Get started with YAML pipelines

Alibaba Cloud DevOps Flow supports pipeline as code, allowing you to define your pipeline configuration with a YAML file. Generally, a pipeline includes elements such as sources, stages, jobs, and steps. This topic describes the structure of a YAML pipeline and the syntax rules for its nodes.

To help you quickly grasp the structure of a YAML pipeline, the following YAML file provides an example on how to build and deploy a Java application to a host group:

name: demo-pipeline                                       # Display name of the pipeline
sources:                                                  # Configure pipeline sources
  my_repo:                                                # Unique identifier of the pipeline source
    type: codeup                                          # Type of pipeline source
    name: Name of the code source                         # Display name of the code source
    endpoint: <your-codeup-repository-url>                # URL of the code source
    branch: master                                        # Default branch of the code source
    triggerEvents: push                                   # Listener events of the code source
    certificate:                                          # Authentication information
      type: serviceConnection                             # Authentication type
      serviceConnection: <your-service-connection-id>      # Service connection ID
stages:                                       # Configure pipeline stages
  build_stage:                                # Identifier of the stage
    name: Build stage                          # Display name of the stage
    jobs:                                     # Configure pipeline jobs
      build_job:                              # Identifier of the job
        name: Build job                        # Display name of the job
        runsOn: public/cn-beijing             # Cluster environment where the job runs
        steps:                                # Configure pipeline steps
          build_step:                         # Identifier of the step
            step: JavaBuild                   # Select execution step
            name: Java build                   # Display name of the execution step
            with:                             # Parameters of the step
              run: |
                mvn -B clean package -Dmaven.test.skip=true -Dautoconfig.skip
          upload_step:
            step: ArtifactUpload
            name: Upload build output
            with:
              uploadType: flowPublic
              artifact: default
              filePath:
                - target/
                - deploy.sh
  deploy_stage:
    name: Deployment stage
    jobs:
      deploy_job:
        name: Host group deployment job
        component: VMDeploy                # Select execution component
        with:                              # Parameters of the execution component
          artifact: $[stages.build_stage.build_job.upload_step.artifacts.default]
          machineGroup: <your-machine-group-id>
          artifactDownloadPath: /home/admin/app/package.tgz
          executeUser: root
          run: |
            mkdir -p /home/admin/application/
            tar zxvf /home/admin/app/package.tgz -C /home/admin/application/
            sh /home/admin/application/deploy.sh restart

The following table describes the key nodes in the YAML file.

Node	Node description
name	The name of the pipeline.
sources	The pipeline sources. You can configure information such as type, URL, and default branch for each pipeline source.
sources.<souce_id>	The ID of the pipeline source. Each ID uniquely identifies a pipeline source.
sources.<souce_id>.type	The type of pipeline source. The following pipeline source types are supported: codeup, github,and gitlab.
sources.<souce_id>.name	The display name of the pipeline source.
stages	The pipeline stages. A pipeline stage consists of one or more jobs.
stages.<stage_id>	The ID of the pipeline stage. Each ID uniquely identifies a pipeline stage.
stages.<stage_id>.name	The display name of the pipeline stage.
stages.<stage_id>.jobs	The pipeline jobs. A pipeline job can either be a combination of multiple steps that share a workspace and collaborate to complete a designated job, or a call to a component that executes a specified job, with component tasks supporting additional operations like retries and skips.
stages.<stage_id>.jobs.<job_id>	The ID of the pipeline job. Each ID uniquely identifies a pipeline job.
stages.<stage_id>.jobs.<job_id>.name	The display name of the pipeline job.
stages.<stage_id>.jobs.<job_id>.runsOn	The cluster environment where the pipeline job runs. You can use the public cluster environment provided by Alibaba Cloud DevOps or use a private build cluster.
stages.<stage_id>.jobs.<job_id>.steps	The pipeline steps. A pipeline job can consist of one or more steps.
stages.<stage_id>.jobs.<job_id>.steps.<step_id>	The ID of the pipeline step. Each ID uniquely identifies a pipeline step.
stages.<stage_id>.jobs.<job_id>.steps.<step_id>.step	The execution step of the pipeline.
stages.<stage_id>.jobs.<job_id>.steps.<step_id>.name	The display name of the pipeline step.
stages.<stage_id>.jobs.<job_id>.steps.<step_id>.with	The parameters of the pipeline step.
stages.<stage_id>.jobs.<job_id>.component	The pipeline component. A pipeline job can invoke a component to perform related operations.
stages.<stage_id>.jobs.<job_id>.with	The parameters of the pipeline component.

Alibaba Cloud DevOps:YAML pipelines

Get started with YAML pipelines

Syntax rules