Upgraded pfSense 2.1 with Quagga OSPF, exiting on signal 6
-
hello,
last friday i updated my main firewall from 2.0.3 to 2.1 (and 2 of my 5 connected alix boards)
since then i have every day at ~21:07 o'clock the problem that the quagga ospfd daemon stops workingi get these messages in the log but can't find any hint what this can be.
System - General
Oct 2 21:07:22 kernel: pid 88002 (ospfd), uid 101: exited on signal 6
System - Routing
Oct 2 21:07:22 zebra[87492]: client 17 disconnected. 21 ospf routes removed from the rib Oct 2 21:07:22 ospfd[88002]: Received signal 11 at 1380740842 (si_addr 0x793ffeacf90); aborting... Oct 2 21:07:22 ospfd[88002]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 2 21:07:22 ospfd[88002]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 2 21:07:22 ospfd[88002]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 2 21:07:22 ospfd[88002]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 2 21:07:22 ospfd[88002]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 2 21:07:22 ospfd[88002]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 2 21:07:22 ospfd[88002]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination
at first i thought this would happen in a 24hour circle, but yesterday i restarted the whole machine at 23:00 and today the error appared again on 21:07 (and so 22hours later, and not 24hours)
i would be really glad if anyone of you has a clue what this could be.
thanks in advance
-
(fyi: if someone gets here with the search function)
after the error appeared again and again, i switched back to 2.0.3 and controlled the log a few days if something appears to be strange
and there i noticed an error on an assigned but inactive interface. after that i unassigned it and did a new upgrade to 2.1 and the last 3 days there was no error that broke the ospfd service -
spoke too soon :(
System - General Oct 15 21:05:42 kernel: pid 48355 (ospfd), uid 101: exited on signal 6 System - Routing Oct 15 21:05:42 zebra[48103]: client 16 disconnected. 21 ospf routes removed from the rib Oct 15 21:05:42 ospfd[48355]: Received signal 11 at 1381863936 (si_addr 0x793ffeacf90); aborting... Oct 15 21:05:36 ospfd[48355]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 15 21:05:36 ospfd[48355]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 15 21:05:36 ospfd[48355]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 15 21:05:36 ospfd[48355]: nsm_change_state(192.168.1xx.xx, Full -> Deleted): scheduling new router-LSA origination Oct 15 21:05:36 ospfd[48355]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 15 21:05:36 ospfd[48355]: nsm_change_state(192.168.1xx.xxx, Full -> Deleted): scheduling new router-LSA origination Oct 15 21:05:36 ospfd[48355]: nsm_change_state(192.168.1xx.xx, Full -> Deleted): scheduling new router-LSA origination Oct 15 21:05:36 ospfd[48355]: SLOW THREAD: task ospf_write (800674c40) ran for 80740ms (cpu time 0ms)
i'm thinking of a cron job to restart the ospfd service at 21:10 every day, but thats just a cheap workaround
the bigger try would be a complete new installation with the settings importeddoes no one know from where this error could come?
edit: at the moment i'm having the problem that i can't find a way to restart or start the ospfd daemon
with "/usr/local/etc/rc.d/quagga stop" i can stop the service by myself, but with "start" there is no output and it won't start
what am i doing wrong here?i would be really thankful for any help in this case!
-
no clue but i've seen the "exit on signal 6' error before on 2.1 installs. (mine runs on site-to-site openvpn interfaces)
i've installed the Service Watchdog package to monitor quagga and restart if needed. (no good fix AFAIK) -
Same problem here on 2.1.2 :D
Full -> Deleted each 1-2 mins and it drops the connection… wonderfull :-\ -
Looks like the dead timer is doing it (40s default), although it gets the hello messages quite regularly:
08:37:50.196698 IP x.x.x.244 > 224.0.0.5: OSPFv2, Hello, length 68 08:37:59.869506 IP x.x.x.244 > 224.0.0.5: OSPFv2, Hello, length 68 08:38:09.148189 IP x.x.x.244 > 224.0.0.5: OSPFv2, Hello, length 68 08:38:18.830889 IP x.x.x.244 > 224.0.0.5: OSPFv2, Hello, length 68