Hard Disk Monitoring

Hi,

I've had HD Sentinel for many years. I built a new PC in late 2018 and have been using it just fine with a boot drive that is a 1 TB Samsung 970 EVO. But I only installed HD Sentinel on this machine today after something happened that concerned me.

I decided to image my boot drive for the first time, long overdue. When I ran Macrium Reflect, it found 3 bad sectors. I ran chkdsk /r, which found a bad sector, I thought that would be the end of it. But I ran Macrium again and it found another bad sector.

HD Sentinel says the drive has only 3% "Percentage Used" but the "Available Spare" is down to 38%, so HD Sentinel is reporting it at 38% Health.

The PC is pushing 3 years old now, and I do use it a lot. I'm just trying to figure out if I need to replace this SSD or not.

I'm pasting the full SMART report below. Appreciate any advice.

--

Critical Warning,,0,00,0,Enabled
Composite Temperature (Kelvin),,321,00,0,Enabled
Available Spare (Percent),,38,00,0,Enabled
Available Spare Threshold,,10,00,0,Enabled
Percentage Used,,3,00,0,Enabled
Data Units Read (512000 b),,"54,372,155",00,0,Enabled
Data Units Written (512000 b),,"157,142,674",00,0,Enabled
Host Read Commands,,"612,878,266",00,0,Enabled
Host Write Commands,,"3,778,938,326",00,0,Enabled
Controller Busy Time (minutes),,"9,533",00,0,Enabled
Power Cycles,,65,00,0,Enabled
Power On Hours,,"21,642",00,0,Enabled
Unsafe Shutdowns,,37,00,0,Enabled
Media and Data Integrity Errors,,142,00,0,Enabled
Number of Error Information Log Entries,,210,00,0,Enabled
Warning Composite Temperature Time (minutes),,0,00,0,Enabled
Critical Composite Temperature Time (minutes),,0,00,0,Enabled
Temperature Sensor 1,,321,00,0,Enabled
Temperature Sensor 2,,340,00,0,Enabled

Hi,

To be short: yes, it is a good time to replace.
As you wrote, 38% Health reported: generally below 50% I'd recommend to plan replacement in the near future - and below 25%, I'd recommend immediate replacement.
If your drive is at 38% Health now, it is expected that it may degrade further with some time/use.

If the Available Spare = 38, it means that only 38% of the total spare area is free, so more than half of the spare area is used instead of the original (bad) sectors.
Yes, SSDs can also have bad sectors, like hard disks (just there are no clicking noises, big delays or so): when the sector fails, it is automatically replaced by the spare sector which originally reserved.
Then the original bad sector is no longer accessible (this is the purpose of the spare area) and all reads/writes targeted to the original sector is redirected to this spare area.
For more information, please check: https://www.hdsentinel.com/ssd_case_bad_sectors.php

Because of the above, the original bad sectors are never causing any more problems - but if their count is high (it is high if only 38% of the spare area is free) then we can expect even more.
Below 10% (see the Available Spare Threshold = 10) the SSD is considered BAD (0 % health reported) and it should be replaced in terms of warranty (if the drive has warranty).

Chkdsk is absolutely useless in this case. It only finds and shows previously undetected issues - but it DOES NOT fix/stabilize/improve the bad sectors, even if we think that.
This described numerous times on this forum, so everybody please STOP THINKING THAT CHKDSK WILL EVER HELP in this case. CHKDSK makes things even worse, especially if you consider cloning because after CHKDSK /R /F, the most common problem is that the drive still can't be cloned due to bad sectors (as they NOT fixed even if we expect that) or the new, perfect cloned drive will also show bad sectors (!)

For example, please check:

https://www.hdsentinel.com/forum/viewto ... 745#p18745
(and the many other links in that topic) or

https://www.hdsentinel.com/forum/viewto ... 651#p19651

or you can check:

https://www.hdsentinel.com/hard_disk_ca ... ectors.php
which gives details/images about the problems with chkdsk.

Testing is generally a good idea, exactly to reveal possible un-detected problems (yes, they may remain undetected until we begin to read/write the appropriate sector) but it is important to use the proper methods, for example the Disk menu -> Surface test -> Read test in Hard Disk Sentinel, as it designed exactly to detect/reveal not only the unreadable sectors but also shows possible retries, slower areas, affected file(s) and generally monitors the operating environment, so shows possible degradatons too.
And the yellow/red blocks can be then really repaired by Disk menu -> Surface test -> Disk Repair function, designed exactly to stabilize them - which allows proper cloning (if required).

However, personally I always recommend cloning only a PERFECT image. Consider the current system as "damaged". While the reallocation (replacing the sectors with the spare area) usually can complete without problems, sometimes it is not true and the original data from the original bad sector lost/corrupted. So generally any kind of reallocation can cause corrupted file(s) and cloning that system may be unstable.
As a general rule, it is better to clone (image) the drive while it is perfect - and then restore the last such perfect image to a new drive.