Kernel 2.2.11 crash...

Mike A. Harris (mharris@meteng.on.ca)
Sun, 3 Oct 1999 19:46:20 -0400 (EDT)


I've heard about flaws in the 2.2.11 kernel, but have not
experienced any of them and I've been using it since the day it
was released. I have had no problems whatsoever, and have had
uptimes as great as 27 days. Of course the uptime ended
voluntarily, not via any kernel trouble or anything like that.

After hearing what caused the problem, and having had 20+ days
of uptime I didn't feel the need to worry much. Many people I know
are using 2.2.11 too, with no problems.

Last night at 8:02 I got an Oops while sleeping. Here is the Oops
report, with details below...

Oct 3 08:02:17 asdf kernel: Unable to handle kernel NULL pointer
dereference at virtual address 00000040
Oct 3 08:02:17 asdf kernel: current->tss.cr3 = 058b7000, %cr3 = 058b7000
Oct 3 08:02:17 asdf kernel: *pde = 00000000
Oct 3 08:02:17 asdf kernel: Oops: 0002
Oct 3 08:02:17 asdf kernel: CPU: 0
Oct 3 08:02:17 asdf kernel: EIP: 0010:[schedule+132/640]
Oct 3 08:02:17 asdf kernel: EFLAGS: 00010046
Oct 3 08:02:17 asdf kernel: eax: 00000000 ebx: c3dd6000 ecx: 00000001
edx: 00000000
Oct 3 08:02:17 asdf kernel: esi: c01cdc00 edi: 00000005 ebp: c3dd7f10
esp: c3dd7f00
Oct 3 08:02:17 asdf kernel: ds: 0018 es: 0018 ss: 0018
Oct 3 08:02:17 asdf kernel: Process kwm (pid: 11536, process nr: 66,
stackpage=c3dd7000)
Oct 3 08:02:17 asdf kernel: Stack: 00000005 00000246 00000286 c01cdc00
c3dd7f38 c010eba8 c3dd7f24 00000000
Oct 3 08:02:17 asdf kernel: 00000010 00000000 c01afe10 0350ad23
c3dd6000 c010e804 00000000 c012ba75
Oct 3 08:02:17 asdf kernel: c553faf4 00000005 c553faf8 00000024
00000104 00000005 c3dd6000 00000005
Oct 3 08:02:17 asdf kernel: Call Trace: [schedule_timeout+108/140]
[process_timeout+0/16] [do_select+157/532] [sys_select+1014/1372]
[system_call+52/56]
Oct 3 08:02:17 asdf kernel: Code: 89 42 40 89 50 3c c7 43 3c 00 00 00 00 c7
43
40 00 00 00 00

I am using a self compiled 2.2.11 kernel which seems to work
flawlessly on my system, at least til now... The kernel is stock
from the tarball. I have VMWARE installed, and had Windows NT 4.0
running in KDE with a game of StarCraft running but on PAUSE. The
VMWARE release is 1.0 build 291 release candidate 1. The VMWARE
modules loaded are the ones that come with it which are for kernel
2.2.10, but which work fine with 2.2.11.

I've been running this beta for quite a while with *NO* trouble
whatsoever. Extreme stability. I've left VM's running for days
on end with no lockups or any other trouble at all.

I have lately been experiencing some KDE problems however, and KWM
(which appears to have caused the Oops) has been acting up from time
to time requiring me to quit X and start up again.

I'm running X in 256 color mode in case that matters. Cirrus Logic
5446 video, K6-200 with 96Mb of RAM, yada yada.

I've no idea if this crash was caused by VMWARE or its modules, but
it is not likely that I can get it to occur again in a reproduceable
manner since 2.2.11 came out a LONG time ago, and this is my first
lockup with LONG uptimes. The video in VMWARE *WAS* messed up however
when I turned my monitor on.

The Oops occured at 8:02 according to "messages", however it was
3:00pm when I turned my monitor on, and saw the messed up video. At
this point I figured that VMWARE had crashed for the first time ever,
and so I switched consoles. It let me switch to a VC, but the screen
was blank. Capslock/Numlock worked for about 20 seconds while I
switched trying to find a visible VC. Then the keyboard stopped
responding entirely. I could not telnet into my machine over the
LAN either. I had a PPP connection that was up, and had to unplug
the phone line to get the thing to disconnect. Very odd.

It has been QUITE a long time since I've encountered an Oops with a
stable kernel, so if I'm missing any required information, please
let me know. I can send my .config, or any other output of commands,
/proc files, etc... if need be.

If this is a known bug, please let me know, and if it is fixed,
let me know where I can get the patch, or if I should use 2.2.12, or
a pre-13 kernel, or just wait for 2.2.13.

Any help would be greatly appreciated.

Thanks.

--
Mike A. Harris                                     Linux advocate     
Computer Consultant                                  GNU advocate  
Capslock Consulting                          Open Source advocate

[insert witty random tagline or cool URL here]

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/