🚀 go-pugleaf

Just before I left for a long vacation this last summer, I had to
move all my cards from one RT chassis to another because my motherboard
went bad.  After doing that, I found that the machine was crashing
regularly several times a day with "panic: machine check."  The
crash usually happened when the machine got busy.  This was bizarre
because the machine had been running constantly for a year and a half
without any problem.

Now, while I was away, there was a AOS43 patch to ca/trap.c which under
certain conditions simply printed "spurious machine check ignored".
I missed any discussion which accompanied this.  This apparently solves
the problem of crashes, but it is the case that I'll get anywhere from
1 to 100 of these benign messages a day now, and this has occurred with
my boards moved into 3 different chassis/motherboard combinations.

What I don't understand is why this "suddenly" started happening,
why it's continued to happen, what are the conditions under which
this happens, and how it is distinguished from a "non-spurious
machine check".

Can anyone shed some light on this?

-- 
Steve Dyer
dyer@ursa-major.spdcc.com aka {ima,harvard,rayssd,linus,m2c}!spdcc!dyer
dyer@arktouros.mit.edu, dyer@hstbme.mit.edu