5.4. Pipeline build

重要

The Pipeline build strategy is deprecated in OpenShift Container Platform 4. Equivalent and improved functionality is present in the OpenShift Container Platform Pipelines based on Tekton.

Jenkins images on OpenShift Container Platform are fully supported and users should follow Jenkins user documentation for defining their jenkinsfile in a job or store it in a Source Control Management system.

The Pipeline build strategy allows developers to define a Jenkins pipeline for use by the Jenkins pipeline plug-in. The build can be started, monitored, and managed by OpenShift Container Platform in the same way as any other build type.

Pipeline workflows are defined in a jenkinsfile, either embedded directly in the build configuration, or supplied in a Git repository and referenced by the build configuration.

5.4.1. Understanding OpenShift Container Platform pipelines

重要

The Pipeline build strategy is deprecated in OpenShift Container Platform 4. Equivalent and improved functionality is present in the OpenShift Container Platform Pipelines based on Tekton.

Jenkins images on OpenShift Container Platform are fully supported and users should follow Jenkins user documentation for defining their jenkinsfile in a job or store it in a Source Control Management system.

Pipelines give you control over building, deploying, and promoting your applications on OpenShift Container Platform. Using a combination of the Jenkins Pipeline build strategy, jenkinsfiles, and the OpenShift Container Platform Domain Specific Language (DSL) provided by the Jenkins Client Plug-in, you can create advanced build, test, deploy, and promote pipelines for any scenario.

OpenShift Container Platform Jenkins Sync Plugin

The OpenShift Container Platform Jenkins Sync Plugin keeps the build configuration and build objects in sync with Jenkins jobs and builds, and provides the following:

  • Dynamic job and run creation in Jenkins.
  • Dynamic creation of agent pod templates from image streams, image stream tags, or config maps.
  • Injecting of environment variables.
  • Pipeline visualization in the OpenShift Container Platform web console.
  • Integration with the Jenkins git plugin, which passes commit information from
  • Synchronizing secrets into Jenkins credential entries OpenShift Container Platform builds to the Jenkins git plugin.

OpenShift Container Platform Jenkins Client Plugin

The OpenShift Container Platform Jenkins Client Plugin is a Jenkins plugin which aims to provide a readable, concise, comprehensive, and fluent Jenkins Pipeline syntax for rich interactions with an OpenShift Container Platform API Server. The plugin uses the OpenShift Container Platform command line tool, oc, which must be available on the nodes executing the script.

The Jenkins Client Plug-in must be installed on your Jenkins master so the OpenShift Container Platform DSL will be available to use within the jenkinsfile for your application. This plug-in is installed and enabled by default when using the OpenShift Container Platform Jenkins image.

For OpenShift Container Platform Pipelines within your project, you will must use the Jenkins Pipeline Build Strategy. This strategy defaults to using a jenkinsfile at the root of your source repository, but also provides the following configuration options:

  • An inline jenkinsfile field within your build configuration.
  • A jenkinsfilePath field within your build configuration that references the location of the jenkinsfile to use relative to the source contextDir.
注意

The optional jenkinsfilePath field specifies the name of the file to use, relative to the source contextDir. If contextDir is omitted, it defaults to the root of the repository. If jenkinsfilePath is omitted, it defaults to jenkinsfile.

5.4.2. Providing the Jenkins file for pipeline builds

重要

The Pipeline build strategy is deprecated in OpenShift Container Platform 4. Equivalent and improved functionality is present in the OpenShift Container Platform Pipelines based on Tekton.

Jenkins images on OpenShift Container Platform are fully supported and users should follow Jenkins user documentation for defining their jenkinsfile in a job or store it in a Source Control Management system.

The jenkinsfile uses the standard groovy language syntax to allow fine grained control over the configuration, build, and deployment of your application.

You can supply the jenkinsfile in one of the following ways:

  • A file located within your source code repository.
  • Embedded as part of your build configuration using the jenkinsfile field.

When using the first option, the jenkinsfile must be included in your applications source code repository at one of the following locations:

  • A file named jenkinsfile at the root of your repository.
  • A file named jenkinsfile at the root of the source contextDir of your repository.
  • A file name specified via the jenkinsfilePath field of the JenkinsPipelineStrategy section of your BuildConfig, which is relative to the source contextDir if supplied, otherwise it defaults to the root of the repository.

The jenkinsfile is run on the Jenkins agent pod, which must have the OpenShift Container Platform client binaries available if you intend to use the OpenShift Container Platform DSL.

Procedure

To provide the Jenkins file, you can either:

  • Embed the Jenkins file in the build configuration.
  • Include in the build configuration a reference to the Git repository that contains the Jenkins file.

Embedded Definition

kind: "BuildConfig"
apiVersion: "v1"
metadata:
  name: "sample-pipeline"
spec:
  strategy:
    jenkinsPipelineStrategy:
      jenkinsfile: |-
        node('agent') {
          stage 'build'
          openshiftBuild(buildConfig: 'ruby-sample-build', showBuildLogs: 'true')
          stage 'deploy'
          openshiftDeploy(deploymentConfig: 'frontend')
        }

Reference to Git Repository

kind: "BuildConfig"
apiVersion: "v1"
metadata:
  name: "sample-pipeline"
spec:
  source:
    git:
      uri: "https://github.com/openshift/ruby-hello-world"
  strategy:
    jenkinsPipelineStrategy:
      jenkinsfilePath: some/repo/dir/filename 1

1
The optional jenkinsfilePath field specifies the name of the file to use, relative to the source contextDir. If contextDir is omitted, it defaults to the root of the repository. If jenkinsfilePath is omitted, it defaults to jenkinsfile.

5.4.3. Using environment variables for pipeline builds

重要

The Pipeline build strategy is deprecated in OpenShift Container Platform 4. Equivalent and improved functionality is present in the OpenShift Container Platform Pipelines based on Tekton.

Jenkins images on OpenShift Container Platform are fully supported and users should follow Jenkins user documentation for defining their jenkinsfile in a job or store it in a Source Control Management system.

To make environment variables available to the Pipeline build process, you can add environment variables to the jenkinsPipelineStrategy definition of the build configuration.

Once defined, the environment variables will be set as parameters for any Jenkins job associated with the build configuration.

Procedure

  • To define environment variables to be used during build, edit the YAML file:

    jenkinsPipelineStrategy:
    ...
      env:
        - name: "FOO"
          value: "BAR"

You can also manage environment variables defined in the build configuration with the oc set env command.

5.4.3.1. Mapping between BuildConfig environment variables and Jenkins job parameters

When a Jenkins job is created or updated based on changes to a Pipeline strategy build configuration, any environment variables in the build configuration are mapped to Jenkins job parameters definitions, where the default values for the Jenkins job parameters definitions are the current values of the associated environment variables.

After the Jenkins job’s initial creation, you can still add additional parameters to the job from the Jenkins console. The parameter names differ from the names of the environment variables in the build configuration. The parameters are honored when builds are started for those Jenkins jobs.

How you start builds for the Jenkins job dictates how the parameters are set.

  • If you start with oc start-build, the values of the environment variables in the build configuration are the parameters set for the corresponding job instance. Any changes you make to the parameters' default values from the Jenkins console are ignored. The build configuration values take precedence.
  • If you start with oc start-build -e, the values for the environment variables specified in the -e option take precedence.

    • If you specify an environment variable not listed in the build configuration, they will be added as a Jenkins job parameter definitions.
    • Any changes you make from the Jenkins console to the parameters corresponding to the environment variables are ignored. The build configuration and what you specify with oc start-build -e takes precedence.
  • If you start the Jenkins job with the Jenkins console, then you can control the setting of the parameters with the Jenkins console as part of starting a build for the job.
注意

It is recommended that you specify in the build configuration all possible environment variables to be associated with job parameters. Doing so reduces disk I/O and improves performance during Jenkins processing.

5.4.4. Pipeline build tutorial

重要

The Pipeline build strategy is deprecated in OpenShift Container Platform 4. Equivalent and improved functionality is present in the OpenShift Container Platform Pipelines based on Tekton.

Jenkins images on OpenShift Container Platform are fully supported and users should follow Jenkins user documentation for defining their jenkinsfile in a job or store it in a Source Control Management system.

This example demonstrates how to create an OpenShift Container Platform Pipeline that will build, deploy, and verify a Node.js/MongoDB application using the nodejs-mongodb.json template.

Procedure

  1. Create the Jenkins master:

      $ oc project <project_name>

    Select the project that you want to use or create a new project with oc new-project <project_name>.

      $ oc new-app jenkins-ephemeral 1

    If you want to use persistent storage, use jenkins-persistent instead.

  2. Create a file named nodejs-sample-pipeline.yaml with the following content:

    注意

    This creates a BuildConfig object that employs the Jenkins pipeline strategy to build, deploy, and scale the Node.js/MongoDB example application.

    kind: "BuildConfig"
    apiVersion: "v1"
    metadata:
      name: "nodejs-sample-pipeline"
    spec:
      strategy:
        jenkinsPipelineStrategy:
          jenkinsfile: <pipeline content from below>
        type: JenkinsPipeline
  3. Once you create a BuildConfig object with a jenkinsPipelineStrategy, tell the pipeline what to do by using an inline jenkinsfile:

    注意

    This example does not set up a Git repository for the application.

    The following jenkinsfile content is written in Groovy using the OpenShift Container Platform DSL. For this example, include inline content in the BuildConfig object using the YAML Literal Style, though including a jenkinsfile in your source repository is the preferred method.

    def templatePath = 'https://raw.githubusercontent.com/openshift/nodejs-ex/master/openshift/templates/nodejs-mongodb.json' 1
    def templateName = 'nodejs-mongodb-example' 2
    pipeline {
      agent {
        node {
          label 'nodejs' 3
        }
      }
      options {
        timeout(time: 20, unit: 'MINUTES') 4
      }
      stages {
        stage('preamble') {
            steps {
                script {
                    openshift.withCluster() {
                        openshift.withProject() {
                            echo "Using project: ${openshift.project()}"
                        }
                    }
                }
            }
        }
        stage('cleanup') {
          steps {
            script {
                openshift.withCluster() {
                    openshift.withProject() {
                      openshift.selector("all", [ template : templateName ]).delete() 5
                      if (openshift.selector("secrets", templateName).exists()) { 6
                        openshift.selector("secrets", templateName).delete()
                      }
                    }
                }
            }
          }
        }
        stage('create') {
          steps {
            script {
                openshift.withCluster() {
                    openshift.withProject() {
                      openshift.newApp(templatePath) 7
                    }
                }
            }
          }
        }
        stage('build') {
          steps {
            script {
                openshift.withCluster() {
                    openshift.withProject() {
                      def builds = openshift.selector("bc", templateName).related('builds')
                      timeout(5) { 8
                        builds.untilEach(1) {
                          return (it.object().status.phase == "Complete")
                        }
                      }
                    }
                }
            }
          }
        }
        stage('deploy') {
          steps {
            script {
                openshift.withCluster() {
                    openshift.withProject() {
                      def rm = openshift.selector("dc", templateName).rollout()
                      timeout(5) { 9
                        openshift.selector("dc", templateName).related('pods').untilEach(1) {
                          return (it.object().status.phase == "Running")
                        }
                      }
                    }
                }
            }
          }
        }
        stage('tag') {
          steps {
            script {
                openshift.withCluster() {
                    openshift.withProject() {
                      openshift.tag("${templateName}:latest", "${templateName}-staging:latest") 10
                    }
                }
            }
          }
        }
      }
    }
    1
    Path of the template to use.
    1 2
    Name of the template that will be created.
    3
    Spin up a node.js agent pod on which to run this build.
    4
    Set a timeout of 20 minutes for this pipeline.
    5
    Delete everything with this template label.
    6
    Delete any secrets with this template label.
    7
    Create a new application from the templatePath.
    8
    Wait up to five minutes for the build to complete.
    9
    Wait up to five minutes for the deployment to complete.
    10
    If everything else succeeded, tag the $ {templateName}:latest image as $ {templateName}-staging:latest. A pipeline build configuration for the staging environment can watch for the $ {templateName}-staging:latest image to change and then deploy it to the staging environment.
    注意

    The previous example was written using the declarative pipeline style, but the older scripted pipeline style is also supported.

  4. Create the Pipeline BuildConfig in your OpenShift Container Platform cluster:

    $ oc create -f nodejs-sample-pipeline.yaml
    1. If you do not want to create your own file, you can use the sample from the Origin repository by running:

      $ oc create -f https://raw.githubusercontent.com/openshift/origin/master/examples/jenkins/pipeline/nodejs-sample-pipeline.yaml
  5. Start the Pipeline:

    $ oc start-build nodejs-sample-pipeline
    注意

    Alternatively, you can start your pipeline with the OpenShift Container Platform web console by navigating to the Builds → Pipeline section and clicking Start Pipeline, or by visiting the Jenkins Console, navigating to the Pipeline that you created, and clicking Build Now.

    Once the pipeline is started, you should see the following actions performed within your project:

    • A job instance is created on the Jenkins server.
    • An agent pod is launched, if your pipeline requires one.
    • The pipeline runs on the agent pod, or the master if no agent is required.

      • Any previously created resources with the template=nodejs-mongodb-example label will be deleted.
      • A new application, and all of its associated resources, will be created from the nodejs-mongodb-example template.
      • A build will be started using the nodejs-mongodb-example BuildConfig.

        • The pipeline will wait until the build has completed to trigger the next stage.
      • A deployment will be started using the nodejs-mongodb-example deployment configuration.

        • The pipeline will wait until the deployment has completed to trigger the next stage.
      • If the build and deploy are successful, the nodejs-mongodb-example:latest image will be tagged as nodejs-mongodb-example:stage.
    • The agent pod is deleted, if one was required for the pipeline.

      注意

      The best way to visualize the pipeline execution is by viewing it in the OpenShift Container Platform web console. You can view your pipelines by logging in to the web console and navigating to Builds → Pipelines.