Hard Disk Monitoring

I have a Western Digital 3TB external drive which died on me a 2 weeks ago, then miraculously returned to the land of the living.
The first thing I did when this happened was I purchased a new drive to back the data up to. The original drive seemed fine at first,
but I soon started noticing errors in the Windows management console.

Errors such as:
The driver detected a controller error on \Device\Harddisk3\DR3.
The device, \Device\Harddisk3\DR8, has a bad block.

I thought it best to use the manufacturer's diagnostic tools instead of just trusting the Windows Logs, as the info might be incorrect.
I tried running the WD Data Lifeguard Quick Test, but it failed in under a minute with the Test Error Code:

06-Quick Test on drive 3 did not complete! Status Code = 07 (Failed read test element), Failure Checkpoint = 97 SMART self-test did
not complete on drive 3!

I then decided RMA would be for the best, after all, if it fails the manufacurer's diagnostic so quickly, something must be wrong. I
ran the Write-Zeros tool to erase my data from the drive and prepare it for return. After 24 hours the erase completed and I ran
the Quick Test again. Now it Passes!

Onto Hard Disk Sentinel. Hard Disk Sentinel tells me that the drive has 2 bad sectors (98% health) and that it's too early to RMA the
drive.

So now I'm torn as to whether I should attempt to return the drive or keep it. I don't like the idea of keeping a drive that's less than
a year old and has already developed bad sectors. But I also don't want to be reliant on only a single copy of my data for too long.
My lesson was learnt about backing up the first time this drive died.

What do you think? am I wasting my time trying to RMA a drive which passes DLG diagnostics or should I do it.

I haven't actually attempted to run the DLG extended test as I know it's going to take another 24 hours to get through.

Thanks in advance for any advice.

> I thought it best to use the manufacturer's diagnostic tools instead of just trusting the Windows Logs, as the info might be incorrect.

Sorry to say, but both are wrong ideas if discuss about the hard disk status in general.

Windows LOGs and tools provided by Windows (chkdsk / scandisk) always verify and report problems related to the current FILE SYSTEM, instead of the actual hard disk. Even if sounds surprising, this may be independent form the actual hard disk status and actual hard disk problems.

The tools provided by the manufacturer checks the hard disks from the manufacturer's viewpoint, but it does not provide any real information about the status of the hard disks, how the disk surface can be read / written and how "good" the drive performs. You may get a quick error, but nothing more.

This is why it is much better to use special tool which is independent from both Windows (and the file systems) and the manufacturers and reveal and fix problems. The tests in Hard Disk Sentinel are the best for this purpose, as it has both the tests included in the tools provided by the manufacturer (these are Disk -> Short self test, Disk -> Extended self test) but also offer much more tests.

Please check: http://www.hdsentinel.com/faq.php#tests
for more information about the _proper_ way to test the hard disks, diagnose them to reveal and fix possible errors.
This is especially recommended before using the hard disks to store important data - and even more recommended for used / RMA-replaced drives which may have further problems.


> I then decided RMA would be for the best, after all, if it fails the manufacurer's diagnostic so quickly, something must be wrong.
> I ran the Write-Zeros tool to erase my data from the drive and prepare it for return. After 24 hours the erase completed
> and I ran the Quick Test again. Now it Passes!

This is completely normal and not surprising
Previously the issues showed by Windows even log were related to "weak sectors". These are sectors which seem damaged and would need to be repaired. Windows (chkdsk or similar) never fixes them and they can cause further problems.

The best way to fix them is using Hard Disk Sentinel -> Disk menu -> Surface test -> Reinitialise disk surface test as this forces the hard disk to examine the status of these sectors and re-allocate them if required. This means that if the sectors are really damaged, then the hard disk marks them and they never accessed in the future: all further reads and writes targeting these sectors are redirected to a spare area.

Sometimes simpler test methods (for example the write-zeroes, which is also included in Hard Disk Sentinel) is also enough for this: the "weak" sectors examined and fixed - and then even the short self test / extended self test (started in Hard Disk Sentinel or in the manufacturer-specific tool) completes without problems.

> Onto Hard Disk Sentinel. Hard Disk Sentinel tells me that the drive has 2 bad sectors (98% health) and that it's too early to RMA the drive.

Yes. The 2 bad sectors found, because they replaced with the spare area. So these will no longer cause problems.
Even if you feel bad about the situation as the drive is not perfect, the drive is still "Excellent". If there are no new problems reported, then I'd recommend to keep the drive as these 2 bad sectors are completely acceptable - and there is risk that you'd even get a worse drive if you start RMA.


For more information about these "bad sectors", please click on the "?" next to the text description (the text area describing the issues) and check

http://www.hdsentinel.com/faq.php#health

For more information about the tests should be used to perform complete verification of the hard disk, to reveal and fix possible problems, please check:
http://www.hdsentinel.com/faq.php#tests

This page describes how it is possible to reveal the problems and if you confirmed with Hard Disk Sentinel that the status is stable, the current disk surface is fully usable, then you can even acknowledge the errors and clear the error counters:
http://www.hdsentinel.com/faq_repair_ha ... _drive.php
Then Hard Disk Sentinel will display only possible new problems.

For more information about the weak sectors, how they affect the logical drives and how this causes that the test(s) may fail, and after using Hard Disk Sentinel to repair the problems, how the problems fixed, how the tests run without problems:
http://www.hdsentinel.com/hard_disk_cas ... ectors.php

I'd recommend to use Report menu -> Send test report to developer option (even many times, now and if you ever see changing status). Then it's possible to check the actual complete status (and change in the status) of the hard disk and advise.