syslog-ng eventually gets to 100% CPU usage

Status
Not open for further replies.

Will Dormann

Explorer
Joined
Feb 10, 2015
Messages
61
Hi folks,

I've encountered this once before, and I'm now seeing it once again since the last reboot. At some point, syslog-ng will start using 100% of a single CPU core. Rebooting the system will bring it back to normal, but obviously that isn't practical.

Before I file a bug, has anybody seen this behavior before? Or more importantly, is there a ticket already filed?
 
Joined
Apr 9, 2015
Messages
1,258
Just updated to 9-10-U3 a couple days ago and noticed the same thing this morning when I logged into the server. This had not happened in previous versions that I am aware of on my system. Have not rebooted as of yet in case something needs to be pulled from the system.

Code:
last pid: 75637;  load averages:  1.16,  1.11,  1.09  up 3+04:38:03	13:17:59
76 processes:  1 running, 75 sleeping

Mem: 98M Active, 1888M Inact, 43G Wired, 20M Cache, 1401M Free
ARC: 40G Total, 20G MFU, 19G MRU, 7012K Anon, 253M Header, 728M Other
Swap: 14G Total, 14G Free


  PID USERNAME	  THR PRI NICE   SIZE	RES STATE   C   TIME	WCPU COMMAND
1671 root			2  20	0 68940K  9264K kqread 13 493:57 100.00% syslog-ng
 
D

dlavigne

Guest
Thanks for following up. Please create a bug at bugs.freenas.org and post the issue number here.
 
Joined
Apr 9, 2015
Messages
1,258
Status
Not open for further replies.
Top