Willard,

dd if=/dev/target bs=512 conv=noerror,sync of=/mnt/destination


Since you're experiencing many I/O errors you'll want to drop back to
the default 512 blocksize.  You'll also want the other options you
passed, reading past errors and padding with null.

You may be experiencing different hash values for the original drive as
a result different data in depending which run.  IE, your md5sum
/dev/hdb may equal 'X' the first run, but 'Y' the second run because
more (or less) data was successfully read the second run.


regards,

farmerdude

http://www.farmerdude.com/




On Thu, 2006-02-16 at 05:31, Willard Van Dyne wrote:
> Hi all.
> 
> I'm trying to properly clone a 4.3GB (it's old, I know) hard disk 
> which unfortunately has a lot of bad sectors.
> I am using Helix 1.7 as an operating environment, not mounting the 
> old drive at all.
> I used the command:
> 
> dd if=/dev/hdb of=/mnt/hda7/image.dd conv=noerror,sync
> 
> My problem is that the md5 hash of the image file is different from 
> that of the original HD (acquired via the command: dd if=/dev/hdb 
> conv=noerror,sync | md5sum > /mnt/hda7/orig_disk_md5sum.txt)
> 
> Can anyone please enlighten me as to what I'm doing wrong?
> 
> Thanks in advance.
> 
> 

Reply via email to