Re: PROBLEM: Linux consistently crashes running grip. (continued)

From: Kenneth Corbin (kencx@peak.org)
Date: Wed Sep 04 2002 - 08:51:53 EST


On Wednesday 04 September 2002 12:02 am, Nuitari wrote:
> On Tue, 3 Sep 2002, Kenneth Corbin wrote:
> > Long description of same problem. Still continuing after dumping the
> > nvidia drivers and upgrading to the latest Redhat kernel. Anything else
> > I can do?
> >
> > I am not subscribed to this list, please cc me with any response.
> > Thanks in advance for wading through this.
> >
> > 1. Linux consistently crashes running grip.
> >
> > 2. Grip 3.0.1-1 is a graphical X frontend for a variety of CD ripper and
> > MP3 encoder programs. When it is only ripping CD's everything is OK.
> > But when it is ripping and encoding tracks, the system will die shortly
> > after the ripping operation finishes while the encoding operation is
> > going on. The time till death is variable, it has ranged from 5 seconds
> > to 5 minutes on one occasion, but it always crashes. It happens with two
> > different encoders (notlame and oggenc) and two different rippers
> > (cdparanoia and cdda2wav) The quickest way to induce a failure is to
> > make one pass ripping tracks and a second pass asking it to rip and
> > encode. It is smart enough to realize the tracks have already been
> > ripped and doesn't do much beyond checking that they are all accounted
> > for. But my system still crashes.
> >
> > Gnu C 2.96
>
> Did you try testing your ram for any defects (even if it means trying
> various combinations of ram) ?
> The only times I had similar problems was when I had faulty ram on my
> motherboard or (in 1 case) a really defective onboard ide controller.
>
> Also be careful with Gcc 2.96, there is a lot of problems related to it.

OK, went looking for hardware diagnostic tools. memtest86 ran all night
without reporting any problems. But cpuburn-1.4 burnK7 locked up almost
immediately. Interestingly it seems to do OK in single user mode, but if
anything else is running things go south very quickly. Usually it is a
silent lockup. In one case I got an Ooops:0002. Indications are pointing
toward some kind of hardware problem. Looks like I need to take this to
someone who has some real diagnostics, or the resources to swap out parts
until things start working. I'll let you know how this turns out.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sat Sep 07 2002 - 22:00:21 EST