Repository navigation
test(platform): wait boundedly for the drained Windows descendant to signal - #116
Merged
Merged
Conversation
…signal suspended_admission_assigns_descendants_to_job_and_drains_them polled the terminated descendant's process handle once with a zero timeout right after terminate_and_drain. Drain completes when the Job reports no active processes, and Windows can drop that count shortly before the terminated process object is signaled, so the poll intermittently returned WAIT_TIMEOUT (258) on CI (#111 and #113 runs on 2026-11-06). Wait up to five seconds instead. The test still proves the descendant was assigned to the Job and terminated by the drain. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Contributor
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars.
|
Contributor
[Low impact] This test-only change appears safe to merge. SummaryThe Windows descendant-drain test now waits up to five seconds for the descendant’s process handle to signal instead of checking it immediately.
Reviews (1) · Last reviewed commit: "test(platform): wait boundedly for the d..." · Reviewed by Greptile |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes an intermittent Windows CI failure that hit #111 and #113 today:
Cause
After
terminate_and_drain, the test checked the descendant's process handle with a 0 msWaitForSingleObject. Drain completes when the Job'sActiveProcessesreaches zero (terminate_and_has_members). Windows can drop that count slightly before the terminated process object is signaled, so the immediate check sometimes caught that gap.Fix
Wait up to 5 s instead of 0 ms. The test still asserts that the descendant was in the admitted Job, that it was terminated by the drain, and that the Job has no members afterwards. No product code changes. Pumas holds no handle to arbitrary descendants, so the Job count stays the drain's completion signal.
Verification
cargo check --tests --target x86_64-pc-windows-gnu -p pumas-librarypasses. This test iscfg(windows), so Windows CI is what runs it.cargo fmt --checkis clean.🤖 Generated with Claude Code