Issue Plesk daily job overwhelming server / MySQL

[OlgaKM]

I am having an issue where the Plesk daily job (/usr/local/psa/admin/plib/DailyMaintainance/script.php) is overwhelming the server, particularly MySQL. This happens for between 5-10 minutes every day when the job is running. As far as I can see what is happening is that the job consumes so many resources that processes connect to MySQL, but are left hanging because they don't have the resources to run and this results in them taking up MySQL connections. Eventually, new processes start receiving a 'Too many connections' error. The max_connection limit for mysql is set to 512.

Is there a way to limit the resources used by the daily job?

 

[serverpoint]

Hello,

You can disable the daily maintenance script From the Tools & Settings menu, click on Scheduled Tasks under the Tools & Resources heading: On the Scheduled Tasks page, click on the desired task. Change the Run sequence.

Also refer the below article to rerun daily maintenance tasks separately for Plesk on Linux

https://support.plesk.com/hc/en-us/...intenance-tasks-separately-for-Plesk-on-Linux

Thanks,

 

[OlgaKM]

Thank you serverpoint, that link was very helpful in diagnosing the issue. Ultimately, the solution I used was to use cgroups to limit disk i/o of the process.

service cgconfig start
cgcreate -g blkio:/limit1M
cgset -r blkio.throttle.write_bps_device="8:0 1048576" limit1M
cgset -r blkio.throttle.read_bps_device="8:0 1048576" limit1M

Then, in the file /etc/cron.daily/50plesk-daily, I changed the following line:

Original line:

/usr/local/psa/bin/sw-engine-pleskrun /usr/local/psa/admin/plib/DailyMaintainance/script.php >/dev/null 2>&1

Modified line:

cgexec --sticky -g blkio:/limit1M /usr/local/psa/bin/sw-engine-pleskrun /usr/local/psa/admin/plib/DailyMaintainance/script.php >/dev/null 2>&1