[RayJob][Status][19/n] Transition to Complete if the K8s Job fails
#1833
Conversation
This file contains 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
kevin85421
changed the title
Complete if the K8s Job fails
|
cc @andrewsykim |
architkulkarni
approved these changes
Jan 16, 2024
2 tasks
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Outdated suggestions cannot be applied.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.
Why are these changes needed?
#1831 addresses the case where the Ray head can receive and create the Ray job successfully, but the Ray job fails. In that case, the RayJob can transition to
JobDeploymentCompletebecause the KubeRay operator can get the job information from the Ray head and the Ray job isFailed, a terminal state.This PR handles a different case compared with #1831. This PR focuses on the case that the submitter K8s Job can't send a request to the Ray head successfully. Hence, the KubeRay operator can't get any Ray job information from the Ray head, and the RayJob can't transition to the
JobDeploymentCompletestatus. This PR transitions the status toJobDeploymentCompletewhen the K8s Job isFailed.Related issue number
Checks
Create a RayJob using this YAML. The submitter Kubernetes Job within the RayJob attempts to submit a request to a random address, which results in the Job failing. Without this PR, the RayCluster and the Kubernetes Job will not shut down, even when the Kubernetes Job reaches its backoff limit.