---
title: Deployment & Operations
weight: 2
---

**Goal:** Configure, deploy, and monitor a Quetza bot instance.

## 1. Environment Preparation

Quetza relies on environment variables for configuration. **Never** commit these to version control.

| Variable        | Description                                   | Required |
| :-------------- | :-------------------------------------------- | :------- |
| `DISCORD_TOKEN` | Bot token from Discord Developer Portal       | **Yes**  |
| `CLIENT_ID`     | Application ID for slash command registration | **Yes**  |
| `LLAMA_API_KEY` | Required if using the AI module               | No |

## 2. Deployment Strategies

{{< tabs >}}
{{< tab name="Docker (Recommended)" >}}
**Docker** ensures a consistent runtime environment.

1.  **Build** the image:

    ```bash
    docker build -t quetza-bot .
    ```

2.  **Run** the container:
    `bash
docker run -d \
  --env-file .env \
  --name quetza \
  quetza-bot
`
    {{< /tab >}}
    {{< tab name="Source" >}}
    **Source** deployment is useful for development.

3.  **Install** dependencies:

    ```bash
    pnpm install
    ```

4.  **Build** the project:

    ```bash
    pnpm build
    ```

5.  **Start** the bot:
    `bash
pnpm start
`
    {{< /tab >}}
    {{< /tabs >}}

## 3. Operational Monitoring

Once running, use these mechanisms to check health:

- **Logs:** Check standard output (Docker logs) for initialization errors.
- **`/ping` Command:** Verifies network latency and API responsiveness.
- **`/modules` Command:** Lists all loaded modules to confirm feature availability.

> [!WARNING]
> **Scaling Constraint:**
> Quetza follows a single-instance-per-token model. Do not run multiple instances with the same token, as this will cause WebSocket conflicts.

---

**See also:**

- [Docker Documentation](https://docs.docker.com/)
- [Discord Developer Portal](https://discord.com/developers/applications)