Data recovery from a failed drive using dd_rescue over ssh:

1. Use Helix3 as a boot disk; ensure the network is connected; Make sure the bad/recovered hard drive is connected, but do not mount it in helix – it causes unnecessary disk activity. I prefer the console mode.
2. Identify the correct disk to recover. You might already know this, but use fdisk -l to be sure.
3. dd_rescue /dev/hda – | ssh huanix@192.168.1.4 ‘dd of=/media/data/test.img’
4. sometimes this hangs and i have to ctrl-z and restart. accept the RSA key and enter the user password.
5. The rescue can take a lot of time depending on how well the head transfer went, or how damaged the platters are. My worst case average is 1gb/hr
6. Now that you have the image on a safe machine, i like to make a copy of the disk image, just to be safe!
7. examine/build the partition table. I struggle here, but i eventually get it right.
8. fdisk -l to check the drive format
9. fdisk /dev/hda
10. enter expert mode (x)
11. set the sectors (s), heads (h), and cylinders (c), then return to home (r) and then write (w)
12. Calculate the offset: fdisk -lu //media/data/test.img
13. look in the “start” column, find the value and multiply it by 512 to get your offset.
14. the following mounts a fat32 partition (vfat)
15. sudo mount -t vfat -o loop,offset=32256 /media/data/test.img /media/test

Primary references:

http://www.debianadmin.com/recover-data-from-a-dead-hard-drive-using-ddrescue.html

http://www.linux-usb.org/gadget/file_storage.html

As I research this method I’m learning more about dd_rescue vs. ddrescue. I’m working to use the newer, better ddrescue… this is just for reference!