Linux 2.2.23-rc2 & an MCE

Linux 2.2.23-rc2 & an MCE

[Rasmus Andersen]

[-- Attachment #1: Type: text/plain, Size: 404 bytes --]

Hi,

I just had an MCE on my aging PPro 200. Before I go out to
buy a replacement I would like to hear if it could be
caused by anything other than the CPU. Googling a bit
gave some indications that sometimes other HW might report
failure through this method.

The MCE (hand copied):

Machine Check Exception: 000000000000004
Bank 4: b200000000040151
Kernel panic: CPU context corrupt

Regards,
  Rasmus

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: Linux 2.2.23-rc2 & an MCE

[Pavel Machek]

Hi!

> I just had an MCE on my aging PPro 200. Before I go out to
> buy a replacement I would like to hear if it could be
> caused by anything other than the CPU. Googling a bit
> gave some indications that sometimes other HW might report
> failure through this method.
> 
> The MCE (hand copied):
> 
> Machine Check Exception: 000000000000004
> Bank 4: b200000000040151
> Kernel panic: CPU context corrupt

Is not it trying to tell you about bad ram?



-- 
Worst form of spam? Adding advertisment signatures ala sourceforge.net.
What goes next? Inserting advertisment *into* email?

Re: Linux 2.2.23-rc2 & an MCE

[Felipe W Damasio]

Pavel Machek wrote:
>>The MCE (hand copied):
>>
>>Machine Check Exception: 000000000000004
>>Bank 4: b200000000040151
>>Kernel panic: CPU context corrupt
> 
> Is not it trying to tell you about bad ram?

	Could be, though this looks like a Instruction fetch error from the 
Level 1 cache, doesn't it? If so, it could be caused by a faulty processor.

	Is this the first time it happened? Could you please check your logs 
and send any more MCE error codes?

	Thanks,

Felipe

Re: Linux 2.2.23-rc2 & an MCE

[Rasmus Andersen]

[-- Attachment #1: Type: text/plain, Size: 924 bytes --]

On Thu, Nov 28, 2002 at 11:03:04PM +0000, Felipe W Damasio wrote:
> Pavel Machek wrote:
> >>The MCE (hand copied):
> >>
> >>Machine Check Exception: 000000000000004
> >>Bank 4: b200000000040151
> >>Kernel panic: CPU context corrupt
> > 
> > Is not it trying to tell you about bad ram?
> 
> 	Could be, though this looks like a Instruction fetch error from the 
> Level 1 cache, doesn't it? If so, it could be caused by a faulty processor.
> 
> 	Is this the first time it happened? Could you please check your logs 
> and send any more MCE error codes?

Hi,

I have nothing in my logs but have had three more chrashes since
my first report. Two of them I couldn't inspect since I was at
work (the machine is at home) and had my girlfriend boot the box,
but the last one was identical to the reported one.

I am getting a new processor now and hope that'll do it.

Thanks for your comments,
  Rasmus

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: Linux 2.2.23-rc2 & an MCE

[Felipe W Damasio]

Rasmus Andersen wrote:
> I have nothing in my logs but have had three more chrashes since
> my first report. Two of them I couldn't inspect since I was at
> work (the machine is at home) and had my girlfriend boot the box,
> but the last one was identical to the reported one.
> 
> I am getting a new processor now and hope that'll do it.

	Since the MCE code is reporting a instruction fetch error from the 
level 1 cache, it could be a bad ram problem...

	Could you try and run the memtest86 on your memory card(s) first (maybe 
in a different machine)?

	Kind Regards,

Felipe

Re: Linux 2.2.23-rc2 & an MCE

[Randy.Dunlap]

| > I just had an MCE on my aging PPro 200. Before I go out to
| > buy a replacement I would like to hear if it could be
| > caused by anything other than the CPU. Googling a bit
| > gave some indications that sometimes other HW might report
| > failure through this method.
| >
| > The MCE (hand copied):
| >
| > Machine Check Exception: 000000000000004
| > Bank 4: b200000000040151
| > Kernel panic: CPU context corrupt

Rasmus,

If you haven't already done so, you should check out the
MCE decoder from Dave Jones at
  http://www.codemonkey.org.uk/cruft/parsemce.c/

-- 
~Randy

Re: Linux 2.2.23-rc2 & an MCE

[Rasmus Andersen]

[-- Attachment #1: Type: text/plain, Size: 560 bytes --]

On Sun, Dec 01, 2002 at 06:39:41PM -0800, Randy.Dunlap wrote:
> | > Machine Check Exception: 000000000000004
> | > Bank 4: b200000000040151
> | > Kernel panic: CPU context corrupt
> 
> Rasmus,
> 
> If you haven't already done so, you should check out the
> MCE decoder from Dave Jones at
>   http://www.codemonkey.org.uk/cruft/parsemce.c/
> 

That gave me: 

Status: (4) Machine Check in progress.
Restart IP invalid.

Not sure what to make of that, though. Further comments
always welcome. And thanks for the pointer.

Regards,
  Rasmus

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]