fault detection

[Scott.Prive at storigen.com (Scott Prive)]

I assume you checked /var/log/messages, looked for the boot sequence, then stepped back in the logs a bit? Did it log a panic?

Using a stock kernel, I don't know what else to suggest. 

There is a "crash" patch to Linux, which upon a panic it will dump all memory to the swapfile, and on the next reboot it will extract a dump file to /var/log/dump. The dump may be too much information for you, but there is an analysys.* file which contains why the system died.

Until you know what happened, you could consider turning on syslogd remote logging, or setup a serial console and send the logs out on that to a nearby gen-purpose system. 

-Scott

-----Original Message-----
From: FRamsay at castelhq.com [mailto:FRamsay at castelhq.com]
Sent: Monday, August 19, 2002 3:42 PM
To: discuss at blu.org
Subject: fault detection


Does anyone know of any tools to help figure out why a box rebooted?  One
of our client boxes rebooted
over the weekend for no apparent reason.  The client claimed there was no
power outage, and a quick look
over the logs verifies the UPS didn't shut the computer down.  Also I
didn't see a shutdown or reboot request
in /var/log/messages.  So what tools do people use to figure out why a
Linux system crashed?

the system is running Redhat 7.2  kernel 2.4.9-13

               -fjr


Frank Ramsay
Systems Programmer
Castel, Inc
14 Summer St, 3rd Floor
Malden, MA 02148
(781) 324-0140 (voice)
(781) 324-0277 (fax)
Emal: framsay at castel.com


_______________________________________________
Discuss mailing list
Discuss at blu.org
http://www.blu.org/mailman/listinfo/discuss