gateway: rewrite transcripts atomically to avoid mid-write corruption - #11083
Closed
Xowiek wants to merge 1 commit into
Closed
gateway: rewrite transcripts atomically to avoid mid-write corruption#11083Xowiek wants to merge 1 commit into
Xowiek wants to merge 1 commit into
Conversation
Contributor
|
Thanks for the contribution, @Xowiek! Closing this as a duplicate of #4985 (by @binhnt92), which targets the same fix/feature. We're consolidating on that PR for review. If you want to help push it over the line, please jump in there — or if you think your approach is better for a specific reason that isn't covered in the other PR, let us know and we can reopen. |
6 tasks
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
SessionStore.rewrite_transcript()was rewriting legacy JSONL transcripts in-place with"w"mode. If Hermes crashed or a serialization error happened mid-rewrite, the transcript file could be left truncated even though the previous version was still valid.This patch makes transcript rewrites atomic by writing to a temporary sibling file, flushing and fsyncing it, and then swapping it into place with
os.replace().Why
This is a small reliability fix, but it closes a real failure mode in a sensitive code path used by
/retry,/undo, and/compress.Before this change:
load_transcript()calls could observe truncated stateAfter this change:
Changes
gateway/session.pysorewrite_transcript()uses:flush()+os.fsync()os.replace()tests/gateway/test_session.pyfor a simulated mid-write serialization failureTest coverage
Ran: