mysql_upgrade 引起的master/slave replication 中断

在生产环境master服务器上处理完《1548-Cannot loadfrom mysql.proc. The table is probably corrupted》后，接到报警信息，slave服务器复制中断查看slave 状态

mysql>show slave status

发现如下语句执行错误

DROP DATABASEIF EXISTS performance_schema

performance_schema是mysql自带的性能信息相关的库，mysql怎么会执行这个操作,看看错误日志吧

[root@db25522]# tail -n 500/data/my2/mysql/db25522.err

13051310:29:54 [Note] Error reading relay log event: slave SQL thread was killed

13051310:29:54 [ERROR] Error reading packet from server: Lost connection to MySQLserver during query ( server_errno=2013)

13051310:29:54 [Note] Slave I/O thread killed while reading event

13051310:29:54 [Note] Slave I/O thread exiting, read up to log 'mysql-bin.002734',position 1017737307

13051310:29:57 [Note] Slave I/O thread: connected to master '[e-mail]:3306',replicationstarted in log 'mysql-bin.002734' at position 1017737307

13051310:29:57 [Note] Slave SQL thread initialized, starting replication in log'mysql-bin.002734' at position 1017729051, relay log'/my/rlog/relay-bin.000764' position: 1017729197

13053015:00:53 [ERROR] Incorrect definition of table mysql.proc: expected column'comment' at position 15 to have type text, found type char(64).

13053015:00:53 [ERROR] Slave SQL: Query caused differenterrors on master and slave.    Error on master: message (format)='Cannot load from mysql.%s. The tableis probably corrupted' error code=1548 ; Error on slave: actual message='noerror', error code=0. Default database: 'performance_schema'. Query: 'DROP DATABASE IF EXISTS performance_schema',Error_code: 0

13053015:00:53 [Warning] Slave: Cannot load from mysql.proc. The table is probablycorrupted Error_code: 1548

13053015:00:53 [ERROR] Error running query, slave SQL thread aborted. Fix theproblem, and restart the slave SQL thread with "SLAVE START". We stoppedat log 'mysql-bin.002947' position 721651903

莫非是数据不一致导致的？发现master服务器有performance_schema这个库，但是slave服务器没有。在执行 mysql_upgrade -uroot 之前，主从复制在运行，判断操作发生在mysql_upgrade-uroot之后，分析master上日志，在这个时间段内mysql进行了那些操作

[root@db25522]# mysqlbinlog  --no-defaults --start-date='2013-05-3015:00:00'  --end-date='2013-05-3015:03:00'   mysql-bin.002947  >/root/tmp.log

查询日志发现

[root@db25522]#vi /root/tmp.log

/DROP

/*!*/;

# at721651876

#13053015:00:25 server id 13084  end_log_pos721651903     Xid = 435509540

COMMIT/*!*/;

# at 721651903

#13053015:00:53 server id 13084  end_log_pos721652022     Query   thread_id=418930    exec_time=0 error_code=1548

SETTIMESTAMP=1369897253/*!*/;

/*!\Cutf8 *//*!*/;

SET@@session.character_set_client=33,@@session.collation_connection=33,@@session.collation_server=28/*!*/;

DROP DATABASE IF EXISTS performance_schema

/*!*/;

# at721652022

#13053015:00:53 server id 13084  end_log_pos721652152     Query   thread_id=418930    exec_time=0 error_code=0

SETTIMESTAMP=1369897253/*!*/;

CREATEDATABASE performance_schema character set utf8

/*!*/;

# at721652152

#13053015:03:19 server id 13084  end_log_pos721652223     Query   thread_id=418956    exec_time=0 error_code=0

SETTIMESTAMP=1369897399/*!*/;

/*!\Cgbk *//*!*/;

SET@@session.character_set_client=28,@@session.collation_connection=28,@@session.collation_server=28/*!*/;

BEGIN

/*!*/;

# at721652223

#13053015:03:19 server id 13084  end_log_pos721652336     Query   thread_id=418956    exec_time=0 error_code=0

usePriceDB/*!*/;

SETTIMESTAMP=1369897399/*!*/;

好吧，上面红色部分，执行了这个操作，再看看slave错误日志

13053015:00:53 [ERROR] Incorrect definition of table mysql.proc: expected column'comment' at position 15 to have type text, found type char(64).

13053015:00:53 [ERROR] Slave SQL: Query caused different errors on master andslave.     Error on master: message(format)='Cannot load from mysql.%s. The table is probably corrupted' errorcode=1548 ; Error on slave: actual message='no error', error code=0. Defaultdatabase: 'performance_schema'. Query: 'DROP DATABASE IF EXISTSperformance_schema', Error_code: 0

13053015:00:53 [Warning] Slave: Cannot load from mysql.proc. The table is probablycorrupted Error_code: 1548

13053015:00:53 [ERROR] Error running query, slave SQL thread aborted. Fix the problem,and restart the slave SQL thread with "SLAVE START". We stopped atlog 'mysql-bin.002947' position 721651903

5.5的日志错误还是很人性化的，slave停止时读取的binlog 日志文件，位置都很清楚。这为我们restart slave提供了方便。既然是DROP DATABASE IFEXISTS performance_schema 导致的错误，那么跳过这条event。

Slave服务器：

mysql>show variables like '%skip%';

mysql>set global sql_slave_skip_counter =1;

mysql>slave start ;

复制正常

总结：复制虽然正常了。为什么 mysql_upgrade 会做 DROP DATABASE IF EXISTSperformance_schema 这个操作？希望遇到类似问题的朋友，一起交流。