Getting started on AWS
The cluster is provisioned in separate stages as follows:
- Terraform to provision the cluster instances, security groups, firewalls, cloud infrastructure and security (SSL) certificates.
- Ansible to configure the cluster, including installing Kubernetes, addons and platform level configuration (files, directories etc...)
- You need an AWS account. Visit http://aws.amazon.com to get started
- You need an AWS instance profile and role with EC2 full access.
- You need to have installed and configured Terraform (>= 0.6.16 recommended). Visit https://www.terraform.io/intro/getting-started/install.html to get started.
- You need to have Python >= 2.7.5 installed along with pip.
- Kubectl installed in and your PATH:
curl -O https://storage.googleapis.com/kubernetes-release/release/v1.2.3/bin/linux/amd64/kubectl
On an OS X workstation, replace linux in the URL above with darwin:
curl -O https://storage.googleapis.com/kubernetes-release/release/v1.2.3/bin/darwin/amd64/kubectl
After downloading the binary, ensure it is executable and move it into your PATH:
chmod +x kubectl mv kubectl /usr/local/bin/kubectl
Download Kubeform (install from source at head)
git clone https://github.com/Capgemini/kubeform.git /tmp/kubeform cd /tmp/kubeform pip install -r requirements.txt
Configuration can be set via environment variables. As a minimum you will need to set these environment variables:
export TF_VAR_access_key=$AWS_ACCESS_KEY_ID export TF_VAR_secret_key=$AWS_SECRET_ACCESS_KEY export TF_VAR_STATE_ROOT=/tmp/kubeform/terraform/aws/public-cloud
Get terraform dependencies
cd /tmp/kubeform/terraform/aws/public-cloud terraform get for i in $(ls .terraform/modules/*/Makefile); do i=$(dirname $i); make -C $i; done
Provision the cluster infrastructure
Configure the cluster
To install the role dependencies for Ansible execute:
cd /tmp/kubeform ansible-galaxy install -r requirements.yml
To run the Ansible playbook (to configure the cluster):
ansible-playbook -u core --ssh-common-args="-F /tmp/kubeform/terraform/aws/public-cloud/ssh.config -i /tmp/kubeform/terraform/aws/public-cloud/id_rsa -q" --inventory-file=inventory site.yml -e kube_apiserver_vip=$(cd /tmp/kubeform/terraform/aws/public-cloud && terraform output master_elb_hostname)
This will run the playbook (using the credentials output by terraform and the terraform state as a dynamic inventory) and inject the AWS ELB (for the master API servers) address as a variable
cd /tmp/kubeform/terraform/aws/public-cloud terraform destroy