Subject: Re: ext2fs_valloc: dup alloc
To: Alexander Becher <firstname.lastname@example.org>
From: Adam Hamsik <email@example.com>
Date: 09/25/2007 11:22:05
-----BEGIN PGP SIGNED MESSAGE-----
On Sep 25, 2007, at 1:14 AM, Alexander Becher wrote:
> * Adam Hamsik:
>> For start include backtrace and error message. Some info about
>> filesystem would be also good. And describe how to reproduce this
> The filesystem is this:
> % mount | grep wd0h
> /dev/wd0h on /data type ext2fs (local)
> % df -h | grep wd0h
> /dev/wd0h 49G 37G 9.7G 79% /data
> To reproduce it, I do this:
> % seq 200 | while read f; do : > /data/WHATEVER/$f; done
> which drops me into the debugger after creating some files
> successfully, sometimes sooner, sometimes later.
> Is there a better way to get the backtrace apart from reading it from
> the screen and typing it in manually on a second computer? I have done
> it like this now (and attached the result), but omitted many of the
> hexadecimal numbers, hoping they are not important. The inum value in
> the error message varies with every panic. I have also attached the
> result of dumpe2fs on the filesystem.
You can dump memory image from ddb with dump. And reboot your computer.
After reboot you can use gdb to get backtrace and past it here.
You have received panic here http://opengrok.netbsd.org/source/xref/
we have to figure why and how to solve it :D.
> Finally, my uname -a gives:
> NetBSD abc 3.1_STABLE NetBSD 3.1_STABLE (abc) #0: Thu May 10
> 03:25:40 CEST 2007 alex@abc:/usr/obj/sys/arch/i386/compile/abc i386
> Anything else that might be useful?
Proud NetBSD user.
We program to have fun.
Even when we program for money, we want to have fun as well.
~ Yukihiro Matsumoto
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)
-----END PGP SIGNATURE-----