One Iceberg MCP Server, Seven Catalogs: What It Takes to Reach Each One
This article provides a step by step guide to one MCP server for Apache Iceberg tables, pointed at seven Iceberg REST catalogs in turn. The server offers four read-only tools, and one environment variable decides which catalog it reads. https://github.com/xbill9/lakehouse-iceberg-2026 What is this project trying to Do? An MCP server gives any MCP client, such as Claude Code or another coding…
This comprehensive guide details the process of setting up and testing an Apache Iceberg MCP server that can access seven different catalogs. The server has four read-only tools: iceberg_list_tables, iceberg_describe_table, iceberg_count_rows, and iceberg_scan_table. It communicates with these catalogs using the Iceberg REST protocol, allowing for consistent interaction across various storage systems.
The guide walks through initializing the server with Docker for a local Apache Polaris catalog, to then individually test each of the seven catalog environments, including Google BigLake, Microsoft OneLake, AWS Glue, AWS S3 Tables, and Snowflake Horizon. Each catalog requires specific login methods and storage package installations, such as OAuth2 for Polaris, gcloud token for BigLake, az token for OneLake, and various AWS and Snowflake credentials.
The server processes MCP calls asynchronously and reports results back as newline-delimited JSON-RPC responses. Failures are logged with a "CATALOG ERROR" prefix, enabling easy identification of issues. This project demonstrates the versatility of Iceberg's MCP framework in accessing a mix of managed data storage solutions with a uniform interface.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
