Kubernetes Jobs and CronJobs: running batch and scheduled workloads

Jobs Run Tasks to Completion

Deployments run services indefinitely. Jobs run a task until it completes successfully. CronJobs schedule Jobs on a cron-style schedule. Use these for database migrations, report generation, cache warming, and data processing pipelines.

Job Manifest

apiVersion: batch/v1
kind: Job
metadata:
  name: db-migrate
spec:
  completions: 1
  parallelism: 1
  backoffLimit: 3       # retry up to 3 times on failure
  template:
    spec:
      restartPolicy: OnFailure
      containers:
      - name: migrate
        image: my-app:2.0
        command: ["python", "manage.py", "migrate"]

CronJob

apiVersion: batch/v1
kind: CronJob
metadata:
  name: cleanup-job
spec:
  schedule: "0 2 * * *"    # every day at 2am
  concurrencyPolicy: Forbid  # skip if previous run still active
  successfulJobsHistoryLimit: 3
  failedJobsHistoryLimit: 1
  jobTemplate:
    spec:
      template:
        spec:
          restartPolicy: OnFailure
          containers:
          - name: cleanup
            image: my-app:2.0
            command: ["python", "cleanup.py"]

Jobs are not deleted automatically. Set ttlSecondsAfterFinished to auto-clean completed Jobs and prevent cluttering your cluster.