fix(test): per-combo retry to absorb startup-race flake in pane-topology harness (#22) - #71
Merged
Merged
Conversation
…ogy harness (#22) The settle gate (822fbc0) reduced but didn't eliminate the flake — under the FULL integration suite's Docker load, direct harness runs still varied (66/0, then 63/3, 64/2 on consecutive runs). Varying counts across identical runs = transient container/pane startup races, not a real defect. Wrap each combo's launch→inspect→teardown in a bounded 2-attempt retry with a PASS/FAIL snapshot-restore: if an attempt logs any NEW failure, discard its tally and re-run the combo once. Crucially this does NOT mask real bugs — a genuine failure recurs on the retry (fails both attempts) and its tally is kept and reported; only transient flake is absorbed. A `RETRY [combo] …` line is printed when a retry fires, so masked flake is still visible in the log. Body left at its original indent (bash ignores it) to keep the wrap a minimal, reviewable diff. Harness-only; not run in CI (integration acceptance is maintainer-run). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Follow-up to the settle gate (#822fbc0), which reduced but didn't eliminate the flake: under the full integration suite's Docker load, consecutive direct harness runs still varied (66/0 → 63/3 → 64/2). Varying counts across identical runs = transient container/pane startup races, not a real defect (every geometry/identity/count assertion is correct — confirmed earlier by a clean 66/0).
Fix
Wrap each combo's launch→inspect→teardown in a bounded 2-attempt retry with a PASS/FAIL snapshot-restore. If an attempt logs any new failure, discard its tally and re-run the combo once.
Does not mask real bugs: a genuine failure recurs on the retry (fails both attempts) and its tally is kept + reported; only transient flake is absorbed. A
RETRY [combo] …line prints whenever a retry fires, so any masked flake stays visible in the log.Harness-only;
bash -nclean, §80 structural guard still matches. Not exercised by CI (integration acceptance is maintainer-run) — validated by a maintainer re-run showing steady 66/0.