What problem does it solve?
Managing Databricks Lakebase Provisioned PostgreSQL instances, handling OAuth token lifecycles, and syncing data between Delta Lake and operational databases is complex and errorāprone. This Skill provides clear patterns and best practices to streamline provisioning, secure connections, and reverseāETL workflows.
Core Features & Use Cases
- Instance Management: Create, list, update, and delete Lakebase Provisioned databases via the Databricks SDK or CLI.
- OAuth Credential Generation: Generate shortālived tokens for secure database connections, with examples for notebooks and production applications.
- Connection Patterns: Simple psycopg scripts, tokenārefreshing connection pools, and environmentāvariable configurations for FastAPI or other services.
- Reverse ETL Sync: Define and manage synced tables from Delta Lake to PostgreSQL using SDK, CLI, or UI, supporting snapshot, triggered, and continuous modes.
- Integration with Databricks Apps and MLflow: Register Lakebase as an app resource and as a model resource for automated credential provisioning.
Quick Start
Use the databricks-lakebase-provisioned skill to create a new Lakebase instance named 'test-db' and obtain its OAuth token.