This blog is part of a multi-part blog series that shows how to run your applications on Kubernetes. It will use the Couchbase, an open source NoSQL distributed document database, as the  Docker

The first part (Couchbase on Kubernetes) explained how to start the Kubernetes cluster using Vagrant. That is a simple and easy way to develop, test, and deploy
Kubernetes cluster on your local machine. But this could be of limited use rather soon as the resources are constrained by the local machine. So what do you do?

Kubernetes cluster can be installed on Amazon as well. This second part will show:

  • How to setup and start the Kubernetes cluster on Amazon Web Services
  • Run Docker container in the Kubernetes cluster
  • Expose Pod on Kubernetes as Service
  • Shutdown the cluster

Here is a quick overview:

Kubernetes Cluster on Amazon with Couchbase

Let's dig into the details!

Setup Kubernetes Cluster on Amazon Web Services

Getting Started on AWS EC2 provide complete instructions to start Kubernetes cluster on Amazon. Make sure to have the pre-requisites (AWS account, AWS CLI, Full EC2 access)
met before you follow these instructions. Kubernetes cluster can be created on Amazon as:

By default, this provisions a new VPC and a 4 node Kubernetes cluster in us-west-2a (Oregon) with t2.micro instances running on Ubuntu. This means 5 AMIs (one for master and 4 for the worker nodes) are created. Some
properties that are worth updating:

  • Set NUM_MINIONS environment variable to whatever number of nodes are required in the cluster. Set it to 2 if you want only two worker nodes to be created.
  • Each instance size is 1.1.x is t2.micro. Set MASTER_SIZE and MINION_SIZE environment variables to m3.medium otherwise the nodes are going to crawl.

If you downloaded Kubernetes from, then all the values can be changed in cluster/aws/ Starting Kubernetes on
Amazon shows the following log:

Amazon Console shows:

Kubernetes Cluster on Amazon

Three instances are created as shown – one for master node and two for worker nodes. Username and password for the Kubernetes master are stored in /Users/arungupta/.kube/config. Look for a section like:

Run Docker container in Kubernetes Cluster on Amazon

Now that the cluster is up and running, get a list of all the nodes:

It shows two worker nodes. Create a new Couchbase pod:

Notice, how the image name can be specified on the CLI. This command creates a Replication Controller with a single pod. The pod uses arungupta/couchbase Docker image
that provides a pre-configured Couchbase server. Any Docker image can be specified here. Get all the RC resources:

This shows the Replication Controller that is created for you. Get all the Pods:

The output shows the Pod that is created as part of the Replication Controller. Get more details about the Pod:

Expose Pod on Kubernetes as Service

Now that our pod is running, how do I access the Couchbase server? You need to expose it outside the Kubernetes cluster. The kubectl expose command takes a pod, service or replication controller and expose it as a Kubernetes Service. Let's
expose the replication controller previously created and expose it:

Get more details about the Service:

The Loadbalancer attribute Ingress gives you the address of the load balancer that is now publicly accessible. Wait for 3 minutes to let the load balancer settle down. Access it using port 8091 and the login page for
Couchbase Web Console shows up:
Kubernetes on Amazon - Couchbase Login Page

Enter the credentials as “Administrator” and “password” to see the Web Console:
Kubernetes on Amazon - Couchbase Web Console

And so you just accessed your pod outside the Kubernetes cluster.

Shutdown Kubernetes Cluster

Finally, shutdown the cluster using cluster/ script.

For a complete clean up, you still need to explicitly delete the S3 bucket where Kubernetes binaries are stored.


Posted by Arun Gupta, VP, Developer Advocacy, Couchbase

Arun Gupta is the vice president of developer advocacy at Couchbase. He has built and led developer communities for 10+ years at Sun, Oracle, and Red Hat. He has deep expertise in leading cross-functional teams to develop and execute strategy, planning and execution of content, marketing campaigns, and programs. Prior to that he led engineering teams at Sun and is a founding member of the Java EE team. Gupta has authored more than 2,000 blog posts on technology. He has extensive speaking experience in more than 40 countries on myriad topics and is a JavaOne Rock Star for three years in a row. Gupta also founded the Devoxx4Kids chapter in the US and continues to promote technology education among children. An author of several books on technology, an avid runner, a globe trotter, a Java Champion, a JUG leader, NetBeans Dream Team member, and a Docker Captain, he is easily accessible at @arungupta.

Leave a reply