K8s_vagrant 🚀

Estimated reading time: 7 minutes ⏱️

This Kubernetes cluster is powered by kubeadm and managed using Vagrant virtual machines. 🎉

Tools Used 🛠️

• Kubernetes: kubeadm
• Container Runtime: containerd
• CNI Plugin: Calico 🌱

Architecture 🏗️

• 1 Control Plane Node: Connected to the public network, accessible from the host 🌍
• 2 Worker Nodes: Connected to a private network 🔒

Prerequisites 📋

Before starting, make sure the following are installed:

• VirtualBox: Download and install VirtualBox 💻
• Vagrant: Download and install Vagrant ⚙️

Setting Up the Vagrant Environment 🛠️

1. Initialize the Vagrant environment:

   vagrant init

   This will create a Vagrantfile in your current directory, which you can edit to define the configuration of your virtual machines.

2. Start the Vagrant machines:

   vagrant up

   This will start and provision your Vagrant virtual machines based on the configuration in the Vagrantfile. 🌱

   You can modify the Vagrantfile to suit your needs (e.g., increase worker nodes, customize resources, etc.). 📝

Setting Up the Cluster with kubeadm ⚡

1. SSH into the master (control plane) node:

   vagrant ssh master

2. Initialize the Kubernetes control plane node with the following command:

   sudo kubeadm init --apiserver-advertise-address=192.168.1.200 --pod-network-cidr=192.168.0.0/16

   The last output displays useful commands for the next steps. Save it!

3. Configure kubectl for your user on the control plane node:

   mkdir -p $HOME/.kube
   sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
   sudo chown $(id -u):$(id -g) $HOME/.kube/config

4. Install the Calico CNI plugin: Quickstart for Calico on Kubernetes

   CALICO_VERSION=3.29.1
   kubectl create -f https://raw.githubusercontent.com/projectcalico/calico/v$CALICO_VERSION/manifests/tigera-operator.yaml
   kubectl create -f https://raw.githubusercontent.com/projectcalico/calico/v$CALICO_VERSION/manifests/custom-resources.yaml

   You can choose a different CNI plugin if needed.

5. Verify that Calico is installed correctly by checking the pods:

   watch kubectl get pods -n calico-system

6. Remove the control plane taint to allow scheduling pods on the control plane node [Optional]:

   kubectl taint nodes --all node-role.kubernetes.io/control-plane-

7. Follow the instructions displayed after running the kubeadm init command to configure and join your worker nodes. 📝

Joining Worker Nodes 🧑‍💻

For each worker node, perform the following steps:

1. SSH into the worker node (e.g., worker1):

   vagrant ssh worker1

2. On the worker node, run the kubeadm join command that was displayed during the kubeadm init process on the control plane node:

   sudo kubeadm join <your-kubeadm-join-command>

Repeat steps 1 and 2 for the remaining worker nodes. 🔄

Log in to the Master and Access Your Cluster 🎉

Once the nodes are joined, you can log in to the master node and enjoy your Kubernetes setup!

Accessing Kubernetes from the Host 🖥️

Your cluster is accessible from the host. To manage it using kubectl on your host machine, copy the kubeconfig from the Vagrant master node and configure it:

1. Copy the kubeconfig file from the master node:

   vagrant ssh master
   sudo cat /etc/kubernetes/admin.conf

2. Set the KUBECONFIG environment variable on your host:

   export KUBECONFIG=/path/to/config

   Replace /path/to/config with the actual path to the admin.conf file you copied.

 $ kubectl get nodes
   NAME          STATUS   ROLES           AGE   VERSION
   k8s-master    Ready    control-plane   20m   v1.31.3
   k8s-worker1   Ready    <none>          12m   v1.31.3
   k8s-worker2   Ready    <none>          11m   v1.31.3

---

Happy Kuberneting! 🐳🎉

---

Clean-Up Steps 🧹

1. Destroy the Vagrant Virtual Machines

To stop and remove the virtual machines created by Vagrant:

vagrant destroy -f

2. Remove the Vagrantfile and .vagrant Directory

To completely clean up Vagrant-related files from your project directory:

rm -rf Vagrantfile .vagrant