Install a Kubernetes cluster with Kubeadm on Ubuntu step by step

Environment

OS: Ubuntu 16.04 LTS

CPU: INTEL CORE I7–7700

MEMORY: 32 GB

Step 1: Prepare necessary tools

Change to root:

sudo su

Turn off swap:

swapoff -a

Edit the swap config:

vi /etc/fstab

Comment the last line:

# swap was on /dev/sda3 during installation
# UUID=9f163eae-1ec4-4ea7-bbbe-2911e8ccd62f none

Configure iptable:

vi /etc/ufw/sysctl.conf

Add following lines to the end:

net/bridge/bridge-nf-call-ip6tables = 1
net/bridge/bridge-nf-call-iptables = 1
net/bridge/bridge-nf-call-arptables = 1

Reboot for changes

Install dependency:

sudo su
apt-get install ebtables ethtool

Reboot again

Step 2: Install kubeadm

Change to root:

sudo su

Install https support:

apt-get update && apt-get install -y apt-transport-https

Get kubernetes repo key:

curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add -

Add kubernetes repo to manifest:

cat <<EOF >/etc/apt/sources.list.d/kubernetes.list
deb http://apt.kubernetes.io/ kubernetes-xenial main
EOF

Install kubeadm and docker:

apt-get update && apt-get install -y kubelet kubeadm kubectl docker.io

Step 3: Create the cluster

Change to root:

sudo su

Initial the kubeadm:

kubeadm init --pod-network-cidr=192.168.0.0/16

After a while, you will get following output:

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 10.91.4.105:6443 --token alpia1.8cjc1yfv5ezganq7 --discovery-token-ca-cert-hash sha256:3f2da2fa1967b8e974b9097fcdd15c66e0d136db5b1f08b3db7fe45c3e2b790b

Change to regular user:

exit

Copy kubectl config:

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

Install network plugin (Calico):

kubectl apply -f https://docs.projectcalico.org/v3.3/getting-started/kubernetes/installation/hosted/rbac-kdd.yaml
kubectl apply -f https://docs.projectcalico.org/v3.3/getting-started/kubernetes/installation/hosted/kubernetes-datastore/calico-networking/1.7/calico.yaml

Taint the master node for allowing deployment:

kubectl taint nodes --all node-role.kubernetes.io/master-

Now let’s try with kubectl:

kubectl get node
NAME              STATUS   ROLES    AGE   VERSION
my-kubernete      Ready    master   83m   v1.13.2

And deploy a pod:

kubectl run hello --image=k8s.gcr.io/echoserver:1.4 --port=8080
kubectl get pod
NAME                     READY   STATUS    RESTARTS   AGE
hello-5975cd9c9d-5pvsn   1/1     Running   0          120m

Step 4: Trouble shooting

If found the core dns is always crashed like that:

$ kubectl get pod -n kube-system
NAME                                      READY     STATUS             RESTARTS   AGE
coredns-86c58d9df4-hrp2w                  0/1       CrashLoopBackOff   542        1d
coredns-86c58d9df4-ptgsk                  0/1       CrashLoopBackOff   543        1d

You need to manually delete the config in /etc/resolv.conf and create new one:

nameserver 8.8.4.4
nameserver 8.8.8.8

Then the DNS service in the cluster is working normally.

Step 5: Join a remote node

To be continued

References

Creating a single master cluster with kubeadm

Installing kubeadm

How to install Kubernetes with Kubeadm: A quick and dirty guide | Mirantis

kubeadm安装踩坑-我心飞扬-51CTO博客