Kubuntu’s disk monitoring feature has recently (last few weeks) started giving me alerts about an imminent disk failure on my NVMe M.2 SSD (a Crucial P2) at boot. I have been running sudo smartctl -a /dev/nvme and get output such as this:
SMART overall-health self-assessment test result: PASSED
SMART/Health Information (NVMe Log 0x02)
Critical Warning: 0x00
Temperature: 3 Celsius
Available Spare: 100%
Available Spare Threshold: 5%
Percentage Used: 1%
Data Units Read: 15,515,158 [7.94 TB]
Data Units Written: 16,500,632 [8.44 TB]
Host Read Commands: 193,875,355
Host Write Commands: 227,509,455
Controller Busy Time: 13,258
Power Cycles: 1,402
Power On Hours: 4,282
Unsafe Shutdowns: 164
Media and Data Integrity Errors: 0
Error Information Log Entries: 2,609
Warning Comp. Temperature Time: 0
Critical Comp. Temperature Time: 0
Error Information (NVMe Log 0x01, 16 of 16 entries)
Num ErrCount SQId CmdId Status PELoc LBA NSID VS
0 2609 0 0x1000 0x4005 0x028 0 0 -
So the problem appears to be low temperature and I am seeing values between 0 and 3°C even after the system has been running for some time; when it’s 0°C then it shows a critical warning code. Both the SSD datasheet and the consensus from the closest question on SuperUser say that 0°C is indeed the lowest intended operating temp. Anecdotal evidence from another question about an SSD not booting suggest that cold temps can be a consequence rather than a cause of disk failure, but my SSD has never failed so far.
The reported temperatures are lower than the outside air temps, never mind the temp is in the room, which gets chilly at night, but not that cold! And I have a bog-standard cooling setup: the fan built into the case and the AMD Wraith Spire cooler that came with the APU. Is it really possible that the airflow could lower the SSD temp by 10-20°?!
What, if anything, should I do about this?:
- Yes, all important data is already backed up frequently and in multiple locations (thank you, Duplicati!).
- Could this be a failure of the temperature sensor? Is there any way to check this or get more detailed info from the SMART data?
- I could try to return the SSD (it’s still in warranty) but I’d have to buy another one anyway if I RMAd it plus the faff of reinstalling everything.
- Searching online for “SSD heating” or “SSD insulation” just gets me cooling solutions. Is there something I could safely buy or make to warm up the SSD?
- Any other thoughts?
