Ext3 users - Dec 2001 - ext3 crash

Hi!

I experienced an ext3 crash a few days ago, the corresponding part of my
kernel log is attached (there were no IDE-errors, just the ext3 stuff).
The /home partition (/dev/hdd2) got remounted ro, so the box was still
usable somehow (was able to log in, kill stuff and reboot it).

I rebooted it, it came back up again w/o any problems, remounted /home ro
and did a e2fsck on it, which reported just one illegal (bad?) block in an
inode (the mentioned block was somewhere above the magical
4294967296 limit, forgot to cut'n'paste the error).

I'm currently running 2.4.13-ac8, IDE Controller is a VT82C586, HDD is a
Seagate ST320430A.

Is there an easy way to resolve the symbols in the call trace?

best regards,
michael

Hi,

On Tue, Dec 11, 2001 at 09:59:54AM +0100, Michael Renner wrote:
 
> I experienced an ext3 crash a few days ago, the corresponding part of my
> kernel log is attached (there were no IDE-errors, just the ext3 stuff).
> The /home partition (/dev/hdd2) got remounted ro, so the box was still
> usable somehow (was able to log in, kill stuff and reboot it).
> 
> I rebooted it, it came back up again w/o any problems, remounted /home ro
> and did a e2fsck on it, which reported just one illegal (bad?) block in an
> inode (the mentioned block was somewhere above the magical
> 4294967296 limit, forgot to cut'n'paste the error).
OK, that in itself looks related to having a bad block group error:
> Dec  9 19:55:30 srck@trottelkunde Kernel panic: EXT3-fs panic (device
ide1(22,66)): load_block_bitmap: block_group >= groups_count - block_group =
131071, groups_count = 94
but the real question is, where did this happen:
> Dec  9 19:55:30 srck@trottelkunde Assertion failure in journal_start() at
transaction.c:227: "handle->h_transaction->t_journal == journal"
Please use "ksymoops" to decode the call trace, and I'll take a
closer
look.

Cheers,
 Stephen