| layout | title |
|---|---|
solution |
Pipeline as Code with Jenkins |
The default interaction model with Jenkins, historically, has been very web UI driven, requiring users to manually create jobs, then manually fill in the details through a web browser. This requires additional effort to create and manage jobs to test and build multiple projects, it also keeps the configuration of a job to build/test/deploy separate from the actual code being built/tested/deployed. This prevents users from applying their existing CI/CD best practices to the job configurations themselves.
With the introduction of the plugin:workflow-aggregator[Pipeline plugin],
users now can implement a project’s entire build/test/deploy pipeline
in a Jenkinsfile and store that alongside their code, treating their
pipeline as another piece of code checked into source control.
The Pipeline plugin was inspired by the Build Flow plugin but aims to improve upon some concepts explored by Build Flow with features like:
-
the ability to suspend/resume of executing jobs.
-
checking the pipeline definition into source control (
Jenkinsfile) -
support for extending the domain specific language with additional, organization specific steps, via the "Shared Libraries" feature.
In an adjacent space is the plugin:job-dsl[Job DSL plugin] which is worth mentioning as well.
-
The plugin:workflow-aggregator[Pipeline plugin] has a fairly comprehensive tutorial checked into its source tree too.
Pipeline tutorials you can run locally.
Demonstrations of Pipeline functionality you can run locally.
-
plugin:workflow-aggregator[Pipeline Plugin page]
A presentation by Kohsuke Kawaguchi, Mark Waite, and Tyler Johnson:
The presentations below were given by Jesse Glick, author of the plugin:workflow-aggregator[Pipeline Plugin]
-
2014 presentation (with slides)
Other presentations: