mosn/holmes

Health v2 · maintenance only

Adoption not part of the score

1147 stars · 134 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Holmes is a self-aware Go library that automatically collects runtime metrics (goroutine count, RSS memory, CPU usage) and dumps pprof-style profiles when anomalies like spikes are detected. It lets developers capture diagnostic profiles of production incidents without being online when they occur.

Use cases

• automatically dump goroutine profile when goroutine number spikes
• capture cpu profile when cpu load spikes in production
• dump heap profile when memory RSS spikes or OOM risk
• diagnose midnight OOM crashes without being woken up
• change profiling configuration at runtime without restart
• run automatic profiling inside docker or cgroup-limited environments
• send dump events to a custom reporter or monitoring system

When to choose

• you run Go services in production and need automatic profile capture on resource spikes
• you want post-mortem diagnostics for OOM kills or goroutine leaks without manual intervention
• you need lightweight, embeddable profiling that triggers only on anomalies

When to avoid

• you need continuous always-on profiling rather than spike-triggered dumps
• your application is not written in Go
• you need full APM/observability with tracing and metrics dashboards rather than profile dumping

Facets

library · maturity active

monitoring logging developer-tools developer-tools monitoring performance backend go cross-platform profiling go-runtime auto-diagnostics goroutine-dump heap-profile cpu-profile oom-debugging linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="mosn/holmes")