[TriLUG] Kernel Panics.......

[Aaron S. Joyner]

Brian Henning wrote:

> Hi Guys and Gals,
>   I have a production server that has been kernel-panicking with very 
> aggrivating regularity over the past week....
>
> Can someone verify if that's caused by a particular sort of hardware 
> failure?  This server was built from aging parts.  I suspect perhaps 
> another bad spot in RAM or a hard drive headed for failure, but I 
> don't really want to keep taking shots in the dark if there's 
> something else at work here that someone can identify.

Usually hardware failures are random problems, they don't repeat in any 
nice or sensible manner, or result in the same types of error messages 
repeatedly.  That's not always true, but it's a good rule of thumb.  It 
sounds like you're not having precisely the same error from time to 
time, which does indeed point towards hardware.  A better question might 
be, has anyone seen this problem associated with a particular software 
problem, and if *not*, then look towards the hardware.  Personally, I'd 
say it sounds like a locking issue in the kernel - which unless you're 
using a poorly-compiled kernel, or have made some odd modifications, 
then it's very likely it's a hardware problem.

> ...Oh yes, here are the system stats:
> CPU: P-III 850MHz

Have you checked for bad capacitors on the motherboard?  They'll appear 
as bulging tops, perhaps leaking a brown or black caustic-looking 
solution (the electrolytic material).  That board is about the right 
age, and from the time period when lots of manufacturers were using a 
bad batch of caps from a manufacturer in Taiwan.  I'd wager a small sum 
of coins that will turn out to be your problem.

Aaron S. Joyner