Skip to content

Kubernetes Level 01 – Day 08: Schedule CronJobs in Kubernetes

This document explains how to create and manage a Kubernetes CronJob to execute a recurring task inside the cluster. The objective is to run a simple command at a fixed interval using native Kubernetes scheduling.





Objective

Create a CronJob with the following configuration:

  • CronJob Name: devops
  • Schedule: */3 * * * *
  • Image: httpd:latest
  • Container Name: cron-devops
  • Command: echo Welcome to xfusioncorp!
  • Restart Policy: OnFailure



Understanding the Goal

A CronJob in Kubernetes is used to run tasks at scheduled times, similar to the traditional Linux cron scheduler. Instead of running scripts on a VM directly, Kubernetes creates and manages Jobs automatically according to the defined schedule.

In this task, the system must execute a container every 3 minutes and print a message.



Step 1: Create the YAML Manifest

Create a manifest file:

vi cronjob.yaml

Manifest Content

apiVersion: batch/v1
kind: CronJob
metadata:
  name: devops
spec:
  schedule: "*/3 * * * *"
  jobTemplate:
    spec:
      template:
        spec:
          containers:
          - name: cron-devops
            image: httpd:latest
            command:
            - /bin/sh
            - -c
            - echo Welcome to xfusioncorp!
          restartPolicy: OnFailure

Important Configuration Details

  • apiVersion: batch/v1 → CronJob API group
  • schedule → Standard cron format (minute hour day month weekday)
  • */3 * * * * → Every 3 minutes
  • jobTemplate → Template used to create a Job
  • template.spec → Pod specification used by the Job
  • restartPolicy: OnFailure → Restart container only if it exits with an error

The CronJob creates a Job object. The Job then creates a Pod to execute the task.



Step 2: Apply the Configuration

Deploy the CronJob into the cluster:

kubectl apply -f cronjob.yaml

Expected output:

cronjob.batch/devops created

Kubernetes now monitors the schedule and triggers Jobs automatically.



Step 3: Verification

Check CronJob Status

kubectl get cronjob

# OUTPUT

thor@jumphost ~$ kubectl get cronjob
NAME     SCHEDULE      SUSPEND   ACTIVE   LAST SCHEDULE   AGE
devops   */3 * * * *   False     0        <none>          6s

Expected output should include:

  • NAME → devops
  • SCHEDULE → */3 * * * *
  • SUSPEND → False

Manual Test (Optional)

Instead of waiting 3 minutes, create a Job immediately:

kubectl create job --from=cronjob/devops test-job

Check Pods

kubectl get pods

Identify the Pod created by the Job.

Check Logs

kubectl logs <pod-name>

Expected output:

Welcome to xfusioncorp!



How CronJobs Work Internally

  1. The CronJob controller checks the schedule.
  2. When the time matches, it creates a Job object.
  3. The Job controller creates a Pod.
  4. The Pod runs the defined command.
  5. When complete, the Pod exits.

The CronJob only manages scheduling. The Job manages execution.



Restart Policy Explanation

For Jobs and CronJobs, valid restart policies are:

  • OnFailure → Restart container if it exits with non-zero status
  • Never → Do not restart container

Always is not allowed because Jobs are expected to complete.



Key Outcome

The CronJob devops is successfully configured to execute every 3 minutes. Kubernetes automatically creates Jobs and Pods according to the defined schedule, ensuring recurring execution without manual intervention.