Book Image

Kubernetes on AWS

By : Ed Robinson
Book Image

Kubernetes on AWS

By: Ed Robinson

Overview of this book

Docker containers promise to radicalize the way developers and operations build, deploy, and manage applications running on the cloud. Kubernetes provides the orchestration tools you need to realize that promise in production. Kubernetes on AWS guides you in deploying a production-ready Kubernetes cluster on the AWS platform. You will then discover how to utilize the power of Kubernetes, which is one of the fastest growing platforms for production-based container orchestration, to manage and update your applications. Kubernetes is becoming the go-to choice for production-grade deployments of cloud-native applications. This book covers Kubernetes from first principles. You will start by learning about Kubernetes' powerful abstractions - Pods and Services - that make managing container deployments easy. This will be followed by a guided tour through setting up a production-ready Kubernetes cluster on AWS, while learning the techniques you need to successfully deploy and manage your own applications. By the end of the book, you will have gained plenty of hands-on experience with Kubernetes on Amazon Web Services. You will also have picked up some tips on deploying and managing applications, keeping your cluster and applications secure, and ensuring that your whole system is reliable and resilient to failure.
Table of Contents (12 chapters)

Jobs

The simplest use case for a job is to launch a single pod and ensure that it successfully runs to completion.

In our next example, we are going to use the Ruby programming language to compute and print out the first 100 Fibonacci numbers:

fib.yaml
apiVersion: batch/v1
kind: Job
metadata:
name: fib
spec:
template:
metadata:
name: fib
spec:
containers:
- name: fib
image: ruby:alpine
command: ["ruby"]
args:
- -e
- |
a,b = 0,1
100.times { puts b = (a = a+b) - b }
restartPolicy: Never

Notice that the contents of spec and template are very similar to the specification we used to launch a pod directly. When we define a pod template for use in a job, we need to choose a restartPolicy of Never or OnFailure.

The reason for this is that the end goal of a job is to run the pod until it...