Repository navigation
Report safe judge retry diagnostics - #517
Merged
vvbandeira merged 1 commit intoOct 13, 2026
Merged
vvbandeira merged 1 commit into
vvbandeira merged 1 commit into
Conversation
Signed-off-by: Vitor Bandeira <vvbandeira@precisioninno.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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Judge timeouts stopped three full evaluations after the backend answered all 100 questions.
The error message reported the batch limit for every TimeoutError, while retry failures had no diagnostics.
Log failed judge attempts with the exception class, integer provider code, count, call limit, elapsed time, and next retry delay.
Report the observed timeout and both configured limits without assigning an unproven cause.
The warning omits prompts, responses, exception text, and credentials.
Retry policy, timing limits, models, dataset, and score gates stay unchanged.
Validation used the locked evaluation environment: mypy checked 43 files, Ruff lint and format passed, 12 unittest cases passed, and 95 evaluation pytest cases passed.
Tests cover temporary failures, exhausted retries, safe output, and cancellation without retries.
The README and script comment now explain both limits.
Risk: merging starts the existing paid full evaluation on master.
Local checks used Python 3.14.7; PR CI uses Python 3.12.