vGPU Manager Container Base Image for OpenShift Virtualization 4.14+ #453
Description
The NVIDIA documentation for building the vGPU Manager container in OpenShift Virtualization still instructs users to use a ubi-8 base image.
However, OpenShift 4.14 and newer are RHEL9-based, and should likely use ubi-9 as a base.
From NVIDIA’s documentation:
OpenShift Virtualization vGPU Manager Guide
Open a terminal and clone the driver container image repository:
git clone https://gitlab.com/nvidia/container-images/driver
cd driver
Change to the vgpu-manager directory for your OS:
cd vgpu-manager/rhel8 <============
[...]
docker build \
--build-arg DRIVER_VERSION=${VERSION} \
-t ${PRIVATE_REGISTRY}/vgpu-manager:${VERSION}-${OS_TAG} .
Example Dockerfile:
FROM nvcr.io/nvidia/cuda:13.0.1-base-ubi8 <===
Suggested fix:
Add a RHEL9 Dockerfile extending from ubi-9.
vpu-manager/rhel9/Dockerfile
- FROM nvcr.io/nvidia/cuda:13.0.1-base-ubi8
+ FROM nvcr.io/nvidia/cuda:XX.Y.Z-base-ubi9
Metadata
Metadata
Assignees
Labels
No labels