Recovering data from failing hard-drive using ddrescue

1

Lately my hard-drive have failed me. I'm trying to rescue some important data from it and right now im stuck doing this... I've read about ddrescue data recovery tool and it seems like it's doing its job but... I think that I'll have to wait indefinitely.

This is information that I got from SMART: http://pastebin.com/yz0ew8zT before trying to rescue any data.

The hard-drive is only visible in bios and from linux distro's - I can't mount it tho.

In this file: http://pastebin.com/xAZj7NBU I provide the current status of invoking 

ddrescue -d /dev/sdd2 partition.img partition.logfile

There are also information that I got from reading logfile...

The problem is that this information is like 6hrs old... and time since last successful read: 7 hr... I don't know if I should let it keep running or not... I want only some data. I don't care about recovering everything from this partition.

I just want to ddrescue finish its job and provide me with .img file that I can mount and explore to recover my data. What should I do? Wait? Try some other options when using ddrescue?

EDIT

In the end I canceled ddrescue because I thought that it recovered enough data. I could let it run but I think that it would end in like one week or so... anyway I will describe my steps after cancelling it.

First command that I need to run was (if I wouldn't do this I would get some nasty errors and not be able to mount the image):

sudo ntfsfix partition.img

Then just mount the image file:

mount -o loop,ro partition.img mountpoint/

And then I was happy to recover my files!

rosko

Posted 2014-05-19T12:03:31.880

Reputation: 111

1You can try SpinRite its expensive but you might have luck and getting the drive to a point where you can create an image of it. Of course it also might be to far gone at this point to do anything. You might have to accept you won't be able to backup your data. – Ramhound – 2014-05-19T12:50:47.670

Can I use some built-in options of ddrescue so it produces the .img file and just omit those bad sectors? There are probably no data in those sectors that I really need but I'm just asking cause I'm a little nooby at this.... – rosko – 2014-05-19T12:55:46.470

You will have to check the documentation for ddrescue to see if there is a switch to skip bad sectors. But you are not guaranteed a complete image if you do that. I can't view your S.M.A.R.T data, it would have to be in your question, if you want an educated guess if SpinRite might be able to help. – Ramhound – 2014-05-19T12:58:01.853

There is one option for this I think and it's: '-n' aka '--no-split' - Skip the splitting phase. Avoids spending a lot of time trying to rescue the most difficult parts of the file. I don't know if I can do this right now when I've already started this with only ddrescue -d /dev/sdd2 partition.img partition.logfile – rosko – 2014-05-19T13:00:36.827

You can't change how the image will be created. You would have to cancel that operation then use the arguments to skip any bad sectors that might exist. Of course the drive might also be dead. – Ramhound – 2014-05-19T13:15:09.537

I think that I will try this... just cancel and retry with --no-split option... – rosko – 2014-05-19T13:22:50.747

let us continue this discussion in chat

– rosko – 2014-05-19T13:48:58.083

Answers

0

You could try changing the HDD mainboard with an exact replacement... the disk platters may be OK but if the board isn't controlling the disk or processing the data correctly then this may be why DDRescure is taking forever...

Another solution may be a professional data recovery specialist... the prices can be astronomical, however.

You could also try TestDisk and PhotoRec by CGSecurity.

Ultimately, if no tools work, you may have to take it as a lesson learned and to always make sure you have backups of your most critical data...

Kinnectus

Posted 2014-05-19T12:03:31.880

Reputation: 9 411

What good would changing the control board do? That should only be done if you actual suspect the control board went bad. Whats more likely is there are sigificant bad sectors on the disk, there really isn't a solution for bad sectors, the only real soltion is to replace the drive. – Ramhound – 2014-05-19T12:49:58.287

The self-tests logs of linux GSmartControl software indicates that the tests passes with electrical failure at 10%. Changing to an exact replacement may be difficult at this point. I would like to know if I can omit somehow those bad sectors and let ddrescue finish it job so I have this .img file which I can later mount or if there are so many bad sectors on disk then this won't be possible without fixing them first...? – rosko – 2014-05-19T12:53:20.750

1@rosko - You would have to stop the current job to create a job that skips the bad sectors ( I only assume there are bad sectors ) because its one of the only reasons for a disk failure of this type. – Ramhound – 2014-05-19T13:30:56.783

Asked: 2014-05-19T12:03:31.880

Viewed: 2 000 times

Active: 2014-05-19T21:21:49.490