⚠️ **Updating from before 30 Mar 2024? See [MIGRATE_PROD.md](./MIGRATE_PROD.md) for instructions.**
Zeppelin's production environment uses Docker. There are a few different ways to run Zeppelin in production:
1.**Standalone**
* The easiest way to get Zeppelin up and running. This setup comes with a built-in database and web server.
2.**Lightweight**
* In case you don't want to use the built-in database and web server. This setup only runs the bot, API, and dashboard themselves. You'll have to provide your own database connection options and set up a proxy server for the API and dashboard.
3.**Manual**
* If you only want to run a specific service, you can use Zeppelin's Dockerfile directly.
## Standalone
### Setup
1. Install Docker on the machine running the bot
2. Make a copy of `.env.example` called `.env`
3. Fill in the missing values in `.env` (including the "PRODUCTION - STANDALONE" section)
**Note:** The dashboard and API are served insecurely over HTTP. It is recommended to set up a proxy with a TLS certificate in front of them. A popular option for this is [Cloudflare Tunnel](https://www.cloudflare.com/products/tunnel/).
1. Build the Zeppelin image: `docker build --tag 'zeppelin' .`
2. Run the service:
* Bot: `docker run zeppelin npm run start-bot`
* API: `docker run zeppelin npm run start-api`
* Dashboard: `docker run zeppelin npm run start-dashboard`
If you're using an application platform such as Railway, you can simply point it to Zeppelin's repository and it should pick up the Dockerfile from there.
For the start command, you can use the same commands as above: `npm run start-bot`, `npm run start-api`, `npm run start-dashboard`.
Make sure to also run migrations when you update the bot.