Western Digital 15EADS w/bad sectors

[Raduque]

My computer was powered down unexpectedly a couple weeks ago (power company changed meter without warning anybody), and my HDDs weren't cleanly unmounted by Windows. They all handled it fine except the 1.5tb WDC15EADS. Now, it's exhibiting bunch of CRC errors and has about 5500 kbytes (per chkdsk) in bad sectors. I've been running chkdsk /r on it, and while it keeps recovering files, the bad sector count doesn't seem to grow.

My question is, should I junk this thing (RMA it) or would it be fine after a zero-fill format?

Edit: oh yeah, it has about 90gb free on it... this is a storage drive for DVRed content, and it was being written to when the power was cut.

 

[F1shF4t]

Check SMART data. If it was being written to, its most likely that the data was corrupted but the drive itself is intact.

Nowdays its almost impossible to kill a drive just by removing power while its operating, as the heads will park properly.

 

[Raduque]

What am I looking for in the SMART data?

edit: Crystaldisk shows "Caution" on the 15EADS, and a yellow dot

"Reallocated Sector Count"
Current 198
Worst 198
Threshold 140
Raw Values 00000000000E

"Current Pending Sector Count"
Current 195
Worst 195
Threshold 0
00000000051E

"Uncorrectable Sector Count"
Current 195
Worst 195
Threshold 0
0000000004E1

 

[RebateMonger]

Those raw values are a lot. I don't have ANY Reallocated Sectors, Pending Sectors, or Uncorrectable Sectors on any of my seven 1, 1.5, and 2 TB disks, some of which have been running continuously for two years now. My big disks include WDC, Seagate, and Hitachi (but mostly Hitachi).

If it was my disk, I'd replace it.

 

[Mark R]

Those smart values show no hard evidence of drive damage, but show a bunch of corruption.

Uncorrectable sectors are sectors that contained garbage data that the drive could not interpret. This could be because the sectors are faulty, or because of a power glitch during saving. Because the drive was commanded to read hose sectors, and couldn't, it told the OS 'unreadable (bad) sector' on 4 occasions.

The 'pending sectors' (81 of them) are ones that are 'probably corrupted' but still readable, pending further analysis by the drive.

There are 0 reallocated (actually bad) sectors on the drive.

You don't need to do anything specific. However, some files may have chunks missing (where the sectors are unreadable). If you run a full chkdsk surface scan, it should identify the damaged files. Alternatively, you could just zero fill the drive, this will force the drive to analyse and repair every sector.

 

[Raduque]

Thanks folks. So I'm guessing the drive will be fine if I zero-fill it?

 

[FishAk]

A 30 or 40 dollar UPS will give you enough battery to run your system and a monitor for a couple minutes without power. Not much, but enough for a crash shut down that will prevent future problems. It also will arrest any power fluctuations that could otherwise hurt your system.

 

[RebateMonger]

There are 0 reallocated (actually bad) sectors on the drive.

What am I missing here?

0E Reallocated Sectors (fourteen) isn't the same as 00 Reallocated Sectors (zero), is it?

I've been running chkdsk /r on it, and while it keeps recovering files, the bad sector count doesn't seem to grow.

My understanding is that "chkdsk /r" looks for read errors and if it has problems reading, it does what it can to read the data and to move the data elsewhere. At that point, that sector probably goes into the "Pending Sector Count" and won't be added to the "Reallocated Sector Count" until a write is attempted to that sector and it doesn't verify properly.

I was wondering how a power outage could generate over one-thousand Uncorrectable Sectors, but I see that a single "bad" sector will cause the entire Cluster to be marked as bad. That obviously multiplies the "bad" Sector count considerably.

 

[JAG87]

Just replace it dude. Do an advance RMA with WD, get your replacement drive first, transfer all recoverable data, and send the trouble maker back.

 

[Mark R]

What am I missing here?

0E Reallocated Sectors (fourteen) isn't the same as 00 Reallocated Sectors (zero), is it?

Umm. No.

My mistake. One of the perils of iphone browsing is that it highlighted the 00000 but not the E so I didn't read the E!

That's quite a lot of pending sectors, when counted correctly!

Zero fill the drive and see what happens. After it's been zeroed run WD's drive fitness test on it.

If it fails, RMA.

 

[razel]

I don't think WD will RMA a drive just because of bad sectors like here where it isn't a manufacturing defect. Of course, you could carefully choose facts to tell WD, but I wouldn't, they've been good to us and karma is a btch.

What I would do is remap those bad sectors with good ones. It should do this automatically, but not always or perhaps you ran out. So I would run WD's Data LifeGuard Diagnostic. Run the short and long test. If it remapped any bad sectors, then you may want to clear those in the file system. If you're on NTFS, do a chkdsk /b. If you're on FAT32, forget about it. It requires a format, unless someone knows otherwise. Just remember next time to reformat it next time when you need to. Have fun.

 

[RebateMonger]

IIf you're on NTFS, do a chkdsk /b.

That's a new one. Looks like it started with Vista.

 

[Raduque]

Well, I managed to actually pull a good 97% or so of the data off. Only about 18gb of ~1100gb couldn't be read.

I'm doing a zero fill using HDD Wipe from hddguru.com. Been running it since 12pm and it's about maybe 35% or so done. It already failed the long DST (passed a short DST) in Seatools, so I'll try the WD tools after this zero fill is done.

 

[VirtualLarry]

I didn't know SeaTools would run on a WD drive...

 

[RebateMonger]

I didn't know SeaTools would run on a WD drive...

As far as I've seen, all of the current disk makers' diagnostics tools will work with any brand of disk. Restrictions usually come with the disk-cloning tools, where one of the two disks has to be made by the distributor of the cloning tool.

Well, I managed to actually pull a good 97% or so of the data off. Only about 18gb of ~1100gb couldn't be read.

I'm doing a zero fill using HDD Wipe from hddguru.com. Been running it since 12pm and it's about maybe 35% or so done. It already failed the long DST (passed a short DST) in Seatools, so I'll try the WD tools after this zero fill is done.

Sounds like you are making some progress. Good luck with that and please let us know what you find. Thanks!

 

[Old Hippie]

It already failed the long DST (passed a short DST) in Seatools,

There's no way I'd be using that drive.

 

[Raduque]

How long is HDD Wipe Tool supposed to take? It's been running for over 30 hours.

edit: Old Hippy, I'm not really trying to USE the drive, was just getting as much of my data off as I can. I'm doing the HDD Wipe now to zero out all the sectors, and force the drive to do any re-allocating as it needs to. Then I'm going to run the WD diag tools, and if it fails them, off to RMA it goes.

I was having some weirdness with the mft and usn journal. Chkdsk was having to repair both every time I ran it.

OK, well the drive is officially toast. Even WINDOWS knows there's something wrong with it.

 

[RebateMonger]

OK, well the drive is officially toast. Even WINDOWS knows there's something wrong with it.

It happens, even with the big WDC disks, which seem to have been reasonably reliable. Hopefully the new Advanced Format disks, with better ECC capability, will give us a bit of reprieve from early disk failures in those big disks.

 

[Raduque]

It happens, even with the big WDC disks, which seem to have been reasonably reliable.

I know it happens, but for me, it only seems to happen when using WD disks.