很多人都分不清死锁和锁等待的区别,也有不同it口的人叫法的差异。在运维侧:
死锁最明显的特征是会自动解开,是需要我们去事后解决逻辑缺陷。
锁等待则是业务卡住了(一般是某个大事务还在执行,或有事务没提交),需要杀掉持有锁的进城让业务正常进行
做几个实验详细演示一下。
表结构及数据情况
mysql> desc ttt; +-------+-------------+------+-----+---------+-------+ | field | type | null | key | default | extra | +-------+-------------+------+-----+---------+-------+ | x | int(11) | no | pri | null | | | y | datetime | yes | | null | | | z | varchar(10) | yes | | null | | +-------+-------------+------+-----+---------+-------+ 3 rows in set (0.00 sec) mysql> select * from ttt; +---+---------------------+------+ | x | y | z | +---+---------------------+------+ | 1 | 2023-05-30 23:50:13 | 123 | | 2 | 2023-04-26 17:58:18 | av3 | | 3 | 2023-05-30 22:52:35 | at | | 4 | 2023-04-26 17:58:29 | attt | | 5 | 2023-05-30 22:52:55 | zxz | +---+---------------------+------+ 5 rows in set (0.00 sec)
死锁(dead lock)
上文说了,死锁会自动解除,这里主要展示一下怎么追查。这里就不演示具体执行顺序了
前台报错
session1,这个先持有
mysql> begin; query ok, 0 rows affected (0.00 sec) mysql> update ttt set y=now() where x=1; query ok, 1 row affected (0.00 sec) rows matched: 1 changed: 1 warnings: 0 mysql> update ttt set y=now() where x=3; query ok, 1 row affected (7.04 sec) rows matched: 1 changed: 1 warnings: 0 mysql>
session2,这个来跳出错误
mysql> begin; query ok, 0 rows affected (0.00 sec) mysql> update ttt set y=now() where x=3; query ok, 1 row affected (0.00 sec) rows matched: 1 changed: 1 warnings: 0 mysql> update ttt set y=now() where x=1; error 1213 (40001): deadlock found when trying to get lock; try restarting transaction mysql>
事后追查
innodb status的last dead lock只会记录上一次,建议是使用innodb_print_all_deadlocks参数,将记录打印到error.log中,该参数默认为off,可以动态修改
mysql> show variables like '%innodb_print_all_deadlocks%'; +----------------------------+-------+ | variable_name | value | +----------------------------+-------+ | innodb_print_all_deadlocks | off | +----------------------------+-------+ 1 row in set (0.00 sec) mysql> set global innodb_print_all_deadlocks=on; query ok, 0 rows affected (0.00 sec) mysql> show variables like '%innodb_print_all_deadlocks%'; +----------------------------+-------+ | variable_name | value | +----------------------------+-------+ | innodb_print_all_deadlocks | on | +----------------------------+-------+ 1 row in set (0.00 sec)
innodb status
show engine innodb status; ********* ------------------------ latest detected deadlock ------------------------ 2023-05-30 22:42:09 0x7f9fd41ba700 *** (1) transaction: transaction 5454, active 26 sec starting index read mysql tables in use 1, locked 1 lock wait 3 lock struct(s), heap size 1136, 2 row lock(s), undo log entries 1 mysql thread id 18, os thread handle 140324434298624, query id 440 localhost root updating update ttt set y=now() where x=3 *** (1) waiting for this lock to be granted: record locks space id 23 page no 3 n bits 80 index primary of table `ddd`.`ttt` trx id 5454 lock_mode x locks rec but not gap waiting record lock, heap no 10 physical record: n_fields 5; compact format; info bits 0 0: len 4; hex 80000003; asc ;; 1: len 6; hex 00000000154f; asc o;; 2: len 7; hex 390000014e0110; asc 9 n ;; 3: len 5; hex 99b03d6a76; asc =jv;; 4: len 2; hex 6174; asc at;; *** (2) transaction: transaction 5455, active 15 sec starting index read mysql tables in use 1, locked 1 3 lock struct(s), heap size 1136, 2 row lock(s), undo log entries 1 mysql thread id 20, os thread handle 140324435109632, query id 441 localhost root updating update ttt set y=now() where x=1 *** (2) holds the lock(s): record locks space id 23 page no 3 n bits 80 index primary of table `ddd`.`ttt` trx id 5455 lock_mode x locks rec but not gap record lock, heap no 10 physical record: n_fields 5; compact format; info bits 0 0: len 4; hex 80000003; asc ;; 1: len 6; hex 00000000154f; asc o;; 2: len 7; hex 390000014e0110; asc 9 n ;; 3: len 5; hex 99b03d6a76; asc =jv;; 4: len 2; hex 6174; asc at;; *** (2) waiting for this lock to be granted: record locks space id 23 page no 3 n bits 80 index primary of table `ddd`.`ttt` trx id 5455 lock_mode x locks rec but not gap waiting record lock, heap no 7 physical record: n_fields 5; compact format; info bits 0 0: len 4; hex 80000001; asc ;; 1: len 6; hex 00000000154e; asc n;; 2: len 7; hex 380000014d0110; asc 8 m ;; 3: len 5; hex 99b03d6a6b; asc =jk;; 4: len 3; hex 313233; asc 123;; *** we roll back transaction (2) ********
error.log
[root@mysql57-1 data]# tail -100f mysql57-1.err 2023-05-25t07:42:37.677851z 14 [note] access denied for user 'root'@'localhost' (using password: yes) 2023-05-25t07:44:15.360390z 15 [note] access denied for user 'roo'@'localhost' (using password: yes) 2023-05-25t07:44:28.477560z 16 [note] access denied for user 'roo'@'localhost' (using password: yes) 2023-05-30t14:53:00.101403z 22 [note] innodb: transactions deadlock detected, dumping detailed information. 2023-05-30t14:53:00.101475z 22 [note] innodb: *** (1) transaction: transaction 5458, active 25 sec starting index read mysql tables in use 1, locked 1 lock wait 3 lock struct(s), heap size 1136, 2 row lock(s), undo log entries 1 mysql thread id 21, os thread handle 140324434298624, query id 473 localhost root updating update ttt set y=now() where x=5 2023-05-30t14:53:00.101516z 22 [note] innodb: *** (1) waiting for this lock to be granted: record locks space id 23 page no 3 n bits 80 index primary of table `ddd`.`ttt` trx id 5458 lock_mode x locks rec but not gap waiting record lock, heap no 6 physical record: n_fields 5; compact format; info bits 0 0: len 4; hex 80000005; asc ;; 1: len 6; hex 000000001553; asc s;; 2: len 7; hex 3b00000130036d; asc ; 0 m;; 3: len 5; hex 99b03d6d32; asc =m2;; 4: len 3; hex 7a787a; asc zxz;; 2023-05-30t14:53:00.101718z 22 [note] innodb: *** (2) transaction: transaction 5459, active 10 sec starting index read mysql tables in use 1, locked 1 3 lock struct(s), heap size 1136, 2 row lock(s), undo log entries 1 mysql thread id 22, os thread handle 140324435109632, query id 474 localhost root updating update ttt set y=now() where x=3 2023-05-30t14:53:00.101748z 22 [note] innodb: *** (2) holds the lock(s): record locks space id 23 page no 3 n bits 80 index primary of table `ddd`.`ttt` trx id 5459 lock_mode x locks rec but not gap record lock, heap no 6 physical record: n_fields 5; compact format; info bits 0 0: len 4; hex 80000005; asc ;; 1: len 6; hex 000000001553; asc s;; 2: len 7; hex 3b00000130036d; asc ; 0 m;; 3: len 5; hex 99b03d6d32; asc =m2;; 4: len 3; hex 7a787a; asc zxz;; 2023-05-30t14:53:00.101912z 22 [note] innodb: *** (2) waiting for this lock to be granted: record locks space id 23 page no 3 n bits 80 index primary of table `ddd`.`ttt` trx id 5459 lock_mode x locks rec but not gap waiting record lock, heap no 10 physical record: n_fields 5; compact format; info bits 0 0: len 4; hex 80000003; asc ;; 1: len 6; hex 000000001552; asc r;; 2: len 7; hex 3a0000012e03d1; asc : . ;; 3: len 5; hex 99b03d6d23; asc =m#;; 4: len 2; hex 6174; asc at;; 2023-05-30t14:53:00.102084z 22 [note] innodb: *** we roll back transaction (2)
锁等待(lock wait)
1个参数
innodb_lock_wait_timeout
行锁等待的时间,如果超过这个时间,session2(后发起那个)会自动跳出
session1
mysql> begin; query ok, 0 rows affected (0.00 sec) mysql> update ttt set y=now() where x=1; query ok, 1 row affected (0.00 sec) rows matched: 1 changed: 1 warnings: 0
session2
mysql> update ttt set y=now() where x=1; error 1205 (hy000): lock wait timeout exceeded; try restarting transaction
怎么处理
省略构造锁
通过information_schema.innodb_lock_waits视图获得锁等待的关系
mysql> select * from information_schema.innodb_lock_waits; +-------------------+-------------------+-----------------+------------------+ | requesting_trx_id | requested_lock_id | blocking_trx_id | blocking_lock_id | +-------------------+-------------------+-----------------+------------------+ | 5467 | 5467:23:3:7 | 5466 | 5466:23:3:7 | +-------------------+-------------------+-----------------+------------------+ 1 row in set, 1 warning (0.00 sec)
再结合information_schema.innodb_trx视图得到语句和线程id
这里提供一个sql
select r.trx_id waiting_trx_id, r.trx_mysql_thread_id waiting_thread, r.trx_query waiting_query, b.trx_id blocking_trx_id, b.trx_mysql_thread_id blocking_thread, b.trx_query blocking_query, b.trx_autocommit_non_locking from information_schema.innodb_lock_waits w inner join information_schema.innodb_trx b on b.trx_id = w.blocking_trx_id inner join information_schema.innodb_trx r on r.trx_id = w.requesting_trx_id; mysql> select r.trx_id waiting_trx_id, -> r.trx_mysql_thread_id waiting_thread, -> r.trx_query waiting_query, -> b.trx_id blocking_trx_id, -> b.trx_mysql_thread_id blocking_thread, -> b.trx_query blocking_query, -> b.trx_autocommit_non_locking -> from information_schema.innodb_lock_waits w -> inner join information_schema.innodb_trx b on -> b.trx_id = w.blocking_trx_id -> inner join information_schema.innodb_trx r on -> r.trx_id = w.requesting_trx_id; +----------------+----------------+----------------------------------+-----------------+-----------------+----------------+----------------------------+ | waiting_trx_id | waiting_thread | waiting_query | blocking_trx_id | blocking_thread | blocking_query | trx_autocommit_non_locking | +----------------+----------------+----------------------------------+-----------------+-----------------+----------------+----------------------------+ | 5470 | 22 | update ttt set y=now() where x=1 | 5466 | 26 | null | 0 | +----------------+----------------+----------------------------------+-----------------+-----------------+----------------+----------------------------+ 1 row in set, 1 warning (0.00 sec)
再对blocking thread的状态进行确认,然后kill
mysql> select * from information_schema.processlist where id=26; +----+------+-----------+------+---------+------+-------+------+ | id | user | host | db | command | time | state | info | +----+------+-----------+------+---------+------+-------+------+ | 26 | root | localhost | ddd | sleep | 1221 | | null | +----+------+-----------+------+---------+------+-------+------+ 1 row in set (0.00 sec)
杀线程
mysql> kill 26; query ok, 0 rows affected (0.00 sec)
级联锁或大量锁
这里构造一个多个争用的情况。
session1,线程id为27,这个先发起
mysql> begin; query ok, 0 rows affected (0.00 sec) mysql> update ttt set y=now() where x=1; query ok, 1 row affected (0.00 sec) rows matched: 1 changed: 1 warnings: 0
session,线程id为22,第二个发起
mysql> update ttt set y=now() where x=1;
session3,线程id为28,最后一个发起
mysql> update ttt set y=now();
锁关系情况
mysql> select * from information_schema.innodb_lock_waits; +-------------------+-------------------+-----------------+------------------+ | requesting_trx_id | requested_lock_id | blocking_trx_id | blocking_lock_id | +-------------------+-------------------+-----------------+------------------+ | 5479 | 5479:23:3:7 | 5478 | 5478:23:3:7 | | 5479 | 5479:23:3:7 | 5473 | 5473:23:3:7 | | 5478 | 5478:23:3:7 | 5473 | 5473:23:3:7 | +-------------------+-------------------+-----------------+------------------+ 3 rows in set, 1 warning (0.01 sec)
查询sql的展示情况
mysql> select r.trx_id waiting_trx_id, r.trx_mysql_thread_id waiting_thread, r.trx_query waiting_query, b.trx_id blocking_trx_id, b.trx_mysql_thread_id blocking_thread, b.trx_query blocking_query, b.trx_autocommit_non_locking from information_schema.innodb_lock_waits w inner join information_schema.innodb_trx b on b.trx_id = w.blocking_trx_id inner join information_schema.innodb_trx r on r.trx_id = w.requesting_trx_id; +----------------+----------------+----------------------------------+-----------------+-----------------+----------------------------------+----------------------------+ | waiting_trx_id | waiting_thread | waiting_query | blocking_trx_id | blocking_thread | blocking_query | trx_autocommit_non_locking | +----------------+----------------+----------------------------------+-----------------+-----------------+----------------------------------+----------------------------+ | 5479 | 28 | update ttt set y=now() | 5478 | 22 | update ttt set y=now() where x=1 | 0 | | 5479 | 28 | update ttt set y=now() | 5473 | 27 | null | 0 | | 5478 | 22 | update ttt set y=now() where x=1 | 5473 | 27 | null | 0 | +----------------+----------------+----------------------------------+-----------------+-----------------+----------------------------------+----------------------------+
这个时候一般的做法是先找到最多的那个blocking_thread
确认他的进程状态后来考虑是否杀
如果他是活动的thread,在干活,就可以删掉释放资源。
如果他是非活动,就要看他这个thread的blocking_thread,找到活动持有资源的,来杀掉。
当然也有执行了完了不提交的情况,innodb_trx的trx_autocommit_non_locking列的值就是为了标记是否是提交了。0表示没有提交。
到此这篇关于mysql死锁(dead lock)与锁等待(lock wait)的出现解决的文章就介绍到这了,更多相关mysql死锁与锁等待内容请搜索代码网以前的文章或继续浏览下面的相关文章希望大家以后多多支持代码网!
发表评论