Alibaba Cloud DevOps:Artifacts

Generate artifacts

Flow pipeline steps, such as image building and artifact uploads, generate the following artifact types:

• Docker image: after an image build job completes, the Docker image is pushed to Alibaba Cloud Container Registry (ACR) or a private image registry. The artifact structure is:

{
  "artifact": "yunxiao-demo",		                                        # Image name
  "type": "dockerImage",    				                        # Artifact type
  "dockerUrl": "registry.cn-beijing.aliyuncs.com/yunxiao/yunxiao-demo:1.0",	# Image address
  "dockerTag": "1.0"								# Image tag
}

• Flow Public Storage artifact: after a build job completes, the build output is uploaded to Flow Public Storage. The artifact structure is:

{
  "artifact": "yunxiao-demo",		# Artifact name
  "type": "flowPublic",			# Artifact type
  "downloadUrl": "https://sdfdff",	# Artifact download address 
  "md5": "dffff"			# Artifact MD5
}

Reference artifacts

Flow provides three ways to reference artifacts, each suited to a different scope:

• Cross-job or cross-stage within the same pipeline: the standard approach for build-then-deploy workflows within a single pipeline, including integrated CI/CD scenarios.

• Cross-pipeline: used when a source pipeline generates artifacts that a target pipeline deploys — for example, a production pipeline that deploys an image validated in a pre-release pipeline. This pattern is common in CI/CD separation scenarios.

• From a pipeline artifact source: used for artifact promotion — for example, after pre-release validation, the artifact is published to an artifact repository, and the production pipeline selects a specific version directly from that repository.

Example 1: Cross-job or cross-stage artifact reference within the same pipeline

The following example shows artifacts flowing between jobs within the same pipeline. The deploy_job depends on build_job completing first and references the Docker image it produces.

sources:
  my_repo:
    ……
stages:
  my_stage:
    name: Build And Deploy
    jobs:
      build_job:
        name: Image Build Job
        runsOn:
          group: public/ap-southeast-1
          container: build-steps-public-registry.ap-southeast-1.cr.aliyuncs.com/build-steps/alinux3:latest
        steps:
          private_registry_docker_setup:
            name: Login Docker Registry
            step: DockerLogin
            with:
              dockerRepository: registry.cn-hangzhou.aliyuncs.com
              serviceConnection: <your-service-connection-id>
          docker_build_push_step:              
            name: Image Build and Push
            step: DockerBuildPush              # DockerBuildPush step will generate a docker image artifact.
            with:
              artifact: my_image
              image: registry.cn-hangzhou.aliyuncs.com/ns/demo:1.0
              dockerfilePath: Dockerfile
              cacheType: remote
      deploy_job:
        name: Kubernetes Deploy Job
        steps:
          kubectl_apply:
            step: KubectlApply
            name: Kubectl Deploy
            with:
              kubernetesCluster: <your-kubernetes-id>
              kubectlVersion: "1.16.4"
              namespace: default
              yamlPath: app-configs/manifest-app
              variables:
                - key: image
                  value: $[jobs.build_job.docker_build_push_step.artifacts.my_image]			   # Reference the artifact generated by the docker_build_step step, which is my-image.

The following example shows artifacts flowing across stages. The deploy_stage references the Flow Public Storage artifact produced by build_stage.

sources:
  my_repo:
    ……
stages:
  build_stage:
    name: Build Stage
    jobs:
      build_job:
        name: Build Job
        runsOn:
          group: public/ap-southeast-1
          container: build-steps-public-registry.ap-southeast-1.cr.aliyuncs.com/build-steps/alinux3:latest
        steps:
          setup_java_step:
            name: Setup Java SDK
            step: SetupJava
            with:
              jdkVersion: "1.8"
              mavenVersion: "3.5.2"
          command_step:
            name: Execute Command
            step: Command
            with:
              run: |
                mvn -B clean package -Dmaven.test.skip=true -Dautoconfig.skip
          upload_artifact_step:
            name: Artifact Upload
            step: ArtifactUpload                 # ArtifactUpload step will generate a artifact.
            with:
              uploadType: flowPublic
              artifact: "Artifacts_${PIPELINE_ID}"
              filePath:
                - target/
  deploy_stage:
    name: Deploy Stage
    jobs:
      deploy_job:
        name: VM Deploy Job
        component: VMDeploy
        with:
          artifact: $[stages.build_stage.build_job.upload_artifact_step.artifacts.default]	 # Reference the artifact generated by the upload_step step, which is 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

Example 2: Cross-pipeline artifact reference

The following example shows Pipeline A generating a Docker image and Pipeline B deploying it. This pattern is commonly used when a pre-release pipeline builds and validates an image, and a separate production pipeline deploys that same image.

• Source Pipeline A

sources:
  my_repo:
    type: gitSample
    name: Java Sample Code Source
    endpoint: https://atomgit.com/flow-example/spring-boot.git
    branch: master
stages:
  build_stage:
    name: Build Stage
    jobs:
      build_job:
        name: Java Image Build Job
        steps:
          private_registry_docker_setup:
            name: Login Docker Registry
            step: DockerLogin
            with:
              dockerRepository: registry.cn-hangzhou.aliyuncs.com
              serviceConnection: <your-service-connection-id>
          docker_build_push_step:
            name: Image Build and Push
            step: DockerBuildPush
            with:
              artifact: my_image         # Source Pipeline A build job will generate a docker image.
              image: registry.cn-hangzhou.aliyuncs.com/ns/demo:1.0
              dockerfilePath: Dockerfile
              cacheType: remote

• Target Pipeline B

sources:
  pipeline_a:
    type: flowPipeline
    name: Source Pipeline A
    flowPipeline: n86bmctmphs84icj       # Source Pipeline A ID 
    build: lastSuccessfulBuild
    triggerEvents: buildSuccess
stages:
  kubectl_apply_stage:
    name: Deploy Stage
    jobs:
      deploy_job:
        name: "Kubectl Apply"
        runsOn:
          group: public/ap-southeast-1
          container: build-steps-public-registry.ap-southeast-1.cr.aliyuncs.com/build-steps/alinux3:latest
        steps:
          kubectl_apply_step:
            name: "Kubectl Apply"
            step: "KubectlApply"
            with:
              kubernetesCluster: <your-kubernetes-id>
              kubectlVersion: "1.27.9"
              namespace: default
              yamlPath: app-configs/manifest-app
              variables:
                - key: image
                  value: $[sources.pipeline_a.stages.build_stage.build_job.docker_build_push_step.artifacts.my_image]   # Reference the image generated by Source Pipeline A, which is my_image.

Reference syntax