kubernetes

How to create a kubernetes cluster of your own

Create a set of VMs to make up your cluster:

• One controller node
• One or more worker nodes
• I used Ubuntu 22.04.1

Clone the repo to each host (as a normal user is ok)

• git clone https://github.com/drq883/kubernetes.git

As root, cd to the cloned repo and run these commands on all nodes

• ./setup-container.sh
• ./setup-kubetools.sh

As root on the controller node run: kubeadm init

• Note the command that is tells you to run on the workers (copy into file/buffer)
  • i.e. kubeadm join 192.168.7.178:6443 --token kk26ym.v4l3o1qk0afau25r \ --discovery-token-ca-cert-hash <...>
• exit

Run this as a regular user on the controller node - setup access to the cluster

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

Run this as a normal user on the controller node - setup Calico as the Network agent

• ./setup-calico.sh
• run kubectl get all -A and wait until all Running

Run the kubeadm join ... command, you saved from step 4 on each worker node as root

After all these run to completion, you should have a working kubernetes cluster. At this point you probably should add an ingress and nfs persistentvolume.

From the controller node, as a normal user, you can run

• ./setup-ingress.sh

From the nfs controller node, as a normal user, you should

• adjust the following values in nfs-pv.yaml:
  • spec.capacity.storage
  • spec.nfs.path
  • spec.nfs.server
• kubectl create -f nfs-pv.yaml