求助：Database was not shutdown normally
上周把一个percona实例升级到了greatsql，3节点PRIMARY-SECONDARY-ARBITRATOR + VIP架构，PRIMARY经常崩溃数据库配置，8c 16g 900G NVME，故障时间点，无任何资源瓶颈，/var/log/message 也无oom日志。
报错关键信息：
2024-04-18T20:38:47.403090+08:00 1 Database was not shutdown normally!
2024-04-18T21:40:24.311136+08:00 1 Database was not shutdown normally!
2024-04-18T23:20:45.409189+08:00 1 Database was not shutdown normally!
2024-04-19T00:24:13.050357+08:00 1 Database was not shutdown normally!
2024-04-19T04:36:34.457112+08:00 1 Database was not shutdown normally!
2024-04-19T17:18:53.836532+08:00 1 Database was not shutdown normally!
2024-04-19T19:41:11.638360+08:00 1 Database was not shutdown normally!
2024-04-20T09:25:32.423128+08:00 1 Database was not shutdown normally!
2024-04-20T10:40:53.207304+08:00 1 Database was not shutdown normally!
2024-04-20T11:18:13.887999+08:00 1 Database was not shutdown normally!
2024-04-20T15:19:26.891691+08:00 1 Database was not shutdown normally!
2024-04-20T15:25:15.887283+08:00 1 Database was not shutdown normally!
2024-04-22T00:40:04.900400+08:00 1 Database was not shutdown normally!
在这个错误之前，均有以下报错
2024-04-21T16:39:45Z UTC - mysqld got signal 11 ;
Most likely, you have hit a bug, but this error can also be caused by malfunctioning hardware.
BuildID=94d74d5128a360d05942861d00eaf4c9a608a665
Build ID: 94d74d5128a360d05942861d00eaf4c9a608a665
Server Version: 8.0.32-24 GreatSQL (GPL), Release 24, Revision c9974813cae
Thread pointer: 0x7f5baf8c6000
Attempting backtrace. You can use the following information to find out
where mysqld died. If you see no messages after this, something went
terribly wrong...
stack_bottom = 7f5c72971b00 thread_stack 0x80000
/usr/sbin/mysqld(my_print_stacktrace(unsigned char const*, unsigned long)+0x2e)
/usr/sbin/mysqld(print_fatal_signal(int)+0x3af)
/usr/sbin/mysqld(handle_fatal_signal+0xc5)
/lib64/libpthread.so.0(+0xf630)
/usr/lib64/mysql/plugin/audit_log.so(+0x11704)
/usr/lib64/mysql/plugin/audit_log.so(+0x15a95)
/usr/sbin/mysqld()
/usr/sbin/mysqld(mysql_audit_notify(THD*, mysql_event_general_subclass_t, char const*, int, char const*, unsigned long)+0x468)
/usr/sbin/mysqld(dispatch_command(THD*, COM_DATA const*, enum_server_command)+0x822)
/usr/sbin/mysqld(do_command(THD*)+0x25a)
/usr/sbin/mysqld()
/usr/sbin/mysqld()
/lib64/libpthread.so.0(+0x7ea5)
/lib64/libc.so.6(clone+0x6d)
Trying to get some variables.
Some pointers may be invalid and cause the dump to abort.
Query (0): Connection ID (thread ID): 234992
Status: NOT_KILLED
Please help us make Percona Server better by reporting any
bugs at https://bugs.percona.com/
You may download the Percona Server operations manual by visiting
http://www.percona.com/software/percona-server/. You may find information
in the manual which will help you identify the cause of the crash.
看起来像是启用了audit（审计）插件触发了未知bug。
几个建议：
1. 尝试升级到最新的GreatSQL 8.0.32-25版本；
2. 尝试关闭audit插件；
另外，还请提供
1. 完整错误日志文件
2. my.cnf配置文件
3. 如有可能，请参考用户手册（https://greatsql.cn/docs/8032-25/user-manual/11-faq/5-faq-others.html#_14-%E4%B8%BA%E4%BB%80%E4%B9%88%E5%9C%A8greatsql%E4%B8%AD%E8%BF%90%E8%A1%8C%E4%B8%80%E4%BA%9Bsql%E5%90%8E%E6%95%B0%E6%8D%AE%E5%BA%93crash%E4%BA%86-%E8%AF%A5%E6%80%8E%E4%B9%88%E5%8A%9E），提供相应的coredump文件，方便我们进一步排查原因 yejr 发表于 2024-4-22 11:15
看起来像是启用了audit（审计）插件触发了未知bug。
几个建议：
我先关掉审计观察验证下，如果确实是审计日志原因，再考虑升级，如果不是这个原因，再提供dump文件
页:
[1]