2.6.19 randomly crashes on boot sometimes

Romashka · 2007-01-09 14:30:15

Splitted from 0.8 Voodoo installation feedback thread.

Misfit138 wrote:

WiLLiE wrote:
Romashka wrote:
Anyone experience random crashes (5% of boots) with the latest kernel (2.6.18 was OK) on early stage (when starting udev from initial image)?
Kernel dumps registers and stack many times (same output), I don't remember where it reported crash (will write down kernel messages when it happens next time) but it was something related to udevd.
That happens here during boot from time to time, totally random. (on my main computer)
After manually resetting, it boots like nothing happened.
Same problem here, very random, output is EXTREMELY FAST...hard reboot works fine.

Romashka · 2007-01-09 14:42:25

It's strange that we don't have many such reports on forums.
I don't have any specific hardware or modules in initcpio. :?

hightower · 2007-01-09 16:03:13

It's strange that we don't have many such reports on forums.

The usual suspects: Memory, hard drive (cables!) and perhaps PSU?

hightower

Avenger · 2007-01-09 16:25:35

I don't know if it's the same, as I use silent gensplash (beyond kernel) at boot, so I cannot see error outputs from the kernel, but I have had a few freezes during the "Initializing kernel"-process with 2.6.19. Very few though, I've only noticed it about 2-3 times I think.

I'm running an asus laptop, intel centrino and I don't use the new pata driver.

As quoted here, a hard reboot workes just fine and the computer boots up as normal.

Romashka · 2007-01-09 17:54:51

I've got it again today.
I took a pen and wrote down all interesting as far as Shift-PgUp allowed me to see messages.

Here's what I've written:

Process cdrom_id (pid: 1369 . . .
Stack: . . .
Call Trace: . . .
. . .
<6>note: cdrom_id[1369] exited with preempt_count 1
BUG: scheduling while atomic: cdrom_id/0x00000001/1369
. . .
kernel BUG at arch/i368/mm/highmem.c:42!
invalid opcode: 0000 [#23]
PREEMPT SMP
Modules linked in: powernow_k8 freq_table usbcore ext3 jbd mbcache sr_mod cdrom sd_mod pata_amd sata_nv ata_generic libata
. . .
Process udevd
. . . the same stuff
Process start_udev
. . . the same stuff

Any ideas?
2.6.18 was OK. Windows XP works fine too.

Romashka · 2007-01-09 18:19:27

OK, here is it: http://lkml.org/lkml/2006/12/5/167
Thanks goes to cactus!

Misfit138 · 2007-01-09 18:20:44

Yes, that looks quite familiar. Same outputs here IIRC, and they scroll down the screen at about 100MPH. I just figured it was something I did wrong somewhere along the line. It does seem quite harmless, though annoying.

timtux · 2007-01-10 14:07:07

Same here :-P

cromo · 2007-01-10 20:06:00

Kinda OT: I had similar problems with rlocate kernel module.

Romashka · 2007-01-11 11:52:28

cromo wrote:

Kinda OT: I had similar problems with rlocate kernel module.

May be the same issue.
There's even a test program that triggers this bug, so it's not boot-only bug.

sokuban · 2007-01-11 12:56:42

Its been happening to me too. Wonder what is up?

The first time it happened I freaked out. "Why is the text moving so fast... Why did it stop... oh crap."

Asaru · 2007-01-11 13:51:08

happend to me too on 2 machines but only once per machine yet

cougar · 2007-01-11 14:42:02

Yep same for me !

soloport · 2007-01-11 20:54:57

Happened only yesterday (first time, only time). But it happened. Not too worried, though, as no data loss apparent.

Romashka · 2007-01-12 10:28:05

OK, I added it to Flyspray:
http://bugs.archlinux.org/task/6202

iphitus · 2007-01-12 11:03:35

Seems like it's fixed for me in 2.6.19.2

I've rebooted a few times already and have yet to experience it. I'm using beyond3pre1, which includes 2.6.19.2
http://iphitus.loudas.com/beyond/2.6.19 … pkg.tar.gz

It'll be in the repos shortly. 2.6.19.2 should be too.

James

Romashka · 2007-01-12 12:30:47

iphitus wrote:

Seems like it's fixed for me in 2.6.19.2

I checked kernel changelog but didn't see a patch for this. :?

iphitus · 2007-01-12 12:55:01

Romashka wrote:

iphitus wrote:
Seems like it's fixed for me in 2.6.19.2
I checked kernel changelog but didn't see a patch for this. :?

Because it's not mentioned in the terms you were lookin for

[PATCH] VM: Fix nasty and subtle race in shared mmap'ed page writeback

James

Romashka · 2007-01-12 13:03:54

iphitus wrote:

Romashka wrote:
I checked kernel changelog but didn't see a patch for this. :?
Because it's not mentioned in the terms you were lookin for
[PATCH] VM: Fix nasty and subtle race in shared mmap'ed page writeback

But this is that bug that leads to filesystem corruption.
Does this cause our boot crash too?

iphitus · 2007-01-13 05:40:28

Romashka wrote:

But this is that bug that leads to filesystem corruption.
Does this cause our boot crash too?

nop, I got it confused. And it turns out im still having problems on boot. Thought it was fixed when I managed to boot 3 times without error.

James

Romashka · 2007-01-13 10:13:38

iphitus wrote:

nop, I got it confused. And it turns out im still having problems on boot. Thought it was fixed when I managed to boot 3 times without error.

Yeah, I got error about 1 per 20 boots.
Need to find patch somewhere.

Romashka · 2007-01-14 12:17:54

http://lkml.org/lkml/2006/12/26/91 - xfs/udev issue? I don't think so.

http://linux-nfs.org/pipermail/nfsv4/20 … 05289.html - same thing.

This is definetely not boot-only issue.

http://permalink.gmane.org/gmane.linux.kernel/103417 - is same fix needed for 2.6.19?

Hmm... the same issue was in 2.6.9: http://sourceware.org/ml/systemtap/2006 … 00485.html

http://lkml.org/lkml/2006/12/12/155 - what did he mean by saying:

With mem=896M highmem-enabled kernels work OK.

?
He set mem=896M in kernel boot option or what?

broch · 2007-01-27 02:39:41

I got this error on shutdown (2.6.19.2) voodoo. So it is not exactly boot up.

Romashka · 2007-01-27 11:46:08

broch wrote:

I got this error on shutdown (2.6.19.2) voodoo. So it is not exactly boot up.

Yes, it can be triggered in different ways, there's even test program for it.

iphitus · 2007-01-27 12:16:39

Romashka wrote:

broch wrote:
I got this error on shutdown (2.6.19.2) voodoo. So it is not exactly boot up.
Yes, it can be triggered in different ways, there's even test program for it.

link?

Arch Linux

#1 2007-01-09 14:30:15

2.6.19 randomly crashes on boot sometimes

#2 2007-01-09 14:42:25

Re: 2.6.19 randomly crashes on boot sometimes

#3 2007-01-09 16:03:13

Re: 2.6.19 randomly crashes on boot sometimes

#4 2007-01-09 16:25:35

Re: 2.6.19 randomly crashes on boot sometimes

#5 2007-01-09 17:54:51

Re: 2.6.19 randomly crashes on boot sometimes

#6 2007-01-09 18:19:27

Re: 2.6.19 randomly crashes on boot sometimes

#7 2007-01-09 18:20:44

Re: 2.6.19 randomly crashes on boot sometimes

#8 2007-01-10 14:07:07

Re: 2.6.19 randomly crashes on boot sometimes

#9 2007-01-10 20:06:00

Re: 2.6.19 randomly crashes on boot sometimes

#10 2007-01-11 11:52:28

Re: 2.6.19 randomly crashes on boot sometimes

#11 2007-01-11 12:56:42

Re: 2.6.19 randomly crashes on boot sometimes

#12 2007-01-11 13:51:08

Re: 2.6.19 randomly crashes on boot sometimes

#13 2007-01-11 14:42:02

Re: 2.6.19 randomly crashes on boot sometimes

#14 2007-01-11 20:54:57

Re: 2.6.19 randomly crashes on boot sometimes

#15 2007-01-12 10:28:05

Re: 2.6.19 randomly crashes on boot sometimes

#16 2007-01-12 11:03:35

Re: 2.6.19 randomly crashes on boot sometimes

#17 2007-01-12 12:30:47

Re: 2.6.19 randomly crashes on boot sometimes

#18 2007-01-12 12:55:01

Re: 2.6.19 randomly crashes on boot sometimes

#19 2007-01-12 13:03:54

Re: 2.6.19 randomly crashes on boot sometimes

#20 2007-01-13 05:40:28

Re: 2.6.19 randomly crashes on boot sometimes

#21 2007-01-13 10:13:38

Re: 2.6.19 randomly crashes on boot sometimes

#22 2007-01-14 12:17:54

Re: 2.6.19 randomly crashes on boot sometimes

#23 2007-01-27 02:39:41

Re: 2.6.19 randomly crashes on boot sometimes

#24 2007-01-27 11:46:08

Re: 2.6.19 randomly crashes on boot sometimes

#25 2007-01-27 12:16:39

Re: 2.6.19 randomly crashes on boot sometimes

Board footer