Re: mapfile when disk needs to be reset

[Antonio Diaz Diaz]

Hello Brian,

Brian J. Murrell wrote:
> I am finding that in trying to use ddrescue, the source disk will
> eventually get into a state where the machine it's in needs to be
> restarted (hard -- with a full power-down/up).
>
> But when the disk gets into this state while ddrescue is running it
> continues to try to read from the disk and fails.  Presumably all of
> the blocks it tried to read while the disk was in this needing-a-reset
> state are being flagged as bad.

Yes but, have you tried ddrescue 1.26-pre1? It includes this change which I 
expect will detect some such needing-a-reset situations:

Ddrescue now exits with status 1 on fatal read errors.
(Suggested by Marco Marques).

You can find 1.26-pre1 at
http://download.savannah.gnu.org/releases/ddrescue/ddrescue-1.26-pre1.tar.lz

Please, try it and tell us if it solves your problem.


> But maybe those blocks that were attempted after the disk got fubar'd
> (and the machine needed a power down/up) are not bad.  Maybe once the
> disk is reset they would be readable.
>
> What is the best way to handle this situation so that those potentially
> readable blocks are actually re-attempted after the machine is
> restarted?

Passing option '-A' to ddrescue the first time it is run after a power up:

'-A'
'--try-again'
     Mark all non-trimmed and non-scraped blocks inside the rescue domain as
     non-tried before beginning the rescue. Try this if the drive stops
     responding and ddrescue immediately starts scraping failed blocks when
     restarted. If '--retrim' is also specified, mark all failed blocks
     inside the rescue domain as non-tried.


Best regards,
Antonio.