[teamleaderleo]

ClassicUpgradeSimulationTests.theFirstLaunchBringsEverythingOnce("multiWindow") fails intermittently on base and on unrelated PRs, most recently #28283's daemon tests. The workspaces show in swapped order at ClassicUpgradeSimulationTests.swift:45 and :52, for example multiWindow: order differs: ["w1 0", "w1 2", "w1 1", "w1 3"] for ["w1 0", "w1 1", "w1 2", "w1 3"].

Tracking: #28425. Cause: the sidebar placements a run queues through PersonalPlacementQueue (#26153) trail it on a loaded runner. A late slot placement undoes the group the run set, and the simulator reads the order before the queue drains. The fix is #28362.

Change

Only that case's verify runs inside withKnownIssue(..., isIntermittent: true), which names #28425. The other seven matrix cases, the other tests and the rest of that case's expectations are unchanged. #28362 removes this quarantine.

No capture: test-only change.

Risk: low. One test case's order and group checks become a known issue until the fix lands.

Agent session: agent session

🤖 Generated with Claude Code

---

^{Need help on this PR? Tag @codesmith-bot with what you need.}

---

Summary by cubic

Quarantines the intermittently failing multiWindow case of ClassicUpgradeSimulationTests.theFirstLaunchBringsEverythingOnce as a known issue (#28425) until the fix in #28362 lands.

The test flakes because sidebar placements a run queues trail it on a loaded runner, showing workspaces in swapped order. Only that case's verify runs inside withKnownIssue(..., isIntermittent: true); the other seven matrix cases and all other expectations are unchanged. Test-only change with low risk.

^{Written for commit f999387. Summary will update on new commits.}

[github-actions]

cmux status _{(one comment, edited in place; each line is updated by the job that owns it)}

Review f999387: HQ feed card · gallery before/after runs · checks

App of this head (app.zip, kept 1 day)

gh workflow run reload-build.yml -R manaflow-ai/cmux --ref feat-cmux-next -f tag=pr28430 -f ref=lane/classic-multiwindow-quarantine -f platform=macos -f backend_mode=local

[blacksmith-sh]

Found 2 test failures on Blacksmith runners:

Failures

Test	View Logs
test_every_listed_hook_is_one_glaeda_released_with_the_check (main.SigningHooksInSt epWithGlaedaTests.test_every_listed_hook_is_one_glaeda_released_with_the_check)/ test_every_listed_hook_is_one_glaeda_released_with_the_check (main.SigningHooksInSt epWithGlaedaTests.test_every_listed_hook_is_one_glaeda_released_with_the_check)	View Logs
test_the_repository_highlights_validate (main.DigestTests.test_the_repository_highl ights_validate)/ test_the_repository_highlights_validate (main.DigestTests.test_the_repository_highl ights_validate)	View Logs

^{Need help on this PR? Tag @codesmith-bot with what you need.}

[github-actions]

Shipped in:

• cmux NEXT nightly 1.0.0-nightly.4110084161401 (92b6fe2118)
• cmux NEXT nightly 1.0.0-nightly.4110335000301 (abe139dc5a)
• cmux NEXT nightly 1.0.0-nightly.4110516029701 (b7196082e9)
• cmux NEXT nightly 1.0.0-nightly.4110699040101 (723153fb7a)