Docker Compose Deployment
Docker Compose is the quickest way to try Luna DevOps on a personal server, in a test environment, or with a small team. It starts every required service together, so PostgreSQL and Redis do not need to be installed separately.
If you want the platform itself to run in Kubernetes, start with Kubernetes (Helm).
Before You Start
You need:
- A machine that can run Docker.
- Docker Compose.
- Network access to pull DockerHub images.
- Host port
8088available.
Choose A Version
The repository root docker-compose.yaml pulls these images by default:
To verify a specific release, set the image tag before starting:
Start
Prepare production settings first:
Edit .env, set SECRET_ENCRYPTION_KEY to a stable random key, and replace the placeholders in BOOTSTRAP_TOKEN and REDIS_PASSWORD. Use URL-safe letters and digits for the Redis password. Compose passes it directly to the built-in Redis server and assembles the complete URI for API and Worker. The complete stack defaults to production mode and does not expose a fixed development administrator.
Run this from the repository root:
This starts PostgreSQL, password-protected Redis, API, and Worker. API completes database migrations first, and Compose starts Worker only after /healthz passes, so Worker cannot access a fresh schema too early. The API image already embeds the web console, so you do not need to start Vite separately. On the first visit, open /bootstrap and use the BOOTSTRAP_TOKEN from .env to create the first administrator, then rotate or remove that one-time token.
To build images from the current source tree:
Open The Console
Visit:
The default Compose stack only exposes API on host port 8088. PostgreSQL and Redis stay inside the container network and do not occupy host ports 5432 and 6379.
Check Services
When API is healthy, the console opens in the browser. Worker must also be healthy for builds, deployments, and status sync to run. If the page opens but tasks never start, check the Worker logs first.
Next
- Open Initialize Console and create or sign in as an administrator.
- Open Connect Cluster and Registry to prepare runtime and image storage.
- Follow Deploy a Web Project to complete the first delivery path.
Stop
To remove data as well: