磁盘写满导致MySQL复制失败的解决方案

吾爱主题阅读：153 2024-04-02 08:01:16 评论：0

案例场景

今天在线上发现一个问题，由于监控没有覆盖到，某台机器的磁盘被写满了，导致线上 MySQL主从复制出现问题。问题如下：

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 localhost.(none)>show slave status\G *************************** 1. row *************************** Slave_IO_State: Master_Host: 10.xx.xx.xx Master_User: replica Master_Port: 5511 Connect_Retry: 60 Master_Log_File: Read_Master_Log_Pos: 4 Relay_Log_File: relay-bin.001605 Relay_Log_Pos: 9489761 Relay_Master_Log_File: Slave_IO_Running: No Slave_SQL_Running: No Last_Errno: 13121 Last_Error: Relay log read failure: Could not parse relay log event entry. The possible reasons are: the master 's binary log is corrupted (you can check this by running ' mysqlbinlog ' on the binary log), the slave' s relay log is corrupted (you can check this by running 'mysqlbinlog' on the relay log), a network problem, the server was unable to fetch a keyring key required to open an encrypted relay log file, or a bug in the master 's or slave' s MySQL code. If you want to check the master 's binary log or slave' s relay log, you will be able to know their names by issuing 'SHOW SLAVE STATUS' on this slave.

于是查看error log，发现error log中的内容如下：

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 2021-03-31T11:34:39.367173+08:00 11 [Warning] [MY-010897] [Repl] Storing MySQL user name or password information in the master info repository is not secure and is therefore not recommended. Please consider using the USER and PASSWORD connection options for START SLAVE; see the 'START SLAVE Syntax' in the MySQL Manual for more information. 2021-03-31T11:34:39.368161+08:00 12 [ERROR] [MY-010596] [Repl] Error reading relay log event for channel '' : binlog truncated in the middle of event; consider out of disk space 2021-03-31T11:34:39.368191+08:00 12 [ERROR] [MY-013121] [Repl] Slave SQL for channel '' : Relay log read failure: Could not parse relay log event entry. The possible reasons are: the master 's binary log is corrupted (you can check this by running ' mysqlbinlog ' on the binary log), the slave' s relay log is corrupted (you can check this by running 'mysqlbinlog' on the relay log), a network problem, the server was unable to fetch a keyring key required to open an encrypted relay log file, or a bug in the master 's or slave' s MySQL code. If you want to check the master 's binary log or slave' s relay log, you will be able to know their names by issuing 'SHOW SLAVE STATUS' on this slave. Error_code: MY-013121 2021-03-31T11:34:39.368205+08:00 12 [ERROR] [MY-010586] [Repl] Error running query, slave SQL thread aborted. Fix the problem, and restart the slave SQL thread with "SLAVE START" . We stopped at log 'mysql-bin.000446' position 9489626

从描述中可以看到，error log是比较智能的，发现了磁盘问题，并提示我们需要"consider out of disk space"

解决问题

登录服务器，很快就发现是MySQL所在的服务器磁盘使用率达到100%了，问题原因跟error log中的内容一致。

现在就解决这个问题。基本的思路就是清理磁盘文件，然后重新搭建复制关系，这个过程似乎比较简单，但是实际操作中，在搭建复制关系的时候出现了下面的报错：

1 2 3 4 5 6 ### 基于gtid的复制，想重新搭建复制关系 localhost.(none)>reset slave; ERROR 1371 (HY000): Failed purging old relay logs: Failed during log reset localhost.(none)>reset slave all ; ERROR 1371 (HY000): Failed purging old relay logs: Failed during log reset

第一步：因为复制是基于gtid进行的，所以直接记录show slave status的状态后，就可以重新reset slave，并利用change master语句来重建复制关系了。

但是却出现上面的报错，从报错信息看是mysql无法完成purge relay log的操作，这看起来不科学。好吧，既然你自己不能完成purge relay logs的操作，那就让我来帮你吧。

第二步：手工rm -f 删除所有的relay log，发现报错变成了：

1 2	`localhost.(none)>reset slave` `all` `;` `ERROR 1374 (HY000): I/O error reading log` `index` `file`

嗯，好吧，问题没有得到解决。

然后思考了下，既然不能通过手工reset slave 来清理relay log，直接stop

slave 然后change master行不行呢？

第三步：直接stop slave，然后change master，不执行reset slave all的语句，结果如下：

1 2 3 4 5 6 localhost.(none)>change master to master_host= '10.13.224.31' , -> master_user= 'replica' , -> master_password= 'eHnNCaQE3ND' , -> master_port=5510, -> master_auto_position=1; ERROR 1371 (HY000): Failed purging old relay logs: Failed during log reset

得，问题依旧。

第四步：反正复制已经报错断开了，执行个start slave看看，结果戏剧性的一幕出现了：

1 2 3 4 5 6 7 8 9 10 11 12 localhost.(none)>start slave; ERROR 2006 (HY000): MySQL server has gone away No connection . Trying to reconnect... Connection id: 262 Current database : *** NONE *** Query OK, 0 rows affected (0.01 sec) localhost.(none)> [root@ ~]#

执行start slave之后，实例直接挂了。

到这里，复制彻底断开了，从库实例已经挂了。

第五步：看看实例还能不能重启，尝试重启实例，发现实例还能起来。实例重新起来后，查看复制关系，结果如下：

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 localhost.(none)>show slave status\G *************************** 1. row *************************** Slave_IO_State: Queueing master event to the relay log Master_Host: 10.xx.xx.xx Master_User: replica Master_Port: 5511 Connect_Retry: 60 Master_Log_File: Read_Master_Log_Pos: 4 Relay_Log_File: relay-bin.001605 Relay_Log_Pos: 9489761 Relay_Master_Log_File: Slave_IO_Running: Yes Slave_SQL_Running: No Last_Errno: 13121 Last_Error: Relay log read failure: Could not parse relay log event entry. The possible reasons are: the master 's binary log is corrupted (you can check this by running ' mysqlbinlog ' on the binary log), the slave' s relay log is corrupted (you can check this by running 'mysqlbinlog' on the relay log), a network problem, the server was unable to fetch a keyring key required to open an encrypted relay log file, or a bug in the master 's or slave' s MySQL code. If you want to check the master 's binary log or slave' s relay log, you will be able to know their names by issuing 'SHOW SLAVE STATUS' on this slave. Skip_Counter: 0

复制关系依旧报错。

第六步：重新reset slave all看看，结果成功了。

1 2 3 4 5 6 localhost.(none)>stop slave; Query OK, 0 rows affected (0.00 sec) localhost.(none)>reset slave all ; Query OK, 0 rows affected (0.03 sec)

第七步：重新搭建复制关系并启动复制

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 localhost.(none)>change master to master_host= '10.xx.xx.xx' , -> master_user= 'replica' , -> master_password= 'xxxxx' , -> master_port=5511, -> master_auto_position=1; Query OK, 0 rows affected, 2 warnings (0.01 sec) localhost.(none)>start slave; Query OK, 0 rows affected (0.00 sec) localhost.(none)>show slave status\G *************************** 1. row *************************** Slave_IO_State: Waiting for master to send event Master_Host: 10.xx.xx.xx Master_User: replica Master_Port: 5511 Connect_Retry: 60 ... Slave_IO_Running: Yes Slave_SQL_Running: Yes