A continuous fairytale
summary
A case study of a public driving‑license system that shifts from eight releases a year to deploying by need, often deploy daily, using an Atlas + OpenShift platform that builds Docker app images and a Jenkins pipeline that automates release packaging and deployments. Team rewrites to Spring Boot, adds Flyway DB migrations, Prometheus/Grafana metrics, SonarQube checks and Artifactory artifacts, introduces feature toggles and branch testing, cuts lead time from 14 weeks to 1 week and ships many more smaller issues (37 in one month). Test bottlenecks and legacy technical debt remain.
AI-written summary · about
chapters (14)
- 0:00Intro
- 0:36Background
- 5:05Why did we want this
- 9:35Villains
- 10:59Atlas
- 15:11Challenges
- 19:56Issues
- 26:28Manual testing
- 32:45Delivery graph
- 35:39Key takeaways
- 39:37Whats next
- 41:26Ownership
- 44:07Questions
- 53:48Feature toggles
related talks
34mEnabling a Continuous Deployment Mindset
35mTaming a beast
38mBuilding a high performing Passenger Information System
54mFrom Code to Customer
49mThe road to continuous deployment: a case study
30mOur Continuous Delivery Journey
53mThe Road to Continuous Deployment: a Case Study
39mOur journey to automated releases