MySQL Community Server 5.1.34がリリースされました。


最も普及しているオープンソースデータベース管理システムの新バージョンであるMySQL Community Server 5.1.34がリリースされました。MySQL 5.1.34は、プロダクションシステムでの使用をお勧めします。

MySQL 5.1.34は、AIX 5.2向けのMySQL 5.1としての最終リリースとなります。AIX 5.2向けのEOL(End of Life) を2009年4月30日に迎える予定です。詳細情報は、以下をご覧ください。


MySQL 5.1の新機能の概要については、以下を参照してください。


新たなサーバにMySQL 5.1.34をインストール、または以前のMySQLリリースからMySQL 5.1.34にアップグレードする際の情報については、以下を参照してください。


MySQL 5.1.34は、http://dev.mysql.com/downloads/から、ソースコード及び多くのプラットフォームのためのバイナリで現在利用可能です。




MySQL 5.1に関するオープンな問題の情報については、以下のエラッタリストを参照してください。






Functionality added or changed:

  * Optimizer: The @@optimizer_switch setting was backported from
    MySQL 6.0, with additional switches to turn on/off use of
    index_merge optimizations or a particular index_merge algorithm.
    For more information please seehttp://s.petrunia.net/blog/?p=52

Bugs fixed:

  * 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.
    See alsoBug#42914:http://bugs.mysql.com/42914.

  * Replication: Assigning an invalid directory for the
    --slave-load-tmpdir caused the replication slave to crash.

  * Replication: The mysql.procs_priv system table was not
    replicated. (Bug#42217:http://bugs.mysql.com/42217)

  * Replication: An INSERT DELAYED into a TIMESTAMP column issued
    concurrently with a 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:http://bugs.mysql.com/41719)

  * Replication: The MIXED binary logging format did not switch to
    row-based mode for statements containing the LOAD_FILE()
    function. (Bug#39701:http://bugs.mysql.com/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.
    See alsoBug#43746:http://bugs.mysql.com/43746.

  * An attempt by a user who did not have the SUPER privilege to
    kill a system thread could cause a server crash.

  * On Windows, incorrectly specified link dependencies in
    CMakeLists.txt resulted in link errors for mysql_embedded,
    mysqltest_embedded, and mysql_client_test_embedded.

  * 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.

  * If the value of the version_comment system variable was too
    long, the mysql client displayed a truncated startup message.

  * Queries of the following form returned an empty result: SELECT
    ... WHERE ... (col=col AND col=col) OR ... (false expression)

  * The strings/CHARSET_INFO.txt file was not included in source
    distributions. (Bug#42937:http://bugs.mysql.com/42937)

  * A dangling pointer in mysys/my_error.c could lead to client
    crashes. (Bug#42675:http://bugs.mysql.com/42675)

  * Passing an unknown time zone specification to CONVERT_TZ()
    resulted in a memory leak.

  * With more than two arguments, LEAST(), GREATEST(), and CASE
    could unnecessarily return Illegal mix of collations errors.

  * The mysql client could misinterpret its input if a line was
    longer than an internal buffer.

  * In the help command output displayed by mysql, the description
    for the \c (clear) command was misleading.

  * 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.

  * If --basedir was specified, mysqld_safe did not use it when
    attempting to locate my_print_defaults.

  * When MySQL was configured with the --with-max-indexes=128
    option, mysqld crashed.

  * Setting the join_buffer_size variable to its minimum value
    produced spurious warnings.

  * 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:http://bugs.mysql.com/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.

  * The Time column for SHOW PROCESSLIST output and the value of
    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.

  * Restoring a mysqldump dump file containing FEDERATED tables
    because the file contained the data for the table. Now only
    the table definition is dumped (because the data is located
    elsewhere). (Bug#21360:http://bugs.mysql.com/21360)