#Pprof

You deploy a Go service that handles 200 req/s in staging just fine. In production, latency climbs...

A parked goroutine burns zero CPU, so flame graphs miss leaks entirely. Read goroutine.pprof, set a NumGoroutine baseline, and gate CI with goleak.

From Dev.to - microservices: Step-by-Step: Optimize a Go 1.24 Microservice for 1M TPS Using pprof 1.24 and Benchstat 1.24

From Dev.to - rust: War Story: How a Rust 1.85 Memory Leak in Production Cost Us $50k in Cloud Costs Before We Debugged It with pprof