Crate pulseengine_mcp_monitoring

Monitoring, metrics, and observability for MCP servers

This crate provides comprehensive monitoring capabilities for MCP servers including:

• Real-time metrics collection and reporting
• Health checks and system monitoring
• Performance profiling and optimization insights
• InfluxDB integration for time-series data
• Prometheus-compatible metrics export

Quick Start

use pulseengine_mcp_monitoring::{MetricsCollector, MonitoringConfig, ServerMetrics};
use std::time::Duration;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create monitoring configuration
    let config = MonitoringConfig {
        enable_metrics: true,
        metrics_port: 9090,
        health_check_interval: Duration::from_secs(30),
        influxdb_url: Some("http://localhost:8086".to_string()),
        influxdb_database: Some("mcp_metrics".to_string()),
        ..Default::default()
    };

    // Create metrics collector
    let mut collector = MetricsCollector::new(config);
    collector.start().await?;

    // Record metrics in your application
    collector.record_request_duration(Duration::from_millis(50)).await;
    collector.increment_request_count("tool_call").await;
    collector.record_error("connection_timeout").await;

    // Get current metrics snapshot
    let metrics = collector.get_metrics().await;
    println!("Total requests: {}", metrics.request_count);
    println!("Average response time: {:?}", metrics.avg_response_time);

    Ok(())
}

Features

• Real-time metrics: Live request/response time tracking
• Health monitoring: System resource and connectivity checks
• Time-series storage: InfluxDB integration for historical data
• Prometheus export: Industry-standard metrics format
• Performance profiling: Identify bottlenecks and optimization opportunities
• Production ready: Low overhead, highly optimized collection

Re-exports

pub use collector::MetricsCollector;

pub use config::MonitoringConfig;

pub use metrics::ServerMetrics;

Modules

collector

Metrics collector implementation

config

Monitoring configuration

metrics

Server metrics types

Functions

default_config

Default monitoring configuration