Repository navigation
Conversation
…nding probe deadlines (HACK-1452)
The cancellation refusal already joins its probe child: createNativeComposeProbe
awaits child.exited on every path before it rethrows, and Bun resolves exited
only after reaping. The leaked-child report came from the test: it read the PID
after the abort, from a "started" file the fake Docker child created and was
killed before writing. Number("") is 0, and process.kill(0, 0) signals the
test's own process group and succeeds.
- The fake Docker child publishes its PID by rename, and tests read it before
the abort with a bounded, named start-up wait that rejects PIDs 0 and 1.
- Both cancellation tests now observe the spawned child and assert at refusal
time that it already exited, was SIGKILLed and is gone.
- The cancellation test always aborts and awaits its acquisition in finally, so
a failed start-up wait cannot reject into a later test.
- The file's acquisitions use a named 45 s fixture probe deadline through
acquireFixtureBinding, unless a test names its own deadline.
- New tests drive the adopted command's refusal and cancel throw sites and
assert the diagnostics' closed tokens.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…HACK-1452) Under heavy load the moving job snapshot case's 20 s ordered operation refused with "did not become ready", and the 3 s shared-deadline case ended before its first bridge read. Name an ordered operation deadline of four fixture probe deadlines, and give both exhausted-deadline cases a 15 s probe deadline with a named 5 s refusal slack. They still require repeated scans and a refusal soon after the deadline. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ed freshness checks (HACK-1452) The moving job snapshot case's ordered start makes 20 assertFresh calls, not about four, so its 180 s aggregate deadline refused with "did not become ready" at a load average of about 115 and above. Count the checks, size the deadline at one fixture probe deadline each, assert the start makes no more, and give the case its own test budget above that deadline. Correct the file budget's comment, which named this case's probes as the heaviest test's. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…get (HACK-1452) The case makes 3 probes before its ordered start and 2 after it, so the old 5-probe allowance left no room for file reads and poll sleeps. Co-Authored-By: Claude Opus 5.5 <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
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.
Integration gate for merge batch b64: origin/next + #495 (HACK-1452). Tree 89b0f94a. Closed after the batch squashes.
🤖 Generated with Claude Code
Need help on this PR? Tag
@codesmith-botwith what you need.