OpenMetadata

OpenMetadata is a comprehensive metadata platform that centralizes data discovery, governance, quality, and collaboration. It provides a web-based UI, a rich set of metadata APIs, a collection of connectors to ingest metadata from 84+ sources, and an ingestion framework to keep your metadata up-to-date. With the OpenMetadata MCP server hosted via Docker, you can quickly spin up the service and begin collecting asset metadata, defining governance policies, and monitoring data quality and lineages through the UI and APIs. The platform supports data discovery across assets like tables, dashboards, pipelines, and services, while enabling data collaboration, data quality tests, and automated lineage visualization. Use the connectors to ingest metadata from your data warehouses, databases, BI tools, messaging systems, and pipeline services, then leverage the UI to organize assets, assign owners, and track data products and domains.

Once running, you can access the OpenMetadata UI to register data sources, configure ingestion pipelines, and manage metadata through the API. The sandbox and local docker deployment options allow you to experiment with sample data and quickly validate your setup. Integrations such as webhooks and role-based access control enable automation and secure collaboration across teams.

Prerequisites:

• Docker installed on your machine (Docker Desktop for Windows/macOS or Docker Engine on Linux)
• Internet access to pull the OpenMetadata image
• Basic familiarity with running containerized applications

Step-by-step installation:

1. Verify Docker is installed

   • macOS/Linux: docker --version
   • Windows: Open a PowerShell/Command Prompt and run docker --version

2. Pull and run the OpenMetadata image (example using latest tag)

   • docker run -d -p 8585:8585 openmetadata/openmetadata:latest
   • This starts the OpenMetadata server in detached mode and exposes port 8585 on localhost.

3. Access the UI

   • Open a browser and navigate to http://localhost:8585
   • Follow on-screen prompts to configure initial settings, data sources, and users.

4. Optional: connect to external databases or upgrade configs

   • If you plan to point OpenMetadata to an external metadata store or database, provide the appropriate database URL, credentials, and environment variables as needed (see environment notes below).

5. For production deployments, consider using docker-compose or Kubernetes for orchestration and persistent storage.

Environment and configuration tips:

• If you run with external databases, set OM_HOST, OM_PORT, and database connection details via environment variables as appropriate for your deployment.
• OpenMetadata supports multiple data connectors; configure connectors in the UI under Ingestion or via APIs to start ingesting metadata from your sources.
• For production security, implement proper authentication/authorization, enable TLS, and consider reverse-proxy termination. Review access controls to assign owners and stakeholders to assets and data products.
• If you encounter port conflicts, change the host port mapping (e.g., -p 8080:8585) accordingly and update OM_PORT.
• Logs are available from docker logs <container_id> to diagnose startup or ingestion issues.
• Backups: regularly back your metadata store and any related configuration to avoid data loss during upgrades.
• If you want to run without a local database, ensure the OpenMetadata image can initialize whether it uses an embedded store or connects to a managed database.