As a DevOps administrator or developer, you might want to create a pipeline in Automation Pipelines by using YAML code, instead of using the user interface. When you create pipelines as code, you can use any editor and insert comments in the pipeline code.

In your pipeline code, you can refer to external configurations such as environment variables and security credentials. When you update variables that you use in your pipeline code, you can update them without having to update the pipeline code.

You can use the pipeline YAML code as a template to clone and create other pipelines, and share the templates with others.

You can store your pipeline code templates in a source control repository, which versions them and tracks updates. By using a source control system, you can easily back up your pipeline code, and restore it if needed.

Prerequisites

  • Verify that you have a code editor.
  • If you plan to store your pipeline code in a source control repository, verify that you can access a working instance.

Procedure

  1. In your code editor, create a file.
  2. Copy and paste the sample pipeline code, and update it to reflect your specific pipeline needs.
  3. To include an endpoint to your pipeline code, copy and paste the example endpoint code, and update it to reflect your endpoint.

    When using a Kubernetes API endpoint in the pipeline workspace, Automation Pipelines creates the necessary Kubernetes resources such as ConfigMap, Secret, and Pod to run the continuous integration (CI) task or custom task. Automation Pipelines communicates with the container by using the NodePort.

    The Automation Pipelines pipeline workspace supports Docker and Kubernetes for continuous integration tasks and custom tasks.

    For more information about configuring the workspace, see Configuring the Pipeline Workspace.

  4. Save the code.
  5. To store and version your pipeline code, check the code into your source control repository.
  6. When you create a continuous integration and delivery pipeline, you must import the Kubernetes YAML file.
    To import the Kubernetes YAML file, select it in the Continuous Delivery area of the smart pipeline template, and click Process. Or, use the API.

Results

By using the code examples, you created the YAML code that represents your pipeline and endpoints.

Example: Example YAML code for a pipeline and endpoints

This example YAML code includes sections that represent the workspace for the Automation Pipelines native build, stages, tasks, notifications, and more in a pipeline.

For examples of code for supported plug-ins, see Connecting Automation Pipelines to endpoints 

---
kind: PIPELINE
name: myPipelineName
tags:
  - tag1
  - tag2
 
 
# Ready for execution
enabled: false
 
#Max number of concurrent executions
concurrency: 10
 
#Input Properties
input:
  input1: '30'
  input2: 'Hello'
 
#Output Properties
output:
  BuildNo: '${Dev.task1.buildNo}'
  Image: '${Dev.task1.image}'
 
#Workspace Definition
ciWorkspace:
  image: docker:maven-latest
  path: /var/tmp
  endpoint: my-k8s
  cache:
    - ~/.m2
 
# Starred Properties
starred:
  input: input1
  output: output1
   
# Stages in order of execution
stageOrder:
  - Dev
  - QA
  - Prod
 
# Task Definition Section
stages:
  Dev:
    taskOrder:
      - Task1, Task6
      - Task2 Long, Task Long Long
      - Task5
    tasks:
      Task1:
        type: jenkins
        ignoreFailure: false
        preCondition: ''
        endpoints:
          jenkinsServer: myJenkins
        input:
          job: Add Two Numbers
          parameters:
            number1: 10
            number2: 20
      Task2:
        type: blah
        # repeats like Task1 above
  QA:
    taskOrder:
      - TaskA
      - TaskB
    tasks:
      TaskA:
        type: ssh
        ignoreFailure: false
        preCondition: ''
        input:
          host: x.y.z.w
          username: abcd
          password: ${var.mypassword}
          script: >
            echo "Hello, remote server"
      TaskB:
        type: blah
        # repeats like TaskA above
 
# Notificatons Section
notifications:
  email:
    - stage: Dev #optional ; if not found - use pipeline scope
      task: Task1 #optional; if not found use stage scope
      event: SUCCESS
      endpoint: default
      to:
         - [email protected]
         - [email protected]
      subject: 'Pipeline ${name} has completed successfully'
      body: 'Pipeline ${name} has completed successfully'
       
  jira:
    - stage: QA #optional ; if not found - use pipeline scope
      task: TaskA #optional; if not found use stage scope
      event: FAILURE
      endpoint: myJiraServer
      issuetype: Bug
      project: Test
      assignee: abc
      summary: 'Pipeline ${name} has failed'
      description: |-
        Pipeline ${name} has failed
        Reason - ${resultsText}
  webhook:
    - stage: QA #optional ; if not found - use pipeline scope
      task: TaskB #optional; if not found use stage scope
      event: FAILURE
      agent: my-remote-agent
      url: 'http://www.abc.com'
      headers: #requestHeaders: '{"build_no":"123","header2":"456"}'
          Content-Type: application/json
          Accept: application/json
      payload: |-
        Pipeline ${name} has failed
        Reason - ${resultsJson}
---

This YAML code represents an example Jenkins endpoint. 

---
name: My-Jenkins
tags:
- My-Jenkins
- Jenkins
kind: ENDPOINT
properties:
  offline: true
  pollInterval: 15.0
  retryWaitSeconds: 60.0
  retryCount: 5.0
  url: http://urlname.yourcompany.com:8080
description: Jenkins test server
type: your.jenkins:JenkinsServer
isLocked: false
---

This YAML code represents an example Kubernetes endpoint. 

---
name: my-k8s
tags: [
  ]
kind: ENDPOINT
properties:
  kubernetesURL: https://urlname.examplelocation.amazonaws.com
  userName: admin
  password: encryptedpassword
description: ''
type: kubernetes:KubernetesServer
isLocked: false
---

What to do next

Run your pipeline, and make any adjustments as needed. See How do I run a pipeline and see results.