Help with finding high resource pages
Hi,
Please move this post if it is in the wrong place. We have a few WordPress sites where i'm trying to improve website resource as I get High 5/10 minute load average alerts from CSF throughout the day. e.g
1 Min Load Avg: 22.43
5 Min Load Avg: 7.52
15 Min Load Avg: 2.80
I know I can stop these notifications; however, I'd like to know if I could find which page or pages are using more resources when clicked. Which I can use a query monitor to see which this may be. I have looked at AWStats but have been unable to tell which page(s) are using more.
Is there a way to find this out? I've inserted an image below of what the ps looks like.
Thanks
-
Hey there! The short answer is "probably not" - the default CSF/LFD settings are often a bit too small, in my experience, to reliably tell if a page is using too many resources. I would check the CSF settings to see what that threshold actually is first before you investigate things on the server.
0 -
Hi Rex, I've moved the alert settings in CSF up already to alert me if there's an avg load of 7 for 5 minutes which I feel is sufficient enough however do let me know If you feel that it should be changed.
0 -
That should be good - how many cores does the CPU have?
I'd work through the details in this document to see if you can further isolate the issue down to a specific account or process: https://support.cpanel.net/hc/en-us/articles/360056001894-How-to-diagnose-high-server-loads
0 -
This server has 8 Cores & 32GB RAM. I believe the issue is down to 1 site on this server from reading the PS emails when sent which runs WordPress.However I've been unable to find what page(s) are making the load go up.
0 -
It may not be something you'll easily find with a specific page. It could be the page itself loads fine, but calls a database that takes more power, or something along those lines. But with 8 cores, your system should easily be able to handle a load of 8 before things start and slow down.
0 -
Yeah, it's more the 1 minute load that is wanting me to look further into it as these increased load periods range from 20-40 average load at a time. I've also looked at MySQL on the in the PS.txt and it was using 3% CPU. I've looked at the link you've sent too but the site doesn't seem to be going down yet. Just trying to be more proactive haha.
Thanks for the help so far.
0 -
Troubleshooting load is almost always best done in real-time while the issue is happening. It's often very difficult to get data after the fact, except in the ways outlined by that article, but that's often not enough to see what the root cause of the issue is.
If the problem creeps up at a certain time each day, it would be best to monitor the system then to see if you can catch something in "top" or with "httpd status" to see if you can narrow things down more.
0
Please sign in to leave a comment.
Comments
7 comments