We classified both e2e failures as environment_state rather than code changes. Neither tanstack-react-start-e2e-rsc nor tanstack-react-start-e2e-virtual-routes are in the set of projects touched by this PR, and the errors — a port-already-in-use conflict and a test-harness SyntaxError during dummy-server startup — have no traceable connection to the lifecycle-callback changes introduced here. Our recommendation is to re-run the affected tasks once the environment is clean.
Failed after
SSheraffView latestfix(router-core): keep lifecycle metadata declaration compatible with Playwright
81654c3bfe9Open on GitHubSelf-Healing CI
Classification: Environment state
An empty commit was applied to the branch to trigger a new CI pipeline execution to resolve the environment issue.
Two e2e tasks failed, neither of which belongs to a project touched by this PR.
Task 1 — tanstack-react-start-e2e-rsc:test:e2e--vite-ssr--shard-2-of-6 Error: SyntaxError: Unexpected end of input
The error surfaces immediately after the local dummy server is started, before any Playwright test has executed. This is a test-harness initialization failure, not a test assertion failure. The project tanstack-react-start-e2e-rsc does not appear in the list of projects touched by the PR. The PR exclusively modifies lifecycle-callback dispatch logic in @tanstack/router-core and @tanstack/react-router — there is no code path from those changes to a JavaScript parse error in the test runner setup.
Task 2 — tanstack-react-start-e2e-virtual-routes:test:e2e Error: Error: listen EADDRINUSE: address already in use 127.0.0.1:45415
Port 45415, mapped to the external dummy server fixture, was already bound by another process when the test tried to start. This is a straightforward infrastructure/environment conflict. The project tanstack-react-start-e2e-virtual-routes is also not in the touched-projects list, and the PR introduces no port configuration changes.
Why not code_change:
- Neither failing project is in
touched_projects. - Neither error message references any symbol, function, or import modified by the PR diff.
- Removing the PR's changes would not logically resolve a pre-occupied port or a parse error in the test runner harness.
Why not flaky_task:
- Both tasks have a recorded flakiness rate of 0%, which rules out this classification per the decision rules.
Conclusion: Both failures are environmental — one is a port conflict caused by a stale or concurrently running process, and the other is an infrastructure-level parse error in the test setup. Neither is attributable to the code changes in this PR.