Keep Ops Happy, Designing for Production
summary
Documentation answers three questions: what it does, what it needs, and how to run it; keep deploys as one-button deploys and push config out of code with tools like figaro. Make code thread safe using thread.current or per-request clears to avoid leaked state, prefer bcrypt for passwords (example: a single GPU brute force drops a full 8‑char space from 16 days to 19 hours with 20 GPUs), throttle with rack-attack plus honeypot, and monitor missed jobs with Dead Man Snitch.
AI-written summary · about
chapters (12)
- 0:00Start
- 2:08Documentation
- 5:20Configuration
- 10:08Thread Safety: Thread.current
- 12:04Security: Attributes
- 13:22Security: Passwords
- 15:47Security: Serialization
- 17:19Security: Other Situations
- 17:59Security Automated Scripts
- 19:23Quick Tips: File Storage
- 19:43Quick Tips: Scheduling Jobs
- 20:27Services
related talks
43mGetting Your System into Production and keeping It There
20mDevOps Without The "Ops" – A Fallacy? A Dream? Or Both?
31mFacepalm to Foolproof: Avoiding Common Production Pitfalls
33mOperations: a developer's guide
01h 39mPostOps: A Non-Surgical Tale of Software, Fragility, and Reliability
31mOperating Rails: What About After You Deploy?
46mPracticalities of Productionizing Distributed Systems
24mWar Stories About the Ops in DevOps