SAMSUNG MZ1LB3T8HMLA-00007 - M.2 NVMe PM983 3.84 TB drive sensor 3 hot

[MarkLambda]

I have a machine crashing.  The OS is running off the:

Model Number: SAMSUNG MZ1LB3T8HMLA-00007

(PM983 M.2 NVMe 3.84TB drive).

All other sensors on the motherboard, GPUs, etc are running cool well below the thresholds.

This NVMe drive has three sensors, and sensor 3 is running hotter consistently.

From smartctl:

Model Number: SAMSUNG MZ1LB3T8HMLA-00007

Available Spare: 100%
Available Spare Threshold: 10%
Percentage Used: 0%

Temperature Sensor 1: 67 Celsius
Temperature Sensor 2: 83 Celsius
Temperature Sensor 3: 94 Celsius

And from lm-sensors:

CPU is between 32C-42C - so that is fine.  But the NVMe drive:
nvme-pci-b300
Adapter: PCI adapter
Composite: temp1_input: 72.850
Sensor 1:  temp2_input: 72.850
Sensor 2: temp3_input: 87.850
Sensor 3: temp4_input: 99.850   <--- NOTE this varies this is the highest I have seen but well above 70-80C.

It passes smartctl and shows only info messages, no errors or other sign of failure.

It is currently just open air cooled. 

    1. Do you think this is causing a system failure?

   2. I was suggesting adding a M.2 FAN as the motherboard supports it and a heat sink or similar. 

[antikythera]

Ignore it, the issue is not your nvme drive and smartctl cannot read the sensors on nvme drives that well. I'd suggest looking into your processor suspend states if you have a Ryzen. My own had issues with crashing because of unsupported suspend states being requested by the Linux kernel. I'm not on the machine at present, there's a command you can add to grub to workaround that issue though. It's to do with c-states. Search for it on Google but include GitHub in the search terms and the answer should be up the top of the list