Applicable to:
- Plesk for Linux
Symptoms
-
The
sw-engineprocess causes high CPU usage:# ps aux | grep sw-engine
root 1242 0.0 0.3 348136 28732 ? Ss Jan14 0:14 /usr/bin/sw-engine -c /usr/local/psa/admin/conf/php.ini /usr/local/psa/admin/bin/modules/watchdog/wdcollect -c /usr/local/psa/etc/modules/watchdog/wdcollect.inc.php
root 2315 0.0 0.1 353260 9908 ? Ss Jan14 0:00 sw-engine-fpm: master process (/etc/sw-engine/sw-engine-fpm.conf)
root 3136 0.0 0.3 363840 32028 ? S Jan14 1:00 /usr/bin/sw-engine -c /usr/local/psa/admin/conf/php.ini /usr/lib64/plesk-9.0/psa-health-monitor-notification.php -
Plesk may fail to load with a 502 error.
Cause
Some sw-engine child processes are stuck and cannot be stopped normally.
Resolution
-
Connect to the Plesk server via SSH.
-
Stop the
sw-engineservice:# service sw-engine stop
Stopping sw-engine-fpm: [ OK ] -
Find stuck sw-engine processes and their PID:
# ps aux | grep sw-engine
psaadm 1597 0.0 2.1 274896 44100 ? Ss 12:39 0:00 /usr/bin/sw-engine -c /usr/local/psa/admin/conf/php.ini /usr/local/psa/admin/plib/WebSocket/bin/ws-server.php
psaadm 1842 0.0 2.3 278952 49564 ? S 12:42 0:00 /usr/bin/sw-engine -c /usr/local/psa/admin/conf/php.ini /usr/local/psa/bin/extension --exec revisium-antivirus ra_executor.php -
Kill these stuck processes using their PID. In the example above, they are 1597 and 1842:
# kill -9 1597 1842
-
Start the
sw-engineservice:# service sw-engine start
Starting sw-engine-fpm: [ OK ]
Comments
5 comments
What is the long term solution to this problem? Is there an upgrade available to fix this problem? Please elaborate on how to prevent this from occurring in the first place.
@Bryan Muthig
There are various causes why fpm service can get stuck.
Such issues should be checked individually.
If you are facing such issue I recommend you submit a Support ticket:
How to submit a request to Plesk support?
I have this problem too with Plesk Obsidian Web Admin Edition Version 18.0.17.
Suddenly at random time, sw-engine hang 100% CPU and crazy high I/O load (I have 10 disks, with several partition, but with atop I see disk usage for all about 70%, something never happen normally, 10% is the maximum). Rester sw-engine correct the problem.
It's a Plesk script, so this is not normal something like this happen. That kill my website 2 times since yesterday and give me headache. Please, update your script for a protection against this. You can check the activity of sw-engine and restart the service when something not normal happen.
Hello @Alexis,
The described behavior isn't typical and required additional investigation. In case the issue is reproduced with the most recent version of Plesk Obsidian (RC4), this required additional investigation.
Please, could you submit a request to Plesk support: https://support.plesk.com/hc/en-us/articles/213608509-How-to-submit-a-request-to-Plesk-support-
I have a recurring problem with this.
Please sign in to leave a comment.