Description
I’m migrating a website from another web hosting company. It’s a WordPress site with about 200 requests on each page load, so that definitely needs to change and be optimized. It takes about 1-2 seconds normally for each page to load on a known good server (not good, I know, bear with me). When I migrated it to ApisCP, the site takes 15 seconds or more to load. Looking at the iostats the disk is hit hard.
The server that runs ApisCP has about 10 sites that are low traffic. The site I’m migrating is low-medium traffic. It’s sitting on a dedicated box, Rocky Linux, AMD EPYC 7642 4-core vCPU, 8 GB RAM. Other sites also feel slow.
When I load this site, it takes 15 seconds or more, and it bad enough where it kills my SSH connection and I have to reconnect, and then shows either 429 Too Many Requests or 503 Service Unavailable in the browser. Restarting php-fpm seems to resolve it, but then there’s also a shield in the mix that I’ve disabled temporarily for troubleshooting. If I migrate the site to my older shared VPS, it loads the same site in 1-2 seconds. So I don’t think it’s a hardware issue. It feels like ApisCP is limiting performance, or something.
I adjusted the max memory in PHP to 768M and increased the php-fpm max processes to 50. The site still has the same issue.
Other websites on the server also feel slow. Even a light WordPress site takes about 500-1000ms to load, whereas on another server it takes 100 ms.
I’ve checked many areas but am not sure what to do to troubleshoot this.
Environment
ApisCP version: cpcmd misc:cp-version
revision: 45fade15eb9c075fc4143fbb67f6dbab116538aa
timestamp: 1768180847
ver_maj: 3
ver_min: 2
ver_patch: 48
ver_pre: 41-g45fade15e-dirty
dirty: true
debug: false
Operating System: uname -r
Additional relevant information (e.g., PHP version, database, etc.):
4.18.0-553.89.1.el8_10.x86_64
MySQL
PHP 8.2.27
I appreciate your time in helping me solve this!
).