48hr panics: Fatal trap 19: non-maskable interrupt trap while in kernel mode

[John Baldwin]

On Monday 04 October 2004 11:21 am, othermark wrote:
> I get these about once every 48hrs for about 2 weeks on a box I build
> -current once or twice a week.
>
> First panic here is from Wednesday, the panic at the end of this message
> is from Friday morning's -current.  Is there any problem currently
> with dumpdev=auto and savecore?  Note the silly lockheld panic when I call
> doadump.

I'm guessing you are getting an NMI due to an ECC error while it is zero'ing 
memory.  Perhaps your memory is going bad or your box is overheating?

-- 
John Baldwin <jhb at FreeBSD.org>  <><  http://www.FreeBSD.org/~jhb/
"Power Users Use the Power to Serve"  =  http://www.FreeBSD.org