router
Failed after Lladybluenotes
View latest

Merge main into stable-serverfn-ids-2

76937e9f5fbOpen on GitHub

Self-Healing CI

Classification: Flaky task

We classified this failure as flaky rather than a code change because the failing project (tanstack-vue-start-e2e-start-manifest) is not touched by this PR, which is scoped entirely to server function ID logic. The test's CSS assertion failure — where the stylesheet was fetched successfully but getComputedStyle returned empty — is a known timing race with an 11.5% historical flakiness rate, unrelated to our changes.

An empty commit was automatically applied to the branch to trigger a new CI pipeline execution to resolve the flaky task.

The failing task tanstack-vue-start-e2e-start-manifest:test:e2e was classified as a flaky task rather than a code change for the following reasons:

  1. Unrelated project: The project tanstack-vue-start-e2e-start-manifest is not present in the list of touched projects (root, tanstack-react-start-e2e-server-functions, @tanstack/start-client-core, @tanstack/start-plugin-core). The PR exclusively modifies server function ID logic and associated tests — none of which are in the Vue start manifest e2e project.

  2. Unrelated failure domain: The PR adds stable IDs for server functions (compiler transforms, ID reservation, deduplication). The failing test verifies that a shared widget's CSS border color (rgb(249, 115, 22)) persists after client-side navigation from a lazy-loaded route to a static route. These are entirely separate concerns.

  3. Timing/race condition signature: The CSS file SharedWidget-BAtMnCRs.css was successfully fetched (HTTP 200) during the test run, but getComputedStyle returned an empty string at assertion time. This is a well-known pattern where the browser has not yet applied loaded styles by the time the assertion executes — a timing race within the test itself.

  4. Established flakiness history: The task has a recorded flakiness rate of ~11.5%, confirming this test has a pre-existing history of non-deterministic failures unrelated to any specific code change.

  5. No cross-branch signal: No similar failures were found in any other branch, ruling out a regression introduced by a shared infrastructure change.