The installation of GPU Operator version v24.6.1 fails to install the /usr/local/nvidia/toolkit - Githubissues

NVIDIA / gpu-operator

NVIDIA GPU Operator creates, configures, and manages GPUs in Kubernetes

https://docs.nvidia.com/datacenter/cloud-native/gpu-operator/latest/index.html

Apache License 2.0

1.83k stars 297 forks source link

The installation of GPU Operator version v24.6.1 fails to install the /usr/local/nvidia/toolkit #925

Open coderRenxy opened 2 months ago

coderRenxy commented 2 months ago

1. Quick Debug Information

Ubuntu20.04
Kernel Version: 5.4.0-192-generic
Container Runtime Type/Version:Containerd/1.7.12
K8s Flavor/Version:k8s/v1.28.2
GPU Operator Version: v24.6.1
GPU version: A40
local GPU driver: 560.28.03

2. Issue or feature description

When installing the GPU Operator version v24.6.1, the nvidia-toolkit fails to install. However, installing version v24.3.0 successfully installs the toolkit.
Customization options are not producing the desired effects.

3. Steps to reproduce the issue

helm install --wait --generate-name \
-n gpu-operator --create-namespace \
nvidia/gpu-operator

helm install --wait --generate-name -n gpu-operator --create-namespace nvidia/gpu-operator --set operator.defaultRuntime=containerd --set driver.enabled=false --set driver.nvidiaDriverCRD.enabled=true --version v24.3.0

cdesiniotis commented 2 months ago

@coderRenxy please provide relevant logs and more details on the issue you are encountering.