One of our servers rebooted last night
reboot system boot 2.6.32-5-amd64 Wed Jan 15 03:24 - 16:07 (1+12:42)
Our hoster didn't do it, nor did I.
syslog does not show any error
Jan 15 03:17:01 big /USR/SBIN/CRON[840]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly >> /dev/null)
Jan 15 03:20:01 big /USR/SBIN/CRON[844]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1)
Jan 15 03:24:44 big kernel: imklog 4.6.4, log source = /proc/kmsg started.
Jan 15 03:24:44 big rsyslogd: [origin software="rsyslogd" swVersion="4.6.4" x-pid="893" x-info="http://www.rsyslog.com"] (re)start
Jan 15 03:24:44 big kernel: [ 0.000000] Initializing cgroup subsys cpuset
Jan 15 03:24:44 big kernel: [ 0.000000] Initializing cgroup subsys cpu
Jan 15 03:24:44 big kernel: [ 0.000000] Linux version 2.6.32-5-amd64 (Debian 2.6.32-48squeeze4) (dannf@debian.org) (gcc version 4.3.5 (Debian 4.3.5-4) ) #1 SMP Mon Sep 23 22:14:43 UTC 2013
Jan 15 03:24:44 big kernel: [ 0.000000] Command line: root=/dev/md1 ro acpi=ht
Jan 15 03:24:44 big kernel: [ 0.000000] KERNEL supported cpus:
auth.log does not show anything in a wide timeframe arround 03:24.
lastlog does not show any connectings besides myself.
faillog has no entries.
uprecords
big:/var/log# uprecords
# Uptime | System Boot up
----------------------------+---------------------------------------------------
1 210 days, 06:33:18 | Linux 2.6.32-5-amd64 Wed Mar 13 09:49:06 2013
2 97 days, 10:58:48 | Linux 2.6.32-5-amd64 Wed Oct 9 17:23:44 2013
-> 3 1 day , 12:54:06 | Linux 2.6.32-5-amd64 Wed Jan 15 03:23:22 2014
I can see log entries for what happened after the reboot, but I cannot find ANY information for what caused the reboot.
The CPU usage was very low during the time so it is surely not a overheating problem.
03:10:01 PM all 0.01 0.00 0.00 0.00 0.00 99.99
03:15:01 PM all 0.02 0.00 0.00 0.00 0.00 99.97
03:20:01 PM all 0.01 0.00 0.00 0.00 0.00 99.99
03:25:01 PM all 0.02 0.00 0.00 0.00 0.00 99.97
03:30:01 PM all 0.01 0.00 0.00 0.00 0.00 99.99
03:35:01 PM all 0.01 0.00 0.00 0.00 0.00 99.99
03:40:01 PM all 0.02 0.00 0.01 0.00 0.00 99.97
Does anyone have an idea where I can go on with the search ?
syslogto a remote host; kernel panics that won't get written to on-system logs (because of the risk of FS corruption) can still get sent over the UDP service to a remote log host, and recovered there. – MadHatter Jan 16 '14 at 15:500means don't wait, not don't reboot, and that just means there won't be any helpful messages left on the screen (as the reboot will clear them away). That's what the syslogging is to help with. – MadHatter Jan 16 '14 at 16:20