Skip to content

Conversation

@Future-Outlier
Copy link
Member

@Future-Outlier Future-Outlier commented Jan 2, 2026

Why are these changes needed?

We should test the history server with both a head node and at least one worker node; this makes it easier to find bugs.

How I test it?

  1. create a raycluster
  2. submit a rayjob using cluster selector
  3. delete head pod and worker pod manually, trigger collectors push logs

This directory contains two subdirectories: one for the head node and the other for the worker node.

image image image

E2E test passed
image

Related issue number

#4274

Checks

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

Copy link
Member Author

@Future-Outlier Future-Outlier left a comment

Choose a reason for hiding this comment

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

  1. cc @rueian to merge
  2. cc @JiangJiaWei1103 to write tests to test both head pod and worker pod, note that in head node's folder, we will have additional logs like gcs_server.out.

Co-authored-by: Jun-Hao Wan <[email protected]>
Signed-off-by: Han-Ju Chen (Future-Outlier) <[email protected]>
Copy link
Contributor

@JiangJiaWei1103 JiangJiaWei1103 left a comment

Choose a reason for hiding this comment

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

Make sense to test scenarios including workers!

I'll cover these cases in e2e tests. Thanks!

@rueian rueian merged commit a01ab4e into ray-project:master Jan 6, 2026
28 checks passed
@github-project-automation github-project-automation bot moved this from can be merged to Done in @Future-Outlier's kuberay project Jan 6, 2026
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