ow to Enjoy Debugging in Production with Karen Tracey
summary
Argues that debugging in production becomes manageable by design not mystique, citing concrete failures and fixes: XML invalidated by ASCII control chars and membership records linking one user to multiple companies, refactors that allow admin fixes, and making a staging system that mirrors production. Recommends monitoring and alerts like Sentry and AWS alarms, health checks, consolidated Django tracebacks, instrumented Celery tasks, and tests/code review so problems caused by concurrency, data surprises, or differing dev vs prod resources surface early and are fixable.
AI-written summary · about
chapters (5)
- 0:00Introduction
- 1:05Debugging stress in production
- 4:18Common production issues
- 14:10Building debug-friendly projects
- 21:35Conclusion and celebrating launch days
related talks
12mDebugging Python in Production: Practical Techniques Beyond Print Statements
15mDebugging debugging; why we must approach debugging differently
11mWhy we must approach debugging differently
41mDetective Hat: Investigating Production Issues
40mProduction Detectives: Solve the Case Before the Crash
26mThe Mindset of Debugging
44mObservability: Superpowers for Developers
26mDebugging Clojure Services Under Stress