weird problem with my external usb hdd

[gringoz]

hi,
i had experienced a input/ouput error when writing to the NTFS disk which was attached to my Raspberry pi 2. its a 1TB disk with 80gb free. The input/output error occured so decided to run a chkdsk on windows which did not detect anything then i tried HDS and got a 95% so i run auto short and extended test but got failed right away. I read the FAQ and didn't worry about it. next i looked at the summary
1,Raw Read Error Rate,51,200,200,OK,000000000000,0,Activé
3,Spin Up Time,21,177,161,OK,000000001014,0,Activé
4,Start/Stop Count,0,100,100,OK (toujours réussi),000000000259,0,Activé
5,Reallocated Sectors Count,140,200,200,OK,000000000000,0,Activé
7,Seek Error Rate,0,100,253,OK (toujours réussi),000000000000,0,Activé
9,Power On Time Count,0,76,76,OK (toujours réussi),0000000046A0,0,Activé
10,Spin Retry Count,0,100,100,OK (toujours réussi),000000000000,0,Activé
11,Drive Calibration Retry Count,0,100,100,OK (toujours réussi),000000000000,0,Activé
12,Drive Power Cycle Count,0,100,100,OK (toujours réussi),000000000242,0,Activé
192,Power off Retract Cycle Count,0,200,200,OK (toujours réussi),00000000020C,0,Activé
193,Load/Unload Cycle Count,0,165,165,OK (toujours réussi),00000001A0A2,0,Activé
194,Disk Temperature,0,112,100,OK (toujours réussi),000000000026,0,Activé
196,Reallocation Event Count,0,200,200,OK (toujours réussi),000000000000,0,Activé
197,Current Pending Sector Count,0,200,200,OK (toujours réussi),000000000008,0,Activé
198,Off-Line Uncorrectable Sector Count,0,100,253,OK (toujours réussi),000000000000,0,Activé
199,Ultra ATA CRC Error Count,0,200,200,OK (toujours réussi),000000000000,0,Activé
200,Write Error Rate,0,100,253,OK (toujours réussi),000000000000,0,Activé

looks like i have 8 bad sectors, decided to run a surface test, everything was fine until i hit bloc 6627 and i got a input/ouput error, decided then to start from 6627 same result. Next i did a reverse test, it went up to 3307 now its doing the second pass.
any idea what to do next?is the I/O error related to the hw or just a bad sector?

[hdsentinel]

Sorry to say, but it is absolutely not a weird problem.
This is a very common situation - and very common that users got confused and uses wrong method (chkdsk) to diagnose.

> I read the FAQ and didn't worry about it.

Sorry, it seems you missed the section which clearly describes the situation.

If you would inspect the text description in Hard Disk Sentinel (without the need of checking the raw values), you may see that "weak sectors" reported about the hard disk.
Exactly to help, there is a complete, detailed page which describes this situation: Support -> Frequently asked questions -> What is a weak sector? How to repair weak sectors?
which available directly at http://www.hdsentinel.com/hard_disk_cas ... ectors.php

There is no need to check the status manually - just inspect the text description and follow the information about how to repair the drive.
As described on that page, chkdsk does not help in this situation, but makes things even worse: users may incorrectly feel there are bad sectors on the drive which is NOT true.

And as described, it is completely normal that short / extended hardware self tests fail in this situation (usually quickly).


> looks like i have 8 bad sectors,

Absolutely NOT.


> decided to run a surface test, everything was fine until i hit bloc 6627 and i
> got a input/ouput error, decided then to start from 6627 same result.
> Next i did a reverse test, it went up to 3307 now its doing the second pass.

Yes, the Disk menu -> Surface test -> Read test shows the location of the problems, the weak sectors which are currently used and can cause instability, data corruption or even data loss.

> any idea what to do next?

As described, the best way is to use Disk menu -> Surface test -> Reinitailise disk surface test (after a complete backup) to repair the problems: completely eliminate the weak sectors and improve the usability. The health of the hard disk will also increase and (how surprising) then chkdsk will find no problems and the short / extended self test will also complete without problems.

I wonder why you missed this page: http://www.hdsentinel.com/hard_disk_cas ... ectors.php
which exactly designed to help, assist and prevent confusion.
Plus it gives information about how these weak sectors originally created and how to avoid them in general, as they may appear even on a perfect, new hard disk - and usually they are completely independent from the drive, they do not mean problem with the drive itself.

[gringoz]

Sorry to say, but it is absolutely not a weird problem.
This is a very common situation - and very common that users got confused and uses wrong method (chkdsk) to diagnose.

> I read the FAQ and didn't worry about it.

Sorry, it seems you missed the section which clearly describes the situation.

If you would inspect the text description in Hard Disk Sentinel (without the need of checking the raw values), you may see that "weak sectors" reported about the hard disk.
Exactly to help, there is a complete, detailed page which describes this situation: Support -> Frequently asked questions -> What is a weak sector? How to repair weak sectors?
which available directly at http://www.hdsentinel.com/hard_disk_cas ... ectors.php

There is no need to check the status manually - just inspect the text description and follow the information about how to repair the drive.
As described on that page, chkdsk does not help in this situation, but makes things even worse: users may incorrectly feel there are bad sectors on the drive which is NOT true.

And as described, it is completely normal that short / extended hardware self tests fail in this situation (usually quickly).


> looks like i have 8 bad sectors,

Absolutely NOT.


> decided to run a surface test, everything was fine until i hit bloc 6627 and i
> got a input/ouput error, decided then to start from 6627 same result.
> Next i did a reverse test, it went up to 3307 now its doing the second pass.

Yes, the Disk menu -> Surface test -> Read test shows the location of the problems, the weak sectors which are currently used and can cause instability, data corruption or even data loss.

> any idea what to do next?

As described, the best way is to use Disk menu -> Surface test -> Reinitailise disk surface test (after a complete backup) to repair the problems: completely eliminate the weak sectors and improve the usability. The health of the hard disk will also increase and (how surprising) then chkdsk will find no problems and the short / extended self test will also complete without problems.

I wonder why you missed this page: http://www.hdsentinel.com/hard_disk_cas ... ectors.php
which exactly designed to help, assist and prevent confusion.
Plus it gives information about how these weak sectors originally created and how to avoid them in general, as they may appear even on a perfect, new hard disk - and usually they are completely independent from the drive, they do not mean problem with the drive itself.

thanks for the support but just to make it clear, i did read everything on the faqs in fact i have the 3 faqs regarding weak sectors and bad sectors opened right now. the problem is the I/O error i get when hitting the bloc 6627 it keeps on going to i just stop it. If i start from the 6630 it shows everything ok so right now i'm backing up my 1tb hdd (takes a long time even if its a USB 3) then i'll do the disk surface reintialisation

[hdsentinel]

> thanks for the support but just to make it clear, i did read everything on the faqs in fact i have the 3 faqs
> regarding weak sectors and bad sectors opened right now.

From the previous self-monitoring status, I see no bad sectors at all.
The bad sector counters are both zero (5 Reallocated Sectors Count, 198 Off-Line Uncorrectable Sector Count).

Do you see any change in status in Hard Disk Sentinel? With other words, does Hard Disk Sentinel displays NOW that you have bad sectors?
If not - then the hard disk surely has no bad sectors (even if chkdsk shows that - causing a major confusion).


> the problem is the I/O error i get when hitting the bloc 6627 it keeps on going to i just stop it.

Yes of course. When the read test reaches the WEAK sector, the drive may slowly responding during processing these sectors.
These may cause general instability in operation of the drive.

This is the problem: the weak sectors are still in use and they cause data corruption / data loss / general instability in operation.
These can and should be fixed (not by chkdsk or other wrong methods, but with the Reinitialise Disk Surface test in Hard Disk Sentinel).


> If i start from the 6630 it shows everything ok

Yes of course. There are no problems in other sectors. On your hard disk, the weak sectors are in the range between blocks 6627-6630, so these would need to be reinitialised. Working on a function to perform this without the need of complete backup and complete clearing.

If you prefer, you can use Report menu -> Send test report to developer option (now and/or after the test) as then it is possible to check the complete status of the drive, verify any change in the reported status after the test.

[gringoz]

After reinitializing the surface of the disk health is now 100% without any weak sectors, i'm running another test of the surface and so far so good.

thanks for the support