Frequently Asked Questions (FAQ) - CData MCP Server for Google BigQuery

Q: What is the CData MCP Server for Google BigQuery? A: It’s a read-only MCP (Model Context Protocol) server that allows Large Language Models (LLMs) like Claude to query live data from Google BigQuery using CData JDBC Drivers.

Q: What is MCP (Model Context Protocol)? A: MCP is an open protocol that standardizes how applications provide context to LLMs, enabling AI models to access and interact with external data sources and tools.

Q: What are the benefits of using the CData MCP Server? A: It enables natural language querying of Google BigQuery data, real-time data analysis, and seamless integration with AI clients like Claude.

Q: Is this a read/write server? A: This particular server is read-only. However, CData offers a free (beta) read/write version available on their website.

Q: What is the CData JDBC Driver for Google BigQuery? A: It’s a driver that exposes Google BigQuery as relational SQL models, allowing the MCP Server to translate natural language queries into SQL queries.

Q: How do I set up the CData MCP Server? A: The setup involves cloning the repository, building the server, installing and licensing the CData JDBC Driver, configuring the connection, creating a .prp file, and configuring your AI client (e.g., Claude Desktop).

Q: Where can I find the CData JDBC Driver for Google BigQuery? A: You can download it from the CData website: https://www.cdata.com/drivers/bigquery/download/jdbc

Q: What is a .prp file? A: It’s a property file that stores the connection details for the JDBC driver, including the prefix, server name, driver path, driver class, JDBC URL, and tables.

Q: How do I configure the connection string? A: Use the Connection String utility provided with the CData JDBC Driver to configure and test the connection to your Google BigQuery instance. Copy the connection string for use in the .prp file.

Q: How do I configure Claude Desktop to use the MCP Server? A: You need to add an entry for the new MCP server in the claude_desktop_config.json file, specifying the path to the Java executable, the JAR file, and the .prp file.

Q: How do I run the MCP Server manually? A: Use the command java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/Salesforce.prp.

Q: What do the servername_get_tables, servername_get_columns, and servername_run_query tools do? A: servername_get_tables retrieves a list of tables. servername_get_columns retrieves a list of columns for a table. servername_run_query executes a SQL SELECT query.

Q: What should I do if I can’t see the CData MCP Server in Claude Desktop? A: Ensure that you have fully quit and restarted the Claude Desktop client.

Q: What should I do if Claude Desktop is unable to retrieve data? A: Double-check that you have configured your connection properly and that the connection string in the .prp file is correct.

Q: Where can I get help with connecting to my data source? A: Contact the CData Support Team.

Q: Where can I get help with using the MCP server? A: Join the CData Community.

Q: What license is the MCP server released under? A: The MCP server is licensed under the MIT License.

Q: How does UBOS enhance the MCP Server experience? A: UBOS is a full-stack AI Agent Development Platform that allows you to orchestrate AI Agents, connect them with your enterprise data, build custom AI Agents with your LLM model and Multi-Agent Systems.

Q: Can I use this MCP server with data sources other than Google BigQuery? A: Yes, CData offers MCP Servers for many other data sources. See the complete table of supported sources.