On Tue, 21 Apr 2015, Peter Viskup wrote:
During the implementation of rsyslog queue monitoring I discovered not all the queue files are removed after flush. They are removed just with the stop/restart of the rsyslog process.
I believe the issue is around race conditions. What happens if you try to delete the files in one thread but then another thread finds that a memory buffer is full and tries to write a log to the file at the same time.
David Lang
I am running older rsyslog version available in official Debian Wheezy. Does anybody know the background of this behavior? root@server:~# ls -latr /var/log/net/forward_queue* -rw------- 1 lognet lognet 990501 Apr 20 04:27 /var/log/net/forward_queue.00000050 -rw-r--r-- 1 lognet lognet 507 Apr 21 14:58 /var/log/net/forward_queue.qi And these are rsyslog statistics: 2015-04-21T15:33:42.086517+02:00 10.X.Y.Z rsyslogd-pstats: main Q[DA]: size=0 enqueued=0 full=0 maxqsize=0 2015-04-21T15:33:42.086519+02:00 10.X.Y.Z rsyslogd-pstats: main Q: size=3 enqueued=85885 full=0 maxqsize=67 2015-04-21T15:34:42.138795+02:00 10.X.Y.Z rsyslogd-pstats: action 6 queue[DA]: size=0 enqueued=0 full=0 maxqsize=0 2015-04-21T15:34:42.138814+02:00 10.X.Y.Z rsyslogd-pstats: action 6 queue: size=0 enqueued=88016 full=0 maxqsize=165
_______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com/professional-services/ What's up with rsyslog? Follow https://twitter.com/rgerhards NOTE WELL: This is a PUBLIC mailing list, posts are ARCHIVED by a myriad of sites beyond our control. PLEASE UNSUBSCRIBE and DO NOT POST if you DON'T LIKE THAT.
