Provisioning Kubernetes Clusters with Kubespray

Kubernetes, an open-source orchestration system, automates the deployment and management of containerized applications. For beginners, the journey into Kubernetes can often start with the daunting question: "Where do I begin?"

In the early days, setting up and managing a Kubernetes cluster was a challenging and time-consuming task. However, with the evolution of Kubernetes, user-friendly solutions have emerged to simplify this process. Among these solutions, Kubespray shines as an invaluable tool.

Kubespray, an open-source solution, facilitates the automated deployment of Kubernetes clusters across nodes. Engineered to be highly customizable, efficient, and lightweight, Kubespray caters to a wide range of requirements, making Kubernetes cluster deployment accessible to all.

Overview of Kubespray
Kubespray is a composition of Ansible playbooks, inventory, provisioning tools, and generic Kubernetes cluster configuration management tasks. In this writeup, I'll demonstrate how to deploy a Kubernetes cluster on 3 nodes (1master and 2 worker nodes) using Kubespray.
While a basic understanding of Ansible and Kubernetes terminologies is assumed, the steps are simple enough for beginners to follow along.

Prerequisites
Before proceeding, ensure the following prerequisites are in place:

• Provision Infrastructure: Set up computing resources, such as 3 servers, for the cluster.
• Install Dependencies: Install the following dependencies on your Ansible server (can be one of your nodes or a dedicated server {recommended} for ansible):
• Git
• Python3
• Pip3
• Ansible

Setting Up the Cluster
Follow these steps to set up your Kubernetes cluster with Kubespray:
Step 1: Set Up SSH Keys
Generate SSH keys on the Ansible node and copy the key to all your cluster nodes:

ssh-keygen # Go with the defaults
ssh-copy-id <user>@<node-IP>

Step 2: Download and Configure Kubespray
Download the Kubespray GitHub repository and checkout the latest version:

git clone git@github.com:Kubernetes-sigs/Kubespray.git
cd Kubespray
git checkout release-2.xx #replace 'xx' with release number

Step 3: Install Python Dependencies
Install the required Python dependencies using pip:

pip3 install -r ./requirements.txt

Step 4: Update Ansible Inventory
Update the Ansible inventory file with the IP addresses of your nodes:

cp -rfp inventory/sample inventory/mycluster
declare -a IPS=(<node1-IP> <node2-IP> <node3-IP>)
CONFIG_FILE=inventory/mycluster/hosts.yaml python3 contrib/inventory_builder/inventory.py ${IPS[@]}

Further customize inventory/mycluster/hosts.yaml to specify your master, worker, and etcd nodes.

Step 5: Review and Customize Configuration
Review and customize parameters under inventory/mycluster/group_vars:

cat inventory/mycluster/group_vars/all/all.yml
cat inventory/mycluster/group_vars/k8s_cluster/k8s-cluster.yml

Step 6: Allow Kubernetes Ports
If behind a firewall, ensure all necessary Kubernetes ports are allowed. These ports include; api-server, etcd, DNS, kubelet, kube-controller-manager, calico ports, kube-scheduler e.t.c.

Step 7: Clean Up Old Kubernetes Cluster
Run the playbook to clean up any old Kubernetes cluster:

ansible-playbook -i inventory/mycluster/hosts.yaml --user=<your-user-with-sudo-access> --ask-become-pass --become reset.yml

Step 8: Deploy Kubernetes with Kubespray
Run the playbook to deploy Kubernetes:

ansible-playbook -i inventory/my-cluster/hosts.yml --user=<your-user-with-sudo-access> --ask-become-pass --become cluster.yml

Step 9: Access the Cluster
To access the cluster using kubectl commands, update your kubeconfig file to be in the ~/.kube directory:

mkdir .kube
cd .kube/
sudo cp /etc/kubernetes/admin.conf config
kubectl get nodes

Note: The playbook will take some time to complete, but once finished, you'll have a highly available and self-managed Kubernetes cluster at your disposal.
Kubespray enables you to manage Kubernetes clusters, you can add/remove nodes (playbooks/scale.yml|remove_node.yml), delete the entire cluster (playbooks/reset.yml), and performing other Kubernetes management tasks.

Troubleshooting

1. Issue with Python Packages Installation: When Ansible is already installed via system packages on the control node, Python packages installed using sudo pip install -r requirements.txt may end up in a different directory tree (e.g., /usr/local/lib/python2.7/dist-packages on Ubuntu) compared to Ansible's directory (e.g., /usr/lib/python2.7/dist-packages/ansible on Ubuntu). Consequently, the ansible-playbook command may fail with the following error:

ERROR! no action detected in task. This often indicates a misspelled module name, or incorrect module path.

This likely indicates that a task depends on a module present in requirements.txt.
2.Ensure Firewall Rules Allow Necessary Ports: Make sure that all necessary ports are allowed through the firewall to ensure proper communication between components.
3.Failure to Run Playbook without --become: The playbook will fail to run if the --become flag is not used. Ensure that you include --become to grant necessary privileges for the playbook to execute successfully.
4.For further troubleshooting on any encountered issue, please refer to the official Kubespray repository for comprehensive troubleshooting steps.