Subject: Re: Decoding machine checks...
To: None <firstname.lastname@example.org>
From: Sean Davis <email@example.com>
Date: 09/12/2003 21:09:07
On Fri, Sep 12, 2003 at 08:42:10PM -0400, firstname.lastname@example.org wrote:
> I've got a 164LX. I've been getting "processor correctable error"
> messages in the past few days. Now mozilla seems to be able to
> crash my box multiple times a day. Big bummer.
> >From the crash that happened just a few minutes ago I got this:
> 97 496 495 494 493 492 491 490 489 488 487 486 485 484 483 482 481 480 479 478 4
> 77 476 475 474 473 472 471 470 469 468 467 466 465 464 463 462 461 460 459
> unexpected machine check:
> mces = 0x1
> vector = 0x660
> param = 0xfffffc0000006068
> pc = 0xfffffc00004fc8c8
> ra = 0xfffffc0000309abc
> code = 0x203
> curproc = 0xfffffc0005124e68
> pid = 467, comm = mozilla-bin
> panic: machine check
> dumping to dev 8,9 offset 4191
> dump device not ready
> Now, I know why the dump device isn't ready: It was dumping because
> of a machine check. When it got a check in the middle of the dump it
> couldn't deal.
> I think the first crash had a vector of 0x670. I didn't write down
> the whole message because I thought I could get it from the dmesg.
> Oops, overflowed the buffer. Is there a ddb command to reboot without
> a dump? I'll have to check.
'reboot' in DDB will reboot without a dump, IIRC.
> Anyway, what do these machine checks mean? Is this a bad CPU? Bad board?
> Bad memory? I've got a gig of memory, should I yank half of it?
Machine checks typically mean hardware is dying somewhere, from everything
I found a URL a while back that had lots of information on misc. machine
check codes and what they meant, but
A '660' vector usually means `memory ECC error'.
> Are there any public docs so I can just look this up myself?
Yes, some, but I can't remember the URLs and I didn't save them anywhere,
/~\ The ASCII
\ / Ribbon Campaign Sean Davis
X Against HTML aka dive
/ \ Email!