[Gllug] Re: ext3 disk recovery?

Wulf Forrester-Barker wulf.f-b at uhl.nhs.uk
Fri Feb 27 10:07:43 UTC 2004

Thanks for all the replies. Skimming through them
(and jumping a few that don't provoke futher
comments or questions from me):

Ben <ben_m_f at yahoo.co.uk> suggested:

> You also described your disk sometimes being
> recognised by the bios and sometimes not. Perhaps
> the drive is flaky and that is the underlying
> cause of your filesystem problems? Consider
> getting a drive stress test utility from your
> manufacturer's website?

That's a good idea. I can't remember what make and
model it is but it's in an easily accessible drive
caddy so I might do some checking and surfing.

> If you have a problem with the first superblock,
> try specifying a different superblock with
> e2fsck.

I did see that in the man page but I don't think I
tried different superblocks. If I try a few block
sizes to no avail, what would be the consequences
of blatting the superblock? When I was installing
Linux on the disk I was having some problems
because of a previous installation of Mandrake that
I'd carelessly wiped out and the left of remnants
of a long distant Win98 install. I followed the
instructions in the fdisk man page to write zeroes
to the first 512 bytes of the disk using dd. Would
that wipe out the superblock and would that be a
good thing (letting me mount it even if not to run
Linux from it) or a bad thing (completely mucking
up all chance of seeing my data again)?

Bernard <bap at shrdlu.com> mentioned:

> It's worth checking the ribbon cable first. I've
> seen similar problems when a cable got too close
> to a cooling fan. The OP has recently installed a
> disk so a faulty or loose cable is quite
> possible. It's worth replacing a 50p cable before
> replacing a £50 disk.

Was that a case of physical (and observable) damage
to the ribbon (cut to, er, ribbons by the fan

Andrew <ah at gnd.com> asked:

> Are you running fsck on /dev/hda... or
> /dev/hda1?  bad superblock can mean you're
> reading the wrong place on the disk.  Make sure
> the partition you're fscking is actually a
> filesystem and not something like an extended or
> swap.

I'm pretty sure I was trying to run it on
/dev/hda1, which was my root Linux partition (the
disk also had a Linux swap partition and another
Linux partition that I think had /home on it... I
know, I know, take more notes in future!).

Chris <chrisbell at overview.demon.co.uk> mentioned:

> You say that it is a dual-boot machine. There are
> some M$ viruses around which corrupt the MBR of
> discs.

It's not been dual boot since I wiped out Win98 in
mid-January. The disk has been repartioned,
formatted and installed with Linux since then, so I
don't think I can blame Bill for this one.

> Another possible problem that has affected more
> than one machine here is a loose power connector
> to a drive. I have been able to poke an old
> broken needle between the metal and the housing
> to tighten the connectors.

That's something I'll look into. The drive is
mounted in a removable disk caddy - it's secure in
there but it is a while since I checked the
connection between the PSU and the caddy housing.
Using a broken needle sounds dodgy (short
circuits?) but I'll look for any signs of

When it comes down to it, I'm not too bothered if I
can't get the data off this disk (although it is a
good learning opportunity) but I do want to get the
machine back up and working to play an
administrative role on my home network and so
physical things like cables will definitely be
worth checking thoroughly.



Wulf Forrester-Barker (Webmaster)
The Lewisham Hospital NHS Trust

wulf.f-b at uhl.nhs.uk


Any opinions expressed in this email are those of the individual and
not necessarily the Trust. This email and any files transmitted with
it are confidential and intended solely for the use of the individual
or entity to whom they are addressed. Any unauthorised disclosure of
the information contained in this email is strictly prohibited.

The contents of this email may contain software viruses which could
damage your own computer system. Whilst we have taken every
reasonable precaution to minimise this risk, we cannot accept liability
for any damage which you sustain as a result of software viruses.
You should therefore carry out your own virus checks before opening
the attachment.

If you have received this email in error please notify the sender or
postmaster at uhl.nhs.uk. Please then delete this email.

University Hospital Lewisham
Tel: 020 8333 3000
Web: http://www.lewisham.nhs.uk/

Gllug mailing list  -  Gllug at gllug.org.uk

More information about the GLLUG mailing list