In certain cases it might be needed to update the Nvidia driver of a GPU VM to a more recent version.

Also, in case of old provisioned VMs still running not supported version of the Nvidia driver and/or after a major upgrade of the underlying infrastructure it might happen that the GPU device is not recognised anymore and the following message is returned when running "nvidia-smi" command:

$ nvidia-smi
No devices were found

GPU Nvidia Driver Update procedure

You may update to the latest compatible driver with the following command:

sudo dnf install -y https://nexus.ecmwf.int/repository/EWC-Common/nvidia-linux-grid-580-580.159.03-1.x86_64.rpm && sudo reboot


After the reboot, once your instance is running, it should be possible to verify the availability of the GPU and the new driver with the "nvidia-smi" command:

$ nvidia-smi
Fri Jun 19 13:16:29 2026       
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 580.159.03             Driver Version: 580.159.03     CUDA Version: 13.0     |
+-----------------------------------------+------------------------+----------------------+
| 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  GRID A100D-2-20C               Off |   00000000:05:00.0 Off |                   On |
| N/A   N/A    P0            N/A  /  N/A  |       3MiB /  20480MiB |     N/A      Default |
|                                         |                        |              Enabled |
+-----------------------------------------+------------------------+----------------------+
 ...



  • No labels