[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
OK... here's an ARGH update. I have swapped out memory with another 256MB/pc
133. I changed harddrives also. Same problem. The box, sitting at the prompt
or login, just dies. Screen goes black and the caps and scroll lock keys
flash. (Please tell me that this isnt some funky suspend mode and I'm just a
moron). However, I have yet to see the kernel panic again, but that doesn't
mean its not right around the corner! Could a faulty ethernet card be doing
this, kinda like a short or something? Every time I boot up it gives me
error:1 overruns:1 dropped:1. All my network settings are correct, nearest I
Anyway, I will continue swapping part out until I'm either out of them, or
it solves the problem. Thanks again everyone for all your help.
Villanova School of Law
ps - Toby, if you read this, it's Roberts. How the hell are ya?
From: "mike.h" <email@example.com>
Subject: Re: [PLUG] ARGH!!!!!!!!!
Date: 26 Feb 2003 09:39:08 -0500
A kernel process, "swapper", is attempting to dereference a NULL pointer
causing a "kernel panic" AKA "fatal crash". The reason that the problem
can "occur at any time for any reason" is because the "swapper" process
takes care of virtual memory management. It literally swaps memory pages
to disk when they're not the active process and more real RAM is
required, and then back into memory when a page fault occurs.
The Possible causes:
1.the kernel image is defective...since you said that you reinstalled
from disk.. I have to ask.. were you running 7.0 successfully for some
time prior to the time when this failure began? If the answer is no,
begin with a different kernel. RH 7.2 is known to be more stable. If the
answer is yes, you ran it for years without a problem....then...
2.hardware could be faulty...I would suspect the disk or RAM first, but
it could also be caused by faulty capacitors, cache, or other MB issues.
for bad sector on the disk..try a different disk drive or reformat the
drive completely and check for bad sectors before repartitioning and
reinstalling. Make sure the swap partition is of adequate size (most
recommend about 2.5 to 3 times the size of ram).
for bad memory, the easiest way to test is to swap it out. Unless you've
got compatible ram lying around that you know is good, this is an
expensive way to test. New RAM is sometimes faulty too, so how do you
test the test RAM? Most computer bios will run a RAM check during the
POST test during boot. Pay attention to it, but it's not guaranteed to
be reliable. Remove the RAM gently, make sure the contacts are clean.
You can clean them with a dry "scrubby", that green stuff that's on the
side of some kitchen sponges. Be gentle. Replace the RAM and make sure
it's well seated.
Finally, there are some motherboard with a lot of bad capacitors
floating around. These could cause random failures and system hangs. The
fix it to replace the faulty caps. There are guys who will do it for you
for about $50 if you're not very handy with a soldering iron. Look over
the motherboard for bad caps. See the bad capacitor story here:
On Tue, 2003-02-25 at 19:44, Dan Roberts wrote:
> Ok folks, Im going to lay it out for you. My linux box is teetering on
> edge of getting thrown out the damn window with this freaking problem.
> Lately it's been acting really strange. Randomly just not letting me log
> (the screen goes black, screen saver like, and then never comes back,
> can i log on remotely through ssh/telnet). Or, it will just crap out in
> middle of a session. Or the kicker is this error message, which seems to
> come at no particular time or for any conceivable reason. Brace
> unable to handle kernel null pointer dereference at virtual address
> current -> tss.cr3 = 00101000, %cr2 = 00101000
> *pde = 00000000
> Oops: 0002
> CPU: 0
> EIP: 0010:[<c01087e1>]
> EFLAGS: 00010246
> eax: 00000000 ebx: 00000000 ecx: c0234000 edx: 00000000
> esi: c0234000 edi: 0000489b ebp: 000058e3 esp: c0235fa4
> ds: 0018 es: 0018 ss: 0018
> Process swapper (pid: 0, process nr: 0, stackpage=c0235000
> Stack c0106000 c0108804 00000000 c0109f58 00000000 00000000 00000001
> c0106000 000058e3 00000070 00000018 00000018 00000070 c010607b
> 00000296 c0236d4f 00000000 c0106000 0fae21d8 c0216cc0 c0100175
> Call trace: [<c0106000>] [<c01088004>] [<c0109f58>] [<c0106000>]
> [<c0106000>] [<c0100175>]
> Code: 8b 5e 14 e8 7f 8a 00 00 e8 4a 22 01 00 eb b4 b8 00 e0 ff ff
> Kernel panic: Attempted to kill the idle task!
> In swapper task - not syncing
> I have not a clue as to what the hell any of this means. I am running a
> stock redhat 7.0 system, same kernel 2.2.16-22 on my Athlon 850, 256MB
> 2 3c905C 3com ethernet cards... I have reinstalled from discs,
> formatting partitions, and the same result. I have experimented running
> every service imaginable or barebones, and one of the above still takes
> place. It even has happened after a very painful upgrade to 8.0. So now
> back to 7.0 again. Prior to this, the machine worked perfectly with the
> exact same hardware setup, so I am at a loss.
> If any of you out there either 1. know what the F this means and would
> to tell me, or 2. just like computers so much that seeing mine fly out
> second story window would be more painful them comical, your assistance
> would be much appreciated. My frustration is reaching "Falling Down"
> Dan Roberts
> Villanova School of Law
> MSN 8 with e-mail virus protection service: 2 months FREE*
> Philadelphia Linux Users Group --
> Announcements -
> General Discussion --
Democracy is the worst form of government;
except for all those other forms that have
been tried from time to time.
GnuPG public key:
<< signature.asc >>
Add photos to your e-mail with MSN 8. Get 2 months FREE*.
Philadelphia Linux Users Group -- http://www.phillylinux.org
Announcements - http://lists.netisland.net/mailman/listinfo/plug-announce
General Discussion -- http://lists.netisland.net/mailman/listinfo/plug