Skip to content

Conversation

@Narwhal-fish
Copy link

Why are these changes needed?

The ray-cluster Helm chart currently does not support setting runtimeClassName for Ray pods. This field is required when running Ray clusters with GPU runtimes.
This PR adds support for specifying runtimeClassName directly in the ymal file.

worker:
  runtimeClassName: nvidia

Related issue number

#3793

Checks

  • I've made sure the tests are passing.
  • Testing Strategy
    • Unit tests
    • Manual tests
    • This PR is not tested :(

@win5923
Copy link
Collaborator

win5923 commented Nov 10, 2025

Could you also install pre-commit to fix linting errors?
https://github.com/ray-project/kuberay/blob/master/ray-operator/DEVELOPMENT.md#pre-commit-hooks

Copy link
Collaborator

@win5923 win5923 left a comment

Choose a reason for hiding this comment

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

LGTM, thank you.

@win5923
Copy link
Collaborator

win5923 commented Nov 13, 2025

Hi @Narwhal-fish, Could you rebase onto the main branch? We’ve just fixed the CI.

@Narwhal-fish Narwhal-fish force-pushed the Support-runtimeClassName-in-values-yaml branch from 5277471 to f864b30 Compare November 13, 2025 17:19
@Narwhal-fish
Copy link
Author

Hi @Narwhal-fish, Could you rebase onto the main branch? We’ve just fixed the CI.

Sure!

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.

2 participants