Re: assistance recovering failed raid6 array

[Brad Campbell <lists2009@fnarfbargle.com>]

On 21/02/17 05:21, Phil Turmel wrote:
> On 02/20/2017 03:45 PM, Wols Lists wrote:
>
>> But there's a request on the linux wiki program for someone to write a
>> utility program that takes a ddrescue log and flags the duff sectors as
>> "soft unreadable". That would mean that if you can recover 35 drives,
>> provided no stripe has lost two sectors across two drives, you wouldn't
>> lose any data.
>>
>> If you want to try and write that utility? Or if you want to email me a
>> ddrescue log with a bunch of failed sectors, I'll have a go at writing
>> it myself :-)
>
> Check out hdparm --make-bad-sector.  You can get what you are describing
> by scripting that.  It's marked very dangerous, but I guess if one has
> nothing to lose....
>

Wol and I have tic-tacced on that a couple of times. He suggested the 
idea and I proved the viability of it by testing hdparm in a RAID to do 
exactly that. Neither of us has had a chance to stitch it all together, 
but the preliminary tests indicate that would do *exactly* what was 
required.

Given the hdparm commands are completely reversible and non-permanent, 
*and* they are being executed on a destination of a ddrescue, there is 
pretty much no risk (as long as the underlying glue code to be written 
gets the sector numbers and offsets right).

Regards,
Brad