MySQL从库 Last_SQL_Errno: 1197 问题处理过程

记录一个遇到过的错误，今天整理一下。

问题

MySQL error code MY-001197 (ER_TRANS_CACHE_FULL): Multi-statement transaction required morethan 'max_binlog_cache_size' bytes of storage; increase this mysqld variable and try again

报错很明显是max_binlog_cache_size值太小了. 解决办法也很简单, 增加就好了

#实际值根据自己业务情况来, 基本上这个值就是限制事务大小
set global max_binlog_cache_size=10*1024*1024*1024;

然后业务验证, 没啥问题，看起来一切都顺利，但突然收到告警, 说从库断开了…

*************************** 1. row ***************************Slave_IO_State: Waiting for source to send eventMaster_Host: 192.168.101.21Master_User: replMaster_Port: 3314Connect_Retry: 60Master_Log_File: m3314.000001Read_Master_Log_Pos: 128860Relay_Log_File: relay.000003Relay_Log_Pos: 362Relay_Master_Log_File: m3314.000001Slave_IO_Running: YesSlave_SQL_Running: NoReplicate_Do_DB: Replicate_Ignore_DB: Replicate_Do_Table: Replicate_Ignore_Table: Replicate_Wild_Do_Table: Replicate_Wild_Ignore_Table: Last_Errno: 1197Last_Error: Coordinator stopped because there were error(s) in the worker(s). The most recent failure being: Worker 1 failed executing transaction 'b68e2434-cd30-11ec-b536-000c2980c11e:7' at source log m3314.000001, end_log_pos 128829. See error log and/or performance_schema.replication_applier_status_by_worker table for more details about this failure or others, if any.Skip_Counter: 0Exec_Master_Log_Pos: 87616Relay_Log_Space: 129473Until_Condition: NoneUntil_Log_File: Until_Log_Pos: 0Master_SSL_Allowed: NoMaster_SSL_CA_File: Master_SSL_CA_Path: Master_SSL_Cert: Master_SSL_Cipher: Master_SSL_Key: Seconds_Behind_Master: NULL
Master_SSL_Verify_Server_Cert: NoLast_IO_Errno: 0Last_IO_Error: Last_SQL_Errno: 1197Last_SQL_Error: Coordinator stopped because there were error(s) in the worker(s). The most recent failure being: Worker 1 failed executing transaction 'b68e2434-cd30-11ec-b536-000c2980c11e:7' at source log m3314.000001, end_log_pos 128829. See error log and/or performance_schema.replication_applier_status_by_worker table for more details about this failure or others, if any.Replicate_Ignore_Server_Ids: Master_Server_Id: 866003314Master_UUID: b68e2434-cd30-11ec-b536-000c2980c11eMaster_Info_File: mysql.slave_master_infoSQL_Delay: 0SQL_Remaining_Delay: NULLSlave_SQL_Running_State: Master_Retry_Count: 86400Master_Bind: Last_IO_Error_Timestamp: Last_SQL_Error_Timestamp: 250111 16:37:21Master_SSL_Crl: Master_SSL_Crlpath: Retrieved_Gtid_Set: b68e2434-cd30-11ec-b536-000c2980c11e:1-7Executed_Gtid_Set: b68e2434-cd30-11ec-b536-000c2980c11e:1-6Auto_Position: 0Replicate_Rewrite_DB: Channel_Name: Master_TLS_Version: Master_public_key_path: Get_master_public_key: 0Network_Namespace: 
1 row in set, 1 warning (0.00 sec)

报错原因：Worker 1 failed executing transaction 'b68e2434-cd30-11ec-b536-000c2980c11e:7' at source log m3314.000001, end_log_pos 128829; Could not execute Write_rows event on table d2025.t1; Multi-statement transaction required more than 'max_binlog_cache_size' bytes of storage; increase this mysqld variable and try again, Error_code: 1197; the event's source log m3314.000001, end_log_pos 128829

翻译了一下，也就是修改完这个参数之后, 会话得重连才行. 虽然不重连也能才看到修改后的值, 但是生效的还是之前的值

解决办法很简单, 重启SQL线程即可, 而SQL线程本身就挂了, 所以直接start slave即可解决

模拟复原

#主库初始化参数
set global max_binlog_cache_size=4096;
exit;#从库初始化参数
set global max_binlog_cache_size=4096;
exit;#主库测试报错
create database db2025;
create table db2025.t1(id int, name text);#会看到报错 ERROR 1197 (HY000)
insert into db2025.t1 values(1,repeat('x',4096)); #主库修改参数后再次测试
set global max_binlog_cache_size=4096000;
show variables like 'max_binlog_cache_size';#还会报错,因为没有重新连接,虽然能查询到新的值了
insert into db2025.t1 values(1,repeat('x',4096));  
exit;#主库重连后插入数据
show variables like 'max_binlog_cache_size';#这次应该就成功了
insert into db2025.t1 values(1,repeat('x',4096)); #从库修改参数(生产环境的时候,数据还没有同步过来, 所以从库看到的应该是正常的,但也得修改参数, 由于本次测试数据量小, 此时已经能看到参数报错了)
set global max_binlog_cache_size=4096000;
exit;#查看主从状态
show slave status\G  #能看到从库报错Last_Errno: 1197#从库启动主从
start slave; -- 直接启动就行. 从库回放应该比主库快好几倍(一样配置的情况下.)