[BlueOnyx:25634] MariaDB 10.9.3 and Active Monitor

[Aaron Greenspan]

Hi there,

I just upgraded MariaDB to release 10.9.3 (not an easy process since yum update got upset, no matter which 10.x version I tried, about the 10.2.13 RPM already being installed). After running rpm -e on the old MariaDB-server RPM, I was finally able to install it cleanly and managed not to delete all of my data in the process. Now, MariaDB 10.9.3 starts just fine, but every 15 minutes it spontaneously shuts down and then restarts. This is from the MariaDB log file:

2022-10-09 23:30:01 0 [Note] /usr/sbin/mariadbd (initiated by: unknown): Normal shutdown
2022-10-09 23:30:01 0 [Note] InnoDB: FTS optimize thread exiting.
2022-10-09 23:30:02 0 [Note] InnoDB: Starting shutdown...
2022-10-09 23:30:02 0 [Note] InnoDB: Dumping buffer pool(s) to /var/lib/mysql/ib_buffer_pool
2022-10-09 23:30:02 0 [Note] InnoDB: Restricted to 194704 pages due to innodb_buf_pool_dump_pct=25
2022-10-09 23:30:02 0 [Note] InnoDB: Buffer pool(s) dump completed at 221009 23:30:02
2022-10-09 23:30:05 0 [Note] InnoDB: Removed temporary tablespace data file: "./ibtmp1"
2022-10-09 23:30:05 0 [Note] InnoDB: Shutdown completed; log sequence number 659909755853; transaction id 2517047825
2022-10-09 23:30:05 0 [Note] /usr/sbin/mariadbd: Shutdown complete

It’s happening every 15 minutes. Naturally, this sounds like an Active Monitor type thing, so I manually ran /usr/sausalito/sbin/swatch.sh and sure enough, it shut down MariaDB:

2022-10-09 23:31:16 0 [Note] InnoDB: Buffer pool(s) load completed at 221009 23:31:16
2022-10-09 23:44:07 0 [Note] /usr/sbin/mariadbd (initiated by: unknown): Normal shutdown
2022-10-09 23:44:07 0 [Note] InnoDB: FTS optimize thread exiting.
2022-10-09 23:44:07 0 [Note] InnoDB: Starting shutdown...
2022-10-09 23:44:07 0 [Note] InnoDB: Dumping buffer pool(s) to /var/lib/mysql/ib_buffer_pool
2022-10-09 23:44:07 0 [Note] InnoDB: Restricted to 194704 pages due to innodb_buf_pool_dump_pct=25
2022-10-09 23:44:07 0 [Note] InnoDB: Buffer pool(s) dump completed at 221009 23:44:07
2022-10-09 23:44:09 0 [Note] InnoDB: Removed temporary tablespace data file: "./ibtmp1"
2022-10-09 23:44:09 0 [Note] InnoDB: Shutdown completed; log sequence number 659909912343; transaction id 2517049473
2022-10-09 23:44:10 0 [Note] /usr/sbin/mariadbd: Shutdown complete

Anyone have any idea why this might be happening? Bonus points if you know why 10.9.3 doesn’t like using /home/tmp as a temporary folder (it says it can’t write to it, even after chmod 1777 /home/tmp), which 10.2.13 was fine with.

Thanks,

Aaron

 <http://www.thinkcomputer.com/>	
Aaron Greenspan
President & CEO
Think Computer Corporation

telephone +1 415 670 9350
fax +1 415 373 3959
e-mail aarong at thinkcomputer.com
web http://www.thinkcomputer.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.blueonyx.it/pipermail/blueonyx/attachments/20221009/a397096e/attachment.html>