Glossary · Phase 11 · Professional practice
Readiness probe
A check Kubernetes runs on each pod; until it passes, the pod receives no traffic, which keeps rollouts from sending users to half-started pods.
Try it now
replicas 3, maxSurge 1, maxUnavailable 0: never fewer than 3 ready pods.
Learn it properly
Kubernetes basics · Desired state, Deployments and Services, labels, probes and rolling updates. About 40 minutes.
Other terms from the same lesson
- kubectl: The command-line tool for talking to a Kubernetes cluster: applying manifests, listing pods, watching rollouts and reading logs.
- Kubernetes: A system that runs containers across a cluster of machines, keeping the running state matching the state you declare.
- label selector: A set of key-value pairs that picks every Kubernetes object carrying all of those labels, which is how Deployments and Services find pods.
- liveness probe: A check Kubernetes runs on each pod and restarts the pod when it fails, for apps that hang without crashing.
- pod: The smallest thing Kubernetes runs: one or more containers scheduled together on one machine.
- reconciliation: A controller repeatedly comparing the desired state with what's actually running, and acting to close the gap.
- rolling update: Replacing pods with a new version a few at a time, within surge and unavailability limits, so the service never goes down.
Never trained an AI?
Teach a small one to write, right in your browser. It takes about 20 seconds and needs no account.