I've googled and lurked 'till my eyes are crossed. Why, I'm telling you, the lenses on my glasses are overheated. But I'm left with a bunch of questions, that are eating at me. Do you guys mind if I post topics, one after another, or should I wait a while between each?


Third old drive acting up... (One crapped out a long time ago, maybe I should take another look at it).

Probably not much point trying to diagnose it online. But the (newer, unbacked-up) data isn't worth $1000.

The drive is (suddenly) a mess, after running pretty hot (doesn't burn the skin, but seems way above normal). I'm wondering if these (selected) symptoms suggest PCB problems? Or corruption of the SA?

1) WD DataLife tools showed a corrupted model name.

2) The drive is seen intermittantly. Computer's BIOS, or Promise U66. Soft reboot, reset, or power cycle. A couple of times, it's dissapeared from Windows.

3) Lots of clicking, but no discernable pattern. During drive recognition (or not). During file access (or not).

4) Room temperature has to be kept below roughly 75 degrees Farenheit, or the drive "goes away". I moved it to an open bay with a 4" ac fan on it.

(I haven't moved it to a machine that reads SMART).


Ok, I should clean the PCB (what kind of solvent, by the way?). (Where's a good place in the USA?).

If that doesn't help, is there any chance these WDC1000 PCBs (Malaysia, early 2002) would be compatible?

Failing Drive
MDL WD1000BB - 00CAA0

DCM dsehnv2aa

Potential Donor (I'd copy everything off, and sacrifice it).
MDL WD1000BB - 00CAA1

DCM dsFhCv2aH


I'm just going to clone, to image(s), with GNU ddrescue, and toss the drive.

Can ddrescue stop those blasted hardware retries? Or any way a layman can do that?

What about shutting off ECC? Layman, or strictly pro?

On the chance that there's bad cache memory, would it be wise to clone a separate image, in reverse, only? I've read that reverse shuts off the cache.

Both pcb are compatible with each other go ahead and replace them .