• 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 Httpd -D FOREGROUND High Memory

cem

New Pleskian
Hello,

About 1 month ago I had a problem accessing the server. I upgraded plesk version to 18.0.54. The configuration of the sites is broken. I reconfigured it via plesk. After that day, httpd started to use high ram from time to time. It started happening twice a day. It was disappearing when I restarted httpd. but it started to lock the server for 3-4 days. sometimes it comes back like a monster every 10 minutes, sometimes every 3-4 hours. For example, nothing happened last night. it started this morning and it comes back every 10 minutes. I tried everything I searched on the internet. I upgraded plesk version to 18.0.55. There are about 1000 domains.

Plesk version: 18.0.55
Ram: 32 GB
Databse: Mariadb
WAF: Closed
Grafana: Closed
php747.4.33-1centos.7.221109.0743
nginx1.24.0-2.centos.7+p18.0.53.0+t230516.1254
nss3.79.0-5.el7_9
openssl1.0.2k-26.el7_9
httpd2.4.6-99.el7.centos.1
CPUIntel(R) Xeon(R) CPU E3-1240 V2 @ 3.40GHz (8 core(s))
VersionPlesk Obsidian v18.0.55_build1800230825.08 os_CentOS 7
OSCentOS Linux 7.9.2009 (Core)
Key numberPLSK.10587643.0001
 

Attachments

  • a5.png
    a5.png
    27.2 KB · Views: 10
  • a4.png
    a4.png
    44.8 KB · Views: 8
  • a3.png
    a3.png
    91.8 KB · Views: 8
  • a2.png
    a2.png
    92.6 KB · Views: 8
  • a1.png
    a1.png
    32.7 KB · Views: 8
  • a6.png
    a6.png
    16.9 KB · Views: 12
The RAM usage seems to be a symptom. Have you also looked at the load average? It is way too high. The high load also indicates an issue, maybe another symptom for the same root cause. The root cause are very likely PHP scripts, maybe bad bot attacks, maybe just a lot of traffic against one of your sites. You can find out more by checking ps aux | grep php-fpm | grep -v master. This will show you the user processes and their cpu load. From there, check the webserver access_ssl_log file of the culprits to find out what's going on with their websites. Solving the cpu load issue will probably solve the RAM issue along.
 
Back
Top