Repository navigation
Bug: quickfiler-tests-depend-on-wall-clock-timing #950
Description
- Work Mode: full-bug
Summary
Several QuickFiler.Test tests fail intermittently under load because they wait on real wall-clock time. Seen during parallel run bugs-2026-09-28:
QfcDatamodelLivenessTests, includingRemainingLoadActive_AcrossAsyncVoidFirstAwait_StaysTrueWhileLoaderProduces. Two failures stopped Bug: engine-toggle-prime-marker-registration-races-removal #944's P3-T8 gate, and one failure occurred during Bug: package-manifest-consistency-residuals #929's local run.QfcItemController.UiThreadDispatcherFixtureTests.Transaction_SecondCallerCannotInstallUntilTheFirstRestores, which failed once in CI on an earlier Bug: package-manifest-consistency-residuals #929 head.
Environment
- OS/version: Windows 11 (local, with concurrent coverage runs) and windows-latest (CI)
- Python version: n/a (C# / MSTest, parallel Workers=0, Scope=ClassLevel)
- Command/flags used:
Invoke-MSTestWithCoverage.ps1, and the CI MSTest-with-coverage required check - Data source or fixture: n/a
Steps to Reproduce
- Run QuickFiler.Test under the parallel regime while the machine is loaded, for example with a second coverage run.
- Observe intermittent failures in the tests named above.
Expected Behavior
The tests are deterministic regardless of machine load.
Actual Behavior
QuickFiler.Test/Controllers/QfcDatamodelLivenessTests.csusesSpinWait.SpinUntil(condition, TimeSpan.FromSeconds(5))(line ~56) and.Task.Wait(TimeSpan.FromSeconds(5))(lines ~103 and ~173). A slow scheduler turns these into assertion failures.QuickFiler.Test/Controllers/QfcItemController.UiThreadDispatcherFixtureTests.cs:206fails intermittently. Its cause is not yet established.
Logs / Screenshots
- Attached minimal logs or screenshot
- Snippet: Bug: engine-toggle-prime-marker-registration-races-removal #944 evidence for P3-T8, first run; Bug: package-manifest-consistency-residuals #929 PR fix(929): remove altcover imports, correct SVGControl binding redirects, pass client-id to the repair workflow token step #949 CI history.
Impact / Severity
- Blocker
- High
- Medium
- Low
Source
From: docs/features/potential/2026-09-30-quickfiler-tests-depend-on-wall-clock-timing.md
Activity
drmoisan commentedon Sep 30, 2026
Restoring sections dropped by the promotion tooling.
Suspected Cause / Notes
Real-time bounded waits violate the repository determinism rules: no wall-clock waits in tests, and use of
FakeTimeProvideror a controllable scheduler.QfcDatamodelLivenessTests.csusesSpinWait.SpinUntil(..., TimeSpan.FromSeconds(5))(line ~56) and.Task.Wait(TimeSpan.FromSeconds(5))(lines ~103 and ~173).QfcItemController.UiThreadDispatcherFixtureTests.cs:206) may be a different root cause. Triage it first, and split it into its own issue if its cause is not timing.This hits a required check. #944 needed a one-time gate re-run because of it.
Proposed Fix / Validation Ideas
TaskCompletionSourceor awaited handles) or an injectedTimeProvider.[DoNotParallelize], Workers=1 or longer timeouts, and find any raced static state.drmoisan commentedon Oct 1, 2026
Scope note from the #950 preparation research.
The same bounded wall-clock wait pattern also exists in:
QuickFiler.Test/Controllers/QfcDatamodelTeardownTests.csQuickFiler.Test/Controllers/QfcInitEmailQueueZeroBatchTests.csBoth waits come from the same cause:
QfcDatamodel.InitEmailQueuestarts its worker on a thread pool the test cannot control. Fix them with the same seam #950 introduces. They can be included in #950's scope, or taken on in a follow-up branch immediately after #950 merges.The research also found that the transaction test failure (
Transaction_SecondCallerCannotInstallUntilTheFirstRestores) has a different cause:UiThreadDispatcherFixture.EnsureDispatcher()replaces the shared dispatcher without taking the lock that guards it, whileQfcItemController_FocusAndThemeTestsruns in parallel. The proposed fix changes only test code and stays within #950.