Crucial MX500 reports (presumably) false pending sectors

How, what, where and why - when using the software.

Crucial MX500 reports (presumably) false pending sectors

Postby dr0 » 2019.08.31. 17:50

Hello. A few weeks ago I bought myself a 1TB version of Crucial MX500 SSD. On 10th day of use, Hard Disk Sentinel reported that the drive health has dropped to 99%. At first, I thought that it was probably due to S.M.A.R.T. attribute #202 (Percentage Of The Rated Lifetime Used) increasing its value, which is normal for SSDs as NAND gets worn out by writes. But that wasn't the case. When the first health drop occurred, the SSD had only 500GB of lifetime writes, but then it got even weirder when the drive health jumped back to 100%. I checked the log and saw this:
Code: Select all
8/22/2019 1:21:07 AM,#197   Current Pending Sector Count  1 -> 0
8/22/2019 1:16:03 AM,#197   Current Pending Sector Count  0 -> 1

After that, the drive had 2 more events of Pending Sector going to 1 and then back to 0:
Code: Select all
8/27/2019 7:59:30 PM,#197   Current Pending Sector Count  1 -> 0
8/27/2019 7:54:25 PM,#197   Current Pending Sector Count  0 -> 1
8/24/2019 10:53:10 PM,#197   Current Pending Sector Count  1 -> 0
8/24/2019 10:48:06 PM,#197   Current Pending Sector Count  0 -> 1

After some googling, I learned that this is a weird quirk that plagues basically every MX500 SSD from Crucial: https://utcc.utoronto.ca/~cks/space/blog/tech/SMARTAlarmingFlakyErrors
Some people say that this is a firmware bug that unfortunately hasn't been fixed to this day.
If I had known this before I would probably have thought twice before purchasing this drive. But I got it for a really good price and this behavior doesn't seem to affect performance or data integrity or system stability in any way. Plus the drive comes with 5 years of warranty. So I decided not to return it, do regular backups and hope that someday Crucial will finally release a firmware that fixes this.

I was wondering if it's possible to configure Hard Disk Sentinel in a way so that it would ignore #197 Pending Sector Count attribute and wouldn't change the reported health percentage because of it and only keep saving the changes to the log and showing in the corresponding tab? Because the current situation doesn't seem to reflect the real condition:

Image

Especially after I was experimenting with the program settings and the reported drive health has dropped to 50% when I tried to select "Analyse vendor-specific values" as health calculation method in Advanced Options:

Image

Afterward, I switched back to the recommended method of health calculation but the health graph hasn't changed.
So, my question is: Is there a way to clear only the Health graph without losing other stats? And prevent the Health graph from changing in the future due to the Pending Sector going to 1 and then back to 0?

Image

For now, I just set an offset of -1 for it, but I'm not sure how it will affect the Health graph and whether the changes still will be logged because I still would like to be able to read the log and see when exactly false pendings occurred.
Last edited by dr0 on 2019.09.03. 13:06, edited 1 time in total.
dr0
 
Posts: 11
Joined: 2018.01.01. 02:05

Re: Crucial MX500 reports (presumably) false pending sectors

Postby hdsentinel » 2019.09.03. 09:42

Thanks for your message.

As I know, the issue is related to the disk controller (motherboard chipset) driver which manages the SSD: in some motherboard/driver combinations the SSD records a pending sector sometimes on different events, usually on shut down. If possible, I'd check if an updated driver is available as it may make things better.

Yes, when Hard Disk Sentinel detects the change of the attribute, it does what it should: add the change to the Log page and also the Health % value reflects the change too. When the health improves back, the graph on that particular day(s) still show the low value, as you can see, because it designed to show the daily lowest health detected.

If you prefer, you can clear ALL statistics (including the Health % graph of all devices) any time if you completely close Hard Disk Sentinel by File -> Exit and delete the file HDSentinel.sta from the folder of the software. This clears all statistics. If you prefer to delete only some values of the appropriate device, it requires manual editing of the file, so please let me know, I can assist.

> For now, I just set an offset of -1 for it, but I'm not sure how it will affect the Health graph and whether the
> changes still will be logged because I still would like to be able to read the log and see when exactly false pendings occurred.

Yes, this is a good solution. Or you may uncheck the mark under "Enable" column on the S.M.A.R.T. page, next to the "Offset" value. This way that particular attribute will be not used to determine health, so its change will not affect the health %.
User avatar
hdsentinel
Site Admin
 
Posts: 2213
Joined: 2008.07.27. 17:00
Location: Hungary

Re: Crucial MX500 reports (presumably) false pending sectors

Postby dr0 » 2019.09.03. 12:58

If you prefer to delete only some values of the appropriate device, it requires manual editing of the file, so please let me know, I can assist.

I'd appreciate that as I really need your guidance here.
The drop to 50% somehow cleared itself after some time. Now I would like to only hide these 3 drops to 99%:

Image

Or you may uncheck the mark under "Enable" column on the S.M.A.R.T. page, next to the "Offset" value. This way that particular attribute will be not used to determine health, so its change will not affect the health %.

Which method is better: setting it to -1 or completely unchecking it, if I know that it only changes by 1 digit and constantly goes up and then back down and if I want it not to affect the Health percentage but keep its changes recorded and displayed on the Log tab?
dr0
 
Posts: 11
Joined: 2018.01.01. 02:05

Re: Crucial MX500 reports (presumably) false pending sectors

Postby dr0 » 2019.09.04. 20:59

Can you elaborate further on the subject of checkmarks on the S.M.A.R.T. tab? If, say, I was to remove a checkmark for #197 attribute and it would change in the future, would the changes still be displayed as a raw data on the S.M.A.R.T. tab? Or would it completely be ignored altogether and just always be displaying 0 even if that is not the case? I mean, what exactly that checkmark disables?
dr0
 
Posts: 11
Joined: 2018.01.01. 02:05

Re: Crucial MX500 reports (presumably) false pending sectors

Postby hdsentinel » 2019.09.05. 10:34

The checkmark controls if the corresponding attribute should be used in determining the health value of the disk drive.
For example, if you un-mark the checkmark for attribute #197, then its possible change will not affect the health %.

The attribute with its current value is always displayed on the S.M.A.R.T. page and you can examine possible change on the graph below the attribute list, so you can see how it changes - just the change will not affect the health value.
User avatar
hdsentinel
Site Admin
 
Posts: 2213
Joined: 2008.07.27. 17:00
Location: Hungary


Return to Questions

Who is online

Users browsing this forum: No registered users