Did My HDD Just Fail??

G

gist901

Weaksauce
Joined
May 4, 2008
Messages
95
So I need a little help.

It seems as though one of my WD Green 1TB drives no longer works. The HDD still shows up in BIOS and after an extremely long boot process, the HDD shows up in My Computer.

However, there is no information of how much storage space there is. I ran Data LifeGuard and did a quick test which it passed. If I go under Disk Management the drive shows up as being Healthy. If I try and run Error-Checking it tells me that "Windows Can't Access the Disk." I also don't and never did hear any Clicking Noises coming from the HDD now, or prior.

I guess it seems as though the drive has failed or something funky is going on. If this is a HDD failure, this will be my first one ever so I really do not know what to do. Right now I am downloading SystemRescue.

So what should my next steps be??
 
Try it in a USB/Esata enclosure to convince yourself that it did indeed fail, then pack it up and RMA it back to WD. Also install and use Crystal Disk Info. It's your friend.
 
techbob said:
Try it in a USB/Esata enclosure to convince yourself that it did indeed fail, then pack it up and RMA it back to WD. Also install and use Crystal Disk Info. It's your friend.
Click to expand...

I download Crystal Disk and it says "Caution" and tells me that there are 842 Uncorrectable Sector Count.

What will be the best way to try and recover as much data from the HDD before I send it back to WD?
 
Western Digital seems to have a problem with bad sectors. I just developed a bunch of them myself on my Caviar Black and my drive is sitting next to me waiting to be sent out. GNU ddrescue (not to be confused with dd_rescue) is not a bad option if you are technically adept enough. I believe it not only requires a linux installation, but also a compilation of the source code with gcc as well. Is that inaccurate, drescherjm? I intend to play around with it once I get around to installing Linux and GCC. Again, if you have the skills it is certainly a nice program and a decent option.

GNU ddrescue attempts to reread bad sectors many times (user selectable) if necessary in a similar way to Spinrite, except I believe that while ddrescue runs you can actually do other things like surf the web in firefox or whatever. With Spinrite your computer will be out of commission for at least 3-5 days while it slowly scans your 1 TB drive. Spinrite is supposed to have some kind of voodoo or secret sauce in addition to just rereads, but not everyone believes in that.

Based on my recent experience, I would advise first downloading a freeware recovery program called testdisk. Go to "advanced". Select "boot" I think and then choose the option regarding $MFT mirroring. There is an example on the web site of exactly how to do it I think. It will check to see if your master file table and its mirror are both identical. If they are not it offers you the option to copy one to the other. The problem that I had with it is it seemed to think the mirror was the one that was bad. After copying over the mirror however, chkdsk /f worked again and fixed a bunch of file system problems. If this happens to you I do not recommend overwriting the mirror unless you have already made a backup of it, but if the original $MFT is bad I would recommend using the mirror to fix it. That's what the mirror is for.

If you are able to restore your MFT from the backup mirror then run chkdsk /f again. If that works then I would try to copy as many files as you can with an improved file copy tool like teracopy/richcopy/robocopy or unstoppable copier. Files that will not copy at that point are likely to not be recoverable if bad sectors is your problem.

If restoring the $MFT doesn't work then I would proceed directly to a recovery tool like GetDataBack, Testdisk, Recuva, Active@, or even ddrescue. But this is just for an initial iteration. A sort of temporary backup in case your drive gets worse with time and reads/writes as it probably will.

If you have data that you feel you really must recover you can attempt to use Spinrite, GNU ddrescue set for many repeat reads, or even the highly questionable HDD Regenerator and then a file copy tool, although this last option didn't work for me and I suspect it of being scamware. I ended up giving up on some of my data and just copying the rest with teracopy. Luckily my bad sectors were in files that I could afford to lose or just redownload. It worked better for me than the attempt at using GetDataBack which just skipped any files with bad sectors. Spinrite would have taken over a week in my case which didn't seem worth it to me since I would have lost the use of the computer for that time.
 
Western Digital seems to have a problem with bad sectors.
Click to expand...

So does Seagate, Samsung, Hitachi ...

I see these unreadable sectors on drives from every single manufacturer at work.


I believe it not only requires a linux installation, but also a compilation of the source code with gcc as well. Is that inaccurate, drescherjm?
Click to expand...

No. Just boot off a live cd containing ddrescue. Like sysrescuecd (which can be installed to a usb stick)

http://www.sysresccd.org/Main_Page

Then in the console determine what drive is your source disk and what drive is your destination.

fdisk -l /dev/sd?

can help by showing you the partition table for all disks in the system.

also if you want to look at the serial number for a drive

hdparm -I /dev/sdX

where X = a, b, c ... the drive you want to look at. Note: The ordering need not be the same as in windows.

smartctl --all /dev/sdX

can give you the SMART for the drive which includes the serial number.

then when you are absolutely certain which drive is the source and which drive is the destination you do

ddrescue source destination /root/logfile.txt

where source is your source drive /dev/sdX (again make sure you get this right the program will not care if you reversed things and are copying an empty disk to a full disk.):

and destination is your destination drive

an example would be

Code: 
ddrescue /dev/sda /dev/sdb /root/log.txt

That assumes /dev/sda is the source and /dev/sdb is the destination.
 
Last edited:
So I don't have enough file space right now to run ddrescue. However what I did do was I booted up Ubuntu and then copied my Music Folder successfully from the bad drive to my 2TB drive :D.

I have RMA'ed the drive with WD, there sending me a new one. Once I get the new one, I will run ddrescue.

Thank you drescherjm for telling me how to program it.
 
You must log in or register to reply here.
Back
Top