Dorin Bosîi

I have installer nvidia container toolkit, but still cant use the device plugin and I don't understand why root@debian:~# ctr run --rm --gpus 0 -t docker.io/nvidia/cuda:11.0.3-base-ubuntu20.04 cuda-11.0.3-base-ubuntu20.04 nvidia-smi Fri May...

You mean this part: https://docs.nvidia.com/datacenter/cloud-native/kubernetes/install-k8s.html#step-4-setup-nvidia-software I have changed the config.toml to match the same configuration from the website and still doesn't work

Hello, at first I couldn't use the device plugin on the 1.24.0 version and went back to version 1.23.6 where I was already using contained. Some weeks later I tried...

Have you installed NVIDIA Container Toolkit and changed docker configuration to run with Nvidia? At ```/etc/docker/daemon.json``` ``` { "default-runtime": "nvidia", "runtimes": { "nvidia": { "path": "/usr/bin/nvidia-container-runtime", "runtimeArgs": [] } }...

can you show the output of ```kubectl describe node``` and ```nvida-smi```?

and logs from nvidia device plugin?

How many nodes do you have? One of them is missing NVML library... try to find the solution to this

``` root@debian:~# kubectl logs -n kubevirt virt-handler-4w99d -c virt-launcher error: failed to get emulator capabilities error: internal error: Failed to start QEMU binary /usr/libexec/qemu-kvm for probing: libvirt: error : cannot...

This is what i get: ``` root@debian:~# journalctl -b | grep DEN May 18 16:42:55 debian audit[655]: AVC apparmor="DENIED" operation="capable" profile="/snap/core/12834/usr/lib/snapd/snap-confine" pid=655 comm="snap-confine" capability=12 capname="net_admin" May 18 16:42:55 debian audit[655]:...

Thanks, now its working ``` root@debian:~# kubectl get pods -n kubevirt NAME READY STATUS RESTARTS AGE virt-api-77df5c4f87-7mqv4 1/1 Running 2 (87m ago) 19h virt-api-77df5c4f87-wcq44 1/1 Running 2 (87m ago) 19h...