Running Jobs and CronJobs in Kubernetes
Overview
Kubernetes Jobs are objects that are designed to run a containerized task successfully to completion. CronJob basically takes that same idea and extends it and allows us to run Jobs periodically, according to a schedule.
Jobs
Create a simple Job that runs the command echo This is a test! .
vi my-job.yml
## Job Definition
apiVersion: batch/v1
kind: Job
metadata:
name: my-job
spec:
template:
spec:
containers:
- name: print
image: busybox:stable
command: ["echo", "This is a test!"]
restartPolicy: Never
backoffLimit: 4
activeDeadlineSeconds: 10
Our container is just going to run its code and then it’s going to stop. It’s not going to continue to run forever.
That’s why we have restartPolicy set to Never, because this container never needs to restart. It just needs to fire and finish.
activeDeadlineSeconds : This is the maximum number of seconds that Kubernetes will allow the Job to run. So if our command gives back an error, and our Job doesn’t actually complete successfully, Kubernetes will try to spin that container up again and retry the Job,
backofflimit: It just limits essentially how many times Kubernetes is going to retry that Job execution.
Run the Job.
kubectl apply -f my-job.yml
Check the status of the Job.
kubectl get jobs
View the Job output. First, you will need to find the name of the Job’s Pod.
kubectl get pods
kubectl logs $JOB_POD_NAME
CronJobs
CronJob basically takes that same idea and extends it and allows us to run Jobs periodically, according to a schedule.
Creating a CronJob that will run the previous Job task every minute is like so: Eerything under the Job template is pretty much everything we used
when we just created the Job. In addition we have the schedule field, the value of which uses the Cron expression syntax.
vi my-cronjob.yml
## Job Definition
apiVersion: batch/v1
kind: CronJob
metadata:
name: my-cronjob
spec:
schedule: "*/1 * * * *"
jobTemplate:
spec:
template:
spec:
containers:
- name: print
image: busybox:stable
command: ["echo", "This is a test!"]
restartPolicy: Never
backoffLimit: 4
activeDeadlineSeconds: 10
Run the Job.
kubectl apply -f my-cronjob.yml
Check the CronJob status.
kubectl get cronjob
View the Job output. First, you will need to find the name of the Job’s Pod.
kubectl get pods
kubectl logs $JOB_POD_NAME
Delete pods
$ kubectl delete -f multi-pod.yml