- Three ubuntu servers with atleast 2vcpu's core on them.
- Ubuntu 16.04 TLS
-
Login into all three servers and download the install_all.sh script.
-
Run it using bash install_all.sh
- 2.a The script will setup docker repository and will install docker-ce packages.
- 2.b The script will setup kubernetes repository and will install required kubernetes packages.
-
Download the install_master.sh on the master node and execute the same using bash install_master.sh
- 3.a The script will initilize the cluster with default pod network CIDR.
- 3.b The script will create the login access for admin user to login from the master.
- 3.c The script will also apply flannel CNI for overlay network connectivity between the cluster nodes.
- 3.d Finally the kubelet service will be restarted.
- 3.e The final command will print the command to add the worker nodes in the cluster.
-
Copy the last command output and run the exact command on the worker nodes to get them joined into the cluster.
-
Once the command is executed, verify the cluster state using kubectl get nodes on the master.