Prev: Cloning SATA Drives?
Next: Why 2 different WD models? 300GB 10k RPM SATA
From: Yousuf Khan on 19 Mar 2010 08:59
Frank Williams wrote:
> Its NOT Bad Sectors its Reallocated Sectors, if you scan the drive there are
> No bad Sectors I never ever use DOS chkdsk as its very buggy and can kill
> hard drives.
>
> The SMART Data no longer lists any thresholds for that drive but does with
> all the others, that's using Hard Disk Sentinel, must be a Smart Bug or a Hard
> Disk Sentinel one.

Yeah, that is kinda weird. Are you saying that this entire column is
missing when you view it on HDS for this drive, but not for others? Or
are you saying that the column is still there, but every row in it are
zeroes?

You might want to ask the HDS developers this question, they'll probably
ask you to run and send their Test Report. It would be interesting to
find out what this is about in any event.

In the meantime, trust but verify. Use a different utility on the drive.
Run HDD Scan, and see what it reports.

HDDScan � free HDD test utility with USB flash and RAID support
http://hddscan.com/

> On the Other Ext Seagate it list 36 for the Threshold Reallocated Sectors
>
>
> The problem is that most SMART monitors do not work on USB drives.


Seagate's SeaTools seem to work on a lot of USB drives. And HDD Scan
seems to work too. At one time, HD Sentinel was the be-all-end-all of
USB-based SMART reporting, but there are others now.

I still like HD Sentinel for its health and performance ranking system,
although its weighting criteria seems to be a bit too conservative. I
have one old 200GB IDE drive that was given 100 days to live, but that
was almost 2 years ago. I've also now converted the drive into a SATA
drive, and it's had no problems since. HDSent has steadily been
increasing its health rankings since then; it's gone from a 40% health
ranking, up to 49%, and now up to 76% and its estimated life is now 336
days.

Yousuf Khan
From: Rod Speed on 19 Mar 2010 12:24
eyes wrote
> Rod Speed wrote
>> eyes wrote
>>> Frank Williams wrote

>>>> Hard Disk Sentinel shows it as 95% good, 103 days of use.

>>>> Now has 5 Reallocated Sectors

>>>> The last one No.4 was in November 2009 and now one today..

>>>> But I did get a Blue Screen what trying to transfer XP Dater using
>>>> the WIN 7 Windows Easy Transfer tool, did show up as a USB
>>>> problem, with copying to this drive, but that was the day before,
>>>> I then decided to use a 4G USB stick and all was OK, 2.4G of data
>>>> transferred. Drive is about 9 months old and SMART no longer lists
>>>> any thresholds for that drive..

>>>> Do I need to worry and what is the limit given to get a replacement..? Thanks

>>> Its a tough one. Most vendors only accept returns if a manufacturer's drive test says it is failing (which is
>>> usually a SMART threshold
>>> check). I'm not sure that a count of 5 would be enough.

>>> Keep an eye on it, but to be fair, you should keep an eye on all of your drives.. you never know when they are about
>>> to fail.

>>> Anecdotal evidence:

>>> 1. I have a 200GB drive with around 20 reallocated sectors. It got
>>> them near the beginning of its life and hasn't changed after 5+ years of use. 2. I have 2 Seagate 1.5TB drives.
>>> These were purchased 1
>>> year ago, and currently show 70 and 107 reallocated sectors.

>>> Seatools says they are still okay. I'm not sure I feel the same - although I guess these big drives have a lot of
>>> sectors, so those counts are a very small percentage of the total.

>> Those are obscene numbers of reallocated sectors.

> Not sure I would call that obscene.

They are anyway.

> On these drives 1 sector = 512 bytes. 107 x 512 bytes = ~54K. 54K in 1,500,000,000K is probably not significant to
> worry about.

Wrong. You shouldnt see more than a very small number
of reallocated sectors on any modern hard drive.

>> That isnt necessarily due to the drive dying tho, when you have so
>> many on multiple drives, its much more likely to be due to something
>> external to the drive like the temperature or the power supply etc.

> Exactly, although I have had a 500GB Seagate and a Samsung 1.5TB in
> the same case. Those drives are not reporting any reallocated sectors.

Particular drives can see quite different temperature regimes even when
in the same case, most obviously with a drive that is on the end of the
stack or one which is in the airflow from a fan on the end of the stack
or with drives that are in the 5.25" drive bay stack in a multidrive housing.

Bet you will find that the SMART temp is much worse with the
drives that have those obscene numbers of reallocated sectors.

> Of course that doesn't mean their firmwares handle reallocated sectors differently and only report in SMART when they
> hit an internal threshold.

Smart doesnt do it like that with reallocated sectors.


From: Christian Franke on 20 Mar 2010 09:25
Yousuf Khan wrote:
> Frank Williams wrote:
>>
>> The problem is that most SMART monitors do not work on USB drives.
>
>
> Seagate's SeaTools seem to work on a lot of USB drives. And HDD Scan
> seems to work too. At one time, HD Sentinel was the be-all-end-all of
> USB-based SMART reporting, but there are others now.
>

This is because many recent USB->SATA bridges support SAT (SCSI/ATA
Translation) standard or vendor specific ATA pass-through commands.

See:
http://sourceforge.net/apps/trac/smartmontools/wiki/USB
http://sourceforge.net/apps/trac/smartmontools/wiki/Supported_USB-Devices

--
Regards,
Christian Franke
First  |  Prev  | 
Pages: 1 2
Prev: Cloning SATA Drives?
Next: Why 2 different WD models? 300GB 10k RPM SATA