Question WP+NGINX=HIGH APACHE MEMORY USAGE??

[cloudbridge]

Hello pleskians,

Yesterday I encountered a problem that one of the server nodes was updated and then problems with RAM usage started. I tried to enable fpm to use only apache, served by NGINX, NGINX + CACHE ON. PHP versions, but I am using just the latest 7.3.1

2 days before node updates

One site (which is a news portal) uses a lot of RAM and a few days ago it wasn't. As soon as you change from NGINX to APACHE, the RAM usage is automatically reduced, but overnight it has risen from 1.5 GB to 7GB.

I upload a picture of what happens when I reset NGINX to APACHE and restore it as it was.

Maybe someone has encountered that and could advise?

Plesk Onyx
Version 17.8.11 Update # 40, last updated on Feb 15, 2019 09:56 PM

 

[Bitpalast]

One suggestion without know the root cause of your issue: PHP-FPM 7.3 is buggy, downgrade to 7.2.
Forwarded to devs - PHP 7.3 FPM dies without a trace

However, you should also analyze your access_log and error_log files of the site that is using much RAM. Frequently this is caused by brute force attacks or long or infinitely long running PHP scripts.

 

[MarkM]

Have you tried adjusting the keepalive_timeout?

 

[GerdSchrewe]

One suggestion without know the root cause of your issue: PHP-FPM 7.3 is buggy, downgrade to 7.2.
Forwarded to devs - PHP 7.3 FPM dies without a trace

However, you should also analyze your access_log and error_log files of the site that is using much RAM. Frequently this is caused by brute force attacks or long or infinitely long running PHP scripts.

Hi Peter,
do you know if 7.3.3. is still "buggy"? (Forwarded to devs - PHP 7.3 FPM dies without a trace).
Thanx a lot in advance!

 

[Bitpalast]

We've not been offering it to customers due to the issues, I don't know what the status is. But I have not read in any release notes about fixes of the issue.