How Trust Worthy is this HDD

[jmone]

I've moved all data off this drive are it was showing all sorts of issues in HDS and then did a couple of Surface Reinitalise. Since then the log has been showing a heath ranging from a low of 16% to currently 97% and the log is below. Should I trust this drive at all?
Thanks
Nathan

=========
23/07/2012 3:02:29 PM,#198 Off-Line Uncorrectable Sector Count 0 -> 2
23/07/2012 7:36:19 AM,#197 Current Pending Sector Count 0 -> 2
22/07/2012 11:33:58 PM,#198 Off-Line Uncorrectable Sector Count 605 -> 0
22/07/2012 7:26:11 AM,#198 Off-Line Uncorrectable Sector Count 15 -> 605
20/07/2012 7:41:48 PM,#197 Current Pending Sector Count 605 -> 0
20/07/2012 7:36:44 PM,#197 Current Pending Sector Count 545 -> 605
20/07/2012 7:31:34 PM,#197 Current Pending Sector Count 450 -> 545
20/07/2012 7:26:29 PM,#197 Current Pending Sector Count 413 -> 450
20/07/2012 7:21:24 PM,#197 Current Pending Sector Count 318 -> 413
20/07/2012 7:16:19 PM,#197 Current Pending Sector Count 242 -> 318
20/07/2012 7:11:15 PM,#197 Current Pending Sector Count 152 -> 242
20/07/2012 7:06:10 PM,#197 Current Pending Sector Count 67 -> 152
20/07/2012 7:01:05 PM,#197 Current Pending Sector Count 15 -> 67
20/07/2012 6:55:58 PM,#197 Current Pending Sector Count 0 -> 15
20/07/2012 3:27:53 AM,#198 Off-Line Uncorrectable Sector Count 20 -> 15
18/07/2012 7:00:42 PM,#197 Current Pending Sector Count 19 -> 0
16/07/2012 7:21:31 PM,#197 Current Pending Sector Count 27 -> 19

[hdsentinel]

It seems first there were lots of issues reported, causing that the "pending" sectors constantly increase (see below for possible causes).
Then the drive reallocated these sectors to prevent any use of these sectors in the future and redirecting all reads/writes to spare area (this is when the "current pending sectors" are cleared and the number of Off-line uncorrectable sectors increased to 605).

However, it is really interesting how the counter of Off-line uncorrectable sectors reset to zero.
In some cases (depending on the type of the drive) this counter may reset to zero if the amount of problems (and their severity) is low and the status of the drive is stable. However, this is not the case because of the high number of issues - and especially because then new problems reported again.

If possible, I'd recommend to perform Disk -> Surface test -> Reinitialise disk surface in a completely different system.
The amount of weak/pending sectors reported previously (and the fact that such pending sectors reported again) may indicate an issue with the operating environment which may cause temporary errors with the drive. For example, the following may cause this issue:
- data or power cable connection issues, connectivity problems
- power supply or power line instability (fluctuation, over-voltage)
- general system instability, overclock, high temperature (not only of the drive, but in general, for example motherboard / chipset)
- system memory failure

To diagnose these, I'd recommend to test the drive in an other system. Maybe perform the Reinit. disk surface test 3-4 times and verify if the status changes in any ways. This would confirm if the status is now stable or if the drive reports new/different problems even during or after these tests.
I'd be happy if you can use Report -> Send test report to developer option (even more reports, before and after tests). This way I can also check the complete status and any changes.

If the drive status gets stable, the drive may be used, but personally I'd use it only for secondary storage, not for sensitive information.

[jmone]

Thanks - this was my WHS OS Drive that I was having issues both backing up and cloning as it had some bad/weak sectors but overall the numer was stable and not growing (and I've now replaced the disc). To test I then moved it first to a cheap e-sata HDD Docking bay on another PC to do the disk reinitalisation but found it very slow (only 5mbs) and unreliable (HDS would seem to hang part way through). I then moved the drive to an internal bay on the intel conroller and did several re-initalisations and you are correct that one of these did behave oddly showing lots of weak sectors but subsequent ones seem to be fine. I've sent a report and will do another reinitalise then send the report again but I think I'll end up just seeing if WDC will RMA it (ED20EADS).....as they have already replaced one and I've two more others with them.

[hdsentinel]

Thanks !

I checked the status and not surprised about the message

"Problems occurred between the communication of the disk and the host 6865 times."

This confirms that lots of problems may be related to communication (or connection) issues, may be related to the docking bay where the drive was working slowly and unreliable. I'm curious about the results and it is possible that RMA will work (some users reported that nowadays WD may provide warranty replacement if the hardware self-tests fail (as I see "Last Extended Self-test Result . . . . . . . . . : Test Failed By Read Element" ).

[jmone]

Thanks - it just finished the disk reinitalisaion and I sent the 2nd report. I will try RMAing the drive when I get the other two back.

[hdsentinel]

Thanks!
As I see, the weak sector is stabilized and the health is slightly improved.
Personally I'm curious to see if they will accept for RMA ....

[jmone]

FYI most of the failing drives I had were the first model using 4K sectors (WD20EARS) and the one replaced so far was with a brand new WD20EARX. Even though the current disc I'm testing is a WD20EADS, in general I've had no issue with this model. From my small sample size of about 10x2TB WD drives I'd suggest avoiding the WD20EARS like the plauge! I'm slowly moving to the Hitachi 7K4000 so I'm hoping these drives are a reliable performer.

I also now have little faith in the JMicron Sata controllers. The 2xInternal ones don't report much info and the e-sata one (while it does report the info) seems to have very slow throughput.

Also I love how when using the WHS 2011 Dashboard, HDS console loads and displays which is great as you don't have to remote into the box to see what is happening - very very good feature IMHO

[jmone]

Thanks!
As I see, the weak sector is stabilized and the health is slightly improved.
Personally I'm curious to see if they will accept for RMA ....

I'll let it run in the system till the other RMA's are delt with then give it a try and report back. I expect I'll be fine as the first one that was sucessfully RMA'ed. All I did was just report that it had bad sectors and had failed the WD Data Lifeguard Test (as per the advice on the WD website) --> http://support.wdc.com/techinfo/general/errorcodes.asp "If you encounter the same error code more than once after re-testing, we recommend for you to create an RMA. "

I was very pleasantly surprised that I got a new and not a "reconditioned" drive

[jmone]

Thanks!
As I see, the weak sector is stabilized and the health is slightly improved.
Personally I'm curious to see if they will accept for RMA ....

Sent another report as this disk is now perfect and passes both the long and short test (eg no more read element errors). So sorry, I guess I will not being trying to RMA this drive for now.

[hdsentinel]

Thanks for the info and the new report !

Yes, it seems the status is now stable and improved, so I agree that RMA is not really required.
Anyway, I'd recommend constant monitoring to check if there is any problem reported (again).