Recovering Data from a problematic HDD

andreuha_

Distinguished
Oct 4, 2006
3
0
18,510
I want to pull about six gigs of data off an HDD (40gig 2.5" Toshiba). It's got lots of bad sectors, but the vast majority of them are affecting system-related files (thankfully).

I've tried to use a bunch of different programs (A few from Hiram's BootCD v.8.5, the one that worked best was R-Studio for NTFS), but they're all SLOW. R-Studio calculates to around 10 days to transfer this data (around 2kb/s).

Is there a faster way to pull this stuff off? I don't mind cutting my losses on a handfull of files if I can get this done within a reasonable amount of time. Are there any better programs I can use? Any other suggestions?

is there a fast drive-clone app which ignores bad-sectors? that would make things very easy for me!


Also, anyone here try Salvation Data HDD Scan & Repair 3? I have it but I don't know if it'll wipe my drive or just make it usable (perhaps allowing me to recover by normally booting into winXP instead of the current DOS or WinXP-PE)


Thanks!!
 

Codesmith

Distinguished
Jul 6, 2003
1,375
0
19,280
No fast method of recovery that I am aware of.

Unstoppable Copier is probably you best bet if you set it to autoskip bad files.

It still repeatedly tries to get a good read from bad sectors, but you can tell it to give up on a file as soon as it has a single unrecoverable read error.

Unfortunately it also copied a few file with unrecoverable errors depsite my settings. Even worse it doesn't log these files as having errors, so I had to check every recoverd file for coruption.

I reran in multiple time and each time files with read errors slippped through the cracks.

Unfortunately I don't have a better alernative.

If the file system isn't intact then simply scanning for files with Get Data Back can take up to 20 hours on a large drive, after that the actual recovery is relatively fast.

Not sure if there is a way to exclude corrupted files. Given the way the data is accessed, error detection might not be possible.