时间:2021-07-01 10:21:17 帮助过:34人阅读
*** (2) TRANSACTION:
TRANSACTION 2067C43F, ACTIVE 7 sec fetching rows, thread declared inside InnoDB 424
mysql tables in use 1, locked 1
2223 lock struct(s), heap size 440760, 192 row lock(s), undo log entries 8
MySQL thread id 20135, OS thread handle 0xf28, query id 276602976 localhost 127.0.0.1 root updating
delete from a93 where patientId=164136 and count=9
*** (2) HOLDS THE LOCK(S):
RECORD LOCKS space id 4609 page no 4 n bits 136 index `PRIMARY` of table `emr_new2`.`a93` trx id 2067C43F lock_mode X locks rec but not gap
Record lock, heap no 4 PHYSICAL RECORD: n_fields 71; compact format; info bits 0
可以看到两个事务 TRANSACTION 2067C318 和 TRANSACTION 2067C43F 分别执行: delete from a93 wehre patientId=164136 and count=9
导致了相互等待。
查看表结构和数据量,发现 patientId未加索引,数据量很大,导致delete语句会全表扫描,容易长时间持有锁。
加上索引: alter table a93 add index patientId(patientId);
MySQL死锁分析一例
标签:show row 事务 fetching com tran 相互 highlight inf