Azure SRE Agent Gets Dynatrace Integration via MCP Server

Microsoft has announced official integration between Azure SRE Agent and Dynatrace's hosted MCP (Model Context Protocol) server, enabling AI-powered observability capabilities directly within Azure's site reliability engineering platform. This integration allows SRE teams to query logs, investigate problems, analyze security vulnerabilities, and execute Dynatrace Query Language (DQL) through natural language conversations with their SRE Agent.

What Changed

The integration leverages Dynatrace's hosted MCP server, which uses Server-Sent Events (SSE) transport to provide real-time observability data to Azure SRE Agent. This eliminates the need for complex local server setups while maintaining secure, authenticated access to Dynatrace's comprehensive monitoring platform.

Key capabilities now available through Azure SRE Agent include:

• Natural language to DQL conversion - Generate complex DQL queries from plain English descriptions
• DQL explanation - Get plain English explanations of existing DQL queries
• Problem investigation - Query and analyze Davis Problems (Dynatrace's AI-powered problem detection)
• Security vulnerability analysis - List and assess open security vulnerabilities by risk level
• Timeseries forecasting - Predict future values using statistical models
• General Dynatrace guidance - Ask questions about workflows, alerts, and platform features

Provider Comparison

This integration positions Azure SRE Agent as a more competitive option in the AI-powered SRE market by adding enterprise-grade observability capabilities. While competitors like Datadog and New Relic offer their own AI assistants, the Dynatrace integration provides unique advantages:

DQL Expertise: Dynatrace's query language is specifically designed for observability data, offering more powerful and flexible querying than generic SQL-based approaches used by some competitors.

AI-Powered Problem Detection: Davis Problems leverage Dynatrace's AI engine to automatically detect and prioritize issues, reducing alert fatigue and improving mean time to resolution (MTTR).

Unified Observability: The integration provides a single interface for logs, metrics, traces, and security data, eliminating the need to switch between multiple tools.

Business Impact

For organizations already invested in both Azure and Dynatrace, this integration delivers several tangible benefits:

Reduced Tool Switching: SRE teams can investigate issues and query observability data without leaving the Azure SRE Agent interface, streamlining workflows and reducing context switching.

Faster Problem Resolution: Natural language querying lowers the barrier to entry for complex observability data, enabling junior engineers to perform advanced investigations without extensive DQL knowledge.

Cost Optimization: By providing timeseries forecasting capabilities, teams can better predict resource needs and optimize cloud spending based on historical patterns.

Enhanced Security Posture: Direct access to security vulnerability data within the SRE workflow ensures that security considerations are integrated into operational decision-making.

Implementation Details

The integration requires several configuration steps:

1. Dynatrace Credentials Setup: Create a Platform Token with specific scopes including mcp-gateway:servers:invoke and mcp-gateway:servers:read for SSE transport functionality.

2. MCP Connector Configuration: Add the Dynatrace MCP server connector to Azure SRE Agent using Streamable-HTTP connection type with the Dynatrace environment URL and platform token.

3. Specialized Subagent Creation: Configure a dedicated "DynatraceExpert" subagent with appropriate system prompts and tool access to ensure optimal interaction with Dynatrace data.

4. Testing and Validation: Verify connectivity and test various query types to ensure proper functionality.

Technical Architecture

The integration uses Dynatrace's hosted MCP server architecture, which provides several advantages over traditional API integrations:

• Real-time Data Streaming: SSE transport ensures immediate data delivery without polling overhead
• Standardized Protocol: MCP provides a consistent interface for AI models to interact with external tools
• Scalable Architecture: Hosted server eliminates local infrastructure management
• Secure Authentication: Bearer token-based authentication with granular scope control

Alternative Configuration

For organizations preferring more control or requiring additional capabilities, Dynatrace also offers an open-source MCP server version that uses Stdio transport. This version supports additional features like entity management, workflows, and document creation, but requires Node.js runtime and local hosting.

Conclusion

The Azure SRE Agent and Dynatrace MCP server integration represents a significant step forward in AI-powered observability. By combining Azure's conversational AI capabilities with Dynatrace's comprehensive monitoring platform, organizations can achieve faster problem resolution, reduced operational overhead, and improved system reliability. As AI continues to transform SRE practices, integrations like this will become increasingly essential for maintaining competitive operational excellence.

For detailed setup instructions and troubleshooting guidance, refer to the official Microsoft Community Hub documentation.