• The APS Catalog has been deprecated and removed from all Plesk Obsidian versions.
    Applications already installed from the APS Catalog will continue working. However, Plesk will no longer provide support for APS applications.
  • Please be aware: with the Plesk Obsidian 18.0.78 release, the support for the ngx_pagespeed.so module will be deprecated and removed from the sw-nginx package.

Issue init at 100% CPU and pleskrc very slow to reload processes

For such a specific question it is best to address it to Plesk support. They can analyze, research and respond in detail.
 
Found the source of this issue and a solid workaround. Apparently Virtuozzo/OpenVZ symlinks /sbin/init to systemd within the container, but then on the node when vzctl launches things it passes init parameters in like: init -z, which results in systemd -z.

For reasons unknown to me, the -z seriously messes with systemd particularly after systemctl reexec-daemon is run, which is exactly what Plesk does when it runs system updates.

There is a patch available for systemd to prevent the unexpected switch from messing with systemd upon reexec, but RedHat hasn't yet backported it. Virtuozzo devs haven't yet taken note, but the original reporter of the issue on the systemd project on Github devised a solid workaround to replace the /sbin/init symlink with a quick script to strip the -z from the parameters that vzctl feeds into init, resulting in systemd never having the -z paramter passed into it.

Full details here: systemctl daemon-reexec forgets running services and tries to restart all services · Issue #28184 · systemd/systemd

Note that the container *must* be rebooted after this workaround is applied.

I also submit a pull request for Virtuozzo to fix libvzctl by removing the -z paramter when systemd is used here, however I doubt they'll look at that. I have also updated my OpenVZ bug report on this issue with this information.
 
Back
Top