False reading from Hard Disk Sentinel?

How, what, where and why - when using the software.
efox31
Posts: 3
Joined: 2021.03.29. 00:42

False reading from Hard Disk Sentinel?

Post by efox31 »

I recently replaced a 1TB Seagate HDD with a 2TB hybrid drive (also Seagate), and it seemed to be a disaster. Hard Disk Sentinel immediately informed me that the drive has over a thousand bytes in bad sectors, has an overall health rating of 9% and is likely to die within two weeks. Yikes. However, every other piece of diagnostic software I've used says that it's fine. I've tried Seagate's own SeaTools application, Aomei Partition Assistant, DiskGenius, PerfectDisk and the inbuilt Windows chkdsk - they all give it a clean bill of health. Who do I believe here? I trust Hard Disk Sentinel but its findings are contradicted by literally everyone else. The only explanation I can think of is that HDS has a problem reading SSHDs (I'm still using the now-outdated version 5.50 which presumably has less support for newer drives).

I'd really appreciate any help anyone can offer - does my new drive need replacing or is HDS in error?
User avatar
hdsentinel
Site Admin
Posts: 3010
Joined: 2008.07.27. 17:00
Location: Hungary
Contact:

Re: False reading from Hard Disk Sentinel?

Post by hdsentinel »

Without seeing/knowing the details, hard the say anything for sure. If you use Report menu -> Send test report to developer option, I can examine the situation, verify the possible errors and advise about how to improve the situation.

Likely there is no "false reading" of course. I can confirm that Hard Disk Sentinel properly supports hybrid hard disks (of course) including your hard disk drive (even version 5.50, but yes, would be better to check with the latest version which has highest possible compatibility).

Please check:

Support -> Frequently Asked Questions -> Why other software displays different health? Which is correct?

https://www.hdsentinel.com/faq.php#difference

which generally describes the situation. I'm afraid it is completely normal and expected that other tools do not report any kind of problems, making their use (at least) questionable. But if other tools show "better" status, it does not mean they are correct - just they are simply not reporting possible errors, reallocations and so.

This is very common for manufacturer-specific tools, as they designed only to minimise warranty replacement and report only if the error-threshold is reached or not. They never report possible issues, degradations - and most other tools work similarly, happily showing "perfect" status for almost failing hard disk drives.

Chkdsk is also not really useful: even if sounds surprising, it does NOT check/report possible problems with the physical hard disk, but checks/fixes the logical drive (partition) only. It can report problems on perfect hard disks - and can never report issues on a failing drive.
The above described numerous times on this forum and on the website...

You did not mention if you used the PROPER ways to test, the hard disk tests offered by Hard Disk Sentnel, described at

Support -> Frequently Asked Questions -> Hard disk health is low or recently changed or I just installed a new (used) hard disk. How can I perform a deep analysis?

https://www.hdsentinel.com/faq.php#tests

as these designed exactly to verify, investigate the situation: reveal / confirm problems OR ensure that the hard disk is now stable or not. It is more sensitive than other solutions in reporting real disk problems, including retries, slower areas and so (which are usually ignored by other tools).

And if the hard disk tests show no errors, complete with perfect results (which means that all possible bad sectors are already reallocated), then it means that the hard disk is now stable, can be used: and in this case, yes, the health can be improved back (even to 100%) in Hard Disk Sentinel of course.

This is also described numerous times, for example at

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


Generally if all bad sectors are already reallocated, then yes, *ideally* the hard disk seems perfectly working, all sectors are readable/writeable, because the spare sectors used instead of the original bad sectors. This is why other test methods/tools can show "perfect" and even the disk tests in Hard Disk Sentinel should show that all sectors are readable/writeable. This is why some (small amount) of bad sectors are even acceptable, as described in the Help and and at https://www.hdsentinel.com/faq.php#health

But it does NOT mean that the hard disk was perfect. It had problems which are already fixed and the status is now stable. However, we can't automatically say "everything is perfect" as this is a bad assumption.

Hard Disk Sentinel designed exactly to reveal and report such issues, to increase attention (and it seems it helped to increase your attention, so did the job).
Ideally after all bad sectors reallocated, the hard disk can be used - but please keep in mind, that sometimes even one single bad sector (which means very high health, eg. 98%, much higher than the health of your hard disk) can cause a broken/degraded RAID array, as just discussed in a different forum topic:

https://www.hdsentinel.com/forum/viewto ... 356#p19356

or data corruption/data loss, degraded performance/functionality etc.

So we'd need to KNOW the real status of the hard disk drive, the real amount of problems (and possible degradations, new issues), perform the tests and then improve the health back to be notified about possible new problems/degradations only.
Usually hard disks with so low health suggests that new and new problems will likely happen with later use, so constant monitoring is essential in this case.

Maybe you can contact the deveoper of other tools and ask why they do NOT report these, why they ignore these completely ...

These are general information, applicable for most hard disk drives, especially for hard disks with low health where (even if things seem fine now) we can expect new and new issues.

And if you use Report menu -> Send test report to developer option, I can check and advise specifically in your case.
efox31
Posts: 3
Joined: 2021.03.29. 00:42

Re: False reading from Hard Disk Sentinel?

Post by efox31 »

Thank you very much for such a detailed and informative response. I see what you're saying about other diagnostic tools; on the other hand, even HDS itself gives the drive a clean bill of health when I run specific tests (screenshots attached, along with one showing the reported condition of the drive). It's very confusing! I've now submitted s report using the 'Send test report to developer' option so if you found time to give it a look I'd be very grateful.
Attachments
2021-03-29_155615.png
2021-03-29_155615.png (66.33 KiB) Viewed 9657 times
2021-03-29_032106.png
2021-03-29_032106.png (29.84 KiB) Viewed 9657 times
2021-03-29_025744.png
2021-03-29_025744.png (56.72 KiB) Viewed 9657 times
User avatar
hdsentinel
Site Admin
Posts: 3010
Joined: 2008.07.27. 17:00
Location: Hungary
Contact:

Re: False reading from Hard Disk Sentinel?

Post by hdsentinel »

Thanks for your answer, the images and the report. I examined them and see the situation.

No, it should be not confusing: in a VERY lucky situation (like this one) things should work exactly as how you can see: even if the hard disk has bad sectors, they are no longer used. When the hard disk detected these, it automatically marked them as "bad" and then all future reads/writes targeting these bad sectors are automatically redirected to the spare area, which originally reserved.

This is described at
https://www.hdsentinel.com/faq.php#health
and
https://www.hdsentinel.com/help/en/57_desc.html


This is exactly the purpose of the spare area: to attempt to "hide" problems and ideally the user should not notice any issue and the disk drive should work perfectly.

But in real life it is very-very rare that a hard disk with over a thousand (1648 in this case) bad sectors works with no errors, even no delays and the tests are completed with no problems reported. This is why I wrote that it is a very lucky situation.

Now, as with the tests you investigated the situation and verified that the hard disk is really correctly working, you can acknowledge these bad sectors: as they are fixed, they should be no longer reported at all. So you can restore the health to 100% and be notified about possible only new problems (if there will be) this way:

1) please select the S.M.A.R.T. page on the hard disk
2) locate the attribute 5 Reallocated Sectors Count and in the Offset column (between the + and - symbols) click on the 0
3) after clicking OK in the confirmation window, please specify

-1648

(the amount of bad sectors reported in the text description with negative sign)

Soon the errors will be no longer reported in the text description and the Health % value will increase back 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 only possible new problems (new bad sectors and/or other kind of issues) will be reported in both the text description and the Health % value, but as long as the hard disk status remains stable, the Health value will be displayed as 100%.
User avatar
hdsentinel
Site Admin
Posts: 3010
Joined: 2008.07.27. 17:00
Location: Hungary
Contact:

Re: False reading from Hard Disk Sentinel?

Post by hdsentinel »

Some notes: I sent details about the report you sent.
Just noticed that on the image, 1640 bad sectors displayed - while in the developer-report, I saw 1648. This is why I wrote that number.
Interesting is that it seems the amount of bad sectors still increased even after you saved the screenshot.

According the experiences, yes, this is expected: when the amount of bad sectors is high, more and more can appear with time / use (or during tests which can reveal further).

Ideally yes, such reallocations (replacing the original sectors with a spare sector) should be completelly transparent and should happen automatically - but it is very rare. Usually (especially if the health was low) such reallocations can cause data corruption/data loss or "just" system instability - this is why it is important to know such events.
efox31
Posts: 3
Joined: 2021.03.29. 00:42

Re: False reading from Hard Disk Sentinel?

Post by efox31 »

I've followed the steps you advised and things are looking good (see attached image).

I really can't thank you enough for all the time you've spent looking into this - it's greatly appreciated.

If any further problems occur I'll let you know - in the meantime, I'll keep my fingers crossed!
Attachments
2021-03-31_144503.png
2021-03-31_144503.png (61.09 KiB) Viewed 9592 times
Yalmaz Khan
Posts: 1
Joined: 2022.01.21. 17:44

Re: False reading from Hard Disk Sentinel?

Post by Yalmaz Khan »

Hi there, i hope i am not late. I have got the same issue, surprisingly, EXACTLY the same. My SSD has 9% health, and i bought it yesterday, it is a 256 GB Samsung SSD that, according to the seller, was checked and its health was 100%. Apart from hard disk sentinel, that shows 9% health, all other softwares, and by that, i mean, every software in the top 10 list, shows GOOD HEALTH. Should i consider Hard disk Sentinel analysis over all of those? Or should i just, let it go and keep using it.
Note: I did all the tests that the thread writer did, and got the same results, Hard disk sentinel shows clean surface test, with successful error free individual reads and writes tests. And yet, the Overview tells a heart panicking story. I attached some of the test results. Do not have time yet to redo all tests to make complete report, its my university exams these days. I just want to ask, is the Thread writer's SSD still working?, has it given any issues yet? If so, please inform me.
Attachments
Capture3.PNG
Capture3.PNG (45.24 KiB) Viewed 6659 times
2
2
Capture2.PNG (84.2 KiB) Viewed 6659 times
1
1
Capture.PNG (253.03 KiB) Viewed 6659 times
User avatar
hdsentinel
Site Admin
Posts: 3010
Joined: 2008.07.27. 17:00
Location: Hungary
Contact:

Re: False reading from Hard Disk Sentinel?

Post by hdsentinel »

Hi,

Can you please use Report menu -> Send test report to developer option?
I'd be more than happy to check the actual status of the drive and advise better by seeing the details.

You added some images - but they do not show the details, for example you did not include a screenshot about the S.M.A.R.T. details displayed in Hard Disk Sentinel.
I always wonder why people did not include that when asking - and why they include screenshots from other tools for no reason ;)

But from the images, as you can see the Overview page shows 2972 errors happened previously and one of the images (from a different tool) confirms these: as you can see the Runtime bad block count = 000B9C (which is 2972 in decimal). So this is why Hard Disk Sentinel displays the errors, as it does exactly what it should: increase attention to problems/degraded status which may be completely ignored by other tools.

If other tools completely ignore this, then yes, I'm afraid this is a serious problem with them - making their use at least questionable.... Please ask the developer(s) of the other tool(s) too, would be nice to hear what they say.

Yes, _ideally_ when all such errors fixed, then the SSD seems perfect: all sectors can be read/written. But we can't automatically say it is "perfect" as it can indicate more problems which can lead to data corruption / data loss. You may trust and use it - and even clear the error counters in Hard Disk Sentinel to acknowledge them and report only further problems/degradations (and increase the Health too, back to 100% if you want to see "perfect"). But this does not make a failing drive perfect of course, so then you may encounter data corruption / data loss anyway, seeing "perfect" status...

Please check: https://www.hdsentinel.com/ssd_case_bad_sectors.php
Post Reply