BLU's server crash and data recovery

[Bill Bogstad]

On Wed, Sep 10, 2008 at 3:07 PM, Tom Metro <blu-5a1Jt6qxUNc at public.gmane.org> wrote:
> Bill Bogstad wrote:
>> Look up ddrescue.  It's a variant of dd which tries hard to skip over
>> bad blocks in order to recover every possible block of data it can.
>
> ddrescue has been mentioned on the list before.
>
> Supposedly Spinrite talks to the IDE controller on the drive and
> disables the internal retries and error correction, so it can access the
> raw data and fully control the recovery process. I'd be curious to know
> if ddrescue tries to do that.

No, I believe it doesn't.  It simply retries reads rather then giving
up.  With some kinds of disk failure
modes, this will allow the drive to make more use of its internal
recover mode.  ddrescue also changes the size of its reads in an
attempt to get all of the data thats available on either side of the
truly bad sectors.

I wasn't aware that Spinwrite was still around.  I thought that kind
of functionality went away when people switched from AT to ATA/IDE
interfaces.  Thanks for the info.

Bill Bogstad