Overview

Metaflow is a workflow framework for data science and machine-learning projects. This DockerCP template deploys the published Metaflow metadata service container, exposes its API endpoint, and prepares the database connection variables used by a production metadata service. A PostgreSQL database is required for durable metadata storage and can be managed as a separate DockerCP service.

Docker image netflixoss/metaflow_metadata_service:latest
Default port 8080
Deployment type docker image
Recommended resources 2 CPU · 2 GB

How deployment works in DockerCP

This template gives DockerCP the deployment metadata needed to create a service, prepare runtime variables, map persistent storage, expose the internal port, and connect a public domain with SSL when required.

  1. Create or select a project where the service belongs
  2. Review the Docker image, tag, default port, environment variables, and volume mappings
  3. Deploy the service and check logs until the application is healthy
  4. Attach a domain or subdomain when the application must be available publicly
  5. Enable HTTPS and verify the final browser response

Configuration

Environment variables

Variable Default / guidance
MF_METADATA_DB_HOST postgres
MF_METADATA_DB_PORT 5432
MF_METADATA_DB_USER metaflow
MF_METADATA_DB_PSWD change_me
MF_METADATA_DB_NAME metaflow

Persistent volumes

No persistent volume is required by the default template profile

Commands

DockerCP handles deployment from the panel, but understanding the equivalent Docker commands helps users debug services with more confidence. 

docker pull netflixoss/metaflow_metadata_service:latest
docker ps --filter "name=metaflow"

docker logs --tail=100 metaflow

DNS and SSL checklist

Step Expected result
Add A or CNAME record The domain points to the DockerCP server or configured target
Attach domain to service The service is reachable through the selected hostname
Enable HTTPS The certificate becomes active and redirect to HTTPS can be enabled

Screenshots

Panel previews for the Metaflow template: how deployment, the running service, and its key details look inside DockerCP.

Metaflow deployment form inside the DockerCP panel
Public URL served through DockerCP domain routing and SSL
Running service with domain and database state

Troubleshoot

  • Confirm the Docker image and tag exist
  • Confirm the selected internal port matches the application port
  • Confirm required variables such as database URL, app URL, and credentials are set
  • Check service logs after the first deployment
  • Verify DNS before requesting or renewing SSL certificates