How to correctly display the temperature of USB SSD in Report UI?

[molay]

Hello everyone!

I plugged in two USB-to-NVMe SSDs as boot pool devices on HPE ProLiant MicroServer Gen10 Plus v2, and they worked fine.
Each USB enclosures use the JMS583 chip and WD SN740 1TB NVMe SSD, which is a very cost-effective combination.

I can use smartctl to view complete information, including data from the two temperature sensors of each USB-to-NVMe SSD.
But in the Report UI of TrueNAS-SCALE-22.12.4.2, the temperature information of them cannot be seen.
The temperature display of the 4 SATA HDDs is normal.

How do I need to set it up to display the USB SSD temperature correctly?

This machine comes with a PCIe 4.0x16 slot that does not support PCIe Bifurcation, and I haven't thought about how to use it yet.
Maybe install a P4800X AIC SSD for slog or a T1000 for jellyfin or an X710-DA2 network card, or simply leave it empty.

 

[NugentS]

I just checked through my server which has NVMe, SATA & SATA SSD's. All temps are shown although to be honest I don't really care about anything other than the HDD temps as SSD's will thermally throttle.

This implies that the USB bridge is blocking how TN do the reporting.

What happens if you use @joeschmuck's multi_report.sh - does that report the temperature of the "missing" devices?

 

[molay]

What happens if you use @joeschmuck's multi_report.sh - does that report the temperature of the "missing" devices?

I ran the multi_report_v2.4.3_2023_06_16.sh script downloaded from the forum, and then it generated a statisticsmartdata.csv file.
Unfortunately, there was no USB SSD information in it, only the status of the SATA HDDs.
The smartctl command can correctly read the USB SSD status. :(

What is particularly strange is that the Report UI will selectively ignore the Disk Temperature graphs of all USB-to-NVMe SSDs, but will display the Disk I/O graphs of them normally.