IBM Developer

Tutorial

Deploy IBM Cloud Paks to Red Hat OpenShift infrastructures

Use the Cloud Pak Deployer tool to automatically deploy IBM Cloud Paks using configurations kept in Git repositories

By Arthur Laimboeck, Markus Wiegleb, Jiří Pětník, Srećko Janjić, Adam Deleeuw, Frank Ketelaars

On this page

To install and continuously manage Red Hat OpenShift infrastructures and deployed IBM Cloud Paks requires automation. The Cloud Pak Deployer tool uses configurations stored in a Git repository (a GitOps approach) to fully automate the deployment of Red Hat OpenShift and IBM Cloud Paks on various cloud infrastructures. Each configuration file defines the desired end-state by defining the infrastructure, OpenShift cluster, and IBM Cloud Pak to be installed. Secrets are kept in a vault, which the Cloud Pak Deployer uses during the deployment. You keep the configuration files and vault updated, and then re-run the Cloud Pak Deployer to deploy the changes.

The Cloud Pak Deployer simplifies the deployment of IBM Cloud Paks by focusing on proven deployment patterns, for POCs or full production environments. The deployment follows a pre-defined flow, but because it relies on the configuration files for deployment, you do not need to understand what happens behind the scenes or "under the hood" (although we do document the full deployment process, if you're curious).

The following figure shows the current high-level architecture for the Cloud Pak Deployer:

High-level architecture diagram for the Cloud Pak Deployer

Prerequisites

The complete list of prerequisites for using the Cloud Pak Deployer are specified in our gitpages. However, in general, you'll need:

  • A Linux server or VM to run the Cloud Pak Deployer tool
  • Docker or Podman on the Linux server
  • Git
  • A Red Hat OpenShift pull secret for some cloud infrastructures
  • Your Cloud Pak entitlement key
  • An OpenShift infrastructure (if you are deploying a Cloud Pak on an existing OpenShift cluster)

Steps

  1. Install the Cloud Pak deployer
  2. Define a Cloud Pak configuration
  3. Run the Cloud Pak deployer

Step 1: Install the Cloud Pak Deployer

To install the Cloud Pak Deployer:

  1. Install the prerequisites
  2. Clone the repository
  3. Build the image

Step 2: Define a Cloud Pak configuration

The Cloud Pak Deployer tool provides a few sample configurations.

You can get started with the simple static configuration to deploy Cloud Pak for Data to OpenShift on IBM Cloud. Or, you can use the more flexible dynamic configuration to select any combination of Cloud Pak and OpenShift infrastructure.

Most importantly, you should copy the sample configuration files out of the cloud-pak-deployer directory to your own working directory so that your changes are not overwritten when you pull down the latest version of the Cloud Pak Deployer tool.

You can also set up some advanced configuration options, such as using dynamic variables or private registries. Or, you can set up continuous adoption using GitOps.

Step 3: Run the Cloud Pak Deployer

The specific steps to run the Cloud Pak Deployer depend on your infrastructure, but in general you need to:

  • Obtain the appropriate infrastructure credentials
  • Acquire an IBM Cloud Pak entitlement key
  • Set environment variables
  • Run Cloud Pak Deployer
  • Verify the deployed Cloud Paks

We provide steps for these supported infrastructures:

Summary

The Cloud Pak Deployer tool helps automate complex deployments of OpenShift infrastructures and IBM Cloud Paks.

The Cloud Pak Deployer tool:

  • Validates and prepares the environment
  • Provisions and configures the OpenShift infrastructures
  • Installs, updates, and configures the Cloud Paks
  • Deploys assets
  • Runs a set of smoke tests