Buglyst Blog
Learn to debug under pressure.
Playbooks for fast pattern recognition, guides for the full investigation, and articles for the engineering judgment around the edges.
18 playbooks · 512 guides · 97 articles · 13 linked practice labs ·skip to practice
Fast pattern recognition for the production failures engineers see most often.
1 playbook in Observability & Performance
Service reliability · from symptoms to evidence
Microsoft 365 outage debugging
Separate Outlook symptoms from service impact, understand Graph retry timing, and follow queued email through acceptance and delivery. Read the guides, then practise the failure patterns in two simulated labs.
Structured investigations for the failure modes engineers meet in real systems.
Python Memory Leak Profiling: Diagnosing Unbounded Growth in Production
A practical guide to identifying and fixing Python memory leaks using tracemalloc, objgraph, and heapy, with production-tested strategies.
Python logging not showing output: Why your log messages vanish
A practical guide to diagnosing silent loggers in Python, covering root logger, handler levels, propagation, and common misconfigurations that cause output to disappear.
Go pprof CPU & Memory Profiling: Real-World Debugging Tactics
A hands-on debugging guide for Go pprof CPU and memory profiling with real commands, root causes, and a war story.
Diagnosing and Fixing Java OutOfMemoryError: Java Heap Space
A practical guide to diagnosing Java heap OutOfMemoryError in production, covering non-obvious causes like GC overhead, native memory leaks, and permgen/metaspace issues.
Custom Metrics Not Appearing in AWS CloudWatch: A Debugging Guide
Diagnose why your custom CloudWatch metrics are not showing up. Covers common causes like namespace mismatch, timestamp issues, and IAM permissions.
HTTP Connection Timeout: Debugging Network Stalls and Misconfigured Timeouts
A practical guide to diagnosing HTTP connection timeouts caused by network latency, firewall drops, TCP backlog exhaustion, and misconfigured timeout stack.
Long-form thinking on debugging habits, observability, and the systems around the bug.
Adding Observability to a 500K-Line Monolith Without a Rewrite
Adding structured logging, distributed tracing, and metrics to a legacy monolith without a full rewrite. Real code examples and a war story from a 500K-line codebase.
Distributed Tracing: Following a Single Request Across Microservices
Distributed tracing lets you follow a request as it hops across services. I'll show you how trace context propagates, why sampling matters, and how tracing helped us debug a 5-second latency spike in production.
Structured Logging in JSON: Fields, Schemas, and Pitfalls from Production
A practical guide to designing JSON log schemas that are queryable, consistent, and actually useful in production — with field recommendations and a war story.
Sentry Error Monitoring: A Practical Setup Guide from a Production Outage
A step-by-step setup guide for Sentry error monitoring, covering source maps, release tracking, grouping rules, and alerting — with a real story of a production outage that taught us the hard way.
Reading Prometheus Alert Rules: A Practical Reference for Debugging Firing Alerts
Alert rules look straightforward until you're staring at a firing alert at 3 AM. This post covers the structure, common traps, and how to extract actionable intent from any rule.
Profiling SQL Queries: Finding and Fixing the Slow 5%
Profiling SQL queries is more than running EXPLAIN. This guide covers practical techniques to find the slowest queries, interpret execution plans, and fix them — with real examples from PostgreSQL and MySQL.