Memory/filesystem corruption

[Robert Arendt]

>> Camilo Mesias wrote:
> [...]
>>> I'm seeing messages like these:
>>>
>>> Uhhuh. NMI received for unknown reason b0 on CPU 0.
>>> You have some hardware problem, likely on the PCI bus.
>>> Dazed and confused, but trying to continue
>>>

This message comes from the kernel, from reading cpu port 61H.
It can have different meanings for different motherboards
and vendors, and is rarely adequately documented.  Interesting
discussion here: http://www.gossamer-threads.com/lists/linux/kernel/969550

For us this error turned out to be the interaction of a pair
of PCI cards.  We regressed a prom on the card, and the problem
vanished.  Basically, the kernel messages won't tell you much.
Assuming the board's OK, it's probably a board - card interaction.

Good luck
-Bob