fix: prevent run directory from being wiped on job requeue #19
Merged
fix: prevent run directory from being wiped on job requeue #19
Conversation
setup_run_directory() would delete the active run directory whenever debug.override_existing=True, including when called from train.py on compute nodes during a running job. Add allow_override=False default so the rmtree only fires when submit.py explicitly opts in.
KonstiNik
approved these changes
Apr 30, 2026
Collaborator
KonstiNik
left a comment
KonstiNik
left a comment
There was a problem hiding this comment.
LGTM. Addition of a small logic layer to prevent submission time wipe.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Gate the destructive run-directory wipe behind an
allow_overrideparameter sotrain.pyon compute nodes can never accidentally delete an active run directory.Previously,
setup_run_directory()calledshutil.rmtreeon the run dir wheneverdebug.override_existing=True, regardless of caller. On a requeue/resume,train.pyrunssetup_run_directory()again — wiping checkpoints mid-job. Fix: addallow_override: bool = False; onlysubmit.py(the submission-time entrypoint) passesTrue.Type of change