router

Failed after

SSheraff
View latest

test(react-router): cover context during pending and not-found loads

8243e1ff33dOpen on GitHub

Self-Healing CI

Classification: Environment state

We classified this failure as an environment issue rather than a code change because the failing task (tanstack-vue-start-e2e-basic:test:e2e--rsbuild-ssr) belongs to a Vue Start e2e project that was not touched by this PR, which only adds React Router unit tests. The root cause is a port conflict (EADDRINUSE :::37965) caused by a stale process in the CI runner, unrelated to any of our changes — a clean rerun should resolve it.

An empty commit was applied to the branch to trigger a new CI pipeline execution to resolve the environment issue.

The failing task tanstack-vue-start-e2e-basic:test:e2e--rsbuild-ssr was not classified as a code change for the following reasons:

  1. Unrelated project: The PR exclusively adds new test files under packages/react-router/tests/. The failing task belongs to the tanstack-vue-start-e2e-basic project — a Vue Start e2e test suite that shares no code path with the React Router changes introduced here.

  2. Empty touched_projects: The touched_projects list is empty ([]), confirming that no files belonging to tanstack-vue-start-e2e-basic were modified by this PR.

  3. Port conflict error, not a code error: The failure is caused by EADDRINUSE: address already in use :::37965. The test web server could not bind to port 37965 because another process in the CI environment was already occupying it. This is an infrastructure-level problem, not a consequence of any code change.

  4. Zero flakiness rate: The task has a recorded flakiness rate of 0%, which rules out a non-deterministic race condition within the test itself (which would otherwise suggest flaky_task).

  5. Not present in the base branch: The similar task failure detector confirmed that the same task does not fail in branch "8165" (the branch this PR stacks on), so this is not a regression introduced by an ancestor branch either.

All evidence points to the CI runner having a stale or conflicting process holding port 37965 at the time this job executed — an environment state problem that resolves itself on a clean rerun.