Re: Ooops galore on 2.2.10-ac11 (with info now ;)

CaT (cat@zip.com.au)
Thu, 22 Jul 1999 10:32:25 +1000


On Wed, Jul 21, 1999 at 08:03:53PM -0400, Steven N. Hirsch wrote:
> On Wed, 21 Jul 1999, CaT wrote:
>
> > Well. this isn't going to be of much use because the oopses were
> > failed to be recorded in any log AGAIN. :/
> >
> > The bestr I can do is describe what I did

Damn. Thought I postponed that. My mistake.

> Actually, the best you can do is to rebuild the kernel with serial console
> enabled. Then connect it to another box with running a terminal program
> in text-capture mode. Works for me.

You supply me with either the cash or the box to do it with and
I'll happily do it. :) Not everyone has more then one box to play
with. Otherwise I'll -actually- describe whereabouts it happened. ;)
Maybe it'll give someone a clue or a hint or something.

Basically I was doing a dummy burn with cdrecord 1.8a23. The first
attempt kinda failed. It froze with the burner being left with
the busy light on. I couldn't eject the cd so I thought that another
run at cdrecord might reset the burner and things would be ok. This
is where it blew up. A whole bunch of rolling oopses (these suckers
were huge too. the calltrace kept getting bigger. maybe something
recursive went gaga?) and then a stop. nothing but alt-sysreq
would work. Since I noticed that the process field in the oops
was corrupted I tried alt-sysreq-T to see the process list. That's
when things exploded again and it looked like the kernel was stuck
in an infinate oops loop.

The way I stopped that was basically with a resethard reset. the
vulcan nerve pinch didn't work. Right now I'm on 2.2.7ac4 which
is the last kernel I consider stable. Everything is rosy again
and I performed a successful burn later that night.

System info:
P-II, 128meg RAM, IDE HD
SCSI Panasonic 4x burner
TekRam dc310 Fast SCSI-2 card
Glibc 2.1.1
egcs 1.1.2
binutils 2.9.4.0.6

2.2.7ac4 and 2.2.10ac11 were both compiled with -march=pentiumpro

Related kernel options are:
SCSI Support
SCSI Disk Support
SCSI CD-ROM Support
SCSI Generic Support
Verbose SCSI error reporting
SCSI logging facility (not turned on at the time AFAIK)
SYM53C8XX SCSI support
default tagged command queue depth: 8
maximum number of queued commands: 32
synchronous transfers frequency in MHz: 20
assume boards are SYMBIOS compatible

2.2.10ac11 also had:
Logical volume manager (LVM) support
LVM information in proc filesystem

Hope this helps. If you need more info and can tell me how to get it
(without involving external h/w which I don't have :/) I'll try
and get it for you.

-- 
CaT (cat@zip.com.au)                       URL: http://www.zip.com.au/dev/null

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