From: Dave Airlie (airlied at domain csn.ul.ie)
Date: Wed 01 Aug 2001 - 00:08:51 IST
I'd go memory as the first option....
run the mem tester mentioned the other day..
Dave.
On Wed, 1 Aug 2001, John Gay wrote:
> I am starting to have major problems with my SmoothWall PC. I am fairly sure
> that these problems are hardware related because if I reset or re-boot the PC
> without giving it plenty of time, The kernel either hangs or panics. These
> chrashes usually happen when I'm on-line and down-loading large files, but
> the other day I had closed the connection and it crashed after the connection
> was closed. I was able to copy only part of the error info before the screen
> went blank.
> <error>
> Unable to handle NULL pointer dereference at address 0000001d
> current-> tss.cr3=00101000, %cr3=00101000
> pde=00000000
> Oops:0000
> CPU:0
> EIP:0010:[<c01252b4>]
> EFLAGS:00010202
> EAX:00000001 EBX:00000001 ECX:00000000 EDX:a0680000
> ESI:00000000 EDI:00000001 EBP:00001000 ESP:c0095???
> </error>
> This was when the screen went blank. I've also seen the screen filled with
> the 'Unable to handle NULL pointer dereference' error and also with a full
> screen of various register data. As I said, this usually happens when I'm
> downloading large files and at first I thought it might be a 'cracker' but
> the fact that turning the system off and back on within several minutes would
> not boot, I figure it must be hardware related. There is no info in
> /var/log/messages. The last entry before the next kernel boot is just <mark>.
>
> My question is, Which piece of hardware is it likely to be?
> I've a 100Mhz Overdrive Pentium processor. I could probably put the original
> 486 back into it. I've got several SIMM's that I could change and I'm
> currently downloading the latest ISO for Smoothwall, since it's only 23M.
> Should I try running the memory tester I saw mentioned in a recent mail? or
> the CPU cooker to see which part is falling over? I was also thinking of just
> putting the harddrive, NIC and modem into another box but I'd realy like to
> know which part is faulty and scrap it. I don't know if any of the above info
> sheds any linght on the subject, but I thought I'd include it incase it could
> identify the culprit.
>
> Thanks for any info.
>
> John Gay
>
>
-- David Airlie, Software Engineer http://www.skynet.ie/~airlied / airlied at domain skynet.ie pam_smb / Linux DecStation / Linux VAX / ILUG person
This archive was generated by hypermail 2.1.6 : Thu 06 Feb 2003 - 13:11:24 GMT