Debian stopped responding all of a sudden

tux75 · 02-18-2014, 03:19 PM

I got a home server setup with Debian 6 which stopped/crashed and stopped responding. It's a headless server with SSH access only.

After waiting for a day I did a hard reboot and the system started working again. Really like to know what caused this?. I am using it without any issues for almost 3 years now.

I checked the syslog and messages logfiles and didn't find anything. Got a cron job which reports the IP address every 5 minutes and according to that it stopped around 8.15am on 17th Feb. The remaining entries are from today.

these are the entries from dmesg, I assume PWRB is for pressing the power button and PWRF for force reboot. Don't know if the entries just before that make any sense. Is there any other place where I can find any information?

Code:

[    1.945052] sd 2:0:0:0: [sda] Attached SCSI disk
[    1.945533] sd 2:0:1:0: [sdb] Attached SCSI disk
[    2.031996] e1000: eth1: e1000_probe: Intel(R) PRO/1000 Network Connection
[    2.503866] PM: Starting manual resume from disk
[    2.503876] PM: Resume from partition 8:7
[    2.503880] PM: Checking hibernation image.
[    2.504143] PM: Error -22 checking image file
[    2.504147] PM: Resume from disk failed.
[    2.542737] EXT3-fs: INFO: recovery required on readonly filesystem.
[    2.542744] EXT3-fs: write access will be enabled during recovery.
[    2.651556] kjournald starting.  Commit interval 5 seconds
[    2.651577] EXT3-fs: recovery complete.
[    2.663125] EXT3-fs: mounted filesystem with ordered data mode.
[    4.032699] udev[381]: starting version 164

[    4.279382] input: Power Button as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0C0C:0                                                                             0/input/input0
[    4.279402] ACPI: Power Button [PWRB]
[    4.279608] input: Power Button as /devices/LNXSYSTM:00/LNXPWRBN:00/input/inp                                                                             ut1
[    4.279622] ACPI: Power Button [PWRF]

bigrigdriver · 02-18-2014, 03:31 PM

From the time stamp on your post, I'm guessing you are in the Eastern U. S. time zone. Did your server experience a power outage or fluctuation as a result of weather conditions?

tux75 · 02-18-2014, 03:35 PM

No, the system is located in Central Europe.

fuorviato · 02-18-2014, 03:43 PM

Do you have syslog installed and enabled? If so. Do

Quote:

tail -n 100 /var/log/syslog

.
Secondarily, you can just browse the log file for the suspected hour and date.

tux75 · 02-18-2014, 04:31 PM

Code:

Feb 17 08:15:01 localhost /USR/SBIN/CRON[6802]: (root) CMD (/home/user/dae/ipdaem.cron.sh)
Feb 17 08:17:01 localhost /USR/SBIN/CRON[6811]: (root) CMD (   cd / && run-parts --report /etc/cron.hourly)

Feb 18 18:11:24 localhost kernel: imklog 4.6.4, log source = /proc/kmsg started.
Feb 18 18:11:24 localhost rsyslogd: [origin software="rsyslogd" swVersion="4.6.4" x-pid="1348" x-info="http://www.rsyslog.com"] (re)start
Feb 18 18:11:24 localhost kernel: [    0.000000] Initializing cgroup subsys cpuset
Feb 18 18:11:24 localhost kernel: [    0.000000] Initializing cgroup subsys cpu
Feb 18 18:11:24 localhost kernel: [    0.000000] Linux version 2.6.32-5-686 (Debian 2.6.32-46) (dannf@debian.org) (gcc version 4.3.5 (Debian 4.3.5-4) ) #1 SMP Sun Sep 23 09:49:36 UTC 2012

In syslog the last is for a cron job which executes every 5 mins, this script is there for almost an year. I don't know what happened after 8.17am. Entries from 18th Feb are after the reboot.

The only suspect is BtSync which I installed the previous day and was still running. This is the only major change to this system.

jlinkels · 02-18-2014, 05:14 PM

I was a transmitter technician for many years. High power shortwave transmitters are very complicated, need a lot of maintenance and anything supernatural can happen. We had one technician among us, Mr. Coffi, who at every fault yelled: "first and forever check the area where you worked last". He was right so often that after 20 years or so, this became known as "Coffi's law".

In your case, if it is not BtSync, it is one of your cooling fans. Or, less probable, a hardware error

jlinkels