Configuring Delivery Pipeline

Delivery Pipeline automates the continuous deployment of your projects through sequences of stages that retrieve input and run jobs, such as builds, tests, and deployments.

Configure Delivery Pipeline to automate the continuous building, testing, and deployment of your apps:

If you are configuring this tool integration as you are creating the toolchain, in the Configurable Integrations section, click Delivery Pipeline. Depending on the template that you use, different fields might be available. Review the default field values and if needed, change those settings.
If you have a toolchain and are adding this tool integration to it, from the IBM Cloud console, click the menu icon and select DevOps. On the Toolchains page, click the toolchain that you want to add Delivery Pipeline to. Alternatively, on your app's Overview page, on the Continuous delivery card, click View toolchain. Then, click Overview.

a. Click Add tool.

b. In the Tool Integrations section, click Delivery Pipeline.
Specify a name for your new pipeline.
Select the type of pipeline that you want to create:
- Classic: Provides an easy-to-use graphical user interface for defining stages and jobs that run on public shared infrastructure, with support for running individual stages on private workers.
- Tekton: Provides a dashboard that you can use to view the output of Tekton pipeline runs on a defined Kubernetes cluster, with support for configuring pipeline definitions repos, the pipeline triggers, where the pipeline runs, and simple secrets.
If you plan to use your pipeline to deploy a user interface, and your pipeline is a Classic pipeline, select the Show apps in the View app menu checkbox. All of the apps that your pipeline creates are shown in the View App list on the toolchain's Overview page.
Click Create Integration to add the Delivery Pipeline to your toolchain.
From your toolchain, on the Delivery pipelines card, click the delivery pipeline to view and configure it. To learn the basics of configuring a pipeline, see Building and deploying pipelines.

If you want the pipeline to automatically run when a commit is pushed to your GitHub or Git repository (repo), follow these steps:

a. Configure GitHub or Git Repos and Issue Tracking for your toolchain before you define the stages for your pipeline. The pipeline stages need the Git URLs for your repos. Each pipeline stage can refer to only one of the GitHub or Git repos that are associated with your toolchain. For instructions to configure GitHub, see the GitHub section. For instructions to configure Git Repos and Issue Tracking, see the Git Repos and Issue Tracking section.

b. Use a webhook. Without a webhook, you can run pipelines manually only. To use a webhook when you link to a GitHub or GitHub Enterprise repo, you need admin privileges. To link to a Git Repos and Issue Tracking repo, you need Master or Owner privileges.
Optional: If you are using a toolchain on IBM Cloud Public and you want Sauce Labs to run tests on your app, configure the Delivery Pipeline to add a Sauce Labs test job. For instructions to configure the test job, see the Configuring a Sauce Labs test job in your pipeline section.

Configuring a Sauce Labs test job in your pipeline

Before you configure a Sauce Labs test job in your pipeline, you need a working pipeline that has stages to build and deploy your app. You must also configure Sauce Labs for your toolchain. For instructions about how to configure Sauce Labs, see the Sauce Labs section.

Configure the Delivery Pipeline to add a Sauce Labs test job:

If you don't have a stage that deploys a test version of your app, create one.
On the stage, add a test job after the deploy job. By placing these jobs in the same stage, they can access the same set of environment properties.

Figure 1. Test job
Configure the stage. On the ENVIRONMENT PROPERTIES tab, create the CF_APP_NAME property.

The Sauce Labs username and access key are available in the test job script as the SAUCE_USERNAME and SAUCE_ACCESS_KEY environment variables. When you write your tests, you must use these environment variables to authenticate with Sauce Labs.
Configure the deploy job. In the Deploy Script field, include this command: export CF_APP_NAME="$CF_APP". That command exports the app name as an environment property.
Configure the test job.

The Service Instance, Target, Organization, and Space fields are populated with the Sauce Labs username, region, org, and space that you are using.

a. For the tester type, select Sauce Labs.

b. For the service instance, select the Sauce Labs username that you used when you configured Sauce Labs for your toolchain.

To see the username and access key that you used when you configured Sauce Labs for your toolchain, click Configure.

c. In the Test Execution Command field, enter the commands that install the dependencies that are required by your tests and then run the tests. For example, for a Node.js app, you might enter these commands:
```
npm install
node_modules/grunt-cli/bin/grunt test:sauce:parallel
```
d. If you want to see your test reports in the test job logs, select the Enable Test Report checkbox, and set the Test Result File Pattern to test/*.xml.
Click SAVE. Whenever your pipeline runs, your Sauce Labs tests run.

Configuring Delivery Pipeline by using the API

The Delivery Pipeline tool integration supports the following configuration parameters that you can use with the Toolchain HTTP API and SDKs when you create, read, and update tool integrations.

You must specify the tool_type_id property in the request body with the pipeline value.

Table 1. Delivery Pipeline tool integration parameters
Parameter	Usage	Type	Terraform argument	Description
name	optional, updatable	String	name	The name of this tool integration.