Very Computer

by **erenio_rei..** » Tue, 08 Jul 1997 04:00:00

Here's the trace output:

Quote:> trace

STACK TRACE FOR PROCESS 3:
STKADDR FRAMEPTR FUNCTION POSSIBLE ARGUMENTS
e0000d7c e0000db8 prf_task (e,f0ff7ca8,1,1)
e0000dc0 e0000dd8 cmn_err (3,f010531c,e,f1368168)
e0000de0 e0000e08 k_trap (e0000e14)
Trap e e0000e14 cmntrap from f00c4928 in bdflush
ax:fffeeca8 cx: 0 dx: 0 bx: 3e8 fl: 10283 ds: 160 fs: 0
sp:e0000e44 bp:e0000e78 si:f1368168 di:f0ff7ca8 err: 0 es: 160 gs: 0
e0000e1c e0000e78 bdflush (0,0,0,0)

The system is really always had errors ever since get switched it to a Pentium
Pro. Before, we had run on a DEC 486/33 for years, very solidly.

Thanks...E.Reiner

______________________________ Reply Separator _________________________________
Subject: Re: Trap 14 on ODT3.0 on HP PentPro 200

Date: 7/4/97 6:24 AM

> I am running ODT 3.0 on an HP VectraVA 200 MHz. The system is dying about
> once a week. Here is the results on a crash , panic command:

> ******************************************************************

> dumpfile = /dev/swap, namelist = /unix, outfile = stdout
> > System Messages:

> mem: total = 32380k, kernel = 7208k, user = 25172k

> Panic String: k_trap - Kernel mode trap type 0x%x

> Kernel Trap. Kernel Registers saved at e0000e14
> ERR=0, TRAPNO=14
> cs:eip=0158:f00c4928 Flags=10283
> ds = 0160 es = 0160 fs = 0000 gs = 0000
> esi= f1368168 edi= f0ff7ca8 ebp= e0000e78 esp= e0000e44
> eax= fffeeca8 ebx= 000003e8 ecx= 00000000 edx= 00000000

> Kernel Stack before Trap:
> STKADDR FRAMEPTR FUNCTION POSSIBLE ARGUMENTS
> e0000e44 e0000e78 bdflush (44a990,0,0,0)

Would you also run the crash "trace" command? And try to write down on
paper what is printed on the screen at the time of the panic. The above
traceback is too short to do much with.

Somewhere, from one of those sources, I hope to see a value for the
register CR2, which is the address which was being accessed when the
panic occurred.

Other details: was the system running without error for a while, then
this started happening? Or is it a new install and has been doing this
since first installed? If it just started happening -- what did you
change just before?

- Show quoted text -

Quote:>Bela<

by **Bela Lubki** » Thu, 10 Jul 1997 04:00:00

> > I am running ODT 3.0 on an HP VectraVA 200 MHz. The system is dying about
> > once a week.

> Here's the trace output:

> > trace

> STACK TRACE FOR PROCESS 3:
> STKADDR FRAMEPTR FUNCTION POSSIBLE ARGUMENTS
> e0000d7c e0000db8 prf_task (e,f0ff7ca8,1,1)
> e0000dc0 e0000dd8 cmn_err (3,f010531c,e,f1368168)
> e0000de0 e0000e08 k_trap (e0000e14)
> Trap e e0000e14 cmntrap from f00c4928 in bdflush
> ax:fffeeca8 cx: 0 dx: 0 bx: 3e8 fl: 10283 ds: 160 fs: 0
> sp:e0000e44 bp:e0000e78 si:f1368168 di:f0ff7ca8 err: 0 es: 160 gs: 0
> e0000e1c e0000e78 bdflush (0,0,0,0)

> The system is really always had errors ever since get switched it to a Pentium
> Pro. Before, we had run on a DEC 486/33 for years, very solidly.

Ok, so that's what changed. There is something wrong with the hardware
on the new machine. It might be bad RAM, bad cache, something wrong
with the host adapter, etc.

Run hardware diagnostics.

Try swapping out parts if the diagnostics don't identify the failing
hardware.

- Show quoted text -

Quote:>Bela<