CertKata

Kubernetes and Cloud Native Associate — Free Essentials

Key facts distilled from CertKata's KCNA question bank. Free to read — no account needed.

Control plane vs. worker nodes

Kubernetes control plane and worker node components, connected only through the API serverCONTROL PLANEetcdschedulercontrollermanagerapi-servervia API serverWORKER NODEkubeletkube-proxycontainerruntimepodsAll control-plane ↔ node traffic goes through the api-server

The control plane manages cluster state; worker nodes run workloads. Control plane: kube-apiserver (the front door — the only component that reads/writes etcd directly), etcd (consistent key-value store holding all cluster state), kube-scheduler (assigns Pods to nodes), kube-controller-manager (runs control loops). Worker node: kubelet (node agent that runs Pods), kube-proxy (implements Service networking), and a container runtime (e.g. containerd).

Workload controllers hierarchy

Kubernetes workload containment: a Deployment contains ReplicaSets, which contain Pods, which contain one or more containersDEPLOYMENTREPLICASETPODContainerPODContainerContainerDeployment → ReplicaSet → Pod → Container(s)nesting = ownership, not data flow

Deployment manages a ReplicaSet, which maintains a set of identical Pods. Deployment adds declarative rolling updates and rollback on top of ReplicaSets. StatefulSet: stable identity + per-Pod persistent storage (Pods named app-0, app-1...). DaemonSet: one Pod per (selected) node — used for node agents like log collectors. Job: run-to-completion; CronJob: scheduled Jobs.

Containers vs. VMs; OCI

Containers share the host OS kernel (lightweight); VMs virtualize hardware and each run a full guest OS. The Open Container Initiative (OCI) standardizes image and runtime formats. Container images are immutable, built from stacked read-only layers, and distributed via registries.

Three pillars: metrics, logs, traces

Metrics (numeric time series, e.g. Prometheus), logs (event records — prefer structured/JSON), and traces (a request followed across services). Monitoring watches for known problems; observability lets you investigate unknown ones from collected telemetry.

CI / CD / Continuous Deployment

CI (Continuous Integration): frequently merge changes with automated build + test. Continuous Delivery: keep code always releasable, release is a deliberate step. Continuous Deployment: every passing change auto-releases to production with no manual gate. Prefer specific/immutable image tags (or digests) over "latest" for reproducible deploys.

Pods are the smallest deployable unit

A Pod wraps one or more tightly-coupled containers that share a network namespace (same IP) and can share storage volumes. You do not deploy a bare container — the smallest thing Kubernetes schedules and manages is a Pod. Pod IPs are ephemeral: a replacement Pod is not guaranteed the same IP, which is why Services exist.

This is a preview. Get the full KCNA cheat sheet — every domain, bundled with the practice question bank.

Sign up for the full version