fix: remove revision==1 gate for scaleDown onsuccess. Fixes #4681

[SimbaKingjoe]

Summary

Remove the revision == 1 gate from promoteStable() so workloadRef.scaleDown: onsuccess triggers on every stable promotion, not just the initial deploy.

Background

When using BlueGreen strategy with workloadRef.scaleDown: onsuccess, the referenced Deployment should be scaled down every time a new ReplicaSet is promoted to stable. However, the check in promoteStable() (rollout/sync.go:1040) included revision == 1, restricting the scale-down to only the very first deployment.

For revision >= 2, shouldFullPromote() correctly runs PostPromotionAnalysis (if configured), returns "Completed blue-green update", and calls promoteStable(), but the scale-down is silently skipped because revision != 1.

Modifications

• rollout/sync.go: Remove revision == 1 && from the scaleDown condition in promoteStable()
• rollout/sync_test.go: Add TestScaleDownDeploymentOnSuccessRevision2 verifying scale-down triggers on revision 2

Verification

go test ./rollout/ -run 'TestScaleDownDeploymentOnSuccess' -v
# PASS: TestScaleDownDeploymentOnSuccess (revision 1)
# PASS: TestScaleDownDeploymentOnSuccessRevision2 (revision 2)

Checklist:

• This is a bug fix
• The title follows conventional commits: fix: remove revision==1 gate for scaleDown onsuccess
• DCO sign-off included
• Unit tests written and passing
• My organization is added to USERS.md (N/A - bug fix)

[sonarqubecloud]

Quality Gate failed

Failed conditions
55.0% Duplication on New Code (required ≤ 40%)

See analysis details on SonarQube Cloud

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 85.07%. Comparing base (2ccdae8) to head (21efebe).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4730      +/-   ##
==========================================
+ Coverage   85.03%   85.07%   +0.03%     
==========================================
  Files         164      164              
  Lines       18989    18989              
==========================================
+ Hits        16148    16154       +6     
+ Misses       1993     1990       -3     
+ Partials      848      845       -3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

Published E2E Test Results

  4 files    4 suites   3h 46m 30s ⏱️
120 tests 107 ✅  7 💤  6 ❌
490 runs  452 ✅ 28 💤 10 ❌

For more details on these failures, see this check.

Results for commit 21efebe.

[github-actions]

Published Unit Test Results

2 470 tests   2 470 ✅  3m 22s ⏱️
  129 suites      0 💤
    1 files        0 ❌

Results for commit 21efebe.

[Hariharasuthan99]

Hello @SimbaKingjoe , scale down on success is expected to be used only during the initial migration from deployment to rollout and so the reason behind hardcoded value of 1. This change would not only try to trigger scale down even after the initial migration it also does not fix the core issue mentioned in #4681 i.e Initial deploy bypasses PostPromotionAnalysis, creating a permanent dead state. Could you please recheck?