Storage Hard drive almost dead

Hi,

I have a 400GB Western Digital hard drive and an assortment of 3x 1TB hard drives (all SATA) in USB enclosures all combined into an lvm2 filesystem.

Yesterday the 400GB hard drive started losing power by itself, the light on my USB hub would go off indicating that it thinks nothing is connected, and I could hear the drive spinning down then spinning back up again a second later and the light coming back on.

It eventually managed to sort itself out and seemed okay until today, now it is doing the same thing again. It cycles about 3 times then the light stays on but the drive is unresponsive. Occasionally it manages to mount the filesystem for a split second until the drive powers down again, but usually it gets as far as reading the partition table then keeping power but not going any further.

What I want to do is try and get all of the data off then replacing it, and eventually moving from lvm2 to RAID when I put these disks into my almost complete media server.

Also, the drive seems to be warmer than usual if that would make any difference. And, yes, I have swapped enclosures, power cables, USB ports etc. etc. etc.

Would it be wise to carry on power cycling the drive in the hope that it will eventually work, or should I take it to someone that knows what they are doing in a shop? Then again, people in shops rarely know what they are doing.

EDIT: This is the dmesg output when the drive becomes unresponsive

Code:

[ 2162.451732] usb-storage: device found at 66
[ 2162.451738] usb-storage: waiting for device to settle before scanning
[ 2167.448327] usb-storage: device scan complete
[ 2167.490423] scsi 66:0:0:0: Direct-Access     WDC WD40 00AAKS-00TMA0         PQ: 0 ANSI: 2 CCS
[ 2167.491098] sd 66:0:0:0: Attached scsi generic sg3 type 0
[ 2167.492903] sd 66:0:0:0: [sdg] 781422768 512-byte logical blocks: (400 GB/372 GiB)
[ 2167.494657] sd 66:0:0:0: [sdg] Write Protect is off
[ 2167.494662] sd 66:0:0:0: [sdg] Mode Sense: 34 00 00 00
[ 2167.494665] sd 66:0:0:0: [sdg] Assuming drive cache: write through
[ 2167.497245] sd 66:0:0:0: [sdg] Assuming drive cache: write through
[ 2167.497253]  sdg: sdg1 sdg2 sdg3

And then nothing happens, but the indicator light on the front of the drive suggests disc activity.

 

Are all the drives running from on the USB bus?

If so, is it worth putting them all on direct IDE/SATA connections to eliminate USB problems as the cause?

 

The other drives work fine, as does the enclosure. Not sure it would help. And I don't have an available direct SATA connection to test it on

 

Taking it to a "shop" (e.g. a data recovery specialist) will be expensive. I assume it contains data that you need?

If you have checked all the connections thoroughly then I would assume the drive is dying. I am not sure what else to suggest, sorry.

 

I have all the important stuff backed up, so I am not going to spend a lot of money getting it back.

Asking around it seems to be the controller that has died, so it should just be a matter of finding an identical drive and swapping the boards.

This is apparently quite common in Western Digital drives.

 

Is it worth it for a 400GB drive? How old is it? If new, surely it is under warranty? If old, why bother when a 500GB Samsung, for example, is £35...

 

If I can't get one cheap I won't bother. Even one which is broken but in a different way would do.

 

Maybe change the thread title to "WD400GB HD wanted"?

 

I am writing a post for the Wanted section as we speak.

EDIT http://forums.bit-tech.net/showthread.php?p=2333831