Cloud PaK for Data 4.0 was released on June 23rd,2021. In this new release, there is an import change and it is about installation.
Key Point 1:
For Cloud Pak for Data 4.0, we will only support OpenShift v 4.6, (we recommend v4.6.29 or above). It no longer support OpenShift 3.11.
Key Point 2:
From Cloud Pak for Data 4.0, control plane and all the services are installed using operators, which simplify the process of upgrading, scaling, and rolling back software on Red Hat OpenShift Container Platform. Even we have lite and some services introduced operator in Cloud Pak for Data 3.5.x, but for 4.0,all services are using operator for installing and upgrading.
Key Point 3:
We introduce the Cloud Pak foundation services in CPD 4.0. In previous releases of Cloud Pak for Data, if you have more than one Cloud Pak installed, you could optionally install the IBM Cloud Pak foundational services on your cluster to integrate IAM Service and the License Service. But In Cloud Pak for Data Version 4.0, the IBM Cloud Pak foundational services are a prerequisite.
Key Point 4:
Cloud Pak for Data Version 4.0 offers two installation methods for installing the control plane and you can see the description for each methods in our Knowledge :
Express installation
This method enables you to install the Cloud Pak for Data control plane with fewer manual steps. It requires elevated permissions and does not enforce strict division between Red Hat OpenShift Container Platform projects (Kubernetes namespaces).
In an express installation, the IBM Cloud Pak foundational services operators and the Cloud Pak for Data operators are in the same projectHowever, it offers less control over namespace scoping. Additionally, you must be a Red Hat OpenShift cluster administrator to use this installation method.
Specialized installation
This method provides more control over your environment. However, it requires additional manual steps. A Red Hat OpenShift project administrator can complete some of the steps after a cluster administrator completes the initial setup.
In a specialized installation, the IBM Cloud Pak foundational services operators are installed in the ibm-common-services project and the Cloud Pak for Data operators are installed in a separate project (typically cpd-operators). Each project has a dedicated:Operator group, which specifies the OwnNamespace installation mode
- NamespaceScope Operator, which allows the operators in the project to manage operators and service workloads in specific projects
Check more information here
Key Point 5: Mirroring Images
In Cloud Pak for Data v3.x, we have cpd-cli command to download CPD images from a repo.yaml for air gap installation. While from 4.0, there is no cpd-cli command any more to download images, it is strongly recommended that you mirror the necessary software images( including all images needed for installation) from the IBM Entitled Registry to a private container registry.
For Cloud Pak for Data 4.0, the following image prefix are used to store Cloud Pak for Data images:
The mirror scenario is similar when we installation Open Shift Container 4.x in air gap environment, we should have a docker registry to store all the images and using mirrors when installing.
We introduce CASE (Container Application Software for Enterprises ) packages from CPD 4.0.
There are CASE packages for:
- IBM Cloud Pak foundational services
- IBM Cloud Pak for Data control plane
- Each IBM Cloud Pak for Data service
- Software dependencies for the control plane and services
Each CASE includes:Metadata about the component
- An inventory of the container images that are required to deploy the component
- References to any software dependencies
- The scripts needed to mirror the images to a private registry
So how to download CASE packages?We also introducing IBM Cloud Pak CLI (cloudctl) to download images from the IBM Entitled Registry.
We are using 'cloudctl cases save" to download package and using 'cloudctl cases lanuch' with different --action parameter to init and start the registry.
For more details, please check
Key Point 6: Node configuration settings
Yes, Cloud Pak for Data 4.0 also have this per-installation work. To ensure that the cluster has the required settings for the services that you will install, an operating system administrator with root privileges must review and adjust the settings on the appropriate nodes in the cluster.
The steps are similar with CPD 3.x. For more details, please check
In the next part II, I will including the assemblies' operator installation related key point. Coming soon.
Usefully Links:
#CloudPakforDataGroup