Two drives at "Critical"

[TheSentinal]

1) I got this message on some drives of mine. I have the Pro version. What is the best method to "repair" or test-out these drives?

2) I started with "reinitialize surface area" while connected to a sata I hub but it would take around 40 hours so would this speed up a lot if connected directly to the motherboard via sata II or even faster connected to may PCI card which is sata III?

3) Can i do two or more drives at the same time to save the overall time taken?

[hdsentinel]

Thanks for your message and the questions.

1)

Please see: Support -> Frequently Asked Questions -> How to repair hard disk drive? How to eliminate displayed hard disk problems?
https://www.hdsentinel.com/faq_repair_h ... _drive.php

This describes how to detect, reveal, fix problems and improve the situation in general.

As mentioned numerous times in this forum (and on the above mentioned page), the bad sectors are already fixed, repaired. So there is no need to repair them again.
But it is possible (considering their count and the relatively low health) that further problems will be detected - and this is exactly why it is required to
perform intensive testing, to reveal possible further problems - or confirm if the drive is now stable and usable.
Then the error-counters can be cleared to show only possible new problems which may happen in the future.



2)

It depends on the speed of the disk drive(s) used and the controller itself.
In case of problems, the test may seems "freezing" for long time and the overall test may take much more time than originally expected.
On a such problematic drive, it is hard to say for sure how long - can be hours, but can be weeks, depending on the issues.

If you prefer a faster solution, you may try Disk menu -> Surface test -> Write test instead of the Reinitialize Disk Surface.
This is still effective and requires approx. 1/5 time compared to the more intensive and robust Reinitialize Disk Surface.

Anyway, for newer, faster drives yes, it may be good idea to connect to faster controllers in general.


3)

Yes of course, you can select Disk menu -> Surface test and by the "Multiple disk drives" button, you can select any number of disks to start the same type of test.
Alternatively, you can select Disk menu -> Surface test numerous times and start (same or even different) test on different drives at the same time.
Just consider that if both drives managed by the same controller, then the total bandwidth of the controller will be divided between the drives, so both may be slower.

But if you connect (for example) one of the drives to the current controller, one drive to the other controller and one on USB connection (to separate them) then there should no such issue and both can be tested with full speed.

If you use Report menu -> Send test report to developer option, I can check the actual situation, help and advise about the best ways to improve the situation.

[TheSentinal]

I Clicked on Surface test - Disk repair which took around 8 hours on my 4TB drive

There were zero bad sectors though there are 9 dark green blocks on the bottom row

When i check the health of the drive it still says 9% which i don't get even the results of the test

https://imgur.com/a/4PrBuD4

[hdsentinel]

Yes, it is completely normal and expected.

The Disk Repair test (as its description shows when you select the test "Reads the disk surface to verify if all sectors are readable and forces the disk drive to repair any problematic (pending, weak) sectors: restore the status to good or reallocate (replace from spare area) if required to prevent further use of the original sector.")

If it does not report any bad block (red) then it means that the hard disk has no new, undetected bad sectors - all bad sectors already reallocated (replaced by spare sectors).


> There were zero bad sectors though there are 9 dark green blocks on the bottom row

The Help describes: Darker green color means that the sectors are slower and harder to process (read or write depending on the test type) but they are still good. Having some (especially random) darker green blocks are completely acceptable unless they form a continuous, larger area (especially with yellow/red blocks near)


> When i check the health of the drive it still says 9%

Yes of course. The health is NOT automatically restored: the bad sectors still present.
You can manually clear the error counter to acknowledge the reported problems (which are already fixed) and by this, restore the health to 100%.

Please check:
https://www.hdsentinel.com/faq_repair_h ... _drive.php
for more information.
if you use Report -> Send test report to developer option, as then I can check the actual situation, actual error counter and advise.


> which i don't get even the results of the test

After the test completed, a window should display the results: should show any changes, degradations and the amount of possible problems (bad, damaged blocks) in addition to the good blocks.

[TheSentinal]

Do you mean this log?

So i gather from your last post that the disk surface area has been repaired/reallocated therefore ready/safe to use?

Is there any physical / mechanical issues like to do with the platters, or nuts and bolts?

[TheSentinal]

Hear are the repair screenshots and logs

[hdsentinel]

Not really...
I mean the Report menu -> Send test report to developer option where I can see the "raw" status provided by the disk drive. It is better for development purposes, especially when the drive health is far from perfect.

Yes: as the surface test shows 0 problems, we can be sure that all bad sectors already reallocated, repaired. They will no longer cause troubles and the hard disk is ready and (ideally) safe to use.
In real world, usually when such high number of problems reported (the health is low) we can expect new and new problems with further use. Maybe not immediately, maybe not tomorrow, but possibly in the future.

So personally I'd recommend to use ONLY for secondary storage (non-critical data) and only with constant monitoring and immediate backup upon any new issue, further health degradation.

As these bad sectors are no longer causing problems, you can clear their count, to restore the health now.

To clear the error counters, please open the S.M.A.R.T. page of the hard disk in Hard Disk Sentinel.

1) Locate the attribute 5 Reallocated Sectors Count
In the Offset column between the + and - symbol, click on the number and acknowledge the notification. Specify
-826
in the small box (the amount of bad sectors reported in the text description. You can also see the correct number on the graph on the bottom: you need to specify that number with negative sign)

2) Repeat the above for attribute 196 Reallocation Event Count
and specify
-513
there too (again, on the bottom graph you can see the current value which need to be specified with - sign).


Soon the problems should vanish from the text description and the health should increase back as well to 100%.
The health graph on the bottom will still show the low value today (as it shows the daily lowest health) but from tomorrow it should increase back.

Then, as the problems are already fixed, they will be no longer counted/reported by Hard Disk Sentinel, just possible future problem(s) will be reported, if there will be any.

If you prefer, you may use Report menu -> Send test report to developer option after adjusting these values (just to confirm if things are fine but if not, I can re-check).

Also any time later, if you'll see any new problem (health begins decreasing again) and you can send new report, I can examine, compare with the current.
To be honest, always good to examine how such non-perfect drives may work, especially after longer period.

[TheSentinal]

1) I did the offset things and the drive health immediately went up to 100%. Just so i know has the SMART data on the actual drive been edited like on the ROM, if it has any or is it just in the HDS software; like when i reformat/re-install would the data go back to 9% drive health?

2) Since the relocated blocks were all at the very end of the drive do you think its prudent to not use the say 100GB or so of the drive by not allocating it to the drive partition? As in the drive size is 3.7TB so ill only create a partition and use 3.6TB to never use the repaired sectors?

3) I am concerned about the other information that it contains in the report like my IP address which i fail to see why you need, which is why i cut and pasted the report info to just the specific drives information rather than sending an official report

[hdsentinel]

> 1) I did the offset things and the drive health immediately went up to 100%.
> Just so i know has the SMART data on the actual drive been edited like on the ROM,
> if it has any or is it just in the HDS software; like when i reformat/re-install would the data go back to 9% drive health?

The offset is stored in the current installation of Hard Disk Sentinel.
It is not stored in the disk drive itself: SMART data internally (in the disk drive) can't be edited/cleared.
If that would be possible, then everybody would sell used, almost dead hard disks as new, perfect drives....

The purpose of the offset is to allow us to clear problems related to the past, after performing testing, fixing/stabilizing issues - and (after understanding the situation and possible risks) restore the health to be notified about possible new problems (if there will be).


> 2) Since the relocated blocks were all at the very end of the drive do you think its prudent
> to not use the say 100GB or so of the drive by not allocating it to the drive partition?
> As in the drive size is 3.7TB so ill only create a partition and use 3.6TB to never use the repaired sectors?

Yes, this is a good idea (I suspect personally I'd do that).
Generally the reallocated area is no longer used, so the sectors would cause no problems.
But the darker green blocks confirm that sectors near the original bad sectors are slower. We can expect degradation (new bad sectors) there in the future, so yes, it is generally good idea to avoid storing important data there by making smaller partition. This way the affected sectors will be outside the partitioned area, so they'd be never read/written at all.


>3) I am concerned about the other information that it contains in the report like my IP address which i
> fail to see why you need, which is why i cut and pasted the report
> info to just the specific drives information rather than sending an official report[/quote]

Of course I'm not interested in your IP address or any similar information. I can confirm that the report does not contain any user data stored on your system.
The developer report contains technical information, for example "raw" response of the disk drive, information about disk controller / motherboard and the installed drivers. These help to check for example "weird" issues, compatiblity problems or similar (which is not the case now).

You may also save, review and send manually if you open Configuration -> Send test report page. There it is possible to create and save a developer-report.

[TheSentinal]

1) I was about to start using the drive again but thought it would be prudent to do another "repair" on the drive and now it looks like there are many more "bad" blocks; there are still the ones right at the bottom like last time but now there are a bunch of them scattered around the first 10% or so of the drive. Is this some kind of software glitch because the second test looks a lot worse than the first? Or what else could this be, because although the drive has been connected i have not written or read a single byte, i just want to have it connected for a week or so to test it out

2) I have sent a report via the software a day or two ago and just sent another one now after doing this test

3) Do you think i could/should use the drive but now, given the extra darker blocks partition out the first 10% of the drive and the last 10% of the drive? How safe would that be?

Thanks

[TheSentinal]

4) Forget to ask... "officially" there are zero "bad" or "damaged" blocks yet there are different colours for different blocks; does that mean they are "semi-bad" or "failing" etc?

[TheSentinal]

bump

[hdsentinel]

Thanks for the images and the details.

> Is this some kind of software glitch because the second test looks a lot worse than the first?

No, of course not. Sorry, do not really know why do you think that?
The test exactly does what it should: detect and reveal that the beginning of the disk drive performed slower than expected. This may remain hidden if we would (for example) simply format the disk drive.

The surface map shows the darker green blocks (where this slowness detected) and the transfer speed graph also confirms that the very first part was slower to process - and then the performance stabilized, jumped back and things worked as should.


> Or what else could this be, because although the drive has been connected i have not written or read a single
> byte, i just want to have it connected for a week or so to test it out

Not sure, but it is possible that the drive may performed an internal self test when the surface test started. This could cause a temporary slowness of some sectors.

In some rare cases, such slowness may be caused by other applications/software running which may intensively use system resources and/or causing high disk I/O on other disk(s) on the same controller.
In theory, those should have no effect, but in worst case, it is possible that an other disk performed intensive I/O (and used the bandwidth of the controller).

Some of such darker green blocks do not mean that the affected sectors would surely fail, these reflect only temporary slower performance.
Maybe you can try the test again after some time, just to be sure, just to verify if you'll get constant results - or not.

> 2) I have sent a report via the software a day or two ago and just sent another one now after doing this test

Thanks! I examined and fortunately see no changes, degradations or new problems.
Alone such slowness is not critical and does not mean any issue, it is acceptable - especially if we consider that the drive was not perfect, already reported problems.


> 3) Do you think i could/should use the drive but now, given the extra darker blocks partition
> out the first 10% of the drive and the last 10% of the drive? How safe would that be?

As the error-counters are stable, I'd say it may be used. But personally (considering the relatively low health) I'd not use for mission critical storage. Maybe for secondary data only.

[hdsentinel]

> 4) Forget to ask... "officially" there are zero "bad" or "damaged" blocks yet there are
> different colours for different blocks; does that mean they are "semi-bad" or "failing" etc?

No. They are not "semi-bad", not "will-surely-fail" or similar.
Those are also GOOD (as there were no problems processing, no retries, no error counter changes, etc), just they could be read a BIT slower than expected (if you move the mouse over the blocks, you can examine the actual performance).
It does not mean VERY slow performance, just slower-than-should.

But alone this does not indicate problems with the affected area.

[TheSentinal]

Hi, after using this drive in question, the 4TB Black it eventually went to 70%% health

I did a full reinitialize disk surface, with one over write, which passed successfully without find any bad blocks which i find weird because the health dropped to 70%?

I have sent the report to you to have a look at

[hdsentinel]

Thanks for the information about the surface test and the image you sent.
I did not receive report. Did you send with Report menu -> Send test report to developer option?
Then I can see the current status.

I see from the image you sent, bad sectors detected and displayed. These cause the 70% health of the hard disk drive.

In general, you do not need to worry about those bad sectors as these are already fixed by the hard disk: now the disk uses the spare area instead of these bad sectors, all read and write operations are using the spare area. So the original sector is reallocated.

This means that disk surface tests (even the surface test in Hard Disk Sentinel) will not access those sectors, but tests the remaining data area and the spare area. This is good, as this way you can be sure that the original (bad) area does not contain important data and can't risk data loss.

So ideally yes, even if the disk drive has bad sectors and reported in Hard Disk Sentinel, the surface test may find no problems at all.
This is not always true - as many times when bad sectors detected, new and new bad sectors appear from time to time. This is (I'm afraid) somehow "normal".

If the test show no errors (which means the hard disk is now stable, all possible bad sectors are already re-allocated and fixed this way), it is possible to acknowledge the reported problems and remove them from the text description, to improve the health back to even 100% as then Hard Disk Sentinel will only report future issues (if there will be).

This is described at http://www.hdsentinel.com/faq_repair_ha ... _drive.php

If you prefer, you can always use Report -> Send test report to developer option any time (now and/or for example if you see new problems, decreased health), so I can check the actual situation and difference and may advise step-by-step how to diagnose and improve the situation.

In most cases (especially if the disk is not perfect) the most important to detect and be advised about any kind of NEW problems, degradations - as this helps us to backup and take the required actions (test, diagnostics or prepare for replacement).

[TheSentinal]

I have re-sent the disk report please take a look at it and let me know what you find

Also can you tell me what i have to do to re-set the SMART counters to make the drive report its health as 100%

Thanks

[hdsentinel]

(I sent answer in e-mail too)

Yes, I see that after the offsets configured (so the errors cleared) previously, there are new problems found: the count of both
5 Reallocated Sectors Count
and
196 Reallocation Event Count

attributes increased again.

This is not surprising, we can say "normal". Generally when a disk has 800+ bad sectors, it is very common that new and new problems will be detected from time to time.

Now, if the test(s) showed no issues, the drive seems stabilized again, all new problems fixed again. So you may re-adjust the offsets again to keep stable for some time, but personally I'd expect more issues in the future.


Now, to clear the error counters, exactly as above, please open the S.M.A.R.T. page of the hard disk in Hard Disk Sentinel.

1) Locate the attribute 5 Reallocated Sectors Count
In the Offset column between the + and - symbol, click on the number and acknowledge the notification. Specify
-847
in the small box. Previously you configured -826, but as the text description says, there are 21 new bad sectors appear.

2) Repeat the above for attribute 196 Reallocation Event Count
and specify
-531
there too (where previously -513 configured, but there are 18 new reallocation events detected, reported)

The health will improve back again for now - but I'm almost sure that you'll see degradation in the near future. Maybe not tomorrow - or a week, but soon.