Docker Deployment

This guide covers deploying Code Search using a single Docker container. This method is suitable for small deployments or testing.

Caution

For production deployments with high availability, consider using Docker Compose or Kubernetes.

Single Container Deployment

For simple deployments where all components run in one container:

docker run -d \
  --name code-search \
  -p 8080:8080 \
  -p 3000:3000 \
  -v code-search-data:/data \
  -e CS_DATABASE_URL="postgres://user:pass@host:5432/codesearch" \
  -e CS_REDIS_URL="redis://host:6379" \
  ghcr.io/techquestsdev/code-search:latest

Environment Variables

Configure Code Search using environment variables:

Variable	Description	Default
CS_SERVER_ADDR	API server address	:8080
CS_DATABASE_URL	PostgreSQL or MySQL connection string	Required
CS_REDIS_ADDR	Redis address	localhost:6379
CS_ZOEKT_URL	Zoekt server URL	http://localhost:6070
CS_INDEXER_INDEX_PATH	Zoekt index directory	./data/index
CS_INDEXER_REPOS_PATH	Git repositories directory	./data/repos
CS_REPOS_BASE_PATH	Base path for repos	./data/repos

See Environment Variables for the complete list.

Volumes

Mount these volumes for data persistence:

docker run -d \
  --name code-search \
  -v /path/to/data:/data \           # All data
  -v /path/to/config:/config \        # Configuration
  ghcr.io/techquestsdev/code-search:latest

Volume structure:

/data/
├── index/      # Zoekt search indexes
├── repos/      # Cloned Git repositories
└── logs/       # Application logs

External Services

PostgreSQL

Code Search supports PostgreSQL 14+ or MySQL 8+. Example with PostgreSQL:

docker run -d \
  --name code-search-postgres \
  -e POSTGRES_USER=codesearch \
  -e POSTGRES_PASSWORD=secret \
  -e POSTGRES_DB=codesearch \
  -v postgres-data:/var/lib/postgresql/data \
  postgres:16-alpine

Redis

Redis 7+ is required for the job queue:

docker run -d \
  --name code-search-redis \
  -v redis-data:/data \
  redis:7-alpine

Docker Networks

Create a network to connect all services:

# Create network
docker network create code-search-net

# Run services on the network
docker run -d --network code-search-net --name postgres ...
docker run -d --network code-search-net --name redis ...
docker run -d --network code-search-net --name code-search \
  -e CS_DATABASE_URL="postgres://codesearch:secret@postgres:5432/codesearch" \
  -e CS_REDIS_URL="redis://redis:6379" \
  ghcr.io/techquestsdev/code-search:latest

Health Checks

The container includes health check endpoints:

# Liveness
curl http://localhost:8080/health

# Readiness (checks all dependencies)
curl http://localhost:8080/ready

Configure Docker health check:

docker run -d \
  --name code-search \
  --health-cmd="curl -f http://localhost:8080/health || exit 1" \
  --health-interval=30s \
  --health-timeout=10s \
  --health-retries=3 \
  ghcr.io/techquestsdev/code-search:latest

Resource Limits

Set resource limits for production:

docker run -d \
  --name code-search \
  --memory=4g \
  --cpus=2 \
  ghcr.io/techquestsdev/code-search:latest

Recommended minimums:

• Memory: 2GB (4GB+ for large codebases)
• CPU: 2 cores
• Disk: 10GB + 2x your total repository size

Logs

View container logs:

# Follow logs
docker logs -f code-search

# Last 100 lines
docker logs --tail 100 code-search

Updating

To update to a new version:

# Pull new image
docker pull ghcr.io/techquestsdev/code-search:latest

# Stop and remove old container
docker stop code-search
docker rm code-search

# Run new container (with same volumes)
docker run -d \
  --name code-search \
  -v code-search-data:/data \
  ... \
  ghcr.io/techquestsdev/code-search:latest

Next Steps

• Docker Compose - Multi-container deployment
• Configuration - All configuration options
• Code Hosts - Connect to GitHub, GitLab, etc.