Fix deployment on arm64 architecture #109
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Fix service build and deployment on arm64 architecture.
There was a problem hiding this comment.
Pull Request Overview
This PR enables deployment on arm64 architecture by updating Docker configurations and Kubernetes deployment templates to support multi-architecture builds and private container registries.
Key changes:
- Adds imagePullSecrets configuration to deployment templates for private registry authentication
- Creates new Dockerfiles for device plugins and kube-scheduler to enable custom image builds
- Updates deployment scripts to use configurable registry prefixes and image tags
Reviewed Changes
Copilot reviewed 13 out of 13 changed files in this pull request and generated 5 comments.
Show a summary per file
| File | Description |
|---|---|
| src/hivedscheduler/deploy/hivedscheduler.yaml.template | Adds imagePullSecrets for private registry access |
| src/hivedscheduler/build/kube-scheduler.k8s.dockerfile | New Dockerfile for kube-scheduler image |
| src/hivedscheduler/build/hivedscheduler.k8s.dockerfile | New Dockerfile replacing the common dockerfile with updated license header |
| src/hivedscheduler/build/hivedscheduler.common.dockerfile | Removed in favor of k8s-specific dockerfile |
| src/frameworkcontroller/deploy/frameworkcontroller.yaml.template | Adds imagePullSecrets for private registry access |
| src/device-plugin/deploy/start.sh.template | Updates to use configurable registry and adds imagePullSecrets injection |
| src/device-plugin/deploy/service.yaml | Adds device-plugin.yaml to template list |
| src/device-plugin/deploy/device-plugin.yaml.template | Replaces hardcoded image with configurable registry and adds imagePullSecrets |
| src/device-plugin/build/k8s-rocm-device-plugin.k8s.dockerfile | New Dockerfile for AMD ROCm device plugin |
| src/device-plugin/build/k8s-rdma-shared-dev-plugin.k8s.dockerfile | New Dockerfile for RDMA device plugin |
| src/device-plugin/build/k8s-nvidia-device-plugin.k8s.dockerfile | New Dockerfile for NVIDIA device plugin |
| src/device-plugin/build/k8s-host-device-plugin.k8s.dockerfile | New Dockerfile for host device plugin |
| contrib/kubespray/script/environment.sh | Adds force flag to symlink commands to prevent errors on re-runs |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| # Copyright (c) Microsoft Corporation. | ||
| # Licensed under the MIT License. | ||
|
|
||
| FROM ghcr.io/mellanox/k8s-rdma-shared-dev-plugin:v1.5.3 |
There was a problem hiding this comment.
The deployment script references v1.4.0 in start.sh.template (line 86), but this Dockerfile uses v1.5.3. This version mismatch could cause deployment issues. Consider aligning both to use the same version.
Suggested change
| FROM ghcr.io/mellanox/k8s-rdma-shared-dev-plugin:v1.5.3 | |
| FROM ghcr.io/mellanox/k8s-rdma-shared-dev-plugin:v1.4.0 |
There was a problem hiding this comment.
the v1.4.0 daemon set yaml uses latest tag, @hippogr to confirm current version
yukirora
reviewed
Oct 27, 2025
Use buildkit for Docker build.
Upgrade NVIDIA k8s device plugin version.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix service build and deployment on arm64 architecture.