LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Slackware - Installation (http://www.linuxquestions.org/questions/slackware-installation-40/)
-   -   Unable to handle kernel paging request Error (http://www.linuxquestions.org/questions/slackware-installation-40/unable-to-handle-kernel-paging-request-error-144618/)

EnigmaX 02-10-2004 07:59 PM

Unable to handle kernel paging request Error
 
I get the following error when trying to load install.1

Usb_control/bulk_msg: timeout
Unable to handle kernel paging request at virtual address e082c388 printing eip:
e082c388
*pde = 01598067
*pte = 00000000
Oops: 0000
CPU: 0
EIP: 0010:[<e982c388>] Not tainted

System:
1.2ghz Athlon AMD
512 DDR 2700 RAM
4GM Samsung Hard Disk

I have ran a memory test, it showed no errors.

Thanks for any and all help in advance =)

theofb 11-28-2007 07:03 PM

same here
 
I routinely get this same error when my system runs run-parts.

How do you run a memory test?

Alien_Hominid 11-29-2007 02:32 AM

http://www.memtest.org/

difrost 11-29-2007 04:18 AM

Is this full crash message from kernel? If not, please provide full output.

theofb 12-07-2007 08:58 PM

Thanks Alien. I was able to download and run memtest and it uncovered some issues with my memory. However intellectual curiosity is still getting the better of me.

This is not specific to run-parts but for any process. In fact I am watching this eip dump repeatedly on sh right now!

In my case this is sometimes a system crash and sometimes not. Usually I can get back my command line, other times not and I have to cold boot. This endless loop is new.

The only constant I think I'm seeing is the pde. If I remember right from my school daze, this is the address in execution and should in theory match what memtest finds (I don't remember the specific addresses.) Is this correct?

Does anyone know where I an get a breakdown of all the fields in the eip dump?

H_TeXMeX_H 12-08-2007 02:53 PM

If memtest reports a problem, then it is time to change out the broken RAM stick and get a new one. No other solutions exist.

theofb 12-11-2007 02:55 PM

uummm...ok...trying not to get sarcastic off that comment.

For my own personal growth...I was hoping someone knew where I could find a breakdown of these eip dumps. What each field represents etc...

uummm...never mind... can't help myself...

anyone know a good software solution or workaround for a cut power cord?

H_TeXMeX_H 12-11-2007 03:25 PM

I don't really get what you're trying to do. Are you trying to troubleshoot a broken RAM stick ? I hope you realize that is futile ... you machine is not running properly without fully functional RAM, so whatever data you will get will likely not mean much, it will likely be nonsense. And, even if it weren't nonsense, you wouldn't be able to do anything with it. Can you repair a broken RAM stick ? If you really want, try cleaning out the RAM slot, there may be dust that's clogging one of the pins (unlikely).

theofb 12-12-2007 10:55 PM

I
 
I'm not really trying to do anything.

I understand that failed memory is failed memory and there is no practical solution other than getting new sticks.

But because I understand hex values and instruction sets and stack traces and memory/register dumps, I would like to at least know what those various fields represent when my system takes a dump. There must be a reference somewhere and there must be a fellow slack user that knows where it is.

And just for argument sake...a real pro might could hack the kernel and tell the system to skip a bad block of addresses. Might come in handy during nuclear winter or something :)

H_TeXMeX_H 12-13-2007 12:01 PM

Oh, ok then, may bad.

Also, what brand are your RAM sticks ? If they are Kingston, then some people say that if you manage to boost the voltage going to the stick, then it'll work fine:
http://www.linuxquestions.org/questi...ddress-267029/

From my experience with Kingston RAM sticks ... I think I'll stick with Corsair or others.


All times are GMT -5. The time now is 08:54 AM.