Problem Statement
What is a Rolling update deployment?
Explanation
Rolling update gradually replaces old version instances with new version in batches while maintaining availability. Update small batch (e.g., 2 servers), wait for health checks, proceed to next batch, repeat until all updated. Load balancer removes unhealthy instances from rotation ensuring zero downtime. If update fails, stop rollout and rollback.
Process: update batch 1 (2 servers), health check passes, update batch 2, continue until complete. Batch size determined by: total instances, acceptable capacity reduction, risk tolerance. Smaller batches slower but safer. Benefits: zero downtime, gradual deployment, no additional infrastructure, automatic via orchestrators (Kubernetes).
Kubernetes rolling update: RollingUpdate strategy with maxSurge (extra instances during update) and maxUnavailable (instances down during update). Use cases: stateless applications, microservices, containerized applications. Understanding rolling updates enables gradual zero-downtime deployments.
