Single click installation of RedHat OpenShift with Contrail SDN (on AWS)

Red Hat OpenShift Origin with Contrail SDN

This tutorial walks you through the installation of Red Hat OpenShift container orchestration platform with Contrail SDN as the CNI on Amazon Web Services (AWS).

It leverages AWS's CloudFormation to launch the stack & takes approximately 30 min for the total installation to complete. The stack builds

• Red Hat OpenShift Origin v3.7
• Contrail Networking CNI 5.0

Prerequisites

• Create an AWS account if you don't have one. Else Login

• Subscribe to CentOS AMI on AWS marketplace

Installation

• Click on the button below to launch the stack in AWS

  

• Once you click on the button, you will be navigated to AWS CloudFormation page. Enter the parameters

  

  NOTE: You can leave most of the parameters set to default

   InstanceType:
     Description: EC2 instance type
     Default: t2.xlarge
  
   VpcCIDR:
     Description: CIDR block for the VPC
     Default: 10.10.0.0/16
  
   SubnetCIDR:
     Description: CIDR block for the VPC subnet
     Default: 10.10.10.0/24
  
   MasterIPv4Address:
     Description: Master instance's IPv4 Address
     Default: 10.10.10.10
  
   MinionIPv4Address:
     Description: Minion instance's IPv4 Address
     Default: 10.10.10.11
  
   SSHLocation:
     Description: Allow access to EC2 instances from
     Default: 0.0.0.0/0
  
   InstancePassword:
     Description: Password for the instances
  
   ContrailBuild:
     Description: Contrail build information
     Default: latest
  

• Wait for the stack to complete. You can monitor the resource creation by clicking on the Events tab

  

• Once complete, navigate to the Outputs tab & copy the ShellURL value. Login to the instance using the ShellURL & the password you set

  

• Run the script from the master instance's /root directory

   (local-instance)# ssh root@ec2-<public-ip>.us-west-1.compute.amazonaws.com
  
   (master-instance)# cd /root
   (master-instance)# ~/run.sh
  

  

• Once the installation is complete, enter the password for "admin" user to login into OpenShift dashboard

   (master-instance)# oadm policy add-cluster-role-to-user cluster-admin admin
   (master-instance)# htpasswd /etc/origin/master/htpasswd admin
  

• Run post-install validations

   (master-instance)# docker restart $(docker ps | grep -i kube-man | cut -d " " -f1 | head -1)
   (master-instance)# ssh root@$(oc get nodes  | grep -i " Ready" | cut -d " " -f1) iptables -I INPUT 4 -j ACCEPT
   (master-instance)# oc get pods -n kube-system -o wide
  

• Login to the dashboards (WebUI) of both OpenShift & Contrail. The URL's are listed in the Outputs tab of AWS CloudFormation

  

  

• Verify all Contrail pods are running healthy, by logging into OpenShift & Contrail dashboards

  OpenShift Dashboard > My Projects > kube-system > Applications > Pods

  

  Contrail Dashboard > Monitor > Infrastructure > Dashboard

  

• Try launching an application using OpenShift dashboard. Follow the example video below