Can't figure this one out, been an admin for a long time.
The system did a graceful shutdown with no-one logged in, no shutdown command issued, no power button pressed, nothing unusual in munin system monitoring, no apcupsd events, just daemons reporting that they received sigterm and were stopping. Server is attached to a 1500VA UPS. Ubuntu 12.04.2 LTS with a 32-bit PAE kernel.
Here are all the related logs I could identify in /var/log:
apache2/error.log:[Mon Apr 29 16:10:33 2013] [notice] caught SIGTERM, shutting down apcupsd.events:2013-04-29 16:10:35 +0300 apcupsd exiting, signal 15 apcupsd.events:2013-04-29 16:10:35 +0300 apcupsd shutdown succeeded mediatomb.log.1:2013-04-29 16:10:33 INFO: MediaTomb shutting down. Please wait... mediatomb.log.1:2013-04-29 16:10:33 INFO: Server terminating munin/munin-node.log:2013/04/29-16:10:33 Server closing! samba/log.nmbd:[2013/04/29 16:10:33.333112, 0] nmbd/nmbd.c:66(terminate) syslog:Apr 29 16:10:33 serengeti acpid: client 31644[0:0] has disconnected syslog:Apr 29 16:10:33 serengeti kernel: Kernel logging (proc) stopped. syslog:Apr 29 16:10:33 serengeti rsyslogd: [origin software="rsyslogd" swVersion="5.8.6" x-pid="1079" x-info="http://www.rsyslog.com"] exiting on signal 15. tinyproxy/tinyproxy.log:INFO Apr 29 16:10:35 [3456]: Shutting down. Questions:
- Did anyone ever see something like this before?
- Is there something I can install (in addition to process accounting, which I added now) that would help me understand what happened, if it happens again?
Any suggestions gratefully received; if I find out what happened I'll report back.
[link][29 comments]