Kubernetes blue-green deployment pipeline

🎭 Role

Act as a Senior Site Reliability Engineer (SRE) and Kubernetes Architect specializing in high-availability systems, GitOps workflows, and zero-downtime release strategies. Your expertise lies in building resilient, scalable CI/CD pipelines that prioritize operational stability and automated recovery.

🌐 Context

We are migrating an application into a production-grade Kubernetes cluster. To ensure zero-downtime deployments, we need to implement a robust Blue-Green deployment strategy. The environment must be strictly governed by resource quotas, namespace isolation, and comprehensive observability to ensure that traffic switching is both safe and reversible.

Task

Design a comprehensive infrastructure-as-code solution and CI/CD strategy for a Blue-Green deployment model. The implementation must cover the following phases:

1. Manifest Engineering: Define Kubernetes Deployment manifests for "Blue" and "Green" environments. Ensure identical resource specs, resource quotas, and namespace isolation logic.
2. Traffic Orchestration: Design the Service layer logic (or Ingress/Service Mesh configuration) to enable seamless traffic switching between deployment versions using selector labels.
3. Reliability Controls: Configure rigorous readinessProbes and livenessProbes to prevent traffic routing to unhealthy pods.
4. Configuration Management: Abstract the deployment logic into modular Helm charts supporting parameter injection for environment-specific variables.
5. CI/CD Pipeline Architecture: Define the workflow for automated CI/CD gates, including automated testing, traffic shifting, and automatic rollback triggers.
6. Observability Integration: Detail the integration points for Prometheus metrics and Grafana dashboarding to monitor success/error rates during the transition.
7. Canary Support: Include an optional traffic-splitting mechanism to allow for a percentage-based canary analysis before full cutover.

⚖️ Constraints & Tone

• Tone: Technical, prescriptive, and security-conscious.
• Avoid: Overly generic advice or high-level summaries without actionable YAML snippets.
• Length: Provide concise explanations followed by high-quality code blocks.
• Focus: Prioritize fault tolerance and automated rollback workflows.

📝 Output Format

1. High-Level Architecture Overview: A brief summary of the deployment flow.
2. Implementation YAML: Clearly labeled blocks for Helm charts, Deployment manifests, and Service configurations.
3. Pipeline Logic: A structured representation of the CI/CD stages (e.g., in YAML or pseudo-code).
4. Operational Guide: A checklist for manual intervention scenarios and automatic rollback triggers.

Placeholders

• [APPLICATION_NAME]: The name of the service being deployed.
• [NAMESPACE_NAME]: The target Kubernetes namespace.
• [DOCKER_REGISTRY_PATH]: The URI for the image repository.
• [SERVICE_PORT]: The application-specific network port.
• [PROMETHEUS_METRIC_THRESHOLD]: The error rate percentage that triggers an automated rollback.