• 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

Issue Plesk server became unreachable.

ajeffcoat

New Pleskian
Hello,

I have a newly deployed VPS with CentOS 8 and Plesk installed. I updated a website in WordPress, I cleared the cache using the plugin WP Rocket. After doing this, the server became completely unresponsive. I had to have my host reboot the VPS. Below are the logs I see for the domain. This same site worked perfectly on my old Plesk server. The only difference is I have enabled the WAP. Any suggestions would be appreciated. After doing some reading yesterday, I did change the PHP FRM Application setting to Apache. I'm just not 100% that is right.

2021-01-25 15:08:51Error108.162.245.144499POST /wp-admin/admin-ajax.php HTTP/1.1
https://lostreschiles.com/wp-admin/post.php?post=1050&action=edit
Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/87.0.4280.141 Safari/537.36
0Apache SSL/TLS access
2021-01-25 15:09:55Error108.162.245.1443349648#0: *56263 upstream timed out (110: Connection timed out) while reading response header from upstreamnginx error
 
One of my plesk servers became completely unresponsive (but still running) two times in the last two months. It's also an CentOS 8 VPS.
Reason absolutely unknown.

Please report back if you find the culprit. I will do too.
 
In the first case @ajeffcoat you might find some 403 errors in your *-error_log files in /logs. These are normally created by ModSecurity due to rule violations. You can add the "ID" of the rules as exceptions to "Web Application Firewall" in the panel. Another issue with WP Rocket can be infinite redirect/rewrite loops that cause PHP to time-out instead of to deliver a result.

In the second case @homberger the first place to check is Fail2Ban (Tools & Settings > Security), because that is probably blocking your local IP address for any type of failed login attempts.
 
The server became unresponsive during or after an kernel update with dracut. Multiple services got stuck in a restart loop as monit was repeatedly starting services that were restarting already (because of the hot/online kernel update).
 
Back
Top