To install a kubernetes cluster from scratch, we use kubeadm.
https://kubernetes.io/docs/setup/independent/create-cluster-kubeadm/
TODO: automate this whole page :)
# ip_forward needs to be enabled before docker starts otherwise docker
# sets the iptables foward poilcy to drop
cat <<EOF >/etc/sysctl.d/05-forwarding.conf
net.ipv4.ip_forward=1
EOF
sysctl --load=/etc/sysctl.d/05-forwarding.conf
curl -sSL https://get.docker.io | sudo bash #yolo
# containerd needs CRI enabled, which is disabled by default
rm /etc/containerd/config.toml
systemctl restart containerd
# kubectl crashloops if swap is enabled... but kubeadm only gives a warning :(
swapoff -a
sed -i '/swap/d' /etc/fstab
curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add -
cat <<EOF >/etc/apt/sources.list.d/kubernetes.list
deb http://apt.kubernetes.io/ kubernetes-xenial main
EOF
apt-get update
apt-get install -y kubelet kubeadm kubectl
sed -i 's/\(ExecStart=.*kubelet.*\)/\1 --cgroup-driver=cgroupfs/' /etc/systemd/system/kubelet.service.d/10-kubeadm.conf
systemctl daemon-reload
systemctl restart kubelet
kubeadm init
If your LAN IP address is not on the same interface as the default route (eg, in VirtualBox where the route is out the private vbox NAT interface), use the --apiserver-advertise-address=x.x.x.x argument to the above command to hard code an address.
The command will output a bunch of garbage, and eventually right at the end give you a line starting with kubeadm join. This is the line you need for the workers. Save it but keep it secure. This token is effectively a root password to your cluster.
And then make the kubectl API easily accessible by making a .kube config directory:
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
Now you can run kubectl commands from your terminal. You can also copy this config file to another PC to use kubectl remotely.
When you initialised the master, it gave you a kubectl join command. Run that on the worker.
Alternatively, you can use kubeadm token create | tail -n 1 on the master to generate a token. Then, on your worker-to-be, run kubeadm join --token <token> 10.0.0.123:6443 (where 10.0.0.123 is your master's IP address)
Now go follow the networking instructions to set up the network for the worker - the node won't be considered as ready until it has network.
Once network is set up, wait for a minute and run (back on the master)
kubectl get nodes
and you should see your worker(s) listed as Ready if it joined and the networking is set up properly.