Install Kubernetes with kubespray

Step-by-step instructions

Page content

Howto: installing Kubernetes using Kubespray, including setting up the environment, configuring the inventory, and running the Ansible playbooks.

workers are sparying the cube in the old factory building

Prerequisites

Hardware and Software Requirements

Ensure you have multiple machines (at least 3 nodes: 1 master and 2 workers) with the following specifications:
- Ubuntu 16.04/18.04 or CentOS 7 installed.
- At least 2 GB of RAM per machine (more RAM is recommended for larger clusters).
- Network connectivity and SSH access from the Kubespray control machine to each node.
- For high availability, consider nodes with 16 GB RAM, 100 GB disk, and 6 CPU cores.

Control Machine Setup

Install Git if not already present:
```
sudo apt install git-all
```

Install Python (2.7 or 3.5+) and pip:

sudo apt update
sudo apt install python3 python3-pip

Install Ansible and other dependencies. Kubespray provides a script to handle this:

git clone https://github.com/kubernetes-sigs/kubespray.git
cd kubespray
git checkout release-2.20  # Optional: Check out a stable release
python3 -m venv venv  # Create a Python virtual environment
source venv/bin/activate  # Activate the virtual environment
pip install -r requirements.txt  # Install Ansible and other dependencies

Configure the Inventory

Copy the example inventory:

cp -r inventory/sample inventory/mycluster

Edit the inventory/mycluster/hosts.ini file to add your nodes:

[all]
node1 ansible_host=10.10.1.3 ansible_user=ubuntu
node2 ansible_host=10.10.1.4 ansible_user=ubuntu
node3 ansible_host=10.10.1.5 ansible_user=ubuntu

[kube-master]
node1

[kube-node]
node2
node3

If you have a different user, replace ubuntu with your user.

You can also use the inventory_builder script to generate the inventory file:

declare -a IPS=(10.10.1.3 10.10.1.4 10.10.1.5)
HOST_PREFIX=dcm-cp- KUBE_CONTROL_HOSTS=3 CONFIG_FILE=inventory/mycluster/hosts.yaml python3 contrib/inventory_builder/inventory.py ${IPS[@]}

Customize Configuration

Customize the Kubernetes cluster settings in the inventory/mycluster/group_vars/k8s_cluster/k8s-cluster.yml file. For example:
```
kube_network_plugin: flannel
kubeconfig_localhost: true
```

For high availability, configure a VIP and load balancer:

# In inventory/mycluster/group_vars/all/all.yml
kube_vip_arp_enabled: true
kube_vip_enabled: true
kube_vip_controlplane_enabled: true
kube_vip_address: 10.7.12.250  # Update with correct VIP
kube_vip_interface: eno1  # Update with correct interface

Deploy the Cluster

Run the Ansible playbook to deploy Kubernetes:

ansible-playbook -i inventory/mycluster/hosts.yaml cluster.yml -b -v --become --become-user=root --private-key=~/.ssh/private_key

Post-Deployment

Once the deployment is complete, you will find a kubeconfig file in the inventory/mycluster/artifacts directory. This file is necessary to interact with your Kubernetes cluster using kubectl.
```
export KUBECONFIG=inventory/mycluster/artifacts/admin.conf
kubectl get nodes
```

Troubleshooting and Access

To troubleshoot or access your cluster, ensure you have the kubectl binary installed on your control machine or any machine where you plan to manage the cluster from.
If you encounter issues like worker nodes not being ready or other pod scheduling problems, check the logs and adjust configurations as necessary.