The CI/CD pipeline is at the heart of the DevOps transformation by providing the underlying infrastructure to automate the workflow of application development and deployment.

Wazi Deploy can be easily integrated into a CI/CD pipeline to automate the deployment of z/OS applications. Technologies used in Wazi Deploy for scripting deployments are identical to non-mainframe deployments like Ansible or Python. Wazi Deploy provides a simple command-line interface (CLI) to integrate into a CI/CD pipeline and stores its environment settings, templates, and deployment methods in Git.

Wazi Deploy is available through IBM Developer for z/OS Enterprise Edition. Its documentation is included in the Deploying to z/OS with IBM Wazi Deploy section of the IBM Developer for z/OS documentation.

Wazi Deploy in the CI/CD pipeline

As you can see in the following overall diagram, Wazi Deploy is integrated into the pipeline to automate the deployment to the test environment (Deploy step in the diagram) and the production environment (Release step in the diagram). All the tools of the pipeline are highly integratable and determine the standards for implementing a unified CI/CD pipeline at the enterprise level.

The Wazi Deploy workflow

You can watch the “Static Deployment with Wazi Deploy” video in the IBM MediaCenter to see a demo of the Wazi Deploy workflow in a pipeline. This video illustrates a  deployment where the artifacts to be deployed are stored in an archive file, named the package file. 

The following diagram breaks down the Wazi Deploy workflow into the main tasks and shows how these tasks fit into the pipeline:

As the diagram shows, you need to complete the following tasks to deploy your applications to z/OS with Wazi Deploy:

1. Build your z/OS application by using IBM Dependency Based Build (DBB) or any build framework used in your organization. 
   The build populates a local folder with the artifacts from your application. The local folder is typically a UNIX System Services partition in a z/OS environment or a standard folder in a distributed environment.
   If you use IBM Dependency Based Build, a build report is generated.
2. Create a package file that contains the build outputs.
   This file is a .tar archive file.
   You can use the Wazi Deploy packager command to automatically create this package file and upload it to an artifact repository.
3. During the deployment step, run the Wazi Deploy generation command by using this package file and a deployment method (a YAML template file that describes the deployment actions and steps to be completed on the artifacts) as input.
   This command produces a package that retrieves the artifacts and a deployment plan that applies the actions and steps of the deployment method to these artifacts.
4. Deploy the package to the target z/OS system by using the deployment command.
   The actions and steps of the generated deployment plan are implemented through the Wazi Deploy building blocks.

The two branches in the diagram constitute two deployment streams:

• To deploy from a z/OS system, use the Python stream. Run the Wazi Deploy generation command from the UNIX System Services file system on z/OS. Then, deploy to your target z/OS system by using the Wazi Deploy deployment command for Python.
• To deploy from a distributed environment, use the Ansible deployment. Run the Wazi Deploy generation command on a Linux environment. Then, deploy to your target z/OS system by using the Ansible deployment command and the deployment scripts based on Red Hat Ansible Certified Content for IBM Z. Ansible communicates with the z/OS system through SSH.

Customizing the deployment

You can create your own Python and Ansible building blocks and call them from the deployment method to implement your own deployment actions and steps.

To experiment a customized deployment, you can follow the scenarios explained in Creating a Python building block or Creating an Ansible building block in the Wazi Deploy section of the IBM Developer for z/OS documentation These scenarios, which are the continuation of the static deployment scenarios, contain sample codes that you can copy and paste for a customized deployment.

The evidence file and requester command

When an Ansible or a Python building block is run to deploy to the target z/OS system, it collects pieces of information that are produced by the steps of the deployment plan. These pieces of information are collected in an evidence file that can be queried after the deployment to analyze the Wazi Deploy deployment process and result.

For more information, see The evidence file in the Wazi Deploy section of the IBM Developer for z/OS documentation.

Closed Beta

The Wazi Deploy Closed Beta program provides early access to features. As a member of this Beta program, you are entitled to download frequently-updated versions of the product and give your feedback on them. You also have access to the entitled documentation where the beta features are explained.

To be involved in the evolution of the product, you can also become a sponsor user.

For any request about the Wazi Deploy Closed Beta program, you can contact Olivier Le Louër (le_louer.o@fr.ibm.com), the Wazi Deploy development manager, or Senthil Nathan (SenthilNathan@ibm.com), the Wazi Deploy product manager.