Kubernetes restart policies

There are several types of workloads in general, including:

• Long-running apps like web servers
• Short-lived apps like batch jobs

Suppose you have a web server, and it just crashed for some reason. It takes time to:

1. Figure out that your service is actually down
2. Log in to your cluster and restart a pod.

Therefore, it’s better to make use of a right restart policy in k8s.

The restart policy for a container is defined within the context of a pod, which is the smallest deployable unit in the Kubernetes architecture. It determines the conditions under which a container should be restarted within a pod.

There are three main restart policies:

• Always. It will restart the container indefinitely, regardless of the exit status.
• OnFailure. The container will be restarted only if it exits with a non-zero status.
• Never. No restarts at all.

The default one is Always.

You can set the value under the corresponding property in your deployment.yaml:

apiVersion: v1
kind: Pod
metadata:
  name: example-pod
spec:
  restartPolicy: Never
  containers:
  - name: my-container
    image: my-image:latest

In the previous article, we figured out how to install k8s on the desktop.

Try it yourself!