Ops and Operability
summary
Argues that operability is concrete: add a phone home heartbeat (single UDP packet ~1500 bytes) carrying name, deps, address, interval and basic metrics so monitoring sees new components instantly; traces use correlation IDs and Zipkin/Dapper-style spans; continuous delivery automation cut one team’s deploy time from weeks to 45 minutes; resilience uses Netflix patterns — chaos monkey, chaos gorilla, data‑center kill tests — to make perceived downtime zero; logs become a UI: timestamps, unique IDs and actionable “what happened / who’s affected / how to fix” entries.
AI-written summary · about
chapters (16)
- 0:00Intro
- 1:13Product companies
- 3:52DevOps
- 8:38Lizzie Bennett
- 9:23Automating deployment
- 15:14Phone Home Packet
- 19:23Support and Support Ability
- 21:00Three Magic Questions
- 27:09Warning vs Error
- 27:56Captains Log
- 29:40Log Messages
- 34:00Use and Usability
- 34:39Supportability
- 36:22Release Engineering
- 45:19Logging
- 47:05Testing
related talks
01h 39mPostOps: A Non-Surgical Tale of Software, Fragility, and Reliability
24mOperationalizing Observability: The Invisible Parts
25mThe History of DevOps in 30 minutes
12mHarnessing Chaos Engineering to Build Robust Systems
33mOperations: the last Mile for DevOps
29mObservability; a gentle introduction
53mObservability Driven Development
52mObservability: it's not just an ops thing