Having WordPress website setup using the 1click script keeps hanging (have to reboot)

[mukeshtechspace]

We have several AWS Lightsail VMs on Ubuntu 20.04 and installing WordPress using the 1 click script for OLS. The site runs fine but every couple of days or randomly will hang requiring a restart of services or reboot. If there are any ideas or areas to investigate, I would appreciate that. Any experts that can help for pay, would be welcome too.
Thanks.

 

[Cold-Egg]

That sounds weird, by the hang mean the page is showing white or? If you visit a static file, does it loadable?

 

[mukeshtechspace]

Hi there, thanks for the reply. All of the instances are running through Cloudflare so the end uses get the "origin" is not reachable error (502 I believe). At that time we reboot the instance. I don't have any logs saved from before the reboot. In one of the case I remember logs showing Reached max children process limit error. Which I did increase the 'LSAPI_CHILDREN' variable but in other instances, it wasn't the case. Are there things to capture when it happens again? Any settings I can set now to get more details in logs?

Thanks in advance,
Mukesh

 

[Cold-Egg]

When the issue happens again, please run
"top" command and share the output here
also, "ps aux | grep lsphp | wc -l" to calculate the lsphp process numbers.

before reboot, try "killall lsphp" and see if it helps to fix the hang

 

[mukeshtechspace]

Thanks. I'll try to post the info you requested. Currently when I run the ps aux command it shows: 4. Is that number supposed to correspond to any of the settings like LSAPI_CHILDREN in the Admin?

Also I was assuming settings like " Auto Fix 503 Error " or " Graceful Restart Timeout " settings will restart the processes if it is down. If not, is there any setting like that available that can restart LSWS if it is down?

Thanks.
Mukesh

 

[Cold-Egg]

Yes, and LSAPI_CHILDREN means the limit. Those are PHP, not really web server related, that's why I asked you to killall lsphp processes.