Skip to content
All recipes
Business intelligence

Deploy Metabase

Dashboards and data exploration with its own application state stored in PostgreSQL.

Why use it

Explore data and share dashboards without building a reporting interface.

Build plan

  1. Pin the official Metabase image in the repository Dockerfile.
  2. Use managed PostgreSQL for Metabase application data instead of the embedded H2 database.
  3. Run on port 3000 and use /api/health for the service health check.

Resources and values

Resources
1 ServicePostgreSQL
Required values
MB_DB_TYPEMB_DB_CONNECTION_URI
Check before you deploy

Metabase can reach the database inside the aidrop Project. Raw connections to databases outside aidrop are not available; HTTP-based data sources still work.

Ask your coding agent

Deploy the open-source Metabase image on aidrop with an explicit version tag. Create managed PostgreSQL for Metabase application data, set MB_DB_TYPE=postgres and configure MB_DB_CONNECTION_URI. Run on port 3000 and use /api/health as the health path. Do not use the embedded H2 database. Before deployment, confirm that every analytics data source is either inside this aidrop Project or reachable over HTTPS.

Put Metabase online

Connect your coding agent, paste the quickstart and keep the build, runtime and public address in one aidrop Project.

7 days free · No credit card required