opendev
/
system-config
Code Issues Proposed changes
system-config/doc/source/jenkins_jobs.rst

7.8 KiB
Raw Blame History

Jenkins Job Builder

Overview

In order to make the process of managing hundreds of Jenkins Jobs easier a Python based utility was designed to take YAML based configurations and convert those into jobs that are injected into Jenkins. The source for this utility can be found on github and it comes with its own documentation.

The documentation below describes how the OpenStack CI team uses the Jenkins Job Builder in their environment.

Configuring Projects

The YAML scripts to make this work are stored in the openstack-ci-puppet repository in the modules/openstack_project/files/jenkins_job_builder/config/ directory. In this directory you can have four different types of yaml config files:

  • Jenkins Jobs Defaults in defaults.yaml.
  • Jenkins Jobs Macros to give larger config sections meaningful names in macros.yaml.
  • Project specific configurations in project_name.yaml.
  • Job template configurations. Need a projects.yaml file to specify how the templates should be filled out and templates go in template_name.yaml.

YAML Format

Defaults

Example defaults config:

- defaults:
    name: global
    description: |
      <p><b>This job is managed by puppet and will be overwritten.</b></p>

      <p><b>Do not edit this job through the web</b></p>

      <p>If you would like to make changes to this job, please see:

        <a href="https://github.com/openstack/openstack-ci-puppet">
          https://github.com/openstack/openstack-ci-puppet
        </a>

      In modules/openstack_project/files/jenkins_jobs
      </p>
    project-type: freestyle
    concurrent: true

    wrappers:
      - timeout:
          timeout: 30
          fail: true
      - timestamps

    logrotate:
      daysToKeep: 1
      numToKeep: -1
      artifactDaysToKeep: -1
      artifactNumToKeep: -1

This config starts with the - defaults:: line. This specifies that this section contains default values rather than job specifications. In this section we specify a useful set of defaults including a default description indicating Puppet manages these jobs, jobs are allowed to run concurrently, and a thirty minute job timeout.

Macros

Macros exist to give meaningful names to blocks of configuration that can be used in job configs in place of the blocks they name. For example:

- builder:
    name: git-prep
    builders:
      - shell: "/slave_scripts/git-prep.sh"

- builder:
    name: docs
    builders:
      - shell: "/slave_scripts/run-docs.sh"

- publisher:
    name: console-log
    publishers:
      - scp:
          site: 'scp-server'
          files:
            - target: 'logs/$JOB_NAME/$BUILD_NUMBER'
              copy-console: true
              copy-after-failure: true

In this block of code we define two builder macros and one publisher macro. Each macro has a name and using that name in a job config is equivalent to having the yaml below the name in place of the name in the job config. The next section shows how you can use these macros.

Job Config

Example job config:

- job:
    name: example-docs
    node: node-label

    triggers:
      - zuul

    builders:
      - git-prep
      - docs

    publishers:
      - scp:
          site: 'scp-server'
          files:
            - target: 'dir/ectory'
              source: 'build/html/foo'
              keep-hierarchy: true
      - console-log

Each job specification begins with -job:. Under this section you can specify the job details like name, node, etc. Any detail defined in the defaults section that is not defined under this job will be included as well. In addition to attribute details you can also specify how jenkins should perform this job. What trigger methods should be used, the build steps, jenkins publishing steps and so on. The macros defined earlier make this easy and simple.

Job Templates

Job templates allow you to specify a job config once with arguments that are replaced with the values specified in projects.yaml. This allows you to reuse job configs across many projects. First you need a templated job config:

- job-template:
    name: '{name}-docs'

    triggers:
      - zuul

    builders:
      - git-prep
      - docs

    publishers:
      - scp:
          site: 'scp-server'
          files:
            - target: 'dir/ectory'
              source: 'build/html/foo'
              keep-hierarchy: true
      - console-log

    node: '{node}'


- job-group:
    name: python-jobs
    jobs:
      - '{name}-docs'

This takes the previous example-docs job and templatizes it. This will allow us to easily create example1-docs and example2-docs jobs. Each job template begins with - job-template: and the job specification is identical to the previous one, but we have introduced variable arguments. In this case {name} is a variable value that will be replaced. The values for name will be defined in the projects.yaml file.

The - job-group: section is not strictly necessary but allows you to group many job templates with the same variable arguments under one name.

The projects.yaml pulls all of the magic together. It specifies the arguemnts to and instantiates the job templates as real jobs. For example:

- project:
    name: example1
    node: precise

    jobs:
      - python-jobs

- project:
    name: example2
    node: oneiric

    jobs:
      - {name}-docs

Each project using templated jobs should have its own - project: section. Under this sections there should be a jobs: section with a list of job templates or job groups to be used by this project. Other values under the - project: section define the arguments to the templates lised under jobs:. In this case we are giving the docs template name and node values.

Notice that example1 makes use of the job group and example2 makes use of the job template.

Job Caching

The Jenkins Jobs builder maintains a special YAML file in ~/.jenkins_jobs_cache.yml. This contains an MD5 of every generated XML that it builds. If it finds the XML is different then it will proceed to send this to Jenkins, otherwise it is skipped. If a job is accidentally deleted then this file should be modified or removed.

Sending a Job to Jenkins

The Jenkins Jobs builder talks to Jenkins using the Jenkins API. This means that it can create and modify jobs directly without the need to restart or reload the Jenkins server. It also means that Jenkins will verify the XML and cause the Jenkins Jobs builder to fail if there is a problem.

For this to work a configuration file is needed. There is an erb template for this configuration file at modules/jenkins/templates/jenkins_jobs.ini.erb. The contents of this erb are:

[jenkins]
user=<%= username %>
password=<%= password %>
url=<%= url %>

The values for user and url are hardcoded in the Puppet repo in modules/openstack_project/manifests/jenkins.pp, but the password is stored in hiera. Make sure you have it defined as jenkins_jobs_password in the hiera DB.

The password can be obtained by logging into the Jenkins user, clicking on your username in the top-right, clicking on Configure and then Show API Token. This API Token is your password for the API.