notify segfault every day. what is notify?

Message

dopey · Post by **dopey** » Wed Sep 12, 2007 6:43 pm

Hi all. I get this:
[ 2188.776086] notify[9582]: segfault at 00000000203a656f rip 00000000f7cd4013 rsp 00000000ffac5de4 error 4
In my kernel log every day at shortly after 4:30 am. I'm using the standard gentoo cronbase with only one user crontab job that runs at 7 am.

Otherwise, this is my /etc/crontab:
# check scripts in cron.hourly, cron.daily, cron.weekly and cron.monthly
0 * * * * root rm -f /var/spool/cron/lastrun/cron.hourly
1 3 * * * root rm -f /var/spool/cron/lastrun/cron.daily
15 4 * * 6 root rm -f /var/spool/cron/lastrun/cron.weekly
30 5 1 * * root rm -f /var/spool/cron/lastrun/cron.monthly
*/10 * * * * root test -x /usr/sbin/run-crons && /usr/sbin/run-crons

I have logrotate.cron, makewhatis and slocate in my cron.daily and all the other cron directories are empty. cron.daily usually starts at 3:10: first */10 minute break after 3:01 which is when teh cron.daily last run file gets deleted. The three tasks don't take over an hour to run.

So I can't figure out what else would cause the notify segfault at the same approximate time every day. I can't find a binary with the name notify. Any idea what this might be that's segfaulting?

Utlo · Post by **Utlo** » Tue Sep 18, 2007 12:08 am

It be helpfull if you can be sure it's not a problem with a cron script. To do this just turn them all off and see if the error remains. If it dissapears you can turn the scripts back on one at time to track it down.

dopey · Post by **dopey** » Thu Sep 20, 2007 3:12 pm

thanks
sometimes you don't think about something that simple

stopped cron completely.

Sep 20 04:30:51 awang [60462.894138] notify[12253]: segfault at 00000000203a656f rip 00000000f7d6c013 rsp 00000000ffa1dd34 error 4

So something else is causing this at 4:30am. That rules out my primary culprit. Now to figure out what else runs every day at around that time.

dopey · Post by **dopey** » Mon Sep 24, 2007 10:28 pm

okay. so on thursday evening (i was out from work on PTO friday) we had a power outage. I disabled cron from starting up automatically, so until monday morning (this morning) my machine was running, and I wasn't logged in.

Sep 21 04:30:57 awang [37616.821358] notify[1596]: segfault at 00000000203a656f rip 00000000f7d76013 rsp 00000000ffaa25b4 error 4
Sep 22 04:30:47 awang [123958.316393] notify[30197]: segfault at 00000000203a656f rip 00000000f7caa013 rsp 00000000ff920c34 error 4
Sep 23 00:30:00 awang [195872.347979] notify[15854]: segfault at 00000000203a656f rip 00000000f7c97013 rsp 00000000ffbca6e4 error 4
Sep 23 04:30:50 awang [210314.355412] notify[26026]: segfault at 00000000203a656f rip 00000000f7d2c013 rsp 00000000ff8be3d4 error 4
Sep 24 04:30:43 awang [296659.164339] notify[21617]: segfault at 00000000203a656f rip 00000000f7d1a013 rsp 00000000fff34a54 error 4

So something that isn't related to a process from me being logged in is resulting in this segfault. Through the rest of this week, I'm going to shut down my machine and bring down as many of my server processes as possible to try to figure this out, but given that it's not a cron job, and it's not something due to my user being logged in overnight, anyone have any clue what might be rcausing this segfault at approximately 4:30 every day with the exception of an extra run at 00:30 on Sunday?