[smartmontools-support] Incorrect Drive Wearout Indicator on WD Blue SSD

[Christian Franke]

Daniel Hellstern wrote:
> Hi there,
>
> I'm getting what I think is an incorrect drive wearout indicator on a relatively new WD Blue 2TB WDS200T2B0A-00SM50 SSD. While I'm not entirely sure where the problem exists, the indicator shows 1% wearout, which sounds correct if the scale is 0-100%. However, from what I can tell drives are supposed to report in 100-0% (based on the other SSDs I've tested), which is making it difficult to parse this with my monitoring tools that expect this format. I filed an issue on GitHub in September (https://github.com/v-zhuravlev/zbx-smartctl/issues/148) with the monitoring tool, who said it was a SMART issue and to check if it can be resolved by smartmontools or the drive manufacturer. However, WD wasn't very useful because they don't officially support the drive under Linux, and I don't really have a way to use their Windows tool at the moment. Am I correct in assuming that the indicator should be standardized to the 100-0% scale, and if so is this an error in smartmontools or the drive firmware, and are there any workarounds?

In general, a drive can return anything in the SMART attributes as these 
were never part of any ATA standard. The SMART READ DATA command was 
declared obsolete in ATA ACS-4 (2015).

The normalized VALUE should decrease if things get worse. So a 
reasonable "wearout" attribute would be VALUE = 100 - WEAROUT%.

Sample outputs from similar drives suggest that attribute 230 depends on 
firmware version:
WDS200T2B0A/X61190WD: 100 (https://www.smartmontools.org/ticket/1198)
WDS200T2B0A-00SM50/X61130WD: 100 (https://www.smartmontools.org/ticket/1169)
WDS200T1R0A-68A4W0/411000WR: 001 (https://www.smartmontools.org/ticket/1321)
WDS200T2B0A-00SM50/411040WD: 001 (your drive)

To disable the interpretation of a specific attribute, it may help to 
create a local drive database entry and change the name of the attribute 
(see -B option on smartctl man page).

The "Percentage Used Endurance Indicator" from ATA Device Statistics 
(smartctl -l devstat, included in -x, but not in -a) is usually more 
reliable. Device Statistics is vendor independent and supported by many 
recent drives. Monitoring tools should IMO use the values from Device 
Statistics if available.

Thanks,
Christian