Pipelines form a single application or a module. You can model a pipeline using multiple stages, and each stage can include multiple tasks. Each stage represents a part of the release cycle for the application or module. You can create, edit, view, copy, and delete the release pipeline.

About this task

Plug-ins can use binding variables as part of input configurations. These variables are replaced at runtime with the values where they are placed. The binding variable always precedes the shared pipeline variable. For example, ${pipeline.PipelineVariable}.


  • Verify that the following items are available to you: predefined vRealize Automation blueprints, workflows, scripts, or test jobs that perform tasks that the pipelines trigger.

  • Verify that the artifacts in the Artifactory server repository are already populated so that you can use the artifact management capabilities.

  • Verify that the applicable plug-ins and endpoints are registered. See the Installation and Configuration guide.

  • Familiarize yourself with the input and output details required to create tasks.


  1. Select Code Stream > Pipelines.
  2. Click Add to create a pipeline.
  3. Enter a name and description for the pipeline.
  4. To define input properties for the pipeline, click Add.

    These properties are only required to pass certain parameters at the time of triggering a pipeline run. You can reference these parameter values across all stages during modeling. The run time values are applied during the pipeline run.




    Enter a property name.


    Enter a description for the property.


    Enter a property value. For example, ABC-876541.

  5. (Optional) To run this pipeline concurrently, select the check box named Allow concurrent pipeline execution.

    The pipeline model reuses the same set of virtual machines to deploy a software change. The concurrent run overwrites the deployed change.

  6. (Optional) To send notifications about the pipeline execution to Socialcast, select the check box named Post updates to Socialcast. Then, enter the Socialcast Webhook Name and Socialcast Webhook URL.

    Socialcast updates are useful to collaborate among team members who work on the releases. The updates notify users about the pipeline execution status such as the pipeline execution start, pause, resume, cancelation, failure, and completion. A new pipeline execution creates a status message on the Socialcast group, and posts the subsequent statuses of the execution as comments to the message. You can add comments to individual threads, and track the progress of a pipeline execution.

    The name of the Webhook you provide is the name of the group created for communication in Socialcast.

    To locate the Webhook URL, navigate to Socialcast > Configured Group > Integrations > Codestream > Add > Step 5.


    To add the integration and copy the Webhook URL, you must have administrator privileges for the group configured on Socialcast.

  7. To send event notifications to users during the pipeline execution, add the email addresses of the recipients.

    The users whose emails you enter must have access to the same vRealize Automation appliance.

  8. Enter the applicable tags.

    A tag is useful in grouping pipeline models or runs.

    For example, you can use a tag to filter a pipeline model or run view.

  9. Click Stages > to continue pipeline creation.
  10. Click Add Stage to add multiple stages to the pipeline.
  11. Double-click the default stage name and enter a name.

    For example, you can create stages named Development, Test, QE, and Production.

  12. (Optional) Select a stage and drag it to a different place in the pipeline.

    For example, if your pipeline consists of stages named Development, Test, and QE, you can move the stage named Test after the stage named QE.

  13. Click Save as Draft to save the pipeline.
  14. Create tasks for every stage in the pipeline.

    You can add multiple tasks to a stage. You can model tasks in a stage to run in parallel, and drag those tasks to run in parallel. Running a parallel task is limited only within a stage.

    The Artifact, Custom, Deploy, Provision, and Test tasks are supported. Certain tasks might depend on tasks that precede them in the workflow. You can drag tasks up or down depending on the workflow. Tasks run either sequentially or in parallel depending on how you modeled the tasks in the stage.

    You can configure the input for a task to depend on the output of an earlier task. When you add parallel tasks in a single group, and configure the input for the task to depend on the output of an earlier task, vRealize Code Stream always considers the variables from the output of the earlier task, which belong to the previous group.