Bug #13252


reduce frequency of php-fpm socket connection attempts from check_reload_status

Added by Royce Williams almost 2 years ago. Updated about 1 year ago.

Target version:
Start date:
Due date:
% Done:


Estimated time:
Plus Target Version:
Release Notes:
Affected Version:
Affected Architecture:


When troubleshooting an issue, I discovered that my system logs were rotating every couple of minutes, due to many of these log entries being generated per second:

May 31 12:56:34 zeb check_reload_status[576]: Could not connect to /var/run/php-fpm.socket
[64 messages in the same second omitted]
May 31 12:56:34 zeb check_reload_status[576]: Could not connect to /var/run/php-fpm.socket

I am not sure what the root cause of this, but it appears to be within the Upgrade category.

If it's possible to reduce the frequency of this, that could help some users in the future.

Actions #1

Updated by → luckman212 almost 2 years ago

I may have also experienced this on an SG-2100 yesterday. Upgraded from 21.05.1 to 22.05-RC.

After the upgrade, CPU was getting absolutely annihilated, and webGUI was unresponsive. After connecting to the console and running top, I saw 4 or 5 bzip2 processes all grinding (Log rotation and compression). I killed them with pkill bzip2 and disabled compression to avoid it. After rebooting, it seems ok.

I noticed a bunch of threads in recent weeks with people complaining of high RAM and/or CPU usage after upgrading to 22.05 betas, so just adding this note here. Seems like it could be caused by this logrotate loop.

Actions #2

Updated by Konstantin Svist about 1 year ago

If it helps, I've experienced something similar in the past few days.
I got a report that "internet is down", but by the time I was able to check pfsense, all system logs had already been rotated full of this message spammed over and over
`Mar 29 18:00:00 pfSense check_reload_status404: Could not connect to /var/run/php-fpm.socket`

This might be happening because the upstream gateway disconnected (there are storms in the area, so the connection may have been broken). By the time I checked the system, the interface was down for some time.
CPU was not overwhelmed (maybe because it's a beefier system than yours?), but access to web UI was broken.

Let me know if any extra logs might be helpful.


Also available in: Atom PDF