A Simple Local Kafka Setup Using Docker Compose

Introduction

This post is about setting up a lightweight Kafka cluster locally using Docker Compose — something that can be reused across projects for quick event-driven testing or development.
I built this setup to make it easier to spin up Kafka without depending on Docker Desktop or heavy scripts. It's minimal, persistent, and ARM-compatible (for Apple M1/M2 users).

One of my intentions for writing this post is to use it as a reference README for myself in the future — whenever I need to re-setup or troubleshoot this local Kafka environment.
If you work with Kafka often, you’ll find this setup incredibly handy as a reusable tool.

Prerequisites

Before you start, make sure your environment has the following tools installed and working correctly:

Docker Engine or Colima
- On macOS, I prefer Colima instead of Docker Desktop.
  Install it with Homebrew:
```
brew install docker colima
colima start
```
Docker Compose
Check installation:
```
docker-compose --version
```

Git (optional)
If you want to clone and reuse this setup:

git clone https://github.com/dimuthnc/kafka-developer-setup.git
cd kafka-developer-setup

Once these are ready, you can directly run the Docker Compose file to start your local Kafka cluster.

Starting the Cluster Cleanly

To ensure a clean environment before running Kafka, I recommend the following three commands:

docker-compose down -v
docker system prune -f
docker-compose up -d

docker-compose down -v removes any existing containers and associated volumes (to ensure a clean state).
docker system prune -f removes unused images and networks.
docker-compose up -d rebuilds and starts everything fresh in detached mode.

After a few seconds, run:

docker ps

You should see three running containers:
zookeeper, kafka, and kafka-ui.

Accessing Kafka UI

Once all containers are up, open your browser and navigate to:

👉 http://localhost:8080

You’ll see the Kafka UI dashboard powered by Provectus Labs Kafka UI.

Here’s what you can do:

View existing topics
Click on the “Topics” tab to see all available topics in the local broker.
Create new topics
Click “Create Topic”, give it a name (e.g., test-topic), and save.
Publish messages
Go to your topic → “Messages” tab → click “Produce Message” → enter your payload and send.
Consume messages
In the same tab, click “View Messages” to read messages from the beginning.

This interface is an excellent way to visualize and interact with Kafka without writing CLI commands.

Troubleshooting

If something doesn’t work as expected, here’s how to debug:

1. Check container statuses

docker ps

If one of them isn’t running, inspect its logs.

2. View logs

docker logs kafka -f
docker logs zookeeper -f
docker logs kafka-ui -f

These will stream live logs and often reveal configuration or connection issues.

3. Restart Colima (if using macOS)

Sometimes Docker socket or network bridge issues occur. Restart Colima:

colima stop
colima start

4. Rebuild from scratch (if all else fails)

docker-compose down -v
docker system prune -f
docker-compose up -d

If the setup still fails, verify ports (2181, 9092, 8080) aren’t already in use, and make sure the advertised listeners in docker-compose.yml are:

KAFKA_ADVERTISED_LISTENERS: PLAINTEXT://kafka:9092

Conclusion

This simple Kafka + Zookeeper + Kafka UI setup is part of my daily development workflow. It helps me quickly test Kafka-based event flows without depending on external infrastructure.

The repository is open source, so others can use it or adapt it as a base for their own Kafka setups.
If you find it useful, feel free to fork, improve, and contribute!

Authored by Dimuth Menikgama
Linked to the open-source Kafka Local Dev setup repository.