Arch Linux

gnome_is_ugly

Member

Registered: 2022-08-12

Posts: 14

Determining actual gpu/ssd temperature and detailed ssd health

I want to know my cpu/gpu/ssd temperature.

sensors

gigabyte_wmi-virtual-0
Adapter: Virtual device
temp1:        +33.0°C
temp2:        +33.0°C
temp3:        +45.0°C
temp4:        +40.0°C
temp5:        +37.0°C
temp6:        +41.0°C

nvme-pci-0400
Adapter: PCI adapter
Composite:    +40.9°C  (low  = -273.1°C, high = +81.8°C)
                       (crit = +84.8°C)
Sensor 1:     +40.9°C  (low  = -273.1°C, high = +65261.8°C)
Sensor 2:     +42.9°C  (low  = -273.1°C, high = +65261.8°C)

acpitz-acpi-0
Adapter: ACPI interface
temp1:        +16.8°C  (crit = +20.8°C)

k10temp-pci-00c3
Adapter: PCI adapter
Tctl:         +45.6°C
Tccd1:        +38.8°C
Tccd2:        +40.5°C

nvme-pci-0100
Adapter: PCI adapter
Composite:    +41.9°C  (low  = -273.1°C, high = +81.8°C)
                       (crit = +84.8°C)
Sensor 1:     +41.9°C  (low  = -273.1°C, high = +65261.8°C)
Sensor 2:     +46.9°C  (low  = -273.1°C, high = +65261.8°C)

What number among tempN is cpu temperature (if any)?
Also what does k10temp-pci-00c3 means? It's not gpu, because nvidia-smi gives different numbers (looks like I should just use data from nvidia-smi, don't know for sure though).
And what temperature (sensor 1 or 2) give the correct one for nvme disks? Also 65261 number doesn't look normal.
Also how can I get fan speed? And what is a good tool for checking ssd "health"? Is the following the only thing I can do?

sudo smartctl -H /dev/nvme0n1

It only give status PASSED. Is there a way to know how health disk is (approximate remaining time until it's dead)?
Sorry for so many questions in one post.

loqs

Member

Registered: 2014-03-06

Posts: 18,964

Re: Determining actual gpu/ssd temperature and detailed ssd health

What number among tempN is cpu temperature (if any)?

Have you read https://docs.kernel.org/hwmon/k10temp.html#description?

Last edited by loqs (2022-08-28 21:30:56)

seth

Member

From: Won't reply 2 private help req

Registered: 2012-09-03

Posts: 76,491

Re: Determining actual gpu/ssd temperature and detailed ssd health

It only give status PASSED. Is there a way to know how health disk is (approximate remaining time until it's dead)?

https://wiki.archlinux.org/title/S.M.A.R.T.

because nvidia-smi gives different numbers

Which is gonna be the relevant one. Try

nvidia-smi --format=csv,noheader,nounits --query-gpu="temperature.gpu"

Sorry for so many questions in one post.

Don't be sorry, just don't do it.
https://wiki.archlinux.org/title/Genera … ow_to_post - item #4

---

How to upload text · How to boot w/o GUI · Disable Windows Fast-Start! · Fix your xinitrc