From a fragile CI job to a staged Jenkins pipeline
A single-stage pipeline with unexplained failures, rebuilt as build, functional tests, integration tests and deploy across 5 services.
- Role
- Software developer at Fortuna Entertainment Group
- Outcome
- Releases became visible and reliable again across 5 services.
- Stack
- Jenkins
- Java
- JUnit 5
- TestNG
- Cucumber
- Docker
Problem
Five services shared a single-stage CI pipeline that failed without clear reasons. When it broke, nobody could see which step was at fault.
Approach
I rebuilt it as a staged, flag-driven Jenkins pipeline: build → functional tests → integration tests → deploy.
Result
- Each stage reports on its own, so failures point at a step.
- Releases across the 5 services became reliable again.