• If you are still using CentOS 7.9, it's time to convert to Alma 8 with the free centos2alma tool by Plesk or Plesk Migrator. Please let us know your experiences or concerns in this thread:
    CentOS2Alma discussion

Question Apache minimum configuration

Seba

Basic Pleskian
Last year I started to manage the company servers and improve the whole project.
Soon I'm going to configure a new server and make some test with HTTP/2 to improve the speed of our mapping service.

I was checking our documentation and our servers configuration and I saw that we use the basic configuration of Plesk, we use Apache event mpm.

This is our mapping project Buchanan Computing map service
As you can see it requires a lot of images to create the whole map, and at the moment we have 60 website connecting to this website indirectly.

Is there some setting that I should implement in our Apache config to optimise the speed a part of switching to HTTP/2 with Nginx?
 
Is there some setting that I should implement in our Apache config to optimise the speed a part of switching to HTTP/2 with Nginx?

Yes, but it's not in the Apache config. Use Apache as little as possible and let Nginx serve all the static content. Turn on Nginx caching, too. This will greatly speed things up.

Regarding your scripts, if they are PHP, switch to 7.2 PHP-FPM. Change Apache Event to Apache Prefork and give it enough spare servers in /etc/httpd/conf settings and anough children per server instances.
 
Regarding your scripts, if they are PHP, switch to 7.2 PHP-FPM. Change Apache Event to Apache Prefork and give it enough spare servers in /etc/httpd/conf settings and anough children per server instances.
I've already switched PHP time ago, why Prefork over Event?
 
Yes, but it's not in the Apache config. Use Apache as little as possible and let Nginx serve all the static content. Turn on Nginx caching, too. This will greatly speed things up.

Regarding your scripts, if they are PHP, switch to 7.2 PHP-FPM. Change Apache Event to Apache Prefork and give it enough spare servers in /etc/httpd/conf settings and anough children per server instances.


This is no longer where things sit if you are running CentOS 7. I keep running into the problem of

[Mon Feb 25 06:00:41.271724 2019] [mpm_prefork:error] [pid 20951] AH00161: server reached MaxRequestWorkers setting, consider raising the MaxRequestWorkers setting

and trying to tune prefork for more MaxRequestWorkers. Not to hijack this thread but if he is being advised to change settings for prefork we should kinda know where to do that. I've been advised to edit /etc/httpd/conf.modules.d/01-cgi.conf as so but can't see where to grep the current MaxRequestWorkers to see if these changes applied.

<IfModule mpm_prefork_module>
LoadModule cgi_module modules/mod_cgi.so
MaxRequestWorkers 4000
ServerLimit 4000
</IfModule>
 
<IfModule mpm_prefork_module>
LoadModule cgi_module modules/mod_cgi.so
MaxRequestWorkers 4000
ServerLimit 4000
</IfModule>
 
Back
Top