DEV Community

Cover image for Installation of Kubernetes (v1.29) on air-gapped Linux system.
Bhikesh Khute
Bhikesh Khute

Posted on

Installation of Kubernetes (v1.29) on air-gapped Linux system.

#kubernetes #devops #cloud #linux

We will be using Ubuntu 20.04 for this demonstration. We will be using a bastion host on which internet is present and download the required packages and tar files required to be transferred to the air-gapped systems.
*Bastion System ==> (Master Node + Worker Node)*


On bastion system

Step 1 - Get the packages list and GPG key to be downloaded by APT. 

curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.29/deb/Release.key 
| sudo gpg --dearmor -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg
Enter fullscreen mode Exit fullscreen mode 
echo 'deb [signed-by=/etc/apt/keyrings/kubernetes-apt-keyring.gpg] 
https://pkgs.k8s.io/core:/stable:/v1.29/deb/ /' 
| sudo tee /etc/apt/sources.list.d/kubernetes.list
Enter fullscreen mode Exit fullscreen mode

Step 2 - Confirm the full version using the following command and note down the version in a variable - 

sudo apt list kubelet

K8S_VERSION=1.29.15-1.1
Enter fullscreen mode Exit fullscreen mode

Step 3 - Update the apt repository and download version specific binaries of kubernetes-

  • kubelet
  • kubeadm
  • kubectl 
sudo apt update
sudo apt install --download-only kubelet=$K8S_VERSION kubeadm=$K8S_VERSION   
kubectl=$K8S_VERSION
Enter fullscreen mode Exit fullscreen mode

Step 4 - We need to also download the engine of kubernetes i.e. containerd, but, the pre-requisite of containerd also needs to be downloaded and installed. Following are the binaries of the same and commands to follow -

  • conntrack
  • ethtool
  • runc
  • cri-tools
  • containerd 
sudo apt list containerd cri-tools runc conntrack ethtool
Enter fullscreen mode Exit fullscreen mode

runc_1.2.5-0ubuntu1
cri-tools_1.29.0-1.1_amd64.deb
containerd_1.7.27-0ubuntu1
conntrack/focal,now 1:1.4.5-2
ethtool/focal,now 1:5.4-1

Step 5 - Now we need networking solution between the pods/containers. We will be using calico for this demo. The yaml and images in tar format needs to be downloaded and transferred. 

curl -O https://raw.githubusercontent.com/projectcalico/calico/v3.27.0/
manifests/calico.yaml
Enter fullscreen mode Exit fullscreen mode

Grepping the images list from the yaml -

grep 'image:' calico.yaml | awk '{print $2}' | sort | uniq
Enter fullscreen mode Exit fullscreen mode

For example -

docker.io/calico/cni:v3.27.0
docker.io/calico/kube-controllers:v3.27.0
docker.io/calico/node:v3.27.0

Step 6 - Download core images of kubernetes as well -

Either run the following command on any system or refer the list below the command -

kubeadm config images list --kubernetes-version=v1.29.0
Enter fullscreen mode Exit fullscreen mode

registry.k8s.io/kube-apiserver:v1.29.0
registry.k8s.io/kube-controller-manager:v1.29.0
registry.k8s.io/kube-scheduler:v1.29.0
registry.k8s.io/kube-proxy:v1.29.0
registry.k8s.io/pause:3.9
registry.k8s.io/etcd:3.5.10-0
registry.k8s.io/coredns/coredns:v1.11.1

Step 7 - Use the following command to save the images(one example) - 

docker save -o kube-api-server.tar registry.k8s.io/kube-apiserver:v1.29.0
Enter fullscreen mode Exit fullscreen mode

Step 8 - Create a folder on the system and keep all the yamls, images and deb files. 

01. conntrack.deb
02. ethtool.deb
03. cri-tools.deb
04. kubeadm.deb
05. kubelet.deb
06. kubectl.deb
07. calico.yaml
08. docker.io/calico/cni:v3.27.0
09. docker.io/calico/kube-controllers:v3.27.0
10. docker.io/calico/node:v3.27.0
11. registry.k8s.io/kube-apiserver:v1.29.0
12. registry.k8s.io/kube-controller-manager:v1.29.0
13. registry.k8s.io/kube-scheduler:v1.29.0
14. registry.k8s.io/kube-proxy:v1.29.0
15. registry.k8s.io/pause:3.9
16. registry.k8s.io/etcd:3.5.10-0
17. registry.k8s.io/coredns/coredns:v1.11.1
18. kubernetes-cni_1.3.0-1.1_amd64.deb
19. containerd
20. runc_1.2.5-0ubuntu1
Enter fullscreen mode Exit fullscreen mode


On Master Node

Step 9 - Install conntrack, ethtool, cri-tools and containerd first before installing kubeadm,kubectl and kubelet 

sudo dpkg -i conntrack.deb runc_1.2.5-0ubuntu1.deb cri-tools.deb ethtool.deb
Enter fullscreen mode Exit fullscreen mode

Step 10 - Now Installing Containerd 

sudo dpkg -i containerd.deb
Enter fullscreen mode Exit fullscreen mode

Step 11 - Once Installed, it won't start due to configuration issue. Below is the fix for the same - 

sudo mkdir -p /etc/containerd
containerd config default | sudo tee /etc/containerd/config.toml > /dev/null
sudo sed -i 
's/SystemdCgroup = false/SystemdCgroup = true/' /etc/containerd/config.toml
sudo systemctl restart containerd
Enter fullscreen mode Exit fullscreen mode

Step 12 - Installing kubeadm, kubelet and kubectl on the master node now. 

sudo dpkg -i kubeadm.deb kubelet.deb kubectl.deb
Enter fullscreen mode Exit fullscreen mode

Step 13 - Now, we need to tune in system's configuration to make initialize the control plane so that pre-flights checks won't fail. 

sudo tee /etc/sysctl.d/k8s.conf <<EOF
net.bridge.bridge-nf-call-iptables  = 1
net.ipv4.ip_forward                 = 1
net.bridge.bridge-nf-call-ip6tables = 1
EOF

# Apply settings
sudo sysctl --system
sudo sysctl -w net.ipv4.ip_forward=1
sudo swapoff -a
sudo sed -i '/ swap / s/^/#/' /etc/fstab
Enter fullscreen mode Exit fullscreen mode

Step 14 - Importing all the tar files we shipped on this air-gapped system(one for example) - 

sudo ctr -n k8s.io images import 
registry.k8s.io_kube-controller-manager_v1.29.0.tar
Enter fullscreen mode Exit fullscreen mode

Step 15 - All pre-requisites checks are done and we can initialize the control plane - 

sudo kubeadm init --pod-network-cidr=192.168.0.0/16 
--kubernetes-version=1.29.0 
--apiserver-advertise-address=192.168.56.101
Enter fullscreen mode Exit fullscreen mode

Wait for about 2-5 minutes for successful installation and save the output for later use.

Step 16 - 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
Enter fullscreen mode Exit fullscreen mode

Alternatively, if you are the root user, you can run:

export KUBECONFIG=/etc/kubernetes/admin.conf
Enter fullscreen mode Exit fullscreen mode

Run the following command to check the status of the cluster -

kubectl get po --all
Enter fullscreen mode Exit fullscreen mode

Step 17 - Install networking plugin to fix the cluster now -

Run the following command to check the pods and node status -

kubectl get po -A
kubectl get no
Enter fullscreen mode Exit fullscreen mode

You might see that coredns and kube-proxy is crashing! In order to fix it, we need to apply the calico yaml on the master node as well as all worker nodes.

Assuming that calico images are imported into the ctr registry.

kubectl apply -f calico.yaml
Enter fullscreen mode Exit fullscreen mode

Once done, wait for 2-5 minutes and keep a watch on the pods of the master node. All the pods will be up and running and worker node will also be in ready status now.


On Worker Node

Step 18 - Repeat the step from 9 to 14 on worker node.

Step 19 - Once the above step is done, run the command which was shown in the output of Step 15 to join the cluster (the below one is the demo) - 

kubeadm join 192.168.56.101:6443 --token 
x06cy2.9tuuzcb0jswvmk6q \
--discovery-token-ca-cert-hash sha256:d0e20c6b3078b9528441cb4e3...
Enter fullscreen mode Exit fullscreen mode
****END OF THE DOCUMENT****

Top comments (0)