---

name: k8s-deploy-workflow
description: Production Kubernetes deployment pipeline

---

Kubernetes Deployment Workflow

Ship to production with confidence.

Pipeline Stages

1. Build → Docker multi-stage build (optimized layers)
2. Test → Unit + integration in CI
3. Scan → Trivy vulnerability scanning
4. Push → Container registry (ECR/GCR/GHCR)
5. Deploy → Helm upgrade with canary strategy
6. Verify → Health checks + smoke tests
7. Monitor → Prometheus alerts + Grafana dashboards

Files Included

.github/workflows/deploy.yml    # GitHub Actions pipeline
Dockerfile                       # Multi-stage optimized build
helm/
  Chart.yaml                     # Helm chart definition
  values.yaml                    # Default values
  values-prod.yaml               # Production overrides
  templates/
    deployment.yaml              # Pod spec with probes
    service.yaml                 # ClusterIP service
    ingress.yaml                 # Nginx ingress with TLS
    hpa.yaml                     # Horizontal Pod Autoscaler
    pdb.yaml                     # Pod Disruption Budget
monitoring/
  alerts.yaml                    # Prometheus alert rules
  dashboard.json                 # Grafana dashboard

Auto-Scaling Config

• Min replicas: 2
• Max replicas: 20
• CPU target: 70%
• Memory target: 80%
• Scale-up cooldown: 60s
• Scale-down cooldown: 300s

Rollback Strategy

Automatic rollback triggers:

• Error rate > 5% for 2 minutes
• P99 latency > 2s for 5 minutes
• Pod crash loop detected
• Health check failures > 3