MySQL Forums
Forum List  »  Quality Assurance

Database can not be shut down properly
Posted by: Masahito Kubota
Date: February 20, 2022 09:51PM

Hello.

In our system we are using MariaDB (Mysql) as the database for Zabbix server.
The Zabbix server is in a cluster, with two servers configured as Active/Standby.
The other day, I restarted the Corosync service on the active server using systemctl command to change the Corosync configuration.
When the Corosync service was restarted, each software cluster resource was automatically restarted and the Zabbix service was restored without any problem.
We assumed that when the Corosync service was restarted, the software that serves as each cluster resource would be restarted automatically and the Zabbix service would be restored without any problems.
In the development environment, the result was as expected, but in the production environment, an error occurred in the database and the
The database could not be shut down properly and a crash recovery occurred.
When the failure occurred, the following log was output on the Active server.
The same log was also output on the Standby server.

211122 12:19:27 [Note] /usr/libexec/mysqld: Normal shutdown
211122 12:19:27 [Note] Event Scheduler: Purging the queue. 0 events
211122 12:19:27 InnoDB: Starting shutdown...
211122 12:19:27 InnoDB: Waiting for XXX pages to be flushed
211122 12:20:07 InnoDB: The InnoDB memory heap is disabled
211122 12:20:07 InnoDB: Mutexes and rw_locks use GCC atomic builtins
211122 12:20:07 InnoDB: Compressed tables use zlib 1.2.7
211122 12:20:07 InnoDB: Using Linux native AIO
211122 12:20:07 InnoDB: Initializing buffer pool, size = X.XG
211122 12:20:08 InnoDB: Completed initialization of buffer pool
211122 12:20:08 InnoDB: highest supported file format is Barracuda.
211122 12:20:08 InnoDB: Database was not shut down normally!
211122 12:20:20 InnoDB: Starting an apply batch of log records to the database...
InnoDB: Progress in percents: 0 211122 12:25:00 InnoDB: The InnoDB memory heap is disabled

The versions of each software are as follows
Red Hat Enterprise Linux release 7.2 (Maipo)
mysql Ver 15.1 Distrib 5.5.52-MariaDB, for Linux (x86_64) using readline 5.1
Pacemaker 1.1.17-1.el7
Corosync Cluster Engine, version '2.4.2'.

From the above information, can you figure out what is causing the problem?
Please let me know if there are any other logs or settings I need to check.

Options: ReplyQuote


Subject
Views
Written By
Posted
Database can not be shut down properly
226
February 20, 2022 09:51PM


Sorry, only registered users may post in this forum.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.