You can find all of the config files on the GitHub page.

Overview

This is the first in a series of posts where I’m going walk through deploying a highly available Kubernetes 1.3 environment. I used the following resources during my research:

By following this guide you can expect the following:

Become more familiar with the Kubernetes components
Learn how to overcome some of the common obstacles you’re likely to run into while deploying Kubernetes
Learn about CloudFlare’s PKI/TLS toolkit to create SSL certificates
Have a Kubernetes lab to experiment with
Learn about flannel for the overlay network
Learn about etcd for service discovery
Have a base from which you can automate the process

There is nothing original here and if you’re familiar with installing open source solutions, you may not find much here, but I hope it helps someone. This is really a dump of my lab notes so I’m not going to into details on each of the components.

Infrastructure

I deployed Kubernetes using my vSphere lab, but there is nothing specific to vSphere in this guide so I imagine it will work in most environments.

Virtual Machines

Component	Value
Operating System	CentOS Linux release 7.2.1511 Minimal
CPUs	1
Memory	2 GB
Hard Disk	1 x 5 GB
Domain Name	vmware.local

Hostname	IP Address
kube-controller0	192.168.3.176
kube-controller1	192.168.3.177
kube-controller2	192.168.3.178
kube-etcd0	192.168.3.179
kube-etcd1	192.168.3.180
kube-etcd2	192.168.3.181
kube-worker0	192.168.3.182
kube-worker1	192.168.3.183
kube-worker2	192.168.3.184

Prerequisites

To make things a little easier, I performed the following. At some point (probably when I deploy Kubernetes 1.4) I’m going to go back and make it so these aren’t necessary.

Disable firewalld and use iptables

Perform the following on each node in the environment.