Re: NMI errors in 2.0.30??, High Availability-Linux

Samuli Kaski (samkaski@cs.Helsinki.FI)
Mon, 28 Apr 1997 19:31:39 +0300 (EET DST)


On Mon, 28 Apr 1997, Rogier Wolff wrote:

> yuri mironoff wrote:
> > On Fri, 25 Apr 1997, Stephen Costaras wrote:
> >
> > > Anyone else notice anything w/ Tyan boards & ECC interacting with the kernel?
> > > (I'm assuming it's probably some kind of reporting problem as the posts that I've
> > > seen are all tyan based. I'm running Tyan's Award BIOS v3.03.
> >
> > Something like that. I've started getting SIGNAL 7 from GCC since
> > upgrading to 2.0.30. This happens when I stress test the machine.
> > (4 concurrent kernel builds with -j8) I'm not getting any NMIs - just
> > SIGNAL 7. I've also gotten a few "Couldn't get a free page." Could that
> > have anything to do with it?
>
> "Couldn't get a free page" means that you are stressing your system to
> the point that the kernel is running low on memory.
>
> The signal 7 sounds to me that you have bad RAM in your system.

Or that you are pushing the limits of your RAM. I had a few signal 7's
when I ran 60ns FP-RAM as 55ns on my Intel Zappa board running
MR BIOS. Same thing happened to me when I was running my p5-75 oc'd
to 120. Since moving on to 60ns EDO-RAM (it runs as 55ns) and clocking
down to 100 (thats about a year ago, or so) I haven't seen any of
signal 7's nor 11's.

--
Samuli Kaski, samkaski@cs.helsinki.fi
Department of Computer Science, University of Helsinki, Finland.