Rollout and Protect
Exam-pace drill: stand up a probed, rolling Deployment; ship a new version zero-downtime while scaling; then front it with a Service and protect it with a PodDisruptionBudget.
Your mission
This is a drill, not a tutorial. It's paced and scored like the real CKAD:
each task fuses several skills, the specs are exact, and hints cost real points.
Work in the prod namespace (already created). The bar is to finish without
opening a hint.
You run web through its life on a busy cluster: stand it up with a rolling
strategy and real health probes, then ship a new version while scaling and
without dropping a request, and finally put a Service in front of it and a
disruption budget behind it so a node drain can't take it down. Every field is
graded — a missed maxUnavailable or a probe on the wrong port is a missed mark.
Save typing
kubectl config set-context --current --namespace=prodk is aliased to kubectl with completion on.