joshua Mouch).
Fixed a nasty bug in the split sql code (thanks everyone!).
Fixed bug in MySqlStream where too much
data could attempt to be read (thanks Peter Belbin)
Implemented HasRows (thanks Nash Pherson).
Fixed bug where tables with more than 252 columns cause an exception (thanks Joshua Kessler).
Fixed bug where SQL statements ending in ; would cause a problem (thanks Shane Krueger).
Fixed bug in driver where error messages were getting truncated by 1 character (thanks Shane Krueger).
Made MySqlException serializable (thanks
Mathias Hasselmann).
Updated some of the character code pages to be more accurate.
Fixed problem where readers could be opened on connections that had readers open.
Moved test to separate assembly
MySqlClientTests.
Fixed stupid problem in driver with sequence out of order (Thanks Peter Belbin).
Added some pipe tests.
Increased default max pool size to 50.
Compiles with Mono 0-24.
Fixed connection and data reader dispose problems.
Added String datatype handling to parameter
serialization.
Fixed sequence problem in driver that occurred after thrown exception (thanks Burkhard Perkens-Golomb).
Added support for CommandBehavior.SingleRow
to DataReader.
Fixed command sql processing so quotes are better handled (thanks Theo Spears).
Fixed parsing of double, single, and decimal values to account for non-English separators. You still have to use the right syntax if you using hard coded sql, but if you use parameters the code will convert floating point types to use '.' appropriately internal both into the server and out.
Added MySqlStream class to simplify
timeouts and driver coding.
Fixed DataReader so that it is closed
properly when the associated connection is closed. [thanks
smishra]
Made client more SqlClient compliant so that DataReaders have to be closed before the connection can be used to run another command.
Improved DBNull.Value handling in the
fields.
Added several unit tests.
Fixed MySqlException base class.
Improved driver coding
Fixed bug where NextResult was returning false on the last resultset.
Added more tests for MySQL.
Improved casting problems by equating unsigned 32bit values to Int64 and unsigned 16bit values to Int32, and so forth.
Added new constructor for MySqlParameter
for (name, type, size, srccol)
Fixed bug in MySqlDataReader where it
didn't check for null fieldlist before returning field count.
Started adding MySqlClient unit tests
(added MySqlClient/Tests folder and some
test cases).
Fixed some things in Connection String handling.
Moved INIT_DB to
MySqlPool. I may move it again, this is in
preparation of the conference.
Fixed bug inside CommandBuilder that
prevented inserts from happening properly.
Reworked some of the internals so that all three execute methods of Command worked properly.
Fixed many small bugs found during benchmarking.
The first cut of CoonectionPooling is
working. "min pool size" and "max pool size" are respected.
Work to enable multiple resultsets to be returned.
Character sets are handled much more intelligently now. The driver queries MySQL at startup for the default character set. That character set is then used for conversions if that code page can be loaded. If not, then the default code page for the current OS is used.
Added code to save the inferred type in the name,value
constructor of Parameter.
Also, inferred type if value of null parameter is changed
using Value property.
Converted all files to use proper Camel case. MySQL is now MySql in all files. PgSQL is now PgSql.
Added attribute to PgSql code to prevent designer from trying to show.
Added MySQLDbType property to Parameter
object and added proper conversion code to convert from
DbType to MySQLDbType).
Removed unused ObjectToString method from
MySQLParameter.cs.
Fixed Add(..) method in
ParameterCollection so that it doesn't use
Add(name, value) instead.
Fixed IndexOf and
Contains in
ParameterCollection to be aware that
parameter names are now stored without @.
Fixed Command.ConvertSQLToBytes so it only
allows characters that can be in MySQL variable names.
Fixed DataReader and
Field so that blob fields read their data
from Field.cs and
GetBytes works right.
Added simple query builder editor to
CommandText property of
MySQLCommand.
Fixed CommandBuilder and
Parameter serialization to account for
Parameters not storing @ in their names.
Removed MySQLFieldType enum from Field.cs.
Now using MySQLDbType enum.
Added Designer attribute to several classes
to prevent designer view when using VS.Net.
Fixed Initial catalog typo in
ConnectionString designer.
Removed 3 parameter constructor for
MySQLParameter that conflicted with (name,
type, value).
Changed MySQLParameter so
paramName is now stored without leading @
(this fixed null inserts when using designer).
Changed TypeConverter for
MySQLParameter to use the constructor with
all properties.
Fixed sequence issue in driver.
Added DbParametersEditor to make parameter
editing more like SqlClient.
Fixed Command class so that parameters can
be edited using the designer
Update connection string designer to support Use
Compression flag.
Fixed string encoding so that European characters will work correctly.
Creating base classes to aid in building new data providers.
Added support for UID key in connection string.
Field, parameter, command now using DBNull.Value instead of null.
CommandBuilder using
DBNull.Value.
CommandBuilder now builds insert command
correctly when an auto_insert field is not present.
Field now uses typeof keyword to return
System.Types (performance).
MySQLCommandBuilder now implemented.
Transaction support now implemented (not all table types support this).
GetSchemaTable fixed to not use xsd (for
Mono).
Driver is now Mono-compatible.
TIME data type now supported.
More work to improve Timestamp data type handling.
Changed signatures of all classes to match corresponding
SqlClient classes.
Protocol compression using SharpZipLib (www.icsharpcode.net).
Named pipes on Windows now working properly.
Work done to improve Timestamp data type
handling.
Implemented IEnumerable on
DataReader so DataGrid
would work.
As of Connector/NET 5.1.2 (14 June 2007), the Visual Studion Plugin is part of the main Connector/NET package. For the change history for the Visual Studio Plugin, see Section C.5, “MySQL Connector/NET Change History”.
Bugs fixed:
Running queries based on a stored procedure would cause the data set designer to terminate. (Bugs #26364)
DataSet wizard would show all tables instead of only the tables available within the selected database. (Bugs #26348)
Bugs fixed:
The Add Connection dialog of the Server Explorer would freeze when accessing databases with capitalized characters in their name. (Bug#24875)
Creating a connection through the Server Explorer when using the Visual Studio Plugin would fail. The installer for the Visual Studio Plugin has been updated to ensure that Connector/NET 5.0.2 must be installed. (Bug#23071)
This is a bug fix release to resolve an incompatibility issue with Connector/NET 5.0.1.
It is critical that this release only be used with Connector/NET
5.0.1. After installing Connector/NET 5.0.1, you will need to
make a small change in your machine.config file. This file
should be located at
%win%\Microsoft.Net\Framework\v2.0.50727\CONFIG\machine.config
(%win% should be the location of your Windows
folder). Near the bottom of the file you will see a line like
this:
<add name="MySQL Data Provider" invariant="MySql.Data.MySqlClient" description=".Net Framework Data Provider for MySQL" type="MySql.Data.MySqlClient.MySqlClientFactory, MySql.Data"/>
It needs to be changed to be like this:
<add name="MySQL Data Provider" invariant="MySql.Data.MySqlClient" description=".Net Framework Data Provider for MySQL" type="MySql.Data.MySqlClient.MySqlClientFactory, MySql.Data, Version=5.0.1.0, Culture=neutral, PublicKeyToken=c5687fc88969c44d"/>
Bugs fixed:
The reported milliseconds since the last server packets were received/sent was incorrect by a factor of 1000. For example, the following method call:
SQLError.createLinkFailureMessageBasedOnHeuristics( (ConnectionImpl) this.conn, System.currentTimeMillis() - 1000, System.currentTimeMillis() - 2000, e, false);
returned the following string:
The last packet successfully received from the server was 2 milliseconds ago. The last packet sent successfully to the server was 1 milliseconds ago.
Calling Connection.serverPrepareStatement()
variants that do not take result set type or concurrency
arguments returned statements that produced result sets with
incorrect defaults, namely
TYPE_SCROLL_SENSITIVE.
(Bug#45171)
The result set returned by getIndexInfo() did
not have the format defined in the JDBC API specifications. The
fourth column, DATA_TYPE, of the result set
should be of type BOOLEAN. Connector/J
however returns CHAR.
(Bug#44869)
The result set returned by getTypeInfo() did
not have the format defined in the JDBC API specifications. The
second column, DATA_TYPE, of the result set
should be of type INTEGER. Connector/J
however returns SMALLINT.
(Bug#44868)
The DEFERRABILITY column in database metadata
result sets was expected to be of type SHORT.
However, Connector/J returned it as INTEGER.
This affected the following methods:
getImportedKeys(),
getExportedKeys(),
getCrossReference().
(Bug#44867)
The result set returned by getColumns() did
not have the format defined in the JDBC API specifications. The
fifth column, DATA_TYPE, of the result set
should be of type INTEGER. Connector/J
however returns SMALLINT.
(Bug#44865)
The result set returned by
getVersionColumns() did not have the format
defined in the JDBC API specifications. The third column,
DATA_TYPE, of the result set should be of
type INTEGER. Connector/J however returns
SMALLINT.
(Bug#44863)
The result set returned by
getBestRowIdentifier() did not have the
format defined in the JDBC API specifications. The third column,
DATA_TYPE, of the result set should be of
type INTEGER. Connector/J however returns
SMALLINT.
(Bug#44862)
Connector/J contains logic to generate a message text
specifically for streaming result sets when there are
CommunicationsException exceptions generated.
However, this code was never reached.
In the CommunicationsException code:
private boolean streamingResultSetInPlay = false;
public CommunicationsException(ConnectionImpl conn, long lastPacketSentTimeMs,
long lastPacketReceivedTimeMs, Exception underlyingException) {
this.exceptionMessage = SQLError.createLinkFailureMessageBasedOnHeuristics(conn,
lastPacketSentTimeMs, lastPacketReceivedTimeMs, underlyingException,
this.streamingResultSetInPlay);
streamingResultSetInPlay was always false,
which in the following code in
SQLError.createLinkFailureMessageBasedOnHeuristics()
never being executed:
if (streamingResultSetInPlay) {
exceptionMessageBuf.append(
Messages.getString("CommunicationsException.ClientWasStreaming")); //$NON-NLS-1$
} else {
...
The
SQLError.createLinkFailureMessageBasedOnHeuristics()
method created a message text for communication link failures.
When certain conditions were met, this message included both
“last packet sent” and “last packet
received” information, but when those conditions were not
met, only “last packet sent” information was
provided.
Information about when the last packet was successfully received should be provided in all cases. (Bug#44587)
Statement.getGeneratedKeys() retained result
set instances until the statement was closed. This caused memory
leaks for long-lived statements, or statements used in tight
loops.
(Bug#44056)
Using useInformationSchema with
DatabaseMetaData.getExportedKeys() generated
the following exception:
com.mysql.jdbc.exceptions.MySQLIntegrityConstraintViolationException: Column 'REFERENCED_TABLE_NAME' in where clause is ambiguous ... at com.mysql.jdbc.PreparedStatement.executeInternal(PreparedStatement.java:1772) at com.mysql.jdbc.PreparedStatement.executeQuery(PreparedStatement.java:1923) at com.mysql.jdbc.DatabaseMetaDataUsingInfoSchema.executeMetadataQuery( DatabaseMetaDataUsingInfoSchema.java:50) at com.mysql.jdbc.DatabaseMetaDataUsingInfoSchema.getExportedKeys( DatabaseMetaDataUsingInfoSchema.java:603)
LoadBalancingConnectionProxy.doPing() did not
have blacklist awareness.
LoadBalancingConnectionProxy implemented
doPing() to ping all underlying connections,
but it threw any exceptions it encountered during this process.
With the global blacklist enabled, it catches these exceptions, adds the host to the global blacklist, and only throws an exception if all hosts are down. (Bug#43421)
When the MySQL Server was upgraded from 4.0 to 5.0, the Connector/J application then failed to connect to the server. This was because authentication failed when the application ran from EBCDIC platforms such as z/OS. (Bug#43071)
When connecting with traceProtocol=true, no
trace data was generated for the server greeting or login
request.
(Bug#43070)
Connector/J generated an unhandled
StringIndexOutOfBoundsException:
java.lang.StringIndexOutOfBoundsException: String index out of range: -1 at java.lang.String.substring(String.java:1938) at com.mysql.jdbc.EscapeProcessor.processTimeToken(EscapeProcessor.java:353) at com.mysql.jdbc.EscapeProcessor.escapeSQL(EscapeProcessor.java:257) at com.mysql.jdbc.StatementImpl.executeUpdate(StatementImpl.java:1546) at com.mysql.jdbc.StatementImpl.executeUpdate(StatementImpl.java:1524)
A ConcurrentModificationException was
generated in LoadBalancingConnectionProxy:
java.util.ConcurrentModificationException at java.util.HashMap$HashIterator.nextEntry(Unknown Source) at java.util.HashMap$KeyIterator.next(Unknown Source) at com.mysql.jdbc.LoadBalancingConnectionProxy.getGlobalBlacklist(LoadBalancingConnectionProxy.java:520) at com.mysql.jdbc.RandomBalanceStrategy.pickConnection(RandomBalanceStrategy.java:55) at com.mysql.jdbc.LoadBalancingConnectionProxy.pickNewConnection(LoadBalancingConnectionProxy.java:414) at com.mysql.jdbc.LoadBalancingConnectionProxy.invoke(LoadBalancingConnectionProxy.java:390)
SQL injection was possible when using a string containing U+00A5 in a client-side prepared statement, and the character set being used was SJIS/Windows-31J. (Bug#41730)
If there was an apostrophe in a comment in a statement that was
being sent through Connector/J, the apostrophe was still
recognized as a quote and put the state machine in
EscapeTokenizer into the
inQuotes state. This led to further parse
errors.
For example, consider the following statement:
String sql = "-- Customer's zip code will be fixed\n" +
"update address set zip_code = 99999\n" +
"where not regexp '^[0-9]{5}([[.-.]])?([0-9]{4})?$'";
When passed through Connector/J, the
EscapeTokenizer did not recognize that the
first apostrophe was in a comment and thus set
inQuotes to true. When that happened, the
quote count was incorrect and thus the regular expression did
not appear to be in quotes. With the parser not detecting that
the regular expression was in quotes, the curly braces were
recognized as escape sequences and were removed from the regular
expression, breaking it. The server thus received SQL such as:
-- Customer's zip code will be fixed update address set zip_code = '99999' where not regexp '^[0-9]([[.-.]])?([0-9])?$'
MySQL Connector/J 5.1.7 was slower than previous versions when
the rewriteBatchedStatements option was set
to true.
The performance regression in
indexOfIgnoreCaseRespectMarker()has been
fixed. It has also been made possible for the driver to
rewrite INSERT statements with ON
DUPLICATE KEY UPDATE clauses in them, as long as the
UPDATE clause contains no reference to
LAST_INSERT_ID(), as that would cause the
driver to return bogus values for
getGeneratedKeys() invocations. This has
resulted in improved performance over version 5.1.7.
When accessing a result set column by name using
ResultSetImpl.findColumn() an exception was
generated:
java.lang.NullPointerException at com.mysql.jdbc.ResultSetImpl.findColumn(ResultSetImpl.java:1103) at com.mysql.jdbc.ResultSetImpl.getShort(ResultSetImpl.java:5415) at org.apache.commons.dbcp.DelegatingResultSet.getShort(DelegatingResultSet.java:219) at com.zimbra.cs.db.DbVolume.constructVolume(DbVolume.java:297) at com.zimbra.cs.db.DbVolume.get(DbVolume.java:197) at com.zimbra.cs.db.DbVolume.create(DbVolume.java:95) at com.zimbra.cs.store.Volume.create(Volume.java:227) at com.zimbra.cs.store.Volume.create(Volume.java:189) at com.zimbra.cs.service.admin.CreateVolume.handle(CreateVolume.java:48) at com.zimbra.soap.SoapEngine.dispatchRequest(SoapEngine.java:428) at com.zimbra.soap.SoapEngine.dispatch(SoapEngine.java:285)
When DatabaseMetaData.getProcedureColumns()
was called, the value for LENGTH was always
returned as 65535, regardless of the column type (fixed or
variable) or the actual length of the column.
However, if you obtained the PRECISION value,
this was correct for both fixed and variable length columns.
(Bug#41269)
PreparedStatement.addBatch() did not check
for all parameters being set, which led to inconsistent behavior
in executeBatch(), especially when rewriting
batched statements into multi-value INSERTs.
(Bug#41161)
Error message strings contained variable values that were not expanded. For example:
Mon Nov 17 11:43:18 JST 2008 WARN: Invalid value {1} for server variable named {0},
falling back to sane default of {2}
When using rewriteBatchedStatements=true
with:
INSERT INTO table_name_values (...) VALUES (...)
Query rewriting failed because “values” at the end of the table name was mistaken for the reserved keyword. The error generated was as follows:
testBug40439(testsuite.simple.TestBug40439)java.sql.BatchUpdateException: You have an
error in your SQL syntax; check the manual that corresponds to your MySQL server version
for the right syntax to use near 'values (2,'toto',2),(id,data, ordr) values
(3,'toto',3),(id,data, ordr) values (' at line 1
at com.mysql.jdbc.PreparedStatement.executeBatchedInserts(PreparedStatement.java:1495)
at com.mysql.jdbc.PreparedStatement.executeBatch(PreparedStatement.java:1097)
at testsuite.simple.TestBug40439.testBug40439(TestBug40439.java:42)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at testsuite.simple.TestBug40439.main(TestBug40439.java:57)
Using Connector/J 5.1.6 the method
ResultSet.getObject returned a
BYTE[] for following:
SELECT TRIM(rowid) FROM tbl
Where rowid had a type of INT(11)
PRIMARY KEY AUTO_INCREMENT.
The expected return type was one of CHAR,
VARCHAR, CLOB, however, a
BYTE[] was returned.
Further, adding
functionsNeverReturnBlobs=true to the
connection string did not have any effect on the return type.
(Bug#38387)
Functionality added or changed:
When statements include ON DUPLICATE UPDATE,
and rewriteBatchedStatements is set to true,
batched statements are not rewritten into the form
INSERT INTO table VALUES (), (), (), instead
the statements are executed sequentially.
Bugs fixed:
Statement.getGeneratedKeys() returned two
keys when using ON DUPLICATE KEY UPDATE and
the row was updated, not inserted.
(Bug#42309)
When using the replication driver with
autoReconnect=true, Connector/J checks in
PreparedStatement.execute (also called by
CallableStatement.execute) to determine if
the first character of the statement is an “S”, in
an attempt to block all statements that are not read-only-safe,
for example non-SELECT
statements. However, this also blocked
CALLs to stored procedures, even
if the stored procedures were defined as SQL READ
DATA or NO SQL.
(Bug#40031)
With large result sets ResultSet.findColumn
became a performance bottleneck.
(Bug#39962)
Connector/J ignored the value of the MySQL Server variable
auto_increment_increment.
(Bug#39956)
Connector/J failed to parse
TIMESTAMP strings for nanos
correctly.
(Bug#39911)
When the LoadBalancingConnectionProxy handles
a SQLException with SQL state starting with
“08”, it calls
invalidateCurrentConnection, which in turn
removes that Connection from
liveConnections and the
connectionsToHostsMap, but it did not add the
host to the new global blacklist, if the global blacklist was
enabled.
There was also the possibility of a
NullPointerException when trying to update
stats, where
connectionsToHostsMap.get(this.currentConn)
was called:
int hostIndex = ((Integer) this.hostsToListIndexMap.get(this.connectionsToHostsMap.get(this.currentConn))).intValue();
This could happen if a client tried to issue a rollback after
catching a SQLException caused by a
connection failure.
(Bug#39784)
When configuring the Java Replication Driver the last slave specified was never used. (Bug#39611)
When an INSERT ON DUPLICATE KEY UPDATE was
performed, and the key already existed, the
affected-rows value was returned as 1 instead
of 0.
(Bug#39352)
When using the random load balancing strategy and starting with
two servers that were both unavailable, an
IndexOutOfBoundsException was generated when
removing a server from the whiteList.
(Bug#38782)
Connector/J threw the following exception when using a read-only connection:
java.sql.SQLException: Connection is read-only. Queries leading to data
modification are not allowed.
Connector/J was unable to connect when using a
non-latin1 password.
(Bug#37570)
Incorrect result is returned from
isAfterLast() in streaming
ResultSet when using
setFetchSize(Integer.MIN_VALUE).
(Bug#35170)
When getGeneratedKeys() was called on a
statement that had not been created with
RETURN_GENERATED_KEYS, no exception was
thrown, and batched executions then returned erroneous values.
(Bug#34185)
The loadBalance
bestResponseTime blacklists did not have a
global state.
(Bug#33861)
Functionality added or changed:
Multiple result sets were not supported when using streaming
mode to return data. Both normal statements and the resul sets
from stored procedures now return multiple results sets, with
the exception of result sets using registered
OUTPUT paramaters.
(Bug#33678)
XAConnections and datasources have been updated to the JDBC-4.0 standard.
The profiler event handling has been made extensible via the
profilerEventHandler connection property.
Add the verifyServerCertificate propery. If
set to "false" the driver will not verify the server's
certificate when useSSL is set to "true"
When using this feature, the keystore parameters should be
specified by the clientCertificateKeyStore*
properties, rather than system properties, as the JSSE doesn't
it straightforward to have a nonverifying trust store and the
"default" key store.
Bugs fixed:
DatabaseMetaData.getColumns() returns
incorrect COLUMN_SIZE value for
SET column.
(Bug#36830)
When trying to read Time values like
“00:00:00” with
ResultSet.getTime(int) an exception is
thrown.
(Bug#36051)
JDBC connection URL parameters is ignored when using
MysqlConnectionPoolDataSource.
(Bug#35810)
When useServerPrepStmts=true and slow query
logging is enabled, the connector throws a
NullPointerException when it encounters a
slow query.
(Bug#35666)
When using the keyword “loadbalance” in the connection string and trying to perform load balancing between two databases, the driver appears to hang. (Bug#35660)
JDBC data type getter method was changed to accept only column name, whereas previously it accepted column label. (Bug#35610)
Prepared statements from pooled connections caused a
NullPointerException when
closed() under JDBC-4.0.
(Bug#35489)
In calling a stored function returning a
bigint, an exception is encountered
beginning:
java.sql.SQLException: java.lang.NumberFormatException: For input string:
followed by the text of the stored function starting after the argument list. (Bug#35199)
The JDBC driver uses a different method for evaluating column
names in
resultsetmetadata.getColumnName() and
when looking for a column in
resultset.getObject(columnName). This
causes Hibernate to fail in queries where the two methods yield
different results, for example in queries that use alias names:
SELECT column AS aliasName from table
MysqlConnectionPoolDataSource does not
support ReplicationConnection. Notice that we
implemented com.mysql.jdbc.Connection for
ReplicationConnection, however, only
accessors from ConnectionProperties are implemented (not the
mutators), and they return values from the currently active
connection. All other methods from
com.mysql.jdbc.Connection are implemented,
and operate on the currently active connection, with the
exception of resetServerState() and
changeUser().
(Bug#34937)
ResultSet.getTimestamp() returns incorrect
values for month/day of
TIMESTAMPs when using server-side
prepared statements (not enabled by default).
(Bug#34913)
RowDataStatic does't always set the
metadata in ResultSetRow, which can lead
to failures when unpacking DATE,
TIME,
DATETIME and
TIMESTAMP types when using
absolute, relative, and previous result set navigation methods.
(Bug#34762)
When calling isValid() on an active
connection, if the timeout is nonzero then the
Connection is invalidated even if the
Connection is valid.
(Bug#34703)
It was not possible to truncate a
BLOB using
Blog.truncate() when using 0 as an argument.
(Bug#34677)
When using a cursor fetch for a statement, the internal prepared statement could cause a memory leak until the connection was closed. The internal prepared statement is now deleted when the corresponding result set is closed. (Bug#34518)
When retrieving the column type name of a geometry field, the
driver would return UNKNOWN instead of
GEOMETRY.
(Bug#34194)
Statements with batched values do not return correct values for
getGeneratedKeys() when
rewriteBatchedStatements is set to
true, and the statement has an ON
DUPLICATE KEY UPDATE clause.
(Bug#34093)
The internal class
ResultSetInternalMethods referenced the
nonpublic class
com.mysql.jdbc.CachedResultSetMetaData.
(Bug#33823)
A NullPointerException could be raised when
using client-side prepared statements and enabled the prepared
statement cache using the cachePrepStmts.
(Bug#33734)
Using server side cursors and cursor fetch, the table metadata information would return the data type name instead of the column name. (Bug#33594)
ResultSet.getTimestamp() would throw a
NullPointerException instead of a
SQLException when called on an empty
ResultSet.
(Bug#33162)
Load balancing connection using best response time would incorrectly "stick" to hosts that were down when the connection was first created.
We solve this problem with a black list that is used during the
picking of new hosts. If the black list ends up including all
configured hosts, the driver will retry for a configurable
number of times (the retriesAllDown
configuration property, with a default of 120 times), sleeping
250ms between attempts to pick a new connection.
We've also went ahead and made the balancing strategy
extensible. To create a new strategy, implement the interface
com.mysql.jdbc.BalanceStrategy (which
also includes our standard "extension" interface), and tell the
driver to use it by passing in the class name via the
loadBalanceStrategy configuration property.
(Bug#32877)
During a Daylight Savings Time (DST) switchover, there was no way to store two timestamp/datetime values , as the hours end up being the same when sent as the literal that MySQL requires.
Note that to get this scenario to work with MySQL (since it
doesn't support per-value timezones), you need to configure your
server (or session) to be in UTC, and tell the driver not to use
the legacy date/time code by setting
useLegacyDatetimeCode to "false". This will
cause the driver to always convert to/from the server and client
timezone consistently.
This bug fix also fixes Bug#15604, by adding entirely new
date/time handling code that can be switched on by
useLegacyDatetimeCode being set to "false" as a
JDBC configuration property. For Connector/J 5.1.x, the default
is "true", in trunk and beyond it will be "false" (that is, the
old date/time handling code will be deprecated)
(Bug#32577, Bug#15604)
When unpacking rows directly, we don't hand off error message packets to the internal method which decodes them correctly, so no exception is raised, and the driver than hangs trying to read rows that aren't there. This tends to happen when calling stored procedures, as normal SELECTs won't have an error in this spot in the protocol unless an I/O error occurs. (Bug#32246)
When using a connection from
ConnectionPoolDataSource, some
Connection.prepareStatement() methods would
return null instead of the prepared statement.
(Bug#32101)
Using CallableStatement.setNull() on a
stored function would throw an
ArrayIndexOutOfBounds exception when setting
the last parameter to null.
(Bug#31823)
MysqlValidConnectionChecker doesn't
properly handle connections created using
ReplicationConnection.
(Bug#31790)
Retrieving the server version information for an active connection could return invalid information if the default character encoding on the host was not ASCII compatible. (Bug#31192)
Further fixes have been made to this bug in the event that a node is nonresponsive. Connector/J will now try a different random node instead of waiting for the node to recover before continuing. (Bug#31053)
ResultSet returned by
Statement.getGeneratedKeys() is not closed
automatically when statement that created it is closed.
(Bug#30508)
DatabaseMetadata.getColumns() doesn't
return the correct column names if the connection character
isn't UTF-8. A bug in MySQL server compounded the issue, but was
fixed within the MySQL 5.0 release cycle. The fix includes
changes to all the sections of the code that access the server
metadata.
(Bug#20491)
Fixed ResultSetMetadata.getColumnName()
for result sets returned from
Statement.getGeneratedKeys() - it was
returning null instead of "GENERATED_KEY" as in 5.0.x.
The following features are new, compared to the 5.0 series of Connector/J
JDBC-4.0 support for setting per-connection client information
(which can be viewed in the comments section of a query via
SHOW PROCESSLIST on a MySQL
server, or can be extended to support custom persistence of the
information via a public interface).
Support for JDBC-4.0 XML processing via JAXP interfaces to DOM, SAX and StAX.
JDBC-4.0 standardized unwrapping to interfaces that include vendor extensions.
Functionality added or changed:
Added autoSlowLog configuration property,
overrides slowQueryThreshold* properties,
driver determines slow queries by those that are slower than 5 *
stddev of the mean query time (outside the 96% percentile).
Bugs fixed:
When a connection is in read-only mode, queries that are wrapped in parentheses were incorrectly identified DML statements. (Bug#28256)
The following features are new, compared to the 5.0 series of Connector/J
JDBC-4.0 support for setting per-connection client information
(which can be viewed in the comments section of a query via
SHOW PROCESSLIST on a MySQL
server, or can be extended to support custom persistence of the
information via a public interface).
Support for JDBC-4.0 XML processing via JAXP interfaces to DOM, SAX and StAX.
JDBC-4.0 standardized unwrapping to interfaces that include vendor extensions.
Functionality added or changed:
Connector/J now connects using an initial character set of
utf-8 solely for the purpose of
authentication to allow user names or database names in any
character set to be used in the JDBC connection URL.
(Bug#29853)
Added two configuration parameters:
blobsAreStrings — Should the driver
always treat BLOBs as Strings. Added specifically to work
around dubious metadata returned by the server for
GROUP BY clauses. Defaults to false.
functionsNeverReturnBlobs — Should
the driver always treat data from functions returning
BLOBs as Strings. Added specifically to
work around dubious metadata returned by the server for
GROUP BY clauses. Defaults to false.
Setting rewriteBatchedStatements to
true now causes CallableStatements with
batched arguments to be re-written in the form "CALL (...); CALL
(...); ..." to send the batch in as few client-server round
trips as possible.
The driver now picks appropriate internal row representation
(whole row in one buffer, or individual byte[]s for each column
value) depending on heuristics, including whether or not the row
has BLOB or
TEXT types and the overall
row-size. The threshold for row size that will cause the driver
to use a buffer rather than individual byte[]s is configured by
the configuration property
largeRowSizeThreshold, which has a default
value of 2KB.
The data (and how it is stored) for ResultSet
rows are now behind an interface which allows us (in some cases)
to allocate less memory per row, in that for "streaming" result
sets, we re-use the packet used to read rows, since only one row
at a time is ever active.
Added experimental support for statement "interceptors" via the
com.mysql.jdbc.StatementInterceptor
interface, examples are in
com/mysql/jdbc/interceptors. Implement this
interface to be placed "in between" query execution, so that it
can be influenced (currently experimental).
The driver will automatically adjust the server session variable
net_write_timeout when it
determines its been asked for a "streaming" result, and resets
it to the previous value when the result set has been consumed.
(The configuration property is named
netTimeoutForStreamingResults, with a unit of
seconds, the value '0' means the driver will not try and adjust
this value).
JDBC-4.0 ease-of-development features including
auto-registration with the DriverManager via
the service provider mechanism, standardized Connection validity
checks and categorized SQLExceptions based on
recoverability/retry-ability and class of the underlying error.
Statement.setQueryTimeout()s now affect the
entire batch for batched statements, rather than the individual
statements that make up the batch.
Errors encountered during
Statement/PreparedStatement/CallableStatement.executeBatch()
when rewriteBatchStatements has been set to
true now return
BatchUpdateExceptions according to the
setting of continueBatchOnError.
If continueBatchOnError is set to
true, the update counts for the "chunk" that
were sent as one unit will all be set to
EXECUTE_FAILED, but the driver will attempt
to process the remainder of the batch. You can determine which
"chunk" failed by looking at the update counts returned in the
BatchUpdateException.
If continueBatchOnError is set to "false",
the update counts returned will contain all updates up-to and
including the failed "chunk", with all counts for the failed
"chunk" set to EXECUTE_FAILED.
Since MySQL doesn't return multiple error codes for
multiple-statements, or for multi-value
INSERT/REPLACE,
it is the application's responsibility to handle determining
which item(s) in the "chunk" actually failed.
New methods on com.mysql.jdbc.Statement:
setLocalInfileInputStream() and
getLocalInfileInputStream():
setLocalInfileInputStream() sets an
InputStream instance that will be used to
send data to the MySQL server for a
LOAD DATA LOCAL
INFILE statement rather than a
FileInputStream or
URLInputStream that represents the path
given as an argument to the statement.
This stream will be read to completion upon execution of a
LOAD DATA LOCAL
INFILE statement, and will automatically be closed
by the driver, so it needs to be reset before each call to
execute*() that would cause the MySQL
server to request data to fulfill the request for
LOAD DATA LOCAL
INFILE.
If this value is set to NULL, the driver
will revert to using a FileInputStream or
URLInputStream as required.
getLocalInfileInputStream() returns the
InputStream instance that will be used to
send data in response to a
LOAD DATA LOCAL
INFILE statement.
This method returns NULL if no such
stream has been set via
setLocalInfileInputStream().
Setting useBlobToStoreUTF8OutsideBMP to
true tells the driver to treat
[MEDIUM/LONG]BLOB columns as
[LONG]VARCHAR columns holding text encoded in
UTF-8 that has characters outside the BMP (4-byte encodings),
which MySQL server can't handle natively.
Set utf8OutsideBmpExcludedColumnNamePattern to
a regex so that column names matching the given regex will still
be treated as BLOBs The regex must follow the
patterns used for the java.util.regexpackage.
The default is to exclude no columns, and include all columns.
Set utf8OutsideBmpIncludedColumnNamePattern to
specify exclusion rules to
utf8OutsideBmpExcludedColumnNamePattern". The regex must follow
the patterns used for the java.util.regex
package.
Bugs fixed:
setObject(int, Object, int, int) delegate in
PreparedStatmentWrapper delegates to wrong method.
(Bug#30892)
NPE with null column values when
padCharsWithSpace is set to true.
(Bug#30851)
Collation on VARBINARY column
types would be misidentified. A fix has been added, but this fix
only works for MySQL server versions 5.0.25 and newer, since
earlier versions didn't consistently return correct metadata for
functions, and thus results from subqueries and functions were
indistinguishable from each other, leading to type-related bugs.
(Bug#30664)
An ArithmeticException or
NullPointerException would be raised when the
batch had zero members and
rewriteBatchedStatements=true when
addBatch() was never called, or
executeBatch() was called immediately after
clearBatch().
(Bug#30550)
Closing a load-balanced connection would cause a
ClassCastException.
(Bug#29852)
Connection checker for JBoss didn't use same method parameters via reflection, causing connections to always seem "bad". (Bug#29106)
DatabaseMetaData.getTypeInfo() for the types
DECIMAL and
NUMERIC will return a precision
of 254 for server versions older than 5.0.3, 64 for versions
5.0.3–5.0.5 and 65 for versions newer than 5.0.5.
(Bug#28972)
CallableStatement.executeBatch() doesn't work
when connection property
noAccessToProcedureBodies has been set to
true.
The fix involves changing the behavior of
noAccessToProcedureBodies,in that the driver
will now report all paramters as "IN" paramters but allow
callers to call registerOutParameter() on them without throwing
an exception.
(Bug#28689)
DatabaseMetaData.getColumns() doesn't contain
SCOPE_* or
IS_AUTOINCREMENT columns.
(Bug#27915)
Schema objects with identifiers other than the connection
character aren't retrieved correctly in
ResultSetMetadata.
(Bug#27867)
Connection.getServerCharacterEncoding()
doesn't work for servers with version >= 4.1.
(Bug#27182)
The automated SVN revisions in
DBMD.getDriverVersion(). The SVN revision of
the directory is now inserted into the version information
during the build.
(Bug#21116)
Specifying a "validation query" in your connection pool that starts with "/* ping */" _exactly_ will cause the driver to instead send a ping to the server and return a fake result set (much lighter weight), and when using a ReplicationConnection or a LoadBalancedConnection, will send the ping across all active connections.
This is a new Beta development release, fixing recently discovered bugs.
Functionality added or changed:
Setting the configuration property
rewriteBatchedStatements to
true will now cause the driver to rewrite
batched prepared statements with more than 3 parameter sets in a
batch into multi-statements (separated by ";") if they are not
plain (that is, without SELECT or
ON DUPLICATE KEY UPDATE clauses)
INSERT or
REPLACE statements.
This is a new Alpha development release, adding new features and fixing recently discovered bugs.
Functionality added or changed:
Incompatible Change:
Pulled vendor-extension methods of Connection
implementation out into an interface to support
java.sql.Wrapper functionality from
ConnectionPoolDataSource. The vendor
extensions are javadoc'd in the
com.mysql.jdbc.Connection interface.
For those looking further into the driver implementation, it is
not an API that is used for plugability of implementations
inside our driver (which is why there are still references to
ConnectionImpl throughout the code).
We've also added server and client
prepareStatement() methods that cover all of
the variants in the JDBC API.
Connection.serverPrepare(String) has been
re-named to
Connection.serverPrepareStatement() for
consistency with
Connection.clientPrepareStatement().
Row navigation now causes any streams/readers open on the result set to be closed, as in some cases we're reading directly from a shared network packet and it will be overwritten by the "next" row.
Made it possible to retrieve prepared statement parameter
bindings (to be used in
StatementInterceptors, primarily).
Externalized the descriptions of connection properties.
The data (and how it is stored) for ResultSet
rows are now behind an interface which allows us (in some cases)
to allocate less memory per row, in that for "streaming" result
sets, we re-use the packet used to read rows, since only one row
at a time is ever active.
Similar to Connection, we pulled out vendor
extensions to Statement into an interface
named com.mysql.Statement, and moved the
Statement class into
com.mysql.StatementImpl. The two methods
(javadoc'd in com.mysql.Statement are
enableStreamingResults(), which already
existed, and disableStreamingResults() which
sets the statement instance back to the fetch size and result
set type it had before
enableStreamingResults() was called.
Driver now picks appropriate internal row representation (whole
row in one buffer, or individual byte[]s for each column value)
depending on heuristics, including whether or not the row has
BLOB or
TEXT types and the overall
row-size. The threshold for row size that will cause the driver
to use a buffer rather than individual byte[]s is configured by
the configuration property
largeRowSizeThreshold, which has a default
value of 2KB.
Added experimental support for statement "interceptors" via the
com.mysql.jdbc.StatementInterceptor
interface, examples are in
com/mysql/jdbc/interceptors.
Implement this interface to be placed "in between" query execution, so that you can influence it. (currently experimental).
StatementInterceptors are "chainable" when
configured by the user, the results returned by the "current"
interceptor will be passed on to the next on in the chain, from
left-to-right order, as specified by the user in the JDBC
configuration property statementInterceptors.
See the sources (fully javadoc'd) for
com.mysql.jdbc.StatementInterceptor for more
details until we iron out the API and get it documented in the
manual.
Setting rewriteBatchedStatements to
true now causes
CallableStatements with batched arguments to
be re-written in the form CALL (...); CALL (...);
... to send the batch in as few client-server round
trips as possible.
This is the first public alpha release of the current Connector/J 5.1 development branch, providing an insight to upcoming features. Although some of these are still under development, this release includes the following new features and changes (in comparison to the current Connector/J 5.0 production release):
Important change: Due to a number of issues with the use of server-side prepared statements, Connector/J 5.0.5 has disabled their use by default. The disabling of server-side prepared statements does not affect the operation of the connector in any way.
To enable server-side prepared statements you must add the following configuration property to your connector string:
useServerPrepStmts=true
The default value of this property is false
(that is, Connector/J does not use server-side prepared
statements).
The disabling of server-side prepared statements does not
affect the operation of the connector. However, if you use the
useTimezone=true connection option and use
client-side prepared statements (instead of server-side
prepared statements) you should also set
useSSPSCompatibleTimezoneShift=true.
Functionality added or changed:
Refactored CommunicationsException into a
JDBC-3.0 version, and a JDBC-4.0 version (which extends
SQLRecoverableException, now that it exists).
This change means that if you were catching
com.mysql.jdbc.CommunicationsException in
your applications instead of looking at the SQLState class of
08, and are moving to Java 6 (or newer),
you need to change your imports to that exception to be
com.mysql.jdbc.exceptions.jdbc4.CommunicationsException,
as the old class will not be instantiated for communications
link-related errors under Java 6.
Added support for JDBC-4.0 categorized
SQLExceptions.
Added support for JDBC-4.0's NCLOB, and
NCHAR/NVARCHAR
types.
com.mysql.jdbc.java6.javac — full path
to your Java-6 javac executable
Added support for JDBC-4.0's SQLXML interfaces.
Re-worked Ant buildfile to build JDBC-4.0 classes separately, as well as support building under Eclipse (since Eclipse can't mix/match JDKs).
To build, you must set JAVA_HOME to
J2SDK-1.4.2 or Java-5, and set the following properties on your
Ant command line:
com.mysql.jdbc.java6.javac — full
path to your Java-6 javac executable
com.mysql.jdbc.java6.rtjar — full
path to your Java-6 rt.jar file
New feature — driver will automatically adjust session
variable net_write_timeout when
it determines it has been asked for a "streaming" result, and
resets it to the previous value when the result set has been
consumed. (configuration property is named
netTimeoutForStreamingResults value and has a
unit of seconds, the value 0 means the driver
will not try and adjust this value).
Added support for JDBC-4.0's client information. The backend
storage of information provided via
Connection.setClientInfo() and retrieved by
Connection.getClientInfo() is pluggable by
any class that implements the
com.mysql.jdbc.JDBC4ClientInfoProvider
interface and has a no-args constructor.
The implementation used by the driver is configured using the
clientInfoProvider configuration property
(with a default of value of
com.mysql.jdbc.JDBC4CommentClientInfoProvider,
an implementation which lists the client information as a
comment prepended to every query sent to the server).
This functionality is only available when using Java-6 or newer.
com.mysql.jdbc.java6.rtjar — full path
to your Java-6 rt.jar file
Added support for JDBC-4.0's Wrapper
interface.
Functionality added or changed:
blobsAreStrings — Should the driver
always treat BLOBs as Strings. Added specifically to work around
dubious metadata returned by the server for GROUP
BY clauses. Defaults to false.
Added two configuration parameters:
blobsAreStrings — Should the driver
always treat BLOBs as Strings. Added specifically to work
around dubious metadata returned by the server for
GROUP BY clauses. Defaults to false.
functionsNeverReturnBlobs — Should
the driver always treat data from functions returning
BLOBs as Strings. Added specifically to
work around dubious metadata returned by the server for
GROUP BY clauses. Defaults to false.
functionsNeverReturnBlobs — Should the
driver always treat data from functions returning
BLOBs as Strings. Added specifically to work
around dubious metadata returned by the server for
GROUP BY clauses. Defaults to false.
XAConnections now start in auto-commit mode (as per JDBC-4.0 specification clarification).
Driver will now fall back to sane defaults for
max_allowed_packet and
net_buffer_length if the server
reports them incorrectly (and will log this situation at
WARN level, since it is actually an error
condition).
Bugs fixed:
Connections established using URLs of the form
jdbc:mysql:loadbalance:// weren't doing
failover if they tried to connect to a MySQL server that was
down. The driver now attempts connections to the next "best"
(depending on the load balance strategy in use) server, and
continues to attempt connecting to the next "best" server every
250 milliseconds until one is found that is up and running or 5
minutes has passed.
If the driver gives up, it will throw the last-received
SQLException.
(Bug#31053)
setObject(int, Object, int, int) delegate in
PreparedStatmentWrapper delegates to wrong method.
(Bug#30892)
NPE with null column values when
padCharsWithSpace is set to true.
(Bug#30851)
Collation on VARBINARY column
types would be misidentified. A fix has been added, but this fix
only works for MySQL server versions 5.0.25 and newer, since
earlier versions didn't consistently return correct metadata for
functions, and thus results from subqueries and functions were
indistinguishable from each other, leading to type-related bugs.
(Bug#30664)
An ArithmeticException or
NullPointerException would be raised when the
batch had zero members and
rewriteBatchedStatements=true when
addBatch() was never called, or
executeBatch() was called immediately after
clearBatch().
(Bug#30550)
Closing a load-balanced connection would cause a
ClassCastException.
(Bug#29852)
Connection checker for JBoss didn't use same method parameters via reflection, causing connections to always seem "bad". (Bug#29106)
DatabaseMetaData.getTypeInfo() for the types
DECIMAL and
NUMERIC will return a precision
of 254 for server versions older than 5.0.3, 64 for versions
5.0.3–5.0.5 and 65 for versions newer than 5.0.5.
(Bug#28972)
CallableStatement.executeBatch() doesn't work
when connection property
noAccessToProcedureBodies has been set to
true.
The fix involves changing the behavior of
noAccessToProcedureBodies,in that the driver
will now report all paramters as "IN" paramters but allow
callers to call registerOutParameter() on them without throwing
an exception.
(Bug#28689)
When a connection is in read-only mode, queries that are wrapped in parentheses were incorrectly identified DML statements. (Bug#28256)
UNSIGNED types not reported via
DBMD.getTypeInfo(), and capitalization of
type names is not consistent between
DBMD.getColumns(),
RSMD.getColumnTypeName() and
DBMD.getTypeInfo().
This fix also ensures that the precision of UNSIGNED
MEDIUMINT and UNSIGNED BIGINT is
reported correctly via DBMD.getColumns().
(Bug#27916)
DatabaseMetaData.getColumns() doesn't contain
SCOPE_* or
IS_AUTOINCREMENT columns.
(Bug#27915)
Schema objects with identifiers other than the connection
character aren't retrieved correctly in
ResultSetMetadata.
(Bug#27867)
Cached metadata with
PreparedStatement.execute() throws
NullPointerException.
(Bug#27412)
Connection.getServerCharacterEncoding()
doesn't work for servers with version >= 4.1.
(Bug#27182)
The automated SVN revisions in
DBMD.getDriverVersion(). The SVN revision of
the directory is now inserted into the version information
during the build.
(Bug#21116)
Specifying a "validation query" in your connection pool that starts with "/* ping */" _exactly_ will cause the driver to instead send a ping to the server and return a fake result set (much lighter weight), and when using a ReplicationConnection or a LoadBalancedConnection, will send the ping across all active connections.
Functionality added or changed:
The driver will now automatically set
useServerPrepStmts to true
when useCursorFetch has been set to
true, since the feature requires server-side
prepared statements in order to function.
tcpKeepAlive - Should the driver set
SO_KEEPALIVE (default true)?
Give more information in EOFExceptions thrown out of MysqlIO (how many bytes the driver expected to read, how many it actually read, say that communications with the server were unexpectedly lost).
Driver detects when it is running in a ColdFusion MX server
(tested with version 7), and uses the configuration bundle
coldFusion, which sets
useDynamicCharsetInfo to
false (see previous entry), and sets
useLocalSessionState and autoReconnect to
true.
tcpNoDelay - Should the driver set
SO_TCP_NODELAY (disabling the Nagle Algorithm, default
true)?
Added configuration property
slowQueryThresholdNanos - if
useNanosForElapsedTime is set to
true, and this property is set to a nonzero
value the driver will use this threshold (in nanosecond units)
to determine if a query was slow, instead of using millisecond
units.
tcpRcvBuf - Should the driver set SO_RCV_BUF
to the given value? The default value of '0', means use the
platform default value for this property.
Setting useDynamicCharsetInfo to
false now causes driver to use static lookups
for collations as well (makes
ResultSetMetadata.isCaseSensitive() much more efficient, which
leads to performance increase for ColdFusion, which calls this
method for every column on every table it sees, it appears).
Added configuration properties to allow tuning of TCP/IP socket parameters:
tcpNoDelay - Should the driver set
SO_TCP_NODELAY (disabling the Nagle Algorithm, default
true)?
tcpKeepAlive - Should the driver set
SO_KEEPALIVE (default true)?
tcpRcvBuf - Should the driver set
SO_RCV_BUF to the given value? The default value of '0',
means use the platform default value for this property.
tcpSndBuf - Should the driver set
SO_SND_BUF to the given value? The default value of '0',
means use the platform default value for this property.
tcpTrafficClass - Should the driver set
traffic class or type-of-service fields? See the
documentation for java.net.Socket.setTrafficClass() for more
information.
Setting the configuration parameter
useCursorFetch to true for
MySQL-5.0+ enables the use of cursors that allow Connector/J to
save memory by fetching result set rows in chunks (where the
chunk size is set by calling setFetchSize() on a Statement or
ResultSet) by using fully-materialized cursors on the server.
tcpSndBuf - Should the driver set SO_SND_BUF
to the given value? The default value of '0', means use the
platform default value for this property.
tcpTrafficClass - Should the driver set
traffic class or type-of-service fields? See the documentation
for java.net.Socket.setTrafficClass() for more information.
Added new debugging functionality - Setting configuration
property
includeInnodbStatusInDeadlockExceptions to
true will cause the driver to append the
output of SHOW
ENGINE INNODB STATUS to deadlock-related exceptions,
which will enumerate the current locks held inside InnoDB.
Added configuration property
useNanosForElapsedTime - for
profiling/debugging functionality that measures elapsed time,
should the driver try to use nanoseconds resolution if available
(requires JDK >= 1.5)?
If useNanosForElapsedTime is set to
true, and this property is set to "0" (or
left default), then elapsed times will still be measured in
nanoseconds (if possible), but the slow query threshold will
be converted from milliseconds to nanoseconds, and thus have
an upper bound of approximately 2000 milliseconds (as that
threshold is represented as an integer, not a long).
Bugs fixed:
Don't send any file data in response to LOAD DATA LOCAL INFILE if the feature is disabled at the client side. This is to prevent a malicious server or man-in-the-middle from asking the client for data that the client is not expecting. Thanks to Jan Kneschke for discovering the exploit and Andrey "Poohie" Hristov, Konstantin Osipov and Sergei Golubchik for discussions about implications and possible fixes. (Bug#29605)
Parser in client-side prepared statements runs to end of statement, rather than end-of-line for '#' comments. Also added support for '--' single-line comments. (Bug#28956)
Parser in client-side prepared statements eats character following '/' if it is not a multi-line comment. (Bug#28851)
PreparedStatement.getMetaData() for statements containing leading one-line comments is not returned correctly.
As part of this fix, we also overhauled detection of DML for
executeQuery() and
SELECTs for
executeUpdate() in plain and prepared
statements to be aware of the same types of comments.
(Bug#28469)
Functionality added or changed:
Added an experimental load-balanced connection designed for use
with SQL nodes in a MySQL Cluster/NDB environment (This is not
for master-slave replication. For that, we suggest you look at
ReplicationConnection or
lbpool).
If the JDBC URL starts with
jdbc:mysql:loadbalance://host-1,host-2,...host-n,
the driver will create an implementation of
java.sql.Connection that load balances
requests across a series of MySQL JDBC connections to the given
hosts, where the balancing takes place after transaction commit.
Therefore, for this to work (at all), you must use transactions, even if only reading data.
Physical connections to the given hosts will not be created until needed.
The driver will invalidate connections that it detects have had communication errors when processing a request. A new connection to the problematic host will be attempted the next time it is selected by the load balancing algorithm.
There are two choices for load balancing algorithms, which may
be specified by the loadBalanceStrategy JDBC
URL configuration property:
random — the driver will pick a
random host for each request. This tends to work better than
round-robin, as the randomness will somewhat account for
spreading loads where requests vary in response time, while
round-robin can sometimes lead to overloaded nodes if there
are variations in response times across the workload.
bestResponseTime — the driver will
route the request to the host that had the best response
time for the previous transaction.
bestResponseTime — the driver will
route the request to the host that had the best response time
for the previous transaction.
Added configuration property
padCharsWithSpace (defaults to
false). If set to true,
and a result set column has the
CHAR type and the value does not
fill the amount of characters specified in the DDL for the
column, the driver will pad the remaining characters with space
(for ANSI compliance).
When useLocalSessionState is set to
true and connected to a MySQL-5.0 or later
server, the JDBC driver will now determine whether an actual
commit or rollback
statement needs to be sent to the database when
Connection.commit() or
Connection.rollback() is called.
This is especially helpful for high-load situations with
connection pools that always call
Connection.rollback() on connection
check-in/check-out because it avoids a round-trip to the server.
Added configuration property
useDynamicCharsetInfo. If set to
false (the default), the driver will use a
per-connection cache of character set information queried from
the server when necessary, or when set to
true, use a built-in static mapping that is
more efficient, but isn't aware of custom character sets or
character sets implemented after the release of the JDBC driver.
This only affects the padCharsWithSpace
configuration property and the
ResultSetMetaData.getColumnDisplayWidth()
method.
New configuration property,
enableQueryTimeouts (default
true).
When enabled, query timeouts set via
Statement.setQueryTimeout() use a shared
java.util.Timer instance for scheduling. Even
if the timeout doesn't expire before the query is processed,
there will be memory used by the TimerTask
for the given timeout which won't be reclaimed until the time
the timeout would have expired if it hadn't been cancelled by
the driver. High-load environments might want to consider
disabling this functionality. (this configuration property is
part of the maxPerformance configuration
bundle).
Give better error message when "streaming" result sets, and the
connection gets clobbered because of exceeding
net_write_timeout on the
server.
random — the driver will pick a random
host for each request. This tends to work better than
round-robin, as the randomness will somewhat account for
spreading loads where requests vary in response time, while
round-robin can sometimes lead to overloaded nodes if there are
variations in response times across the workload.
com.mysql.jdbc.[NonRegistering]Driver now
understands URLs of the format
jdbc:mysql:replication:// and
jdbc:mysql:loadbalance:// which will create a
ReplicationConnection (exactly like when using
[NonRegistering]ReplicationDriver) and an
experimental load-balanced connection designed for use with SQL
nodes in a MySQL Cluster/NDB environment, respectively.
In an effort to simplify things, we're working on deprecating
multiple drivers, and instead specifying different core behavior
based upon JDBC URL prefixes, so watch for
[NonRegistering]ReplicationDriver to
eventually disappear, to be replaced with
com.mysql.jdbc[NonRegistering]Driver with the
new URL prefix.
Fixed issue where a failed-over connection would let an
application call setReadOnly(false), when
that call should be ignored until the connection is reconnected
to a writable master unless failoverReadOnly
had been set to false.
Driver will now use INSERT INTO ... VALUES
(DEFAULT)form of statement for updatable result sets
for ResultSet.insertRow(), rather than
pre-populating the insert row with values from
DatabaseMetaData.getColumns()(which results
in a SHOW FULL
COLUMNS on the server for every result set). If an
application requires access to the default values before
insertRow() has been called, the JDBC URL
should be configured with
populateInsertRowWithDefaultValues set to
true.
This fix specifically targets performance issues with ColdFusion and the fact that it seems to ask for updatable result sets no matter what the application does with them.
More intelligent initial packet sizes for the "shared" packets are used (512 bytes, rather than 16K), and initial packets used during handshake are now sized appropriately as to not require reallocation.
Bugs fixed:
More useful error messages are generated when the driver thinks a result set is not updatable. (Thanks to Ashley Martens for the patch). (Bug#28085)
Connection.getTransactionIsolation() uses
"SHOW VARIABLES LIKE" which is very
inefficient on MySQL-5.0+ servers.
(Bug#27655)
Fixed issue where calling getGeneratedKeys()
on a prepared statement after calling
execute() didn't always return the generated
keys (executeUpdate() worked fine however).
(Bug#27655)
CALL /* ... */
doesn't work.
As a side effect of this fix, you can now use some_proc()/*
*/ and # comments when preparing
statements using client-side prepared statement emulation.
If the comments happen to contain parameter markers
(?), they will be treated as belonging to the
comment (that is, not recognized) rather than being a parameter
of the statement.
The statement when sent to the server will contain the
comments as-is, they're not stripped during the process of
preparing the PreparedStatement or
CallableStatement.
ResultSet.get*() with a column index < 1
returns misleading error message.
(Bug#27317)
Using ResultSet.get*() with a column index
less than 1 returns a misleading error message.
(Bug#27317)
Comments in DDL of stored procedures/functions confuse procedure parser, and thus metadata about them can not be created, leading to inability to retrieve said metadata, or execute procedures that have certain comments in them. (Bug#26959)
Fast date/time parsing doesn't take into account
00:00:00 as a legal value.
(Bug#26789)
PreparedStatement is not closed in
BlobFromLocator.getBytes().
(Bug#26592)
When the configuration property
useCursorFetch was set to
true, sometimes server would return new, more
exact metadata during the execution of the server-side prepared
statement that enables this functionality, which the driver
ignored (using the original metadata returned during
prepare()), causing corrupt reading of data
due to type mismatch when the actual rows were returned.
(Bug#26173)
CallableStatements with
OUT/INOUT parameters that are "binary"
(BLOB,
BIT,
(VAR)BINARY, JAVA_OBJECT)
have extra 7 bytes.
(Bug#25715)
Whitespace surrounding storage/size specifiers in stored
procedure parameters declaration causes
NumberFormatException to be thrown when
calling stored procedure on JDK-1.5 or newer, as the Number
classes in JDK-1.5+ are whitespace intolerant.
(Bug#25624)
Client options not sent correctly when using SSL, leading to stored procedures not being able to return results. Thanks to Don Cohen for the bug report, testcase and patch. (Bug#25545)
Statement.setMaxRows() is not effective on
result sets materialized from cursors.
(Bug#25517)
BIT(> 1) is returned as
java.lang.String from
ResultSet.getObject() rather than
byte[].
(Bug#25328)
Functionality added or changed:
Usage Advisor will now issue warnings for result sets with large
numbers of rows. You can configure the trigger value by using
the resultSetSizeThreshold parameter, which
has a default value of 100.
The rewriteBatchedStatements feature can now
be used with server-side prepared statements.
Important change: Due to a number of issues with the use of server-side prepared statements, Connector/J 5.0.5 has disabled their use by default. The disabling of server-side prepared statements does not affect the operation of the connector in any way.
To enable server-side prepared statements you must add the following configuration property to your connector string:
useServerPrepStmts=true
The default value of this property is false
(that is, Connector/J does not use server-side prepared
statements).
Improved speed of datetime parsing for
ResultSets that come from plain or nonserver-side prepared
statements. You can enable old implementation with
useFastDateParsing=false as a configuration
parameter.
Usage Advisor now detects empty results sets and does not report on columns not referenced in those empty sets.
Fixed logging of XA commands sent to server, it is now
configurable via logXaCommands property
(defaults to false).
Added configuration property
localSocketAddress, which is the host name or
IP address given to explicitly configure the interface that the
driver will bind the client side of the TCP/IP connection to
when connecting.
We've added a new configuration option
treatUtilDateAsTimestamp, which is
false by default, as (1) We already had
specific behavior to treat java.util.Date as a
java.sql.Timestamp because it is useful to many folks, and (2)
that behavior will very likely be required for drivers
JDBC-post-4.0.
Bugs fixed:
Connection property socketFactory wasn't
exposed via correctly named mutator/accessor, causing data
source implementations that use JavaBean naming conventions to
set properties to fail to set the property (and in the case of
SJAS, fail silently when trying to set this parameter).
(Bug#26326)
A query execution which timed out did not always throw a
MySQLTimeoutException.
(Bug#25836)
Storing a java.util.Date object in a
BLOB column would not be
serialized correctly during setObject.
(Bug#25787)
Timer instance used for
Statement.setQueryTimeout() created
per-connection, rather than per-VM, causing memory leak.
(Bug#25514)
EscapeProcessor gets confused by multiple
backslashes. We now push the responsibility of syntax errors
back on to the server for most escape sequences.
(Bug#25399)
INOUT parameters in
CallableStatements get doubly-escaped.
(Bug#25379)
When using the rewriteBatchedStatements
connection option with
PreparedState.executeBatch() an internal
memory leak would occur.
(Bug#25073)
Fixed issue where field-level for metadata from
DatabaseMetaData when using
INFORMATION_SCHEMA didn't have references to
current connections, sometimes leading to Null Pointer
Exceptions (NPEs) when introspecting them via
ResultSetMetaData.
(Bug#25073)
StringUtils.indexOfIgnoreCaseRespectQuotes()
isn't case-insensitive on the first character of the target.
This bug also affected
rewriteBatchedStatements functionality when
prepared statements did not use uppercase for the
VALUES clause.
(Bug#25047)
Client-side prepared statement parser gets confused by in-line
comments /*...*/ and therefore cannot rewrite
batch statements or reliably detect the type of statements when
they are used.
(Bug#25025)
Results sets from UPDATE
statements that are part of multi-statement queries would cause
an SQLException error, "Result is from
UPDATE".
(Bug#25009)
Specifying US-ASCII as the character set in a
connection to a MySQL 4.1 or newer server does not map
correctly.
(Bug#24840)
Using DatabaseMetaData.getSQLKeywords() does
not return a all of the of the reserved keywords for the current
MySQL version. Current implementation returns the list of
reserved words for MySQL 5.1, and does not distinguish between
versions.
(Bug#24794)
Calling Statement.cancel() could result in a
Null Pointer Exception (NPE).
(Bug#24721)
Using setFetchSize() breaks prepared
SHOW and other commands.
(Bug#24360)
Calendars and timezones are now lazily instantiated when required. (Bug#24351)
Using DATETIME columns would
result in time shifts when useServerPrepStmts
was true. The reason was due to different behavior when using
client-side compared to server-side prepared statements and the
useJDBCCompliantTimezoneShift option. This is
now fixed if moving from server-side prepared statements to
client-side prepared statements by setting
useSSPSCompatibleTimezoneShift to
true, as the driver can't tell if this is a
new deployment that never used server-side prepared statements,
or if it is an existing deployment that is switching to
client-side prepared statements from server-side prepared
statements.
(Bug#24344)
Connector/J now returns a better error message when server doesn't return enough information to determine stored procedure/function parameter types. (Bug#24065)
A connection error would occur when connecting to a MySQL server
with certain character sets. Some collations/character sets
reported as "unknown" (specifically cias
variants of existing character sets), and inability to override
the detected server character set.
(Bug#23645)
Inconsistency between getSchemas and
INFORMATION_SCHEMA.
(Bug#23304)
DatabaseMetaData.getSchemas() doesn't return
a TABLE_CATALOG column.
(Bug#23303)
When using a JDBC connection URL that is malformed, the
NonRegisteringDriver.getPropertyInfo method
will throw a Null Pointer Exception (NPE).
(Bug#22628)
Some exceptions thrown out of
StandardSocketFactory were needlessly
wrapped, obscuring their true cause, especially when using
socket timeouts.
(Bug#21480)
When using a server-side prepared statement the driver would send timestamps to the server using nanoseconds instead of milliseconds. (Bug#21438)
When using server-side prepared statements and timestamp columns, value would be incorrectly populated (with nanoseconds, not microseconds). (Bug#21438)
ParameterMetaData throws
NullPointerException when prepared SQL has a
syntax error. Added
generateSimpleParameterMetadata configuration
property, which when set to true will
generate metadata reflecting
VARCHAR for every parameter (the
default is false, which will cause an
exception to be thrown if no parameter metadata for the
statement is actually available).
(Bug#21267)
Fixed an issue where XADataSources couldn't
be bound into JNDI, as the DataSourceFactory
didn't know how to create instances of them.
Other changes:
Avoid static synchronized code in JVM class libraries for dealing with default timezones.
Performance enhancement of initial character set configuration, driver will only send commands required to configure connection character set session variables if the current values on the server do not match what is required.
Re-worked stored procedure parameter parser to be more robust.
Driver no longer requires BEGIN in stored
procedure definition, but does have requirement that if a stored
function begins with a label directly after the "returns"
clause, that the label is not a quoted identifier.
Throw exceptions encountered during timeout to thread calling
Statement.execute*(), rather than
RuntimeException.
Changed cached result set metadata (when using
cacheResultSetMetadata=true) to be cached
per-connection rather than per-statement as previously
implemented.
Reverted back to internal character conversion routines for single-byte character sets, as the ones internal to the JVM are using much more CPU time than our internal implementation.
When extracting foreign key information from
SHOW CREATE TABLE in
DatabaseMetaData, ignore exceptions relating
to tables being missing (which could happen for cross-reference
or imported-key requests, as the list of tables is generated
first, then iterated).
Fixed some Null Pointer Exceptions (NPEs) when cached metadata
was used with UpdatableResultSets.
Take localSocketAddress property into account
when creating instances of
CommunicationsException when the underyling
exception is a java.net.BindException, so
that a friendlier error message is given with a little internal
diagnostics.
Fixed cases where ServerPreparedStatements
weren't using cached metadata when
cacheResultSetMetadata=true was used.
Use a java.util.TreeMap to map column names
to ordinal indexes for ResultSet.findColumn()
instead of a HashMap. This allows us to have case-insensitive
lookups (required by the JDBC specification) without resorting
to the many transient object instances needed to support this
requirement with a normal HashMap with either
case-adjusted keys, or case-insensitive keys. (In the worst case
scenario for lookups of a 1000 column result set, TreeMaps are
about half as fast wall-clock time as a HashMap, however in
normal applications their use gives many orders of magnitude
reduction in transient object instance creation which pays off
later for CPU usage in garbage collection).
When using cached metadata, skip field-level metadata packets
coming from the server, rather than reading them and discarding
them without creating com.mysql.jdbc.Field
instances.
Bugs fixed:
DBMD.getColumns() does not return expected COLUMN_SIZE for the SET type, now returns length of largest possible set disregarding whitespace or the "," delimitters to be consistent with the ODBC driver. (Bug#22613)
Added new _ci collations to CharsetMapping - utf8_unicode_ci not working. (Bug#22456)
Driver was using milliseconds for Statement.setQueryTimeout() when specification says argument is to be in seconds. (Bug#22359)
Workaround for server crash when calling stored procedures via a server-side prepared statement (driver now detects prepare(stored procedure) and substitutes client-side prepared statement). (Bug#22297)
Driver issues truncation on write exception when it shouldn't (due to sending big decimal incorrectly to server with server-side prepared statement). (Bug#22290)
Newlines causing whitespace to span confuse procedure parser when getting parameter metadata for stored procedures. (Bug#22024)
When using information_schema for metadata, COLUMN_SIZE for getColumns() is not clamped to range of java.lang.Integer as is the case when not using information_schema, thus leading to a truncation exception that isn't present when not using information_schema. (Bug#21544)
Column names don't match metadata in cases where server doesn't return original column names (column functions) thus breaking compatibility with applications that expect 1–1 mappings between findColumn() and rsmd.getColumnName(), usually manifests itself as "Can't find column ('')" exceptions. (Bug#21379)
Driver now sends numeric 1 or 0 for client-prepared statement
setBoolean() calls instead of '1' or '0'.
Fixed configuration property
jdbcCompliantTruncation was not being used
for reads of result set values.
DatabaseMetaData correctly reports true for
supportsCatalog*() methods.
Driver now supports {call sp} (without "()"
if procedure has no arguments).
Functionality added or changed:
Added configuration option
noAccessToProcedureBodies which will cause
the driver to create basic parameter metadata for
CallableStatements when the user does not
have access to procedure bodies via SHOW
CREATE PROCEDURE or selecting from
mysql.proc instead of throwing an exception.
The default value for this option is false
Bugs fixed:
Fixed Statement.cancel() causes
NullPointerException if underlying connection
has been closed due to server failure.
(Bug#20650)
If the connection to the server has been closed due to a server
failure, then the cleanup process will call
Statement.cancel(), triggering a
NullPointerException, even though there is no
active connection.
(Bug#20650)
Bugs fixed:
MysqlXaConnection.recover(int flags) now
allows combinations of
XAResource.TMSTARTRSCAN and
TMENDRSCAN. To simulate the
“scanning” nature of the interface, we return all
prepared XIDs for TMSTARTRSCAN, and no new
XIDs for calls with TMNOFLAGS, or
TMENDRSCAN when not in combination with
TMSTARTRSCAN. This change was made for API
compliance, as well as integration with IBM WebSphere's
transaction manager.
(Bug#20242)
Fixed MysqlValidConnectionChecker for JBoss
doesn't work with MySQLXADataSources.
(Bug#20242)
Added connection/datasource property
pinGlobalTxToPhysicalConnection (defaults to
false). When set to true,
when using XAConnections, the driver ensures
that operations on a given XID are always routed to the same
physical connection. This allows the
XAConnection to support XA START ...
JOIN after
XA END
has been called, and is also a workaround for transaction
managers that don't maintain thread affinity for a global
transaction (most either always maintain thread affinity, or
have it as a configuration option).
(Bug#20242)
Better caching of character set converters (per-connection) to remove a bottleneck for multibyte character sets. (Bug#20242)
Fixed ConnectionProperties (and thus some
subclasses) are not serializable, even though some J2EE
containers expect them to be.
(Bug#19169)
Fixed driver fails on non-ASCII platforms. The driver was
assuming that the platform character set would be a superset of
MySQL's latin1 when doing the handshake for
authentication, and when reading error messages. We now use
Cp1252 for all strings sent to the server during the handshake
phase, and a hard-coded mapping of the
language systtem variable to
the character set that is used for error messages.
(Bug#18086)
Fixed can't use XAConnection for local
transactions when no global transaction is in progress.
(Bug#17401)
Bugs fixed:
Added support for Connector/MXJ integration via url subprotocol
jdbc:mysql:mxj://....
(Bug#14729)
Idle timeouts cause XAConnections to whine
about rolling themselves back.
(Bug#14729)
When fix for Bug#14562 was merged from 3.1.12, added
functionality for CallableStatement's
parameter metadata to return correct information for
.getParameterClassName().
(Bug#14729)
Added service-provider entry to
META-INF/services/java.sql.Driver for
JDBC-4.0 support.
(Bug#14729)
Fuller synchronization of Connection to avoid
deadlocks when using multithreaded frameworks that multithread a
single connection (usually not recommended, but the JDBC spec
allows it anyways), part of fix to Bug#14972).
(Bug#14729)
Moved all SQLException constructor usage to a
factory in SQLError (ground-work for JDBC-4.0
SQLState-based exception classes).
(Bug#14729)
Removed Java5-specific calls to BigDecimal
constructor (when result set value is '',
(int)0 was being used as an argument
indirectly via method return value. This signature doesn't exist
prior to Java5.)
(Bug#14729)
Implementation of Statement.cancel() and
Statement.setQueryTimeout(). Both require
MySQL-5.0.0 or newer server, require a separate connection to
issue the KILL
QUERY statement, and in the case of
setQueryTimeout() creates an additional
thread to handle the timeout functionality.
Note: Failures to cancel the statement for
setQueryTimeout() may manifest themselves as
RuntimeExceptions rather than failing
silently, as there is currently no way to unblock the thread
that is executing the query being cancelled due to timeout
expiration and have it throw the exception instead.
(Bug#14729)
Return "[VAR]BINARY" for
RSMD.getColumnTypeName() when that is
actually the type, and it can be distinguished (MySQL-4.1 and
newer).
(Bug#14729)
Attempt detection of the MySQL type
BINARY (it is an alias, so this
isn't always reliable), and use the
java.sql.Types.BINARY type mapping for it.
Added unit tests for XADatasource, as well as
friendlier exceptions for XA failures compared to the "stock"
XAException (which has no messages).
If the connection useTimezone is set to
true, then also respect time zone conversions
in escape-processed string literals (for example, "{ts
...}" and "{t ...}").
Don't allow .setAutoCommit(true), or
.commit() or .rollback()
on an XA-managed connection as per the JDBC specification.
XADataSource implemented (ported from 3.2
branch which won't be released as a product). Use
com.mysql.jdbc.jdbc2.optional.MysqlXADataSource
as your datasource class name in your application server to
utilize XA transactions in MySQL-5.0.10 and newer.
Moved -bin-g.jar file into separate
debug subdirectory to avoid confusion.
Return original column name for
RSMD.getColumnName() if the column was
aliased, alias name for .getColumnLabel() (if
aliased), and original table name for
.getTableName(). Note this only works for
MySQL-4.1 and newer, as older servers don't make this
information available to clients.
Setting useJDBCCompliantTimezoneShift=true
(it is not the default) causes the driver to use GMT for
all
TIMESTAMP/DATETIME
time zones, and the current VM time zone for any other type that
refers to time zones. This feature can not be used when
useTimezone=true to convert between server
and client time zones.
PreparedStatement.setString() didn't work
correctly when sql_mode on
server contained
NO_BACKSLASH_ESCAPES and no
characters that needed escaping were present in the string.
Add one level of indirection of internal representation of
CallableStatement parameter metadata to avoid
class not found issues on JDK-1.3 for
ParameterMetadata interface (which doesn't
exist prior to JDBC-3.0).
Important change: Due to a number of issues with the use of server-side prepared statements, Connector/J 5.0.5 has disabled their use by default. The disabling of server-side prepared statements does not affect the operation of the connector in any way.
To enable server-side prepared statements you must add the following configuration property to your connector string:
useServerPrepStmts=true
The default value of this property is false
(that is, Connector/J does not use server-side prepared
statements).
Bugs fixed:
Specifying US-ASCII as the character set in a
connection to a MySQL 4.1 or newer server does not map
correctly.
(Bug#24840)
Bugs fixed:
Check and store value for continueBatchOnError property in constructor of Statements, rather than when executing batches, so that Connections closed out from underneath statements don't cause NullPointerExceptions when it is required to check this property. (Bug#22290)
Fixed Bug#18258 - DatabaseMetaData.getTables(), columns() with bad catalog parameter threw exception rather than return empty result set (as required by spec). (Bug#22290)
Driver now sends numeric 1 or 0 for client-prepared statement setBoolean() calls instead of '1' or '0'. (Bug#22290)
Fixed bug where driver would not advance to next host if roundRobinLoadBalance=true and the last host in the list is down. (Bug#22290)
Driver issues truncation on write exception when it shouldn't (due to sending big decimal incorrectly to server with server-side prepared statement). (Bug#22290)
Fixed bug when calling stored functions, where parameters weren't numbered correctly (first parameter is now the return value, subsequent parameters if specified start at index "2"). (Bug#22290)
Removed logger autodetection altogether, must now specify logger explicitly if you want to use a logger other than one that logs to STDERR. (Bug#21207)
DDriver throws NPE when tracing prepared statements that have been closed (in asSQL()). (Bug#21207)
ResultSet.getSomeInteger() doesn't work for BIT(>1). (Bug#21062)
Escape of quotes in client-side prepared statements parsing not respected. Patch covers more than bug report, including NO_BACKSLASH_ESCAPES being set, and stacked quote characters forms of escaping (that is, '' or ""). (Bug#20888)
Fixed can't pool server-side prepared statements, exception raised when re-using them. (Bug#20687)
Fixed Updatable result set that contains a BIT column fails when server-side prepared statements are used. (Bug#20485)
Fixed updatable result set throws ClassCastException when there is row data and moveToInsertRow() is called. (Bug#20479)
Fixed ResultSet.getShort() for UNSIGNED TINYINT returns incorrect values when using server-side prepared statements. (Bug#20306)
ReplicationDriver does not always round-robin load balance depending on URL used for slaves list. (Bug#19993)
Fixed calling toString() on ResultSetMetaData for driver-generated (that is, from DatabaseMetaData method calls, or from getGeneratedKeys()) result sets would raise a NullPointerException. (Bug#19993)
Connection fails to localhost when using timeout and IPv6 is configured. (Bug#19726)
ResultSet.getFloatFromString() can't retrieve values near Float.MIN/MAX_VALUE. (Bug#18880)
Fixed memory leak with profileSQL=true. (Bug#16987)
Fixed NullPointerException in MysqlDataSourceFactory due to Reference containing RefAddrs with null content. (Bug#16791)
Bugs fixed:
Fixed PreparedStatement.setObject(int, Object,
int) doesn't respect scale of BigDecimals.
(Bug#19615)
Fixed ResultSet.wasNull() returns incorrect
value when extracting native string from server-side prepared
statement generated result set.
(Bug#19282)
Fixed invalid classname returned for
ResultSetMetaData.getColumnClassName() for
BIGINT type.
(Bug#19282)
Fixed case where driver wasn't reading server status correctly when fetching server-side prepared statement rows, which in some cases could cause warning counts to be off, or multiple result sets to not be read off the wire. (Bug#19282)
Fixed data truncation and getWarnings() only
returns last warning in set.
(Bug#18740)
Fixed aliased column names where length of name > 251 are corrupted. (Bug#18554)
Improved performance of retrieving
BigDecimal, Time,
Timestamp and Date values
from server-side prepared statements by creating fewer
short-lived instances of Strings when the
native type is not an exact match for the requested type.
(Bug#18496)
Added performance feature, re-writing of batched executes for
Statement.executeBatch() (for all DML
statements) and
PreparedStatement.executeBatch() (for INSERTs
with VALUE clauses only). Enable by using
"rewriteBatchedStatements=true" in your JDBC URL.
(Bug#18041)
Fixed issue where server-side prepared statements don't cause truncation exceptions to be thrown when truncation happens. (Bug#18041)
Fixed
CallableStatement.registerOutParameter() not
working when some parameters pre-populated. Still waiting for
feedback from JDBC experts group to determine what correct
parameter count from getMetaData() should be,
however.
(Bug#17898)
Fixed calling clearParameters() on a closed
prepared statement causes NPE.
(Bug#17587)
Map "latin1" on MySQL server to CP1252 for MySQL > 4.1.0. (Bug#17587)
Added additional accessor and mutator methods on ConnectionProperties so that DataSource users can use same naming as regular URL properties. (Bug#17587)
Fixed ResultSet.wasNull() not always reset
correctly for booleans when done via conversion for server-side
prepared statements.
(Bug#17450)
Fixed Statement.getGeneratedKeys() throws
NullPointerException when no query has been
processed.
(Bug#17099)
Fixed updatable result set doesn't return
AUTO_INCREMENT values for
insertRow() when multiple column primary keys
are used. (the driver was checking for the existence of
single-column primary keys and an autoincrement value > 0
instead of a straightforward
isAutoIncrement() check).
(Bug#16841)
lib-nodist directory missing from package
breaks out-of-box build.
(Bug#15676)
Fixed issue with ReplicationConnection
incorrectly copying state, doesn't transfer connection context
correctly when transitioning between the same read-only states.
(Bug#15570)
No "dos" character set in MySQL > 4.1.0. (Bug#15544)
INOUT parameter does not store
IN value.
(Bug#15464)
PreparedStatement.setObject() serializes
BigInteger as object, rather than sending as
numeric value (and is thus not complementary to
.getObject() on an UNSIGNED
LONG type).
(Bug#15383)
Fixed issue where driver was unable to initialize character set
mapping tables. Removed reliance on
.properties files to hold this information,
as it turns out to be too problematic to code around class
loader hierarchies that change depending on how an application
is deployed. Moved information back into the
CharsetMapping class.
(Bug#14938)
Exception thrown for new decimal type when using updatable result sets. (Bug#14609)
Driver now aware of fix for BIT
type metadata that went into MySQL-5.0.21 for server not
reporting length consistently .
(Bug#13601)
Added support for Apache Commons logging, use "com.mysql.jdbc.log.CommonsLogger" as the value for the "logger" configuration property. (Bug#13469)
Fixed driver trying to call methods that don't exist on older and newer versions of Log4j. The fix is not trying to auto-detect presence of log4j, too many different incompatible versions out there in the wild to do this reliably.
If you relied on autodetection before, you will need to add "logger=com.mysql.jdbc.log.Log4JLogger" to your JDBC URL to enable Log4J usage, or alternatively use the new "CommonsLogger" class to take care of this. (Bug#13469)
LogFactory now prepends "com.mysql.jdbc.log" to log class name if it can't be found as-specified. This allows you to use "short names" for the built-in log factories, for example "logger=CommonsLogger" instead of "logger=com.mysql.jdbc.log.CommonsLogger". (Bug#13469)
ResultSet.getShort() for UNSIGNED
TINYINT returned wrong values.
(Bug#11874)
Bugs fixed:
Process escape tokens in
Connection.prepareStatement(...). You can
disable this behavior by setting the JDBC URL configuration
property processEscapeCodesForPrepStmts to
false.
(Bug#15141)
Usage advisor complains about unreferenced columns, even though they've been referenced. (Bug#15065)
Driver incorrectly closes streams passed as arguments to
PreparedStatements. Reverts to legacy
behavior by setting the JDBC configuration property
autoClosePStmtStreams to
true (also included in the 3-0-Compat
configuration “bundle”).
(Bug#15024)
Deadlock while closing server-side prepared statements from multiple threads sharing one connection. (Bug#14972)
Unable to initialize character set mapping tables (due to J2EE classloader differences). (Bug#14938)
Escape processor replaces quote character in quoted string with string delimiter. (Bug#14909)
DatabaseMetaData.getColumns() doesn't return
TABLE_NAME correctly.
(Bug#14815)
storesMixedCaseIdentifiers() returns
false
(Bug#14562)
storesLowerCaseIdentifiers() returns
true
(Bug#14562)
storesMixedCaseQuotedIdentifiers() returns
false
(Bug#14562)
storesMixedCaseQuotedIdentifiers() returns
true
(Bug#14562)
If lower_case_table_names=0 (on server):
storesLowerCaseIdentifiers() returns
false
storesLowerCaseQuotedIdentifiers()
returns false
storesMixedCaseIdentifiers() returns
true
storesMixedCaseQuotedIdentifiers()
returns true
storesUpperCaseIdentifiers() returns
false
storesUpperCaseQuotedIdentifiers()
returns true
storesUpperCaseIdentifiers() returns
false
(Bug#14562)
storesUpperCaseQuotedIdentifiers() returns
true
(Bug#14562)
If lower_case_table_names=1 (on server):
storesLowerCaseIdentifiers() returns
true
storesLowerCaseQuotedIdentifiers()
returns true
storesMixedCaseIdentifiers() returns
false
storesMixedCaseQuotedIdentifiers()
returns false
storesUpperCaseIdentifiers() returns
false
storesUpperCaseQuotedIdentifiers()
returns true
storesLowerCaseQuotedIdentifiers() returns
true
(Bug#14562)
Fixed DatabaseMetaData.stores*Identifiers():
If lower_case_table_names=0 (on server):
storesLowerCaseIdentifiers() returns
false
storesLowerCaseQuotedIdentifiers()
returns false
storesMixedCaseIdentifiers() returns
true
storesMixedCaseQuotedIdentifiers()
returns true
storesUpperCaseIdentifiers() returns
false
storesUpperCaseQuotedIdentifiers()
returns true
If lower_case_table_names=1 (on server):
storesLowerCaseIdentifiers() returns
true
storesLowerCaseQuotedIdentifiers()
returns true
storesMixedCaseIdentifiers() returns
false
storesMixedCaseQuotedIdentifiers()
returns false
storesUpperCaseIdentifiers() returns
false
storesUpperCaseQuotedIdentifiers()
returns true
storesMixedCaseIdentifiers() returns
true
(Bug#14562)
storesLowerCaseQuotedIdentifiers() returns
false
(Bug#14562)
Java type conversion may be incorrect for
MEDIUMINT.
(Bug#14562)
storesLowerCaseIdentifiers() returns
false
(Bug#14562)
Added configuration property
useGmtMillisForDatetimes which when set to
true causes
ResultSet.getDate(),
.getTimestamp() to return correct
millis-since GMT when .getTime() is called on
the return value (currently default is false
for legacy behavior).
(Bug#14562)
Extraneous sleep on autoReconnect.
(Bug#13775)
Reconnect during middle of executeBatch()
should not occur if autoReconnect is enabled.
(Bug#13255)
maxQuerySizeToLog is not respected. Added
logging of bound values for execute() phase
of server-side prepared statements when
profileSQL=true as well.
(Bug#13048)
OpenOffice expects
DBMD.supportsIntegrityEnhancementFacility()
to return true if foreign keys are supported
by the datasource, even though this method also covers support
for check constraints, which MySQL doesn't
have. Setting the configuration property
overrideSupportsIntegrityEnhancementFacility
to true causes the driver to return
true for this method.
(Bug#12975)
Added com.mysql.jdbc.testsuite.url.default
system property to set default JDBC url for testsuite (to speed
up bug resolution when I'm working in Eclipse).
(Bug#12975)
logSlowQueries should give better info.
(Bug#12230)
Don't increase timeout for failover/reconnect. (Bug#6577)
Fixed client-side prepared statement bug with embedded
? characters inside quoted identifiers (it
was recognized as a placeholder, when it was not).
Don't allow executeBatch() for
CallableStatements with registered
OUT/INOUT parameters (JDBC
compliance).
Fall back to platform-encoding for
URLDecoder.decode() when parsing driver URL
properties if the platform doesn't have a two-argument version
of this method.
Bugs fixed:
The configuration property sessionVariables
now allows you to specify variables that start with the
“@” sign.
(Bug#13453)
URL configuration parameters don't allow
“&” or
“=” in their values. The JDBC
driver now parses configuration parameters as if they are
encoded using the application/x-www-form-urlencoded format as
specified by java.net.URLDecoder
(http://java.sun.com/j2se/1.5.0/docs/api/java/net/URLDecoder.html).
If the “%” character is present
in a configuration property, it must now be represented as
%25, which is the encoded form of
“%” when using
application/x-www-form-urlencoded encoding.
(Bug#13453)
Workaround for Bug#13374:
ResultSet.getStatement() on closed result set
returns NULL (as per JDBC 4.0 spec, but not
backward-compatible). Set the connection property
retainStatementAfterResultSetClose to
true to be able to retrieve a
ResultSet's statement after the
ResultSet has been closed via
.getStatement() (the default is
false, to be JDBC-compliant and to reduce the
chance that code using JDBC leaks Statement
instances).
(Bug#13277)
ResultSetMetaData from
Statement.getGeneratedKeys() caused a
NullPointerException to be thrown whenever a
method that required a connection reference was called.
(Bug#13277)
Backport of VAR[BINARY|CHAR] [BINARY] types
detection from 5.0 branch.
(Bug#13277)
Fixed NullPointerException when converting
catalog parameter in many
DatabaseMetaDataMethods to
byte[]s (for the result set) when the
parameter is null. (null
isn't technically allowed by the JDBC specification, but we've
historically allowed it).
(Bug#13277)
Backport of Field class,
ResultSetMetaData.getColumnClassName(), and
ResultSet.getObject(int) changes from 5.0
branch to fix behavior surrounding VARCHAR
BINARY/VARBINARY and
related types.
(Bug#13277)
Read response in MysqlIO.sendFileToServer(),
even if the local file can't be opened, otherwise next query
issued will fail, because it is reading the response to the
empty LOAD DATA
INFILE packet sent to the server.
(Bug#13277)
When gatherPerfMetrics is enabled for servers
older than 4.1.0, a NullPointerException is
thrown from the constructor of ResultSet if
the query doesn't use any tables.
(Bug#13043)
java.sql.Types.OTHER returned for
BINARY and
VARBINARY columns when using
DatabaseMetaData.getColumns().
(Bug#12970)
ServerPreparedStatement.getBinding() now
checks if the statement is closed before attempting to reference
the list of parameter bindings, to avoid throwing a
NullPointerException.
(Bug#12970)
Tokenizer for = in URL properties was causing
sessionVariables=.... to be parameterized
incorrectly.
(Bug#12753)
cp1251 incorrectly mapped to
win1251 for servers newer than 4.0.x.
(Bug#12752)
getExportedKeys()
(Bug#12541)
Specifying a catalog works as stated in the API docs. (Bug#12541)
Specifying NULL means that catalog will not
be used to filter the results (thus all databases will be
searched), unless you've set
nullCatalogMeansCurrent=true in your JDBC URL
properties.
(Bug#12541)
getIndexInfo()
(Bug#12541)
getProcedures() (and thus indirectly
getProcedureColumns())
(Bug#12541)
getImportedKeys()
(Bug#12541)
Specifying "" means “current”
catalog, even though this isn't quite JDBC spec compliant, it is
there for legacy users.
(Bug#12541)
getCrossReference()
(Bug#12541)
Added Connection.isMasterConnection() for
clients to be able to determine if a multi-host master/slave
connection is connected to the first host in the list.
(Bug#12541)
getColumns()
(Bug#12541)
Handling of catalog argument in
DatabaseMetaData.getIndexInfo(), which also
means changes to the following methods in
DatabaseMetaData:
getBestRowIdentifier()
getColumns()
getCrossReference()
getExportedKeys()
getImportedKeys()
getIndexInfo()
getPrimaryKeys()
getProcedures() (and thus indirectly
getProcedureColumns())
getTables()
The catalog argument in all of these methods
now behaves in the following way:
Specifying NULL means that catalog will
not be used to filter the results (thus all databases will
be searched), unless you've set
nullCatalogMeansCurrent=true in your JDBC
URL properties.
Specifying "" means
“current” catalog, even though this isn't quite
JDBC spec compliant, it is there for legacy users.
Specifying a catalog works as stated in the API docs.
Made Connection.clientPrepare() available
from “wrapped” connections in the
jdbc2.optional package (connections built
by ConnectionPoolDataSource instances).
getBestRowIdentifier()
(Bug#12541)
Made Connection.clientPrepare() available
from “wrapped” connections in the
jdbc2.optional package (connections built by
ConnectionPoolDataSource instances).
(Bug#12541)
getTables()
(Bug#12541)
getPrimaryKeys()
(Bug#12541)
Connection.prepareCall() is database name
case-sensitive (on Windows systems).
(Bug#12417)
explainSlowQueries hangs with server-side
prepared statements.
(Bug#12229)
Properties shared between master and slave with replication connection. (Bug#12218)
Geometry types not handled with server-side prepared statements. (Bug#12104)
maxPerformance.properties mis-spells
“elideSetAutoCommits”.
(Bug#11976)
ReplicationConnection won't switch to slave,
throws “Catalog can't be null” exception.
(Bug#11879)
Pstmt.setObject(...., Types.BOOLEAN) throws
exception.
(Bug#11798)
Escape tokenizer doesn't respect stacked single quotes for escapes. (Bug#11797)
GEOMETRY type not recognized when using
server-side prepared statements.
(Bug#11797)
Foreign key information that is quoted is parsed incorrectly
when DatabaseMetaData methods use that
information.
(Bug#11781)
The sendBlobChunkSize property is now clamped
to max_allowed_packet with
consideration of stream buffer size and packet headers to avoid
PacketTooBigExceptions when
max_allowed_packet is similar
in size to the default sendBlobChunkSize
which is 1M.
(Bug#11781)
CallableStatement.clearParameters() now
clears resources associated with
INOUT/OUTPUT parameters as
well as INPUT parameters.
(Bug#11781)
Fixed regression caused by fix for Bug#11552 that caused driver to return incorrect values for unsigned integers when those integers where within the range of the positive signed type. (Bug#11663)
Moved source code to Subversion repository. (Bug#11663)
Incorrect generation of testcase scripts for server-side prepared statements. (Bug#11663)
Fixed statements generated for testcases missing
; for “plain” statements.
(Bug#11629)
Spurious ! on console when character encoding
is utf8.
(Bug#11629)
StringUtils.getBytes() doesn't work when
using multi-byte character encodings and a length in
characters is specified.
(Bug#11614)
DBMD.storesLower/Mixed/UpperIdentifiers()
reports incorrect values for servers deployed on Windows.
(Bug#11575)
Reworked Field class,
*Buffer, and MysqlIO to be
aware of field lengths >
Integer.MAX_VALUE.
(Bug#11498)
Escape processor didn't honor strings demarcated with double quotes. (Bug#11498)
Updated DBMD.supportsCorrelatedQueries() to
return true for versions > 4.1,
supportsGroupByUnrelated() to return
true and
getResultSetHoldability() to return
HOLD_CURSORS_OVER_COMMIT.
(Bug#11498)
Lifted restriction of changing streaming parameters with
server-side prepared statements. As long as
all streaming parameters were set before
execution, .clearParameters() does not have
to be called. (due to limitation of client/server protocol,
prepared statements can not reset
individual stream data on the server side).
(Bug#11498)
ResultSet.moveToCurrentRow() fails to work
when preceded by a call to
ResultSet.moveToInsertRow().
(Bug#11190)
VARBINARY data corrupted when
using server-side prepared statements and
.setBytes().
(Bug#11115)
Statement.getWarnings() fails with NPE if
statement has been closed.
(Bug#10630)
Only get char[] from SQL in
PreparedStatement.ParseInfo() when needed.
(Bug#10630)
Bugs fixed:
Initial implemention of ParameterMetadata for
PreparedStatement.getParameterMetadata().
Only works fully for CallableStatements, as
current server-side prepared statements return every parameter
as a VARCHAR type.
Fixed connecting without a database specified raised an
exception in MysqlIO.changeDatabaseTo().
Bugs fixed:
Production package doesn't include JBoss integration classes. (Bug#11411)
Removed nonsensical “costly type conversion” warnings when using usage advisor. (Bug#11411)
Fixed PreparedStatement.setClob() not
accepting null as a parameter.
(Bug#11360)
Connector/J dumping query into SQLException
twice.
(Bug#11360)
autoReconnect ping causes exception on
connection startup.
(Bug#11259)
Connection.setCatalog() is now aware of the
useLocalSessionState configuration property,
which when set to true will prevent the
driver from sending USE ... to the server if
the requested catalog is the same as the current catalog.
(Bug#11115)
3-0-Compat — Compatibility with
Connector/J 3.0.x functionality
(Bug#11115)
maxPerformance — maximum performance
without being reckless
(Bug#11115)
solarisMaxPerformance — maximum
performance for Solaris, avoids syscalls where it can
(Bug#11115)
Added maintainTimeStats configuration
property (defaults to true), which tells the
driver whether or not to keep track of the last query time and
the last successful packet sent to the server's time. If set to
false, removes two syscalls per query.
(Bug#11115)
VARBINARY data corrupted when
using server-side prepared statements and
ResultSet.getBytes().
(Bug#11115)
Added the following configuration bundles, use one or many via
the useConfigs configuration property:
maxPerformance — maximum
performance without being reckless
solarisMaxPerformance — maximum
performance for Solaris, avoids syscalls where it can
3-0-Compat — Compatibility with
Connector/J 3.0.x functionality
Try to handle OutOfMemoryErrors more
gracefully. Although not much can be done, they will in most
cases close the connection they happened on so that further
operations don't run into a connection in some unknown state.
When an OOM has happened, any further operations on the
connection will fail with a “Connection closed”
exception that will also list the OOM exception as the reason
for the implicit connection close event.
(Bug#10850)
Setting cachePrepStmts=true now causes the
Connection to also cache the check the driver
performs to determine if a prepared statement can be server-side
or not, as well as caches server-side prepared statements for
the lifetime of a connection. As before, the
prepStmtCacheSize parameter controls the size
of these caches.
(Bug#10850)
Don't send COM_RESET_STMT for each execution
of a server-side prepared statement if it isn't required.
(Bug#10850)
0-length streams not sent to server when using server-side prepared statements. (Bug#10850)
Driver detects if you're running MySQL-5.0.7 or later, and does
not scan for LIMIT ?[,?] in statements being
prepared, as the server supports those types of queries now.
(Bug#10850)
Reorganized directory layout. Sources now are in
src folder. Don't pollute parent directory
when building, now output goes to ./build,
distribution goes to ./dist.
(Bug#10496)
Added support/bug hunting feature that generates
.sql test scripts to
STDERR when
autoGenerateTestcaseScript is set to
true.
(Bug#10496)
SQLException is thrown when using property
characterSetResults with
cp932 or eucjpms.
(Bug#10496)
The datatype returned for TINYINT(1) columns
when tinyInt1isBit=true (the default) can be
switched between Types.BOOLEAN and
Types.BIT using the new configuration
property transformedBitIsBoolean, which
defaults to false. If set to
false (the default),
DatabaseMetaData.getColumns() and
ResultSetMetaData.getColumnType() will return
Types.BOOLEAN for
TINYINT(1) columns. If
true, Types.BOOLEAN will
be returned instead. Regardless of this configuration property,
if tinyInt1isBit is enabled, columns with the
type TINYINT(1) will be returned as
java.lang.Boolean instances from
ResultSet.getObject(...), and
ResultSetMetaData.getColumnClassName() will
return java.lang.Boolean.
(Bug#10485)
SQLException thrown when retrieving
YEAR(2) with
ResultSet.getString(). The driver will now
always treat YEAR types as
java.sql.Dates and return the correct values
for getString(). Alternatively, the
yearIsDateType connection property can be set
to false and the values will be treated as
SHORTs.
(Bug#10485)
Driver doesn't support {?=CALL(...)} for
calling stored functions. This involved adding support for
function retrieval to
DatabaseMetaData.getProcedures() and
getProcedureColumns() as well.
(Bug#10310)
Unsigned SMALLINT treated as
signed for ResultSet.getInt(), fixed all
cases for UNSIGNED integer values and
server-side prepared statements, as well as
ResultSet.getObject() for UNSIGNED
TINYINT.
(Bug#10156)
Made ServerPreparedStatement.asSql() work
correctly so auto-explain functionality would work with
server-side prepared statements.
(Bug#10155)
Double quotes not recognized when parsing client-side prepared statements. (Bug#10155)
Made JDBC2-compliant wrappers public in order to allow access to vendor extensions. (Bug#10155)
DatabaseMetaData.supportsMultipleOpenResults()
now returns true. The driver has supported
this for some time, DBMD just missed that fact.
(Bug#10155)
Cleaned up logging of profiler events, moved code to dump a
profiler event as a string to
com.mysql.jdbc.log.LogUtils so that third
parties can use it.
(Bug#10155)
Made enableStreamingResults() visible on
com.mysql.jdbc.jdbc2.optional.StatementWrapper.
(Bug#10155)
Actually write manifest file to correct place so it ends up in the binary jar file. (Bug#10144)
Added createDatabaseIfNotExist property
(default is false), which will cause the
driver to ask the server to create the database specified in the
URL if it doesn't exist. You must have the appropriate
privileges for database creation for this to work.
(Bug#10144)
Memory leak in ServerPreparedStatement if
serverPrepare() fails.
(Bug#10144)
com.mysql.jdbc.PreparedStatement.ParseInfo
does unnecessary call to toCharArray().
(Bug#9064)
Driver now correctly uses CP932 if available on the server for Windows-31J, CP932 and MS932 java encoding names, otherwise it resorts to SJIS, which is only a close approximation. Currently only MySQL-5.0.3 and newer (and MySQL-4.1.12 or .13, depending on when the character set gets backported) can reliably support any variant of CP932.
Overhaul of character set configuration, everything now lives in a properties file.
Bugs fixed:
Should accept null for catalog (meaning use
current) in DBMD methods, even though it is not JDBC-compliant
for legacy's sake. Disable by setting connection property
nullCatalogMeansCurrent to
false (which will be the default value in C/J
3.2.x).
(Bug#9917)
Fixed driver not returning true for
-1 when
ResultSet.getBoolean() was called on result
sets returned from server-side prepared statements.
(Bug#9778)
Added a Manifest.MF file with
implementation information to the .jar
file.
(Bug#9778)
More tests in Field.isOpaqueBinary() to
distinguish opaque binary (that is, fields with type
CHAR(n) and CHARACTER SET
BINARY) from output of various scalar and aggregate
functions that return strings.
(Bug#9778)
DBMD.getTables() shouldn't return tables if
views are asked for, even if the database version doesn't
support views.
(Bug#9778)
Should accept null for name patterns in DBMD
(meaning “%”), even though it
isn't JDBC compliant, for legacy's sake. Disable by setting
connection property nullNamePatternMatchesAll
to false (which will be the default value in
C/J 3.2.x).
(Bug#9769)
The performance metrics feature now gathers information about number of tables referenced in a SELECT. (Bug#9704)
The logging system is now automatically configured. If the value
has been set by the user, via the URL property
logger or the system property
com.mysql.jdbc.logger, then use that,
otherwise, autodetect it using the following steps:
Log4j, if it is available,
Then JDK1.4 logging,
Then fallback to our STDERR logging.
(Bug#9704)
Statement.getMoreResults() could throw NPE
when existing result set was .close()d.
(Bug#9704)
Stored procedures with DECIMAL
parameters with storage specifications that contained
“,” in them would fail.
(Bug#9682)
PreparedStatement.setObject(int, Object, int type, int
scale) now uses scale value for
BigDecimal instances.
(Bug#9682)
Added support for the c3p0 connection pool's
(http://c3p0.sf.net/) validation/connection
checker interface which uses the lightweight
COM_PING call to the server if available. To
use it, configure your c3p0 connection pool's
connectionTesterClassName property to use
com.mysql.jdbc.integration.c3p0.MysqlConnectionTester.
(Bug#9320)
PreparedStatement.getMetaData() inserts blank
row in database under certain conditions when not using
server-side prepared statements.
(Bug#9320)
Better detection of LIMIT inside/outside of
quoted strings so that the driver can more correctly determine
whether a prepared statement can be prepared on the server or
not.
(Bug#9320)
Connection.canHandleAsPreparedStatement() now
makes “best effort” to distinguish
LIMIT clauses with placeholders in them from
ones without in order to have fewer false positives when
generating work-arounds for statements the server cannot
currently handle as server-side prepared statements.
(Bug#9320)
Fixed build.xml to not compile
log4j logging if log4j not
available.
(Bug#9320)
Added finalizers to ResultSet and
Statement implementations to be JDBC
spec-compliant, which requires that if not explicitly closed,
these resources should be closed upon garbage collection.
(Bug#9319)
Stored procedures with same name in different databases confuse the driver when it tries to determine parameter counts/types. (Bug#9319)
A continuation of Bug#8868, where functions used in queries
that should return nonstring types when resolved by temporary
tables suddenly become opaque binary strings (work-around for
server limitation). Also fixed fields with type of
CHAR(n) CHARACTER SET BINARY to return
correct/matching classes for
RSMD.getColumnClassName() and
ResultSet.getObject().
(Bug#9236)
Cannot use UTF-8 for characterSetResults
configuration property.
(Bug#9206)
PreparedStatement.addBatch() doesn't work
with server-side prepared statements and streaming
BINARY data.
(Bug#9040)
ServerPreparedStatements now correctly
“stream”
BLOB/CLOB data
to the server. You can configure the threshold chunk size using
the JDBC URL property blobSendChunkSize (the
default is 1MB).
(Bug#8868)
DATE_FORMAT() queries returned as
BLOBs from
getObject().
(Bug#8868)
Server-side session variables can be preset at connection time
by passing them as a comma-delimited list for the connection
property sessionVariables.
(Bug#8868)
BlobFromLocator now uses correct identifier
quoting when generating prepared statements.
(Bug#8868)
Fixed regression in ping() for users using
autoReconnect=true.
(Bug#8868)
Check for empty strings ('') when converting
CHAR/VARCHAR
column data to numbers, throw exception if
emptyStringsConvertToZero configuration
property is set to false (for
backward-compatibility with 3.0, it is now set to
true by default, but will most likely default
to false in 3.2).
(Bug#8803)
DATA_TYPE column from
DBMD.getBestRowIdentifier() causes
ArrayIndexOutOfBoundsException when accessed
(and in fact, didn't return any value).
(Bug#8803)
DBMD.supportsMixedCase*Identifiers() returns
wrong value on servers running on case-sensitive file systems.
(Bug#8800)
DBMD.supportsResultSetConcurrency() not
returning true for forward-only/read-only
result sets (we obviously support this).
(Bug#8792)
Fixed ResultSet.getTime() on a
NULL value for server-side prepared
statements throws NPE.
Made Connection.ping() a public method.
Added support for new precision-math
DECIMAL type in MySQL 5.0.3 and
up.
Fixed DatabaseMetaData.getTables() returning
views when they were not asked for as one of the requested table
types.
Bugs fixed:
PreparedStatements not creating streaming
result sets.
(Bug#8487)
Don't pass NULL to
String.valueOf() in
ResultSet.getNativeConvertToString(), as it
stringifies it (that is, returns null), which
is not correct for the method in question.
(Bug#8487)
Fixed NPE in ResultSet.realClose() when using
usage advisor and result set was already closed.
(Bug#8428)
ResultSet.getString() doesn't maintain format
stored on server, bug fix only enabled when
noDatetimeStringSync property is set to
true (the default is
false).
(Bug#8428)
Added support for BIT type in
MySQL-5.0.3. The driver will treat BIT(1-8)
as the JDBC standard BIT type
(which maps to java.lang.Boolean), as the
server does not currently send enough information to determine
the size of a bitfield when < 9 bits are declared.
BIT(>9) will be treated as
VARBINARY, and will return
byte[] when getObject() is
called.
(Bug#8424)
Added useLocalSessionState configuration
property, when set to true the JDBC driver
trusts that the application is well-behaved and only sets
autocommit and transaction isolation levels using the methods
provided on java.sql.Connection, and
therefore can manipulate these values in many cases without
incurring round-trips to the database server.
(Bug#8424)
Added enableStreamingResults() to
Statement for connection pool implementations
that check Statement.setFetchSize() for
specification-compliant values. Call
Statement.setFetchSize(>=0) to disable the
streaming results for that statement.
(Bug#8424)
ResultSet.getBigDecimal() throws exception
when rounding would need to occur to set scale. The driver now
chooses a rounding mode of “half up” if nonrounding
BigDecimal.setScale() fails.
(Bug#8424)
Fixed synchronization issue with
ServerPreparedStatement.serverPrepare() that
could cause deadlocks/crashes if connection was shared between
threads.
(Bug#8096)
Emulated locators corrupt binary data when using server-side prepared statements. (Bug#8096)
Infinite recursion when “falling back” to master in failover configuration. (Bug#7952)
Disable multi-statements (if enabled) for MySQL-4.1 versions prior to version 4.1.10 if the query cache is enabled, as the server returns wrong results in this configuration. (Bug#7952)
Removed dontUnpackBinaryResults
functionality, the driver now always stores results from
server-side prepared statements as is from the server and
unpacks them on demand.
(Bug#7952)
Fixed duplicated code in
configureClientCharset() that prevented
useOldUTF8Behavior=true from working
properly.
(Bug#7952)
Added holdResultsOpenOverStatementClose
property (default is false), that keeps
result sets open over statement.close() or new execution on same
statement (suggested by Kevin Burton).
(Bug#7715)
Detect new sql_mode variable in
string form (it used to be integer) and adjust quoting method
for strings appropriately.
(Bug#7715)
Timestamps converted incorrectly to strings with server-side prepared statements and updatable result sets. (Bug#7715)
Timestamp key column data needed _binary
stripped for UpdatableResultSet.refreshRow().
(Bug#7686)
Choose correct “direction” to apply time
adjustments when both client and server are in GMT time zone
when using ResultSet.get(..., cal) and
PreparedStatement.set(...., cal).
(Bug#4718)
Remove _binary introducer from parameters
used as in/out parameters in
CallableStatement.
(Bug#4718)
Always return byte[]s for output parameters
registered as *BINARY.
(Bug#4718)
By default, the driver now scans SQL you are preparing via all
variants of Connection.prepareStatement() to
determine if it is a supported type of statement to prepare on
the server side, and if it is not supported by the server, it
instead prepares it as a client-side emulated prepared
statement. You can disable this by passing
emulateUnsupportedPstmts=false in your JDBC
URL.
(Bug#4718)
Added dontTrackOpenResources option (default
is false, to be JDBC compliant), which helps
with memory use for nonwell-behaved apps (that is, applications
that don't close Statement objects when they
should).
(Bug#4718)
Send correct value for “boolean”
true to server for
PreparedStatement.setObject(n, "true",
Types.BIT).
(Bug#4718)
Fixed bug with Connection not caching statements from
prepareStatement() when the statement wasn't
a server-side prepared statement.
(Bug#4718)
Bugs fixed:
DBMD.getProcedures() doesn't respect catalog
parameter.
(Bug#7026)
Fixed hang on SocketInputStream.read() with
Statement.setMaxRows() and multiple result
sets when driver has to truncate result set directly, rather
than tacking a LIMIT
on the end of it.
n
Bugs fixed:
Use 1MB packet for sending file for
LOAD DATA LOCAL
INFILE if that is <
max_allowed_packet on server.
(Bug#6537)
SUM() on
DECIMAL with server-side prepared
statement ignores scale if zero-padding is needed (this ends up
being due to conversion to DOUBLE
by server, which when converted to a string to parse into
BigDecimal, loses all “padding”
zeros).
(Bug#6537)
Use
DatabaseMetaData.getIdentifierQuoteString()
when building DBMD queries.
(Bug#6537)
Use our own implementation of buffered input streams to get
around blocking behavior of
java.io.BufferedInputStream. Disable this
with useReadAheadInput=false.
(Bug#6399)
Make auto-deserialization of
java.lang.Objects stored in
BLOB columns configurable via
autoDeserialize property (defaults to
false).
(Bug#6399)
ResultSetMetaData.getColumnDisplaySize()
returns incorrect values for multi-byte charsets.
(Bug#6399)
Re-work Field.isOpaqueBinary() to detect
CHAR( to support fixed-length binary fields for
n) CHARACTER SET
BINARYResultSet.getObject().
(Bug#6399)
Failing to connect to the server when one of the addresses for
the given host name is IPV6 (which the server does not yet bind
on). The driver now loops through all IP
addresses for a given host, and stops on the first one that
accepts() a
socket.connect().
(Bug#6348)
Removed unwanted new Throwable() in
ResultSet constructor due to bad merge
(caused a new object instance that was never used for every
result set created). Found while profiling for Bug#6359.
(Bug#6225)
ServerSidePreparedStatement allocating
short-lived objects unnecessarily.
(Bug#6225)
Use null-safe-equals for key comparisons in updatable result sets. (Bug#6225)
Fixed too-early creation of StringBuffer in
EscapeProcessor.escapeSQL(), also return
String when escaping not needed (to avoid
unnecessary object allocations). Found while profiling for Bug#6359.
(Bug#6225)
UNSIGNED BIGINT unpacked incorrectly from
server-side prepared statement result sets.
(Bug#5729)
Added experimental configuration property
dontUnpackBinaryResults, which delays
unpacking binary result set values until they're asked for, and
only creates object instances for nonnumerical values (it is set
to false by default). For some usecase/jvm
combinations, this is friendlier on the garbage collector.
(Bug#5706)
Don't throw exceptions for
Connection.releaseSavepoint().
(Bug#5706)
Inefficient detection of pre-existing string instances in
ResultSet.getNativeString().
(Bug#5706)
Use a per-session Calendar instance by
default when decoding dates from
ServerPreparedStatements (set to old, less
performant behavior by setting property
dynamicCalendars=true).
(Bug#5706)
Fixed batched updates with server prepared statements weren't looking if the types had changed for a given batched set of parameters compared to the previous set, causing the server to return the error “Wrong arguments to mysql_stmt_execute()”. (Bug#5235)
Handle case when string representation of timestamp contains
trailing “.” with no numbers
following it.
(Bug#5235)
Server-side prepared statements did not honor
zeroDateTimeBehavior property, and would
cause class-cast exceptions when using
ResultSet.getObject(), as the all-zero string
was always returned.
(Bug#5235)
Fix comparisons made between string constants and dynamic
strings that are converted with either
toUpperCase() or
toLowerCase() to use
Locale.ENGLISH, as some locales
“override” case rules for English. Also use
StringUtils.indexOfIgnoreCase() instead of
.toUpperCase().indexOf(), avoids creating a
very short-lived transient String instance.
Bugs fixed:
Fixed ServerPreparedStatement to read
prepared statement metadata off the wire, even though it is
currently a placeholder instead of using
MysqlIO.clearInputStream() which didn't work
at various times because data wasn't available to read from the
server yet. This fixes sporadic errors users were having with
ServerPreparedStatements throwing
ArrayIndexOutOfBoundExceptions.
(Bug#5032)
Added three ways to deal with all-zero datetimes when reading
them from a ResultSet:
exception (the default), which throws an
SQLException with an SQLState of
S1009; convertToNull,
which returns NULL instead of the date; and
round, which rounds the date to the nearest
closest value which is '0001-01-01'.
(Bug#5032)
The driver is more strict about truncation of numerics on
ResultSet.get*(), and will throw an
SQLException when truncation is detected. You
can disable this by setting
jdbcCompliantTruncation to
false (it is enabled by default, as this
functionality is required for JDBC compliance).
(Bug#5032)
You can now use URLs in
LOAD DATA LOCAL
INFILE statements, and the driver will use Java's
built-in handlers for retreiving the data and sending it to the
server. This feature is not enabled by default, you must set the
allowUrlInLocalInfile connection property to
true.
(Bug#5032)
ResultSet.getObject() doesn't return type
Boolean for pseudo-bit types from prepared
statements on 4.1.x (shortcut for avoiding extra type conversion
when using binary-encoded result sets obscured test in
getObject() for “pseudo” bit
type).
(Bug#5032)
Use com.mysql.jdbc.Message's classloader when
loading resource bundle, should fix sporadic issues when the
caller's classloader can't locate the resource bundle.
(Bug#5032)
ServerPreparedStatements dealing with return
of DECIMAL type don't work.
(Bug#5012)
Track packet sequence numbers if
enablePacketDebug=true, and throw an
exception if packets received out-of-order.
(Bug#4689)
ResultSet.wasNull() does not work for
primatives if a previous null was returned.
(Bug#4689)
Optimized integer number parsing, enable “old”
slower integer parsing using JDK classes via
useFastIntParsing=false property.
(Bug#4642)
Added useOnlyServerErrorMessages property,
which causes message text in exceptions generated by the server
to only contain the text sent by the server (as opposed to the
SQLState's “standard” description, followed by the
server's error message). This property is set to
true by default.
(Bug#4642)
ServerPreparedStatement.execute*() sometimes
threw ArrayIndexOutOfBoundsException when
unpacking field metadata.
(Bug#4642)
Connector/J 3.1.3 beta does not handle integers correctly
(caused by changes to support unsigned reads in
Buffer.readInt() ->
Buffer.readShort()).
(Bug#4510)
Added support in DatabaseMetaData.getTables()
and getTableTypes() for views, which are now
available in MySQL server 5.0.x.
(Bug#4510)
ResultSet.getObject() returns wrong type for
strings when using prepared statements.
(Bug#4482)
Calling MysqlPooledConnection.close() twice
(even though an application error), caused NPE. Fixed.
(Bug#4482)
Bugs fixed:
Support new time zone variables in MySQL-4.1.3 when
useTimezone=true.
(Bug#4311)
Error in retrieval of mediumint column with
prepared statements and binary protocol.
(Bug#4311)
Support for unsigned numerics as return types from prepared
statements. This also causes a change in
ResultSet.getObject() for the bigint
unsigned type, which used to return
BigDecimal instances, it now returns
instances of java.lang.BigInteger.
(Bug#4311)
Externalized more messages (on-going effort). (Bug#4119)
Null bitmask sent for server-side prepared statements was incorrect. (Bug#4119)
Added constants for MySQL error numbers (publicly accessible,
see com.mysql.jdbc.MysqlErrorNumbers), and
the ability to generate the mappings of vendor error codes to
SQLStates that the driver uses (for documentation purposes).
(Bug#4119)
Added packet debuging code (see the
enablePacketDebug property documentation).
(Bug#4119)
Use SQL Standard SQL states by default, unless
useSqlStateCodes property is set to
false.
(Bug#4119)
Mangle output parameter names for
CallableStatements so they will not clash
with user variable names.
Added support for INOUT parameters in
CallableStatements.
Bugs fixed:
Don't enable server-side prepared statements for server version 5.0.0 or 5.0.1, as they aren't compatible with the '4.1.2+' style that the driver uses (the driver expects information to come back that isn't there, so it hangs). (Bug#3804)
getWarnings() returns
SQLWarning instead of
DataTruncation.
(Bug#3804)
getProcedureColumns() doesn't work with
wildcards for procedure name.
(Bug#3540)
getProcedures() does not return any
procedures in result set.
(Bug#3539)
Fixed DatabaseMetaData.getProcedures() when
run on MySQL-5.0.0 (output of SHOW
PROCEDURE STATUS changed between 5.0.0 and 5.0.1.
(Bug#3520)
Added connectionCollation property to cause
driver to issue set collation_connection=...
query on connection init if default collation for given charset
is not appropriate.
(Bug#3520)
DBMD.getSQLStateType() returns incorrect
value.
(Bug#3520)
Correctly map output parameters to position given in
prepareCall() versus. order implied during
registerOutParameter().
(Bug#3146)
Cleaned up detection of server properties. (Bug#3146)
Correctly detect initial character set for servers >= 4.1.0. (Bug#3146)
Support placeholder for parameter metadata for server >= 4.1.2. (Bug#3146)
Added gatherPerformanceMetrics property,
along with properties to control when/where this info gets
logged (see docs for more info).
Fixed case when no parameters could cause a
NullPointerException in
CallableStatement.setOutputParameters().
Enabled callable statement caching via
cacheCallableStmts property.
Fixed sending of split packets for large queries, enabled nio ability to send large packets as well.
Added .toString() functionality to
ServerPreparedStatement, which should help if
you're trying to debug a query that is a prepared statement (it
shows SQL as the server would process).
Added logSlowQueries property, along with
slowQueriesThresholdMillis property to
control when a query should be considered “slow.”
Removed wrapping of exceptions in
MysqlIO.changeUser().
Fixed stored procedure parameter parsing info when size was
specified for a parameter (for example,
char(), varchar()).
ServerPreparedStatements weren't actually
de-allocating server-side resources when
.close() was called.
Fixed case when no output parameters specified for a stored procedure caused a bogus query to be issued to retrieve out parameters, leading to a syntax error from the server.
Bugs fixed:
Use DocBook version of docs for shipped versions of drivers. (Bug#2671)
NULL fields were not being encoded correctly
in all cases in server-side prepared statements.
(Bug#2671)
Fixed rare buffer underflow when writing numbers into buffers for sending prepared statement execution requests. (Bug#2671)
Fixed ConnectionProperties that weren't
properly exposed via accessors, cleaned up
ConnectionProperties code.
(Bug#2623)
Class-cast exception when using scrolling result sets and server-side prepared statements. (Bug#2623)
Merged unbuffered input code from 3.0. (Bug#2623)
Enabled streaming of result sets from server-side prepared statements. (Bug#2606)
Server-side prepared statements were not returning datatype
YEAR correctly.
(Bug#2606)
Fixed charset conversion issue in
getTables().
(Bug#2502)
Implemented multiple result sets returned from a statement or stored procedure. (Bug#2502)
Implemented Connection.prepareCall(), and
DatabaseMetaData.
getProcedures() and
getProcedureColumns().
(Bug#2359)
Merged prepared statement caching, and
.getMetaData() support from 3.0 branch.
(Bug#2359)
Fixed off-by-1900 error in some cases for years in
TimeUtil.fastDate/TimeCreate()
when unpacking results from server-side prepared statements.
(Bug#2359)
Reset long binary parameters in
ServerPreparedStatement when
clearParameters() is called, by sending
COM_RESET_STMT to the server.
(Bug#2359)
NULL values for numeric types in binary
encoded result sets causing
NullPointerExceptions.
(Bug#2359)
Display where/why a connection was implicitly closed (to aid debugging). (Bug#1673)
DatabaseMetaData.getColumns() is not
returning correct column ordinal info for
non-'%' column name patterns.
(Bug#1673)
Fixed NullPointerException in
ServerPreparedStatement.setTimestamp(), as
well as year and month descrepencies in
ServerPreparedStatement.setTimestamp(),
setDate().
(Bug#1673)
Added ability to have multiple database/JVM targets for
compliance and regression/unit tests in
build.xml.
(Bug#1673)
Fixed sending of queries larger than 16M. (Bug#1673)
Merged fix of datatype mapping from MySQL type
FLOAT to
java.sql.Types.REAL from 3.0 branch.
(Bug#1673)
Fixed NPE and year/month bad conversions when accessing some
datetime functionality in
ServerPreparedStatements and their resultant
result sets.
(Bug#1673)
Added named and indexed input/output parameter support to
CallableStatement. MySQL-5.0.x or newer.
(Bug#1673)
CommunicationsException implemented, that
tries to determine why communications was lost with a server,
and displays possible reasons when
.getMessage() is called.
(Bug#1673)
Detect collation of column for
RSMD.isCaseSensitive().
(Bug#1673)
Optimized Buffer.readLenByteArray() to return
shared empty byte array when length is 0.
Fix support for table aliases when checking for all primary keys
in UpdatableResultSet.
Unpack “unknown” data types from server prepared
statements as Strings.
Implemented Statement.getWarnings() for
MySQL-4.1 and newer (using SHOW
WARNINGS).
Ensure that warnings are cleared before executing queries on prepared statements, as-per JDBC spec (now that we support warnings).
Correctly initialize datasource properties from JNDI Refs, including explicitly specified URLs.
Implemented long data (Blobs, Clobs, InputStreams, Readers) for server prepared statements.
Deal with 0-length tokens in EscapeProcessor
(caused by callable statement escape syntax).
DatabaseMetaData now reports
supportsStoredProcedures() for MySQL versions
>= 5.0.0
Support for mysql_change_user().
See the changeUser() method in
com.mysql.jdbc.Connection.
Removed useFastDates connection property.
Support for NIO. Use useNIO=true on platforms
that support NIO.
Check for closed connection on delete/update/insert row
operations in UpdatableResultSet.
Support for transaction savepoints (MySQL >= 4.0.14 or 4.1.1).
Support “old” profileSql
capitalization in ConnectionProperties. This
property is deprecated, you should use
profileSQL if possible.
Fixed character encoding issues when converting bytes to ASCII when MySQL doesn't provide the character set, and the JVM is set to a multi-byte encoding (usually affecting retrieval of numeric values).
Centralized setting of result set type and concurrency.
Fixed bug with UpdatableResultSets not using
client-side prepared statements.
Default result set type changed to
TYPE_FORWARD_ONLY (JDBC compliance).
Fixed IllegalAccessError to
Calendar.getTimeInMillis() in
DateTimeValue (for JDK < 1.4).
Allow contents of PreparedStatement.setBlob()
to be retained between calls to .execute*().
Fixed stack overflow in
Connection.prepareCall() (bad merge).
Refactored how connection properties are set and exposed as
DriverPropertyInfo as well as
Connection and DataSource
properties.
Reduced number of methods called in average query to be more efficient.
Prepared Statements will be re-prepared on
auto-reconnect. Any errors encountered are postponed until first
attempt to re-execute the re-prepared statement.
Bugs fixed:
Added useServerPrepStmts property (default
false). The driver will use server-side
prepared statements when the server version supports them (4.1
and newer) when this property is set to true.
It is currently set to false by default until
all bind/fetch functionality has been implemented. Currently
only DML prepared statements are implemented for 4.1 server-side
prepared statements.
Added requireSSL property.
Track open Statements, close all when
Connection.close() is called (JDBC
compliance).
Bugs fixed:
Workaround for server Bug#9098: Default values of
CURRENT_* for
DATE,
TIME,
DATETIME, and
TIMESTAMP columns can't be
distinguished from string values, so
UpdatableResultSet.moveToInsertRow()
generates bad SQL for inserting default values.
(Bug#8812)
NON_UNIQUE column from
DBMD.getIndexInfo() returned inverted value.
(Bug#8812)
EUCKR charset is sent as SET NAMES
euc_kr which MySQL-4.1 and newer doesn't understand.
(Bug#8629)
Added support for the EUC_JP_Solaris
character encoding, which maps to a MySQL encoding of
eucjpms (backported from 3.1 branch). This
only works on servers that support eucjpms,
namely 5.0.3 or later.
(Bug#8629)
Use hex escapes for
PreparedStatement.setBytes() for double-byte
charsets including “aliases”
Windows-31J, CP934,
MS932.
(Bug#8629)
DatabaseMetaData.supportsSelectForUpdate()
returns correct value based on server version.
(Bug#8629)
Which requires hex escaping of binary data when using multi-byte charsets with prepared statements. (Bug#8064)
Fixed duplicated code in
configureClientCharset() that prevented
useOldUTF8Behavior=true from working
properly.
(Bug#7952)
Backported SQLState codes mapping from Connector/J 3.1, enable
with useSqlStateCodes=true as a connection
property, it defaults to false in this
release, so that we don't break legacy applications (it defaults
to true starting with Connector/J 3.1).
(Bug#7686)
Timestamp key column data needed _binary
stripped for UpdatableResultSet.refreshRow().
(Bug#7686)
MS932, SHIFT_JIS, and
Windows_31J not recognized as aliases for
sjis.
(Bug#7607)
Handle streaming result sets with more than 2 billion rows properly by fixing wraparound of row number counter. (Bug#7601)
PreparedStatement.fixDecimalExponent() adding
extra +, making number unparseable by MySQL
server.
(Bug#7601)
Escape sequence {fn convert(..., type)} now supports ODBC-style
types that are prepended by SQL_.
(Bug#7601)
Statements created from a pooled connection were returning
physical connection instead of logical connection when
getConnection() was called.
(Bug#7316)
Support new protocol type MYSQL_TYPE_VARCHAR.
(Bug#7081)
Added useOldUTF8Behavior' configuration
property, which causes JDBC driver to act like it did with
MySQL-4.0.x and earlier when the character encoding is
utf-8 when connected to MySQL-4.1 or newer.
(Bug#7081)
DatabaseMetaData.getIndexInfo() ignored
unique parameter.
(Bug#7081)
PreparedStatement.fixDecimalExponent() adding
extra +, making number unparseable by MySQL
server.
(Bug#7061)
PreparedStatements don't encode Big5 (and
other multi-byte) character sets correctly in static SQL
strings.
(Bug#7033)
Connections starting up failed-over (due to down master) never retry master. (Bug#6966)
Timestamp/Time conversion
goes in the wrong “direction” when
useTimeZone=true and server time zone differs
from client time zone.
(Bug#5874)
Bugs fixed:
Made TINYINT(1) ->
BIT/Boolean
conversion configurable via tinyInt1isBit
property (default true to be JDBC compliant
out of the box).
(Bug#5664)
Off-by-one bug in
Buffer.readString(.
(Bug#5664)string)
ResultSet.updateByte() when on insert row
throws ArrayOutOfBoundsException.
(Bug#5664)
Fixed regression where useUnbufferedInput was
defaulting to false.
(Bug#5664)
ResultSet.getTimestamp() on a column with
TIME in it fails.
(Bug#5664)
Fixed DatabaseMetaData.getTypes() returning
incorrect (this is, nonnegative) scale for the
NUMERIC type.
(Bug#5664)
Only set character_set_results
during connection establishment if server version >= 4.1.1.
(Bug#5664)
Fixed ResultSetMetaData.isReadOnly() to
detect nonwritable columns when connected to MySQL-4.1 or newer,
based on existence of “original” table and column
names.
Re-issue character set configuration commands when re-using
pooled connections and/or
Connection.changeUser() when connected to
MySQL-4.1 or newer.
Bugs fixed:
ResultSet.getMetaData() should not return
incorrectly initialized metadata if the result set has been
closed, but should instead throw an
SQLException. Also fixed for
getRow() and getWarnings()
and traversal methods by calling
checkClosed() before operating on
instance-level fields that are nullified during
.close().
(Bug#5069)
Use _binary introducer for
PreparedStatement.setBytes() and
set*Stream() when connected to MySQL-4.1.x or
newer to avoid misinterpretation during character conversion.
(Bug#5069)
Parse new time zone variables from 4.1.x servers. (Bug#5069)
ResultSet should release
Field[] instance in
.close().
(Bug#5022)
RSMD.getPrecision() returning 0 for
nonnumeric types (should return max length in chars for
nonbinary types, max length in bytes for binary types). This fix
also fixes mapping of RSMD.getColumnType()
and RSMD.getColumnTypeName() for the
BLOB types based on the length
sent from the server (the server doesn't distinguish between
TINYBLOB,
BLOB,
MEDIUMBLOB or
LONGBLOB at the network protocol
level).
(Bug#4880)
“Production” is now “GA” (General Availability) in naming scheme of distributions. (Bug#4860, Bug#4138)
DBMD.getColumns() returns incorrect JDBC type
for unsigned columns. This affects type mappings for all numeric
types in the RSMD.getColumnType() and
RSMD.getColumnTypeNames() methods as well, to
ensure that “like” types from
DBMD.getColumns() match up with what
RSMD.getColumnType() and
getColumnTypeNames() return.
(Bug#4860, Bug#4138)
Calling .close() twice on a
PooledConnection causes NPE.
(Bug#4808)
Added FLOSS license exemption. (Bug#4742)
Removed redundant calls to checkRowPos() in
ResultSet.
(Bug#4334)
Failover for autoReconnect not using port
numbers for any hosts, and not retrying all hosts.
This required a change to the SocketFactory
connect() method signature, which is now
public Socket connect(String host, int portNumber,
Properties props); therefore, any third-party socket
factories will have to be changed to support this signature.
(Bug#4334)
Logical connections created by
MysqlConnectionPoolDataSource will now issue
a rollback() when they are closed and sent
back to the pool. If your application server/connection pool
already does this for you, you can set the
rollbackOnPooledClose property to
false to avoid the overhead of an extra
rollback().
(Bug#4334)
StringUtils.escapeEasternUnicodeByteStream
was still broken for GBK.
(Bug#4010)
Bugs fixed:
Bugs fixed:
Inconsistent reporting of data type. The server still doesn't return all types for *BLOBs *TEXT correctly, so the driver won't return those correctly. (Bug#3570)
UpdatableResultSet not picking up default
values for moveToInsertRow().
(Bug#3557)
Not specifying database in URL caused
MalformedURL exception.
(Bug#3554)
Auto-convert MySQL encoding names to Java encoding names if used
for characterEncoding property.
(Bug#3554)
Use junit.textui.TestRunner for all unit
tests (to allow them to be run from the command line outside of
Ant or Eclipse).
(Bug#3554)
Added encoding names that are recognized on some JVMs to fix case where they were reverse-mapped to MySQL encoding names incorrectly. (Bug#3554)
Made StringRegressionTest 4.1-unicode aware.
(Bug#3520)
Fixed regression in
PreparedStatement.setString() and eastern
character encodings.
(Bug#3520)
DBMD.getSQLStateType() returns incorrect
value.
(Bug#3520)
Renamed StringUtils.escapeSJISByteStream() to
more appropriate
escapeEasternUnicodeByteStream().
(Bug#3511)
StringUtils.escapeSJISByteStream() not
covering all eastern double-byte charsets correctly.
(Bug#3511)
Return creating statement for ResultSets
created by getGeneratedKeys().
(Bug#2957)
Use SET character_set_results during
initialization to allow any charset to be returned to the driver
for result sets.
(Bug#2670)
Don't truncate BLOB or
CLOB values when using
setBytes() and/or
setBinary/CharacterStream(). .
(Bug#2670)
Dynamically configure character set mappings for field-level
character sets on MySQL-4.1.0 and newer using
SHOW COLLATION when connecting.
(Bug#2670)
Map binary character set to
US-ASCII to support
DATETIME charset recognition for
servers >= 4.1.2.
(Bug#2670)
Use charsetnr returned during connect to
encode queries before issuing SET NAMES on
MySQL >= 4.1.0.
(Bug#2670)
Add helper methods to ResultSetMetaData
(getColumnCharacterEncoding() and
getColumnCharacterSet()) to allow end-users
to see what charset the driver thinks it should be using for the
column.
(Bug#2670)
Only set character_set_results
for MySQL >= 4.1.0.
(Bug#2670)
Allow url parameter for
MysqlDataSource and
MysqlConnectionPool
DataSource so that passing of other
properties is possible from inside appservers.
Don't escape SJIS/GBK/BIG5 when using MySQL-4.1 or newer.
Backport documentation tooling from 3.1 branch.
Added failOverReadOnly property, to allow
end-user to configure state of connection (read-only/writable)
when failed over.
Allow java.util.Date to be sent in as
parameter to PreparedStatement.setObject(),
converting it to a Timestamp to maintain full
precision. .
(Bug#103)
Add unsigned attribute to
DatabaseMetaData.getColumns() output in the
TYPE_NAME column.
Map duplicate key and foreign key errors to SQLState of
23000.
Backported “change user” and “reset server
state” functionality from 3.1 branch, to allow clients of
MysqlConnectionPoolDataSource to reset server
state on getConnection() on a pooled
connection.
Bugs fixed:
Return java.lang.Double for
FLOAT type from
ResultSetMetaData.getColumnClassName().
(Bug#2855)
Return [B instead of
java.lang.Object for
BINARY,
VARBINARY and
LONGVARBINARY types from
ResultSetMetaData.getColumnClassName() (JDBC
compliance).
(Bug#2855)
Issue connection events on all instances created from a
ConnectionPoolDataSource.
(Bug#2855)
Return java.lang.Integer for
TINYINT and
SMALLINT types from
ResultSetMetaData.getColumnClassName().
(Bug#2852)
Added useUnbufferedInput parameter, and now
use it by default (due to JVM issue
http://developer.java.sun.com/developer/bugParade/bugs/4401235.html)
(Bug#2578)
Fixed failover always going to last host in list. (Bug#2578)
Detect on/off or
1, 2, 3
form of lower_case_table_names
value on server.
(Bug#2578)
AutoReconnect time was growing faster than
exponentially.
(Bug#2447)
Trigger a SET NAMES utf8 when encoding is
forced to utf8 or
utf-8 via the
characterEncoding property. Previously, only
the Java-style encoding name of utf-8 would
trigger this.
Bugs fixed:
Enable caching of the parsing stage of prepared statements via
the cachePrepStmts,
prepStmtCacheSize, and
prepStmtCacheSqlLimit properties (disabled by
default).
(Bug#2006)
Fixed security exception when used in Applets (applets can't
read the system property file.encoding which
is needed for LOAD
DATA LOCAL INFILE).
(Bug#2006)
Speed up parsing of PreparedStatements, try
to use one-pass whenever possible.
(Bug#2006)
Fixed exception Unknown character set
'danish' on connect with JDK-1.4.0
(Bug#2006)
Fixed mappings in SQLError to report deadlocks with SQLStates of
41000.
(Bug#2006)
Removed static synchronization bottleneck from instance factory
method of SingleByteCharsetConverter.
(Bug#2006)
Removed static synchronization bottleneck from
PreparedStatement.setTimestamp().
(Bug#2006)
ResultSet.findColumn() should use first
matching column name when there are duplicate column names in
SELECT query (JDBC-compliance).
(Bug#2006)
maxRows property would affect internal
statements, so check it for all statement creation internal to
the driver, and set to 0 when it is not.
(Bug#2006)
Use constants for SQLStates. (Bug#2006)
Map charset ko18_ru to
ko18r when connected to MySQL-4.1.0 or newer.
(Bug#2006)
Ensure that Buffer.writeString() saves room
for the \0.
(Bug#2006)
ArrayIndexOutOfBounds when parameter number
== number of parameters + 1.
(Bug#1958)
Connection property maxRows not honored.
(Bug#1933)
Statements being created too many times in
DBMD.extractForeignKeyFromCreateTable().
(Bug#1925)
Support escape sequence {fn convert ... }. (Bug#1914)
Implement ResultSet.updateClob().
(Bug#1913)
Autoreconnect code didn't set catalog upon reconnect if it had been changed. (Bug#1913)
ResultSet.getObject() on
TINYINT and
SMALLINT columns should return
Java type Integer.
(Bug#1913)
Added more descriptive error message Server
Configuration Denies Access to DataSource, as well as
retrieval of message from server.
(Bug#1913)
ResultSetMetaData.isCaseSensitive() returned
wrong value for
CHAR/VARCHAR
columns.
(Bug#1913)
Added alwaysClearStream connection property,
which causes the driver to always empty any remaining data on
the input stream before each query.
(Bug#1913)
DatabaseMetaData.getSystemFunction()
returning bad function VResultsSion.
(Bug#1775)
Foreign Keys column sequence is not consistent in
DatabaseMetaData.getImported/Exported/CrossReference().
(Bug#1731)
Fix for ArrayIndexOutOfBounds exception when
using Statement.setMaxRows().
(Bug#1695)
Subsequent call to ResultSet.updateFoo()
causes NPE if result set is not updatable.
(Bug#1630)
Fix for 4.1.1-style authentication with no password. (Bug#1630)
Cross-database updatable result sets are not checked for updatability correctly. (Bug#1592)
DatabaseMetaData.getColumns() should return
Types.LONGVARCHAR for MySQL
LONGTEXT type.
(Bug#1592)
Fixed regression of
Statement.getGeneratedKeys() and
REPLACE statements.
(Bug#1576)
Barge blobs and split packets not being read correctly. (Bug#1576)
Backported fix for aliased tables and
UpdatableResultSets in
checkUpdatability() method from 3.1 branch.
(Bug#1534)
“Friendlier” exception message for
PacketTooLargeException.
(Bug#1534)
Don't count quoted IDs when inside a 'string' in
PreparedStatement parsing.
(Bug#1511)
Bugs fixed:
ResultSet.get/setString mashing char 127.
(Bug#1247)
Added property to “clobber” streaming results, by
setting the clobberStreamingResults property
to true (the default is
false). This will cause a
“streaming” ResultSet to be
automatically closed, and any oustanding data still streaming
from the server to be discarded if another query is executed
before all the data has been read from the server.
(Bug#1247)
Added com.mysql.jdbc.util.BaseBugReport to
help creation of testcases for bug reports.
(Bug#1247)
Backported authentication changes for 4.1.1 and newer from 3.1 branch. (Bug#1247)
Made databaseName,
portNumber, and serverName
optional parameters for
MysqlDataSourceFactory.
(Bug#1246)
Optimized CLOB.setChracterStream().
(Bug#1131)
Fixed CLOB.truncate().
(Bug#1130)
Fixed deadlock issue with
Statement.setMaxRows().
(Bug#1099)
DatabaseMetaData.getColumns() getting
confused about the keyword “set” in character
columns.
(Bug#1099)
Clip +/- INF (to smallest and largest representative values for
the type in MySQL) and NaN (to 0) for
setDouble/setFloat(), and
issue a warning on the statement when the server does not
support +/- INF or NaN.
(Bug#884)
Don't fire connection closed events when closing pooled
connections, or on
PooledConnection.getConnection() with already
open connections.
(Bug#884)
Double-escaping of '\' when charset is SJIS
or GBK and '\' appears in nonescaped input.
(Bug#879)
When emptying input stream of unused rows for
“streaming” result sets, have the current thread
yield() every 100 rows in order to not
monopolize CPU time.
(Bug#879)
Issue exception on
ResultSet.get
on empty result set (wasn't caught in some cases).
(Bug#848)XXX()
Don't hide messages from exceptions thrown in I/O layers. (Bug#848)
Fixed regression in large split-packet handling. (Bug#848)
Better diagnostic error messages in exceptions for “streaming” result sets. (Bug#848)
Don't change timestamp TZ twice if
useTimezone==true.
(Bug#774)
Don't wrap SQLExceptions in
RowDataDynamic.
(Bug#688)
Don't try and reset isolation level on reconnect if MySQL doesn't support them. (Bug#688)
The insertRow in an
UpdatableResultSet is now loaded with the
default column values when moveToInsertRow()
is called.
(Bug#688)
DatabaseMetaData.getColumns() wasn't
returning NULL for default values that are
specified as NULL.
(Bug#688)
Change default statement type/concurrency to
TYPE_FORWARD_ONLY and
CONCUR_READ_ONLY (spec compliance).
(Bug#688)
Fix UpdatableResultSet to return values for
get when on
insert row.
(Bug#675)XXX()
Support InnoDB contraint names when
extracting foreign key information in
DatabaseMetaData (implementing ideas from
Parwinder Sekhon).
(Bug#664, Bug#517)
Backported 4.1 protocol changes from 3.1 branch (server-side SQL states, new field information, larger client capability flags, connect-with-database, and so forth). (Bug#664, Bug#517)
refreshRow didn't work when primary key
values contained values that needed to be escaped (they ended up
being doubly escaped).
(Bug#661)
Fixed ResultSet.previous() behavior to move
current position to before result set when on first row of
result set.
(Bug#496)
Fixed Statement and
PreparedStatement issuing bogus queries when
setMaxRows() had been used and a
LIMIT clause was present in the query.
(Bug#496)
Faster date handling code in ResultSet and
PreparedStatement (no longer uses
Date methods that synchronize on static
calendars).
Fixed test for end of buffer in
Buffer.readString().
Bugs fixed:
Fixed SJIS encoding bug, thanks to Naoto Sato. (Bug#378)
Fix problem detecting server character set in some cases. (Bug#378)
Allow multiple calls to Statement.close().
(Bug#378)
Return correct number of generated keys when using
REPLACE statements.
(Bug#378)
Unicode character 0xFFFF in a string would cause the driver to
throw an ArrayOutOfBoundsException. .
(Bug#378)
Fix row data decoding error when using very large packets. (Bug#378)
Optimized row data decoding. (Bug#378)
Issue exception when operating on an already closed prepared statement. (Bug#378)
Optimized usage of EscapeProcessor.
(Bug#378)
Use JVM charset with file names and LOAD DATA [LOCAL]
INFILE.
Fix infinite loop with Connection.cleanup().
Changed Ant target compile-core to
compile-driver, and made testsuite
compilation a separate target.
Fixed result set not getting set for
Statement.executeUpdate(), which affected
getGeneratedKeys() and
getUpdateCount() in some cases.
Return list of generated keys when using multi-value
INSERTS with
Statement.getGeneratedKeys().
Allow bogus URLs in Driver.getPropertyInfo().
Bugs fixed:
Fixed charset issues with database metadata (charset was not getting set correctly).
You can now toggle profiling on/off using
Connection.setProfileSql(boolean).
4.1 Column Metadata fixes.
Fixed MysqlPooledConnection.close() calling
wrong event type.
Fixed StringIndexOutOfBoundsException in
PreparedStatement.setClob().
IOExceptions during a transaction now cause
the Connection to be closed.
Remove synchronization from Driver.connect()
and Driver.acceptsUrl().
Fixed missing conversion for YEAR
type in
ResultSetMetaData.getColumnTypeName().
Updatable ResultSets can now be created for
aliased tables/columns when connected to MySQL-4.1 or newer.
Fixed LOAD DATA LOCAL
INFILE bug when file >
max_allowed_packet.
Don't pick up indexes that start with pri as
primary keys for DBMD.getPrimaryKeys().
Ensure that packet size from
alignPacketSize() does not exceed
max_allowed_packet (JVM bug)
Don't reset Connection.isReadOnly() when
autoReconnecting.
Fixed escaping of 0x5c ('\') character for
GBK and Big5 charsets.
Fixed ResultSet.getTimestamp() when
underlying field is of type DATE.
Throw SQLExceptions when trying to do
operations on a forcefully closed Connection
(that is, when a communication link failure occurs).
Bugs fixed:
Backported 4.1 charset field info changes from Connector/J 3.1.
Fixed Statement.setMaxRows() to stop sending
LIMIT type queries when not needed
(performance).
Fixed DBMD.getTypeInfo() and
DBMD.getColumns() returning different value
for precision in TEXT and
BLOB types.
Fixed SQLExceptions getting swallowed on
initial connect.
Fixed ResultSetMetaData to return
"" when catalog not known. Fixes
NullPointerExceptions with Sun's
CachedRowSet.
Allow ignoring of warning for “non transactional
tables” during rollback (compliance/usability) by setting
ignoreNonTxTables property to
true.
Clean up Statement query/method mismatch
tests (that is, INSERT not
allowed with .executeQuery()).
Fixed ResultSetMetaData.isWritable() to
return correct value.
More checks added in ResultSet traversal
method to catch when in closed state.
Implemented Blob.setBytes(). You still need
to pass the resultant Blob back into an
updatable ResultSet or
PreparedStatement to persist the changes,
because MySQL does not support “locators”.
Add “window” of different NULL
sorting behavior to
DBMD.nullsAreSortedAtStart (4.0.2 to 4.0.10,
true; otherwise, no).
Bugs fixed:
Fixed ResultSet.isBeforeFirst() for empty
result sets.
Added missing LONGTEXT type to
DBMD.getColumns().
Implemented an empty TypeMap for
Connection.getTypeMap() so that some
third-party apps work with MySQL (IBM WebSphere 5.0 Connection
pool).
Added update options for foreign key metadata.
Fixed Buffer.fastSkipLenString() causing
ArrayIndexOutOfBounds exceptions with some
queries when unpacking fields.
Quote table names in
DatabaseMetaData.getColumns(),
getPrimaryKeys(),
getIndexInfo(),
getBestRowIdentifier().
Retrieve TX_ISOLATION from database for
Connection.getTransactionIsolation() when the
MySQL version supports it, instead of an instance variable.
Greatly reduce memory required for
setBinaryStream() in
PreparedStatements.
Bugs fixed:
Streamlined character conversion and byte[]
handling in PreparedStatements for
setByte().
Fixed PreparedStatement.executeBatch()
parameter overwriting.
Added quoted identifiers to database names for
Connection.setCatalog.
Added support for 4.0.8-style large packets.
Reduce memory footprint of PreparedStatements
by sharing outbound packet with MysqlIO.
Added strictUpdates property to allow control
of amount of checking for “correctness” of
updatable result sets. Set this to false if
you want faster updatable result sets and you know that you
create them from SELECT
statements on tables with primary keys and that you have
selected all primary keys in your query.
Added support for quoted identifiers in
PreparedStatement parser.
Bugs fixed:
Allow user to alter behavior of Statement/
PreparedStatement.executeBatch() via
continueBatchOnError property (defaults to
true).
More robust escape tokenizer: Recognize --
comments, and allow nested escape sequences (see
testsuite.EscapeProcessingTest).
Fixed Buffer.isLastDataPacket() for 4.1 and
newer servers.
NamedPipeSocketFactory now works (only
intended for Windows), see README for
instructions.
Changed charsToByte in
SingleByteCharConverter to be nonstatic.
Use nonaliased table/column names and database names to fully
qualify tables and columns in
UpdatableResultSet (requires MySQL-4.1 or
newer).
LOAD DATA LOCAL INFILE ... now works, if your
server is configured to allow it. Can be turned off with the
allowLoadLocalInfile property (see the
README).
Implemented Connection.nativeSQL().
Fixed ResultSetMetaData.getColumnTypeName()
returning BLOB for
TEXT and
TEXT for
BLOB types.
Fixed charset handling in Fields.java.
Because of above, implemented
ResultSetMetaData.isAutoIncrement() to use
Field.isAutoIncrement().
Substitute '?' for unknown character
conversions in single-byte character sets instead of
'\0'.
Added CLIENT_LONG_FLAG to be able to get more
column flags (isAutoIncrement() being the
most important).
Honor lower_case_table_names
when enabled in the server when doing table name comparisons in
DatabaseMetaData methods.
DBMD.getImported/ExportedKeys() now handles
multiple foreign keys per table.
More robust implementation of updatable result sets. Checks that all primary keys of the table have been selected.
Some MySQL-4.1 protocol support (extended field info from selects).
Check for connection closed in more
Connection methods
(createStatement,
prepareStatement,
setTransactionIsolation,
setAutoCommit).
Fixed ResultSetMetaData.getPrecision()
returning incorrect values for some floating-point types.
Changed SingleByteCharConverter to use lazy
initialization of each converter.
Bugs fixed:
Implemented Clob.setString().
Added com.mysql.jdbc.MiniAdmin class, which
allows you to send shutdown command to MySQL
server. This is intended to be used when
“embedding” Java and MySQL server together in an
end-user application.
Added SSL support. See README for
information on how to use it.
All DBMD result set columns describing
schemas now return NULL to be more compliant
with the behavior of other JDBC drivers for other database
systems (MySQL does not support schemas).
Use SHOW CREATE TABLE when
possible for determining foreign key information for
DatabaseMetaData. Also allows cascade options
for DELETE information to be
returned.
Implemented Clob.setCharacterStream().
Failover and autoReconnect work only when the
connection is in an autoCommit(false) state,
in order to stay transaction-safe.
Fixed DBMD.supportsResultSetConcurrency() so
that it returns true for
ResultSet.TYPE_SCROLL_INSENSITIVE and
ResultSet.CONCUR_READ_ONLY or
ResultSet.CONCUR_UPDATABLE.
Implemented Clob.setAsciiStream().
Removed duplicate code from
UpdatableResultSet (it can be inherited from
ResultSet, the extra code for each method to
handle updatability I thought might someday be necessary has not
been needed).
Fixed UnsupportedEncodingException thrown
when “forcing” a character encoding via properties.
Fixed incorrect conversion in
ResultSet.getLong().
Implemented ResultSet.updateBlob().
Removed some not-needed temporary object creation by smarter use
of Strings in
EscapeProcessor,
Connection and
DatabaseMetaData classes.
Escape 0x5c character in strings for the SJIS
charset.
PreparedStatement now honors stream lengths
in setBinary/Ascii/Character Stream() unless you set the
connection property
useStreamLengthsInPrepStmts to
false.
Fixed issue with updatable result sets and
PreparedStatements not working.
Fixed start position off-by-1 error in
Clob.getSubString().
Added connectTimeout parameter that allows
users of JDK-1.4 and newer to specify a maximum time to wait to
establish a connection.
Fixed various non-ASCII character encoding issues.
Fixed ResultSet.isLast() for empty result
sets (should return false).
Added driver property useHostsInPrivileges.
Defaults to true. Affects whether or not
@hostname will be used in
DBMD.getColumn/TablePrivileges.
Fixed
ResultSet.setFetchDirection(FETCH_UNKNOWN).
Added queriesBeforeRetryMaster property that
specifies how many queries to issue when failed over before
attempting to reconnect to the master (defaults to 50).
Fixed issue when calling
Statement.setFetchSize() when using arbitrary
values.
Properly restore connection properties when autoReconnecting or
failing-over, including autoCommit state, and
isolation level.
Implemented Clob.truncate().
Bugs fixed:
Charsets now automatically detected. Optimized code for single-byte character set conversion.
Fixed RowDataStatic.getAt() off-by-one bug.
Fixed ResultSet.getRow() off-by-one bug.
Massive code clean-up to follow Java coding conventions (the time had come).
Implemented ResultSet.getCharacterStream().
Added limited Clob functionality
(ResultSet.getClob(),
PreparedStatemtent.setClob(),
PreparedStatement.setObject(Clob).
Connection.isClosed() no longer
“pings” the server.
Connection.close() issues
rollback() when
getAutoCommit() is false.
Added socketTimeout parameter to URL.
Added LOCAL TEMPORARY to table types in
DatabaseMetaData.getTableTypes().
Added paranoid parameter, which sanitizes
error messages by removing “sensitive” information
from them (such as host names, ports, or user names), as well as
clearing “sensitive” data structures when possible.
Bugs fixed:
General source-code cleanup.
The driver now only works with JDK-1.2 or newer.
Fix and sort primary key names in DBMetaData
(SF bugs 582086 and 582086).
ResultSet.getTimestamp() now works for
DATE types (SF bug 559134).
Float types now reported as
java.sql.Types.FLOAT (SF bug 579573).
Support for streaming (row-by-row) result sets (see
README) Thanks to Doron.
Testsuite now uses Junit (which you can get from http://www.junit.org.
JDBC Compliance: Passes all tests besides stored procedure tests.
ResultSet.getDate/Time/Timestamp now
recognizes all forms of invalid values that have been set to all
zeros by MySQL (SF bug 586058).
Added multi-host failover support (see
README).
Repackaging: New driver name is
com.mysql.jdbc.Driver, old name still works,
though (the driver is now provided by MySQL-AB).
Support for large packets (new addition to MySQL-4.0 protocol),
see README for more information.
Better checking for closed connections in
Statement and
PreparedStatement.
Performance improvements in string handling and field metadata creation (lazily instantiated) contributed by Alex Twisleton-Wykeham-Fiennes.
JDBC-3.0 functionality including
Statement/PreparedStatement.getGeneratedKeys()
and ResultSet.getURL().
Overall speed improvements via controlling transient object
creation in MysqlIO class when reading
packets.
!!! LICENSE CHANGE !!! The
driver is now GPL. If you need non-GPL licenses, please contact
me <[email protected]>.
Performance enchancements: Driver is now 50–100% faster in most situations, and creates fewer temporary objects.
Bugs fixed:
ResultSet.getDouble() now uses code built
into JDK to be more precise (but slower).
Fixed typo for relaxAutoCommit parameter.
LogicalHandle.isClosed() calls through to
physical connection.
Added SQL profiling (to STDERR). Set
profileSql=true in your JDBC URL. See
README for more information.
PreparedStatement now releases resources on
.close(). (SF bug 553268)
More code cleanup.
Quoted identifiers not used if server version does not support
them. Also, if server started with
--ansi or
--sql-mode=ANSI_QUOTES,
“"” will be used as an
identifier quote character, otherwise
“'” will be used.
Bugs fixed:
Fixed unicode chars being read incorrectly. (SF bug 541088)
Faster blob escaping for PrepStmt.
Added setURL() to
MySQLXADataSource. (SF bug 546019)
Added set/getPortNumber()
to DataSource(s). (SF bug 548167)
PreparedStatement.toString() fixed. (SF bug
534026)
More code cleanup.
Rudimentary version of
Statement.getGeneratedKeys() from JDBC-3.0
now implemented (you need to be using JDK-1.4 for this to work,
I believe).
DBMetaData.getIndexInfo() - bad PAGES fixed.
(SF BUG 542201)
ResultSetMetaData.getColumnClassName() now
implemented.
Bugs fixed:
Fixed testsuite.Traversal
afterLast() bug, thanks to Igor Lastric.
Added new types to getTypeInfo(), fixed
existing types thanks to Al Davis and Kid Kalanon.
Fixed time zone off-by-1-hour bug in
PreparedStatement (538286, 528785).
Added identifier quoting to all
DatabaseMetaData methods that need them
(should fix 518108).
Added support for BIT types
(51870) to PreparedStatement.
ResultSet.insertRow() should now detect
auto_increment fields in most cases and use that value in the
new row. This detection will not work in multi-valued keys,
however, due to the fact that the MySQL protocol does not return
this information.
Relaxed synchronization in all classes, should fix 520615 and 520393.
DataSources - fixed setUrl
bug (511614, 525565), wrong datasource class name (532816,
528767).
Added support for YEAR type
(533556).
Fixes for ResultSet updatability in
PreparedStatement.
ResultSet: Fixed updatability (values being
set to null if not updated).
Added getTable/ColumnPrivileges() to DBMD
(fixes 484502).
Added getIdleFor() method to
Connection and
MysqlLogicalHandle.
ResultSet.refreshRow() implemented.
Fixed getRow() bug (527165) in
ResultSet.
General code cleanup.
Bugs fixed:
Full synchronization of Statement.java.
Fixed missing DELETE_RULE value in
DBMD.getImported/ExportedKeys() and
getCrossReference().
More changes to fix Unexpected end of input
stream errors when reading
BLOB values. This should be the
last fix.
Bugs fixed:
Fixed null-pointer-exceptions when using
MysqlConnectionPoolDataSource with Websphere
4 (bug 505839).
Fixed spurious Unexpected end of input stream
errors in MysqlIO (bug 507456).
Bugs fixed:
Fixed extra memory allocation in
MysqlIO.readPacket() (bug 488663).
Added detection of network connection being closed when reading packets (thanks to Todd Lizambri).
Fixed casting bug in PreparedStatement (bug
488663).
DataSource implementations moved to
org.gjt.mm.mysql.jdbc2.optional package, and
(initial) implementations of
PooledConnectionDataSource and
XADataSource are in place (thanks to Todd
Wolff for the implementation and testing of
PooledConnectionDataSource with IBM WebSphere
4).
Fixed quoting error with escape processor (bug 486265).
Removed concatenation support from driver (the
|| operator), as older versions of VisualAge
seem to be the only thing that use it, and it conflicts with the
logical || operator. You will need to start
mysqld with the
--ansi flag to use the
|| operator as concatenation (bug 491680).
Ant build was corrupting included
jar files, fixed (bug 487669).
Report batch update support through
DatabaseMetaData (bug 495101).
Implementation of
DatabaseMetaData.getExported/ImportedKeys()
and getCrossReference().
Fixed off-by-one-hour error in
PreparedStatement.setTimestamp() (bug
491577).
Full synchronization on methods modifying instance and class-shared references, driver should be entirely thread-safe now (please let me know if you have problems).
Bugs fixed:
XADataSource/ConnectionPoolDataSource
code (experimental)
DatabaseMetaData.getPrimaryKeys() and
getBestRowIdentifier() are now more robust in
identifying primary keys (matches regardless of case or
abbreviation/full spelling of Primary Key in
Key_type column).
Batch updates now supported (thanks to some inspiration from Daniel Rall).
PreparedStatement.setAnyNumericType() now
handles positive exponents correctly (adds +
so MySQL can understand it).
Bugs fixed:
Character sets read from database if
useUnicode=true and
characterEncoding is not set. (thanks to
Dmitry Vereshchagin)
Initial transaction isolation level read from database (if available). (thanks to Dmitry Vereshchagin)
Fixed PreparedStatement generating SQL that
would end up with syntax errors for some queries.
PreparedStatement.setCharacterStream() now
implemented
Captialize type names when
captializeTypeNames=true is passed in URL or
properties (for WebObjects. (thanks to Anjo Krank)
ResultSet.getBlob() now returns
null if column value was
null.
Fixed ResultSetMetaData.getPrecision()
returning one less than actual on newer versions of MySQL.
Fixed dangling socket problem when in high availability
(autoReconnect=true) mode, and finalizer for
Connection will close any dangling sockets on
GC.
Fixed time zone issue in
PreparedStatement.setTimestamp(). (thanks to
Erik Olofsson)
PreparedStatement.setDouble() now uses full-precision doubles (reverting a fix made earlier to truncate them).
Fixed
DatabaseMetaData.supportsTransactions(), and
supportsTransactionIsolationLevel() and
getTypeInfo()
SQL_DATETIME_SUB and
SQL_DATA_TYPE fields not being readable.
Updatable result sets now correctly handle
NULL values in fields.
PreparedStatement.setBoolean() will use 1/0 for values if your MySQL version is 3.21.23 or higher.
Fixed ResultSet.isAfterLast() always
returning false.
Bugs fixed:
Fixed PreparedStatement parameter checking.
Fixed case-sensitive column names in
ResultSet.java.
Bugs fixed:
ResultSet.insertRow() works now, even if not
all columns are set (they will be set to
NULL).
Added Byte to
PreparedStatement.setObject().
Fixed data parsing of TIMESTAMP
values with 2-digit years.
Added ISOLATION level support to
Connection.setIsolationLevel()
DataBaseMetaData.getCrossReference() no
longer ArrayIndexOOB.
ResultSet.getBoolean() now recognizes
-1 as true.
ResultSet has +/-Inf/inf support.
getObject() on ResultSet
correctly does
TINYINT->Byte
and
SMALLINT->Short.
Fixed ArrayIndexOutOfBounds when sending
large BLOB queries. (Max size
packet was not being set)
Fixed NPE on
PreparedStatement.executeUpdate() when all
columns have not been set.
Fixed ResultSet.getBlob()
ArrayIndex out-of-bounds.
Bugs fixed:
Fixed composite key problem with updatable result sets.
Faster ASCII string operations.
Fixed off-by-one error in java.sql.Blob
implementation code.
Fixed incorrect detection of
MAX_ALLOWED_PACKET, so sending large blobs
should work now.
Added detection of -/+INF for doubles.
Added ultraDevHack URL parameter, set to
true to allow (broken) Macromedia UltraDev to
use the driver.
Implemented getBigDecimal() without scale
component for JDBC2.
Bugs fixed:
Columns that are of type TEXT now
return as Strings when you use
getObject().
Cleaned up exception handling when driver connects.
Fixed RSMD.isWritable() returning wrong
value. Thanks to Moritz Maass.
DatabaseMetaData.getPrimaryKeys() now works
correctly with respect to key_seq. Thanks to
Brian Slesinsky.
Fixed many JDBC-2.0 traversal, positioning bugs, especially with respect to empty result sets. Thanks to Ron Smits, Nick Brook, Cessar Garcia and Carlos Martinez.
No escape processing is done on
PreparedStatements anymore per JDBC spec.
Fixed some issues with updatability support in
ResultSet when using multiple primary keys.
Fixes to ResultSet for insertRow() - Thanks to Cesar Garcia
Fix to Driver to recognize JDBC-2.0 by loading a JDBC-2.0 class, instead of relying on JDK version numbers. Thanks to John Baker.
Fixed ResultSet to return correct row numbers
Statement.getUpdateCount() now returns rows matched, instead of rows actually updated, which is more SQL-92 like.
10-29-99
Statement/PreparedStatement.getMoreResults() bug fixed. Thanks to Noel J. Bergman.
Added Short as a type to PreparedStatement.setObject(). Thanks to Jeff Crowder
Driver now automagically configures maximum/preferred packet sizes by querying server.
Autoreconnect code uses fast ping command if server supports it.
Fixed various bugs with respect to packet sizing when reading from the server and when alloc'ing to write to the server.
Now compiles under JDK-1.2. The driver supports both JDK-1.1 and JDK-1.2 at the same time through a core set of classes. The driver will load the appropriate interface classes at runtime by figuring out which JVM version you are using.
Fixes for result sets with all nulls in the first row. (Pointed out by Tim Endres)
Fixes to column numbers in SQLExceptions in ResultSet (Thanks to Blas Rodriguez Somoza)
The database no longer needs to specified to connect. (Thanks to Christian Motschke)
Better Documentation (in progress), in doc/mm.doc/book1.html
DBMD now allows null for a column name pattern (not in spec), which it changes to '%'.
DBMD now has correct types/lengths for getXXX().
ResultSet.getDate(), getTime(), and getTimestamp() fixes. (contributed by Alan Wilken)
EscapeProcessor now handles \{ \} and { or } inside quotes correctly. (thanks to Alik for some ideas on how to fix it)
Fixes to properties handling in Connection. (contributed by Juho Tikkala)
ResultSet.getObject() now returns null for NULL columns in the table, rather than bombing out. (thanks to Ben Grosman)
ResultSet.getObject() now returns Strings for types from MySQL that it doesn't know about. (Suggested by Chris Perdue)
Removed DataInput/Output streams, not needed, 1/2 number of method calls per IO operation.
Use default character encoding if one is not specified. This is a work-around for broken JVMs, because according to spec, EVERY JVM must support "ISO8859_1", but they do not.
Fixed Connection to use the platform character encoding instead of "ISO8859_1" if one isn't explicitly set. This fixes problems people were having loading the character- converter classes that didn't always exist (JVM bug). (thanks to Fritz Elfert for pointing out this problem)
Changed MysqlIO to re-use packets where possible to reduce memory usage.
Fixed escape-processor bugs pertaining to {} inside quotes.
Fixed character-set support for non-Javasoft JVMs (thanks to many people for pointing it out)
Fixed ResultSet.getBoolean() to recognize 'y' & 'n' as well as '1' & '0' as boolean flags. (thanks to Tim Pizey)
Fixed ResultSet.getTimestamp() to give better performance. (thanks to Richard Swift)
Fixed getByte() for numeric types. (thanks to Ray Bellis)
Fixed DatabaseMetaData.getTypeInfo() for DATE type. (thanks to Paul Johnston)
Fixed EscapeProcessor for "fn" calls. (thanks to Piyush Shah at locomotive.org)
Fixed EscapeProcessor to not do extraneous work if there are no escape codes. (thanks to Ryan Gustafson)
Fixed Driver to parse URLs of the form "jdbc:mysql://host:port" (thanks to Richard Lobb)
Fixed Timestamps for PreparedStatements
Fixed null pointer exceptions in RSMD and RS
Re-compiled with jikes for valid class files (thanks ms!)
Fixed escape processor to deal with unmatched { and } (thanks to Craig Coles)
Fixed escape processor to create more portable (between DATETIME and TIMESTAMP types) representations so that it will work with BETWEEN clauses. (thanks to Craig Longman)
MysqlIO.quit() now closes the socket connection. Before, after many failed connections some OS's would run out of file descriptors. (thanks to Michael Brinkman)
Fixed NullPointerException in Driver.getPropertyInfo. (thanks to Dave Potts)
Fixes to MysqlDefs to allow all *text fields to be retrieved as Strings. (thanks to Chris at Leverage)
Fixed setDouble in PreparedStatement for large numbers to avoid sending scientific notation to the database. (thanks to J.S. Ferguson)
Fixed getScale() and getPrecision() in RSMD. (contrib'd by James Klicman)
Fixed getObject() when field was DECIMAL or NUMERIC (thanks to Bert Hobbs)
DBMD.getTables() bombed when passed a null table-name pattern. Fixed. (thanks to Richard Lobb)
Added check for "client not authorized" errors during connect. (thanks to Hannes Wallnoefer)
Result set rows are now byte arrays. Blobs and Unicode work bidriectonally now. The useUnicode and encoding options are implemented now.
Fixes to PreparedStatement to send binary set by setXXXStream to be sent untouched to the MySQL server.
Fixes to getDriverPropertyInfo().
Changed all ResultSet fields to Strings, this should allow Unicode to work, but your JVM must be able to convert between the character sets. This should also make reading data from the server be a bit quicker, because there is now no conversion from StringBuffer to String.
Changed PreparedStatement.streamToString() to be more efficient (code from Uwe Schaefer).
URL parsing is more robust (throws SQL exceptions on errors rather than NullPointerExceptions)
PreparedStatement now can convert Strings to Time/Date values via setObject() (code from Robert Currey).
IO no longer hangs in Buffer.readInt(), that bug was introduced in 1.1d when changing to all byte-arrays for result sets. (Pointed out by Samo Login)
Fixes to DatabaseMetaData to allow both IBM VA and J-Builder to work. Let me know how it goes. (thanks to Jac Kersing)
Fix to ResultSet.getBoolean() for NULL strings (thanks to Barry Lagerweij)
Beginning of code cleanup, and formatting. Getting ready to branch this off to a parallel JDBC-2.0 source tree.
Added "final" modifier to critical sections in MysqlIO and Buffer to allow compiler to inline methods for speed.
9-29-98
If object references passed to setXXX() in PreparedStatement are null, setNull() is automatically called for you. (Thanks for the suggestion goes to Erik Ostrom)
setObject() in PreparedStatement will now attempt to write a serialized representation of the object to the database for objects of Types.OTHER and objects of unknown type.
Util now has a static method readObject() which given a ResultSet and a column index will re-instantiate an object serialized in the above manner.
Got rid of "ugly hack" in MysqlIO.nextRow(). Rather than catch an exception, Buffer.isLastDataPacket() was fixed.
Connection.getCatalog() and Connection.setCatalog() should work now.
Statement.setMaxRows() works, as well as setting by property maxRows. Statement.setMaxRows() overrides maxRows set via properties or url parameters.
Automatic re-connection is available. Because it has to "ping" the database before each query, it is turned off by default. To use it, pass in "autoReconnect=true" in the connection URL. You may also change the number of reconnect tries, and the initial timeout value via "maxReconnects=n" (default 3) and "initialTimeout=n" (seconds, default 2) parameters. The timeout is an exponential backoff type of timeout; for example, if you have initial timeout of 2 seconds, and maxReconnects of 3, then the driver will timeout 2 seconds, 4 seconds, then 16 seconds between each re-connection attempt.
Fixed handling of blob data in Buffer.java
Fixed bug with authentication packet being sized too small.
The JDBC Driver is now under the LPGL
8-14-98
Fixed Buffer.readLenString() to correctly read data for BLOBS.
Fixed PreparedStatement.stringToStream to correctly read data for BLOBS.
Fixed PreparedStatement.setDate() to not add a day. (above fixes thanks to Vincent Partington)
Added URL parameter parsing (?user=... and so forth).
Big news! New package name. Tim Endres from ICE Engineering is starting a new source tree for GNU GPL'd Java software. He's graciously given me the org.gjt.mm package directory to use, so now the driver is in the org.gjt.mm.mysql package scheme. I'm "legal" now. Look for more information on Tim's project soon.
Now using dynamically sized packets to reduce memory usage when sending commands to the DB.
Small fixes to getTypeInfo() for parameters, and so forth.
DatabaseMetaData is now fully implemented. Let me know if these drivers work with the various IDEs out there. I've heard that they're working with JBuilder right now.
Added JavaDoc documentation to the package.
Package now available in .zip or .tar.gz.
Implemented getTypeInfo(). Connection.rollback() now throws an SQLException per the JDBC spec.
Added PreparedStatement that supports all JDBC API methods for PreparedStatement including InputStreams. Please check this out and let me know if anything is broken.
Fixed a bug in ResultSet that would break some queries that only returned 1 row.
Fixed bugs in DatabaseMetaData.getTables(), DatabaseMetaData.getColumns() and DatabaseMetaData.getCatalogs().
Added functionality to Statement that allows executeUpdate() to store values for IDs that are automatically generated for AUTO_INCREMENT fields. Basically, after an executeUpdate(), look at the SQLWarnings for warnings like "LAST_INSERTED_ID = 'some number', COMMAND = 'your SQL query'". If you are using AUTO_INCREMENT fields in your tables and are executing a lot of executeUpdate()s on one Statement, be sure to clearWarnings() every so often to save memory.
Split MysqlIO and Buffer to separate classes. Some ClassLoaders gave an IllegalAccess error for some fields in those two classes. Now mm.mysql works in applets and all classloaders. Thanks to Joe Ennis <[email protected]> for pointing out the problem and working on a fix with me.
Fixed DatabaseMetadata problems in getColumns() and bug in switch statement in the Field constructor. Thanks to Costin Manolache <[email protected]> for pointing these out.
Incorporated efficiency changes from Richard Swift
<[email protected]> in
MysqlIO.java and
ResultSet.java:
We're now 15% faster than gwe's driver.
Started working on DatabaseMetaData.
The following methods are implemented:
getTables()
getTableTypes()
getColumns()
getCatalogs()
Functionality added or changed:
Updated internal jar file names to include version information
and be more consistent with Connector/J jar naming. For example,
connector-mxj.jar is now
mysql-connector-mxj-${mxj-version}.jar.
Updated commercial license files.
Added copyright notices to some classes which were missing them.
Added InitializeUser and
QueryUtil classes to support new feature.
Added new tests for initial-user & expanded some existing tests.
ConnectorMXJUrlTestExample and
ConnectorMXJObjectTestExample now
demonstrate the initialization of user/password and creating the
initial database (rather than using "test").
Added new connection property initialize-user
which, if set to true will remove the
default, un-passworded anonymous and root users, and create the
user/password from the connection url.
Removed obsolete field
SimpleMysqldDynamicMBean.lastInvocation.
Clarified code in DefaultsMap.entrySet().
Removed obsolete
PatchedStandardSocketFactory java file.
Added main(String[]) to
com/mysql/management/AllTestsSuite.java.
Errors reading portFile are now reported
using stacktrace(err), previously
System.err was used.
portFile now contains a new-line to be
consistent with pidFile.
Fixed where versionString.trim() was
ignored.
Removed references to File.deleteOnExit,
a warning is printed instead.
Bugs fixed:
Changed tests to shutdown mysqld prior to deleting files.
Fixed port file to always be writen to datadir.
Added os.name-os.arch to resource directory mapping properties file.
Swapped out commercial binaries for v5.0.40.
Delete portFile on shutdown.
Moved platform-map.properties into
db-files.jar.
Clarified the startup max wait numbers.
Updated build.xml in preperation for next
beta build.
Removed use-default-architecture property
replaced.
Added null-check to deal with C/MXJ being loaded by the
bootstrap classloaders with JVMs for which
getClassLoader() returns null.
Added robustness around reading portfile.
Removed PatchedStandardSocketFactory (fixed
in Connetor/J 5.0.6).
Refactored duplication from tests and examples to
QueryUtil.
Removed obsolete
InitializePasswordExample
Bugs fixed:
Moved MysqldFactory to main package.
Reformatting: Added newlines some files which did not end in them.
Swapped out commercial binaries for v5.0.36.
Found and removed dynamic linking in mysql_kill; updated solution.
Changed protected constructor of
SimpleMysqldDynamicMBean from taking a
MysqldResource to taking a
MysqldFactory, in order to lay groundwork for
addressing BUG discovered by Andrew Rubinger. See:
MySQL
Forums (Actual testing with JBoss, and filing a bug, is
still required.)
build.xml: usage now
slightly more verbose; some reformatting.
Now incoporates Reggie Bernett's
SafeTerminateProcess and only calls the
unsafe TerminateProcess as a final last resort.
New windows kill.exe fixes bug where mysqld
was being force terminated. Issue reported by bruno haleblian
and others, see:
MySQL
Forums.
Replaced Boolean.parseBoolean with JDK 1.4
compliant valueOf.
Changed connector-mxj.properties default
mysql version to 5.0.37.
In testing so far mysqld reliably shuts down cleanly much faster.
Added testcase to
com.mysql.management.jmx.AcceptanceTest which
demonstrats that dataDir is a mutable MBean
property.
Updated build.xml in prep for next release.
Changed SimpleMysqldDynamicMBean to create
MysqldResource on demand in order to allow
setting of datadir. (Rubinger bug
groundwork).
Clarified the synchronization of
MysqldResource methods.
SIGHUP is replaced with
MySQLShutdown<PID> event.
Clarified the immutability of baseDir,
dataDir, pidFile,
portFile.
Added 5.1.15 binaries to the repository.
Removed 5.1.14 binaries from the repository.
Added getDataDir() to interface
MysqldResourceI.
Added 5.1.14 binaries to repository.
Replaced windows kill.exe resource with re-written version specific to mysqld.
Added Patched StandardSocketFactory from
Connector/J 5-0 HEAD.
Ensured 5.1.14 compatibility.
Swapped out gpl binaries for v5.0.37.
Removed 5.0.22 binaries from the repository.
Bugs fixed:
Allow multiple calls to start server from URL connection on non-3306 port. (Bug#24004)
Updated build.xml to build to handle with
different gpl and commercial mysld version numbers.
Only populate the options map from the help text if specifically requested or in the MBean case.
Introduced property for Linux & WinXX to default to 32bit versions.
Swapped out gpl binaries for v5.0.27.
Swapped out commercial binaries for v5.0.32.
Moved mysqld binary resourced into separate jar file NOTICE:
CLASSPATH will now need to
connector-mxj-db-files.jar.
Minor test robustness improvements.
Moved default version string out of java class into a text
editable properties file
(connector-mxj.properties) in the resources
directory.
Fixed test to be tollerant of /tmp being a
symlink to /foo/tmp.
Bugs fixed:
Removed unused imports, formatted code, made minor edits to tests.
Removed "TeeOutputStream" - no longer needed.
Swapped out the mysqld binaries for MySQL v5.0.22.
Bugs fixed:
Replaced string parsing with JDBC connection attempt for
determining if a mysqld is "ready for connections"
CLASSPATH will now need to include
Connector/J jar.
"platform" directories replace spaces with underscores
extracted array and list printing to ListToString utility class
Swapped out the mysqld binaries for MySQL v5.0.21
Added trace level logging with Aspect/J.
CLASSPATH will now need to include
lib/aspectjrt.jar
reformatted code
altered to be "basedir" rather than "port" oriented.
help parsing test reflects current help options
insulated users from problems with "." in basedir
swapped out the mysqld binaries for MySQL v5.0.18
Made tests more robust be deleting the /tmp/test-c.mxj directory before running tests.
ServerLauncherSocketFactory.shutdown API change: now takes File parameter (basedir) instead of port.
socket is now "mysql.sock" in datadir
added ability to specify "mysql-version" as an url parameter
Extended timeout for help string parsing, to avoid cases where the help text was getting prematurely flushed, and thus truncated.
swapped out the mysqld binaries for MySQL v5.0.19
MysqldResource now tied to dataDir as well as basedir (API CHANGE)
moved PID file into datadir
ServerLauncherSocketFactory.shutdown now works across JVMs.
extracted splitLines(String) to Str utility class
ServerLauncherSocketFactory.shutdown(port) no longer throws, only reports to System.err
ServerLauncherSocketFactory now treats URL parameters in the
form of &server.foo=null as
serverOptionMap.put("foo", null)
ServerLauncherSocketFactory.shutdown API change: now takes 2 File parameters (basedir, datadir)
Bugs fixed:
Removed HelpOptionsParser's need to reference a MysqldResource.
Reorganized utils into a single "Utils" collaborator.
Minor test tweaks
Altered examples and tests to use new Connector/J 5.0 URL syntax for launching Connector/MXJ ("jdbc:mysql:mxj://")
Swapped out the mysqld binaries for MySQL v5.0.16.
Ditched "ClassUtil" (merged with Str).
Minor refactorings for type casting and exception handling.
Bugs fixed:
Security Enhancement: Accessing mysql-proxy using a client or backend with a MySQL protocol less than MySQL 5.0 would result in mysql-proxy aborting with an assertion. This is because mysql-proxy only supports MySQL Protocol 5.0 or higher. The proxy will now report a fault. (Bug#31419)
Using mysql-proxy with very large return datasets from queries, with or without manipulate of the dataset within the Lua engine could cause a crash. (Bug#39332)
If a submitted packet was smaller than expected by the protocol, MySQL Proxy would terminate. (Bug#36743)
When using mysql-proxy in a master-master replication scenario, a failure in one of the replication masters would fail to be identified by the proxy and connections would not be redirected to the other master. (Bug#35295)
Functionality added or changed:
Fixed assertions on write-errors
Fixed sending fake server-greetings in
connect_server().
Fixed error handling for socket functions on Windows.
Added new features to run-tests.lua.
Functionality added or changed:
When using read/write splitting and the
rw-splitting.lua example script, connecting
a second user to the proxy returns an error message.
(Bug#30867)
Added support in read_query_result() to
overwrite the result-set.
Added --no-daemon and
--pid-file.
Added hooks for read_auth(),
read_handshake() and
read_auth_result().
Added handling of
proxy.connection.backend_ndx in
connect_server() and
read_query() to support read/write
splitting.
Added support for proxy.response.packets.
Added testcases.
Added --no-proxy to disable the proxy.
Added support for listening UNIX sockets.
Added a global lua-scope proxy.global.*.
Added connection pooling.
Bugs fixed:
Fixed assertion on COM_BINLOG_DUMP.
(Bug#29764)
Fixed assertion on result-packets like [ field-len |
fields | EOF | ERR ].
(Bug#29732)
Fixed assertion at login with empty password + empty default db. (Bug#29719)
Fixed assertion at COM_SHUTDOWN.
(Bug#29719)
Fixed crash if proxy.connection is used in
connect_server().
Fixed check for glib2 to require at least
2.6.0.
Fixed assertion when all backends are down and we try to connect.
Fixed connection-stalling if
read_query_result() throws an
assert()ion.
Fixed len-encoding on proxy.resulsets.
Fixed compilation on win32.
Fixed assertion when connecting to the MySQL 6.0.1.
Fixed decoding of len-encoded ints for 3-byte notation.
Fixed inj.resultset.affected_rows on
SELECT queries.
Fixed handling of (SQL) NULL in result-sets.
Fixed mem-leak with proxy.response.* is used.
Functionality added or changed:
Added resultset.affected_rows and
resultset.insert_id.
Changed --proxy.profiling to
--proxy-skip-profiling.
Added missing dependency to
libmysqlclient-dev to the INSTALL file.
Added inj.query_time and
inj.response_time into the lua scripts.
Added support for pre-4.1 passwords in a 4.1 connection.
Added script examples for rewriting and injection.
Added proxy.VERSION.
Added support for UNIX sockets.
Added protection against duplicate resultsets from a script.
Bugs fixed:
Fixed mysql check in configure to die when mysql.h isn't detected.
Fixed handling of duplicate ERR on
COM_CHANGE_USER in MySQL 5.1.18+.
Fixed compile error with MySQL 4.1.x on missing COM_STMT_*.
Fixed crash on fields > 250 bytes when the resultset is inspected.
Fixed warning if connect_server() is not
provided.
Fixed assertion when a error occurs at initial script exec time.
Fixed assertion when read_query_result() is
not provided when PROXY_SEND_QUERY is used.