Skip to content

Latest commit

 

History

History
99 lines (51 loc) · 6.08 KB

02createRedHatOpenShiftCluster.md

File metadata and controls

99 lines (51 loc) · 6.08 KB

Step 2: Create new RedHat OpenShift Cluster

Overview

  1. Access https://cloud.ibm.com with your IBM ID

  2. Select the account under which to create the new RedHat OpenShift Cluster

    Select Account

  3. In the top-left corner, click on the Navigation Menu icon

    Navigation Menu

  4. In the Navigation Menu select OpenShift -> Clusters

    Clusters

  5. Click Create cluster

  6. Select the appropriate OpenShift Version, 4.8.26 or a newer 4.8.x version

    Note: IBM Automation SWAT team tested only with OCP version 4.8.x, we recommend to not use older OCP versions (4.6.x is still supported by CP4BA v21.0.3 but does not contain important OCP performance fixes), CP4BA v21.0.3 is not supported any more on OCP 4.7, no tests made yet by IBM Automation SWAT team with OCP 4.9.x

  7. As OCP entitlement, select Apply my Cloud Pak OCP entitlement to this worker pool (as this will reduce the costs for the cluster)

    OpenShift Version and entitlement

  8. Select the Infrastructure, e.g., Classic

  9. Select the Geography where you want to run your cluster, e.g., Europe

  10. Set the Availabilty to Single zone (Multizone is only required for disaster recovery)

  11. Select the Worker zone (datacenter) where the cluster should be created, e.g., Frankfurt 04

    Location

  12. Select the worker pool flavor, the amount of workers needed and deselect Encrypt local disk

    • Select a worker pool flavor that fulfills the OCP Cluster sizing requirements of the previously selected CP4BA template (see Step 0: Select the CP4BA template for deployment)
    • Select the amount of workers that fulfills the OCP Cluster sizing requirements of the previously selected CP4BA template. If you plan to scale up the deployment (deploy with a pod replica size of 2 or 3 for High Availabilty) more workers/resources are needed. Optionally you can start with a replica size of 1 first and then scale up when needed - first add more workers and then increase the pod replica size, see also Step 11: Optional: Scale up the deployment
    • Local disk encryption is not needed as we only work with sample / test data

    Notes: To save costs, you also can select Virtual - shared nodes, this is sufficient for running the Client Onboarding Demo with one or only a few users in parallel for demo purposes. Same is true for the amount of workers, when you plan to run the Demo only with one user, usually no HA is needed. If you plan to use this OpenShift cluster for higher workloads / multiple users, means you require High Availability and better/best performance, you might want to select Virtual - dedicated or even Bare metal machines and also select an higher amount of workers / a flavor that gives you more resources per worker.

    Select flavor

    Select amount of workers

  13. Make sure the permissions checker reports no errors and provide the Cluster name

    Cluster name

    Optionally deselect available integrations as this is only a Demo / Enablement environment

    Integrations

  14. Review the Summary on the right hand side and click Create

  15. Monitor the creation of your new ROKS cluster, it usually will take up to one hour to fully provision it (if you selected Bare metal, it also can take longer, dependant on the HW availability in the selected data center)

As you by default will be running DB2 containerized on the OpenShift cluster, you might need - dependant on the previously selected CP4BA template - one worker node with more memory (see Step 0: Select the CP4BA template for deployment) - if this is the case for your selected CP4BA template, complete the remaining sub-steps, otherwise you can proceed with Step 3: Create new VM for LDAP, install IBM SDS & import ldif files with users and groups

  1. Once the first worker nodes are up and running, navigate to your cluster, in the Navigation Menu select OpenShift -> Clusters

    Clusters

  2. Select the cluster, e.g., tech-jam-emea, and open the Worker pools

    Worker pools

  3. Click Add

  4. Provide the Worker pool name, e.g., db2worker, and select the same Worker zone as selected before, e.g., Frankfurt 04

    Worker pool name and zone

  5. Select a woker pool flavor that provides you the amount of memory needed for DB2 according to the selected CP4BA template

    Worker pool flavor for DB2

  6. Set the amount of workers to 1, deselect Encrypt local disk, select Apply my Cloud Pak OCP entitlement to this worker pool and make sure the permissions checker reports no errors

    Worker amount

  7. Review the Summary on the right hand side and click Create

  8. Monitor the creation of your additional worker

What to do next

Once the ROKS cluster is created, you can proceed with the next step: Step 3: Create new VM for LDAP, install IBM SDS & import ldif files with users and groups

Here you can get back to the overview page

Issues or questions? IBMers can use this IBM internal Slack channel: #dba-swat-asset-qna (https://ibm-cloud.slack.com/archives/C026TD1SGCA)

Everyone else can open a new issue in this github.