MySQL Community Server 5.0.82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 5.0.82がリリースされました。MySQL Community Server 5.0リリースシリーズにおける本リリース以降、MySQL Enterprise Serverのカウンターパートに対応するバージョン番号を共有します。
MySQL 5.0.82は、http://dev.mysql.com/downloads/およびミラーサイトから、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。




AIX 5.2向けのEOL(End of Life) を2009年4月30日に迎え、AIX 5.2向けのサポートを廃止しましたのでご注意ください。MySQL 5.0.82は、MySQL 5.0系では、AIX 5.2用のバイナリを含まない、初めてのリリースです。

以下のセクションは、以前のリリースであるMySQL Community Server 5.0.81以来のMySQLソースコードにおけるバージョン間の変更を記載しています。より多くの修正ログ修正を含む変更ログは以下で確認できます。より詳細な個別の修正情報は以下を参照してください。



  * Replication: Restarting the replication slave --- either by
    using STOP SLAVE plus START SLAVE, or by restarting the slave
    mysqld process --- could sometimes cause the slave to crash
    when using a debug version of the server.

  * 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.
    See alsoBug#27571:http://bugs.mysql.com/27571,

  * Replication: Column aliases used inside subqueries were ignored
    in the binary log. (Bug#35515:http://bugs.mysql.com/35515)

  * Replication: The statements DROP PROCEDURE IF EXISTS and DROP
    FUNCTION IF EXISTS were not written to the binary log if the
    procedure or function to be dropped did not exist.
   See alsoBug#25705:http://bugs.mysql.com/25705.

  * Use of HANDLER statements with INFORMATION_SCHEMA tables
    caused a server crash. Now HANDLER is prohibited with such
    tables. (Bug#44151:http://bugs.mysql.com/44151)

  * myisamchk could display a negative Max keyfile length value.

  * 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.

  * UNION of floating-point numbers did unnecessary rounding.

  * 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

  * The InnoDB btr_search_drop_page_hash_when_freed() function had
    a race condition. (Bug#42279:http://bugs.mysql.com/42279)

  * Compressing a table with the myisampack utility caused the
    server to produce Valgrind warnings when it opened the table.

  * 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.

  * Multiple-table UPDATE statements did not properly activate
    triggers. (Bug#39953:http://bugs.mysql.com/39953)

  * An UPDATE statement that updated a column using the same
    DES_ENCRYPT() value for each row actually updated different
    rows with different values.

  * 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:http://bugs.mysql.com/33899)

  * CHECKSUM TABLE was not killable with KILL QUERY.

  * myisamchk and myisampack were not being linked with the
    library that enabled support for * filename pattern expansion.

  * COMMIT did not delete savepoints if there were no changes in
    the transaction. (Bug#26288:http://bugs.mysql.com/26288)

  * Several memory allocation functions were not being checked for
    out-of-memory return values.