Snort stops working after snort update (newest 2.0 RELEASE)
-
same here.
Hi I noticed someone else recommended you change your snort interface settings to AC-BNFA mode to lower memory usage. I was set to AC-STD prior to upgrade and everything worked fine. Now that I changed to AC-BNFA after upgrade it appears to be working so far. Will keep monitoring.
I just changed it but I dont see how this can be an issue. I have 4Gigs of ram…
Hopefully this does the job :)Thanks
-
I too have 4 GB of RAM but somehow it seems to have resolved my issue and dramatically reduced my memory usage.
-
I made a post about this as well. However I just started trying to use it so don't know if it was an issue in the past. However I just tried to adjust my memory usage to AC-BNFA & my WAN interface still will not start.
Perhaps I spoke too soon. It would appear that, after, Snort performs its scheduled definitions update (and attempts to restart), it crashes. I must then manually start each snort interface but monitor periodically to ensure they didn't stop. Haven't yet identified any pattern in the logs to indicate what might be causing it to stop. Also, am not quite convinced that changing memory setting from AC-STD to AC-BNFA has any impact at all.
-
I made a post about this as well. However I just started trying to use it so don't know if it was an issue in the past. However I just tried to adjust my memory usage to AC-BNFA & my WAN interface still will not start.
Perhaps I spoke too soon. It would appear that, after, Snort performs its scheduled definitions update (and attempts to restart), it crashes. I must then manually start each snort interface but monitor periodically to ensure they didn't stop. Haven't yet identified any pattern in the logs to indicate what might be causing it to stop. Also, am not quite convinced that changing memory setting from AC-STD to AC-BNFA has any impact at all.
As I thought.
I think the issue is that snort has some how change there definition pattern once again. -
How is this thread not firing off sirens? We surely aren't the only ones having this issue. Would expect the majority of users are experiencing this but simply haven't checked their snort interface statuses to ensure they stay running (without a restart)
-
Well probably a bug in snort which does not like to reload the rulesets.
Maybe changing snort to stop and start again would be a better choice i guess. -
BTW can you post the system logs here for review and hopefully fixing that?
-
ermal,
I been trying to catch it but is so random that I have been unable to… usually happens overnight...
-
Ermal here is the system log.
Sep 30 09:03:55 snort[5170]: Snort exiting
Sep 30 09:03:55 snort[5170]: Snort exiting
Sep 30 09:03:54 snort[6986]: Snort exiting
Sep 30 09:03:54 snort[6986]: Snort exiting
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2408044 type=Limit tracking=src count=1 seconds=60 filtered=8
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2408044 type=Limit tracking=src count=1 seconds=60 filtered=8
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2406628 type=Limit tracking=src count=1 seconds=60 filtered=18
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2406628 type=Limit tracking=src count=1 seconds=60 filtered=18
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2403304 type=Limit tracking=src count=1 seconds=3600 filtered=1
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2403304 type=Limit tracking=src count=1 seconds=3600 filtered=1
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2001569 type=Both tracking=src count=70 seconds=60 filtered=331
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2001569 type=Both tracking=src count=70 seconds=60 filtered=331
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2406036 type=Limit tracking=src count=1 seconds=60 filtered=4
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2406036 type=Limit tracking=src count=1 seconds=60 filtered=4
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2001579 type=Both tracking=src count=70 seconds=60 filtered=330
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2001579 type=Both tracking=src count=70 seconds=60 filtered=330
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2408002 type=Limit tracking=src count=1 seconds=60 filtered=31
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2408002 type=Limit tracking=src count=1 seconds=60 filtered=31
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2406690 type=Limit tracking=src count=1 seconds=60 filtered=2
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2406690 type=Limit tracking=src count=1 seconds=60 filtered=2
Sep 30 09:03:53 snort[5170]: | gen-id=1 sig-id=2403312 type=Limit tracking=src count=1 seconds=3600 filtered -
Just to add confirmation – I, too, have noticed this behavior of Snort not restarting upon a rules update. However, you can immediately manually start it without issue. Like one of the other posters, the only message in the log of note is "...snort exiting." Not much help for troubleshooting.
I discovered my event by kicking off a manual update and then coming back an hour later to check that everything was cool. Found pfSense running fine but Snort not running. I looked in the log and saw only the "snort exiting" message. I made no changes to anything and simply clicked the little green arrow icon to start Snort. It then started up successfully.
This seems to have only occurred after my upgrade to the 2.0-Release. Prior to that I was running the Release Candidate (RC3) and did not see this issue.
-
I'm experiencing the same problem since upgrading to 2.0-Release. The system log shows the usual Snort initialization messages but with the last one indicating 'Snort exiting' and it generally occurs just after midnight EST (I assume after a ruleset update …which I have set for once a day).
I also used to run with the AC-STD memory setting, but can no longer do so since the upgrade (Snort won't run for more than an hour or two with AC-STD, so I changed to AC-SPARSEBANDS). Last, Snort seems to take quite a bit longer to start up (several minutes) since the upgrade.
I hope that this problem is going to be investigated ... thx.
-
I hate to post a "me too" reply, but unfortunately, me too. Snort 2.9.0.5 pkg 2.0, pfsense 2.0-RELEASE (i386)
built on Tue Sep 13 17:28:43 EDT 2011.All I can find in the kernel log for snort is "snort exiting". And as with the others, this always happens overnight. Unlike the others, I had upgraded/updated to 2.0 final, a few weeks ago. But only noticed this behaviour since, maybe end of last week?
Prior to that, snort had been running ok.
So I'm suspecting maybe it's one of the rules that's causing a problem? From memory, one of the snort categories always caused me problems (e.g. snort_exploit.rules), so maybe another is doing the same thing?
Update: I just tried a manual update of rules, and restarted snort. Now the snort instance running on my WAN interface (AC-BNFA) has died a few minutes later (15 minutes), and with an error message "kernel: pid 23257 (snort), uid 920: exited on signal 11". More logs:
Oct 5 11:32:43 snort[35687]: PID path stat checked out ok, PID path set to /var/log/snort/run Oct 5 11:32:43 snort[35687]: PID path stat checked out ok, PID path set to /var/log/snort/run Oct 5 11:32:43 snort[35687]: Writing PID "35687" to file "/var/log/snort/run/snort_em0_vlan78035309.pid" Oct 5 11:32:43 snort[35687]: Writing PID "35687" to file "/var/log/snort/run/snort_em0_vlan78035309.pid" Oct 5 11:32:43 snort[35687]: Set gid to 920 Oct 5 11:32:43 snort[35687]: Set gid to 920 Oct 5 11:32:43 snort[35687]: Set uid to 920 Oct 5 11:32:43 snort[35687]: Set uid to 920 Oct 5 11:32:43 snort[35687]: Oct 5 11:32:43 snort[35687]: Oct 5 11:32:43 snort[35687]: --== Initialization Complete ==-- Oct 5 11:32:43 snort[35687]: --== Initialization Complete ==-- Oct 5 11:32:43 snort[35687]: Commencing packet processing (pid=35687) Oct 5 11:32:43 snort[35687]: Commencing packet processing (pid=35687) Oct 5 11:48:26 kernel: pid 23257 (snort), uid 920: exited on signal 11 -
And famous last words, now snort on the WAN interface is crashing regularly after a restart:
Oct 5 12:52:44 snort[14024]: Set uid to 920 Oct 5 12:52:44 snort[14024]: Oct 5 12:52:44 snort[14024]: Oct 5 12:52:44 snort[14024]: --== Initialization Complete ==-- Oct 5 12:52:44 snort[14024]: --== Initialization Complete ==-- Oct 5 12:52:44 snort[14024]: Commencing packet processing (pid=14024) Oct 5 12:52:44 snort[14024]: Commencing packet processing (pid=14024) Oct 5 13:42:06 kernel: pid 14024 (snort), uid 920: exited on signal 11I'll have to start disabling rules and see if it is indeed the rules causing the problem…
-
FYI I removed snort_exploit.so.rules from the categories I was using on the WAN interface (as well as snort_exploit.rules that I always had issues with crashing snort pretty much from startup), yesterday afternoon. Monitored last night and this morning, snort hasn't crash yet.
So might be worth your while to test as well to see if it is this set of rules causing your snort problem.
-
Just got this in my system logs after what appears to be an update; things have been running really solid up until now.
Oct 6 12:06:43 snort[39831]: FATAL ERROR: fpcreate.c(1557) Failed to compile port group patterns.
Oct 6 12:06:43 snort[39831]: FATAL ERROR: fpcreate.c(1557) Failed to compile port group patterns.
Oct 6 12:04:14 snort[39831]: 377 out of 1024 flowbits in use.
Oct 6 12:04:14 snort[39831]: 377 out of 1024 flowbits in use.Now Snort service is stopped. I forced an update and it's back up and going again.
We'll see what happens next. So far so good. :)
-
Got this error again; once this morning and once now. I don't know if it's related to updates, because I have it set for 12 hours and it hasn't been yet.
Oct 7 12:06:20 snort[11782]: FATAL ERROR: Could not create rule maps
Oct 7 12:06:20 snort[11782]: FATAL ERROR: Could not create rule maps
Oct 7 12:03:36 snort[11782]: 377 out of 1024 flowbits in use.
Oct 7 12:03:36 snort[11782]: 377 out of 1024 flowbits in use.Doing manual update of rules starts it up again.
-
And got this again with a slightly different error.
Oct 7 20:27:08 kernel: pid 27248 (snort), uid 920: exited on signal 11
Oct 7 20:27:07 snort[27248]: FATAL ERROR: fpcreate.c(1557) Failed to compile port group patterns.
Oct 7 20:27:07 snort[27248]: FATAL ERROR: fpcreate.c(1557) Failed to compile port group patterns.
Oct 7 20:25:09 snort[27248]: 377 out of 1024 flowbits in use.
Oct 7 20:25:09 snort[27248]: 377 out of 1024 flowbits in use. -
I will try to give it a look and reproduce it somehow depending on availble time.
Seems like something has changed somewhere…. -
Okay, an update on what I'm seeing. I have the snort.exploits running on both servers with no problem(I don't know if we should compare rules running.), but I did have spamd installed/running and I uninstalled it and also did an install of County Block. So far the Snort service has not quit; ermal… I don't know if that helps with anything you are looking at, but I know the one p3 system I have running has not quit out at all, where my xeon system was exiting almost every 6 hours or so and seemed consistant. My updates are set for 12 hours and all I would have to do is run the update check and it would somehow repair and start the service again. It didn't even download any new rules, it would run through some of the other standard processes and then say things were updated.
Let you know if anything changes. I'm sure it's tough trying to pin point issues when so many other software packages have to be considered. Thanks for your work on this ermal.
Update 10/9/11@7:05am So far so good; no exits or errors after update. Still keeping an eye on things.
-
Stopped again -
Oct 9 12:05:57 kernel: pid 44648 (snort), uid 920: exited on signal 11
Oct 9 12:05:56 snort[44648]: FATAL ERROR: Could not create rule maps
Oct 9 12:05:56 snort[44648]: FATAL ERROR: Could not create rule maps
Oct 9 12:03:39 snort[44648]: 377 out of 1024 flowbits in use.
Oct 9 12:03:39 snort[44648]: 377 out of 1024 flowbits in use.** UPDATE **
11/10/2011 - Error and exit again today.
Oct 10 12:05:44 snort[56669]: FATAL ERROR: fpcreate.c(1557) Failed to compile port group patterns.
Oct 10 12:05:44 snort[56669]: FATAL ERROR: fpcreate.c(1557) Failed to compile port group patterns.
Oct 10 12:03:42 snort[56669]: 377 out of 1024 flowbits in use.
Oct 10 12:03:42 snort[56669]: 377 out of 1024 flowbits in use.
