Optimizing Your CI: Lessons Learned from a Successful Jenkins Rebuild
summary
Rebuilt a chaotic Jenkins setup into a cloud-native CI that uses Ravello Systems for dynamic environments, Docker-based Jenkins agents via the Jenkins Docker plugin, Artifactory for artifacts, Google SSO plus metrics authorization, and job-config-in-SCM for full revision history. Result: jobs drop from ~137 to ~50–56, builds shrink from ~70 to ~43 minutes, and agent count scales from 0 to 50 dynamically. Also adds automated log collection to S3, a Groovy-updatable global parameter service for cross-job coordination and a sandbox Jenkins cloned from prod for safe changes.
AI-written summary · about
chapters (22)
- 0:00Intro
- 0:23Welcome
- 3:34Jobs left over
- 4:26Resource distribution
- 6:11Docker
- 7:22Security
- 8:31Traveler System
- 9:36Scripts
- 10:01Big Scripts
- 10:46OneSem
- 11:23Job Configuration History
- 12:58Job Data Sharing
- 13:48Jenkins Global Variables
- 15:39How CI works
- 17:57Creating dynamic environment
- 19:09How to present the environment to developers
- 22:16Adding new jobs
- 24:49Dynamic environment
- 25:22Nested Jenkins
- 26:43Developers
- 27:47RiveroWhip
- 31:21Job naming conventions
related talks
35mJenkins Scaling and Organization for an Efficient CI
41mContinuous Integration Tips & Tricks
55mContinuous Integration on Steroids
13mEvolution of containers in continuous integration
01h 01mOpening Keynote
30mContinuous Integration; Tips & Tricks
26mFaster, Cheaper, Leaner: Horizontally Scaling a CI Pipeline
36mCI/CD BoF: Delivering Quality Quickly