Vagrant Kubernetes Cluster

Kubernetes cluster deployment with a single control plane node and an arbitrary number of nodes, using Vagrant (VirtualBox provider, Ansible provisioner).

What's in the Box

The default configuration will build a 4 node cluster --- a master and 3 workers:

$ kubectl get nodes
NAME     STATUS   ROLES    AGE    VERSION
master   Ready    master   176m   v1.17.0
node01   Ready    <none>   115m   v1.17.0
node02   Ready    <none>   79m    v1.17.0
node03   Ready    <none>   74m    v1.17.0

Prerequisites

You will need the following installed:

• VirtualBox
• Vagrant
• kubectl (Optional)
• Git (Optional)

Getting Started

1. Download the zip file or git clone this repository:

git clone https://github.com/craigforr/vagrant-kubernetes.git

2. Change into the project directory and run Vagrant:

cd vagrant-kubernetes
vagrant up

3. Once the cluster is built, you can either run kubectl against the master from your local system or SSH into the master, as follows:

vagrant ssh

4. When you're done using the cluster, you can either stop the nodes or destroy the cluster:

• Power down all the nodes:

$ vagrant halt
==> node03: Attempting graceful shutdown of VM...
==> node02: Attempting graceful shutdown of VM...
==> node01: Attempting graceful shutdown of VM...
==> master: Attempting graceful shutdown of VM...

• Delete all the nodes:

$ vagrant destroy
# vagrant destroy --force
==> node03: Destroying VM and associated drives...
==> node02: Destroying VM and associated drives...
==> node01: Destroying VM and associated drives...
==> master: Destroying VM and associated drives...
==> master: Running action triggers after destroy ...
==> master: Running trigger: Remove join_command.sh...
    master: Running local: Inline script
    master: rm -v playbooks/join_command.sh
    master: VERBOSE: Performing the operation "Remove File" on target
    master: "/home/user/vagrant-kubernetes/playbooks/join_command.sh".

To see the status of the cluster nodes at any time, use vagrant status:

$ vagrant status
Current machine states:

master                    poweroff (virtualbox)
node01                    poweroff (virtualbox)
node02                    poweroff (virtualbox)
node03                    poweroff (virtualbox)