Kubernetes for Beginners: Pods, Deployments, and Services

Kubernetes (K8s) has a reputation for complexity, but the core mental model is simple: you describe the state you want, and Kubernetes continuously works to make reality match it. Let's cover the three objects you'll use every day.

Table of Contents

• Pods: the smallest unit
• Deployments: keeping N copies alive
• Services: a stable address
• Essential kubectl commands
• How to learn without drowning

Pods: the smallest unit

A Pod is one or more containers that share a network and storage. You almost never create Pods directly — they're managed by higher-level objects. Think of a Pod as the atom of Kubernetes: indivisible, ephemeral, disposable.

Deployments: keeping N copies alive

A Deployment declares: "run 3 replicas of this image." If a Pod dies, Kubernetes replaces it. If you change the image tag, it performs a rolling update — gradually swapping old Pods for new ones with no downtime.

apiVersion: apps/v1
kind: Deployment
metadata:
  name: web
spec:
  replicas: 3
  selector:
    matchLabels:
      app: web
  template:
    metadata:
      labels:
        app: web
    spec:
      containers:
        - name: web
          image: nginx:1.27
          ports:
            - containerPort: 80

Services: a stable address

Pods come and go and their IPs change. A Service gives a set of Pods a stable DNS name and virtual IP. Other apps just call web:80 and Kubernetes load-balances across healthy Pods.

apiVersion: v1
kind: Service
metadata:
  name: web
spec:
  selector:
    app: web
  ports:
    - port: 80
      targetPort: 80

Essential kubectl commands

kubectl get pods                  # what's running
kubectl logs deploy/web           # stream logs
kubectl describe pod <name>      # why is it broken?
kubectl rollout status deploy/web
kubectl rollout undo deploy/web   # instant rollback

How to learn without drowning

• Start with kind or minikube locally — no cloud bill needed.
• Master Deployments and Services before touching Ingress, ConfigMaps, or Helm.
• Always set resources.requests and resources.limits — unbounded Pods are the #1 cause of noisy-neighbor incidents.
• Add a readinessProbe early; without it, traffic hits Pods before they're ready.

Kubernetes rewards incremental learning. Get one Deployment healthy end-to-end, and everything else builds on that.

Related Articles

• Kubernetes Ingress and TLS: Exposing Services Properly
• The Complete kubectl Cheat Sheet: Every Command That Matters

---

Last Updated: October 2026 Author: CloudOpsGuide Team Difficulty: Beginner Estimated Reading Time: 12 minutes