如何解决由于 hibernate envers 审计导致的 MySQL 中的死锁？
How to resolve deadlock in MySQL due to hibernate envers auditing?
虽然 运行 很少有并行事务，但大多数时候我会遇到死锁：
------------------------
LATEST DETECTED DEADLOCK
------------------------
2019-09-04 06:19:12 0x2b01917c7700
*** (1) TRANSACTION:
TRANSACTION 14470484, ACTIVE 0 sec starting index read
mysql tables in use 1, locked 1
LOCK WAIT 13 lock struct(s), heap size 1136, 7 row lock(s), undo log entries 4
MySQL thread id 69372, OS thread handle 47285779531520, query id 10366178979 172.31.19.11 master updating
update `VerificationActionLog_AUD` set `REVEND`=427956 where `id`=138136 and `REV`<> 427956 and `REVEND` is null
*** (1) WAITING FOR THIS LOCK TO BE GRANTED:
RECORD LOCKS space id 7307 page no 1108 n bits 128 index PRIMARY of table `TestDB`.`VerificationActionLog_AUD` trx id 14470484 lock_mode X waiting
Record lock, heap no 60 PHYSICAL RECORD: n_fields 27; compact format; info bits 0
...
*** (2) TRANSACTION:
TRANSACTION 14470485, ACTIVE 0 sec starting index read
mysql tables in use 1, locked 1
11 lock struct(s), heap size 1136, 5 row lock(s), undo log entries 4
MySQL thread id 69395, OS thread handle 47285735814912, query id 10366178981 172.31.19.11 master updating
update `VerificationActionLog_AUD` set `REVEND`=427957 where `id`=138137 and `REV`<> 427957 and `REVEND` is null
*** (2) HOLDS THE LOCK(S):
RECORD LOCKS space id 7307 page no 1108 n bits 128 index PRIMARY of table `TestDB`.`VerificationActionLog_AUD` trx id 14470485 lock_mode X locks rec but not gap
Record lock, heap no 60 PHYSICAL RECORD: n_fields 27; compact format; info bits 0
...
*** (2) WAITING FOR THIS LOCK TO BE GRANTED:
RECORD LOCKS space id 7307 page no 1108 n bits 128 index PRIMARY of table `TestDB`.`VerificationActionLog_AUD` trx id 14470485 lock_mode X waiting
Record lock, heap no 60 PHYSICAL RECORD: n_fields 27; compact format; info bits 0
...
*** WE ROLL BACK TRANSACTION (2)
我试图推断这些陈述在解释什么。据我了解，事务2锁定了TestDB.VerificationActionLog_AUD的主索引。同时事务 2 也在等待同一个锁。一个事务怎么可能持有并等待同一个锁？
我是不是从这些陈述中推断错了？我怎样才能继续解决这些僵局。死锁仅适用于由 envers 在幕后维护的 AUD 表，如何解决？
这是因为间隙锁而发生的。间隙锁是在索引记录之间的间隙上的锁，或者是在第一个或最后一个索引记录之前或之后的间隙上的锁
假设您有相邻的 id 、 1 和 2。当过程从 2 个不同的会话同时执行时，每个会话都在两个索引记录上放置一个间隙锁（id 值为 1 和2 - 也可能是 0 ,4,5 ，但为了简单起见，我们假设只有 2 个），并且它们中的每一个都必须等待另一个释放锁才能执行插入。
Gap locks in InnoDB are “purely inhibitive”, which means they only
stop other transactions from inserting to the gap. They do not prevent
different transactions from taking gap locks on the same gap. Thus, a
gap X-lock has the same effect as a gap S-lock *."
解决方案：
Gap locking can be disabled explicitly. This occurs if you change the
transaction isolation level to READ COMMITTED or enable the
innodb_locks_unsafe_for_binlog system variable (which is now
deprecated). Under these circumstances, gap locking is disabled for
searches and index scans and is used only for foreign-key constraint
checking and duplicate-key checking.
参考文献：
虽然 运行 很少有并行事务，但大多数时候我会遇到死锁：
------------------------
LATEST DETECTED DEADLOCK
------------------------
2019-09-04 06:19:12 0x2b01917c7700
*** (1) TRANSACTION:
TRANSACTION 14470484, ACTIVE 0 sec starting index read
mysql tables in use 1, locked 1
LOCK WAIT 13 lock struct(s), heap size 1136, 7 row lock(s), undo log entries 4
MySQL thread id 69372, OS thread handle 47285779531520, query id 10366178979 172.31.19.11 master updating
update `VerificationActionLog_AUD` set `REVEND`=427956 where `id`=138136 and `REV`<> 427956 and `REVEND` is null
*** (1) WAITING FOR THIS LOCK TO BE GRANTED:
RECORD LOCKS space id 7307 page no 1108 n bits 128 index PRIMARY of table `TestDB`.`VerificationActionLog_AUD` trx id 14470484 lock_mode X waiting
Record lock, heap no 60 PHYSICAL RECORD: n_fields 27; compact format; info bits 0
...
*** (2) TRANSACTION:
TRANSACTION 14470485, ACTIVE 0 sec starting index read
mysql tables in use 1, locked 1
11 lock struct(s), heap size 1136, 5 row lock(s), undo log entries 4
MySQL thread id 69395, OS thread handle 47285735814912, query id 10366178981 172.31.19.11 master updating
update `VerificationActionLog_AUD` set `REVEND`=427957 where `id`=138137 and `REV`<> 427957 and `REVEND` is null
*** (2) HOLDS THE LOCK(S):
RECORD LOCKS space id 7307 page no 1108 n bits 128 index PRIMARY of table `TestDB`.`VerificationActionLog_AUD` trx id 14470485 lock_mode X locks rec but not gap
Record lock, heap no 60 PHYSICAL RECORD: n_fields 27; compact format; info bits 0
...
*** (2) WAITING FOR THIS LOCK TO BE GRANTED:
RECORD LOCKS space id 7307 page no 1108 n bits 128 index PRIMARY of table `TestDB`.`VerificationActionLog_AUD` trx id 14470485 lock_mode X waiting
Record lock, heap no 60 PHYSICAL RECORD: n_fields 27; compact format; info bits 0
...
*** WE ROLL BACK TRANSACTION (2)
我试图推断这些陈述在解释什么。据我了解，事务2锁定了TestDB.VerificationActionLog_AUD的主索引。同时事务 2 也在等待同一个锁。一个事务怎么可能持有并等待同一个锁？
我是不是从这些陈述中推断错了？我怎样才能继续解决这些僵局。死锁仅适用于由 envers 在幕后维护的 AUD 表，如何解决？
这是因为间隙锁而发生的。间隙锁是在索引记录之间的间隙上的锁，或者是在第一个或最后一个索引记录之前或之后的间隙上的锁
假设您有相邻的 id 、 1 和 2。当过程从 2 个不同的会话同时执行时，每个会话都在两个索引记录上放置一个间隙锁（id 值为 1 和2 - 也可能是 0 ,4,5 ，但为了简单起见，我们假设只有 2 个），并且它们中的每一个都必须等待另一个释放锁才能执行插入。
Gap locks in InnoDB are “purely inhibitive”, which means they only stop other transactions from inserting to the gap. They do not prevent different transactions from taking gap locks on the same gap. Thus, a gap X-lock has the same effect as a gap S-lock *."
解决方案：
Gap locking can be disabled explicitly. This occurs if you change the transaction isolation level to READ COMMITTED or enable the innodb_locks_unsafe_for_binlog system variable (which is now deprecated). Under these circumstances, gap locking is disabled for searches and index scans and is used only for foreign-key constraint checking and duplicate-key checking.
参考文献：