UNIX for Dummies Questions & Answers

Hi ,
My Sun Machine Rebooted by itself 2 days back . Its running fine now , But i wanted to find out wat caused it to reboot...
This is wat the var/adm/messages show .
Kern.notice:- System booting after fatal error FATAL...
Wat causes this message ...
And wat tasks should i do to ensure it dosent happen again

Thanx
DPAI

This type of problem is typical of some
sort of hardware problem (but not always).
Could be memory (in the lower address range) or
disk error in the swap region or maybe even a
loose CPU board. The possibilities are numerous.
Can you determine a "best guess" as to what the
machine was doing just before the error?
You may also want to run system diagnostics
which may give you some indication.

No , I am not sure wat was goin on when it rebooted . I mean it was idle , Nobody had logged into the m/c for quite some time

You gotta love a mistery
I can tell you that over the years, I have
expierenced these type of "gremlins" on
several occasions on all types of hardware.
In one case, we expierenced unknown crashes
about once every 30 or 40 days.
I just started pulling each board
one by one to examine them closely and I found
one with a deep gash across several of the onnector pins at the card edge. I actually
dabbed a bit of solder on each scratched
connector and the problem was never expierenced
again.
When ever I see these unexplained crashes, I
just shut the machine down and go into it and
reseat all the boards, memory and connectors
and the results are more often that the problem
never surfaces again. That's not to say you
should skip running diags. though. Running them
can point you in the right direction if not
actually find the problem.

Good hunting

Ya , Its sure is a mystery ...
I tried to run prtdiag -v ...
But it shows nothing Just No failures found on system

Are ther any other diag routines i can use ..

How dirty is the motherboard and hardware? Sometimes the years of dust and other grime can get between connections, especially host adapters and other end cards.

Is the inside of the machine clean? Have you taken the parts apart, cleaned and reassembled (if old)?

Also, is this a SCSI bus? Wierd things can happen with SCSI.

At the ">" (system monitor) prompt, type n
That will get you into the on-board diagnostics
in the PROM. You can run tests like Ethernet,
SCSI, Memory and Disk diagnostics from here.
You can also run tests from the SunDiagnostic
Executive.
Refer to your SunDiagnostic Executive
User's Guide for the SPARCstations for more information. If I can think of any more, I'll
let you know.