fix(scheduler): retry failed nodes when resuming a job#146
Draft
SavinRazvan wants to merge 1 commit into
Draft
Conversation
On resume, FAILED checkpoints were added to the terminal failed set, so BackgroundRuntime.resume_job() could not re-execute failed nodes. Only COMPLETED checkpoints are now treated as skipped on resume. Author: Savin I. Razvan GitHub-User: @SavinRazvan Assisted-by: Cursor Co-authored-by: Razvan <SavinRazvan@users.noreply.github.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
BackgroundRuntime.resume_job()is documented to recover failed and cancelled jobs, but the scheduler treatedFAILEDcheckpoints as terminal on resume. Failed nodes were added to thefailedset and never re-executed, so a resumed job would immediately finish as failed without retrying the node.The background-runtime test suite worked around this by swapping the graph handler before calling
resume_job().Fix
On
resume=True, onlyCOMPLETEDcheckpoints are skipped.FAILED,RUNNING, andCANCELLEDcheckpoints no longer block re-execution — their handlers run again so operators can recover failed jobs.Tests
test_scheduler_resume_retries_failed_checkpoint_nodesto assert a failed checkpoint is retried and completes on resume.Verification
Slack Thread