I have a failed Western Digital WD5000AAKS. It worked fine for a year in D-Link DNS-323 NAS. The other WD5000AAKS is fine, but I guess it has not been abused enough to die.
The symptoms/troubleshooting:
- some files became inaccessible, or the access would drop during disk I/O transaction. I was mainly using this drive to store movies and music to share it accross the network. My media centre box is connected to router through wireless AP. DNS-323 NAS is connected with ethernet cable to the same router. So when the video playback issues occurred I've attributed it to the bad wireless connection. By the time I've figured it out it was too late... I was trying access files from a PC which is connected the same router - some file were not showing.
- shut down the DNS-323 then restarted it - the drive disappeared.
- pulled the drive out of DNS-323 enclosure; hooked it up to my PC's motherboard with the SATA cable, hooked up power to HDD; powered on PC; drive seemed to power up but it had been detected by BIOS as unreadable; restarted computer again with my ear on the HDD; I could here spin up cycle (3x) and then drive would just shut down; the heads do not make much noise; the drive is actually quiet when it tries to spin up. I've recycled the power few times and eventually the drive came up. I keep the PC powered up and booted into Linux distro (DNS-323 formats drives as ext2) since then (it's been 5 days already).
- I was able to recover some (a lot of files) but not all. Some of files are unrecoverable, i.e. copy cycle start and then I see bunch of hardware errors in dmesg log related to the disk, or reading slows to down to a crawl 10 - 25 kBytes/second; try to copy 380MBytes file like that.

What are my options to recover the rest of the files. Is there anything I can do to improve "readability" of the affected areas?

As a aftermath of the problem I will:
- User RAID1 (mirror) set-up in this enclosure.
- Pull drives periodically and check smart codes once in a few months. DNS-323 is suppossed to detect drive failure, but I don't think it's doing a good job; when I ran smartctl -all /dev/sda once the drive has been running hooked up to PC, it showed quite a few errors.

I have dealt with a lot of drives with that problem.

Your drive must have a lot of PRON on it and many viruses from that.

I would recommend using brand new drives for your server, because drives get infected very often with those viruses.

Pro. help is out of the question, as I understand?

Yes, it would be expensive to recover your infected files, if anyone would even agree to do it. It takes a lot of time, effort and drives to get those files back to new and healthy drive.

P.S. yes! you have posted to wrong forum, try a Dr. Aurora section instead.

_________________
www.datarecoveryne.com

PRON?? or PORN ??

You know exactly what i mean

_________________
www.datarecoveryne.com

Sorry, but no PRON and no viruses; mostly cartoons, mp3 and pdf files.
Drive has a skyrocketing count of Raw_Read_Error_Rate - 184576 (Raw value) and UDMA_CRC_Error_Count - 245 (Raw value) according to smartclt scan.
Viruses "usually" do not cause drive start-up problems (takes up to 10 power ups for the drive to kick in). This does not sound like logical explanation.
Other takes on the issue?

Drive is failing. Maybe failing head(s). It's probably only going to get worse.

_________________
You don't have to backup all of your files, just the ones you want to keep.

Does not matter how often you check your drives, the fact that electronics fail is in tact!

You either pay for data storage or you store your data, its a cost issue.

You have an option to take to a pro ASAP and pay few hundred bucks or........ if you will wait....
take to a pro and pay $1,000 plus.....

_________________
www.datarecoveryne.com