On Mon, 25 Mar 2002, Jan Schmidt wrote:
> > > I had a similar problem years ago, but that was giving me SIG11's and also
> > > confusing the compiler so code was breaking as it was compiling. Not sure
> > > what an error 139 is so this may not be relevant. Anyway, basically it was
> > > due to my CPU overheating. I pulled the CPU and found no heatsink grease.
> > > dabbed a bit on and also down clocked it. It worked without fault for years
> > > afterwards.
> >
> > The motherboard has an on-board temperature monitor. According to it, both
> > CPU's are running around the 40 degree mark - which, since they're both
> > P3's, isn't unacceptable.
>
> Return code 139 *is* a SIG11/SEGV. The default exit codes for signals which
> terminate a program is 128+the signal no.
This much I figured out for myself. :-) Which is why I started looking at
the RAM.
> You are running an SMP system. Which kernel are you running? In the past, I
> have occasionally had trouble with isolated kernel versions and SMP. It may
> be worth booting a non-SMP kernel and trying it. Check /var/log/messages
> for any kernel oopses as well.
I'm currently running 2.4.7 - and I would boot to a uniprocessor kernel,
except that I can't, because I stupidly put everything except / {/usr,
/var, /home} on reiserfs partitions, and the only uniprocessor kernel I've
got doesn't have reiserfs in it except as a module - and it can't find the
modules
> What else? If you had bad RAM, I suppose it is possible that the files
> installed on your hard drive were corrupted when they installed and may need
> reinstalling? That one is a wild throw, though.
That's my next option - reinstall the mongrel completely. I hesitate to do
that, though, because I spent a lot of time getting things just right, and
I don't want to lose my setup.
DaZZa
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://lists.slug.org.au/listinfo/slug
