1

This morning MySQL crashed on my server, and after further investigation, it looks like systemd reloaded which caused MySQL to go down. I found these lines in the syslog at the moment everything went down:

May  5 06:34:15 s1 systemd[1]: Reloading.
May  5 06:34:15 s1 systemd[1]: Starting Message of the Day...
May  5 06:34:15 s1 systemd[1]: Starting Daily apt download activities...
May  5 06:34:15 s1 systemd[1]: Started Message of the Day.
May  5 06:34:15 s1 systemd[1]: Stopping MySQL Community Server...

Would I be correct in saying that systemd going down is what caused MySQL to crash? Is there explanation for why systemd reloaded -- no one with access to the server ran daemon-reload, nor does that line appear in "history".

Additionally, we have munin installed, and there does not appear do have been any cpu or memory issues at that time, although the disk may have had high i/o, although that might have been caused by mysql restarting and not the other way around.

Are there any logs or directions I can look as to why systemd and mysql restarted?

4
  • 1
    Anything you can share from the error log for the 60 minutes prior to 06:34 on May 5? May 6, 2020 at 16:17
  • @WilsonHauck - there were some (ordinary) postfix, ufw blocks, and cron jobs. I also found these, but not sure if they are related: ``` May 5 06:33:35 s1 systemd[1]: Starting Daily apt upgrade and clean activities... May 5 05:56:22 s1 systemd[1]: Reloading The Apache HTTP Server. May 5 05:56:23 s1 systemd[1]: Reloaded The Apache HTTP Server. May 5 06:09:00 s1 systemd[1]: Starting Clean php session files... May 5 06:09:01 s1 systemd[1]: Started Clean php session files.```
    – danielb
    May 6, 2020 at 21:58
  • The May 5 06:33:35 s1 systemd[1]: Starting Daily apt upgrade and clean activities... happened 40 seconds exactly before
    – danielb
    May 6, 2020 at 22:02
  • @WilsonHauck - I found this: askubuntu.com/questions/1037285/… THANK you for pointing me in the right direction...
    – danielb
    May 6, 2020 at 22:12

1 Answer 1

1

The issue was that apt.daily-upgrade was ran and momentarily reset mysql. I changed the timer settings so that it runs in the middle of the night (the default is 6am-7am but those were high load times for us as the server is in US-WEST and that is 9am on the east coast).

H/T @WilsonHauck

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .