Kubernete Cluster on Ubuntu

Table of Contents

•  
• Step 1: Install Docker
• Step 2: Start and Enable Docker
• Step 3: Add Kubernetes Signing Key
• Step 4: Add Software Repositories
• Step 5: Kubernetes Installation Tools
• Step 6: Begin Kubernetes Deployment
• Step 7: Assign Unique Hostname for Each Server Node 
• Step 8: Initialize Kubernetes on Master Node
• Step 9: Deploy Pod Network to Cluster
• Step 10: Join Worker Node to Cluster
• Step 11: Create Docker Registry
• Troubleshoot:

Step 1: Install Docker

1. Update the package list with the command:

 sudo apt-get update

2. Next, install Docker with the command:

 sudo apt-get install docker.io

3. Repeat the process on each server that will act as a node.

4. Check the installation (and version) by entering the following:

 docker ??version

Step 2: Start and Enable Docker

1. Set Docker to launch at boot by entering the following:

 sudo systemctl enable docker

2. Verify Docker is running:

 sudo systemctl status docker

To start Docker if it?s not running:

 sudo systemctl start docker

3. Repeat on all the other nodes.

Step 3: Add Kubernetes Signing Key

1. Enter the following to add a signing key:

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

If you get an error that curl is not installed, install it with:

 sudo apt-get install curl

2. Then repeat the previous command to install the signing keys. Repeat for each server node.

Step 4: Add Software Repositories

Kubernetes is not included in the default repositories. To add them, enter the following:

 sudo apt-add-repository "deb http://apt.kubernetes.io/ kubernetes-xenial main"

Repeat on each server node.

Step 5: Kubernetes Installation Tools

1. Install Kubernetes tools with the command:

 sudo apt-get install kubeadm kubelet kubectl

 sudo apt-mark hold kubeadm kubelet kubectl

Allow the process to complete.

2. Verify the installation with:

 kubeadm version

3. Repeat for each server node.

Step 6: Begin Kubernetes Deployment

Start by disabling the swap memory on each server:

 sudo swapoff -a

Step 7: Assign Unique Hostname for Each Server Node 

Decide which server to set as the master node. Then enter the command:

 sudo hostnamectl set-hostname master-node

sudo vi /etc/hosts and add the entry: 

     127.0.0.1       master-node

Next, set a worker node hostname by entering the following on the worker server:

 sudo hostnamectl set-hostname worker1

sudo vi /etc/hosts and add the entry: 

     127.0.0.1       worker1

 You need to do for the worker2 node too.

 You may need to rerun:

 sudo swapoff -a

Step 8: Initialize Kubernetes on Master Node

Switch to the master server node, and enter the following:

 sudo kubeadm init --pod-network-cidr=10.244.0.0/16

Once this command finishes, it will display a kubeadm join message at the end. Make a note of the whole entry. This will be used to join the worker nodes to the cluster.

Next, enter the following to create a directory for the cluster:

 kubernetes-master:~$ mkdir -p $HOME/.kube

 kubernetes-master:~$ sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config

 kubernetes-master:~$ sudo chown $(id -u):$(id -g) $HOME/.kube/config

Step 9: Deploy Pod Network to Cluster

sudo kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/ku...

Verify that everything is running and communicating: 

 kubectl get pods --all-namespaces

Step 10: Join Worker Node to Cluster

Switch to the worker1 system and enter the command you noted from Step 7:

kubeadm join 10.114.95.77:6443 --token mgnqdd.uw3929bzbft7n87h     --discovery-token-ca-cert-hash sha256:2508d1d3312488effc9bf73ea13c0863efc49a7e77de64864c9d884a2e0e9e1a

*** Node joining:

1. Go to master-node, run 

   kubeadm token create --print-join-command

2. Go to worker1 node. Run the command return from 1. Example:

   kubeadm join 10.114.95.77:6443 --token mgnqdd.uw3929bzbft7n87h     --discovery-token-ca-cert-hash sha256:2508d1d3312488effc9bf73ea13c0863efc49a7e77de64864c9d884a2e0e9e1a

Step 11: Create Docker Registry

1. start docker registry 

    docker run -d -p 5000:5000 -e REGISTRY_STORAGE_DELETE_ENABLED=true --restart=always --name registry registry:2

2. on each node machine, change insecure docker registry setting on all nodes. Create or update /etc/docker/daemon.json. And restart the docker service: service docker restart (Or systemctl restart docker)

   {"insecure-registries" : ["10.114.95.77:5000"]}

3. sudo docker run -d -p 5000:5000 --restart=always --name registry registry:2

Troubleshoot:

1. If the kubeadm and kubectl version is not consistent on master and worker nodes, we will get some error like:

   [kubelet-start] Downloading configuration for the kubelet from the "kubelet-config-1.15" ConfigMap in the kube-system namespace error execution phase kubelet-start: configmaps "kubelet-config-1.15" is forbidden: User "system:bootstrap:1qn44v" cannot get resource "configmaps" in API group "" in the namespace "kube-system"

2. change docker image storage location to  avoid /var/lib disk run out.  https://kubernetes.io/docs/tasks/administer-cluster/out-of-resource/
   • Open the /etc/default/docker file, uncomment the following line and add the new path to DOCKER_OPTS variable like ?-g /yourfolder/docker_storage?
   • Or you can add the new path into /etc/docker/daemon.json file like the following.

      

     {

      

     //...

      

     "data-root": "/yourfolder/docker_storage",

      

     //...

      

     }

3.  After the machine reboot, we have to turn off the swap.

    

   Run: swapoff -a

    

   systemctl start kubelet

4. Warning  FailedCreatePodSandBox  1s               kubelet, worker-node-154  Failed to create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "3a0c7cc1ed01d73fe779af669d3ac1c4d5fb7758a23350bf43089ad49c5f8d9e" network for pod "bcce-cms-5b5d6dd69c-bh8b8": networkPlugin cni failed to set up pod "bcce-cms-5b5d6dd69c-bh8b8_default" network: failed to set bridge addr: "cni0" already has an IP address different from 10.244.3.1/24

    

   On worker node, delete cni0 and rejoin the cluster

    

   sudo ip link delete cni0 

Please contact TIBCO support or presales for the scripts.