Forums | MacLife

Pariah wrote:

[snip]
A drive that bad needs to be tossed after you recover data. I sure wouldnt ever trust it again.

Agreed.

If the data are important and/or valuable, you should stop the search for bad blocks, turn off the computer, and consider sending the drive into one of the data recovery services. They aren't cheap: the last time I checked (which was a number of years ago), one such service charged US$150 just to look at the drive in order to determine if the data were recoverable, then $10 per MB of recovered data. The final bill was just under $2000.

macnuke wrote:

you need to have words with you client about backups.

damn straight.   he said he didn't do it as he didn't know how.   there was a case of blank discs on the desk!

I downloaded Data Rescue II (it's running in demo mode), so we'll see.   I ended with 113780 bad blocks.

macaddictguy wrote:

Have you tried the data recovery feature of TT Pro?

I did.   Didn't seem to get anywhere though.   I may try again, though I'm thinking that the drive is stuffed and will need to be changed - which was my plan.   This is a lesson to all those who kill the power while the drive is reading/writing.

Blandford Fly wrote:

macaddictguy wrote:

Have you tried the data recovery feature of TT Pro?

I did.   Didn't seem to get anywhere though.   I may try again, though I'm thinking that the drive is stuffed and will need to be changed - which was my plan.   This is a lesson to all those who kill the power while the drive is reading/writing.

Yikes! However it does happen.

As you know TTPro does take a long time to finalize any job.

Hope you keep us up to date as to how this turns out for you and your client.

macnuke wrote:

Blandford Fly wrote:

This is a lesson to all those who kill the power while the drive is reading/writing.

ouch
another topic for you and your client. perchance he did it intentionally, but he does have a suitable UPS installed as well eh?

Story goes like this:

Attached an attachment to an E-mail, got the beachball

waited

waited

Hit the power button

History

Blandford Fly wrote:

macnuke wrote:

Blandford Fly wrote:

This is a lesson to all those who kill the power while the drive is reading/writing.

ouch
another topic for you and your client. perchance he did it intentionally, but he does have a suitable UPS installed as well eh?

Story goes like this:

Attached an attachment to an E-mail, got the beachball

waited

waited

Hit the power button

History

While thats a generaly bad thing to do I have never known doing that to hose a drive on a Mac.
I suspect an underlieing problem.

Pariah wrote:

I suspect an underlieing problem.

Like?

I'm with you, sort of.   This is one of the most destroyed drives I've ever seen, but it's not uncommon for this to happen.   Last year a friend of mine had a 'friend' thought it a good idea to pull the plug whilst the beachball was up.   That took about 15 hours to get back up and running.   Massive damage.

Pariah wrote:

Blandford Fly wrote:

Pariah wrote:

I suspect an underlieing problem.

Like?

I'm with you, sort of.   This is one of the most destroyed drives I've ever seen, but it's not uncommon for this to happen.   Last year a friend of mine had a 'friend' thought it a good idea to pull the plug whilst the beachball was up.   That took about 15 hours to get back up and running.   Massive damage.

With so much damage I would suspect the drive was failing along the lines of the head boom alowing the head to touch the platters and the beach balling was the drive going down and the unplugging just finishing it off.

just guessing.

Sounds good enough to me.

You can attempt to use the command line utility dd to recover the data on the disk. Boot from an external drive, open up Terminal.app and type the following to get the device node for the dying hard drive:

df

You'll get output like this:

Filesystem              512-blocks     Used    Avail Capacity  Mounted on
/dev/disk0s9              78132000 55304120 22315880    71%    /
devfs                          203      203        0   100%    /dev
fdesc                            2        2        0   100%    /dev
<volfs>                       1024     1024        0   100%    /.vol
automount -nsl [243]             0        0        0   100%    /Network
automount -fstab [248]           0        0        0   100%    /automount/Servers
automount -static [248]          0        0        0   100%    /automount/static

What you looking for is the name of the dying disk under the 'Mounted On' column, and you'll pair that up with the /dev/ node under 'Filesystem' column. Lets say that your dying disk is /dev/disk1s1.

Next, issue the dd command to make a block copy of the disk, padding any bad sectors with zeros:

dd bs=512 if=/dev/disk1s1 of=/Recovery_Folder/Disk_Image.dmg conv=noerr,sync

(note, Recovery_Folder would be a folder on the disk you booted from; you create a folder with any name you wanted)

After a long, long while, dd will have made a block-for-block image of the damaged disk which you can then attempt to mount with Disk Utility. Hopefully, the bad sectors (which were zeroed on the image) don't extend to the underlying filesystem on the image, and you will be able to extract some data from the image.

Drive Genius? it's what I use for last resort.