Bug #9859
closed
Memory exhaustion by hundreds of minicron and php-cgi processes.
0%
Description
After repeated gateway failovers I noticed I wasn't able to login any more using https or ssh.
I would then get an email that a gateway is or was down.
After days (I'm not on location but routing mostly worked) I finally was able to get into a shell and run top - I saw hundreds of php-cgi processes taking up about 3.5 of 4 GB ram
killall php-cgi would only help for seconds, after which the processes were respawned.
Only after using killall on the minicron processes, this behavior stopped.
Updated by Jim Pingle about 5 years ago
- Category set to Gateway Monitoring
- Status changed from New to Rejected
- Affected Architecture added
- Affected Architecture deleted (
amd64)
There is not enough solid information here to classify this as an identifiable or reproducible bug. This site is not for support or diagnostic discussion.
For assistance in solving problems, please post on the Netgate Forum or the pfSense Subreddit .
See Reporting Issues with pfSense Software for more information.