Skip to main content

Deploy on a Remote Server

This guide walks you through deploying Agenta on a remote server, such as a cloud instance or dedicated machine.

Prerequisites​

Before starting, ensure you have:

  • A remote server with:
    • Docker installed
    • Docker Compose installed
    • Root access or sudo privileges
    • Open ports for HTTP traffic (default: 80)
  • Git installed on the server

Step 1: Clone the Repository​

Clone the Agenta repository and navigate to its directory:

git clone https://github.com/agenta-ai/agenta
cd agenta

Step 2: Configure Environment Variables​

Edit the environment configuration file at /hosting/docker-compose/oss/.env.oss.gh with your server details:

# Core configuration
AGENTA_RUNTIME_PREFIX=
AGENTA_PORT=80                         # The port Agenta will run on
BARE_DOMAIN_NAME=mydomain.ai           # Domain name without protocol
DOMAIN_NAME=http://mydomain.ai         # Domain name with protocol
WEBSITE_DOMAIN_NAME=http://mydomain.ai # Same as DOMAIN_NAME unless web container is hosted elsewhere
AGENTA_HOST=http://mydomain.ai:80      # API domain name
SERVICE_URL_TEMPLATE=http://mydomain.ai:80/services/{path} # URL template for services

You can also customize database connection settings for Redis, PostgreSQL, and RabbitMQ in the same file if needed.

Using an IP address instead of a domain name: If you're using an IP address rather than a domain name, use the same format:

DOMAIN_NAME=http://192.168.1.100
BARE_DOMAIN_NAME=192.168.1.100

Step 3: Build and Deploy​

Run the following command to build and deploy Agenta:

docker compose -f hosting/docker-compose/oss/docker-compose.gh.yml --env-file=hosting/docker-compose/oss/.env.oss.gh --profile with-web up -d --build
info

At the moment, you need to rebuild the web image because the domain name is hardcoded at build time in Next.js. Other services can use the pre-built images from Docker Hub.

Step 4: Verify Your Deployment​

After deployment, verify that Agenta is running correctly:

  1. Check container status:

    docker compose ps

  2. Access the web interface in your browser:

    http://mydomain.ai

    Or if using an IP address:

    http://192.168.1.100

Troubleshooting​

If you encounter issues with your deployment:

  1. Check container logs for errors:

    docker compose logs

    To view logs for a specific service:

    docker compose logs [service-name]

  2. Verify environment variables are set correctly in the .env.oss.gh file

  3. Ensure the server firewall allows traffic on port 80 (or your custom port)

  4. Check that all required Docker services are running:

    docker compose ps

Important Notes​

  • This guide uses HTTP for initial setup. For production environments, you should configure HTTPS.
  • For production deployments, set up HTTPS with SSL/TLS certificates (documentation for this is planned).
  • Resource requirements will vary based on your usage patterns. Monitor system resources after deployment.

Getting Help​

If you need assistance, you can: