As you can see in the sensor graph(teal color) and min/max in the table the composite and temp1 sensors sometimes shoot straight to exactly 84°C and back down again . Generally these are small spikes but 84°C is well beyond the specified operating temperature(0°C to 70°C) for the 980.
The temp2 sensor stays around 28 to 55°C
I have a thermal grizzly M.2 SSD Cooler installed on the SSD. I removed all 4 of the plastic strips from the thermal pads.
Is this detrimental to my SSD's longevity and performance? And what could be causing this?
I've had this issue too. I'm honestly not sure if it's an actual issue with the drive or a bug in the reporting. Temp2 always stays the same but temp1 will shoot up. Also, it'll likely be OK since the drive is thermal throttling (if it is a real value, but I'm skeptical)
Also, getting the *exact* same 84c value as you. Also on Linux. So yeah, I'm going to toss it up to a bug. One day I'll check on Windows. I've had this drive for several months but only used Pop! and Windows on it, when I recently installed Arch the NVME Log showed 0 temp warnings. I've had Arch installed for ~a week and it's up to 101 warnings in the NVMe log.
Have the same issue. Running Arch linux. I noticed this since KDE/Plasma popped up notification about SMART for Samsung NVME 980. Exactly this issue with spikes to 84°C. It does not look realistic to me, though what do I know?
Would be interesting if Samsung would comment on this. If this is actual spikes or some issue with sensor readings.
One thing I'd urge to try: check if issue persists on an LTS Linux kernel. On Arch Linux I switched to linux-lts kernel version 5.10.77 and I do not observer any spikes. Previously I was running kernel version 5.14.16 and spikes in temperature were occasionally appearing.
Per your post I also dropped back to lts 5.10 kernel and I have only had one or two spikes in the past 2 weeks. With the 5.17 kernel I had them multiple times per day.
I have this issue as well with my 980nvme on 5.14.13 kernel. I checked with psensors and iotop running and there was very little IO happening when it spiked to 84c. There doesn't seem to be much performance hit either, this really sounds like sensor error.
Same 84c problem. Opensuse Tumbleweed (20211116), jumping from say 34c to 84c and then back to 34c. I logged it and it would stay at a noromal lower temp and then spike for 84c for 10 to 80 seconds before jumping back to the lower temp, no pattern I could see. There was no temperature transition in-between and it did not feel hot to the touch when it was logging 84c. Very low disk activity during logging.
I have a 970 with win11 on the same machine. No problem with spiking. I installed Opensuse Leap 15.3 on the 980 and did not see the problem occur. I decided to migrate the win11 os to the 980 and install Tumbleweed on the 970. Tumbleweed is not spiking to 84c on the 970 and I don't see any spiking on the 980 with windows.
Seems to be a problem with the 980 and the newest cutting edge linux operating systems.
I am running Debian 11 on kernel 5.14.xx (don't remember the full number currently) and am having the same 84°c issue with my 980 1tb nvme.
I am worried this is causing issues as the "Thermal Temperature 2 Transition Count" is slowly increasing. From nvme specs this is when the drive needs to heavily thermally throttle to reduce heat. I'm not sure if it's bad for the drive to be doing this all the time if the temperature reading is actually wrong.
I am going to check the actual temperature of the drive later today with an infrared thermometer gun and physically touching it with my finger when it's reporting 84° to see what the temperature really is.
Edit: want to add that my drive seems to sit at 84c constantly after a short time after reboot. I am using lm-sensors and smartctl to get the temp.
I am experiencing this as well with an UNRAID install which is built on Slackware i believe.
I noticed that there is a new firmware out for this drive labelled 2B4QFXO7 and I am going to try that out. By looking at the contents it was dated Nov 30, 2021 so I'm hopeful.
My current drive with the issue is a 980 500GB drive with firmware 1B4QFXO7.
Might be useful to identify the firmware levels that are experiencing this.