Containerized Deployment

Hugr is distributed as Docker containers hosted in the GitHub Container Registry. This page describes how to deploy hugr using Docker and Docker Compose.

Container Images

Hugr provides three official Docker images:

1. Automigrate Image (Recommended for Single-Node Deployments)

Image: ghcr.io/hugr-lab/automigrate

Hugr server with automatic database schema migration for the core database. This image automatically applies schema updates on startup, making updates transparent and seamless. The automigrate image includes the same functionality as the server image, plus automatic schema migration on startup.

Use this image for:

• Single-node (non-clustered) deployments
• Development environments
• Production deployments without clustering

docker pull ghcr.io/hugr-lab/automigrate:latest

2. Server Image (For Cluster Work Nodes)

Image: ghcr.io/hugr-lab/server

The main hugr server that provides the GraphQL API and admin interface. This is the base server image without automatic migrations.

Use this image for:

• Work nodes in clustered deployments
• Custom migration workflows

docker pull ghcr.io/hugr-lab/server:latest

3. Management Image (For Cluster Management Node)

Image: ghcr.io/hugr-lab/management

Management node for cluster mode deployments. This node coordinates schema synchronization, data source configuration, object storage, and authentication settings across multiple work nodes. The management node does not provide a GraphQL API; all queries must go through work nodes. The management node also applies core database migrations at startup if the core database is not in read-only mode.

Use this image for:

• Management node in clustered deployments only

docker pull ghcr.io/hugr-lab/management:latest

Quick Start with Docker

Basic Deployment (Recommended)

Run a basic hugr server with local storage using the automigrate image:

docker run -d \
  --name hugr \
  -p 15000:15000 \
  -v $(pwd)/data:/data \
  -e CORE_DB_PATH=/data/core.db \
  -e ADMIN_UI=true \
  ghcr.io/hugr-lab/automigrate:latest

The automigrate image ensures that database schema migrations are automatically applied on startup, making updates seamless.

Access the admin interface at http://localhost:15000/admin

With Environment File

Create a .env file:

BIND=:15000
ADMIN_UI=true
DEBUG=false
CORE_DB_PATH=/data/core.db
CACHE_L1_ENABLED=true
CACHE_L1_MAX_SIZE=512

Run with environment file:

docker run -d \
  --name hugr \
  -p 15000:15000 \
  -v $(pwd)/data:/data \
  --env-file .env \
  ghcr.io/hugr-lab/automigrate:latest

Docker Compose Deployment

Docker Compose provides a convenient way to manage multi-container deployments. The hugr-lab/docker repository contains several example configurations.

Basic Deployment

Create a docker-compose.yml file using the automigrate image for automatic schema migrations:

version: '3.8'

services:
  hugr:
    image: ghcr.io/hugr-lab/automigrate:latest
    ports:
      - "15000:15000"
    environment:
      - BIND=:15000
      - ADMIN_UI=true
      - CORE_DB_PATH=/data/core.db
      - CACHE_L1_ENABLED=true
      - CACHE_L1_MAX_SIZE=512
    volumes:
      - ./data:/data
      - ./workspace:/workspace
    restart: unless-stopped

Start the service:

docker-compose up -d

Deployment with PostgreSQL

Use PostgreSQL as the core database for production deployments with the automigrate image:

version: '3.8'

services:
  hugr:
    image: ghcr.io/hugr-lab/automigrate:latest
    ports:
      - "15000:15000"
    environment:
      - BIND=:15000
      - ADMIN_UI=true
      - CORE_DB_PATH=postgres://hugr:hugr_password@postgres:5432/hugr_core
      - CACHE_L1_ENABLED=true
      - CACHE_L1_MAX_SIZE=512
    volumes:
      - ./workspace:/workspace
    depends_on:
      - postgres
    restart: unless-stopped

  postgres:
    image: postgres:15-alpine
    environment:
      - POSTGRES_DB=hugr_core
      - POSTGRES_USER=hugr
      - POSTGRES_PASSWORD=hugr_password
    volumes:
      - postgres_data:/var/lib/postgresql/data
    restart: unless-stopped

volumes:
  postgres_data:

Deployment with Caching (Redis)

Add Redis for distributed caching with the automigrate image:

version: '3.8'

services:
  hugr:
    image: ghcr.io/hugr-lab/automigrate:latest
    ports:
      - "15000:15000"
    environment:
      - BIND=:15000
      - ADMIN_UI=true
      - CORE_DB_PATH=/data/core.db

      # L1 Cache
      - CACHE_L1_ENABLED=true
      - CACHE_L1_MAX_SIZE=512

      # L2 Cache (Redis)
      - CACHE_L2_ENABLED=true
      - CACHE_L2_BACKEND=redis
      - CACHE_L2_ADDRESSES=redis:6379
      - CACHE_L2_PASSWORD=redis_password

      - CACHE_TTL=10m
    volumes:
      - ./data:/data
      - ./workspace:/workspace
    depends_on:
      - redis
    restart: unless-stopped

  redis:
    image: redis:7-alpine
    command: redis-server --requirepass redis_password
    ports:
      - "6379:6379"
    volumes:
      - redis_data:/data
    restart: unless-stopped

volumes:
  redis_data:

Deployment with S3 Storage (MinIO)

Use S3-compatible storage for the core database with the automigrate image:

version: '3.8'

services:
  hugr:
    image: ghcr.io/hugr-lab/automigrate:latest
    ports:
      - "15000:15000"
    environment:
      - BIND=:15000
      - ADMIN_UI=true
      - CORE_DB_PATH=s3://hugr-bucket/core.db

      # S3 Configuration (MinIO)
      - CORE_DB_S3_ENDPOINT=http://minio:9000
      - CORE_DB_S3_REGION=us-east-1
      - CORE_DB_S3_KEY=minioadmin
      - CORE_DB_S3_SECRET=minioadmin
      - CORE_DB_S3_USE_SSL=false

      # DuckDB S3 Access
      - DB_ALLOWED_PATHS=s3://hugr-bucket/*
    volumes:
      - ./workspace:/workspace
    depends_on:
      - minio
    restart: unless-stopped

  minio:
    image: minio/minio:latest
    command: server /data --console-address ":9001"
    ports:
      - "9000:9000"
      - "9001:9001"
    environment:
      - MINIO_ROOT_USER=minioadmin
      - MINIO_ROOT_PASSWORD=minioadmin
    volumes:
      - minio_data:/data
    restart: unless-stopped

volumes:
  minio_data:

Complete Production Setup

A comprehensive production-ready setup with all components:

version: '3.8'

services:
  hugr:
    image: ghcr.io/hugr-lab/automigrate:latest
    ports:
      - "15000:15000"
      - "15001:15001"
    environment:
      # Server Configuration
      - BIND=:15000
      - SERVICE_BIND=:15001
      - ADMIN_UI=true
      - DEBUG=false
      - MAX_DEPTH=7
      - ALLOW_PARALLEL=true
      - MAX_PARALLEL_QUERIES=20

      # Core Database (PostgreSQL)
      - CORE_DB_PATH=postgres://hugr:hugr_password@postgres:5432/hugr_core

      # DuckDB Configuration
      - DB_ALLOWED_DIRECTORIES=/workspace
      - DB_MAX_MEMORY=8GB
      - DB_WORKER_THREADS=4

      # L1 Cache
      - CACHE_L1_ENABLED=true
      - CACHE_L1_MAX_SIZE=1024

      # L2 Cache (Redis)
      - CACHE_L2_ENABLED=true
      - CACHE_L2_BACKEND=redis
      - CACHE_L2_ADDRESSES=redis:6379
      - CACHE_L2_PASSWORD=redis_password
      - CACHE_TTL=10m

      # CORS
      - CORS_ALLOWED_ORIGINS=https://app.example.com

      # Authentication
      - ALLOWED_ANONYMOUS=false
      - SECRET_KEY=${SECRET_KEY}
    volumes:
      - ./workspace:/workspace
    depends_on:
      - postgres
      - redis
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:15001/health"]
      interval: 30s
      timeout: 10s
      retries: 3

  postgres:
    image: postgres:15-alpine
    environment:
      - POSTGRES_DB=hugr_core
      - POSTGRES_USER=hugr
      - POSTGRES_PASSWORD=hugr_password
    volumes:
      - postgres_data:/var/lib/postgresql/data
    restart: unless-stopped
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U hugr"]
      interval: 10s
      timeout: 5s
      retries: 5

  redis:
    image: redis:7-alpine
    command: redis-server --requirepass redis_password --maxmemory 512mb --maxmemory-policy allkeys-lru
    volumes:
      - redis_data:/data
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "redis-cli", "--raw", "incr", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5

volumes:
  postgres_data:
  redis_data:

Building Custom Images

If you need to customize the Docker images, you can build them from source. Clone the hugr-lab/docker repository:

git clone https://github.com/hugr-lab/docker.git
cd docker

Build Server Image

docker build -t ghcr.io/hugr-lab/server:latest -f server.dockerfile .

Build Automigrate Image

docker build -t ghcr.io/hugr-lab/automigrate:latest -f automigrate.dockerfile .

Build Management Image

docker build -t ghcr.io/hugr-lab/management:latest -f management.dockerfile .

Using Docker Compose Build

The repository includes a docker-compose.yml file for building:

docker-compose -f example.build.docker-compose.yml build

Image Selection Guide

Choose the right image based on your deployment scenario:

Deployment Scenario	Recommended Image	Notes
Single-node production	automigrate	Automatic migrations on startup
Development environment	automigrate	Simplifies schema updates
Cluster work nodes	server	No automatic migrations needed
Cluster management node	management	Manages cluster operations

Kubernetes Deployment

For Kubernetes deployments, see the Cluster Deployment guide which includes Kubernetes manifests and configuration details.

The hugr-lab/docker repository contains Kubernetes templates in the k8s/cluster directory.

Volume Mounts

Required Volumes

• /data - Core database and local DuckDB files
• /workspace - Schema definitions and data source catalogs

Note: Volume mounting is not strictly required, but provides several benefits:

• Easy editing: Mount /workspace to edit data source schema definition files from the host
• Database access: If using DuckDB for core database, mount /data directory and set the corresponding CORE_DB_PATH to access the database file
• Persistent secrets: Mount a persistent volume to DB_HOME_DIRECTORY to make secrets storage independent and avoid losing credentials when the container is removed

Optional Volumes

• /tmp - Temporary files (configure with DB_TEMP_DIRECTORY)
• /credentials - Stored credentials (configure with DB_HOME_DIRECTORY)

Example with all volumes:

volumes:
  - ./data:/data
  - ./workspace:/workspace
  - ./tmp:/tmp
  - ./credentials:/credentials

Networking

Ports

• 15000 - Main GraphQL API and admin interface (default)
• 15001 - Metrics and health check endpoint (optional, configure with SERVICE_BIND)

Network Configuration

For multi-container deployments, create a custom network:

services:
  hugr:
    # ... service config ...
    networks:
      - hugr-network

  postgres:
    # ... service config ...
    networks:
      - hugr-network

networks:
  hugr-network:
    driver: bridge

Health Checks

Configure health checks for monitoring and orchestration:

healthcheck:
  test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:15000/admin"]
  interval: 30s
  timeout: 10s
  retries: 3
  start_period: 40s

For the service endpoint (when SERVICE_BIND is configured):

healthcheck:
  test: ["CMD", "wget", "--quiet", "--tries=1", "--spider", "http://localhost:15001/health"]
  interval: 30s
  timeout: 10s
  retries: 3

Resource Limits

Configure resource limits to prevent container resource exhaustion:

services:
  hugr:
    # ... other config ...
    deploy:
      resources:
        limits:
          cpus: '4'
          memory: 8G
        reservations:
          cpus: '2'
          memory: 4G

Logging

Configure logging drivers for centralized log management:

services:
  hugr:
    # ... other config ...
    logging:
      driver: "json-file"
      options:
        max-size: "10m"
        max-file: "3"

For structured logging with external systems:

logging:
  driver: "syslog"
  options:
    syslog-address: "tcp://logserver:514"
    tag: "hugr"

Offline Deployment

For environments without internet access, you can save and load images:

Save Images

# Pull images
docker pull ghcr.io/hugr-lab/server:latest
docker pull postgres:15-alpine
docker pull redis:7-alpine

# Save to tar files
docker save ghcr.io/hugr-lab/server:latest -o hugr-server.tar
docker save postgres:15-alpine -o postgres.tar
docker save redis:7-alpine -o redis.tar

Load Images

docker load -i hugr-server.tar
docker load -i postgres.tar
docker load -i redis.tar

Troubleshooting

Container Won't Start

1. Check container logs:

   docker logs hugr

2. Verify environment variables:

   docker exec hugr env

3. Check volume permissions:

   docker exec hugr ls -la /data

Connection Issues

1. Verify port bindings:

   docker ps

2. Test network connectivity:

   docker exec hugr ping postgres

3. Check firewall rules on the host

Performance Issues

1. Monitor resource usage:

   docker stats hugr

2. Review DuckDB memory settings (DB_MAX_MEMORY)

3. Adjust cache configuration

4. Consider scaling to cluster mode

Example Repositories

For more examples and templates, visit:

• Docker Repository: https://github.com/hugr-lab/docker
  • compose/ - Docker Compose examples
  • k8s/ - Kubernetes templates
  • Example configurations for various deployment scenarios

Next Steps

• Learn about Cluster Deployment for high availability
• Review Configuration for detailed environment variables
• Configure Caching for improved performance
• Explore Authentication for securing your deployment