从库某个表损坏导致复制中断(mariadb10.2)

一 从库中断信息

               Slave_IO_State: Waiting for master to send event 故障表名已经替换
                  Master_Host: 10.10.10.3
                  Master_User: rep
                  Master_Port: 3306
                Connect_Retry: 10
              Master_Log_File: mysql-bin.017486
          Read_Master_Log_Pos: 337817815
               Relay_Log_File: relay-bin.020353
                Relay_Log_Pos: 63272569
        Relay_Master_Log_File: mysql-bin.017486
             Slave_IO_Running: Yes
            Slave_SQL_Running: No
              Replicate_Do_DB: 
          Replicate_Ignore_DB: 
           Replicate_Do_Table: 
       Replicate_Ignore_Table: 
      Replicate_Wild_Do_Table: 
  Replicate_Wild_Ignore_Table: 
                   Last_Errno: 1034
                   Last_Error: Could not execute Write_rows_v1 event on table school.student; 
                               Table school/student in tablespace 5065 corrupted., Error_code: 126; 
                               Index for table 'student' is corrupt; 
                               try to repair it, Error_code: 1034; handler error HA_ERR_CRASHED; the event's master log mysql-bin.017486, end_log_pos 112187504
                 Skip_Counter: 0
          Exec_Master_Log_Pos: 112185362
              Relay_Log_Space: 2549518298
              Until_Condition: None
               Until_Log_File: 
                Until_Log_Pos: 0
        Seconds_Behind_Master: NULL
Master_SSL_Verify_Server_Cert: No
                Last_IO_Errno: 0
                Last_IO_Error: 
               Last_SQL_Errno: 1034
               Last_SQL_Error: Could not execute Write_rows_v1 event on table school.student; 
                               Table school/student in tablespace 5065 corrupted., Error_code: 126; 
                               Index for table 'student' is corrupt; 
                               try to repair it, Error_code: 1034; handler error HA_ERR_CRASHED; the event's master log mysql-bin.017486, end_log_pos 112187504
  Replicate_Ignore_Server_Ids: 
             Master_Server_Id: 2541816
                   Using_Gtid: Slave_Pos
                  Gtid_IO_Pos: 0-2541816-46995033790
      Replicate_Do_Domain_Ids: 
  Replicate_Ignore_Domain_Ids: 
                Parallel_Mode: conservative
                    SQL_Delay: 0
          SQL_Remaining_Delay: NULL
      Slave_SQL_Running_State:

二 修复过程

1.非常重要,从库先停止复制(虽然sql进程已停止),设置过滤复制表 student

https://mariadb.com/kb/en/replication-and-binary-log-system-variables/#replicate_ignore_table

stop slave; 
set global REPLICATE_IGNORE_TABLE='school.student';

动态调整检查

SELECT VARIABLE_NAME, SESSION_VALUE, GLOBAL_VALUE FROM  INFORMATION_SCHEMA.SYSTEM_VARIABLES WHERE    VARIABLE_NAME like '%ignore%';
+-----------------------------+---------------+-----------------+
| VARIABLE_NAME               | SESSION_VALUE | GLOBAL_VALUE    |
+-----------------------------+---------------+-----------------+
| IGNORE_BUILTIN_INNODB       | NULL          | OFF             |
| REPLICATE_IGNORE_TABLE      | NULL          | school.student |
| GTID_IGNORE_DUPLICATES      | NULL          | OFF             |
| REPLICATE_IGNORE_DB         | NULL          |                 |
| IGNORE_DB_DIRS              | NULL          |                 |
| REPLICATE_WILD_IGNORE_TABLE | NULL          |                 |
+-----------------------------+---------------+-----------------+
6 rows in set (0.00 sec)

 

2.去主库或者从库备份表,得到同步位置,并拷贝到故障从库

  --在从库备份表 /apps/svr/mariadb10/bin/mysqldump --opt --single-transaction --dump-slave=2 -S  /tmp/mysql3309.sock -udba -p***  school student  > /apps/tmp/student.sql
  在主库备份表  /apps/svr/mariadb10/bin/mysqldump --opt --single-transaction --master-data=2 -S  /tmp/mysql3309.sock -udba -p***  school student  > /apps/tmp/student.sql
  -- CHANGE MASTER TO MASTER_LOG_FILE='mysql-bin.000006', MASTER_LOG_POS=41947;

 

3.从库执行复制

START SLAVE UNTIL MASTER_LOG_FILE = 'mysql-bin.000006', MASTER_LOG_POS = 41947;

当sql 线程停止时检查停止位置是否正确

        Relay_Master_Log_File: mysql-bin.000006
          Exec_Master_Log_Pos: 41947
             Slave_IO_Running: Yes
            Slave_SQL_Running: No    
Replicate_Ignore_Table: school.student
              Until_Condition: Master
               Until_Log_File: mysql-bin.000006
                Until_Log_Pos: 41947    

从库导入备份数据(删除旧表,重建表导入数据)

set session sql_log_bin=0;          
source /apps/tmp/student.sql;

关闭过滤复制

stop slave;set global REPLICATE_IGNORE_TABLE='';
show slave status \G

 

4.检查是否正常同步

start slave;
show slave status \G

 

延伸:

动态设置

-- mysql复制过滤
stop slave sql_thread;
CHANGE REPLICATION FILTER REPLICATE_WILD_DO_TABLE = ('school_5.%','school_6.%','school_7.%','school_8.%');
--CHANGE REPLICATION FILTER REPLICATE_WILD_DO_TABLE = ();
--CHANGE REPLICATION FILTER REPLICATE_IGNORE_DB  = (school);
--CHANGE REPLICATION FILTER REPLICATE_IGNORE_TABLE =(); 过滤复制某个表
--CHANGE REPLICATION FILTER REPLICATE_WILD_IGNORE_TABLE = ('db1.new%', 'db2.new%'); 过滤复制一批表
CHANGE MASTER TO master_host='10.10.10.10',master_port=3309,master_user='repl',master_password='***' , MASTER_AUTO_POSITION=1;
start slave;

-- mariadb 复制过滤
stop slave;
set global REPLICATE_IGNORE_TABLE='school.student';  --具体某个表
-- set global replicate_wild_ignore_table = 'school.student%'; 某个格式的表,以 student 为表头
start slave;

 低版本只能配置文件设置,重启生效

#replicate-do-db= 
#replicate-ignore-db= 
#replicate-do-table= 
#replicate-ignore-table= 
#replicate-wild-do-table= 
#replicate-wild-ignore-table= 
#replicate-same-server-id 
#replicate-rewrite-db= 
#innodb_replication_delay=0