maxtor low level format

[hodw]

Hi

My Maxtor drive failed with bad sectors.
I also get SMART error condition relocatable sector count on this
drive.
The company support told me to do a low level format with maxblast and
then I can use the drive again.

Isnt a SMART error and indication that the drive is at the end of life?
Is it worth to do the low level format?

Hod

 

[Arno Wagner]

Hi

My Maxtor drive failed with bad sectors.
I also get SMART error condition relocatable sector count on this
drive.
The company support told me to do a low level format with maxblast and
then I can use the drive again.

Hahaha! Until it develops new bad sectors....

Isnt a SMART error and indication that the drive is at the end of life?
Is it worth to do the low level format?

Generally no. Unless it is a small number of bad sectors, say < 200
and the number does not increase or only increases very slowly.
If you have the SMART attribute on relocated sectors fail,
then you have a very large number od allready reallocated ones,
and the defects you see may be due to the spare tables being
exhausted. I would say this drive is a paperweight.

Arno

 

[Rod Speed]

My Maxtor drive failed with bad sectors.
I also get SMART error condition relocatable sector count on this drive.
The company support told me to do a low level format with maxblast and
then I can use the drive again.

Isnt a SMART error and indication that the drive is at the end of life?

Not necessarily, particularly with maxtors, an occasional
relocateable sector isnt that uncommon, they appear to
ship with an optimistic bad sector list and rely on the
drive to add some dubious ones to the bad sector list.

Is it worth to do the low level format?

Yes, as long as its only a small number of reallocated sectors.

 

[hodw]

Hi
I do not understand.
The bad cluster problem started before the SMART error.
The SMART error has a treshhold value which was exceeded.
My understaning is that it found BAD CLUSTERS a lot more them it can
handle.

Even if it is a small number is it worth doing a low level format?
Won't I get new bad sectors afterwards?
The low level format does not fix them just maps them as bad?

Hod

 

[Rod Speed]

I do not understand.
The bad cluster problem started before the SMART error.
The SMART error has a treshhold value which was exceeded.

OK, you didnt make that clear initially.

My understaning is that it found BAD
CLUSTERS a lot more them it can handle.

SMART knows nothing about clusters, it only knows about sectors.

If you have a SMART threshold being exceeded, you have
seen a lot more reallocated sectors than is acceptible.

Even if it is a small number

Thats unlikely with the SMART threshold being exceeded.

is it worth doing a low level format?
Won't I get new bad sectors afterwards?

Depends on what has caused the bad sectors. If it was because
the drive got severely overheated and thats what produced the bad
sectors, and its not overheated anymore, the LLF may well be useful.

Same with a bad power supply once the bad power supply has been replaced.

The low level format does not fix them just maps them as bad?

It checks if they are bad and maps them as bad if they really are bad NOW.

 

[Arno Wagner]

Hi
I do not understand.
The bad cluster problem started before the SMART error.

A sucessfully reallocated sector can heve two effects on the
visible data: If it was recoverd by retries or ECC during reading
or found while writing, the reallocation will be transparent and
no bad clusters will be visible. However if it was found during
reading and the data was not recoverd from it, there will be a
bad cluster as result, until this specific disk area is written
to. If you have a lot of reallocated sectors, you oftehn will have
some of both types.

If the reallocation was unsuccessful, e.g. because no spares are left,
then you also will get bad clusters.

The SMART error has a treshhold value which was exceeded.
My understaning is that it found BAD CLUSTERS a lot more them it can
handle.
Agreed.

Even if it is a small number is it worth doing a low level format?

Well, it wis worth overwriting the area were the secors are.
ll-format is not done today.

Won't I get new bad sectors afterwards?

Depends. Some faults are transient, e.g. mechanical shock or abad
and now replaced PSU. Some are not and you will indeed get more
bad secotrs.

The low level format does not fix them just maps them as bad?

Yes.

Arno

 

[hodw]

Hi
Ok so I do the llf and find no bad clusters.
Is this drive OK for use?
If I will get Bad clusters again it is useless to me.
Ive already spend to much time on it.

Hod

 

[Arno Wagner]

Hi
Ok so I do the llf and find no bad clusters.
Is this drive OK for use?

No. If you llf and hace a non-failed SMART status and
have a relatively low number of reallocated sectors
in the SMART attributes (I would say < 200), then you are
likely ok. Even if the SMART status goes to non-failed,
but you have >200 reallocated sectors, I would not trust
it anymore. Maxtor is awfully generous with the SMART
thresholds. I had a 200GB Maxtor disk with already > 1400
reallocated secotrs where reads took forever, but the
SMART status was ''good''.

If I will get Bad clusters again it is useless to me.

Agreed.

Just run a long SMART self-test or the long test with
power-max. That is what amounts to a llf today. Then
check the SMART status and the reallocated secot number,
e.g. with Sandra or the smartmontools.

If they are bad or give real reason for concern, discard
the drive.

Ive already spend to much time on it.

I found that a good rule of thumw was to look at how much
money I could reasonably earn with not too unpleasant work
in the time I spent so far. If that exceeds the residual
value of the drive (estimate e.g. at 50% current retail,
if unknown), I stop working on it. Then I might have
some free fun in destroying the bad beast physically. ;-)

Arno

 

[Al Dykes]

Hi
Ok so I do the llf and find no bad clusters.
Is this drive OK for use?
If I will get Bad clusters again it is useless to me.
Ive already spend to much time on it.

Hod

I'd download the Maxtor drive test/certification software and run it.

 

[Folkert Rienstra]

You didn't understand a word of it, didn't you.

 

[Folkert Rienstra]

A sucessfully reallocated sector can heve

two effects

on the visible data: If it was recoverd by retries or ECC during
reading or found while writing, the reallocation will be transparent
and no bad clusters will be visible. However if it was found during
reading and the data was not recoverd from it, there will be a
bad cluster as result, until this specific disk area is written to.
If you have a lot of reallocated sectors, you oftehn will have
some of both types.

So what's the second one, babblemouth?

 

[Alexi]

Here's a spin. I have one OneTouch I. No smart, no diagnostics. I saved
what I could, reformatted and back in business. When my II failed, I took
it back to Costco and now have 2 Onetouch III's. But what to do with an
aging I?

What about opening it up and replacing the hard disk. I hav not opened the
two phillips, but Let's assume I can. Swap it, junk it or ebay it for $30
or so probably

 

[Jon Forrest]

A sucessfully reallocated sector can heve two effects on the
visible data: If it was recoverd by retries or ECC during reading
or found while writing, the reallocation will be transparent and
no bad clusters will be visible. However if it was found during
reading and the data was not recoverd from it, there will be a
bad cluster as result, until this specific disk area is written
to. If you have a lot of reallocated sectors, you oftehn will have
some of both types.

This is very interesting information. I'd like to learn
more about bad block handling in xATA disks. Could you
post some links to reference material that covers this
in technical detail.

Thanks,
Jon Forrest

P.S. I'm not doubting your claim. In fact, it explains
some things I've been wondering about for a long time.

 

[Arno Wagner]

Arno Wagner wrote:

This is very interesting information. I'd like to learn
more about bad block handling in xATA disks. Could you
post some links to reference material that covers this
in technical detail.

Well, I don't really have one reference for that. Vendor whitepapers
may help. A lot of my information is from various sources and
accumulated over the years, also from this newsgroup. I have
personally observed the reallocation on overwrite several times. In
fact I have done it for some drives manually, by identifying the file
the defective sector was in and then overwriting that.

Thanks,
Jon Forrest

P.S. I'm not doubting your claim. In fact, it explains
some things I've been wondering about for a long time.

Always glad to help in better understanding of technology.

Arno

 

[Folkert Rienstra]

Well, *I* don't really have one reference for that.

That's nicer words for having none.

Vendor whitepapers may help.

And if you knew of any you would tell, right babblemouth?

A lot of *my* information is from various sources and accumulated over the years,

also from this newsgroup.

*Only* this newsgroup.

*I* have personally observed the reallocation on overwrite several times.

You did, huh.

In fact *I* have done it for some drives manually, by identifying
the file the defective sector was in and then overwriting that.

An idea that you stole from me, Arnie.

Always glad to help in better understanding of technology.

Yes babblemouth, by hijacking other peoples ideas as your own.

 

[Jon Forrest]

Well, I don't really have one reference for that. Vendor whitepapers
may help. A lot of my information is from various sources and
accumulated over the years, also from this newsgroup. I have
personally observed the reallocation on overwrite several times. In
fact I have done it for some drives manually, by identifying the file
the defective sector was in and then overwriting that.

This explains why vendors get away with calling their complete
disk writing utility "low level" formatting. You can't do true
low level formatting on xATA drives but by writing to every sector
you'll detect and correct all the bad sectors, which almost
has the same affect as true low level formatting, especially
if every write creates a sector containing all 0s.

Jon Forrest

 

[Arno Wagner]

Arno Wagner wrote:

This explains why vendors get away with calling their complete
disk writing utility "low level" formatting. You can't do true
low level formatting on xATA drives but by writing to every sector
you'll detect and correct all the bad sectors, which almost
has the same affect as true low level formatting, especially
if every write creates a sector containing all 0s.

Not quite. You cannot detect all bad sectors on writing. Only
those that have some problem that makes them difficult to
find reliably. The actual defect identification is usually
done by a full surface scan (long SMART self-test) before.
The drive remembers which sectors had reading unrecoverable
troubles and need to be realloceted on the next write.

So you need one read and one write for each defective
sector for it to be reallocated reliably.

Arno

 

[Folkert Rienstra]

This is very interesting information. I'd like to learn
more about bad block handling in xATA disks. Could you
post some links to reference material that covers this
in technical detail.

http://www.hgst.com/tech/techlib.ns...27F35A86256F4F006B8AD4/$file/7K500_spv1.3.pdf

10.11.1 Auto Reassign function
The sectors which show some errors may be reallocated automatically when specific conditions are met. The spare
tracks for reallocation are located at regular intervals from Cylinder 0. The conditions for auto- reallocation are
described below.
Nonrecovered write errors
• When a write operation cannot be completed after the Error Recovery Procedure (ERP) is fully carried
out, the sector(s) are reallocated to the spare location. An error is reported to the host system only when
the write cache is disabled and the auto reallocation has failed. If the Write Cache function is
ENABLED when the number of available spare sectors reaches 0 sector, both Auto Reassign function
and Write Cache function are automatically disabled.
Nonrecovered read errors
• When a read operation has failed after defined *ERP* is fully carried out, a hard error is reported to the
host system. This location is registered internally as a candidate for reallocation. When a registered
location is specified as a target of a write operation, a sequence of media verification is performed auto-matically.
When the result of this verification meets the criteria, this sector is reallocated.
Recovered read errors
• When a read operation for a sector has failed once and then has recovered at the specific ERP step, this
sector of data is automatically reallocated. A media verification sequence may be run prior to the reallo-cation
according to the predefined conditions.


Re *ERP* (error recovery procedure):

http://www.hitachigst.com/tech/tech...8006A31E587256A9C0064975D/$file/dchs_spiw.pdf

Appendix B. Recovery Procedures
also see ARRE

And of further interest:
4.18 Bring-Up Sequence
4.20 Reserved Area

DCHS is a SCSI drive so you should read between the lines but most
of it applies to any drive if you leave out the SCSI specific detail.