A Homelab
Setting up a homelab is a great way to learn about technologies.
This homelab is set up using KinD, the Kubernetes in Docker version of K8s, on an Ubuntu 24.04 VM. All nodes will be on the same VM in docker containers.
Networking is on a Tailscale VPN, so each machine has an IP address of 100.x.x.x.
The Github repo for this configuration contains the yaml files referenced in this article.
This is more of a template than tutorial. For information on how all of this ties together, see the excellent book Learn Kubernetes in a Month of Lunches.
Install Docker for Containers
Docker is required for this configuration, so install it if it's not available. See the official Docker documentation for instructions on installing Docker.
Install NFS for Persistent Volumes
NFS is a common provider of K8s persistent volumes. Azure and AWS (and probably Google) use NFS to mount persistent volumes (among other tech).
This homelab will create an NFS server and mount it as a client on the same VM.
Creating this server is only to approximate a cloud environment somewhat.
Export the NFS directory
Create the directory that will serve the files via NFS:
mkdir -p /srv/nfs/k8s
Create or add to the /etc/exports file the entry:
/srv/nfs/k8s 172.0.0.0/8(rw,sync,no_subtree_check,no_root_squash)
This tells the NFS server that IP addresses in Docker's IP range (172.x.x.x) are allowed to mount this directory, along with some security options.
Restart the NFS server with
sudo systemctl restart nfs-kernel-server.(Instructions taken from LearnLinux.TV.)
Create the cluster
To bring up the cluster, use the
config.kind.yaml file in the yaml directory.kind create cluster --config yaml/config.kind.yaml
This will bring up the cluster with a control plane and two workers (as defined
at the time of this writing).
Install the Ingress
Next, install the
ingress-nginx ingress controller to handle network traffic into the cluster.kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/main/deploy/static/provider/kind/deploy.yamlThis will install the most recent version of
ingress-nginx from the github repo directly.Install NFS Provisioner
Helm is required to install the provisioner for NFS, so install it first with
sudo apt install helm.Once helm is installed, the NFS provisioner can be installed with configuration information about the NFS server and export path to use.
In this cluster, the command would look like
helm repo add nfs-subdir-external-provisioner https://kubernetes-sigs.github.io/nfs-subdir-external-provisioner/ helm install nfs-subdir-external-provisioner nfs-subdir-external-provisioner/nfs-subdir-external-provisioner \ --set nfs.server=100.x.x.x \ --set nfs.path=/srv/nfs/k8s
since we have previously exported the
/srv/nfs/k8s directory in the /etc/exports file.(Instructions from their Github repo.)
Test the website
Now it's time to test the configuration of the cluster.
Let's install a web server and see if it can read a file off the NFS share. Look in the
yaml directory for files to deploy an nginx web server configured to read files off the NFS share.Deploy these with the following commmands:
sudo kubectl apply -f nfs-pv.yaml sudo kubectl apply -f nfs-pvc.yaml sudo kubectl apply -f nfs-web-deployment.yaml sudo kubectl apply -f nfs-web-service.yaml
Put a test
index.html file in /srv/nfs/k8s. Save the following (or whatever you want) to that path.<html> <head><title>NGINX in Kind</title></head> <body> <center><h1>NGINX in Kind</h1></center> <hr><center>nginx</center> </body> </html>
Use the command
curl http://localhost to see if the web server works, and it should return the contents of the file that we put in the directory.If this is what you see, congratulations, you have set up Kubernetes cluster
with an NFS share.
Next Up
A web server is a good start. However, an actual application, like the Gitea VCS server, will be a better test for the cluster.
Eventual plans are to set up a K8s cluster on various small machines to play with.