[NBLUG/talk] Recovering data from a failing hard disk

[Kyle Rankin]

On Mon, Aug 23, 2004 at 11:10:21PM -0700, Lincoln Peters wrote:
> On Mon, 2004-08-23 at 22:45, Kyle Rankin wrote:
> > I'm going to go out on a limb and say that for what you are needing to do
> > (image a failing drive to an identical drive) that dd /is/ what you want to
> > use--or at least a dd-like program. You want to do block-for-block copies
> > and not worry about the filesystem underneath, but with something like
> > dd_rescue that can ignore any errors your hard drive might be spitting out.
> 
> Unfortunately, I've already started copying the data off the old hard
> disk with cpio.  And I've copied 92/96GB at this point.

Ahh well if you far along already might as well go with that, but if your
hard drive was spouting errors throughtout the copy, you might want to use
something like dd_rescue.

> 
> I don't think that dd_rescue would work in this case anyway because the
> failing hard disk has a capacity of 160GB, and the new hard disk that
> Maxtor sent me (as per the warranty) is 250GB.  I guess that, for
> whatever reason, it was easier for them to send me a significantly
> larger hard disk than one of the same size as the one that failed.  Not
> that I'm complaining!

Well, you can actually use it for partition-to-partition copying, and then
grow the filesystem (reiser, ext3 and XFS all have grow tools) or just dd
to a single file, mount it loopback, and then copy stuff over. Lots of
options really.

> 
> ---
> Lincoln Peters
> <sampln at sbcglobal.net>
> 
> "You can't have filenames longer than 14 chars.
> You can't even think about them!"
>              -- Larry Wall in Configure from the perl distribution
> 
> 

Basically, dd_rescue is useful for making bit-for-bit copies when the hard
drive is sometimes spitting out errors that make other tools exit.

-- 
Kyle Rankin
NBLUG President
The North Bay Linux Users Group
http://nblug.org
IRC: greenfly at irc.freenode.net #nblug 
kyle at nblug.org