• We value your experience with Plesk during 2024
    Plesk strives to perform even better in 2025. To help us improve further, please answer a few questions about your experience with Plesk Obsidian 2024.
    Please take this short survey:

    https://pt-research.typeform.com/to/AmZvSXkx
  • The Horde webmail has been deprecated. Its complete removal is scheduled for April 2025. For details and recommended actions, see the Feature and Deprecation Plan.
  • We’re working on enhancing the Monitoring feature in Plesk, and we could really use your expertise! If you’re open to sharing your experiences with server and website monitoring or providing feedback, we’d love to have a one-hour online meeting with you.

Question CPU Usage "Capped" at 25%

thealex

New Pleskian
We released tickets for sale on one of our sites today, we dont have Cgroups or any other "limiting plugins" on our plesk instance, we also had speedkit (which is a waste of time, basically was instantly used its 15GB bandwidth then was doing nothing.

We had about 300k page requests from cloudflare over the past 24 hours aswell.

Running on 16 Thread, 32GB Memory instance, but CPU would not go over 25%, plesk control loads quick as, but site was crashing all the time, is there some inbuilt rate limiter on each individual site? or some other thing causing a virtual limit of 25% usage.

Shows the same in Htop aswell.
 
No. What does the actual per thread breakdown look like? Is this a VPS, where the host may be throttling? Do you have a proc limit/PHP worker bottleneck?
 
You might not have the Plesk cgroups plugin, but you can still have cgroup limiting in place in your operating system, because that's a function of your OS, not of Plesk. What's the output of
# systemctl show httpd | grep CPU

I would not recommend to allow httpd to get 100% cpu share anyway, so maybe you could also just run
# systemctl set-property httpd CPUAccounting=true
# systemctl set-property httpd CPUQuota=90%
and see if this solves the issue.

The above assumes that your Apache runs as httpd. On Debian, Ubuntu it is probably named apache2.
 
what is your iowait, and load average? If all your tasks are waiting for I/O, there won't be much CPU time used.
 
It could also be that your Apache's ServerLimit (and other related parameters) are too low. Once the ServerLimit or max. number of threads/childs has been reached, it will no longer accept new connections until others are closed.

That might explain the 'limit' of 25% CPU usage.
 
It could also be that your Apache's ServerLimit (and other related parameters) are too low. Once the ServerLimit or max. number of threads/childs has been reached, it will no longer accept new connections until others are closed.
Doesn't apache log something in that case?
 
Back
Top