Hi,

I am a graduate student and recently, one of the two 3ware Escalade
7500-8 cards in our filserver bonked out. We got a replacement but it
seems that the previous card corrupted the filesystem somehow. Now I
cannot mount/e2fsck one of the two RAID arrays.
Luckily the root partition RAID is alive and working. But the other
RAID absolutely needs to be recovered. It has 600GB of not completely
backed up data.
I have tried the obvious solutions already:
1. e2fsck : does not work -> cannot find superblock
2. e2fsck -b with ALL alternate superblocks: does not work

The errors are as follows:

e2fsck 1.27 (8-Mar-2002)
Couldn't find ext2 superblock, trying backup blocks...
fsck.ext3: Filesystem has unexpected block size while trying to open /dev/sdb1

The superblock could not be read or does not describe a correct ext2
filesystem. If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
e2fsck -b 8193 <device>

I have not tried mke2fs -S yet as that is supposed to be
destructive. Also I have not yet tried data recovery tools like
e2retrieve, e2restore or debugfs.
The situation is pretty much desperate and my advisor is willing to
throw in a printer or playstation as a reward to whosoever helps us
get the filesystem back.

Any *quick* help would be really appreciated.

On Wed, 10 Dec 2003 23:47:31 -0800, Puneet Gupta wrote:

Have you ensured that the RAID state is OK? Used whatever reporting tools
that come with the card to ensure that the RAID is not corrupted on more
than one disk? If you have tried every superblock backup copy, and none
work, then I would think that whatever data was written during this
corruption episode would also be toast. A 600GB filesystem has a _lot_ of
backup superblocks, and if they are all corrupt then you probably have
corrupted all your data. Conversely, if the superblock backups are not
the problem, then the filesystem is in such a hosed state that recovery is
not likely to be possible. Either way, I think you may be out of luck.

Puneet Gupta <puneet@steiner.ucsd.edu> writes:

[...]

[...]

Somehow I do doubt that this is indeed a completely hosed RAID
filesystem, although this is possible. You didn 't mention what happened
to the original RAID card, and whether or not the system was running
with the faulty card still in action for some time. IF the RAID controller went
havoc and continued to write junk across the whole 600GB, then you're
out of luck. There's probably no data left to restore the filesystem.

OTOH, it could be as simple as finding the correct blocksize
used to create the filesystem in the first place. Try using the
-B option with e2fsck.

Sidenote: You did reinstall the replacement card with the
precisely same settings as the original card used, did you?

Michael
--
Michael Buchenrieder * mibu@scrum.greenie.muc.de * http://www.muc.de/~mibu
Lumber Cartel Unit #456 (TINLC) & Official Netscum
Note: If you want me to send you email, don't munge your address.