Set up IBM Cloud Pak for Data on IBM Z and LinuxONE

• Hardware: The setup in this quick start guide uses the minimum hardware recommendation suggested by IBM Cloud Pak for Data, IBM Watson Studio and IBM Watson Machine Learning. Information about the hardware requirements is available here.

• Red Hat OpenShift Container Platform (OCP) cluster installed on IBM Z and LinuxONE architecture. Use this setup for a non-production environment. To set up your environment for production follow the official Red Hat documentation.

• Red Hat OpenShift Data Foundation (ODF) for storage. The installation information is available here. For more information about the supported storage, check the official IBM Cloud Pak for Data document.

• Make sure you have the Cluster administrator role in Red Hat OpenShift Container Platform before you start the installation.

• Set up a client workstation that can connect to the Red Hat OpenShift Container Platform cluster and the Internet to download the necessary images for IBM Cloud Pak for Data installation. The example in this quick start guide uses the Bastion as the Client Workstation. Make sure the client workstation can access the IBM Entitled Registry to pull the required images for the IBM Cloud Pak for Data installation.

  curl -v https://icr.io

  Output:

  Connected to icr.io (169.60.98.86) port 443 (#0)

Make sure the prerequisites are met before you start the installation procedure.

Step 1. Install IBM Cloud Pak for Data and the Red Hat OpenShift Container Platform CLI on the client workstation

1. Install the packages:

   dnf install wget tar docker podman

2. Install the Red Hat OpenShift Container Platform CLI by following the instructions in the official Red Hat documentation.

3. Install the IBM Cloud Pak for Data command line interface (cpd-cli).

   wget https://github.com/IBM/cpd-cli/releases/download/v14.0.2/cpd-cli-s390x-EE-14.0.2.tgz

   1. Unzip cpd-cli-s390x-EE-14.0.2.tgz.

      tar xvfz cpd-cli-s390x-EE-14.0.2.tgz

      Output:

      [root@xxxxx cpd-cli-s390x-EE-14.0.2-xxx]# ls
       LICENSES  cpd-cli  plugins

      To run the cpd-cli from any directory, add the fully qualified name to your ~/.bashrc file**

      export PATH=<fully-qualified-path-to-the-cpd-cli>:$PATH

   2. Verify the installed cpd-cli version:

      cpd-cli version 

      Output:

      Version: 14.0.2
      Build Date: 2024-08-08T05:24:32
      Build Number: 236
      CPD Release Version: 5.0.2

      IBM Cloud Pak for Data v5.0.2 uses v14.0.2 cpd-cli in the example described in this paper. For other releases, check the official IBM Cloud Pak for Data cpd-cli documentation.

Step 2. Obtain the IBM Cloud Pak for Data entitlement API key

1. This guide uses the 60-day trial version of the Enterprise Edition license for the non-production installation.

   Note: Skip this step if you already have an IBM Cloud Pak for Data license.

   Follow the link to sign up for the trial (Enterprise Edition license).

2. After you have signed up for the 60-day trial version, copy the IBM entitlement API key by following the steps described in the official IBM Cloud Pak for Data documentation.

Step 3. Prepare the environment variables

1. Provide the variables required for the IBM Cloud Pak for Data installation. Save the variable file as a shell script with the name cpd_vars.sh.

   #===============================================================================
        # Cloud Pak for Data installation variables
        #===============================================================================
   
        # ------------------------------------------------------------------------------
        # Cluster
        # ------------------------------------------------------------------------------
   
        export OCP_URL=<enter your Red Hat OpenShift Container Platform URL>
        export OPENSHIFT_TYPE=self-managed
        export IMAGE_ARCH=s390x
        export OCP_TOKEN=<enter your token>
        export SERVER_ARGUMENTS="--server=${OCP_URL}"
        export LOGIN_ARGUMENTS="--token=${OCP_TOKEN}"
        export CPDM_OC_LOGIN="cpd-cli manage login-to-ocp ${SERVER_ARGUMENTS} ${LOGIN_ARGUMENTS}"
        export OC_LOGIN="oc login ${OCP_URL} ${LOGIN_ARGUMENTS}"
   
        # ------------------------------------------------------------------------------
        # Projects
        # ------------------------------------------------------------------------------
   
        export PROJECT_CERT_MANAGER=ibm-cert-manager
        export PROJECT_LICENSE_SERVICE=ibm-licensing
        export PROJECT_CPD_INST_OPERATORS=my-cpd-operators
        export PROJECT_CPD_INST_OPERANDS=my-cpd-operands
   
        # ------------------------------------------------------------------------------
        # Storage
        # ------------------------------------------------------------------------------
   
        export STG_CLASS_BLOCK=ocs-storagecluster-ceph-rbd
        export STG_CLASS_FILE=ocs-storagecluster-cephfs
   
        # ------------------------------------------------------------------------------
        # IBM Entitled Registry
        # ------------------------------------------------------------------------------
   
        export IBM_ENTITLEMENT_KEY=<enter your IBM entitlement API key from Step 2>
   
        # ------------------------------------------------------------------------------
        # Cloud Pak for Data version
        # ------------------------------------------------------------------------------
   
        export VERSION=5.0.2

   • Cluster: Enter the information related to your Red Hat OpenShift Container Platform cluster.

   • Projects: Enter the Red Hat OpenShift Container Platform project names in which your IBM Cloud Pak for Data is to be installed.

   • Storage: Enter your Red Hat OpenShift Data Foundation storage classes.

   • IBM Entitled Registry: Enter your IBM entitlement API key from Step 2

   • IBM Cloud Pak for Data version: Enter your IBM Cloud Pak for Data version. This quick start guide installs v5.0.2.

     For more information, check the official cpd-cli documentation

2. Run the script to verify that it has no errors.

   bash ./cpd_vars.sh

3. Load the environment variable by sourcing the cpd_vars.sh file to make it accessible for the installation.

   source ./cpd_vars.sh

Step 4. Prepare the Red Hat OpenShift Container Platform cluster

1. Log in to cpd-cli.

   ${CPDM_OC_LOGIN}

   Note: Make sure you log in to CPDM_OC_LOGIN before you run any cpd-cli commands.

2. Add the entitlement key to the global pull secret so that you have the access to the cluster to pull the images required for the IBM Cloud Pak for Data installation.

   cpd-cli manage add-icr-cred-to-global-pull-secret \
   --entitled_registry_key=${IBM_ENTITLEMENT_KEY}

3. After the command has run, verify that the nodes are in Ready status.

   cpd-cli manage oc get nodes

   Output:

   
   
   

4. Install the IBM Cloud Pak Foundational Services Certificate Manager and License Service. Enter the cpd-cli manage command:

   cpd-cli manage apply-cluster-components \
   --release=${VERSION} \
   --license_acceptance=true \
   --cert_manager_ns=${PROJECT_CERT_MANAGER} \
   --licensing_ns=${PROJECT_LICENSE_SERVICE} 

   Wait for the success message before you proceed:

5. Create project permissions. Create and provide the permissions for the projects in which the IBM Cloud Pak for Data instance will be installed.

   1. Log in to your Red Hat OpenShift Container Platform cluster.

      ${OC_LOGIN}  

   2. Create a CPD Instance Operator.

      oc new-project ${PROJECT_CPD_INST_OPERATORS}

   3. Create a CPD Instance Operands.

      oc new-project ${PROJECT_CPD_INST_OPERANDS}

   4. Apply the required permissions to the projects. Use the authorize-instance-topology command:

       cpd-cli manage authorize-instance-topology \
       --cpd_operator_ns=${PROJECT_CPD_INST_OPERATORS} \
       --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS}

      Wait for the successful authentication message to be displayed.

Step 5. Install the IBM Cloud Pak Foundational Service. Use the cpd-cli manage setup-instance-topology command to install the IBM Cloud Pak for Foundation Service.

cpd-cli manage setup-instance-topology \
        --release=${VERSION} \
        --cpd_operator_ns=${PROJECT_CPD_INST_OPERATORS} \
        --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
        --license_acceptance=true \
        --block_storage_class=${STG_CLASS_BLOCK}

Wait for the success message to be displayed.

Step 6. Install IBM Cloud Pak for Data

Make sure you complete the above steps before you start the IBM Cloud Pak for Data installation.

1. Check the license status before you start the installation.

   cpd-cli manage get-license \
   --release=${VERSION} \
   --license-type=EE

   Wait for the success message to be displayed.

2. Install the IBM Cloud Pak for Data control plane. The control plane provides a central user experience for managing users, creating and accessing service instances, customizing the Web client, monitoring platform health and resource use, and managing storage volumes.

   cpd-cli manage apply-olm \
   --release=${VERSION} \
   --cpd_operator_ns=${PROJECT_CPD_INST_OPERATORS} \
   --components=cpd_platform

   Wait for the success message to be displayed, before proceeding with the next step.

3. Set up Red Hat OpenShift Data Foundation storage and install a control plane in the operator projects.

   cpd-cli manage apply-cr \
   --release=${VERSION} \
   --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
   --components=cpd_platform \
   --block_storage_class=${STG_CLASS_BLOCK} \
   --file_storage_class=${STG_CLASS_FILE} \
   --license_acceptance=true

   Wait for the success message to be displayed.

4. Verify the installation.

   • Verification 1: Run the get-cr-status command to confirm the status of the component.

     cpd-cli manage get-cr-status \
     --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
     --components=cpd_platform

     Output: Verify the cpd_platform status and ensure it completed.

   • Verification 2:

     cpd-cli manage get-cr-status \
     --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS}

     Output: Verify the status of all the components and ensure they are completed or successful.

5. Get the IBM Cloud Pak for Data URL and login credential.

    cpd-cli manage get-cpd-instance-details \
    --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
    --get_admin_initial_credentials=true

   Output:

    CPD Url: cpd-my-cpd-operands.xx.xx.xxx.xx
    CPD Username: cpadmin
    CPD Password: xxxxxxxxxxxxxxxxxxxxxxxx
    [SUCCESS]... You may find output and logs in the /root/cp4d-installation/installation_scripts/work directory.
    [SUCCESS]... The get-cpd-instance-details command ran successfully.

The IBM Cloud Pak for Data installation is complete.

Step 7. Optional: Install IBM Watson Studio and IBM Watson Machine Learning service. See the procedures in the next sections.

1. Create the required Operator Lifecycle Manager (OLM) object for IBM Watson Studio:

      cpd-cli manage apply-olm \
      --release=${VERSION} \
      --cpd_operator_ns=${PROJECT_CPD_INST_OPERATORS} \
      --components=ws

   Wait for the success message to be displayed.

2. Create a custom resource for IBM Watson Studio on Red Hat OpenShift Data Foundation storage.

   cpd-cli manage apply-cr \
       --components=ws \
       --release=${VERSION} \
       --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
       --block_storage_class=${STG_CLASS_BLOCK} \
       --file_storage_class=${STG_CLASS_FILE} \
       --license_acceptance=true

   Wait for the success message to be displayed.

3. Verify the installation.

   1. Run the get-cr-status command to confirm the status of the component.

      cpd-cli manage get-cr-status \
          --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
          --components=ws

   2. Verify the status of IBM Watson Studio. For more information about IBM Watson Studio, check the IBM Watson Studio documentation.

1. Create the required Operator Lifecycle Manager (OLM) object for IBM Watson Machine Learning:

   cpd-cli manage apply-olm \
      --release=${VERSION} \
      --cpd_operator_ns=${PROJECT_CPD_INST_OPERATORS} \
      --components=wml

   Wait for the success message to be displayed.

2. Create a custom resource for IBM Watson Machine Learning on OpenShift Data Foundation storage.

   cpd-cli manage apply-cr \
      --components=wml \
      --release=${VERSION} \
      --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
      --block_storage_class=${STG_CLASS_BLOCK} \
      --file_storage_class=${STG_CLASS_FILE} \
      --license_acceptance=true

   Wait for the success message to be displayed.

3. Verify the installation:

   1. Run the get-cr-status command to confirm the status of the component.

      cpd-cli manage get-cr-status \
      --cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
      --components=wml

   2. Verify that the IBM Watson Machine Learning installation is complete. For more information about IBM Watson Machine Learning, check the IBM Watson Machine Learning documentation.

IBM

• Ansible OpenShift Provisioning

• cpd-cli documentation

• Planning Enterprise Edition license entitlements

• IBM Cloud Pak for Data - Obtaining your IBM entitlement API key

• IBM Watson Studio documentation

• IBM Watson Machine Learning documentation

Red Hat

• Red Hat documentation

• Deploying OpenShift Data Foundation using IBM Z

• Getting started with the OpenShift CLI