Deploying the IBM Cloud Private Pattern

Complete these steps to deploy the IBM Cloud Private Pattern.

Procedure

  1. From PureApplication® Software, go to Patterns > Virtual System Patterns.
  2. Select the IBM Cloud Private Pattern template that you want to deploy.
  3. Click Deploy.
  4. Select the deploy options.
    Installation type
    The type of IBM Cloud Private installation: IBM Cloud private-ce or IBM Cloud private-ee. This information is used to locate the IBM Cloud Private Docker images when the product is installed. It is also used to locate the product binary files in the PureApplication Software Storehouse.
    IBM Cloud private version
    The version of the IBM Cloud Private product. Together with the Installation type, it is used to locate the product binary files that are uploaded to storehouse. Because several versions of the IBM Cloud Private binary files can be uploaded to storehouse and available to use for new deployments, the binary file name is located at deployment time by using the following rules:
    • If the Installation type is set to Cloud private-ce, the system looks for the binary file in the /admin/files/ICp/ce directory in storehouse. Otherwise, the system looks for the binary file in the /admin/files/ICp/ee directory in storehouse.
    • The name of the product binary file in storehouse must be ibm-cloud-private<os>-<IBM Cloud private version>.tar.gz. For example, for an Intel system, if the IBM Cloud Private version is set to 3.1.1 and the Installation type is IBM Cloud Private-ce, at deployment time the pattern tries to download the product binary file from /admin/files/ICp/ce/ibm-cloud-private-x86-64-3.1.1.tar.gz in storehouse.
    Notes:
    • For only the IBM Cloud Private-CE (Community Edition) installation, if the binary file is not found in storehouse in one of the directories as described in the preceding section, the pattern tries to install the product online from Docker Hub. If your system has internet access, the pattern can install IBM Cloud Private-CE directly from Docker Hub if the binary file for the specified version is not uploaded to storehouse.
    • In the scenario where the IBM Cloud Private binary file that you want to use doesn't follow the naming rule that is described in the preceding section, specify the full name of the binary file instead of the IBM Cloud Private version. The pattern uses that binary file name directly instead of trying to process it from the specified version and installation type parameters.
    System administrator password
    The password for the IBM Cloud Private system administrator that is used to access the IBM Cloud Private console. The default value for the password is admin.
    Restriction: IBM Cloud Private doesn't provide a method to change the password after installation.
    Installation Directory for IBM Cloud Private
    The location of the IBM Cloud Private installation on the host virtual machine.
    Kubernetes API Insecure port
    The port that is used by Kubernetes.
    PureApp Maestro port
    The port that is used by the PureApplication Software internal inlet application. Update this parameter only if you changed your system to use a different port. The default value is 8888.
    Notes:
    • The Installation type (IBM Cloud private-ce or IBM Cloud private-ee) must match the type of the product binary file you want to use (Community Edition (CE) or Enterprise Edition (EE)). If you use the default binary file that is packaged with the pattern, select IBM Cloud private-ce.
    • The IBM Cloud private version must match the version of the binary file that you are using. For the IBM Cloud Private version, provide the value after the ibm-cloud-private-x86_64- and before .tar.gz. For example, ibm-cloud-private-x86_64-3.1.1.tar.gz. If you use the default binary file that is packaged with the pattern, keep the default value.
    • If the template uses a scaling policy for the worker node, you can configure that policy to dynamically scale in or out based on CPU or memory use. You can also update the number of initial worker hosts that are deployed. The worker nodes can be scaled in or out after deployment by using the scaling policy operations if you enable the scaling policy at deployment. Depending on the template that you choose, you might be able to select the number of master and proxy nodes at deployment. Master and proxy nodes cannot be dynamically scaled in or out after deployment.
  5. After you populate all required fields and modify the default values as needed, click Quick Deploy, which is near the top of the screen.

    A link appears, which you can click to take you to the deploying instance.

Results

After you complete the deployment steps and the instance state is Launching, the instance takes about 40 minutes to complete the configurations and reach the Running status.

If you choose to deploy more than one master node, you must specify the virtual IP addresses for the cluster master and the cluster proxy. Make sure that the virtual IP addresses for the cluster master and the cluster proxy are different. The Ethernet interface name must match the interface on which these virtual IP addresses are available.

What to do next

You can check the instance deployment status by looking at the History section. In this section, you can also see the type of the installation (CE or EE), the binary file name, and the storehouse location (if that file was uploaded to storehouse).