(III) Building a Kubernetes Cluster with Kubespray: A Learning Journey

If you haven’t checked the previous blog post, Part II covers how the hardware was prepared and Ubuntu Server was successfully installed on all nodes. With the servers finally running, the next challenge was setting up a stable network so the cluster could communicate reliably.

If the hardware setup was the first battle, networking was definitely the second. To run a Kubernetes cluster, we needed a clean lab network and static IPs for each node. Sounds simple, right? Not when every reboot wiped the configuration. In this post, I’ll explain how we assigned IPs, the commands we used, and the lessons we learned about keeping things stable.

The Lab Network

After setting up the hardware, the next step was to connect all nodes to a stable network. Our lab was already connected to the university internet via Ethernet, but to run a Kubernetes cluster, we needed isolated static IPs for each node. Static IPs are important because Kubernetes nodes must have predictable addresses to communicate reliably. Dynamic IPs (assigned via DHCP) can change after a reboot, which would break the cluster.

We were assigned five consecutive IP addresses:

192.168.1.201 – 192.168.1.206

With the guidance of a technical officer, we mapped each workstation to one of these static IPs.

Understanding the Technical Terms

1. IP Address:
An IP (Internet Protocol) address is a unique number assigned to each device on a network, allowing them to communicate. In our case, we used private IP addresses in the 192.168.x.x range, common for local networks.

2. Network Interface:
Every device has a network interface, such as eth0 or enp0s3, representing a physical or virtual network adapter.

3. Subnet Mask (/24):
The /24 in 192.168.1.201/24 represents the subnet mask 255.255.255.0. This defines the range of IPs that are part of the same local network. All devices in this subnet can communicate directly without routing.

4. Static vs Dynamic IPs:

• Dynamic IPs are assigned by a DHCP server and can change after a reboot.
• Static IPs remain constant, which is essential for nodes in a Kubernetes cluster.

Assigning Static IPs

We used the following Linux commands to configure each node:

// Display all network interfaces and their current IPs
ip addr

This shows all network adapters (like eth0) and their assigned IP addresses.

//Assign a static IP to the network interface eth0
sudo ip addr add 192.168.1.201/24 dev eth0

Here:

• sudo runs the command with administrative privileges.
• ip addr add assigns the IP address to the interface.
• /24 is the subnet mask.
• dev eth0 specifies the network interface.

//Bring the interface up to activate the new IP
sudo ip link set eth0 up

Finally, to verify the IP assignment:

ip addr

This confirms that eth0 now has the static IP 192.168.1.201. We repeated this process for all workstations, assigning them consecutive IPs.

Whats Next

Once static IPs were assigned and verified, the next post is about the step to enable passwordless SSH between the master and worker nodes. This is required for Kubespray to deploy Kubernetes across the cluster without manual intervention on each node.