Kubernetes Test Execution

Run Playwright tests as Kubernetes Jobs and CronJobs for scalable, scheduled execution.

Why Kubernetes for Tests?

Kubernetes (k8s) enables:

• Scalable execution — spin up N pods for parallel sharding
• Scheduled runs — CronJob for nightly regression suites
• Resource isolation — CPU/memory limits per test pod
• Cloud-native CI — tests run inside the same cluster as the app under test

Playwright Test Job

# playwright-job.yaml
apiVersion: batch/v1
kind: Job
metadata:
  name: playwright-tests
  namespace: testing
spec:
  completions: 1
  backoffLimit: 0               # don't retry on failure — report clearly
  template:
    spec:
      restartPolicy: Never
      containers:
        - name: playwright
          image: mcr.microsoft.com/playwright:v1.45.0-jammy
          workingDir: /app
          command: ["npx", "playwright", "test"]
          env:
            - name: BASE_URL
              value: "http://app-service.default.svc.cluster.local"
            - name: CI
              value: "true"
            - name: API_KEY
              valueFrom:
                secretKeyRef:
                  name: test-secrets
                  key: api-key
          resources:
            requests:
              memory: "2Gi"
              cpu: "1"
            limits:
              memory: "4Gi"
              cpu: "2"
          volumeMounts:
            - name: test-code
              mountPath: /app
            - name: reports
              mountPath: /app/playwright-report
      volumes:
        - name: test-code
          configMap:
            name: test-config
        - name: reports
          emptyDir: {}

Parallel Sharding with Jobs

# playwright-sharded-job.yaml
apiVersion: batch/v1
kind: Job
metadata:
  name: playwright-sharded
spec:
  completions: 4                 # total shards
  parallelism: 4                 # run all shards simultaneously
  completionMode: Indexed        # JOB_COMPLETION_INDEX = 0,1,2,3
  backoffLimit: 0
  template:
    spec:
      restartPolicy: Never
      containers:
        - name: playwright
          image: my-pw-tests:latest
          command:
            - sh
            - -c
            - |
              SHARD=$((JOB_COMPLETION_INDEX + 1))
              npx playwright test --shard=$SHARD/4 --reporter=blob
          env:
            - name: JOB_COMPLETION_INDEX
              valueFrom:
                fieldRef:
                  fieldPath: metadata.annotations['batch.kubernetes.io/job-completion-index']

Nightly CronJob

# playwright-cronjob.yaml
apiVersion: batch/v1
kind: CronJob
metadata:
  name: playwright-nightly
  namespace: testing
spec:
  schedule: "0 2 * * *"         # 2 AM UTC every day
  concurrencyPolicy: Forbid     # don't overlap runs
  failedJobsHistoryLimit: 3
  successfulJobsHistoryLimit: 3
  jobTemplate:
    spec:
      backoffLimit: 1
      template:
        spec:
          restartPolicy: Never
          containers:
            - name: playwright
              image: my-pw-tests:latest
              command: ["npx", "playwright", "test", "--reporter=html,junit"]
              env:
                - name: BASE_URL
                  value: "https://production.example.com"

Secrets in Kubernetes

# Create secret from literal values
kubectl create secret generic test-secrets \
  --from-literal=api-key=mysecretkey \
  --from-literal=db-password=dbpassword \
  -n testing

# Reference in Pod spec
env:
  - name: API_KEY
    valueFrom:
      secretKeyRef:
        name: test-secrets
        key: api-key

Never commit secrets to YAML files. Use kubectl create secret or sealed-secrets / external-secrets operator.

Running and Monitoring

# Apply job
kubectl apply -f playwright-job.yaml

# Watch pod status
kubectl get pods -n testing -w

# Stream logs
kubectl logs -f job/playwright-tests -n testing

# Get exit code
kubectl get job playwright-tests -n testing -o jsonpath='{.status.conditions[0].type}'
# "Complete" = passed, "Failed" = failed

# Cleanup
kubectl delete job playwright-tests -n testing

When to Use Kubernetes vs Docker Compose