Continuous Stupidities - Myths of DevOps
summary
Argues that blanket automation often backfires: many pipelines create slower, fragile environments rather than help; recommends test automation first (unit, integration, behavior with Maven/Gradle, Jenkins), storing artifacts in Nexus or Artifactory, and nightly deploys as a full integration test. Claims pull requests are a bottleneck in commercial teams and urges pairing devs with ops, semantic versioning, Docker reuse, and feature flags. Notes rollback pain when DB migrations lack reversible scripts and cites surprising overheads like teams scaling config managers 5→15.
AI-written summary · about
chapters (20)
- 0:00Introduction to DevOps
- 1:07Defining DevOps culture
- 3:59Roles and boundaries in DevOps
- 6:51Standards and automation paradox
- 8:20Test automation necessity
- 10:27Clarifying release terms
- 13:18Continuous integration basics
- 16:28Continuous delivery explained
- 17:53Continuous deployment strategies
- 22:29Pull requests and quality gates
- 25:03Build frequency and duration
- 26:29Nightly deployments importance
- 27:54Release candidates vs production candidates
- 30:19Conway's law and process design
- 33:50Reducing complexity in DevOps
- 34:36Semantic versioning benefits
- 38:08Overhead in configuration management
- 40:31Rollback strategies and risks
- 42:23Prerequisites for DevOps success
- 43:27Q&A and closing remarks
related talks
27mThe curious case of the DevOps engineer and other things ...
55mPatterns & Anti Patterns for Scaling & Enabling DevOps Adoption
32mTales from the DevOps Transformation Trenches
35mMore Buzzwords Won't Help: The Long History of DevOps Failures
54mEffective DevOps for Organizations
40mContinuous Delivery Anti-Patterns
11mDevOps Pitfalls, Beyond The Technology
25mDevOps: Why do we Keep Messing Up?