duende-observe

Observability integration for the Duende daemon framework.

Overview

This crate provides observability features:

• Renacer integration: Syscall tracing with source correlation
• ttop integration: Real-time resource monitoring via trueno-viz collectors
• Metrics export: Prometheus and OTLP format support

Usage

use duende_observe::{DaemonTracer, DaemonMonitor};

// Attach tracer to daemon
let mut tracer = DaemonTracer::new();
tracer.attach(daemon_pid).await?;

// Collect syscall trace
let report = tracer.collect().await?;
println!("Critical path: {:?}", report.critical_path);

// Monitor daemon resources
let mut monitor = DaemonMonitor::new();
let snapshot = monitor.collect(daemon_pid)?;
println!("CPU: {}%, Memory: {} bytes", snapshot.cpu_percent, snapshot.memory_bytes);

Iron Lotus Framework

• Genchi Genbutsu (現地現物): Direct observation via syscall tracing
• Visual Management (目で見る管理): Real-time metrics dashboards
• Kaizen (改善): Continuous improvement via metrics collection

License

MIT OR Apache-2.0