Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

One of my nfs servers disappeared this morning from monitoring. I checked it out and the console was hung and non-responsive, and it was apparently crashed.

I power-cycled and checked out the syslog, and it very much appears to have just crashed with no indication of why.

Is there any kernel or debug settings that I can apply, to try and trap any future re-occurence or this problem. (or any recommendation on how to proceed)

share|improve this question
add comment

2 Answers

up vote 3 down vote accepted

If it completely hard-crashed, nothing in the logs, I'd strongly suspect it was hardware related. I'd reseat memory, check that the fans are running properly to cool the server, and if it's a server-grade system, use the diagnostics to check the equipment out (I know that Dell servers usually have a series of tests that can be run, but it depends on the model whether it's in BIOS or a startup partition or a bootable CD)

Rarely, rarely, rarely have I had Linux crash completely in an unresponsive manner without a kernel dump or something in the logs. I have had systems go nuts due to a dying controller, memory creeping, or something else hardware-related, which can easily do what you're describing.

share|improve this answer
    
yeah. I have rarely seen such a complete cessation of activity –  Tom H Jul 19 '10 at 12:39
add comment

Check your hardware, as Bart said. Also, sometimes an unresponsive machine may be on that state because of a stupidly large load. I saw some mailservers do that. Check your network, NFS can crash badly if the networks goes away when it is doing something.

If you ever need to do that to a machine again, remember the Magic Sysrq key and the Raising Elephants Is So Utterly Boring phrase. The ALT+SysRQ+ can do wonders on a linux box that is to all aspects dead. The prase is to remember the commands to use with ALT+SysRQ:

R: take control of the keyboard
E: sends SIGTERM to all processes
I: sends SIGKILL to all processes
S: Sync (flush caches to disk, very important)
U: remount all FS read only
B: reboot!
share|improve this answer
    
thats really interesting, I didn't know about the sysrq key. I am looking forward to something crashing again so I can try it ;-) –  Tom H Jul 20 '10 at 9:14
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.