The 2.2TB partition is the corrupt partition? I thought the entire 3TB partition was already corrupt even before running testdisk.

I don't think I will have enough disk space to fit a 3TB image file for now unless I buy another equally-sized hard drive.

Can anybody think of any alternatives to this? Like for example, does ddrescue and/or photorec have an option to split the image into smaller filesizes?

You keep posting data from the attempts to do a recovery. What you need to keep track of is the device size, not the corrupt data being reported.

Ddrescue does not split images. Its output is a full image of the device.
I have not used photorec but since the output is recovered files it may be possible to use a smaller output device.

Regardless, the recommendation is still to do a full image with ddrescue then use photorec on that image, not on the failing drive. The more you use that failing drive the worse the damage becomes.

A new 4TB drive at newegg.com is only ~$105. That is an NAS drive that uses CMR recording tech so it does not have the problems with data that cheaper desktop drives (like the WD Blue or Seagate Barracuda series) using the SMR tech have.

Absolutely true. And ddrescue is indeed much, much better than dd. Especially for damaged disks!
ddrescue rules!

Okay so my new hard drive finally arrived and am running ddrescue right now.

So I bought this new 4TB hard drive and it is recognized as /dev/sdc (I named the recently created Ext4 partition as "BACKUP"). The faulty 3TB hard drive is recognized as /dev/sdb:

Now I am running ddrescue and the application says it's going to take 14 hours and 13 minutes?

Can you guys check if all settings are correct? I didn't know what to do with the 'Recovery Map File (previously called logfile)' setting in ddrescue.

14 hours is good for the size drive.

The map file is for use later in case you have some interruption to the process; ddrescue can use it to restart where it let off instead of at the beginning.

I am not sure about the destination in that screenshot image. It looks like you are writing directly to /dev/sdc and not to an image file in /media/ubuntubox/BACKUP/. If so then it will destroy the partition you created and /dev/sdc will look exactly like /dev/sdb when it is done.

You need to select the location and give it a file name to write to. After all, you are trying to create an image file, not copy the failing drive to another.

Man, I don't know what's going on right now.

I tried to abort the process, reboot the PC (because ddrescue wouldn't shutdown) and now I am trying to start the entire process again but the new 4TB hard drive is giving me some very concerning issues.

So I deleted the existing partition on the new 4TB drive (named 'BACKUP') and tried to create another clean partition but GParted is giving me this error message:

From what I see, after the reboot ddrescue created a 2,2TB partition on the new 4TB drive just like the faulty 3TB drive.

Does that mean I damaged the partition table of the new 4TB drive or something? I can't create new 4TB partitions on the new 4TB hard drive!

If gparted will not let you erase the partition table then you may need to wipe the beginning of the new drive. I would first try with gparted to create a new blank partition table then recreate your large partition.

If gparted fails then you can wipe the first part of the drive with a simple dd command to zero the beginning of the drive
That will zero out the first 10M of the drive. You then should be able to use gparted and recreate the partition table and the large partition.

Thank you man! I tried to create a new GPT partition table using GParted and apparently the process went fine. I was then able to successfully create a 4TB Ext4 partition on the new drive oh god!

So I am back at square one with ddrescue now.

What did I do wrong? Why is ddrescue doing an exact copy of the faulty 3TB drive onto the new 4TB drive? I know it's something related to ddrescue settings.

How should the settings in ddrescue look like in order for it to create an image file of the faulty 3TB onto the new 4TB drive?

Let me test it here.

It seems you are using ddrescue-gui which has some limitations.

If you use just ddrescue from the command line it will do what you want. You have to enter the command similar to what is used with dd as dd is the underlying structure.

I think that using the command line, a command such as
will create a file named sdb.img and the accompanying mapfile in the mounted partition filesystem /media/ubuntubox/BACKUP/. Use the man page for ddrescue to see what options are available and adjust the command above to your needs.

It appears the gui does not give you the option to create the image file, but writes directly to the device.

Hey!

I think I got it working correctly now upon fiddling with ddrescue settings.

The incorrect setting was the 'Image Destination' option where I have to specify a location for the image file manually.

Let's see what it can do...

I'll post the results when it's done.

Thank you man!

looks good so far

Okay. Ddrescue finished the recovery process:

But I still can't mount or access files from the image file that ddrescue created:

What should I do now?

Now you use photorec on that image file you created and hopefully most of the data can be recovered.

If not already done then disconnect the failing drive and leave it aside. It should no longer be needed.

Notice that ddrescue told you it found 3 bad sectors. Probably photorec could have recovered all but the data in those 3 sectors easily had you not changed the partition table.

Wait.

What does that mean exactly? Does it mean ddrescue could not recover all the data in the failing drive? But it says "Recovered Data: 3000.0GB" and "pct rescued: 99.99%".

And just out curiosity, how much data in GBytes equals 3 bad sectors approximately?