home / mcp / apache hadoop hive mcp server

Apache Hadoop Hive MCP Server

Provides live access to Hive data via MCP tools for querying and data exploration.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "cdatasoftware-apache-hadoop-hive-mcp-server-by-cdata": {
      "command": "java",
      "args": [
        "-jar",
        "/PATH/TO/CDataMCP-jar-with-dependencies.jar",
        "/PATH/TO/apache-hadoop-hive.prp"
      ]
    }
  }
}

You have a read-only MCP Server that exposes live data from Apache Hadoop Hive through the CData JDBC Driver. It lets you ask natural language questions and get up-to-date information without writing SQL. This server is designed for local, read-only access to Hive data and is easy to run alongside your data sources using a single MCP client workflow.

How to use

You connect with an MCP client by loading the hive MCP server configuration and starting the local MCP server. Once running, you can ask questions like “What is the total revenue by region this quarter?” or “Show me all open support tickets.” The server exposes a small set of endpoints for listing tables, listing columns, and running queries. You interact with these capabilities indirectly through the MCP client’s built-in tools, so you do not need to write raw SQL unless you choose to.

How to install

# Prerequisites
# Java must be installed to run the MCP server jar
java -version

# 1) Clone the repository
git clone https://github.com/cdatasoftware/apache-hadoop-hive-mcp-server-by-cdata.git
cd apache-hadoop-hive-mcp-server-by-cdata

# 2) Build the server
mvn clean install
# This creates the JAR file: CDataMCP-jar-with-dependencies.jar

# 3) License the CData JDBC Driver (example path shown; adjust to your install)
# Windows example
# java -jar cdata.jdbc.apachehive.jar --license

# Mac/Linux example
# java -jar cdata.jdbc.apachehive.jar --license

# 4) Configure your connection to the data source using the JDBC Driver
# Open the Connection String utility to create a connection string
java -jar cdata.jdbc.apachehive.jar
# Test the connection, then copy the resulting connection string

# 5) Create a .prp file for the JDBC connection (example values)
# You would replace PATH TO and values with your setup
Prefix=apachehive
ServerName=CDataApacheHive
ServerVersion=1.0
DriverPath=PATH\TO\cdata.jdbc.apachehive.jar
DriverClass=cdata.jdbc.apachehive.ApacheHiveDriver
JdbcUrl=jdbc:apachehive:InitiateOAuth=GETANDREFRESH;
Tables=

Additional setup for Claude Desktop or other MCP clients

To enable Claude Desktop to load this MCP server, create a config file that points to the MCP jar and the .prp file, then place that config where Claude Desktop expects it and reconnect. The exact entries look like this for Windows and Linux/macOS.

# Windows example (claude_desktop_config.json)
{
  "mcpServers": {
    "hivemcp": {
      "command": "PATH\\TO\\java.exe",
      "args": [
        "-jar",
        "PATH\\TO\\CDataMCP-jar-with-dependencies.jar",
        "PATH\\TO\\apache-hadoop-hive.prp"
      ]
    },
    "...": {}
  }
}

Running the server

Start the MCP server from the command line and keep it running on the same machine as your client. The MCP server operates in stdio mode, so the client and server reside on the same host.

java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/apache-hadoop-hive.prp

Usage details and tools

Once configured, you can use the built-in MCP tools to explore and query the data. The server exposes these tool endpoints, named with your configured server name as a prefix (for example, hive_get_tables, hive_get_columns, and hive_run_query). The tools return results in a structured format (tables, columns, and query results) and you typically issue a query through your MCP client rather than calling the tools directly.

Available tools

apachehive_get_tables

Retrieves a list of tables available in the Hive data source. The output is returned in CSV format with the first line containing column headers.

apachehive_get_columns

Retrieves a list of columns for a specific table. The output is returned in CSV format with the first line containing column headers.

apachehive_run_query

Executes a SQL SELECT query against the Hive data source.