Failed after Nnx-cloud[bot]
View latest

fix(protect): update-signals-sdk [Self-Healing CI Rerun]

863e47c448Open on GitHub

Self-Healing CI

Classification: Flaky task

We looked at the webauthn-device e2e failure, where the 'Logout' button didn't appear within 5s after WebAuthn authentication. Our changes only update the Protect signals SDK bundle and regenerate API reports, none of which touch the WebAuthn flow. The task has a non-zero flakiness rate and no matching failure on other branches, so we are treating it as flaky and a rerun is appropriate.

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

This was not classified as a code change, for these reasons:

  1. The failure is a Playwright timeout.

    • After the authenticate step was submitted, getByRole('button', { name: 'Logout' }) was not visible within 5000ms (webauthn-device.test.ts:100).
    • The other test in the file passed.
    • The registration steps earlier in the failing test also succeeded.
  2. The PR changes do not relate to the failing flow.

    • The PR adds a changeset, regenerates api-report markdown files, and replaces the bundled packages/protect/src/lib/signals-sdk.js.
    • None of these touch the WebAuthn register/authenticate journey.
    • The failing project (@forgerock/journey-suites) is not in touched_projects.
  3. The evidence points to flakiness.

    • The task has a flakiness rate of about 1.4%, so it is not deterministic.
    • No similar failure was found on other branches.
    • The symptom is a timing issue in the live browser and virtual authenticator flow.

A rerun is appropriate. No code changes are proposed.