- Create a namespace.
- Deploy an Argo Rollout with a blue/green strategy.
- Verify the application’s /health endpoint — suitable for use in an AnalysisTemplate and AnalysisRun to gate promotion.
rollout-initial.yaml) runs five replicas of the “highway-animation” app (blue variant), sets a POD_COUNT environment variable, and configures blue/green with manual promotion (autoPromotionEnabled: false) so promotion can be done manually or via an analysis.
argo-analysis-lab namespace to confirm the Rollout and blue/green details.

/health endpoint that returns a simple JSON payload:
This
/health endpoint is a good target for an Argo Rollouts AnalysisTemplate. An AnalysisRun can query the endpoint and assert that "status" == "OK" before promoting the preview to active.The Rollout is configured with
autoPromotionEnabled: false. Promotion must be performed manually via the Argo Rollouts UI/CLI or gated using an AnalysisRun. Also be cautious exposing NodePorts in production clusters — prefer Ingress or LoadBalancer for controlled external access.| Resource Type | Purpose |
|---|---|
| Rollout | Manages blue/green deployment of highway-bluegreen with 5 replicas and manual promotion. |
| Service (active) | NodePort service routed to the active replica set (port 32079). |
| Service (preview) | NodePort service routed to the preview replica set (port 31058). |
- Argo Rollouts blue/green traffic management: https://argoproj.github.io/argo-rollouts/features/traffic-management/#blue-green
- Argo Rollouts analysis feature (AnalysisTemplate & AnalysisRun): https://argoproj.github.io/argo-rollouts/features/analysis/
- Kubernetes documentation: https://kubernetes.io/docs/
/health, run an AnalysisRun during promotion, and use its result to gate the transition from preview to active.