Make step_latest symlink update atomic - #62
Merged
Hannibal046 merged 1 commit intoJul 9, 2026
Conversation
safe_symlink unlinks the old symlink before creating the new one. A crash between the two calls deletes step_latest entirely, so the next launch finds no checkpoint and silently retrains from scratch. Create the symlink under a temporary name and os.replace it over the target instead, so step_latest always points at a complete checkpoint.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
safe_symlinkunlinks the old symlink before creating the new one. A crash between the two calls deletesstep_latestentirely, sodiscover_latest_checkpoint()returns None and the next launch silently retrains from scratch — even though a complete checkpoint is sitting on disk.Change
Create the symlink under a temporary name and
os.replaceit over the target. The replace is atomic, sostep_latestalways points at a complete checkpoint.