so my old 16tb drive has decided to damn near die, I pulled everything I could off of it that would be a nightmare to re acquire, but an itemized list of what is missing would be very nice to have.
I have written the largest folder, and consequently the easiest one to require the files off as a lost cause, it has something along the lines of 9tb in it with I want to say 5tb of stuff that doesn’t matter, i’m going to cross fingers that that’s the only problem area I have yet to transfer over.
none of the data on it is required, not worth the cost of professionals pulling data, but a list of everything would go a long way especially for wanted things.
the problem I am encountering is this, the hdd will hit 100% use and spike seak time in certain files, so craps not good, problem is I have around 9tb still on it I could pull but 5tb of that I can write off and not care, I just need the file names.
if there is a good way to copy data over that once seak time spikes it just moves onto the next file without needing a 3-20 minute cool down period, I am all ears, I only have this pc, and this replacement hard drive to work with, i’m in the process of changing its drive letter… which fun of fun, is not going well, i’m going to swap in the new drive and rebuild as best I can.
as far as I can tell, I got everything of importance off the drive, what’s left is a pile of nice to have but not required.
ddrescue is designed to talk to the drive directly and skip bad reads by default. Once you have all of the readable data you can then ask it to go back and try the parts it cant read. The output can go right to another drive or an image of your drive for later recovery shenanigans.
If you have a place to copy all 16tb to, ddrescue would be the preference for pulling everything you can off what little life the drive has left.
I know for NTFS, there are file tables at the beginning and end of the partition which should be readable by some applications, but I’m not certain how you would dump those into something human readable. Windows has fsutil which can read and dump info about the file table from the drive but I would advise against doing anything without trying to take a recovery image first. I have not used fsutil to know exactly what you’d need to get that working.
As a professional who specializes in large enterprise recoveries you need exactly 3 things:
-1 drive LARGER than your current failing drive (1TB =/= 1TB across manufacturers or generations)
-a workstation that can have both the old and new drive connected simultaneously
-a bootable Linux image to run ddrescue as stated by @Adubs
What you are doing is the only way to guarantee you pay me a visit.
STOP using the drive, copy the contents in whole then prune from there.
Copy the ENTIRE disc using the command:
ddrescue -d -D -f -r /dev/sdX /dev/sdY
-d direct IO input
-D direct IO output
-f force overwrite anything on new drive
-r read from end of disc and work backwards
X - dying drive
Y - new drive
Like I said, I pulled damn near everything I need off it, and the rest is nice to have, im going to pull the drive out and put it to the side, ill probably pick up another hdd at some point and use it as a backup for what’s in their now, but ill put it in an older pc and just let it have a crack at getting more off it, worst case drive is fully dead/dies, best case I get a few more files back, put it on the new main storage drive and work from that to back up to the other drive.