32. Observability: metrics, logs, and traces

Collect Kubernetes and application metrics, logs, traces, events, and audit records. Build dashboards and alerts around workload health, control-plane health, saturation, latency, errors, and service objectives.