28Kubernetes for Testers

28.1 Mental model

• Cluster = control plane + worker nodes
• Node = VM/host running pods
• Pod = one or more containers sharing network + storage. Smallest deployable unit.
• Deployment = manages a set of identical pods (replicas) and rolling updates
• Service = stable network endpoint pointing at pods (load balancer)
• Ingress = HTTP router exposing services externally
• ConfigMap / Secret = injected config (env vars or files)
• Namespace = logical partition for grouping/quota/access

28.2 kubectl essentials

kubectl config use-context staging
kubectl get nodes
kubectl get pods -A                    # all namespaces
kubectl get pods -n my-app -w           # watch
kubectl describe pod my-pod
kubectl logs my-pod -f                  # follow
kubectl logs my-pod -c container-name --previous
kubectl exec -it my-pod -- /bin/sh
kubectl port-forward svc/web 8080:80    # local 8080 → cluster service
kubectl apply -f deploy.yaml
kubectl delete -f deploy.yaml
kubectl rollout restart deployment web
kubectl rollout undo deployment web
kubectl top pod                         # CPU/RAM

28.3 A test Job

apiVersion: batch/v1
kind: Job
metadata:
  name: e2e-tests-pr-123
spec:
  backoffLimit: 1
  template:
    spec:
      restartPolicy: Never
      containers:
      - name: pw
        image: mcr.microsoft.com/playwright:v1.50.0-jammy
        env:
        - name: BASE_URL
          value: https://staging.acme.com
        - name: ADMIN_PASS
          valueFrom: { secretKeyRef: { name: e2e-secrets, key: admin-pass } }
        command: ["npx","playwright","test","--shard=1/4"]

kubectl apply -f e2e-job.yaml
kubectl logs -f job/e2e-tests-pr-123
kubectl wait --for=condition=complete --timeout=600s job/e2e-tests-pr-123

28.4 Deployment + Service + Ingress (mini stack)

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: acme/web:1.2.0
        ports: [{ containerPort: 3000 }]
        readinessProbe:
          httpGet: { path: /health, port: 3000 }
        env:
        - name: DATABASE_URL
          valueFrom: { configMapKeyRef: { name: app-config, key: db-url } }
---
apiVersion: v1
kind: Service
metadata: { name: web }
spec:
  selector: { app: web }
  ports: [{ port: 80, targetPort: 3000 }]
---
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata: { name: web }
spec:
  rules:
  - host: app.acme.com
    http:
      paths:
      - path: /
        pathType: Prefix
        backend:
          service: { name: web, port: { number: 80 } }

28.5 ConfigMap & Secret

kubectl create configmap app-config --from-literal=db-url=postgres://...
kubectl create secret generic e2e-secrets --from-literal=admin-pass=hunter2
kubectl get secret e2e-secrets -o jsonpath='{.data.admin-pass}' | base64 -d

28.6 Helm — package manager

helm repo add bitnami https://charts.bitnami.com/bitnami
helm install pg bitnami/postgresql --namespace test --create-namespace
helm upgrade --install web ./charts/web --values prod-values.yaml
helm uninstall pg
helm template ./charts/web   # render YAML without applying

A Helm chart templatises YAML. values.yaml holds defaults; values-prod.yaml overrides for prod. Common for the SDET's test stack (DB + app + mocks).

28.7 Observability for tests

• kubectl logs --tail=100 -f pod — live logs
• kubectl get events --sort-by=.metadata.creationTimestamp — recent cluster events
• Datadog / New Relic / Grafana — production-grade observability
• Lens / k9s — TUI alternatives to kubectl

28.8 Common SDET use cases

• Per-PR test environments — spin a namespace per PR, deploy app + dependencies, run E2E, destroy on close
• CronJob nightly — schedule full regression at 2 AM
• Long-running smoke — Deployment that runs synthetic monitoring against prod
• Test infra at scale — Selenium Grid / Playwright workers as deployments, autoscaled by queue depth