小编给大家分享一下MySQL中sql_mode修改不生效怎么办,相信大部分人都还不怎么了解,因此分享这篇文章给大家参考一下,希望大家阅读完这篇文章后大有收获,下面让我们一起去了解一下吧!
基于业务敏感性的考虑,下面涉及的表、存储过程等均非真实数据,但并不影响排查过程。
(1)客户侧开发童鞋创建了一个存储过程,该存储过程没有严格遵守group by标准语法
session 1: mysql> delimiter // mysql> create procedure test_for_group_by() -> begin -> select k,pad,count(*) from test.test group by k; -> end // Query OK, 0 rows affected (0.01 sec) mysql> delimiter ;
(2)客户侧开发童鞋调用该存储过程,报错ERROR 1140;因为当时存储过程比较复杂,改造起来比较麻烦,所以客户侧选择修改sql_mode
session 1: mysql> call test_for_group_by(); ERROR 1140 (42000): In aggregated query without GROUP BY, expression #1 of SELECT list contains nonaggregated column 'test.test.k'; this is incompatible with sql_mode=only_full_group_by
(3)客户侧修改完sql_mode,再次执行,发现仍然报错ERROR 1140
session 2: mysql> set global sql_mode='STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION'; Query OK, 0 rows affected (0.00 sec) session 1: mysql> call test_for_group_by(); ERROR 1140 (42000): In aggregated query without GROUP BY, expression #1 of SELECT list contains nonaggregated column 'test.test.k'; this is incompatible with sql_mode=only_full_group_by
(4)此时想到,修改系统变量,只对新建连接有效,对已有连接不起作用;于是,让客户侧重新建立连接,确认系统变量已生效,再次调用存储过程,但仍然报错ERROR 1140,重复尝试几次都是这个结果
session 3: mysql> show variables like 'sql_mode'; +---------------+------------------------------------------------------------------------------------------------------------------------+ | Variable_name | Value | +---------------+------------------------------------------------------------------------------------------------------------------------+ | sql_mode | STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION | +---------------+------------------------------------------------------------------------------------------------------------------------+ 1 row in set (0.01 sec) mysql> call test_for_group_by(); ERROR 1140 (42000): In aggregated query without GROUP BY, expression #1 of SELECT list contains nonaggregated column 'test.test.k'; this is incompatible with sql_mode=only_full_group_by
(5)进一步排查,让客户侧在该会话,执行非标准的group by语句,发现可以正常执行
session 3: mysql> select user,host,count(*) From mysql.user group by user; +---------------+-----------+----------+ | user | host | count(*) | +---------------+-----------+----------+ | mysql.session | localhost | 1 | | mysql.sys | localhost | 1 | | root | localhost | 1 | | rpl_user | % | 1 | | test | % | 1 | +---------------+-----------+----------+ 5 rows in set (0.00 sec)
(6)继续排查发现,该存储过程的sql_mode,还是包括ONLY_FULL_GROUP_BY,因此执行报错
session 2: mysql> select routine_catalog,routine_schema,routine_name,routine_type,created,last_altered,sql_mode from routines where routine_name='test_for_group_by'; +-----------------+----------------+-------------------+--------------+---------------------+---------------------+-------------------------------------------------------------------------------------------------------------------------------------------+ | routine_catalog | routine_schema | routine_name | routine_type | created | last_altered | sql_mode | +-----------------+----------------+-------------------+--------------+---------------------+---------------------+-------------------------------------------------------------------------------------------------------------------------------------------+ | def | test | test_for_group_by | PROCEDURE | 2020-12-24 12:12:10 | 2020-12-24 12:12:10 | ONLY_FULL_GROUP_BY,STRICT_TRANS_TABLES,NO_ZERO_IN_DATE,NO_ZERO_DATE,ERROR_FOR_DIVISION_BY_ZERO,NO_AUTO_CREATE_USER,NO_ENGINE_SUBSTITUTION | +-----------------+----------------+-------------------+--------------+---------------------+---------------------+-------------------------------------------------------------------------------------------------------------------------------------------+ 1 row in set (0.00 sec)
(7)这里我们也可以知道,系统变量修改只对新建对象有效,对已有对象不生效;解决办法很简单,重建该存储过程即可
session 3: mysql> drop procedure test_for_group_by; Query OK, 0 rows affected (0.01 sec) mysql> delimiter // mysql> create procedure test_for_group_by() -> begin -> select k,pad,count(*) from test.test group by k; -> end // Query OK, 0 rows affected (0.01 sec) mysql> delimiter ; mysql> call test_for_group_by(); +--------+-------------------------------------------------------------+----------+ | k | pad | count(*) | +--------+-------------------------------------------------------------+----------+ | 393975 | 35227182905-15234265621-59793845249-15413569710-23749555118 | 1 | | 495688 | 09512147864-77936258834-40901700703-13541171421-15205431759 | 1 | | 497896 | 13152283289-69561545685-52868757241-04245213425-69280254356 | 1 | | 498573 | 43131080328-59298106536-35954612339-97546855884-75769514803 | 1 | | 500775 | 27590239742-20204899609-34345212327-79811525340-24267764271 | 1 | | 501885 | 63188288836-92351140030-06390587585-66802097351-49282961843 | 1 | | 503330 | 01495266405-82925129145-92643983850-90243995398-18709399387 | 1 | | 503666 | 40929980986-33813039690-13155419391-97985458477-39771362212 | 1 | | 504353 | 00505722282-72931248925-57037623248-81117963809-88658076981 | 1 | | 514246 | 21979564480-87492594656-60524686334-78820761788-57684966682 | 1 | +--------+-------------------------------------------------------------+----------+ 10 rows in set (0.00 sec) Query OK, 0 rows affected (0.00 sec)
通过这个案例,我们可以知道,修改sql_mode系统变量,只对新建连接和新建对象(主要包括函数和存储过程)有效,对已有连接和已有对象不生效。
以上是“MySQL中sql_mode修改不生效怎么办”这篇文章的所有内容,感谢各位的阅读!相信大家都有了一定的了解,希望分享的内容对大家有所帮助,如果还想学习更多知识,欢迎关注亿速云行业资讯频道!
免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:is@yisu.com进行举报,并提供相关证据,一经查实,将立刻删除涉嫌侵权内容。