kubernetes-deployment-patterns

Zero-downtime Kubernetes deployments use rolling updates, blue-green, or canary strategies to shift traffic gradually while keeping services available. Rolling updates replace pods incrementally; blue-green maintains two parallel environments; canary routes a percentage of traffic to new versions before full promotion, enabling safe rollbacks if issues arise.

Rolling updates gradually replace old pods with new ones. Blue-green maintains two identical production environments and switches traffic between them instantly. Canary releases route a small traffic percentage to the new version first, monitoring metrics before expanding to all users. Each trades deployment speed, resource overhead, and rollback complexity differently.

Deployment handles stateless applications with rolling updates; StatefulSet manages stateful services requiring stable identity and ordered scaling; DaemonSet runs one pod per node for logging or monitoring; Job executes batch tasks to completion. Match the workload type to whether your app maintains state and requires ordered or distributed execution.

Yes. Configure liveness and readiness probes to ensure only healthy pods receive traffic, then pair them with Horizontal Pod Autoscaler (HPA) or Vertical Pod Autoscaler (VPA) to scale based on metrics. Health checks prevent traffic to failing pods; autoscaling adjusts replica counts or resource requests based on demand, enabling safe high-availability deployments.

Kubernetes tracks deployment revisions automatically. Use `kubectl rollout history` to view past versions, then `kubectl rollout undo` to revert to the previous revision or specify an exact revision number. Define rollback criteria upfront—monitor metrics like error rates or latency during canary phases—and trigger rollback before traffic reaches all replicas.

Use ConfigMaps to store non-sensitive configuration data and Secrets for credentials, API keys, and certificates. Mount both as volumes or inject as environment variables into pods. Separate configuration from code, rotate secrets regularly, and use RBAC to limit access. This enables safe, version-controlled, environment-specific deployments.