Fix flaky E2E test in specs/modules/tagmanager/setup.test.js.

[techanvil]

Feature Description

We still have a number of E2E tests failing with some regularity in our CI test runs. The task of fixing them has been split into four issues in order to make it easier to investigate and fix. See #10108, #10109, #10110.

Tag Manager module setup › Setup with AMP active › with Secondary AMP › when validating › validates homepage AMP for non-logged-in users

Example: https://github.com/google/site-kit-wp/actions/runs/12871930956/job/35886293222#step:11:245

Details

FAIL specs/modules/tagmanager/setup.test.js (47.115 s)
  ● Tag Manager module setup › Setup with AMP active › with Secondary AMP › when validating › validates homepage AMP for non-logged-in users

    TimeoutError: Navigation timeout of 5000 ms exceeded

      at ../../node_modules/puppeteer/src/common/LifecycleWatcher.ts:204:18

---

Do not alter or remove anything below. The following sections will be managed by moderators only.

Acceptance criteria

• The test listed in the Feature Description should pass every time in CI.

Implementation Brief

Looking at the failures of these runs the other tests in this file pass and this failing test is failing with a call to fetch page and the AMP validator. Draft PR here where I have rerun the E2Es multiple times without the issue recurring.

• Update allowed timeout for toHaveValidAMPForUser and toHaveValidAMPForVisitor to 10000 and test repeatidly to confirm the issue is resolve, otherwise increase the timeout until the issue no longer occurs over multiple CR runs:

  • site-kit-wp/tests/e2e/matchers/to-have-valid-amp-for-visitor.js

    Line 48 in 9eb1215

    const html = await fetchPageContent( urlToFetch, { credentials: 'omit' } );

Test Coverage

• N/A

QA Brief

• QA:Eng, validate results of CR/MR.

Changelog entry

• N/A

[eugene-manuilov]

IB ✔

[zutigrm]

QA:Eng ✅