DevOps Patterns & Antipatterns for Continuous Updates
summary
Argues for tiny, frequent updates with automated rollbacks, progressive delivery and strong observability, showing concrete failures: instrumentation from New Relic finds 80% still run Java 8 despite six releases, a Google OnHub update reverted routers to factory state without a local rollback, and a trading firm deployment triggers a cascade that lost $440 million in 40 minutes. Recommends save previous images on devices, deploy to small cohorts, monitor logs/traces, and automate rollbacks to shrink blast radius.
AI-written summary · about
chapters (6)
- 0:00Introduction and motivation
- 4:19Java release acceleration
- 8:42Vicious cycle of updates
- 10:53Server update failures
- 13:02Progressive delivery concept
- 14:28Liquid software vision
related talks
30mDevOps Patterns and Antipatterns for Continuous Software Updates
50mDevOps Patterns & Antipatterns for Continuous Software Updates
48mDevOps Patterns and Antipatterns for Continuous Software Updates
27mDevOps Patterns and Antipatterns for Continuous Software Updates
42mPatterns and Antipatterns of Continuous Deployments in DevOps Practice
46mContinuous delivery antipatterns
05mDo this, not that: Antipatterns in Continuous Deployment
40mContinuous Delivery Anti-Patterns
