test: race between disabling interception and network request in chromium

[Dmitry Gozman (dgozman)]

References #41802.

[github-actions]

Test results for "tests 1"

2 failed
❌ [playwright-test] › reporter.spec.ts:251 › merged › should not have internal error when steps are finished after timeout @windows-latest-node22
❌ [playwright-test] › ui-mode-trace.spec.ts:827 › should update state on subsequent run @windows-latest-node22

4 flaky

⚠️ [playwright-test] › ui-mode-test-update.spec.ts:202 › should update test locations `@ubuntu-latest-node22`
⚠️ [playwright-test] › ui-mode-test-watch.spec.ts:185 › should watch new file `@windows-latest-node22`
⚠️ [playwright-test] › ui-mode-trace.spec.ts:189 › should show snapshots for steps `@windows-latest-node22`
⚠️ [playwright-test] › ui-mode-trace.spec.ts:715 › should indicate current test status `@windows-latest-node22`

51123 passed, 1220 skipped

---

Merge workflow run.

[github-actions]

Hi, I'm the Playwright bot and I took a first look at the CI failures.

🟢 Both failures are pre-existing flakes — this PR is clear

The two reds are windows-latest-node22 [playwright-test] tests. This PR only adds one test to tests/page/page-route.spec.ts and touches no product code, so it can't reach the reporter or UI-mode suites. Both tests have a failure history on unrelated SHAs.

Details

Pre-existing flake / infra

• [playwright-test] › reporter.spec.ts:251 › merged › should not have internal error when steps are finished after timeout — pre-existing Windows flake. Across the test-results DB it times out on the first attempt on windows-latest-node22 on runs the PR can't be responsible for: PR test: test for issue 41348 #41930 (run 29911466278), a main push (run 29940740435), PR docs(locator): drop type markup from ariaSnapshotJSON details #42136 (run 31179887339), and PR feat(chromium): roll to r1240 #42246 (run 31748384334). Final verdict flips green after retry in nearly every run (0 failed of 1031 final verdicts). It's a slow-Windows timeout, not a real failure.

• [playwright-test] › ui-mode-trace.spec.ts:827 › should update state on subsequent run — pre-existing cross-run flake: failed 9 of 1028 runs (~1%) on SHAs unrelated to this PR, windows-latest-node22 again. UI-mode trace state isn't exercised by a page-route test.

The 4 flaky entries in the report (ui-mode-*) are the same story — UI-mode/watch tests wobbling on Windows, untouched by this diff.

_{Triaged by the Playwright bot - agent run}

[github-actions]

Test results for "MCP"

5 failed
❌ [firefox] › mcp/annotate.spec.ts:316 › should annotate via direct browser_annotate MCP call @mcp-windows-latest-firefox
❌ [firefox] › mcp/annotate.spec.ts:349 › should annotate when context has no fixed viewport @mcp-windows-latest-firefox
❌ [firefox] › mcp/cli-core.spec.ts:57 › click link @mcp-windows-latest-firefox
❌ [firefox] › mcp/cli-core.spec.ts:140 › uncheck @mcp-windows-latest-firefox
❌ [firefox] › mcp/cli-save-as.spec.ts:26 › screenshot @mcp-windows-latest-firefox

8096 passed, 1311 skipped

---

Merge workflow run.

[github-actions]

Hi, I'm the Playwright bot and I took a look at the CI failures.

🟢 All five failures are pre-existing Windows/Firefox flakes — this PR is clear

The five reds are all on mcp-windows-latest-firefox. This PR only adds one test to tests/page/page-route.spec.ts (gated fixme on Chromium) and touches no product code, so it can't reach the MCP suite. Every one of these tests has a failure history on mcp-windows-latest-firefox across unrelated PRs, and none of them ever fails on the ubuntu or macos Firefox bots.

Details

Pre-existing flake / infra

• [firefox] › mcp/cli-core.spec.ts:57 › click link — pre-existing Windows-only flake: failed 20 of 540 runs (~4%) on mcp-windows-latest-firefox, across 15+ unrelated PRs (fix(types): support branded primitives in evaluate arguments #42009, fix(mcp): do not run heartbeat for clients without the event stream #42189, fix(trace): record failed request durations, print request start times #42191, …). Passed 520/540; 0 failures on ubuntu (719 runs) and macos (713 runs).
• [firefox] › mcp/cli-core.spec.ts:140 › uncheck — pre-existing Windows-only flake: failed 5 of 540 runs on mcp-windows-latest-firefox (feat(trace): add action screenshot and aria snapshot events #42167, feat(test-runner): annotate serial suites for custom sharding #42164, feat(mcp): default codegen language from PW_LANG_NAME #42150, …); 0/719 ubuntu, 0/713 macos.
• [firefox] › mcp/annotate.spec.ts:316 › should annotate via direct browser_annotate MCP call — Windows-only flake: failed 2 of 540 (chore: mark v1.62.1 #42020, Revert "chore(deps-dev): bump fast-uri from 3.1.3 to 3.1.4" #41978); clean on ubuntu/macos.
• [firefox] › mcp/annotate.spec.ts:349 › should annotate when context has no fixed viewport — Windows-only flake: failed 3 of 540 (a main push, Revert "chore(deps-dev): bump fast-uri from 3.1.3 to 3.1.4" #41978, fix(mcp): bump firefox action timeout for flaky dialog cli tests #42088); clean on ubuntu/macos.
• [firefox] › mcp/cli-save-as.spec.ts:26 › screenshot <ref> — Windows-only flake: failed 2 of 540 (feat(webkit): roll to r2341 #42061, fix(mcp): do not run heartbeat for clients without the event stream #42189); clean on ubuntu/macos.

The diff is a single new page-route test with it.fixme(browserName === 'chromium', ...); it exercises page.route(...).abort() on Chromium and has nothing to do with the MCP CLI/annotate flows on Firefox.

_{Triaged by the Playwright bot.}

_{Triaged by the Playwright bot - agent run}