[linux-lvm] Server Crashes Sometimes

[Stuart D. Gathman]

On Sun, 3 Apr 2011, Jonathan Tripathy wrote:

> still respond to ping), and dmseg is flooded with
> 
> Uhhuh. NMI received. Dazed and confused, but trying to continue
> You probably have a hardware problem with your RAM chips
> ...
> This issue is very rare, and has only happened to me maybe 3 times over the
> past 7 months, each time being when I issued an LVM command
> 
> Has anybody experienced this before?

Yes.  The message means just what it says.  NMI is a hardware interrupt
usually reserved for machine errors such as an uncorrectable memory error.
In my case, it was a defective PCI card (with USB ports) raising the NMI.
(Which I determined by process of elimination.)  Note that many system buses,
including PCI, have error checking and will raise NMI on failure.

I have heard of hardware that raised NMI in normal operation as a kind
of highest priority interrupt.  However, such hardware is generally
equivalent to broken.

--
 	      Stuart D. Gathman <stuart at bmsi.com>
     Business Management Systems Inc.  Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flammis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.