/var/log/messages(dmesg) ?‏

hadimotamedi · 02-22-2010, 05:44 AM

Dear All
My Debian server got hung and when I tried to power cycle it , I checked its /var/log/messages & /var/log/dmesg with respect to the time of fault occurance to see what was really happened . But its 'messages' log does not show anything recorded for about 30 minutes before its recovery after power cycling . Can you please let me know where I can check to find the root cause of failure ?
Thank you

evo2 · 02-22-2010, 06:46 AM

Sometimes hardware failures will occur in such a way that nothing will/can be written to the log files. Also, when you reboot, the time stamps on the log files will be updated which can make it difficult to work out what was the last thing written before the crash. One thing you can do after such a lock up is boot to a live/rescue cd, and then mount the file system read only: this should make it easier to see what was going on just before the crash.

Cheers,

Evo2.

hadimotamedi · 02-22-2010, 07:00 AM

Quote:

Originally Posted by evo2

Sometimes hardware failures will occur in such a way that nothing will/can be written to the log files. Also, when you reboot, the time stamps on the log files will be updated which can make it difficult to work out what was the last thing written before the crash. One thing you can do after such a lock up is boot to a live/rescue cd, and then mount the file system read only: this should make it easier to see what was going on just before the crash.

Cheers,

Evo2.

Thank you for your reply . Can you please provide more details on this procedure ? After booting from /rescue mode and mounting the file system read only , where I can check for what was going on before the crash ?

evo2 · 02-22-2010, 07:30 AM

Start with the files in /var/log/ you may be able to find something that you would otherwise overlook if the machine had booted up again normally. Another place to look in is /tmp/, which will normally be completely deleted on a reboot.

Cheers,

Evo2.

onebuck · 02-22-2010, 08:33 AM

Hi,

You could try a another console while the system is still up. Meaning try to login via another console then look at the system messages or information to diagnose without re-booting. Or attempt to login via 'ssh' to console then look at the system conditions.

hadimotamedi · 02-23-2010, 12:06 AM

Quote:

Originally Posted by onebuck

Hi,

You could try a another console while the system is still up. Meaning try to login via another console then look at the system messages or information to diagnose without re-booting. Or attempt to login via 'ssh' to console then look at the system conditions.

Thank you . But my Red Hat server hard drive LED was continously lit and no login was possible on any console/SSH session . I want to find why this is occured .