Dealing with Data Driven Test Combinatorial Complexity using Jenkins Job DSL
summary
Uses Jenkins Job DSL to replace dozens of hand-copied Jenkins jobs with generated, per-configuration jobs: a seed job produces template-backed child jobs so each data combination has its own build, logs and history. Handles about 60 combinations (doubling to 120 jobs for multiple test beds) and generates a multi-job runner, preserves execution history after regeneration, centralizes steps (Python 2.7→3 change in one place) and significantly reduces maintenance and debugging effort.
AI-written summary · about
chapters (20)
- 0:00Start
- 3:06tion: Create a Jenkins Job
- 3:19Extend Our Task a Little
- 4:25Solution 1: Just Copy the Original Job
- 5:29Could Have Possibly Go Wrong?
- 7:02tion 1 Drawbacks
- 8:01Improve It: MultiJob Step with Parameters
- 9:06Job Project Final Configuration
- 9:52to Debug. Unreliable Execution Trend
- 10:56to Update the Configurations
- 11:50to Re-run the Failing Configuration
- 12:30Nutshell...
- 12:40Could Help Us?
- 13:27ducing Jenkins DSL
- 16:18rate a Job from the Seed Job
- 17:16Add Some Parameters
- 19:33a Template Job
- 28:21Little Bit Messy
- 28:58Organize It Better
- 30:03Did We Achieve?
related talks
10mConfiguration as Code: The Job DSL Plugin
42mSeven Habits of Highly Effective Jenkins Users
34mContinuous Delivery with Jenkins Workflow
35mJenkins Scaling and Organization for an Efficient CI
40mBuilding Pipelines at Scale
44mScaling Your Jenkins Jobs
28mHiding in Plain Sight: Unmasking the Jenkins DSL
51mJenkins Workflows with Parallel Steps Boosts Productivity and Quality