[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: sig11 on as



On Sat, Apr 08, 2000 at 09:44:14PM -0500, Adrianne Rutledge wrote:
> on 08/04/2000 9:36 PM, Ethan Benson at erbenson@alaska.net wrote:
> 
> *snip*
> 
> > i just recompiled my kernel to see, and no i had no problems at all,
> > what you describe sounds very much like `the signal 11 problem' you
> > should go read the document that describes it and how to debug it if
> > you have not already:
> > 
> > http://www.bitwizard.nl/sig11/
> > 
> > and yes you macos crashes could be a result of bad hardware too (as
> > sig11's usually are in compiling on linux) its just much harder to
> > notice bad hardware on macos since crashes are really are not
> > uncharacteristic.
> 
> The MacOS one is just a bug in IE5, its the only one that causes it, the
> rest of the programs don't, but its gotten consistant, I'm going to blank my
> PRAM, see if that fixes it... and unless its heat damaging something, I
> don't see any reason why stuff coulda gone bad...  But IE5 is the only
> program that causes it under MacOS, but then, its consistant and generally
> happens after a certain amount of time every time I try to run IE5.  Hrml...
> I know this is off topic, but it might help in debugging the problem...
> GaugePro doesn't say anything is wrong with my RAM.  I may play RAM roulette
> later tonight and see if it starts working, but it happens consistantly, not
> just every now and then, no matter how many RAM eating programs I've got
> loaded... I'm really starting to get scared that this processor overheated
> too... my room is very bad about that.  I recently sold a K6-3 because it
> ran at unacceptable temperatures in my room, and I had it underclocked to
> try and cool it down, the guy I sold it too accidentally overclocked it and
> it ran a good 30F cooler than it did underclocked in this room.  ;_;

read that doc anyway, it has alot of tips on tracking down these
problems, compiling stuff is very ram and processor intensive,
compiling the kernel is a good way to stress test hardware.  as for IE
it may be a bug, or it might be the only thing intensive enough to
trigger the problem.  signal 11s compiling stuff and in configure
script tests sounds exactly like what is described in the sig11 doc
above.


-- 
Ethan Benson
http://www.alaska.net/~erbenson/

Attachment: pgp2KW053YoQP.pgp
Description: PGP signature


Reply to: