HDD suddenly failed, please help

"Rod Speed" <rod.speed.aaa.DeleteThis@gmail.com> wrote in message
news:49ea3gFo6nqoU1@individual.net...
> Gary Read <guffygumpster.DeleteThis@hotmail.nospam.com> wrote:
>
>> I have an IBM IC35L120 HDD, set up as slave on IDE2, running Windows
>> XP SP2. I have an Asus A7V333 mobo. All was well until yesterday when
>> my system slowed then ground to a halt. I rebooted and then did not
>> have the drive available to me in My Computer. The drive shows up in
>> Device Manager with no warnings but in the Computer Management
>> console it is shown as 'Offline'. The first warnings I can see in
>> Event Manager is this: ('Gary' being the name of my computer)
>>
>> "Disk group GaryDg0: Errors in some configuration copies: Disk
>> Harddisk2, copy 1: Block 1: Invalid magic number in the configuration
>> copy"
>> followed by:
>>
>> "INTERNAL Error - The disk group contains no valid configuration
>> copies (C10000B6)"
>>
>> then this:
>>
>> "The driver detected a controller error on \Device\Harddisk2\D."
>>
>> The only thing I can do in the Disk Management console is attempt to
>> reactivate the disk but Windows tells me that 'The operation did not
>> complete'.
>>
>> The disk is just over a year old - I would be very grateful if anyone
>> could help me to get this thing going again.
>
> Looks like the drive has died. See what Hitachi's
> Drive Fitness Test says about the drive.
>
> Hitachi bought the IBM hard drive business.

Thank you for your reply. I downloaded the utility and ran both the basic
and advanced tests, both of which returned a disposition code of 0x00 which
translates to 'No errors'. I wonder if this is good or bad news?
 >> Stay informed about: HDD suddenly failed, please help 

Svend Olaf Mikkelsen wrote:
> On Wed, 5 Apr 2006 15:29:52 +1000, "Rod Speed"
> <rod.speed.aaa.DeleteThis@gmail.com> wrote:
> >> 05 Reallocated Sector Count 5 100 100 1245192
> >> OK: Value is normal
> >
> >Urk, thats the problem, thats an utterly obscene reallocated sector count.
>
> The value 1245192 is hexadecimal 0x130008. The value are from the 8
> vendor specific bytes, which typically are divided into an 8 bit field
> (the reported "low" value of 100), two 16 bit fields and a 24 bit
> field.
>
> I would say that the number interprets two Smart 16 bit fields. One of
> 0x0013 = 19, and one of 0x0008 = 8.
>
> This seems to be a logical data recovery problem only.
> --
> Svend Olaf

Hi Svend

Things are getting a bit over my head now Is there something I can
do to repair the disk?

Many thanks

Gary
 >> Stay informed about: HDD suddenly failed, please help 

On Wed, 5 Apr 2006 15:29:52 +1000, "Rod Speed"
<rod.speed.aaa.TakeThisOut@gmail.com> wrote:
>> 05 Reallocated Sector Count 5 100 100 1245192
>> OK: Value is normal
>
>Urk, thats the problem, thats an utterly obscene reallocated sector count.

The value 1245192 is hexadecimal 0x130008. The value are from the 8
vendor specific bytes, which typically are divided into an 8 bit field
(the reported "low" value of 100), two 16 bit fields and a 24 bit
field.

I would say that the number interprets two Smart 16 bit fields. One of
0x0013 = 19, and one of 0x0008 = 8.

This seems to be a logical data recovery problem only.
--
Svend Olaf
 >> Stay informed about: HDD suddenly failed, please help 

Svend Olaf Mikkelsen <svolaf.TakeThisOut@partitionsupport.com> wrote
> Rod Speed <rod.speed.aaa.TakeThisOut@gmail.com> wrote

>>> 05 Reallocated Sector Count 5 100 100 1245192
>>> OK: Value is normal

>> Urk, thats the problem, thats an utterly
>> obscene reallocated sector count.

> The value 1245192 is hexadecimal 0x130008. The value are from the
> 8 vendor specific bytes, which typically are divided into an 8 bit field
> (the reported "low" value of 100), two 16 bit fields and a 24 bit field.

What makes you think that 05 is vendor specific ?

> I would say that the number interprets two Smart 16 bit fields.
> One of 0x0013 = 19, and one of 0x0008 = 8.

Thats not how 05 is defined.

> This seems to be a logical data recovery problem only.

It isnt the only evidence of a drive problem, there are other
SMART fields that also indicate that the drive has a problem.

> 05 Reallocated Sector Count 5 100 100 1245192
> OK: Value is normal
> C4 Reallocation Event Count 0 100 100 9
> OK: Always passing
> C5 Current Pending Sector Count 0 100 100 1
> OK: Always passing

And its a DeathStar.
 >> Stay informed about: HDD suddenly failed, please help 

On Thu, 6 Apr 2006 03:50:26 +1000, "Rod Speed"
<rod.speed.aaa.RemoveThis@gmail.com> wrote:

>Svend Olaf Mikkelsen <svolaf.RemoveThis@partitionsupport.com> wrote
>> Rod Speed <rod.speed.aaa.RemoveThis@gmail.com> wrote
>
>>>> 05 Reallocated Sector Count 5 100 100 1245192
>>>> OK: Value is normal
>
>>> Urk, thats the problem, thats an utterly
>>> obscene reallocated sector count.
>
>> The value 1245192 is hexadecimal 0x130008. The value are from the
>> 8 vendor specific bytes, which typically are divided into an 8 bit field
>> (the reported "low" value of 100), two 16 bit fields and a 24 bit field.
>
>What makes you think that 05 is vendor specific ?

The only documented values are the threshold value 5, and the
attribute value 100. When threshold value is lower than the attribute
value, the status for the attribute is considered OK, by definition.
Most likely the number of reallocated sectors is not 1245192, a number
which comes from the vendor specific field for attribute 05 and is not
documented in the Hitachi specifications for the disk.

>> I would say that the number interprets two Smart 16 bit fields.
>> One of 0x0013 = 19, and one of 0x0008 = 8.
>
>Thats not how 05 is defined.

No, we have to guess.

>> This seems to be a logical data recovery problem only.
>
>It isnt the only evidence of a drive problem, there are other
>SMART fields that also indicate that the drive has a problem.
>
>> 05 Reallocated Sector Count 5 100 100 1245192
>> OK: Value is normal
>> C4 Reallocation Event Count 0 100 100 9
>> OK: Always passing
>> C5 Current Pending Sector Count 0 100 100 1
>> OK: Always passing
>
>And its a DeathStar.

A pending sector may be a bad sector. In this case Drive Fitness Test
however did not find any bad sectors. In my system I also have had a
pending sector for some years, but no bad sectors. It may be the Smart
software that is not reliable in this matter.
--
Svend Olaf
 >> Stay informed about: HDD suddenly failed, please help 

Svend Olaf Mikkelsen <svolaf.TakeThisOut@partitionsupport.com> wrote
> Rod Speed <rod.speed.aaa.TakeThisOut@gmail.com> wrote
>> Svend Olaf Mikkelsen <svolaf.TakeThisOut@partitionsupport.com> wrote
>>> Rod Speed <rod.speed.aaa.TakeThisOut@gmail.com> wrote

>>>>> 05 Reallocated Sector Count 5 100 100 1245192
>>>>> OK: Value is normal

>>>> Urk, thats the problem, thats an utterly
>>>> obscene reallocated sector count.

>>> The value 1245192 is hexadecimal 0x130008. The value are from the
>>> 8 vendor specific bytes, which typically are divided into an 8 bit
>>> field
>>> (the reported "low" value of 100), two 16 bit fields and a 24 bit
>>> field.

>> What makes you think that 05 is vendor specific ?

> The only documented values are the threshold value 5, and the
> attribute value 100. When threshold value is lower than the attribute
> value, the status for the attribute is considered OK, by definition.
> Most likely the number of reallocated sectors is not 1245192, a
> number which comes from the vendor specific field for attribute 05

No evidence that that is vendor specific.

> and is not documented in the Hitachi specifications for the disk.

But is documented in the SMART spec which dont say
anything about any separate fields as you are suggesting.

And it isnt clear why any manufacturer would need
separate fields for that particular value anyway.

>>> I would say that the number interprets two Smart 16 bit fields.
>>> One of 0x0013 = 19, and one of 0x0008 = 8.

>> Thats not how 05 is defined.

> No, we have to guess.

No reason to guess like that tho. And even if your guess
is correct, it still indicates that the drive has a problem,
when combined with those other fields below.

>>> This seems to be a logical data recovery problem only.

>> It isnt the only evidence of a drive problem, there are other
>> SMART fields that also indicate that the drive has a problem.

>>> 05 Reallocated Sector Count 5 100 100 1245192
>>> OK: Value is normal
>>> C4 Reallocation Event Count 0 100 100 9
>>> OK: Always passing
>>> C5 Current Pending Sector Count 0 100 100 1
>>> OK: Always passing

>> And its a DeathStar.

> A pending sector may be a bad sector.

Correct, but we already know that there are many more reallocated
sectors, even using your novel interpretation of the 05 value.

> In this case Drive Fitness Test however did not find any bad sectors.

Yes, pending isnt the same as bad.

> In my system I also have had a pending
> sector for some years, but no bad sectors.

Sure, but he has some number of reallocated sectors too
and clearly something got lost in some sector somewhere
to produce the error messages he is seeing.

> It may be the Smart software that is not reliable in this matter.

Or more likely that DFT isnt. Its been known for its problems.

We do know that something got lost somewhere to
produce the error messages he saw and since there
are problems clearly visible in the SMART data, its
very unlikely to be purely a software problem.
 >> Stay informed about: HDD suddenly failed, please help 

"Rod Speed" <rod.speed.aaa DeleteThis @gmail.com> wrote in message news:49ivs9ForbupU1@individual.net
> Svend Olaf Mikkelsen <svolaf DeleteThis @partitionsupport.com> wrote
> > Rod Speed <rod.speed.aaa DeleteThis @gmail.com> wrote
> > > Svend Olaf Mikkelsen <svolaf DeleteThis @partitionsupport.com> wrote
> > > > Rod Speed <rod.speed.aaa DeleteThis @gmail.com> wrote
>
> > > > > > 05 Reallocated Sector Count 5 100 100 1245192
> > > > > > OK: Value is normal
>
> > > > > Urk, thats the problem, thats an utterly obscene reallocated sector count.
>
> > > > The value 1245192 is hexadecimal 0x130008. The value are from the
> > > > 8 vendor specific bytes, which typically are divided into an 8 bit field
> > > > (the reported "low" value of 100), two 16 bit fields and a 24 bit field.
>
> > > What makes you think that 05 is vendor specific ?

> > The only documented values are the threshold value 5,

Don't think so.

> > and the attribute value 100.

Right.

> > When threshold value is lower than the attribute value, the status
> > for the attribute is considered OK, by definition. Most likely
> > the number of reallocated sectors is not 1245192, a number
> > which comes from the vendor specific fields for attribute 05

> No evidence that that is vendor specific.

Right, Hitachi/IBM obviously are lieing about that in their drive manual.

>
> > and is not documented in the Hitachi specifications for the disk.

> But is documented in the SMART spec which dont say
> anything about any separate fields

Yes, they do.
There are 12 bytes in the attribute table for each attribute entry of
which only the first four are defined. The other eight are vendor specific.

> as you are suggesting.

Right, Hitachi ar obviously lieing about what their disks support:

11.42.2.2 Figure 160. Individual Attribute Data Structure

offset 04h, 8 bytes, Vendor Specific

>
> And it isnt clear why any manufacturer would need
> separate fields for that particular value anyway.

They were left the choice how to use them.

>
> > > > I would say that the number interprets two Smart 16 bit fields.
> > > > One of 0x0013 = 19, and one of 0x0008 = 8.

Which 2 numbers still have no significance.
Both are too small for the available number of spares.

> > > Thats not how 05 is defined.

Right. that's what he said.

>
> > No, we have to guess.
>
> No reason to guess like that tho.

Right, damn what Hitachi have to say about it.
Obviously they don't have a clue about their own product.

> And even if your guess is correct, it still indicates that the drive
> has a problem, when combined with those other fields below.

Which too are vendor specific.

>
> > > > This seems to be a logical data recovery problem only.
>
> > > It isnt the only evidence of a drive problem, there are other
> > > SMART fields that also indicate that the drive has a problem.
>
> > > > 05 Reallocated Sector Count 5 100 100 1245192 OK: Value is normal
> > > > C4 Reallocation Event Count 0 100 100 9 OK: Always passing
> > > > C5 Current Pending Sector Count 0 100 100 1 OK: Always passing
>
> > > And its a DeathStar.

> > A pending sector may be a bad sector.

It is. It's not readable currently (supposedly, sin-
ce it wasn't readable the last time the drive tried).

>
> Correct, but we already know that there are many more realloca-
> ted sectors, even using your novel interpretation of the 05 value.

Nope. Same goes for the other attributes.

>
> > In this case Drive Fitness Test however did not find any bad sectors.

> Yes, pending isnt the same as bad.

It is on the current status: it can't be read currently (assuming it's
status didn't change -unnoticed- after it was discovered unreadable).

>
> > In my system I also have had a pending
> > sector for some years, but no bad sectors.
>
> Sure, but he has some number of reallocated sectors too

Which aren't necessarily bad sectors too.
What counts is whether they can be read or not.

> and clearly something got lost in some sector somewhere
> to produce the error messages he is seeing.

No data get's lost when sectors are reallocated.
If something got lost it was lost for other reasons/causes.

>
> > It may be the Smart software that is not reliable in this matter.
>
> Or more likely that DFT isnt.

> Its been known for its problems.

Nope.

> We do know that something got lost somewhere to
> produce the error messages he saw and since there
> are problems clearly visible in the SMART data, its
> very unlikely to be purely a software problem.

Nope, unless the eror log shows that all the events occurred very recently.
 >> Stay informed about: HDD suddenly failed, please help