Databricks FAQs

Question 1

What are the key features of this tool?

Accepted Answer

Key features include executing SQL queries, listing schemas and tables within catalogs, describing table schemas, and handling long-running queries through polling.

Question 2

What permissions are required to use this server?

Accepted Answer

The user associated with the Databricks token needs appropriate permissions to access the SQL warehouse, catalogs, schemas, and tables. Use a dedicated token with read-only permissions whenever possible for enhanced security.

Question 3

What is the Databricks MCP Server?

Accepted Answer

It's a Model Context Protocol (MCP) server that allows you to execute SQL queries against Databricks using the Statement Execution API. It simplifies data access and management within Databricks.

Question 4

How do I set up the Databricks MCP Server?

Accepted Answer

First, install the required dependencies using pip or uv. Then, configure your Databricks credentials using a .env file or by setting environment variables (DATABRICKS_HOST, DATABRICKS_TOKEN, DATABRICKS_SQL_WAREHOUSE_ID). Finally, run the main.py file.

Question 5

How does this server handle long-running queries?

Accepted Answer

The server uses a polling mechanism to handle long-running queries. It periodically checks the Databricks API until the query completes or times out, with a default timeout of 10 minutes.

Databricks

About

Key Features

Use Cases