System requirements: A machine running Linux, macOS, or Windows with Docker installed. The Devset container requires at least 512 MB of available memory. For building from source you also need Java 25+ and Node.js 22+.
Option 1: Docker (recommended)
Basic setup
Pull and start the latest Devset Community Edition image with a single command:| Flag | Purpose |
|---|---|
-p 8082:8082 | Maps container port 8082 to your host on the same port |
-v devset-data:/data | Persists all application data in a named Docker volume |
To run Devset on a different host port — for example
9090 — change the left side of the port mapping: -p 9090:8082. Devset itself always listens on 8082 inside the container.Docker Compose with Kafka and RabbitMQ
If you don’t already have broker infrastructure running locally, use the following Compose file to bring up Kafka, RabbitMQ, and Devset together as a single stack.docker-compose.yml
- Kafka bootstrap server:
kafka:9092(from within the Compose network) orlocalhost:9092(from your host) - RabbitMQ host:
rabbitmq(from within the Compose network) orlocalhost(from your host)
Environment variables
You can configure Devset’s server binding at startup using standard Spring Boot environment variable overrides:| Variable | Default | Description |
|---|---|---|
SERVER_PORT | 8082 | The port Devset listens on inside the container |
-e flag:
Option 2: Build from source
Build Devset from source when you want to modify the codebase, run the test suite, or develop new features.Prerequisites
Before you begin, make sure the following are installed and available on yourPATH:
- Java 25 or later — the Devset backend is a Spring Boot application built with Gradle
- Node.js 22 or later — the frontend is built with a Node-based toolchain
- Docker — required to run the Docker image or the end-to-end test suite
Start the backend
8082. With the dev profile active, Devset uses a local SQLite database instead of the default data directory.
In
dev mode the SQLite database is created at devset-ce-be/data/devset.db. You can inspect or reset it directly using any SQLite client. Delete the file to start with a clean database on the next run.Start the frontend
Open a second terminal and run:8082. Open http://localhost:5173 in your browser during development.