Strange bug on a 1 TB WD Caviar Green

Hi,

I have been using a caviar green 1 TB on a Windows XP SP3 system for 2-3 years (WD10 EARS-00Y5B1).

This is not my system disk.

This is the disk where I’m storing my pictures, downloads aso.

Lately I have noticed that one file is making my PC freeze. I mean that each time I’m trying to access this file, to copy it on a USB key for exemple, or to make a copy of it on the same HDD (or on an other), the explorer process that tries this freezes and the OS becomes unstable after this even if I kill that explorer process.

When I ask the OS to check that drive at reboot, the check (chkdsk) freezes at 23 % (step 4 of 5).

So my question is : what should I do ? Is that disk jeopardizing my data or can I still be confident in him ?

There is only 1 buggy file (at this time) : an AVI of 700 KB  (a single film that is not important at all to me).

I don’t want to erase this file, since this will make the sectors available for another new file (and this is surely a hardware bug). But I also cannot keep this file in my file system, since this may lead copy processes of the whole disk to freeze (and virus scans also).

Has anybody got an advice ? Is this a known bug ? Why is the OS freezing and not “correcting” the bug ? Why is even CHKDSK freezing ? Is this a firmware bug ? Should I try to flash firmware ?

Is there a utility I could use to better check this drive or to mark this file as a set of bad sectors (NTFS) ?

Thank you.

In fact I’m not sure that I really manage to kill the explorer process that freezes while copying the file …

I even cannot shutdown after that : it looks like an kernal mode freeze of a process.

Hi you need to run a test on the drive you can use WD data lifeguard program. Or get a copy of Hard Disk Sentinel and test with it. You can get the windows Data Lifeguard program here http://support.wd.com/product/download.asp?groupid=608&lang=en or the Hard Disk Sentinel trial here http://www.hdsentinel.com/  see what report it gives and post back the results.

Ok, I did the test with Lifeguard.

The Quick Test finished without any error.

Résultat test rapide.png

But the Extended Test froze (time continues to be updated but the sector number freezes) at sector 1482752255 of 1953525167. This is at 75,90 %.

So it looks like CHKDSK is reading the sectors the opposite way (CHKDSK froze at 23%).

After that I had to hard-reset the computer (as usual, when the fault is showing).

So there is no report of the Extended Test since the program froze.

There are two faulty files on that disk. I did a complete backup of that drive on another drive of the same kind, and so, I can now say that every single file has been read successfully … except the two faulty movies.

SMART Disk Info :

SMART Disk Info.jpg

I did a copy of the SMART info before the Extended Test. The numbers are exactly the same before and after the freeze and reboot.

Hi your smart data picture does not show if it has the serial number in the picture, that is why it wont show otherwise we wait for a mod to fix it. Please give the Hard Disk Sentinel program a try, it will give a report as soon as it starts no long test and see what it says

There is no OS or software installed on that disk. I wiped the serial number in my latest upload.

Downloading sentinal …

Sorry for the french user interface.

This doesn’t look good. I’m quite disapointed. This disk is out of warranty.

I think I’m going to swap the two WD disks I have : the one in my desktop and the one in the external USB enclosure.

Is there a tool that i can use to mark the 975 bad sectors as bad (from the OS point of view) ? (so I can still use the disk to save some non critical data ? It looks like there is a scratch on the plate that affects less than 2% of the capacity : CHKDSK and LifeGuard went both through the disk without any problem, until they reached a specific zone between 23 and 25 %)

Is there a tool I can use to see the sectors used by the two faulty files ? They were both writen the same day (two episodes of a TV show that I downloaded the same day, and that I wrote on that disk. These sectors were “brand new” before that day)

The important point is that, even when using the WD tool “lifeguard” to test the disk, the disk freezes and makes the whole computer freeze. I mean: isn’t there a quality problem with the firmware ? Is this a known problem ? Is there an update ?

It’s the firmware that is freezing, not the OS.

What about those tools that even cannot handle an asynchronous read with timeout ?

There must be a lack of ingeneering somewhere if you cannot map the faulty sectors of a hard disk, and I’m not sure it is at the OS level this time.

Every disk manufacturer has to manage plate scratches, read errors, aso. But the way this issue rises to the OS level and in the manufacturer tools is absolutely disapointing.

It looks like you jeopardize the whole system when you have one of those Caviar Green disks installed.

In fact, I bought four pieces the same day, 3 years ago. I was a Western Digital preacher at this time.

Now I have to check them all because the way this issue freezes the PC is not acceptable.