I am using on a Centos7.9 Unix HPC Cluster, that uses NVIDIA cards (various ones, but a P100 is a common one) in a Virtualized Desktop Envrionment.
One of our commercial software packages uses nvidia-settings to get values, but fails as nvidia-settings throws an error. From a shell you can see this clearly.
nvidia-settings
ERROR: Unable to load info from any available system
Can anyone advise us on how to fix this? nvidia-smi returns correct values, and everything else appears to work normally with respect to the graphics card. The solutions for Prime and Optimus (mentioned in this Forum) do not apply to our use-case.
thanks
Simon
nvidia-smi
Wed Mar 2 15:13:51 2022
±----------------------------------------------------------------------------+
| NVIDIA-SMI 470.82.01 Driver Version: 470.82.01 CUDA Version: 11.4 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|===============================+======================+======================|
| 0 Tesla P4 On | 00000000:00:07.0 Off | 0 |
| N/A 35C P8 6W / 75W | 15MiB / 7611MiB | 0% Default |
| | | N/A |
±------------------------------±---------------------±---------------------+