Nightmare Before Best Practices
summary
Five real production disasters show concrete causes: a payments script triggers refunds and causes €180,000 loss because an action ID isn't changed; an events site allows unauthorized check‑in and crashes under voting load; swapping photo providers breaks location IDs and aborts the project; an SSH tunnel on a developer laptop halts payments; a fragile newsletter system fails at 250,000 recipients due to ad‑hoc fixes. Fixes recommended: code review, tests, docs, deployment practices, and language‑specific linters.
AI-written summary · about
chapters (10)
- 0:00Start
- 1:32Nightmare Before Best Practices
- 2:35Nightmare #1
- 7:30Nightmare #2
- 10:10Chapter #2
- 10:56Chapter #3
- 12:40Nightmare #3
- 16:44Nightmare #4
- 21:00Nightmare #5
- 30:11Best Practices change across different languages
related talks
28mHalloween Horror Stories from the Trenches
39mPEBCAK
37mFear Driven Development
15mIncident Ready: Chaos Engineer Your Incident Response
47mThreat Modelling Stories From The Trenches
43mIf security is hard, you are doing it wrong
50mLearning in Production (or Why the Apollo 11 Landing Nearly Failed)
44mThe role of catastrophic failure in software design