Skip to content

[Helm] Supports HostNetwork Mode for each components - #1365

Merged
rahulait merged 2 commits into
NVIDIA:mainfrom
ssup2:main
Apr 15, 2026
Merged

[Helm] Supports HostNetwork Mode for each components#1365
rahulait merged 2 commits into
NVIDIA:mainfrom
ssup2:main

Conversation

@ssup2

@ssup2 ssup2 commented Aug 13, 2025

Copy link
Copy Markdown
Contributor

This PR adds support for hostNetwork mode in the Helm Chart.

Using Host Network mode can be useful in environments where the host and pods share the same network CIDR (non-overlay network), as it prevents unnecessary IP allocation for daemon pods, thereby reducing overall IP usage. This option is beneficial in cloud environments such as AWS, Azure, and GCP.

@copy-pr-bot

copy-pr-bot Bot commented Aug 13, 2025

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@ssup2
ssup2 force-pushed the main branch 2 times, most recently from ccb3b11 to 40d61f2 Compare August 31, 2025 03:49
@ssup2

ssup2 commented Aug 31, 2025

Copy link
Copy Markdown
Contributor Author

@elezar Hello. Please review this pr?

@github-actions

Copy link
Copy Markdown

This PR is stale because it has been open 90 days with no activity. This PR will be closed in 30 days unless new comments are made or the stale label is removed. To skip these checks, apply the "lifecycle/frozen" label.

@github-actions github-actions Bot added the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Nov 29, 2025
@ssup2

ssup2 commented Dec 4, 2025

Copy link
Copy Markdown
Contributor Author

@elezar Hi! I'd appreciate it if you could take a look at this PR when you get a moment.

@github-actions github-actions Bot removed the lifecycle/stale Denotes an issue or PR has remained open with no activity and has become stale. label Dec 5, 2025
@rahulait

rahulait commented Apr 6, 2026

Copy link
Copy Markdown
Contributor

Thanks @ssup2 . Can you please also add default value to values.yaml so that its easier to find what one can set. So something like:

# In the devicePlugin section (~line 103):
devicePlugin:
  enabled: true
  # enableHostNetwork when set to true runs the pod with hostNetwork: true.
  # This can be useful in non-overlay network environments (e.g., AWS, Azure, GCP)
  # to avoid unnecessary IP allocation for daemon pods.
  enableHostNetwork: false

Similar change for gfd and mps as well.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Adds optional per-component HostNetwork support to the Helm chart so daemon pods can run on the node network and avoid allocating pod IPs in non-overlay environments.

Changes:

  • Add hostNetwork: true + dnsPolicy: ClusterFirstWithHostNet to the device plugin DaemonSet when devicePlugin.enableHostNetwork is set.
  • Add hostNetwork: true + dnsPolicy: ClusterFirstWithHostNet to the GFD DaemonSet when gfd.enableHostNetwork is set.
  • Add hostNetwork: true + dnsPolicy: ClusterFirstWithHostNet to the MPS control daemon DaemonSet when mps.enableHostNetwork is set.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File Description
deployments/helm/nvidia-device-plugin/templates/daemonset-device-plugin.yml Conditionally enables host networking for the device plugin pod.
deployments/helm/nvidia-device-plugin/templates/daemonset-gfd.yml Conditionally enables host networking for the GFD pod.
deployments/helm/nvidia-device-plugin/templates/daemonset-mps-control-daemon.yml Conditionally enables host networking for the MPS control daemon pod.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comment thread deployments/helm/nvidia-device-plugin/templates/daemonset-gfd.yml
Signed-off-by: Jungsub Shin <supsup5642@gmail.com>
@rahulait

Copy link
Copy Markdown
Contributor

Gentle reminder @ssup2 . Please update the values.yaml with default values so that its easier to find how it can be configured.

Signed-off-by: Rahul Sharma <rahulsharm@nvidia.com>
@rahulait

Copy link
Copy Markdown
Contributor

/ok to test 20f2c29

@rahulait
rahulait merged commit b4cf0a1 into NVIDIA:main Apr 15, 2026
11 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants