Important Change: Replication: RESET MASTER and RESET SLAVE now reset the values shown for Last_IO_Error, Last_IO_Errno, Last_SQL_Error, and Last_SQL_Errno in the output of SHOW SLAVE STATUS. (Bug#44270)

See also Bug#34654.

Important Change: Replication: When using STATEMENT or MIXED binary logging format, a statement that changes both non-transactional and transactional tables must be written to the binary log whenever there are changes to non-transactional tables. This means that the statement goes into the binary log even when the changes to the transactional tables fail. In particular, in the event of a failure such statement is annotated with the error number and wrapped inside a pair of BEGIN and ROLLBACK statements.

On the slave, while applying the statement, it is expected that the same failure and the rollback prevent the transactional changes from persisting. However, statements that fail due to concurrency issues such as deadlocks and timeouts are logged in the same way, causing the slave to stop since the statements are applied sequentially by the SQL thread.

To address this issue, we ignore concurrency failures on the slave. Specifically, the following failures are now ignored: ER_LOCK_WAIT_TIMEOUT, ER_LOCK_DEADLOCK, and ER_XA_RBDEADLOCK. (Bug#44581)

Partitioning: Truncating a partitioned MyISAM table did not reset the AUTO_INCREMENT value. (Bug#35111)

Replication: The SHOW SLAVE STATUS connection thread competed with the slave SQL thread for use of the error message buffer. As a result, the connection thread sometimes received incomplete messages. This issue was uncovered with valgrind when message strings were passed without NULL terminators, causing the error Conditional jump or move depends on uninitialised value(s). (Bug#45511)

See also Bug#43076.

Replication: Large transactions and statements could corrupt the binary log if the size of the cache (as set by max_binlog_cache_size) was not large enough to store the changes.

Now, for transactions that do not fit into the cache, the statement is not logged, and the statement generates an error instead.

For non-transactional changes that do not fit into the cache, the statement is also not logged — an incident event is logged after committing or rolling back any pending transaction, and the statement then raises an error.

(Bug#43929)

See also Bug#37148.

Replication: The --database option for mysqlbinlog was ignored when using the row-based logging format. (Bug#42941)

Replication: Shutting down the server while executing FLUSH LOGS, CHANGE MASTER TO, or STOP SLAVE could sometimes cause mysqld to crash. (Bug#38240)

Replication: When reading a binary log that was in use by a master or that had not been properly closed (possibly due to a crash), the following message was printed: Warning: this binlog was not closed properly. Most probably mysqld crashed writing it. This message did not take into account the possibility that the file was merely in use by the master, which caused some users concern who were not aware that this could happen.

To make this clear, the original message has been replaced with Warning: this binlog is either is use or was not closed properly. (Bug#34687)

The server crashed if evaluation of GROUP_CONCAT(... ORDER BY) required allocation of a sort buffer but allocation failed. (Bug#46080)

When creating tables using the IBMDB2I storage engine with the ibmdb2i_create_index_option option set to 1, creating an IBMDB2I table with a primary key should produce an additional index that uses EBCDIC hexadecimal sorting, but this index was not created. (Bug#45983)

With InnoDB tables, MySQL used a less-selective secondary index to avoid a filesort even if a prefix of the primary key was much more selective.

The fix for this problem might cause other queries to run more slowly. (Bug#45828)

The server crashed for attempts to use REPLACE or INSERT ... ON DUPLICATE KEY UPDATE with a view defined using a join. (Bug#45806)

Some collations were causing IBMDB2I to report inaccurate key range estimations to the optimizer for LIKE clauses that select substrings. This can be seen by running EXPLAIN. This problem primarily affects multi-byte and unicode character sets. (Bug#45803)

Invalid memory reads and writes were generated when altering merge and base tables. This could lead to a crash or Valgrind errors:

==28038== Invalid write of size 1
at: memset (mc_replace_strmem.c:479)
by: myrg_attach_children (myrg_open.c:433)
by: ha_myisammrg::attach_children() (ha_myisammrg.cc:546)
by: ha_myisammrg::extra(ha_extra_function) (ha_myisammrg.cc:944)
by: attach_merge_children(TABLE_LIST*) (sql_base.cc:4147)
by: open_tables(THD*, TABLE_LIST**, unsigned*, unsigned) (sql_base.cc:4709)
by: open_and_lock_tables_derived(THD*, TABLE_LIST*, bool) (sql_base.cc:4977)
by: open_n_lock_single_table (mysql_priv.h:1550)
by: mysql_alter_table(sql_table.cc:6428)
by: mysql_execute_command(THD*) (sql_parse.cc:2860)
by: mysql_parse(THD*, char const*, unsigned, char const**) (sql_parse.cc:5933)
by: dispatch_command (sql_parse.cc:1213)

(Bug#45796)

Inserting data into a table using the macce character set with the IBMDB2I storage engine would fail. (Bug#45793)

There was a race condition when changing innodb_commit_concurrency at runtime to the value DEFAULT. (Bug#45749)

See also Bug#42101.

If autocommit was enabled, InnoDB did not roll back DELETE or UPDATE statements if the statement was killed. (Bug#45309)

Use of DECIMAL constants with more than 65 digits in CREATE TABLE ... SELECT statements led to spurious errors or assertion failures. (Bug#45262)

The mysql client could misinterpret some character sequences as commands under some circumstances. (Bug#45236)

Use of CONVERT() with an empty SET value could cause an assertion failure. (Bug#45168)

InnoDB recovery could hang due to redo logging of doublewrite buffer pages. (Bug#45097)

when reading binary data, the concatenation function for geometry data collections did not rigorously check for available data, leading to invalid reads and server crashes. (Bug#44684)

If an error occurred during the creation of a table (for example, the table already existed) having an AUTO_INCREMENT column and a BEFORE trigger that used the INSERT ... SELECT construct, an internal flag was not reset properly. This led to a crash the next time that the table was opened again. (Bug#44653)

For queries with a sufficient number of subqueries in the FROM clause of this form:

SELECT * FROM (SELECT 1) AS t1,
              (SELECT 2) AS t2,
              (SELECT 3) AS t3, ...

The query failed with a Too high level of nesting for select error, as those the query had this form:

SELECT * FROM (SELECT 1 FROM (SELECT 2 FROM (SELECT 3 FROM ...

(Bug#44156)

configure.in contained references to literal instances of nm and libc, rather than to variables parameterized for the proper values on the current platform. (Bug#42721)

configure.in did not properly check for the pthread_setschedprio() function. (Bug#42599)

A workaround for a Sun Studio bug was instituted. (Bug#41710)

Valgrind warnings that occurred for SHOW TABLE STATUS with InnoDB tables were silenced. (Bug#38479)

In the mysql client, if the server connection was lost during repeated status commands, the client would fail to detect this and command output would be inconsistent. (Bug#37274)

When invoked to start multiple server instances, mysqld_multi sometimes would fail to start them all due to not changing location into the base directory for each instance. (Bug#36654)

Renaming a column that appeared in a foreign key definition did not update the foreign key definition with the new column name. (Bug#21704)

Important Change: Replication: Previously, incident log events were represented as comments in the output from mysqlbinlog, making them effectively silent when playing back the binlog.

(An incident log event represents an incident that could cause the contents of the database to change without that event being recorded in the binary log.)

This meant that, if the SQL were applied to a server, it could potentially lead to the master and the slave having different data. To make it possible to handle incident log events without breaking applications that expect the previous behavior, the nonsense statement RELOAD DATABASE is added to the SQL output for that incident log event, which causes an error.

To use this functionality currently requires hand editing of the dump file and handling of each case on an individual basis by a database administrator before applying the output to a server. (Bug#44442)

mysql_upgrade now displays a message indicating the connection parameters it uses when invoking mysqlcheck. (Bug#44638)

The time zone tables for Windows available at http://dev.mysql.com/downloads/timezones.html have been updated. (Bug#39923)

The mysqltest program now has a move_file from_file to_file command for renaming files. This should be used in test cases rather than invoking an external command that might be platform specific. (Bug#39542)

Performance: The InnoDB adaptive hash latch is released (if held) for serveral potentially long-running operations. This improves throughput for other queries if the current query is removing a temporary table, changing a temporary table from memory to disk, using CREATE TABLE ... SELECT, or performing a MyISAM repair on a table used within a transaction. (Bug#32149)

Security Fix: Four potential format string vulnerabilities were fixed (discovered by the Veracode code analysis). (Bug#44166)

Incompatible Change: The server can load plugins under the control of startup options. For example, many storage engines can be built in pluggable form and loaded when the server starts. In the following descriptions, plugin_name stands for a plugin name such as innodb.

Previously, plugin options were handled like other boolean options (see Section 4.2.3.2, “Program Option Modifiers”). That is, any of these options enabled the plugin:

--plugin_name
--plugin_name=1
--enable-plugin_name

And these options disabled the plugin:

--plugin_name=0
--disable-plugin_name
--skip-plugin_name

However, use of a boolean option for plugin loading did not provide control over what to do if the plugin failed to start properly: Should the server exit, or start with the plugin disabled? The actual behavior has been that the server starts with the plugin disabled, which can be problematic. For example, if InnoDB fails to start, existing InnoDB tables become inaccessible, and attempts to create new InnoDB tables result in tables that use the default storage engine unless the NO_ENGINE_SUBSTITUTION SQL mode has been enabled to cause an error to occur instead.

Now, there is a change in the options used to control plugin loading, such that they have a tristate format:

The values OFF, ON, and FORCE are not case sensitive.

Suppose that CSV and InnoDB have been built as pluggable storage engines and that you want the server to load them at startup, subject to these conditions: The server is allowed to run if CSV initialization fails, but must require that InnoDB initialization succeed. To accomplish that, use these lines in an option file:

[mysqld]
csv=ON
innodb=FORCE

This change is incompatible with the previous implementation if you used options of the form --plugin_name=0 or --plugin_name=1, which should be changed to --plugin_name=OFF or --plugin_name=ON, respectively.

--enable-plugin_name is still supported and is the same as --plugin_name=ON. --disable-plugin_name and --skip-plugin_name are still supported and are the same as --plugin_name=OFF. (Bug#19027)

Important Change: Replication: BEGIN, COMMIT, and ROLLBACK statements are no longer affected by --replicate-do-db or --replicate-ignore-db rules. (Bug#43263)

Partitioning: Queries using DISTINCT on multiple columns or GROUP BY on multiple columns did not return correct results with partitioned tables. (Bug#44821)

See also Bug#41136.

Replication: When using row-based logging, the length of an event for which the field metadata exceeded 255 bytes in size was incorrectly calculated. This could lead to corruption of the binary log, or cause the server to hang. (Bug#42749)

See also Bug#44548, Bug#44672, Bug#44752.

Replication: The warning Statement is not safe to log in statement format, issued in situations when it cannot be determined that a statement or other database event can be written reliably to the binary log using the statement-based format, has been changed to Statement may not be safe to log in statement format. (Bug#42415)

Replication: The Query_log_event used by replication to transfer a query to the slave has been refactored. Query_log_event also stores and sends the error code resulting from the execution since it, in some cases, is necessary to execute the statement on the slave as well, which should result in the same error code. The Query_log_event constructor previously worked out for itself the error code using a complex routine, the result of which was often set aside within the constructor itself. This was also involved with at least 2 known bugs relating to invalid errors, and taken as a clear sign that the constructor was not well-designed and needed to be re-written. (Bug#41948)

See also Bug#37145.

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)

The combination of MIN() or MAX() in the select list with WHERE and GROUP BY clauses could lead to incorrect results. (Bug#45386)

Running a SELECT query over an IBMDB2I table using the cp1250 character set would produce an error

ibmdb2i error 2027: Error converting single-byte sort sequence to UCS-2 

(Bug#45197)

Use of ROUND() on a LONGTEXT or LONGBLOB column of a derived table could cause a server crash. (Bug#45152)

DROP USER could fail to drop all privileges for an account if the PAD_CHAR_TO_FULL_LENGTH SQL mode was enabled. (Bug#45100)

GROUP BY on a constant (single-row) InnoDB table joined to other tables caused a server crash. (Bug#44886)

ALTER TABLE on a view crashed the server. (Bug#44860)

When using partitioning with the IBMDB2I storage engine, the engine could report that a valid character set was not supported. (Bug#44856)

Running queries on tables with the IBMDB2I storage engine using the utf8 character would fail when using the 64-bit version of MySQL. (Bug#44811)

Index Merge followed by a filesort could result in a server crash if sort_buffer_size was not large enough for all sort keys. (Bug#44810)

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)

Selecting RAND(N) function where N is a column of a constant table (table with a single row) failed with a SIGFPE signal. (Bug#44768)

The PASSWORD() and OLD_PASSWORD() functions could read memory outside of an internal buffer when used with BLOB arguments. (Bug#44767)

Conversion of a string to a different character set could use the same buffer for input and output, leading to incorrect results or warnings. (Bug#44743, Bug#44766)

mysqld_safe could fail to find the logger program. (Bug#44736)

Code that optimized a read-only XA transaction failed to reset the XID once the transaction was no longer active. (Bug#44672)

A Valgrind warning related to transaction processing was silenced. (Bug#44664)

When creating tables using the IBMDB2I storage engine, the RCDFMT (record format) that would be applied to the corresponding files within the IBM i would be set according to the table name. During whole table operations, the name could get modified to a value inconsistent with the table name. In addition, the record format would be inconsistent compared to the file content. The IBMDB2I storage engine now adds an explicit RCDFMT clause to the CREATE TABLE statement passed down to the DB2 storage engine layer. (Bug#44610)

innochecksum could incorrectly determine the input file name from the arguments. (Bug#44484)

Incorrect time was reported at the end of mysqldump output. (Bug#44424)

Caching of GROUP BY expressions could lead to mismatches between compile-time and runtime calculations and cause a server crash. (Bug#44399)

Lettercase conversion in multibyte cp932 or sjis character sequences could produce incorrect results. (Bug#44352)

InnoDB was missing DB_ROLL_PTR information in Table Monitor COLUMNS output. (Bug#44320)

Assertion failure could occur for duplicate-key errors in INSERT INTO ... SELECT statements. (Bug#44306)

Trying to use an unsupported character set on an IBMDB2I table would produce DB2 error 2501 or 2511. The error has been updated to produce Error 2504 (Character set is unsupported). (Bug#44232)

On 64-bit Windows systems, myisamchk did not handle key_buffer_size values larger than 4GB. (Bug#43940)

For user-defined utf8 collations, attempts to store values too long for a column could cause a server crash. (Bug#43827)

Invalidation of query cache entries due to table modifications could cause threads to hang inside the query cache with state “freeing items”. (Bug#43758)

EXPLAIN EXTENDED could crash for UNION queries in which the last SELECT was not parenthesized and included an ORDER BY clause. (Bug#43612)

Multiple-table updates for InnoDB tables could produce unexpected results. (Bug#43580)

For DELETE statements with ORDER BY var, where var was a global system variable with a NULL value, the server could crash. (Bug#42778)

Builds linked against OpenSSL had a memory leak in association with use of X509 certificates. (Bug#42158)

There was a race condition when changing innodb_commit_concurrency at runtime from zero to nonzero or from nonzero to zero. Now this variable cannot be changed at runtime from zero to nonzero or vice versa. The value can still be changed from one nonzero value to another. (Bug#42101)

See also Bug#45749.

SELECT ... INTO @var could produce values different from SELECT ... without the INTO clause. (Bug#42009)

mysql_zap did not work on Mac OS X. (Bug#41883)

A crash occurred due to a race condition between the merge table and table_cache evictions.

00000001403C452F    mysqld.exe!memcpy()[memcpy.asm:151]
00000001402A275F    mysqld.exe!ha_myisammrg::info()[ha_myisammrg.cc:854]
00000001402A2471    mysqld.exe!ha_myisammrg::attach_children()[ha_myisammrg.cc:488]
00000001402A2788    mysqld.exe!ha_myisammrg::extra()[ha_myisammrg.cc:863]
000000014015FC5D    mysqld.exe!attach_merge_children()[sql_base.cc:4135]
000000014016A4C1    mysqld.exe!open_tables()[sql_base.cc:4697]
000000014016A898    mysqld.exe!open_and_lock_tables_derived()[sql_base.cc:4956]
000000014018BB54    mysqld.exe!mysql_insert()[sql_insert.cc:613]
000000014019EDD3    mysqld.exe!mysql_execute_command()[sql_parse.cc:3066]
00000001401A2F06    mysqld.exe!mysql_parse()[sql_parse.cc:5791]
00000001401A3C1A    mysqld.exe!dispatch_command()[sql_parse.cc:1202]
00000001401A4CD7    mysqld.exe!do_command()[sql_parse.cc:857]
0000000140246327    mysqld.exe!handle_one_connection()[sql_connect.cc:1115]
00000001402B82C5    mysqld.exe!pthread_start()[my_winthread.c:85]
00000001403CAC37    mysqld.exe!_callthreadstart()[thread.c:295]
00000001403CAD05    mysqld.exe!_threadstart()[thread.c:275]
0000000077D6B69A    kernel32.dll!BaseThreadStart()
Trying to get some variables.
Some pointers may be invalid and cause the dump to abort...

(Bug#41212)

Shared-memory connections did not work in Vista if mysqld was started from the command line. (Bug#41190)

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)

In the mysql client, using a default character set of binary caused internal commands such as DELIMITER to become case sensitive. (Bug#37268)

If the MYSQL_HISTFILE environment variable was set to /dev/null, the mysql client overwrote the /dev/null device file as a normal file. (Bug#34224)

mysqld_safe mishandled certain parameters if they contained spaces. (Bug#33685)

mysqladmin kill did not work for thread IDs larger than 32 bits. (Bug#32457)

Several client programs failed to interpret --skip-password as “send no password.” (Bug#28479)

Output from mysql --html did not encode the <, >, or & characters. (Bug#27884)

mysql_convert_table_format did not prevent converting tables to MEMORY or BLACKHOLE tables, which could result in data loss. (Bug#27149)

The .msi installer does not detect an existing root password on the initial configuration attempt. To work around this, install and configure MySQL as normal, but skip any changes to security. (There is a checkbox that allows this on the security screen of the configuration wizard.) Then check your settings:

This issue has been filed as Bug#45200 for correction in a future release.

The Windows configuration wizard allows changes to InnoDB settings during a reconfiguration operation. For an upgrade, this may cause difficulties. To work around this, use one of the following alternatives:

This issue has been filed as Bug#45201 for correction in a future release.

Important Change: Replication: The transactional behavior of STOP SLAVE has changed. Formerly, it took effect immediately, even inside a transaction; now, it waits until the current replication event group (if any) has finished executing, or until the user issues a KILL QUERY or KILL CONNECTION statement.

This was done in order to solve the problem encountered when replication was stopped while a nontransactional slave was replicating a transaction on the master. (It was impossible to roll back a mixed-engines transaction when one of the engines was nontransactional, which meant that the slave could not safely re-apply any transaction that had been interrupted by STOP SLAVE.) (Bug#319, Bug#38205)

See also Bug#43217.

Partitioning: When a value was equal to a PARTITION ... VALUES LESS THAN (value) value other than MAXVALUE, the corresponding partition was not pruned. (Bug#42944)

Replication: Unrelated errors occurring during the execution of RESET SLAVE could cause the slave to crash. (Bug#44179)

Replication: The --slave-skip-errors option had no effect when using row-based logging format. (Bug#39393)

Replication: The following erors were not correctly reported:

Information about these types of failures can now be found in the output of SHOW SLAVE STATUS. (Bug#38197)

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)

See also Bug#27571, Bug#22725.

Replication: Column aliases used inside subqueries were ignored in the binary log. (Bug#35515)

Valgrind warnings for the DECODE(), ENCRYPT(), and FIND_IN_SET() functions were corrected. (Bug#44358, Bug#44365, Bug#44367)

On Windows, entries for build-vs9.bat and build-vs9_x64.bat were missing in win/Makefile.am. (Bug#44353)

Incomplete cleanup of JOIN_TAB::select during the filesort of rows for a GROUP BY clause inside a subquery caused a server crash. (Bug#44290)

Not all lock types had proper descriptive strings, resulting in garbage output from mysqladmin debug. (Bug#44164)

Use of HANDLER statements with INFORMATION_SCHEMA tables caused a server crash. Now HANDLER is prohibited with such tables. (Bug#44151)

MySQL Server allowed the creation of a merge table based on views but crashed when attempts were made to read from that table. The following example demonstrates this:

#Create a test table
CREATE TABLE tmp (id int, c char(2));                                           

#Create two VIEWs upon it
CREATE VIEW v1 AS SELECT * FROM tmp;                                            
CREATE VIEW v2 AS SELECT * FROM tmp;                                            

#Finally create a MERGE table upon the VIEWs
CREATE TABLE merge (id int, c char(2))
ENGINE=MERGE UNION(v1, v2);

#Reading from the merge table lead to a crash
SELECT * FROM merge;

The final line of the code generated the crash. (Bug#44040)

Some schema names longer than 8 characters were not supported by IBMDB2I. The engine has been updated to allow digits and underscore characters to be used in names longer than 8 characters. (Bug#44025)

In some circumstances, when a table is created with the IBMDB2I engine, the CREATE TABLE statement will return successfully but the table will not exist. (Bug#44022)

The ucs2_swedish_ci and utf8_swedish_ci collations did not work with indexes using the IBMDB2I storage engine. Support is now provided for MySQL when running on IBM i 6.1 or higher. (Bug#44020)

Invoking SHOW TABLE STATUS from within a stored procedure could cause a Packets out of order error. (Bug#43962)

myisamchk could display a negative Max keyfile length value. (Bug#43950)

On 64-bit systems, a key_buffer_size value larger than 4GB could couse MyISAM index corruption. (Bug#43932)

mysqld_multi incorrectly passed --no-defaults to mysqld_safe. (Bug#43876)

SHOW VARIABLES did not properly display the value of slave_skip_errors. (Bug#43835)

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)

Incorrect initialization of MyISAM table indexes could cause incorrect query results. (Bug#43737)

libmysqld crashed when it was reinitialized. (Bug#43706, Bug#44091)

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)

ALTER DATABASE ... UPGRADE DATA DIRECTORY NAME failed when the database contained views. (Bug#43385)

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)

Using an XML function such as ExtractValue() more than once in a single query could produce erroneous results. (Bug#43183)

See also Bug#43937.

Full-text prefix searches could hang the connection and cause 100% CPU consumption. (Bug#42907)

InnoDB had excessive contention for a character set mutex. (Bug#42649)

Incorrect elevation of warning messages to error messages for unsafe statements caused a server crash. (Bug#42640)

CHECK TABLE suggested use of REPAIR TABLE for corrupt tables for storage engines not supported by REPAIR TABLE. Now CHECK TABLE suggests that the user dump and reload the table. (Bug#42563)

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)

For some queries, an equality propagation problem could cause a = b and b = a to be handled differently. (Bug#40925)

Killing an INSERT ... SELECT statement for a MyISAM table could cause table corruption if the table had indexes. (Bug#40827)

A multiple-table DELETE IGNORE statement involving a foreign key constraint caused an assertion failure. (Bug#40127)

Multiple-table UPDATE statements did not properly activate triggers. (Bug#39953)

The mysql_setpermission operation for removing database privileges removed global privileges instead. (Bug#39852)

A stored routine contain a C-style comment could not be dumped and reloaded. (Bug#39559)

In an UPDATE or DELETE via a secondary index, InnoDB did not store the cursor position. This made InnoDB crash in semi-consistent read while attempting to unlock a nonmatching record. (Bug#39320)

The functions listed in Section 11.13.4.2.3, “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 11.13.4.2.2, “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)

On WIndows, running the server with myisam_use_mmap enabled caused MyISAM table corruption. (Bug#38848)

CHECK TABLE did not properly check whether MyISAM tables created by servers from MySQL 4.0 or older needed to be upgraded. This could cause problems upgrading to MySQL 5.1 or higher. (Bug#37631)

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)

For InnoDB tables that have their own .ibd tablespace file, a superfluous ibuf cursor restoration fails! message could be written to the error log. This warning has been suppressed. (Bug#27276)

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)

Incomplete cleanup of JOIN_TAB::select during the filesort of rows for a GROUP BY clause inside a subquery caused a server crash. (Bug#44290)

Use of HANDLER statements with INFORMATION_SCHEMA tables caused a server crash. Now HANDLER is prohibited with such tables. (Bug#44151)

On 64-bit systems, a key_buffer_size value larger than 4GB could couse MyISAM index corruption. (Bug#43932)

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)

libmysqld crashed when it was reinitialized. (Bug#43706, Bug#44091)

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)

Using an XML function such as ExtractValue() more than once in a single query could produce erroneous results. (Bug#43183)

See also Bug#43937.

Incorrect elevation of warning messages to error messages for unsafe statements caused a server crash. (Bug#42640)

In an UPDATE or DELETE via a secondary index, InnoDB did not store the cursor position. This made InnoDB crash in semi-consistent read while attempting to unlock a nonmatching record. (Bug#39320)

The functions listed in Section 11.13.4.2.3, “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 11.13.4.2.2, “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)

The optimizer_switch system variable is now available to control optimizations that can be switched on and off. See Section 7.2.18, “Using optimizer_switch to Control the Optimizer”.

Replication: Important Note: Binary logging with --binlog_format=ROW failed when a change to be logged included more than 251 columns. This issue was not known to occur with mixed-format or statement-based logging. (Bug#42977)

See also Bug#42914.

Replication: Assigning an invalid directory for the --slave-load-tmpdir caused the replication slave to crash. (Bug#42861)

Replication: The mysql.procs_priv system table was not replicated. (Bug#42217)

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)

Replication: The MIXED binary logging format did not switch to row-based mode for statements containing the LOAD_FILE() function. (Bug#39701)

Replication: When the server SQL mode included IGNORE_SPACE, statement-based replication of LOAD DATA INFILE ... INTO tbl_name failed because the statement was read incorrectly from the binary log; a trailing space was omitted, causing the statement to fail with a syntax error when run on the slave. (Bug#22504)

See also Bug#43746.

An attempt by a user who did not have the SUPER privilege to kill a system thread could cause a server crash. (Bug#43748)

On Windows, incorrectly specified link dependencies in CMakeLists.txt resulted in link errors for mysql_embedded, mysqltest_embedded, and mysql_client_test_embedded. (Bug#43715)

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)

If the value of the version_comment system variable was too long, the mysql client displayed a truncated startup message. (Bug#43153)

Queries of the following form returned an empty result:

SELECT ... WHERE ... (col=col AND col=col) OR ... (false expression)

(Bug#42957)

The strings/CHARSET_INFO.txt file was not included in source distributions. (Bug#42937)

A dangling pointer in mysys/my_error.c could lead to client crashes. (Bug#42675)

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 load_defaults(), my_search_option_files() and my_print_default_files() functions in the C client library were subject to a race condition in multi-threaded operation. (Bug#40552)

If --basedir was specified, mysqld_safe did not use it when attempting to locate my_print_defaults. (Bug#39326)

When MySQL was configured with the --with-max-indexes=128 option, mysqld crashed. (Bug#36751)

Setting the join_buffer_size variable to its minimum value produced spurious warnings. (Bug#36446)

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)

An attempt to check or repair an ARCHIVE table that had been subjected to a server crash returned a 144 internal error. The data appeared to be irrecoverable. (Bug#32880)

The Time column for SHOW PROCESSLIST output and the value of the TIME column of the INFORMATION_SCHEMA.PROCESSLIST table 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 = value 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)

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)

mysql-test-run.pl now supports an --experimental=file_name option. It enables you to specify a file that contains a list of test cases that should be displayed with the [ exp-fail ] code rather than [ fail ] if they fail. (Bug#42888)

The MD5 algorithm now uses the Xfree implementation. (Bug#42434)

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)

Partitioning: A duplicate key error raised when inserting into a partitioned table used a different error code from that returned by such an error raised when inserting into a table that was not partitioned. (Bug#38719)

See also Bug#28842.

Partitioning: Several error messages relating to partitioned tables were incorrect or missing. (Bug#36001)

Replication: When --binlog_format was set to STATEMENT, a statement unsafe for statement-based logging caused an error or warning to be issued even if sql_log_bin was set to 0. (Bug#41980)

Replication: When using MIXED replication format and temporary tables were created in statement-based mode, but a later operation in the same session caused a switch to row-based mode, the temporary tables were not dropped on the slave at the end of the session. (Bug#40013)

See also Bug#43046.

This regression was introduced by Bug#20499.

Replication: When using the MIXED replication format, UPDATE and DELETE statements that searched for rows where part of the key had nullable BIT columns failed. This occurred because operations that inserted the data were replicated as statements, but UPDATE and DELETE statements affecting the same data were replicated using row-based format.

This issue did not occur when using statement-based replication (only) or row-based replication (only). (Bug#39753)

See also Bug#39648.

Replication: The server SQL mode in effect when a stored procedure was created was not retained in the binary log. This could cause a CREATE PROCEDURE statement that succeeded on the master to fail on the slave.

This issue was first noticed when a stored procedure was created when ANSI_QUOTES was in effect on the master, but could possibly cause failed CREATE PROCEDURE statements and other problems on the slave when using other server SQL modes as well. (Bug#39526)

Replication: If --secure-file-priv was set on the slave, it was unable to execute LOAD DATA INFILE statements sent from the master when using mixed-format or statement-based replication.

As a result of this fix, this security restriction is now ignored on the slave in such cases; instead the slave checks whether the files were created and should be read by the slave in its --slave-load-tmpdir. (Bug#38174)

Replication: Server IDs greater than 2147483647 (2³² – 1) were represented by negative numbers in the binary log. (Bug#37313)

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:

(Bug#22082)

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.

The IBM DB2i storage engine has been added to this release for the IBM i Series platform. For more information, see Section 13.7, “The IBMDB2I Storage Engine”. (Bug#44217)

On 64-bit debug builds, code in safemalloc resulted in errors due to use of a 32-bit value for 64-bit allocations. (Bug#43885)

make distcheck failed to properly handle subdirectories of storage/ndb. (Bug#43614)

Use of USE INDEX hints could cause EXPLAIN EXTENDED to crash. (Bug#43354)

For InnoDB tables, overflow in an AUTO_INCREMENT column could cause a server crash. (Bug#43203)

On 32-bit Windows, mysqld could not use large buffers due to a 2GB user mode address limit. (Bug#43082)

stderr should be unbuffered, but when the server redirected stderr to a file, it became buffered. (Bug#42790)

The DATA_TYPE column of the INFORMATION_SCHEMA.COLUMNS table displayed the UNSIGNED attribute for floating-point data types. (The column should contain only the data type name.) (Bug#42758)

For InnoDB tables, spurious duplicate-key errors could occur when inserting into an AUTO_INCREMENT column. (Bug#42714)

mysqldump included views that were excluded with the --ignore-table option. (Bug#42635)

An earlier bug fix resulted in the problem that the InnoDB plugin could not be used with a server that was compiled with the built-in InnoDB. To handle this two changes were made:

(Bug#42610)

This regression was introduced by Bug#29263.

With the ONLY_FULL_GROUP_BY SQL mode enabled, some legal queries failed. (Bug#42567)

Tables could enter open table cache for a thread without being properly cleaned up, leading to a server crash. (Bug#42419)

For InnoDB tables, inserting into floating-point AUTO_INCREMENT columns failed. (Bug#42400)

The InnoDB btr_search_drop_page_hash_when_freed() function had a race condition. (Bug#42279)

For InnoDB tables, there was a race condition for ALTER TABLE, OPTIMIZE TABLE, CREATE INDEX, and DROP INDEX operations when periodically checking whether table copying can be committed. (Bug#42152)

Parsing of the optional microsecond component of DATETIME values did not fail gracefully when that component width was larger than the allowed six places. (Bug#42146)

In InnoDB recovery after a server crash, table lookup could fail and corrupt the data dictionary cache. (Bug#42075)

mysqldumpslow parsed the --debug and --verbose options incorrectly. (Bug#42027)

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)

The error message for a too-long column comment was Unknown error rather than a more appropriate message. (Bug#41465)

Use of SELECT * allowed users with rights to only some columns of a view to access all columns. (Bug#41354)

If the tables underlying a MERGE table had a primary key but the MERGE table itself did not, inserting a duplicate row into the MERGE table caused a server crash. (Bug#41305)

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)

SELECT statements executed concurrently with INSERT statements for a MyISAM table could cause incorrect results to be returned from the query cache. (Bug#41098)

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)

Deprecation warnings that referred to MySQL 5.2 were changed to refer to MySQL 6.0. (Bug#41077)

For user-defined variables in a query result, incorrect length values were returned in the result metadata. (Bug#41030)

On Windows, starting the server with an invalid value for innodb_flush_method caused a crash. (Bug#40757)

MySQL 5.1 crashed with index merge algorithm and merge tables.

A query in the MyISAM merge table caused a crash if the index merge algorithm was being used. (Bug#40675)

With strict SQL mode enabled, setting a system variable to an out-of-bounds value caused an assertion failure. (Bug#40657)

Table temporary scans were slower than necessary due to use of mmap rather than caching, even with the myisam_use_mmap system variable disabled. (Bug#40634)

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)

On platforms where long and pointer variables have different sizes, MyISAM could copy key statistics incorrectly, resulting in a server crash or incorrect cardinality values. (Bug#40321)

DELETE tried to acquire write (not read) locks for tables accessed within a subquery of the WHERE clause. (Bug#39843)

perror did not produce correct output for error codes 153 to 163. (Bug#39370)

Several functions in libmysqld called exit() when an error occurred rather than returning an error to the caller. (Bug#39289)

The innodb_log_arch_dir system variable is no longer available but was present in some of the sample option files included with MySQL distributions (such as my-huge.cnf). The line was present as a comment but uncommenting it would cause server startup failure so the line has been removed. (Bug#38249)

Setting a savepoint with the same name as an existing savepoint incorrectly deleted any other savepoints that had been set in the meantime. For example, setting savepoints named a, b, c, b resulted in savepoints a, b, rather than the correct savepoints a, c, b. (Bug#38187)

--help output for myisamchk did not list the --HELP option. (Bug#38103)

Comparisons between row constructors, such as (a, b) = (c, d) resulted in unnecessary Illegal mix of collations errors for string columns. (Bug#37601)

If a user created a view that referenced tables for which the user had disjoint privileges, an assertion failure occurred. (Bug#37191)

An argument to the MATCH() function that was an alias for an expression other than a column name caused a server crash. (Bug#36737)

The event, general_log, and slow_log tables in the mysql database store server_id values, but did not use an UNSIGNED column and thus were not able to store the full range of ID values. (Bug#36540)

On Windows, the _PC macro in my_global.h was causing problems for modern compilers. It has been removed because it is no longer used. (Bug#34309)

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)

Detection by configure of several functions such as setsockopt(), bind(), sched_yield(), and gtty() could fail. (Bug#31506)

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)

SHOW PRIVILEGES listed the CREATE ROUTINE privilege as having a context of Functions,Procedures, but it is a database-level privilege. (Bug#30305)

mysqld --help did not work as root. (Bug#30261)

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)

SHOW TABLE STATUS could fail to produce output for tables with non-ASCII characters in their name. (Bug#25830)

Allocation of stack space for error messages could be too small on HP-UX, leading to stack overflow crashes. (Bug#21476)

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)

The libedit library was upgraded to version 2.11. (Bug#42433)

Security Fix: Using an XPath expression employing a scalar expression as a FilterExpr with ExtractValue() or UpdateXML() caused the server to crash. Such expressions now cause an error instead. (Bug#42495)

Incompatible Change: The fix for Bug#33699 introduced a change to the UPDATE statement such that assigning NULL to a NOT NULL column caused an error even when strict SQL mode was not enabled. The original behavior before was that such assignments caused an error only in strict SQL mode, and otherwise set the column to the implicit default value for the column data type and generated a warning. (For information about implicit default values, see Section 10.1.4, “Data Type Default Values”.)

The change caused compatibility problems for applications that relied on the original behavior. It also caused replication problems between servers that had the original behavior and those that did not, for applications that assigned NULL to NOT NULL columns in UPDATE statements without strict SQL mode enabled. This change has been reverted so that UPDATE again had the original behavior. Problems can still occur if you replicate between servers that have the modified UPDATE behavior and those that do not. (Bug#39265)

Replication: START SLAVE UNTIL did not work correctly with --replicate-same-server-id enabled; when started with this option, the slave did not perform events recorded in the relay log and that originated from a different master.

Log rotation events are automatically generated and written when rotating the binary log or relay log. Such events for relay logs are usually ignored by the slave SQL thread because they have the same server ID as that of the slave. However, when --replicate-same-server-id was enabled, the rotation event for the relay log was treated as if it originated on the master, because the log's name and position were incorrectly updated. This caused the MASTER_POS_WAIT() function always to return NULL and thus to fail. (Bug#38734, Bug#38934)

Replication: TRUNCATE statements failed to replicate when statement-based binary logging mode was not available. The issue was observed when using InnoDB with the transaction isolation level set to READ UNCOMMITTED (thus forcing InnoDB not to allow statement-based logging). However, the same behavior could be reproduced using any transactional storage engine supporting only row-based logging, regardless of the isolation level. This was due to two separate problems:

As a result of this fix, TRUNCATE is now treated as DDL for purposes of binary logging and replication; that is, it is always logged as a statement and so no longer causes an error when replicated using a transactional storage engine such as InnoDB. (Bug#36763)

See also Bug#42643.

Replication: mysqlbinlog replay of CREATE TEMPORARY TABLE ... LIKE statements and of TRUNCATE statements used on temporary tables failed with Error 1146 (Table ... doesn't exist). (Bug#35583)

Replication: In statement mode, mysqlbinlog failed to issue a SET @@autommit statement when the autocommit mode was changed. (Bug#34541)

Replication: LOAD DATA INFILE statements did not replicate correctly from a master running MySQL 4.1 to a slave running MySQL 5.1 or later. (Bug#31240)

The use by libedit of the __weak_reference() macro caused compilation failure on FreeBSD. (Bug#42817)

A '%' character in SQL statements could cause the server to crash. (Bug#42634)

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)

On the IBM i5 platform, the MySQL configuration process caused the system version of pthread_setschedprio() to be used. This function returns SIGILL on i5 because it is not supported, causing the server to crash. Now the my_pthread_setprio() function in the mysys library is used instead. (Bug#42524)

The SSL certficates included with MySQL distributions were regenerated because the previous ones had expired. (Bug#42366)

User variables within triggers could cause a crash if the mysql_change_user() C API function was invoked. (Bug#42188)

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);

(Bug#42037)

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)

String reallocation could cause memory overruns. (Bug#41868)

mysql_install_db did not pass some relevant options to mysqld. (Bug#41828)

Setting innodb_locks_unsafe_for_binlog should be equivalent to setting the transaction isolation level to READ COMMITTED. However, if both of those things were done, nonmatching semi-consistently read rows were not unlocked when they should have been. (Bug#41671)

REPAIR TABLE crashed for compressed MyISAM tables. (Bug#41574)

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 server cannot execute INSERT DELAYED statements when statement-based binary logging is enabled, but the error message displayed only the table name, not the entire statement. (Bug#41121)

FULLTEXT indexes did not work for Unicode columns that used a custom UCA collation. (Bug#41084)

The Windows installer displayed incorrect product names in some images. (Bug#40845)

Changing innodb_thread_concurrency at runtime could cause errors. (Bug#40760)

SELECT statements could be blocked by INSERT DELAYED statements that were waiting for a lock, even with low_priority_updates enabled. (Bug#40536)

For InnoDB tables that used ROW_FORMAT=REDUNDANT, storage size of NULL columns could be determined incorrectly. (Bug#40369)

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)

When a MEMORY table became full, the error generated was returned to the client but was not written to the error log. (Bug#39886)

With row-based binary logging, replication of InnoDB tables containing NULL-valued BIT columns could fail. (Bug#39648)

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)

The innodb_stats_on_metadata system variable was not displayed by SHOW VARIABLES and was not settable at runtime. (Bug#38189)

Enabling the sync_frm system variable had no effect on the handling of .frm files for views. (Bug#38145)

For comparison of NULL to a subquery result inside IS NULL, the comparison could evaluate to NULL rather than to TRUE or FALSE. This occurred for expressions such as:

SELECT ... WHERE NULL IN (SELECT ...) IS NULL

(Bug#37822)

Setting myisam_repair_threads greater than 1 caused a server crash for table repair or alteration operations for MyISAM tables with multiple FULLTEXT indexes. (Bug#37756)

The mysql client sometimes improperly interpreted string escape sequences in nonstring contexts. (Bug#36391)

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)

libmysqld was not built with all character sets. (Bug#32831)

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)

For Solaris package installation using pkgadd, the postinstall script failed, causing the system tables in the mysql database not to be created. (Bug#31164)

If the default database was dropped, the value of character_set_database was not reset to character_set_server as it should have been. (Bug#27208)

The libedit library was upgraded to version 2.11. (Bug#42433)

Security Fix: Using an XPath expression employing a scalar expression as a FilterExpr with ExtractValue() or UpdateXML() caused the server to crash. Such expressions now cause an error instead. (Bug#42495)

On the IBM i5 platform, the MySQL configuration process caused the system version of pthread_setschedprio() to be used. This function returns SIGILL on i5 because it is not supported, causing the server to crash. Now the my_pthread_setprio() function in the mysys library is used instead. (Bug#42524)

The SSL certficates included with MySQL distributions were regenerated because the previous ones had expired. (Bug#42366)

User variables within triggers could cause a crash if the mysql_change_user() C API function was invoked. (Bug#42188)

Some queries using NAME_CONST(.. COLLATE ...) led to a server crash due to a failed type cast. (Bug#42014)

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)

Performance of SELECT * retrievals from INFORMATION_SCHEMA.COLUMNS was improved slightly. (Bug#38918)

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 with FOR ORDER BY or FOR GROUP BY are silently ignored. Index hints with FOR JOIN or no FOR modifier are honored. In contrast to how hints apply for non-FULLTEXT searches, the hint is used for all phases of query execution (finding rows and retrieval, grouping, and ordering). This is true even if the hint is given for a non-FULLTEXT index. (Bug#38842)

Important Change: Replication: If a trigger was defined on an InnoDB table and this trigger updated a nontransactional table, changes performed on the InnoDB table were replicated and were visible on the slave before they were committed on the master, and were not rolled back on the slave after a successful rollback of those changes on the master.

As a result of the fix for this issue, the semantics of mixing nontransactional and transactional tables in a transaction in the first statement of a transaction have changed. Previously, if the first statement in a transaction contained nontransactional changes, the statement was written directly to the binary log. Now, any statement appearing after a BEGIN (or immediately following a COMMIT if AUTOCOMMIT = 0) is always considered part of the transaction and cached. This means that nontransactional changes do not propagate to the slave until the transaction is committed and thus written to the binary log.

See Section 16.3.1.25, “Replication and Transactions”, for more information about this change in behavior. (Bug#40116)

Partitioning: Replication: Changing the transaction isolation level while replicating partitioned InnoDB tables could cause statement-based logging to fail. (Bug#39084)

Partitioning: A comparison with an invalid DATE value in a query against a partitioned table could lead to a crash of the MySQL server.

(Bug#40972)

Partitioning: This bug was introduced in MySQL 5.1.29. (Bug#40954)

This regression was introduced by Bug#30573, Bug#33257, Bug#33555.

Partitioning: With READ COMMITTED transaction isolation level, InnoDB uses a semi-consistent read that releases nonmatching rows after MySQL has evaluated the WHERE clause. However, this was not happening if the table used partitions. (Bug#40595)

Partitioning: A query that timed out when run against a partitioned table failed silently, without providing any warnings or errors, rather than returning Lock wait timeout exceeded. (Bug#40515)

Partitioning: ALTER TABLE ... REORGANIZE PARTITION could crash the server when the number of partitions was not changed. (Bug#40389)

See also Bug#41945.

Partitioning: For a partitioned table having an AUTO_INCREMENT column: If the first statement following a start of the server or a FLUSH TABLES statement was an UPDATE statement, the AUTO_INCREMENT column was not incremented correctly. (Bug#40176)

Partitioning: The server attempted to execute the statements ALTER TABLE ... ANALYZE PARTITION, ALTER TABLE ... CHECK PARTITION, ALTER TABLE ... OPTIMIZE PARTITION, and ALTER TABLE ... REORGANIZE PARTITION on tables that were not partitioned. (Bug#39434)

See also Bug#20129.

Partitioning: The value of the CREATE_COLUMNS column in INFORMATION_SCHEMA.TABLES was not partitioned for partitioned tables. (Bug#38909)

Partitioning: When executing an ORDER BY query on a partitioned InnoDB table using an index that was not in the partition expression, the results were sorted on a per-partition basis rather than for the table as a whole. (Bug#37721)

Partitioning: Dropping or creating an index on a partitioned table managed by the InnoDB Plugin locked the table. (Bug#37453)

Partitioning: Partitioned table checking sometimes returned a warning with an error code of 0, making proper response to errors impossible. The fix also renders the error message subject to translation in non-English deployments. (Bug#36768)

Partitioning: SHOW TABLE STATUS could show a nonzero value for the Mean record length of a partitioned InnoDB table, even if the table contained no rows. (Bug#36312)

Partitioning: When SHOW CREATE TABLE was used on a partitioned table, all of the table's PARTITION and SUBPARTITION clauses were output on a single line, making it difficult to read or parse. (Bug#14326)

Replication: Per-table AUTO_INCREMENT option values were not replicated correctly for InnoDB tables. (Bug#41986)

Replication: Some log_event types did not skip the post-header when reading. (Bug#41961)

Replication: Attempting to read a binary log containing an Incident_log_event having an invalid incident number could cause the debug server to crash. (Bug#40482)

Replication: When using row-based replication, an update of a primary key that was rolled back on the master due to a duplicate key error was not rolled back on the slave. (Bug#40221)

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)

Replication: With row-based replication, UPDATE and DELETE statements using LIMIT and a table's primary key could produce different results on the master and slave. (Bug#38230)

resolve_stack_dump was unable to resolve the stack trace format produced by mysqld in MySQL 5.1 and up (see Section 22.5.1.5, “Using a Stack Trace”). (Bug#41612)

In example option files provided in MySQL distributions, the thread_stack value was increased from 64K to 128K. (Bug#41577)

The optimizer could ignore an error and rollback request during a filesort, causing an assertion failure. (Bug#41543)

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 a repair operation was carried out on a CSV table, the debug server crashed. (Bug#41441)

When substituting system constant functions with a constant result, the server was not expecting NULL function return values and could crash. (Bug#41437)

Queries such as SELECT ... CASE AVG(...) WHEN ... that used aggregate functions in a CASE expression crashed the server. (Bug#41363)

INSERT INTO .. SELECT ... FROM and CREATE TABLE ... SELECT ... FROM a TEMPORARY table could inadvertently change the locking type of the temporary table from a write lock to a read lock, causing statement failure. (Bug#41348)

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)

For debug servers, OPTIMIZE TABLE on a compressed table caused a server crash. (Bug#40949)

Accessing user variables within triggers could cause a server crash. (Bug#40770)

IF(..., CAST(longtext_val AS UNSIGNED), signed_val) as an argument to an aggregate function could cause an assertion failure. (Bug#40761)

For single-table UPDATE statements, an assertion failure resulted from a runtime error in a stored function (such as a recursive function call or an attempt to update the same table as in the UPDATE statement). (Bug#40745)

TRUNCATE TABLE for an InnoDB table did not flush cached queries for the table. (Bug#40386)

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)

The ':' character was incorrectly disallowed in table names. (Bug#40104)

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)

Retrieval speed from the following INFORMATION_SCHEMA tables was improved by shortening the VARIABLE_VALUE column to 1024 characters: GLOBAL_VARIABLES, SESSION_VARIABLES, GLOBAL_STATUS, and SESSION_STATUS.

As a result of this change, any variable value longer than 1024 characters will be truncated with a warning. This affects only the init_connect system variable. (Bug#39955)

For an InnoDB table, DROP TABLE or ALTER TABLE ... DISCARD TABLESPACE could take a long time or cause a server crash. (Bug#39939)

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)

For a server started with the --temp-pool option on Windows, temporary file creation could fail. This option now is ignored except on Linux systems, which was its original intended scope. (Bug#39750)

ALTER TABLE on a table with FULLTEXT index that used a pluggable FULLTEXT parser could cause debug servers to crash. (Bug#39746)

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)

The do_abi_check program run during the build process depends on mysql_version.h but that file was not created first, resulting in build failure. (Bug#39571)

CHECK TABLE failed for MyISAM INFORMATION_SCHEMA tables. (Bug#39541)

On 64-bit Windows systems, the server accepted key_buffer_size values larger than 4GB, but allocated less. (For example, specifying a value of 5GB resulted in 1GB being allocated.) (Bug#39494)

InnoDB could hang trying to open an adaptive hash index. (Bug#39483)

Following ALTER TABLE ... DISCARD TABLESPACE for an InnoDB table, an attempt to determine the free space for the table before the ALTER TABLE operation had completely finished could cause a server crash. (Bug#39438)

Use of the PACK_KEYS or MAX_ROWS table option in ALTER TABLE should have triggered table reconstruction but did not. (Bug#39372)

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)

A server built using yaSSL for SSL support would crash if configured to use an RSA key and a client sent a cipher list containing a non-RSA key as acceptable. (Bug#39178)

When built with Valgrind, the server failed to access tables created with the DATA DIRECTORY or INDEX DIRECTORY table option. (Bug#39102)

With binary logging enabled CREATE VIEW was subject to possible buffer overwrite and a server crash. (Bug#39040)

The fast mutex implementation was subject to excessive lock contention. (Bug#38941)

Use of InnoDB monitoring (SHOW ENGINE INNODB STATUS or one of the InnoDB Monitor tables) could cause a server crash due to invalid access to a shared variable in a concurrent environment. (Bug#38883)

InnoDB could fail to generate AUTO_INCREMENT values after an UPDATE statement for the table. (Bug#38839)

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)

On Windows, a five-second delay occurred at shutdown of applications that used the embedded server. (Bug#38522)

On Solaris, a scheduling policy applied to the main server process could be unintentionally overwritten in client-servicing threads. (Bug#38477)

Building MySQL on FreeBSD would result in a failure during the gen_lex_hash phase of the build. (Bug#38364)

On Windows, the embedded server would crash in mysql_library_init() if the language file was missing. (Bug#38293)

A mix of TRUNCATE TABLE with LOCK TABLES and UNLOCK TABLES for an InnoDB could cause a server crash. (Bug#38231)

The ExtractValue() function did not work correctly with XML documents containing a DOCTYPE declaration. (Bug#38227)

Queries with a HAVING clause could return a spurious row. (Bug#38072)

The Event Scheduler no longer logs “started in thread” or “executed” successfully messages to the error log. (Bug#38066)

Use of spatial data types in prepared statements could cause memory leaks or server crashes. (Bug#37956, Bug#37671)

An error in a debugging check caused crashes in debug servers. (Bug#37936)

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)

The presence of a /* ... */ comment preceding a query could cause InnoDB to use unnecessary gap locks. (Bug#37885)

Use of an uninitialized constant in EXPLAIN evaluation caused an assertion failure. (Bug#37870)

When using ALTER TABLE on an InnoDB table, the AUTO_INCREMENT value could be changed to an incorrect value. (Bug#37788)

Primary keys were treated as part of a covering index even if only a prefix of a key column was used. (Bug#37742)

Renaming an ARCHIVE table to the same name with different lettercase and then selecting from it could cause a server crash. (Bug#37719)

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.

SHOW PROCESSLIST displayed “copy to tmp table” when no such copy was occurring. (Bug#37550)

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)

Certain boolean-mode FULLTEXT searches that used the truncation operator did not return matching records and calculated relevance incorrectly. (Bug#37245)

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)

For an InnoDB table with a FOREIGN KEY constraint, TRUNCATE TABLE may be performed using row by row deletion. If an error occurred during this deletion, the table would be only partially emptied. Now if an error occurs, the truncation operation is rolled back and the table is left unchanged. (Bug#37016)

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)

The mysql client, when built with Visual Studio 2005, did not display Japanese characters. (Bug#36279)

CREATE INDEX for InnoDB tables could under very rare circumstances cause the server to crash.. (Bug#36169)

A read past the end of the string could occur while parsing the value of the --innodb-data-file-path option. (Bug#36149)

Setting the slave_compressed_protocol system variable to DEFAULT failed in the embedded server. (Bug#35999)

For upgrades to MySQL 5.1 or higher, mysql_upgrade did not re-encode database or table names that contained nonalphanumeric characters. (They would still appear after the upgrade with the #mysql50# prefix described in Section 8.2.3, “Mapping of Identifiers to File Names”.) To correct this problem, it was necessary to run mysqlcheck --all-databases --check-upgrade --fix-db-names --fix-table-names manually. mysql_upgrade now runs that command automatically after performing the initial upgrade. (Bug#35934)

SHOW CREATE TABLE did not display a printable value for the default value of BIT columns. (Bug#35796)

The columns that store character set and collation names in several INFORMATION_SCHEMA tables were lengthened because they were not long enough to store some possible values: SCHEMATA, TABLES, COLUMNS, CHARACTER_SETS, COLLATIONS, and COLLATION_CHARACTER_SET_APPLICABILITY. (Bug#35789)

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)

InnoDB was not updating the Handler_delete or Handler_update status variables. (Bug#35537)

InnoDB could fail to generate AUTO_INCREMENT values if rows previously had been inserted containing literal values for the AUTO_INCREMENT column. (Bug#35498, Bug#36411, Bug#39830)

The CREATE_OPTIONS column for INFORMATION_SCHEMA.TABLES did not display the KEY_BLOCK_SIZE option. (Bug#35275)

Selecting from an INFORMATION_SCHEMA table into an incorrectly defined MERGE table caused an assertion failure. (Bug#35068)

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)

SHOW GLOBAL STATUS shows values that aggregate the session status values for all threads. This did not work correctly for the embedded server. (Bug#34517)

mysqldumpslow did not aggregate times. (Bug#34129)

mysql_config did not output -ldl (or equivalent) when needed for --libmysqld-libs, so its output could be insufficient to build applications that use the embedded server. (Bug#34025)

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 5.1.26. (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)

Three conditions were discovered that could cause an upgrade from MySQL 5.0 to 5.1 to fail: 1) Triggers associated with a table that had a #mysql50# prefix in the name could cause assertion failure. 2) ALTER DATABASE ... UPGRADE DATA DIRECTORY NAME failed for databases that had a #mysql50# prefix if there were triggers in the database. 3) mysqlcheck --fix-table-name didn't use UTF8 as the default character set, resulting in parsing errors for tables with nonlatin symbols in their names and trigger definitions. (Bug#33094, Bug#41385)

Execution of a prepared statement that referred to a system variable caused a server crash. (Bug#32124)

Some division operations produced a result with incorrect precision. (Bug#31616)

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)

Server variables could not be set to their current values on Linux platforms. (Bug#31177)

See also Bug#6958.

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)

Static storage engines and plugins that were disabled and dynamic plugins that were installed but disabled were not listed in the INFORMATION_SCHEMA appropriate PLUGINS or ENGINES table. (Bug#29263)

Some SHOW statements and retrievals from the INFORMATION_SCHEMA TRIGGERS and EVENTS tables used a temporary table and incremented the Created_tmp_disk_tables status variable, due to the way that TEXT columns are handled. The TRIGGERS.SQL_MODE, TRIGGERS.DEFINER, and EVENTS.SQL_MODE columns now are VARCHAR to avoid this problem. (Bug#29153)

For several read only system variables that were viewable with SHOW VARIABLES, attempting to view them with SELECT @@var_name or set their values with SET resulted in an unknown system variable error. Now they can be viewed with SELECT @@var_name and attempting to set their values results in a message indicating that they are read only. (Bug#28234)

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)

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)

Setting the session value of the max_allowed_packet or net_buffer_length system variable was allowed but had no effect. The session value of these variables is now read only. (Bug#22891)

See also Bug#32223.

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)

On Windows, moving an InnoDB .ibd file and then symlinking to it in the database directory using a .sym file caused a server crash. (Bug#11894)

Partitioning: A SELECT using a range WHERE condition with an ORDER BY on a partitioned table caused a server crash. (Bug#40494)

Replication: Row-based replication failed with nonpartitioned MyISAM tables having no indexes. (Bug#40004)

With statement-based binary logging format and a transaction isolation level of READ COMMITTED or stricter, InnoDB printed an error because statement-based logging might lead to inconsistency between master and slave databases. However, this error was printed even when binary logging was not enabled (in which case, no such inconsistency can occur). (Bug#40360)

The CHECK TABLE ... FOR UPGRADE statement did not check for incompatible collation changes made in MySQL 5.1.24 (Bug#27877). This also affects mysqlcheck and mysql_upgrade, which cause that statement to be executed. See Section 2.12.3, “Checking Whether Table Indexes Must Be Rebuilt”.

Prior to this fix, a binary upgrade (performed without dumping tables with mysqldump before the upgrade and reloading the dump file after the upgrade) would corrupt tables that have indexes that use the utf8_general_ci or ucs2_general_ci collation for columns that contain 'ß' LATIN SMALL LETTER SHARP S (German). After the fix, CHECK TABLE ... FOR UPGRADE properly detects the problem and warns about tables that need repair.

However, the fix is not backward compatible and can result in a downgrading problem under these circumstances:

The solution is to dump tables with mysqldump before the downgrade and reload the dump file after the downgrade. Alternatively, drop and recreate affected indexes. (Bug#40053)

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 TABLE ... SELECT and INSERT INTO ... SELECT failed if the source table was a log table. (Bug#34306)

XA transaction rollbacks could result in corrupted transaction states and a server crash. (Bug#28323)

ALTER TABLE for an ENUM column could change column values. (Bug#23113)

Important Change: The --skip-thread-priority option is now deprecated in MySQL 5.1 and is removed in MySQL 6.0 such that the server won't change the thread priorities by default. Giving threads different priorities might yield marginal improvements in some platforms (where it actually works), but it might instead cause significant degradation depending on the thread count and number of processors. Meddling with the thread priorities is a not a safe bet as it is very dependent on the behavior of the CPU scheduler and system where MySQL is being run. (Bug#35164, Bug#37536)

Important Change: The --log option now is deprecated and will be removed (along with the log system variable) in the future. Instead, use the --general_log option to enable the general query log and the --general_log_file=file_name option to set the general query log file name. The values of these options are available in the general_log and general_log_file system variables, which can be changed at runtime.

Similar changes were made for the --log-slow-queries option and log_slow_queries system variable. You should use the --slow_query_log and --slow_query_log_file=file_name options instead (and the slow_query_log and slow_query_log_file system variables).

The BUILD/compile-solaris-* scripts now compile MySQL with the mtmalloc library rather than malloc. (Bug#38727)

Incompatible Change: Replication: The default binary logging mode has been changed from MIXED to STATEMENT for compatibility with MySQL 5.0. (Bug#39812)

Incompatible Change: CHECK TABLE ... FOR UPGRADE did not check for incompatible collation changes made in MySQL 5.1.21 (Bug#29499) and 5.1.23 (Bug#27562, Bug#29461). This also affects mysqlcheck and mysql_upgrade, which cause that statement to be executed. See Section 2.12.3, “Checking Whether Table Indexes Must Be Rebuilt”. (Bug#39585)

See also Bug#40984.

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:

As a workaround to avoid this problem, use either of these approaches:

The downgrade problem introduced by the fix for this bug has been addressed as Bug#40021. (Bug#17823)

Important Change: Replication: The SUPER privilege is now required to change the session value of binlog_format as well as its global value. For more information about binlog_format, see Section 16.1.2, “Replication Formats”. (Bug#39106)

Partitioning: Replication: Replication to partitioned MyISAM tables could be slow with row-based binary logging. (Bug#35843)

Partitioning: If an error occurred when evaluating a column of a partitioned table for the partitioning function, the row could be inserted anyway. (Bug#38083)

Partitioning: Using INSERT ... SELECT to insert records into a partitioned MyISAM table could fail if some partitions were empty and others are not. (Bug#38005)

Partitioning: Ordered range scans on partitioned tables were not always handled correctly. In some cases this caused some rows to be returned twice. The same issue also caused GROUP BY query results to be aggregated incorrectly. (Bug#30573, Bug#33257, Bug#33555)

Replication: Server code used in binary logging could in some cases be invoked even though binary logging was not actually enabled, leading to asserts and other server errors. (Bug#38798)

Replication: Replication of BLACKHOLE tables did not work with row-based binary logging. (Bug#38360)

Replication: In some cases, a replication master sent a special event to a reconnecting slave to keep the slave's temporary tables, but they still had references to the “old” slave SQL thread and used them to access that thread's data. (Bug#38269)

Replication: Replication filtering rules were inappropiately applied when executing BINLOG pseudo-queries. One way in which this problem showed itself was that, when replaying a binary log with mysqlbinlog, RBR events were sometimes not executed if the --replicate-do-db option was specified. Now replication rules are applied only to those events executed by the slave SQL thread. (Bug#36099)

Replication: For a CREATE TABLE ... SELECT statement that creates a table in a database other than the current one, the table could be created in the wrong database on replication slaves if row-based binary logging is used. (Bug#34707)

Replication: A statement did not always commit or roll back correctly when the server was shut down; the error could be triggered by having a failing UPDATE or INSERT statement on a transactional table, causing an implicit rollback. (Bug#32709)

See also Bug#38262.

The Sun Studio compiler failed to build debug versions of the server due to use of features specific to gcc. (Bug#39451)

For a TIMESTAMP column in an InnoDB table, testing the column with multiple conditions in the WHERE clause caused a server crash. (Bug#39353)

References to local variables in stored procedures are replaced with NAME_CONST(name, value) 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)

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 col_name = DEFAULT could result in a server crash. (Bug#39002)

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)

For deadlock between two transactions that required a timeout to resolve, all server tables became inaccessible for the duration of the deadlock. (Bug#38804)

When inserting a string into a duplicate-key error message, the server could improperly interpret the string, resulting in a crash. (Bug#38701)

A race condition between threads sometimes caused unallocated memory to be addressed. (Bug#38692)

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)

An uninitialized variable in the query profiling code was corrected (detected by Valgrind). (Bug#38560)

A server crash resulted from execution of an UPDATE that used a derived table together with FLUSH TABLES. (Bug#38499)

Stored procedures involving substrings could crash the server on certain platforms due to invalid memory reads. (Bug#38469)

The handlerton-to-plugin mapping implementation did not free handler plugin references when the plugin was uninstalled, resulting in a server crash after several install/uninstall cycles. Also, on Mac OS X, the server crashed when trying to access an EXAMPLE table after the EXAMPLE plugin was installed. (Bug#37958)

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)

Access checks were skipped for SHOW PROCEDURE STATUS and SHOW FUNCTION STATUS, which could lead to a server crash or insufficient access checks in subsequent statements. (Bug#37908)

The <=> operator could return incorrect results when comparing NULL to DATE, TIME, or DATETIME values. (Bug#37526)

The combination of a subquery with a GROUP BY, an aggregate function calculated outside the subquery, and a GROUP BY on the outer SELECT could cause the server to crash. (Bug#37348)

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)

In some cases, references to views were confused with references to anonymous tables and privilege checking was not performed. (Bug#36086)

For crash reports on Windows, symbol names in stack traces were not correctly resolved. (Bug#35987)

ALTER EVENT changed the PRESERVE attribute of an event even when PRESERVE was not specified in the statement. (Bug#35981)

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)

With the PAD_CHAR_TO_FULL_LENGTH SQL mode enabled, a ucs2 CHAR column returned additional garbage after trailing space characters. (Bug#35720)

A trigger for an InnoDB table activating multiple times could lead to AUTO_INCREMENT gaps. (Bug#31612)

mysqldump could fail to dump views containing a large number of columns. (Bug#31434)

The server could improperly type user-defined variables used in the select list of a query. (Bug#26020)

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)

Important Change: mysqlbinlog now supports --verbose and --base64-output=DECODE-ROWS options to display row events as commented SQL statements. (The default otherwise is to display row events encoded as base-64 strings using BINLOG statements.) See Section 4.6.7.2, “mysqlbinlog Row Event Display”. (Bug#31455)

MySQL source distributions are now available in Zip format. (Bug#27742)

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 12.5.5.33, “SHOW PROFILES Syntax”, and Section 20.26, “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)

Important Change: Security Fix: Additional corrections were made for the symlink-related privilege problem originally addressed in MySQL 5.1.24. 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:

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: Some performance problems of SHOW ENGINE INNODB STATUS were reduced by removing used cells and Total number of lock structs in row lock hash table from the output. Now these values are present only if the UNIV_DEBUG symbol is defined at MySQL build time. (Bug#36941, Bug#36942)

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: An additional correction to the original MySQL 5.1.23 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.

Partitioning: When a partitioned table had a TIMESTAMP column defined with CURRENT_TIMESTAMP as the default but with no ON UPDATE clause, the column's value was incorrectly set to CURRENT_TIMESTAMP when updating across partitions. (Bug#38272)

Partitioning: myisamchk failed with an assertion error when analyzing a partitioned MyISAM table. (Bug#37537)

Partitioning: A LIST partitioned MyISAM table returned erroneous results when an index was present on a column in the WHERE clause and NOT IN was used on that column.

Searches using the index were also much slower then if the index were not present. (Bug#35931)

Partitioning: SELECT COUNT(*) was not correct for some partitioned tables using a storage engine that did not support HA_STATS_RECORDS_IS_EXACT. Tables using the ARCHIVE storage engine were known to be affected.

This was because ha_partition::records() was not implemented, and so the default handler::records() was used in its place. However, this is not correct behavior if the storage engine does not support HA_STATS_RECORDS_IS_EXACT.

The solution was to implement ha_partition::records() as a wrapper around the underlying partition records.

As a result of this fix, the rows column in the output of EXPLAIN PARTITIONS now includes the total number of records in the partitioned table. (Bug#35745)

Partitioning: MyISAM recovery enabled with the --myisam-recover option did not work for partitioned MyISAM tables. (Bug#35161)

Partitioning: When one user was in the midst of a transaction on a partitioned table, a second user performing an ALTER TABLE on this table caused the server to hang. (Bug#34604)

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: Row-based replication did not correctly copy TIMESTAMP values from a big-endian storage engine to a little-endian storage engine. (Bug#37076)

Replication: INSTALL PLUGIN and UNINSTALL PLUGIN caused row-based replication to fail.

(Bug#35807)

Server-side cursors were not initialized properly, which could cause a server crash. (Bug#38486)

A server crash or Valgrind warnings could result when a stored procedure selected from a view that referenced a function. (Bug#38291)

A failure to clean up binary log events was corrected (detected by Valgrind). (Bug#38290)

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)

Over-aggressive lock acquisition by InnoDB when calculating free space for tablespaces could result in performance degradation when multiple threads were executing statements on multi-core machines. (Bug#38185)

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)

If the table definition cache contained tables with many BLOB columns, much memory could be allocated to caching BLOB values. Now a size limit on the cached BLOB values is enforced. (Bug#38002)

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 = constant) AND c2 ..., the query could return an unexpected result set. (Bug#37799)

The server returned unexpected results if a right side of the NOT IN clause consisted of the NULL value and some constants of the same type. For example, this query might return 3, 4, 5, and so forth if a table contained those values:

SELECT * FROM t WHERE NOT t.id IN (NULL, 1, 2);

(Bug#37761)

Setting the session value of the innodb_table_locks system variable caused a server crash. (Bug#37669)

Nesting of IF() inside of SUM() could cause an extreme server slowdown. (Bug#37662)

Killing a query that used an EXISTS subquery as the argument to SUM() or AVG() caused a server crash. (Bug#37627)

When using indexed ORDER BY sorting, incorrect query results could be produced if the optimizer switched from a covering index to a noncovering index. (Bug#37548)

After TRUNCATE TABLE for an InnoDB table, inserting explicit values into an AUTO_INCREMENT column could fail to increment the counter and result in a duplicate-key error for subsequent insertion of NULL. (Bug#37531)

Within stored programs or prepared statements, REGEXP could return incorrect results due to improper initialization. (Bug#37337)

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)

If the length of a field was 3, internal InnoDB to integer type conversion didn't work on big-endian machines in the row_search_autoinc_column() function. (Bug#36793)

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)

The CSV storage engine returned success even when it failed to open a table's data file. (Bug#36638)

SELECT DISTINCT from a simple view on an InnoDB table, where all selected columns belong to the same unique index key, returned incorrect results. (Bug#36632)

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)

If initialization of an INFORMATION_SCHEMA plugin failed, INSTALL PLUGIN freed some internal plugin data twice. (Bug#36399)

For InnoDB tables, the DATA_FREE column of the INFORMATION_SCHEMA.TABLES displayed free space in kilobytes rather than bytes. Now it displays bytes. (Bug#36278)

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)

The mysql client failed to recognize comment lines consisting of -- followed by a newline. (Bug#36244)

The server could crash with an assertion failure (or cause the client to get a “Packets out of order” error) when the expected query result was that it should terminate with a “Subquery returns more than 1 row” error. (Bug#36135)

The UUID() function returned UUIDs with the wrong time; this was because the offset for the time part in UUIDs was miscalculated. (Bug#35848)

The configure script did not allow utf8_hungarian_ci to be specified as the default collation. (Bug#35808)

On 64-bit systems, assigning values of 2 ⁶³ – 1 or larger to key_buffer_size caused memory overruns. (Bug#35616)

For InnoDB tables, REPLACE statements used “traditional” style locking, regardless of the setting of innodb_autoinc_lock_mode. Now REPLACE works the same way as “simple inserts” instead of using the old locking algorithm. (REPLACE statements are treated in the same way as as INSERT statements.) (Bug#35602)

Freeing of an internal parser stack during parsing of complex stored programs caused a server crash. (Bug#35577, Bug#37269, Bug#37228)

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)

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)

Table checksum calculation could cause a server crash for FEDERATED tables with BLOB columns containing NULL values. (Bug#34779)

A significant slowdown occurred when many SELECT statements that return many rows from InnoDB tables were running concurrently. (Bug#34409)

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)

Fast ALTER TABLE operations were not fast for columns that used multibyte character sets. (Bug#33873)

The internal functions my_getsystime(), my_micro_time(), and my_micro_time_and_time() did not work correctly on Windows. One symptom was that uniqueness of UUID() values could be compromised. (Bug#33748)

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)

mysql_install_db failed if the default storage engine was NDB. Now it explicitly uses MyISAM as the storage engine when running mysqld --bootstrap. (Bug#31315)

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)

TRUNCATE TABLE for InnoDB tables returned a count showing too many rows affected. Now the statement returns 0 for InnoDB tables. (Bug#29507)

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 FLUSH PRIVILEGES statement did not produce an error when it failed. (Bug#21226)

After executing a prepared statement that accesses a stored function, the next execution would fail to find the function if the stored function cache was flushed in the meantime. (Bug#12093, Bug#21294)

mysqltest now installs signal handlers and generates a stack trace if it crashes. (Bug#37003)

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)

Partitioning: Incompatible Change: On Mac OS X with lower_case_table_names = 2, the server could not read partitioned tables whose names contained uppercase letters.

Partitioned tables using mixed case names should be renamed or dropped before upgrading to this version of the server on Mac OS X. (Bug#37402)

Important Change: Partitioning: The statements ANALYZE TABLE, CHECK TABLE, OPTIMIZE TABLE, and REPAIR TABLE are now supported for partitioned tables.

Also as a result of this fix, the following statements which were disabled in MySQL 5.1.24 have been re-enabled:

(Bug#20129)

See also Bug#39434.

Replication: Issuing a DROP DATABASE while any temporary tables were open caused the server to switch to statement-based mode. (Bug#38773)

Replication: The --replicate-*-table options were not evaluated correctly when replicating multi-table updates.

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)

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)

Important Change: Incompatible Change: The FEDERATED storage engine is now disabled by default in binary distributions. The engine is still available and can be enabled by starting the server with the --federated option. (Bug#37069)

mysqltest was changed to be more robust in the case of a race condition that can occur for rapid disconnect/connect sequences with the server. The account used by mysqltest could reach its allowed simultaneous-sessions user limit if the connect attempt occurred before the server had fully processed the preceding disconnect. mysqltest now checks specificaly for a user-limits error when it connects; if that error occurs, it delays briefly before retrying. (Bug#23921)

Replication: Row-based replication broke for utf8 CHAR columns longer than 85 characters. (Bug#37426)

Replication: Performing an insert on a table having an AUTO_INCREMENT column and an INSERT trigger that was being replicated from a master running MySQL 5.0 or any version of MySQL 5.1 up to and including MySQL 5.1.11 to a slave running MySQL 5.1.12 or later caused the replication slave to crash. (Bug#36443)

See also Bug#33029.

Some binary distributions had a duplicate “-64bit” suffix in the file name. (Bug#37623)

NOT IN subqueries that selected MIN() or MAX() values but produced an empty result could cause a server crash. (Bug#37004)

ha_innodb.so was incorrectly installed in the lib/mysql directory rather than in lib/mysql/plugin. (Bug#36434)

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)

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 5.1.27. (Bug#33812)

Incompatible Change: A change has been made to the way that the server handles prepared statements. This affects prepared statements processed at the SQL level (using the PREPARE statement) and those processed using the binary client-server protocol (using the mysql_stmt_prepare() C API function).

Previously, changes to metadata of tables or views referred to in a prepared statement could cause a server crash when the statement was next executed, or perhaps an error at execute time with a crash occurring later. For example, this could happen after dropping a table and recreating it with a different definition.

Now metadata changes to tables or views referred to by prepared statements are detected and cause automatic repreparation of the statement when it is next executed. Metadata changes occur for DDL statements such as those that create, drop, alter, rename, or truncate tables, or that analyze, optimize, or repair tables. Repreparation also occurs after referenced tables or views are flushed from the table definition cache, either implicitly to make room for new entries in the cache, or explicitly due to FLUSH TABLES.

Repreparation is automatic, but to the extent that it occurs, performance of prepared statements is diminished.

Table content changes (for example, with INSERT or UPDATE) do not cause repreparation, nor do SELECT statements.

An incompatibility with previous versions of MySQL is that a prepared statement may now return a different set of columns or different column types from one execution to the next. For example, if the prepared statement is SELECT * FROM t1, altering t1 to contain a different number of columns causes the next execution to return a number of columns different from the previous execution.

Older versions of the client library cannot handle this change in behavior. For applications that use prepared statements with the new server, an upgrade to the new client library is strongly recommended.

Along with this change to statement repreparation, the default value of the table_definition_cache system variable has been increased from 128 to 256. The purpose of this increase is to lessen the chance that prepared statements will need repreparation due to referred-to tables/views having been flushed from the cache to make room for new entries.

A new status variable, Com_stmt_reprepare, has been introduced to track the number of repreparations. (Bug#27420, Bug#27430, Bug#27690)

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.

(Bug#36055)

mysql_upgrade now has a --tmpdir option to enable the location of temporary files to be specified. (Bug#36469)

mysqldump now adds the LOCAL qualifier to the FLUSH TABLES statement that is sent to the server when the --master-data option is enabled. This prevents the FLUSH TABLES statement from replicating to slaves, which is disadvantageous because it would cause slaves to block while the statement executes. (Bug#35157)

See also Bug#38303.

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 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: When flushing tables, there was a slight chance that the flush occurred between the processing of one table map event and the next. Since the tables were opened one by one, subsequent locking of tables would cause the slave to crash. This problem was observed when replicating NDBCLUSTER or InnoDB tables, when executing multi-table updates, and when a trigger or a stored routine performed an (additional) insert on a table so that two tables were effectively being inserted into in the same statement. (Bug#36197)

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.

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)

When updating an existing instance (for example, from MySQL 5.0 to 5.1, or 5.1 to 6.0), the Instance Configuration Wizard unnecessarily prompted for a root password when there was an existing root password. (Bug#36305)

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)

mysqltest was performing escape processing for the --replace_result command, which it should not have been. (Bug#36041)

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 MERGE storage engine did a table scan for SELECT COUNT(*) statements when it could calculate the number of records from the underlying tables. (Bug#36006)

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)

The Event Scheduler was not designed to work under the embedded server. It is now disabled for the embedded server, and the event_scheduler system variable is not displayed. (Bug#35997)

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 “prefer full scan on clustered primary key over full scan of any secondary key” optimizer rule introduced by Bug#26447 caused a performance regression for some queries, so it has been disabled. (Bug#35850)

The server ignored any covering index used for ref access of a table in a query with ORDER BY if this index was incompatible with the ORDER BY list and there was another covering index compatible with this list. As a result, suboptimal execution plans were chosen for some queries that used an ORDER BY clause. (Bug#35844)

mysql_upgrade did not properly update the mysql.event table. (Bug#35824)

An incorrect error and message was produced for attempts to create a MyISAM table with an index (.MYI) file name that was already in use by some other MyISAM table that was open at the same time. For example, this might happen if you use the same value of the INDEX DIRECTORY table option for tables belonging to different databases. (Bug#35733)

Enabling the read_only system variable while autocommit mode was enabled caused SELECT statements for transactional storage engines to fail. (Bug#35732)

The combination of GROUP_CONCAT(), DISTINCT, and LEFT JOIN could crash the server when the right table is empty. (Bug#35298)

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)

The log_output system variable could be set to an illegal value. (Bug#34820)

On Windows 64-bit builds, an apparent compiler bug caused memory overruns for code in innobase/mem/*. Removed optimizations so as not to trigger this problem. (Bug#34297)

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 2.10.3, “Installing from the Development Source Tree”, for information about what they do.) (Bug#34291)

Executing a FLUSH PRIVILEGES statement after creating a temporary table in the mysql database with the same name as one of the MySQL system tables caused the server to crash.

(Bug#33275)

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)

Assignment of relative path names to general_log_file or slow_query_log_file did not always work. (Bug#32748)

The mysql.servers table was not created during installation on Windows. (Bug#28680, Bug#32797)

The jp test suite was not working. (Bug#28563)

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)

Important Change: MySQL Cluster: Packaging: Beginning with this release, standard MySQL 5.1 binaries are no longer built with support for the NDBCLUSTER storage engine, and the NDBCLUSTER code included in 5.1 mainline sources is no longer guaranteed to be maintained or supported. Those using MySQL Cluster in MySQL 5.1.23 and earlier MySQL 5.1 mainline releases should upgrade to MySQL Cluster NDB 6.2.15 or a later MySQL Cluster NDB 6.2 or 6.3 release. (Bug#36193)

Important Change: The FEDERATED storage engine is not included in binary distributions of MySQL 5.1.24. (It will be included again in 5.1.25.)

Replication: Introduced the slave_exec_mode system variable to control whether idempotent or strict mode is used for replication conflict resolution. Idempotent mode suppresses duplicate-key, no-key-found, and some other errors, and is needed for circular replication, multi-master replication, and some other complex replication setups when using MySQL Cluster. Strict mode is the default. (Bug#31609)

Replication: When running the server with --binlog-format=MIXED or --binlog-format=STATEMENT, a query that referred to a system variable used the slave's value when replayed on the slave. This meant that, if the value of a system variable was inserted into a table, the slave differed from the master. Now, statements that refer to a system variable are marked as “unsafe”, which means that:

(Bug#31168)

See also Bug#34732.

The PROCESS privilege now is required to start or stop the InnoDB monitor tables (see Section 13.6.13.2, “SHOW ENGINE INNODB STATUS and the InnoDB Monitors”). Previously, no privilege was required. (Bug#34053)

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)

Formerly, when the MySQL server crashed, the generated stack dump was numeric and required external tools to properly resolve the names of functions. This is not very helpful to users having a limited knowledge of debugging techniques. In addition, the generated stack trace contained only the names of functions and was formatted differently for each platform due to different stack layouts.

Now it is possible to take advantage of newer versions of the GNU C Library provide a set of functions to obtain and manipulate stack traces from within the program. On systems that use the ELF binary format, the stack trace contains important information such as the shared object where the call was generated, an offset into the function, and the actual return address. Having the function name also makes possible the name demangling of C++ functions.

The library generates meaningful stack traces on the following platforms: i386, x86_64, PowerPC, IA64, Alpha, and S390. On other platforms, a numeric stack trace is still produced, and the use of the resolve_stack_dump utility is still required. (Bug#31891)

mysqltest now has mkdir and rmdir commands for creating and removing directories. (Bug#31004)

The server uses less memory when loading privileges containing table grants. (Patch provided by Google.) (Bug#25175)

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)

SHOW OPEN TABLES now supports FROM and LIKE clauses. (Bug#12183)

The new read-only global system variables report_host, report_password, report_port, and report_user system variables provide runtime access to the values of the corresponding --report-host, --report-password, --report-port, and --report-user options.

Formerly it was possible to specify an innodb_flush_method value of fdatasync to obtain the default flush behavior of using fdatasync() for flushing. This is no longer possible because it can be confusing that a value of fdatasync causes use of fsync() rather than fdatasync().

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 13.6.10.4, “Adaptive Hash Indexes”.

Performance: 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)

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 is now disallowed. This is now also true of these options when used with partitioned tables and individual partitions of such tables.

(Bug#32167, CVE-2008-2079)

See also Bug#39277.

Security Fix: A client that connects to a malicious server could be tricked by the server into sending files from the client host to the server. This occurs because the libmysqlclient client library would respond to a FETCH LOCAL FILE request from the server even if the request is sent for statements from the client other than LOAD DATA LOCAL INFILE. The client library has been modified to respond to a FETCH LOCAL FILE request from the server only if is is sent in response to a LOAD DATA LOCAL INFILE statement from the client.

The client library now also checks whether CLIENT_LOCAL_FILE is set and refuses to send a local file if not.

(Bug#29605)

Important Change: Security Enhancement: On Windows Vista and Windows Server 2008, a user without administrative privileges does not have write permissions to the Program Files directory where MySQL and the associated data files are normally installed. Using data files located in the standard Program Files installation directory could therefore cause MySQL to fail, or lead to potential security issues in an installed instance.

To address the problem, on Windows XP, Windows Vista and Windows Server 2008, the datafiles and data file configuration are now set to the Microsoft recommended AppData folder. The AppData folder is typically located within the user's home directory.

(Bug#34593)

Incompatible Change: In MySQL 5.1.23, the last_errno and last_error members of the NET structure in mysql_com.h were renamed to client_last_errno and client_last_error. This was found to cause problems for connectors that use the internal NET structure for error handling. The change has been reverted. (Bug#34655)

See also Bug#12713.

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: The UPDATE statement allowed NULL to be assigned to NOT NULL columns (the implicit default value for the column data type was assigned). This was changed so that on error occurs.

This change was reverted, because the original report was determined not to be a bug: Assigning NULL to a NOT NULL column in an UPDATE statement should produce an error only in strict SQL mode and set the column to the implicit default with a warning otherwise, which was the original behavior. See Section 10.1.4, “Data Type Default Values”, and Bug#39265. (Bug#33699)

Incompatible Change: For packages that are built within their own prefix (for example, /usr/local/mysql) the plugin directory will be lib/plugin. For packages that are built to be installed into a system-wide prefix (such as RPM packages with a prefix of /usr), the plugin directory will be lib/mysql/plugin to ensure a clean /usr/lib hierarchy. In both cases, the $pkglibdir configuration setting is used at build time to set the plugin directory.

The current plugin directory location is available as the value of the plugin_dir system variable as before, but the mysql_config script now has a --plugindir option that can be used externally to the server by third-party plugin writers to obtain the default plugin directory path name and configure their installation directory appropriately. (Bug#31736)

Incompatible Change: The utf8_general_ci and ucs2_general_ci collations did not sort the letter "U+00DF SHARP S" equal to 's'.

As a result of this bug fix, indexes must be rebuilt for columns that use the utf8_general_ci or ucs2_general_ci collation for columns that contain SHARP S. See Section 2.12.3, “Checking Whether Table Indexes Must Be Rebuilt”. (Bug#27877)

See also Bug#37046.

Important Change: Partitioning: The following statements did not function correctly with corrupted or crashed tables and have been disabled:

ALTER TABLE ... REBUILD PARTITION is unaffected by this change and continues to be available. This statement and ALTER TABLE ... REORGANIZE PARTITION may be used to analyze and optimize partitioned tables, since these operations cause the partition files to be rebuilt. (Bug#20129)

See also Bug#39434.

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: InnoDB free space information is now shown in the Data_free column of SHOW TABLE STATUS and in the DATA_FREE column of the INFORMATION_SCHEMA.TABLES table. (Bug#32440)

See also Bug#11379.

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)

Important Change: An AFTER UPDATE trigger was not invoked when the UPDATE did not make any changes to the table for which the trigger was defined. Now AFTER UPDATE triggers behave the same in this regard as do BEFORE UPDATE triggers, which are invoked whether the UPDATE makes any changes in the table or not. (Bug#23771)

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)

Partitioning: In some cases, matching rows from a partitioned MyISAM using a BIT column as the primary key were not found by queries. (Bug#34358)

Partitioning: Enabling innodb_file_per_table produced problems with partitioning and tablespace operations on partitioned InnoDB tables, in some cases leading to corrupt partitions or causing the server to crash. (Bug#33429)

Partitioning: A table defined using PARTITION BY KEY and having a BIT column referenced in the partitioning key did not behave correctly; some rows could be inserted into the wrong partition, causing wrong results to be returned from queries. (Bug#33379)

Partitioning: For InnoDB tables, there was a race condition involving the data dictionary and repartitioning. (Bug#33349)

Partitioning: When ALTER TABLE DROP PARTITION was executed on a table on which there was a trigger, the statement failed with an error. This occurred even if the trigger did not reference any tables. (Bug#32943)

Partitioning: Currently, all partitions of a partitioned table must use the same storage engine. One may optinally specify the storage engine on a per-partition basis; however, where this is the done, the storage engine must be the same as used by the table as a whole. ALTER TABLE did not enforce these rules correctly, the result being that incaccurate error messages were shown when trying to use the statement to change the storage engine used by an individual partition or partitions. (Bug#31931)

Partitioning: Using the DATA DIRECTORY and INDEX DIRECTORY options for partitions with CREATE TABLE or ALTER TABLE statements appeared to work on Windows, although they are not supported by MySQL on Windows systems, and subsequent attempts to use the tables referenced caused errors. Now these options are disabled on Windows, and attempting to use them generates a warning. (Bug#30459)

Replication: The failure of a CREATE TABLE ... ENGINE=InnoDB ... SELECT statement caused the slave to lose data. (Bug#35762)

Replication: When using row-based replication, a slave could crash at startup because it received a row-based replication event that InnoDB could not handle due to an incorrect test of the query string provided by MySQL, which was NULL for row-based replication events. (Bug#35226)

Replication: insert_id was not written to the binary log for inserts into BLACKHOLE tables. (Bug#35178)

Replication: When using statement-based replication and a DELETE, UPDATE, or INSERT ... SELECT statement using a LIMIT clause is encountered, a warning that the statement is not safe to replicate in statement mode is now issued; when using MIXED mode, the statement is now replicated using the row-based format. (Bug#34768)

Replication: mysqlbinlog did not output the values of auto_increment_increment and auto_increment_offset when both were equal to their default values (for both of these variables, the default is 1). This meant that a binary log recorded by a client using the defaults for both variables and then replayed on another client using its own values for either or both of these variables produced erroneous results. (Bug#34732)

See also Bug#31168.

Replication: When the Windows version of mysqlbinlog read 4.1 binlogs containing LOAD DATA INFILE statements, it output backslashes as path separators, causing problems for client programs expecting forward slashes. In such cases, it now converts \\ to / in directory paths. (Bug#34355)

Replication: SHOW SLAVE STATUS failed when slave I/O was about to terminate. (Bug#34305)

Replication: The character sets and collations used for constant identifiers in stored procedures were not replicated correctly. (Bug#34289)

Replication: mysqlbinlog from a 5.1 or later MySQL distribution could not read binary logs generated by a 4.1 server when the logs contained LOAD DATA INFILE statements. (Bug#34141)

This regression was introduced by Bug#32407.

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.

(Bug#33862)

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: mysqlbinlog failed to release all of its memory after terminating abnormally. (Bug#33247)

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: The error message generated due to lack of a default value for an extra column was not sufficiently informative. (Bug#32971)

Replication: When a user variable was used inside an INSERT statement, the corresponding binlog event was not written to the binlog correctly. (Bug#32580)

Replication: When using row-based replication, deletes from a table with a foreign key constraint failed on the slave. (Bug#32468)

Replication: The --base64-output option for mysqlbinlog was not honored for all types of events. This interfered in some cases with performing point-in-time recovery. (Bug#32407)

Replication: SQL statements containing comments using -- syntax were not replayable by mysqlbinlog, even though such statements replicated correctly. (Bug#32205)

Replication: When using row-based replication from a master running MySQL 5.1.21 or earlier to a slave running 5.1.22 or later, updates of integer columns failed on the slave with Error in Unknown event: row application failed. (Bug#31583)

This regression was introduced by Bug#21842.

Replication: Replicating write, update, or delete events from a master running MySQL 5.1.15 or earlier to a slave running 5.1.16 or later caused the slave to crash. (Bug#31581)

Replication: When using row-based replication, the slave stopped when attempting to delete nonexistent rows from a slave table without a primary key. In addition, no error was reported when this occurred. (Bug#31552)

Replication: Errors due to server ID conflicts were reported only in the slave's error log; now these errors are also shown in the Server_IO_State column in the output of SHOW SLAVE STATUS. (Bug#31316)

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: Replication of LOAD DATA INFILE could fail when read_buffer_size was larger than max_allowed_packet. (Bug#30435)

Replication: Replication crashed with the NDB storage engine when mysqld was started with --character-set-server=ucs2. (Bug#29562)

Replication: The effects of scheduled events were not always correctly reproduced on the slave when using row-based replication. (Bug#29020)

Replication: Setting server_id did not update its value for the current session. (Bug#28908)

Replication: Some older servers wrote events to the binary log using different numbering from what is currently used, even though the file format number in the file is the same. Slaves running MySQL 5.1.18 and later could not read these binary logs properly. Binary logs from these older versions now are recognized and event numbers are mapped to the current numbering so that they can be interpreted properly. (Bug#27779, Bug#32434)

This regression was introduced by Bug#22583.

Replication: MASTER_POS_WAIT() did not return NULL when the server was not a slave. (Bug#26622)

Replication: The nonspecific 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:

(Bug#22989)

See also Bug#19328.

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. (Bug#18199, Bug#18453)

Replication: START SLAVE UNTIL MASTER_LOG_POS=position issued on a slave that was using --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)

Manually replacing a binary log file with a directory having the same name caused an error that was not handled correctly. (Bug#35675)

Using LOAD DATA INFILE with a view could crash the server. (Bug#35469)

Selecting from INFORMATION_SCHEMA.REFERENTIAL_CONSTRAINTS could cause a server crash. (Bug#35406)

See also Bug#35108.

For a TEMPORARY table, DELETE with no WHERE clause could fail when preceded by DELETE statements with a WHERE clause. (Bug#35392)

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)

In some cases, when too many clients tried to connect to the server, the proper SQLSTATE code was not returned. (Bug#35289)

Memory-allocation failures for attempts to set key_buffer_size to large values could result in a server crash. (Bug#35272)

For InnoDB tables, ALTER TABLE DROP failed if the name of the column to be dropped began with “foreign”. (Bug#35220)

Queries could return different results depending on whether ORDER BY columns were indexed. (Bug#35206)

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)

SELECT ... FROM INFORMATION_SCHEMA.REFERENTIAL_CONSTRAINTS caused the server to crash if the table referenced by a foreign key had been dropped. This issue was observed on Windows platforms only. (Bug#35108)

See also Bug#35406.

Debugging symbols were missing for some executables in Windows binary distributions. (Bug#35104)

Nonconnection threads were being counted in the value of the Max_used_connections status variable. (Bug#35074)

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.

For some queries, the optimizer used an ordered index scan for GROUP BY or DISTINCT when it was supposed to use a loose index scan, leading to incorrect results. (Bug#34928)

Creating a foreign key on an InnoDB table that was created with an explicit AUTO_INCREMENT value caused that value to be reset to 1. (Bug#34920)

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)

If fsync() returned ENOLCK, InnoDB could treat this as fatal and cause abnormal server termination. InnoDB now retries the operation. (Bug#34823)

CREATE SERVER and ALTER SERVER could crash the server if out-of-memory conditions occurred. (Bug#34790)

DROP SERVER does not release memory cached for server structures created by CREATE SERVER, so repeated iterations of these statements resulted in a memory leak. FLUSH PRIVILEGES now releases the memory allocated for CREATE SERVER. (Bug#34789)

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 col_name < N failed to return rows where the value of col_name is N - 1. (Bug#34731)

A server running with the --debug option could attempt to dereference a null pointer when opening tables, resulting in a crash. (Bug#34726)

Assigning an “incremental” value to the debug system variable did not add the new value to the current value. For example, if the current debug value was 'T', the statement SET debug = '+P' resulted in a value of 'P' rather than the correct value of 'P:T'. (Bug#34678)

For debug builds, reading from INFORMATION_SCHEMA.TABLES or INFORMATION_SCHEMA.COLUMNS could cause assertion failures. This could happen under rare circumstances when INFORMATION_SCHEMA fails to get information about a table (for example, when a connection is killed). (Bug#34656)

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)

Dangerous pointer arithmetic crashed the server on some systems. (Bug#34598)

Creating a view inside a stored procedure could lead to a crash of the MySQL Server. (Bug#34587)

A server crash could occur if INFORMATION_SCHEMA tables built in memory were swapped out to disk during query execution. (Bug#34529)

CAST(AVG(arg) AS DECIMAL) produced incorrect results for non-DECIMAL arguments. (Bug#34512)

The per-thread debugging settings stack was not being deallocated before thread termination, resulting in a stack memory leak. (Bug#34424)

Executing an ALTER VIEW statement on a table crashed the server. (Bug#34337)

InnoDB could crash if overflow occurred for an AUTO_INCREMENT column. (Bug#34335)

For InnoDB, exporting and importing a table could corrupt TINYBLOB columns, and a subsequent ALTER TABLE could corrupt TINYTEXT columns as well. (Bug#34300)

DEFAULT 0 was not allowed for the YEAR data type. (Bug#34274)

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)

Use of stored functions in the WHERE clause for SHOW OPEN TABLES caused a server crash. (Bug#34166)

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 an integer argument 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='string1' AND column2=constant1) OR (column1='string2' AND column2=constant2), where 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=constant1 OR column2=constant2. (Bug#33833)

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)

ORDER BY ... DESC sorts could produce misordered results. (Bug#33697)

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)

Threads created by the event scheduler were incorrectly counted against the max_connections thread limit, which could lead to client lockout. (Bug#33507)

Dropping a function after dropping the function's creator could cause the server to crash. (Bug#33464)

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.

ENUM- or SET-valued plugin variables could not be set from the command line. (Bug#33358)

Loading plugins via command-line options to mysqld could cause an assertion failure. (Bug#33345)

SLEEP(0) failed to return on 64-bit Mac OS X due to a bug in pthread_cond_timedwait(). (Bug#33304)

Using Control-R in the mysql client caused it to crash. (Bug#33288)

For MyISAM tables, CHECK TABLE (non-QUICK) and any form of REPAIR TABLE incorrected treated rows as corrupted under the combination of the following conditions:

This problem affected CHECK TABLE, REPAIR TABLE, OPTIMIZE TABLE, ALTER TABLE. CHECK TABLE reported and marked the table as crashed if any row was present that fulfilled the third condition. The other statements deleted these rows. (Bug#33222)

Granting the UPDATE privilege on one column of a view caused the server to crash. (Bug#33201)

For DECIMAL columns used with the ROUND(X,D) or TRUNCATE(X,D) function with a nonconstant value of D, adding an ORDER BY for the function result produced misordered output. (Bug#33143)

See also Bug#33402, Bug#30617.

The CSV engine did not honor update requests for BLOB columns when the new column value had the same length as the value to be updated. (Bug#33067)

After receiving a SIGHUP signal, the server could crash, and user-specified log options were ignored when reopening the logs. (Bug#33065)

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, MySQL 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)

Repeated creation and deletion of views within prepared statements could eventually crash the server. (Bug#32890)

See also Bug#34587.

Incorrect assertions could cause a server crash for DELETE triggers for transactional tables. (Bug#32790)

In some cases where setting a system variable failed, no error was sent to the client, causing the client to hang. (Bug#32757)

Enabling the PAD_CHAR_TO_FULL_LENGTH SQL mode caused privilege-loading operations (such as FLUSH PRIVILEGES) to include trailing spaces from grant table values stored in CHAR columns. Authentication for incoming connections failed as a result. Now privilege loading does not include trailing spaces, regardless of SQL mode. (Bug#32753)

The SHOW ENGINE INNODB STATUS and SHOW ENGINE INNODB MUTEX statements incorrectly required the SUPER privilege rather than the PROCESS privilege. (Bug#32710)

Inserting strings with a common prefix into a table that used the ucs2 character set corrupted the table. (Bug#32705)

Tables in the mysql database that stored the current sql_mode value as part of stored program definitions were not updated with newer mode values (NO_ENGINE_SUBSTITUTION, PAD_CHAR_TO_FULL_LENGTH). This causes various problems defining stored programs if those modes were included in the current sql_mode value. (Bug#32633)

A view created with a string literal for one of the columns picked up the connection character set, but not the collation. Comparison to that field therefore used the default collation for that character set, causing an error if the connection collation was not compatible with the default collation. The problem was caused by text literals in a view being dumped with a character set introducer even when this was not necessary, sometimes leading to a loss of collation information. Now the character set introducer is dumped only if it was included in the original query. (Bug#32538)

See also Bug#21505.

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)

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)

On Windows, mysqltest_embedded.exe did not properly execute the send command. (Bug#32044)

A variable named read_only could be declared even though that is a reserved word. (Bug#31947)

On Windows, the build process failed with four parallel build threads. (Bug#31929)

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 mysqld crash handler failed on Windows. (Bug#31745)

When upgrading from MySQL 5.1.19 to any version between MySQL 5.1.20 to MySQL 5.1.23, the MySQL Instance Configuration Wizard would fail to account for the change in name for the mysqld-nt.exe to mysqld.exe, causing MySQL to fail to start properly after the upgrade. (Bug#31674)

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 table having an index that included a BLOB or TEXT column, and that was originally created with a MySQL server using version 4.1 or earlier, could not be opened by a 5.1 or later server. (Bug#31331)

The -, *, and / operators and the functions POW() and EXP() could misbehave when used with floating-point numbers. Previously they might return +INF, -INF, or NaN in cases of numeric overflow (including that caused by division by zero) or when invalid arguments were used. Now NULL is returned in all such cases. (Bug#31236)

The mysql_change_user() C API function caused global Com_xxx status variable values to be incorrect. (Bug#31222)

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)

ROUND(X,D) or TRUNCATE(X,D) for nonconstant values of D could crash the server if these functions were used in an ORDER BY that was resolved using filesort. (Bug#30889)

Resetting the query cache by issuing a SET GLOBAL query_cache_size=0 statement caused the server to crash if it concurrently was saving a new result set to the query cache. (Bug#30887)

Manifest problems prevented MySQLInstanceConfig.exe from running on Windows Vista. (Bug#30823)

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 Table_locks_waited waited variable was not incremented in the cases that a lock had to be waited for but the waiting thread was killed or the request was aborted. (Bug#30331)

The Com_create_function status variable was not incremented properly. (Bug#30252)

View metadata returned from INFORMATION_SCHEMA.VIEWS was changed by the fix for Bug#11986, causing the information returned in MySQL 5.1 to differ from that returned in 5.0. (Bug#30217)

mysqld displayed the --enable-pstack option in its help message even if MySQL was configured without --with-pstack. (Bug#29836)

The mysql_config command would output CFLAGS values that were incompatible with C++ for the HP-UX platform. (Bug#29645)

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)

myisamchk always reported the character set for a table as latin1_swedish_ci (8) regardless of the table' actual character set. (Bug#29182)

InnoDB could return an incorrect rows-updated value for UPDATE statements. (Bug#29157)

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)

No information was written to the general query log for the COM_STMT_CLOSE, COM_STMT_RESET, and COM_STMT_SEND_LONG_DATA commands. (These occur when a client invokes the mysql_stmt_close(), mysql_stmt_reset() and mysql_stmt_send_long_data() C API functions.) (Bug#28386)

Previously, the parser accepted the ODBC { OJ ... LEFT OUTER JOIN ...} syntax for writing left outer joins. The parser now allows { OJ ... } to be used to write other types of joins, such as INNER JOIN or RIGHT OUTER JOIN. This helps with compatibility with some third-party applications, but is not official ODBC syntax. (Bug#28317)

The FEDERATED storage engine did not perform identifier quoting for column names that are reserved words when sending statements to the remote server. (Bug#28269)

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.

The parser rules for the SHOW PROFILE statement were revised to work with older versions of bison. (Bug#27433)

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)

Memory corruption, a crash of the MySQL server, or both, could take place if a low-level I/O error occurred while an ARCHIVE table was being opened. (Bug#26978)

DROP DATABASE failed for attempts to drop databases with names that contained the legacy #mysql50# name prefix. (Bug#26703)

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)

The XPath boolean() function did not cast string and nodeset values correctly in some cases. It now returns TRUE for any nonempty string or nodeset and 0 for a NULL string, as specified in the XPath standard.. (Bug#26051)

When symbolic links were disabled, either with a server startup option or by enabling the NO_DIR_IN_CREATE SQL mode, CREATE TABLE silently ignored the DATA DIRECTORY and INDEX DIRECTORY table options. Now the server issues a warning if symbolic links are disabled when these table options are used. (Bug#25677)

Attempting to create an index with a prefix on a DECIMAL column appeared to succeed with an inaccurate warning message. Now, this action fails with the error Incorrect prefix key; the used key part isn't a string, the used length is longer than the key part, or the storage engine doesn't support unique prefix keys. (Bug#25426)

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)

The Qcache_free_blocks status variable did not display a value of 0 if the query cache was disabled. (Bug#25132)

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)

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)

Logging of statements to log tables was incorrect for statements that contained utf8-incompatible binary strings. Incompatible sequences are hex-encoded now. (Bug#23924)

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)

perror reported incomplete or inaccurate information. (Bug#23028, Bug#25177)

InnoDB exhibited thread thrashing with more than 50 concurrent connections under an update-intensive workload. (Bug#22868)

After stopping and starting the event scheduler, disabled events could remain in the execution queue. (Bug#22738)

The server produced a confusing error message when attempting to open a table that required a storage engine that was not loaded. (Bug#22708)

For views or stored programs created with an invalid DEFINER value, the error message was confusing (did not tie the problem to the DEFINER clause) and has been improved. (Bug#21854)

Warnings for deprecated syntax constructs used in stored routines make sense to report only when the routine is being created, but they were also being reported when the routine was parsed for loading into the execution cache. Now they are reported only at routine creation time. (Bug#21801)

On Mac OS X, mysqld did not react to Ctrl-C when run under gdb, even when run with the --gdb option. (Bug#21567)

CREATE ... SELECT did not always set DEFAULT column values in the new table. (Bug#21380)

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)

The BENCHMARK() function, invoked with more than 2147483648 iterations (the size of a signed 32-bit integer), terminated prematurely. (Bug#20752)

mysqldumpslow returned a confusing error message when no configuration file was found. (Bug#20455)

MySQLInstanceConfig.exe could lose the innodb_data_home_dir setting when reconfiguring an instance. (Bug#19797)

DROP DATABASE did not drop orphaned FOREIGN KEY constraints. (Bug#18942)

CREATE TABLE allowed 0 as the default value for a TIMESTAMP column when the server was running in NO_ZERO_DATE mode. (Bug#18834)

A SET column whose definition specified 64 elements could not be updated using integer values. (Bug#15409)

If a SELECT calls a stored function in a transaction, and a statement within the function fails, that statement should roll back. Furthermore, if ROLLBACK is executed after that, the entire transaction should be rolled back. Before this fix, the failed statement did not roll back when it failed (even though it might ultimately get rolled back by a ROLLBACK later that rolls back the entire transaction). (Bug#12713)

See also Bug#34655.

The parser incorrectly allowed SQLSTATE '00000' to be specified for a condition handler. (This is incorrect because the condition must be a failure condition and '00000' indicates success.) (Bug#8759)

MySQLInstanceConfig.exe did not save the innodb_data_home_dir value to the my.ini file under certain circumstances. (Bug#6627)

Important Change: Partitioning: Security Fix: It was possible, by creating a partitioned table using the DATA DIRECTORY and INDEX DIRECTORY options to gain privileges on other tables having the same name as the partitioned table. As a result of this fix, any table-level DATA DIRECTORY or INDEX DIRECTORY options are now ignored for partitioned tables. (Bug#32091, CVE-2007-5970)

See also Bug#29325, Bug#32111.

Incompatible Change: In MySQL 5.1.6, when log tables were implemented, the default log destination for the general query and slow query log was TABLE. This default has been changed to FILE, which is compatible with MySQL 5.0, but incompatible with earlier releases of MySQL 5.1 from 5.1.6 to 5.1.20. If you are upgrading from MySQL 5.0 to 5.1.21 or higher, no logging option changes should be necessary. However, if you are upgrading from 5.1.6 through 5.1.20 to 5.1.21 or higher and were using TABLE logging, use the --log-output=TABLE option explicitly to preserve your server's table-logging behavior.

The MySQL 5.1.23 fix is in addition to a fix in 5.1.21 because it turned out that the default was set in two places, only one of which was fixed the first time. (Bug#29993)

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. (Bug#28779)

MySQL Cluster: The following improvements have been made in the ndb_size.pl utility:

Support for a --socket option was also added.

For more information, see Section 17.6.21, “ndb_size.pl — NDBCLUSTER Size Requirement Estimator”. (Bug#28683, Bug#28253)

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)

Partitioning: Error messages for partitioning syntax errors have been made more descriptive. (Bug#29368)

Replication: Replication of the following SQL functions now switches to row-based logging in MIXED mode, and generates a warning in STATEMENT mode:

See Section 5.2.4.3, “Mixed Binary Logging Format”, for more information. (Bug#12092, Bug#28086, Bug#30244)

mysqldump information at the top of the output now shows the same information as mysqldump invoked with the -V option, namely the mysqldump version number, the MySQL server version, and the distribution. (Bug#32350)

mysqltest now has a change_user command to change the user for the current connection. (It invokes the mysql_change_user() C API function.) (Bug#31608)

mysql-test-run.pl now allows a suite name prefix to be specified in command-line arguments that name test cases. The test name syntax now is [suite_name.]test_name[.suffix]. For example, mysql-test-run.pl binlog.mytest runs the mytest.test test in the binlog test suite. (Bug#31400)

The --event-scheduler option without a value disabled the event scheduler. Now it enables the event scheduler. (Bug#31332)

mysqldump produces a -- Dump completed on DATE comment at the end of the dump if --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)

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 identifier lists, expression lists, and UDF expression lists. (Bug#30333)

Server parser performance was improved for boolean expressions. (Bug#30237)

The LAST_EXECUTED column of the INFORMATION_SCHEMA.EVENTS table now indicates when the event started executing rather than when it finished executing. As a result, the ENDS column is never less than LAST_EXECUTED. (Bug#29830)

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)

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)

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 'XXX', system error: errno. (Bug#28359)

MySQL now can be compiled with gcc 4.2.x. There was a problem involving a conflict with the min() and max() macros in my_global.h. (Bug#28184)

mysql-test-run.pl now supports a --combination option for specifying options to the mysqld server. This option is similar to --mysqld but should be given two or more times. mysql-test-run.pl executes multiple test runs, using the options for each instance of --combination in successive runs.

For test runs specific to a given test suite, an alternative to the use of --combination is to create a combinations file in the suite directory. The file should contain a section of options for each test run.

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.

Security Fix: Replication: It was possible for any connected user to issue a BINLOG statement, which could be used to escalate privileges.

Use of the BINLOG statement now requires the SUPER privilege. (Bug#31611, CVE-2007-6313)

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.

(Bug#33814, CVE-2008-0226, CVE-2008-0227)

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 is no longer possible to create CSV tables with NULL columns. However, for backwards compatibility, you can continue to use such tables that were created in previous MySQL releases. (Bug#32050)

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:

This behavior differs from previous versions as follows:

(Bug#31794)

Incompatible Change: Inserting a row with a NULL value for a DATETIME column results in a CSV file that the storage engine cannot read.

All CSV tables now need to be defined with each column marked as NOT NULL. An error is raised if you try to create a CSV table with columns that are not defined with NOT NULL. (Bug#31473, Bug#32817)

Incompatible Change: SET PASSWORD statements now cause an implicit commit, and thus are prohibited within stored functions and triggers. (Bug#30904)

Incompatible Change: The mysql_install_db script could fail to locate some components (including resolveip) during execution if the --basedir option was specified on the command-line or within the my.cnf file. This was due to a conflict when comparing the compiled-in values and the supplied values.

The --source-install command-line option to the script has been removed and replaced with the --srcdir option. mysql_install_db now locates components either using the compiled-in options, the --basedir option or --srcdir option. (Bug#30759)

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;

Now alias declarations can be declared only in the table_references part. Elsewhere in the statement, alias references are allowed but not alias declarations. Statements containing aliases that are no longer allowed must be rewritten. (Bug#30234)

See also Bug#27525.

Incompatible Change: Within a stored routine, it is no longer allowable to declare a cursor for a SHOW or DESCRIBE statement. This happened to work in some instances, but is no longer supported. In many cases, a workaround for this change is to use the cursor with a SELECT query to read from an INFORMATION_SCHEMA table that produces the same information as the SHOW statement. (Bug#29223)

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: GRANT and REVOKE statements now cause an implicit commit, and thus are prohibited within stored functions and triggers. (Bug#21975, Bug#21422, Bug#17244)

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 was read last. The latter file now is read earlier so that ~/.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)

Incompatible Change: A number of problems existed in the implementation of MERGE tables that could cause problems. The problems are summarized below:

The fix introduces the following changes in behavior:

(Bug#19627, Bug#25038, Bug#25700, Bug#26377, Bug#26379, Bug#26867, Bug#27660, Bug#30275, Bug#30491)

Important Change: MySQL Cluster: AUTO_INCREMENT columns had the following problems when used in NDB tables:

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)

Partitioning: Important Note: An apostrophe or single quote character (') used in the DATA DIRECTORY, INDEX DIRECTORY, or COMMENT for a PARTITION clause caused the server to crash. When used as part of a CREATE TABLE statement, the crash was immediate. When used in an ALTER TABLE statement, the crash did not occur until trying to perform a SELECT or DML statement on the table. In either case, the server could not be completely restarted until the .FRM file corresponding to the newly created or altered table was deleted.