Skip to content
A collection of scripts, custom steps that can be used in a Jenkins pipeline as a shared library
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

jenkins-util-scripts jj

A collection of scripts, custom steps that can be used in a Jenkins pipeline (declarative and scripted) as a shared library.

Some of the API's provided:

  • Get a list of the failed stages
  • Extract details about test results (junit)
  • Add/Remove environment variabels to a node
  • Delete jobs based on a condition
  • Fetch plotting data from the current job/pipeline
  • Run a script on multiple slaves
  • ...


Add this repository as a shared library in your jenkins instance


See the example for more information.


It is possible to get a list of the failed stages. However this is only possible when the stages use the super_stage wrapper instead of directly using the stage API by jenkins.

try {
    super_stage("stage1") {
        // Do stuff
    super_stage("stage2") {
        error "oei"
} catch(e) {
    println "Failed stages: ${get_failed_stages().join(',')}"

This will print out => Failed stages: stage2


Jenkins default locking API only allows the locking of one resource at the time. This becomes especially an issue if the number of locks is dynamicly programmed. Super_lock is a wrapper arround that API that allows the locking of multiple resources at the same time.

 super_lock(['lock1', 'lock2']) {
     // Do stuff


Running scripts is a fundamental stone of jenkins, however sometimes a script is not critical for a jenkins pipeline but it still can pollute the logging of the pipeline because of the amount of output it prints to stdout. With the run_script API you can prevent this from happening. The run_script is also os independent (dynamic scripted pipelines, anyone?).

 def directories = run_script(script: "ls -w1", quiet: true, returnStdout: true)
 println directories.split("\n").join(",")


Jenkins supports the parsing of junit fail to determine the build result. However the results are not directly available to be used in a various number of reporting plugins. For example if you want to share the test results through e-mail, slack, ...

 junit "test.xml"

 def results = get_junit_results()

 println "FAILED TC:"
 println "=========="

 for (def failed: results.failed) {
     println "Name ${}"
     println "Log: ${failed.log}"
     println "-----"


The plot plugin of jenkins is very minimalistic, it does exactly one thing and one thing only: plotting. However sometimes you would want to plot based on the previous data from the previous builds or you want to calculate the average:

  sh "echo 'job_duration\n${get_job_duration()}'>test.csv"
  plot csvFileName: 'plot-jobduration.csv', group: 'group', style: 'line', title: 'Job duration', 
       csvSeries: [[displayTableFlag: false, exclusionValues: '', file: 'test.csv', inclusionFlag: 'OFF', url: '']]
def plot_data = get_plot_data(plot: "jobduration", serie: "job_duration").collect() { it.toBigDecimal() }
println "Average job duration: ${plot_data.sum()/plot_data.size()}"


See the example for more examples ...


Feel free to fork and add your own pipeline 'hacks'

You can’t perform that action at this time.