Skip to main content
In this guide, you’ll learn how to seamlessly scale your Docker Universal Control Plane (UCP) cluster by adding a new worker node. We’ll cover:
  1. Installing Docker Enterprise Engine on the new CentOS host
  2. Retrieving the Swarm join command from the UCP web console
  3. Running the join command on the new server
  4. Verifying the node in both the UCP console and via the Docker CLI

​
1. Install Docker Enterprise Engine on the New Node

Begin by installing Docker Enterprise Engine on your CentOS machine. Follow the official Docker Engine Enterprise Documentation and select Linux → CentOS. Once installation completes, SSH into your new host (for example, ucp-worker) and start Docker:
You should see output similar to:
Ensure your Docker Enterprise Engine version matches the versions running on existing UCP managers to avoid compatibility issues.

​
2. Retrieve the Swarm Join Command from UCP

  1. Log in to your UCP web console.
  2. Go to Shared Resources → Nodes → Add Node.
  3. Under Node Type, choose Linux.
  4. Under Node Role, select Worker Node.
UCP will generate a docker swarm join command similar to:
Keep your Swarm join token secret. Treat it like a password and do not expose it in public repositories.

​
3. Join the Node to the Docker Swarm

On your new CentOS host, copy and run the command provided by UCP:
A successful join will return:

​
4. Verify the Worker Node in UCP

​
In the UCP Web Console

  1. Navigate back to Shared Resources → Nodes.
  2. Confirm your new host appears with Role: worker and Status: Ready.

​
Via the Docker CLI

On any existing manager node, run:

​
Next Steps

Repeat these steps for each additional worker or manager you wish to add. Scaling your UCP cluster enhances fault tolerance and distributes workload effectively.

​
References

Watch Video