8 weak sectors on HDD
8 weak sectors on HDD
Hello, today when starting my PC I saw that windows needs to do chkdsk on one of my HDD's. I didn't skip it and let it do it. I saw some lines mentioning deleting some indexes related to few files. Once it was finished and my system loaded up, I checked HDD Sentinel and it's showing 8 weak sectors on that HDD and the health dropped to 95%. Should I be worried and do something like additional scans with HDD Sentinel?
Re: 8 weak sectors on HDD
Few hours later now it reports 2 bad sectors as well
-
hdsentinel
- Site Admin
- Posts: 3115
- Joined: 2008.07.27. 17:00
- Location: Hungary
- Contact:
Re: 8 weak sectors on HDD
Yes, usually problems start exactly this way....
Usually the first step is to perform backup - at least the important data, but ideally it is best to copy all files to a different drive, to be safe.
DO NOT attempt to clone or similar as it can make things worse.
Then it is good idea to check cables, connections, power cable, power supply etc. Make sure to have stable operating environment: avoid overclock or similar which may also affect the disk drive.
Then perform disk testing, exactly to reveal possible further issues, attempt to stabilize them.
I recommend to check the Support -> Frequently Asked Questions page, as it contains different sections about what to do in such similar situation:
How to repair hard disk drive? How to eliminate displayed hard disk problems?
https://www.hdsentinel.com/faq_repair_hard_disk_drive.php
What is a weak sector? How to repair weak sectors?
https://www.hdsentinel.com/hard_disk_case_weak_sectors.php
Ideally the Disk menu -> Surface test -> Disk Repair test can help to stabilize and improve the situation.
Usually the first step is to perform backup - at least the important data, but ideally it is best to copy all files to a different drive, to be safe.
DO NOT attempt to clone or similar as it can make things worse.
Then it is good idea to check cables, connections, power cable, power supply etc. Make sure to have stable operating environment: avoid overclock or similar which may also affect the disk drive.
Then perform disk testing, exactly to reveal possible further issues, attempt to stabilize them.
I recommend to check the Support -> Frequently Asked Questions page, as it contains different sections about what to do in such similar situation:
How to repair hard disk drive? How to eliminate displayed hard disk problems?
https://www.hdsentinel.com/faq_repair_hard_disk_drive.php
What is a weak sector? How to repair weak sectors?
https://www.hdsentinel.com/hard_disk_case_weak_sectors.php
Ideally the Disk menu -> Surface test -> Disk Repair test can help to stabilize and improve the situation.
Re: 8 weak sectors on HDD
Hello, I already had a backup on the most important files on other drive. Just wanted to make sure if this issue means that HDD is going to break soon or if it is nothing serious yet. However I already replaced it today. I read up today about how this software calculates it's health and remaining lifetime values and yeah it's not magic as it can't predict when something is going to happen and it will update it's values only after something did happen. Yesterday it was still 100% health and like 947 days. Today once weak sectors had appeared it dropped to 95% health and 773 days. Few hours later now there are 2 bad sectors with 93% health and 741 days. Few minutes later 5 bad sectors with 90% health and like 654 days. I don't know if it could actually stop there and this HDD could actually still run for those ~ 600 days, but I just didn't want to risk it and decided to replace it right away. I managed to copy everything to a new drive and it all went smoothly thanks to your software since I was aware of this happening early on. I remember few years ago when I bought this HDD, it's predecessor was going to break in the same day I found out about it by pure accident when copying some video files between disk's and 1 video file failed to copy because of some error. I quickly checked with HDTune and indeed there were few yellow values and few damaged blocks on Error Scan. I still managed to copy everything to a new drive back then with the exception of that 1 video file, but had I waited any longer there could be more lost files as just when I was done copying everything and checked with the HDTune one last time before plugging it off, there were many more damaged blocks and yellow values. I didn't want to risk this again with this one. It was my oldest disk in my PC with 23k hours and I remember all my HDD's breaking near this number, so I am not really surprised. In fact I was more surprised of HD Sentinel still showing those 900 days lifetime for this HDD. I mean it did acknowledge the fact that it was pretty old, because for all my newest HDD's and SSD'd it shows more than 1000 days and for this it did show an actual number, but as I learned today that lifetime value can also go down by 200 or 300 in a 1 day just in few hours based on what's happening to it. Still thank you for your reply and for your amazing software that really helps with detecting those kind of issues early on and saves us from losing data.
-
hdsentinel
- Site Admin
- Posts: 3115
- Joined: 2008.07.27. 17:00
- Location: Hungary
- Contact:
Re: 8 weak sectors on HDD
> I read up today about how this software calculates it's health and remaining lifetime values and yeah it's not magic
Yes, the Help explains how version 1.00 (released 15+ years ago) calculated the Health % value, based on the S.M.A.R.T. attributes.
Yes, that was not magic, and things changed, become much more sophisticated since that, just it is not really possible to describe with all details.
> as it can't predict when something is going to happen and it will update it's values only after something did happen.
Yes, of course, this is how things work: until a sensor and/or error counter do not detect a problem - we can't show that there is an issue with the disk drive
And this is why generally it is recommended to perform some tests (even on a new drive and especially if there is any minor change, degradation - which can indicate more problems) to REVEAL possible problems - or confirm that the drive is stable, really error free:
https://www.hdsentinel.com/faq.php#tests
Until the drive reads/writes a particular sector, the problem may remain unnoticed for long time (even for years) as explained at
https://www.hdsentinel.com/hard_disk_case_bad_sectors.php
This is why the Health % graph on the bottom designed to show any (even minor) change/drop so dramatically, to increase attention to the change.
Ideally there are no more problems - so then the drive can be stable - but in many cases, this is just the first step and (exactly as you can see) more and more problems can appear sooner or later.
> Yesterday it was still 100% health and like 947 days. Today once weak sectors had appeared it dropped to 95% health and 773 days.
> Few hours later now there are 2 bad sectors with 93% health and 741 days. Few minutes later 5 bad sectors with 90% health and like 654 days.
Yes, this is completely "normal" and expected: as more problems detected, the Health % reflects those and the estimated remaining lifetime also decreases. The estimated remaining lifetime value is not a strict countdown, just can indicate how long we can use the drive (if its status is stable), when we should plan replacement because of its age / usage and/or the actual problems with it.
> I don't know if it could actually stop there and this HDD could actually still run for those ~ 600 days,
> but I just didn't want to risk it and decided to replace it right away.
I can completely understand. For a mission critical environment, I'd probably replace too (as you do). Many users replace a hard disk drive even on 99% Health because the value of the data is much more than the disk drive itself (plus possible downtime / maintenance).
For non-critical data, the drive may be still used with constant monitoring (and after testing, exactly to reveal possible further issues).
> I managed to copy everything to a new drive and it all went smoothly thanks to your software since I was aware of this happening early on.
Thanks, good to hear that the data could be saved. Usually yes, in this case, this is possible, if we do not wait more.
Yes, generally this should be the main purpose: as you can see, the Health % may remain stable for long time - until it suddently starts decreasing. This is the point which we need to take action: backup first - and then decide what to do
- start testing, repairing
- or do not risk but replace the disk drive
So it is important to be notified about this point, when the Health % of the hard disk drive starts decreasing.
> In fact I was more surprised of HD Sentinel still showing those 900 days lifetime for this HDD.
May sound weird, but usually some bad sectors can be acceptable, as described at
https://www.hdsentinel.com/faq.php#health
so the first step is to increase attention about the change, degradation and do not show "replace immediately" until really required.
More problems (more drops in the Health) would probably trigger even lower Health % and even (much) lower estimated remaining lifetime to suggest replacement.
> because for all my newest HDD's and SSD'd it shows more than 1000 days and for this it did show an actual number,
Yes of course. If the estimated remaining lifetime would be (much) more, "more than 1000 days" displayed.
I do not feel real / comfortable to say "30 years" or so (even if displayed by other tools), this is why the "1000 days" displayed, as it is more reasonable.
If the Health and/or the power on time value indicates - then the number changes (can drop) as you can see.
> but as I learned today that lifetime value can also go down by 200 or 300 in a 1 day just in few hours based on what's happening to it.
Yes, exactly as you can see.
> Still thank you for your reply and for your amazing software that really helps with detecting those kind of issues early on and saves us from losing data.
Thanks for your kind words and support!
Yes, the Help explains how version 1.00 (released 15+ years ago) calculated the Health % value, based on the S.M.A.R.T. attributes.
Yes, that was not magic, and things changed, become much more sophisticated since that, just it is not really possible to describe with all details.
> as it can't predict when something is going to happen and it will update it's values only after something did happen.
Yes, of course, this is how things work: until a sensor and/or error counter do not detect a problem - we can't show that there is an issue with the disk drive
And this is why generally it is recommended to perform some tests (even on a new drive and especially if there is any minor change, degradation - which can indicate more problems) to REVEAL possible problems - or confirm that the drive is stable, really error free:
https://www.hdsentinel.com/faq.php#tests
Until the drive reads/writes a particular sector, the problem may remain unnoticed for long time (even for years) as explained at
https://www.hdsentinel.com/hard_disk_case_bad_sectors.php
This is why the Health % graph on the bottom designed to show any (even minor) change/drop so dramatically, to increase attention to the change.
Ideally there are no more problems - so then the drive can be stable - but in many cases, this is just the first step and (exactly as you can see) more and more problems can appear sooner or later.
> Yesterday it was still 100% health and like 947 days. Today once weak sectors had appeared it dropped to 95% health and 773 days.
> Few hours later now there are 2 bad sectors with 93% health and 741 days. Few minutes later 5 bad sectors with 90% health and like 654 days.
Yes, this is completely "normal" and expected: as more problems detected, the Health % reflects those and the estimated remaining lifetime also decreases. The estimated remaining lifetime value is not a strict countdown, just can indicate how long we can use the drive (if its status is stable), when we should plan replacement because of its age / usage and/or the actual problems with it.
> I don't know if it could actually stop there and this HDD could actually still run for those ~ 600 days,
> but I just didn't want to risk it and decided to replace it right away.
I can completely understand. For a mission critical environment, I'd probably replace too (as you do). Many users replace a hard disk drive even on 99% Health because the value of the data is much more than the disk drive itself (plus possible downtime / maintenance).
For non-critical data, the drive may be still used with constant monitoring (and after testing, exactly to reveal possible further issues).
> I managed to copy everything to a new drive and it all went smoothly thanks to your software since I was aware of this happening early on.
Thanks, good to hear that the data could be saved. Usually yes, in this case, this is possible, if we do not wait more.
Yes, generally this should be the main purpose: as you can see, the Health % may remain stable for long time - until it suddently starts decreasing. This is the point which we need to take action: backup first - and then decide what to do
- start testing, repairing
- or do not risk but replace the disk drive
So it is important to be notified about this point, when the Health % of the hard disk drive starts decreasing.
> In fact I was more surprised of HD Sentinel still showing those 900 days lifetime for this HDD.
May sound weird, but usually some bad sectors can be acceptable, as described at
https://www.hdsentinel.com/faq.php#health
so the first step is to increase attention about the change, degradation and do not show "replace immediately" until really required.
More problems (more drops in the Health) would probably trigger even lower Health % and even (much) lower estimated remaining lifetime to suggest replacement.
> because for all my newest HDD's and SSD'd it shows more than 1000 days and for this it did show an actual number,
Yes of course. If the estimated remaining lifetime would be (much) more, "more than 1000 days" displayed.
I do not feel real / comfortable to say "30 years" or so (even if displayed by other tools), this is why the "1000 days" displayed, as it is more reasonable.
If the Health and/or the power on time value indicates - then the number changes (can drop) as you can see.
> but as I learned today that lifetime value can also go down by 200 or 300 in a 1 day just in few hours based on what's happening to it.
Yes, exactly as you can see.
> Still thank you for your reply and for your amazing software that really helps with detecting those kind of issues early on and saves us from losing data.
Thanks for your kind words and support!