> ## Documentation Index
> Fetch the complete documentation index at: https://docs.knowledgestack.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Docker Deployment

## Overview

This guide covers deploying Knowledge Stack using Docker in a self-hosted environment. All Knowledge Stack services are available as Docker images that you can pull from the container registry.

## Prerequisites

* Docker and Docker Compose installed
* Access to the Knowledge Stack container images
* A configured LLM provider (e.g., OpenAI API key)

## Required Images

A complete Knowledge Stack deployment uses the following images:

| Image          | Description                              |
| -------------- | ---------------------------------------- |
| **API**        | The main Knowledge Stack API server      |
| **Worker**     | Document ingestion and processing worker |
| **Agent**      | AI assistant worker                      |
| **Frontend**   | Web application                          |
| **PostgreSQL** | Database (TimescaleDB recommended)       |
| **Temporal**   | Workflow orchestration                   |
| **MinIO**      | S3-compatible object storage             |
| **LiteLLM**    | LLM gateway proxy                        |
| **Qdrant**     | Vector search engine                     |
| **Redis**      | Real-time notifications and streaming    |
| **Nginx**      | Reverse proxy and TLS termination        |

## Deployment Steps

### 1. Prepare Your Environment

Set up your environment configuration files with the appropriate values for your deployment:

* Database connection strings
* API keys for your LLM provider
* JWT secret keys
* S3/MinIO credentials

### 2. Pull Container Images

Pull all required images for your target version. Each release is tagged with a version number (e.g., `v1.67.0`).

### 3. Start the Stack

Start the production stack with your version tags:

```bash theme={null}
BE_VERSION_TAG=v1.67.0 FE_VERSION_TAG=v1.0.0 docker compose up -d
```

### 4. Run Database Migrations

After the database is running, apply migrations to set up the schema:

```bash theme={null}
# Run via the API container
docker compose exec api alembic upgrade head
```

### 5. Verify the Deployment

* Check that all containers are running: `docker compose ps`
* Verify the API health endpoint: `curl http://localhost:8000/health`
* Access the API documentation at `/api/docs`

## Upgrading

To upgrade to a new version:

1. Pull the new images with the updated version tag
2. Stop the running stack
3. Start the stack with the new version tags
4. Run any pending database migrations

## Air-Gapped Deployments

For environments without internet access, you can transfer images between registries:

1. Pull images from the primary registry on a connected machine
2. Tag images for your local registry
3. Push images to your local registry
4. Deploy from your local registry

Consult your infrastructure team for specific registry configuration.