Last updated on February 10, 2018
Before installing Kubernetes, you have to install dependencies of Kubernetes. You have to install those dependencies in all of the nodes that you will join in the Kubernetes cluster.
Installing Dependencies
The first piece to be installed is apt-transport-https (a package that allows using HTTPS well as HTTP in apt repository sources). You can install it with below commond
root@arjun:~$ apt-get update && apt-get install -y apt-transport-https
We gonna use Docker as container engine, so let’s install it with below command
root@arjun:~$ apt install docker.io
Once Docker is installed, start and enable the Docker service with the commands
root@arjun:~$ systemctl start docker root@arjun:~$ systemctl enable docker
Here enable marks the service for starting up on boot, and start actually start the service immediately.
Install kubeadm
In the previous step, we have installed all the dependencies now install kubeadm.
Retrieve the key for the Kubernetes repo and add it to your key manager
root@arjun:~$ curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add -
Add the kubernetes repo to your system
root@arjun:~$ cat </etc/apt/sources.list.d/kubernetes.list deb http://apt.kubernetes.io/ kubernetes-xenial main EOF
Update repository list
root@arjun:~$ apt-get update
All set, now Install kubelet, kubeadm, kubectl and kubernetes-cni (a package that enables cni network on your machine)
root@arjun:~$ apt-get install -y kubelet kubeadm kubectl kubernetes-cni
You need to do all the above steps on all of your machines that you want to run kubernetes
Initialize your master
To initialize the master, pick one of the machines you previously installed kubeadm on, and run:
root@arjun:~$ kubeadm init
you will get similar output in your terminal…
... ... Your Kubernetes master has initialized successfully! To start using your cluster, you need to run the following as a regular user: mkdir -p $HOME/.kube sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config sudo chown $(id -u):$(id -g) $HOME/.kube/config You should now deploy a pod network to the cluster. Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at: https://kubernetes.io/docs/concepts/cluster-administration/addons/ You can now join any number of machines by running the following on each node as root: kubeadm join --token 24fb38.98e1d033bd7d4e9c 172.31.9.30:6443 --discovery-token-ca-cert-hash sha256:f1fffda887f34f62afcb9306b13926e6c009eb5c74ef4e1f8949bb54df797bcd
To make kubectl work for your non-root user, you might want to run these commands (which is also a part of the kubeadm init output):
ubuntu@arjun:~$ sudo mkdir -p $HOME/.kube ubuntu@arjun:~$ sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config ubuntu@arjun:~$ sudo chown $(id -u):$(id -g) $HOME/.kube/config
Make a record of the kubeadm join command that kubeadm init outputs. You will need this in a moment.
Deploying a pod network
You must deploy a pod network before anything will actually function properly. I’ll demonstrate this by installing the Romana pod network. This can be done with two commands (run on the master):
root@arjun:~$ sysctl net.bridge.bridge-nf-call-iptables=1 root@arjun:~$ kubectl apply -f https://raw.githubusercontent.com/romana/romana/master/containerize/specs/romana-kubeadm.yml
You can check your current node status with command
ubuntu@arjun:~$ kubectl get nodes
Setup kubernetes on other node
Just do all the steps on the node but not kubeadm init and pod network setup and run the join command,
root@node-01:~$ kubeadm join --token 24fb38.98e1d033bd7d4e9c 172.31.9.30:6443 --discovery-token-ca-cert-hash sha256:f1fffda887f34f62afcb9306b13926e6c009eb5c74ef4e1f8949bb54df797bcd
If you forgot the cluster token, you can generate a new one with the command:
root@arjun: kubeadm token generate 24fb38.98e1d033bd7d4e9c