xfs
[Top] [All Lists]

Re: Core dump when mounting disk

To: lordvader@xxxxxxxxxxxxxxxxxx
Subject: Re: Core dump when mounting disk
From: Eric Sandeen <sandeen@xxxxxxx>
Date: 11 Nov 2003 09:52:41 -0600
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <3fb0d46ed017a2.75422683@xxxxxxxxxxxxxxxxxx>
Organization: Eric Conspiracy Secret Labs
References: <3fb0d46ed017a2.75422683@xxxxxxxxxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
On Tue, 2003-11-11 at 06:22, Michael Mamone wrote:
> My system recently locked up while I was compiling a program. My
> system has two partitions, and the files I was compiling lived on the
> second partition. When I reboot, the attempt to remount that partition
> fails with a kernel panic, and any subsequent attempts to mount or
> call xfs_repair on that disk just hang.

If the filesystem is small enough, it would be nice if you could dd the
partition into a file so that you have a backup of the problem
filesystem, before you try other things.

If the filesystem is not a system disk, I'd comment it out of the fstab
so that you can boot without an oops.  Or, add "ro,norecovery" to the
mount options.  The fs will then be inconsistent, but it won't oops
during recovery since that won't run.

In a pinch, you can xfs_repair -L to zero out the log, but that will
lose both the metadata still in your log, and any forensic evidence
about what has happened here.

Another xfs developer here has just worked out a log copy utility, if we
can get that over to linux fairly quickly perhaps you could make a copy
of your log and we could take a look.

> I've attached the kernel panic output. Note that this is with Mandrake
> 9.1's default kernel (2.4.21-something-or-other). The same thing
> happens with kerne; 2.4.22, but the dump from the mandrake kernel had
> more information.

Not sure what to make of the backtrace at the moment...

-Eric

-- 
Eric Sandeen      [C]XFS for Linux   http://oss.sgi.com/projects/xfs
sandeen@xxxxxxx   SGI, Inc.          651-683-3102


<Prev in Thread] Current Thread [Next in Thread>