Datadog MCP Server

🚀 Datadog MCP Server Documentation

This documentation provides a comprehensive guide to the Datadog MCP Server, including its features, tool usage, installation, configuration, and troubleshooting.

✨ Features

• Log Search: Supports querying log data by keywords, time ranges, and other conditions.
• Trace Span Search: Allows filtering trace data based on service names, operations, and timestamps.
• Aggregation Function: Enables statistical analysis of log and trace data, such as calculating the occurrence count of specific events.

📦 Installation and Configuration

Quick Installation

Install globally via npm:

npm install -g @your-org/datadog-mcp-server

Environment Variable Configuration

Set the following environment variables to connect to the Datadog service:

• DD_API_KEY: Required. Datadog API key.
• DD_APP_KEY: Required. Datadog application key.

📚 Documentation

Tool Guide

1. search_logs Tool

Function Description

Used to query and retrieve log data, supporting multiple filtering conditions and sorting options.

Input Parameters

• query: Required. Search keyword or phrase.
• start_time: Optional, default is the current time minus 24 hours. Specify the start time in the format YYYY-MM-DDTHH:MM:SSZ.
• end_time: Optional, default is the current time. Specify the end time in the same format.
• sort_by: Optional, default is sorting by time. Supported fields include timestamp, log_level, and message_length.

Return Results

• logs: A list of matching logs, each containing a timestamp, log level, source, and message content.
• total_count: Total number of matches.
• duration: Search time in seconds.

2. search_spans Tool

Function Description

Used to query and retrieve trace data, supporting filtering based on service names and service operations.

Input Parameters

• service_name: Optional. Specify the service name to filter.
• operation_name: Optional. Specify the operation name to filter.
• start_time: Same as above.
• end_time: Same as above.
• sort_by: Supports timestamp, span_id, and duration.

Return Results

• spans: A list of matching traces, each containing a timestamp, span ID, duration, and related information.
• total_count: Total number of matches.
• duration: Search time in seconds.

3. aggregate Tool

Function Description

Performs statistical aggregation on log or trace data, supporting various grouping and summarization methods.

Input Parameters

• resource: Required. Specify the resource type to aggregate, such as logs or spans.
• group_by: Optional, default is grouping by time granularity (5 minutes). Supports custom fields for grouping.
• aggregations: Optional, default is count. Supports other statistical methods, such as sum, avg, min, and max.

Return Results

• result: A list of aggregated data, each entry containing grouping information and corresponding statistical values.
• stats: Provides global statistical data such as the total number of records and the average value.
• duration: Operation time in seconds.

💻 Usage Examples

Query Logs

search_logs --query="error" --start_time="2023-10-01T00:00:00Z"

Aggregate Trace Data

aggregate --resource=spans --group_by=service_name --aggregations=count,duration_avg

Integration with Different Environments

Integration with Claude Desktop

Add the following configuration to claude_desktop_config.json:

{
  "mcpServers": {
    "datadog": {
      "command": "node",
      "args": [
        "/path/to/datadog-mcp-server/index.js"
      ],
      "env": {
        "DD_API_KEY": "<YOUR_DATADOG_API_KEY>",
        "DD_APP_KEY": "<YOUR_DATADOG_APP_KEY>"
      }
    }
  }
}

Docker Usage

If running with Docker, configure as follows:

{
  "mcpServers": {
    "datadog": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "DD_API_KEY=your_api_key_here",
        "-e",
        "DD_APP_KEY=your_app_key_here",
        "datadog-mcp-server:latest"
      ]
    }
  }
}

VS Code Configuration

Add the following to user settings (JSON format):

{
  "mcpServers": {
    "datadog": {
      "command": "node",
      "args": [
        "/path/to/datadog-mcp-server/index.js"
      ],
      "env": {
        "DD_API_KEY": "<YOUR_DATADOG_API_KEY>",
        "DD_APP_KEY": "<YOUR_DATADOG_APP_KEY>"
      }
    }
  }
}

🔧 Troubleshooting

If you encounter issues, please check the following:

1. Ensure that the environment variables are correctly set.
2. Ensure a normal network connection and the ability to access the Datadog service.
3. Ensure that log and trace data are correctly indexed.

If you have other questions or need further assistance, please contact the technical support team.