Overview

Skill Level: Intermediate

IBM Cloud Pak System can reduce effort and make it easier to manage your OCP cluster post deployment. It comes with support for automated capability to add worker nodes to cluster and restart cluster. This tutorial focuses on the management of Red Hat OpenShift Container Platform 4.6.

This tutorial walks you through the steps to manage Red Hat OpenShift Container Platform 4.6.0 Cluster instance on IBM Cloud Pak System V2.3.3.3

Ingredients

IBM Cloud Pak System comes with a support for automated capability to add or delete worker nodes to a cluster and restart a cluster.

This tutorial focuses on the management of Red Hat OpenShift Container Platform 4.6. For details on the deployment of OpenShift Container Platform cluster in IBM Cloud Pak System, see the IBM Developer article Deploying RedHat OpenShift Container Platform 4.6 on IBM Cloud Pak System.

Step-by-step

  1. View nodes and pods summary of OpenShift Container Platform cluster on IBM Cloud Pak System 2.3.3.3

    You can see a summary of cluster nodes, pods, PVCs, and resource utilization in the Managed accelerator instances >Your cluster page of IBM Cloud Pak System console as shown in Figure 1.

    OCP-Instance-1

    Figure 1

  2. Node details for Red Hat OpenShift Container Platform cluster on IBM Cloud Pak System 2.3.3.3

    You can see a complete virtual machine list for the OpenShift Container Platform cluster in the Managed accelerator instances >Your cluster page > Nodes tab of IBM Cloud Pak System console as shown in Figure 2.

    CP4I-nodes-1

    Figure 2

  3. Perform actions on your cluster on IBM Cloud Pak System 2.3.3.3

    IBM Cloud Pak system simplifies OpenShift Container Platform cluster management on IBM Cloud Pak System by providing one click operations for life cycle operations. These operations are available in the Actions menu of Managed accelerator instances > Your cluster page as shown in Figure 3.

    CPS-Actions-highlighted-blue

    Figure 3

    Stopping and Starting your cluster in IBM Cloud Pak System 2.3.3.3

    To stop the cluster, click Stop cluster in the Actions menu as shown in Figure 3. A message box appears prompting you for confirmation as shown in Figure 4. Type STOP to confirm.

    CPS-StopCluster-Blue

    Figure 4

    Cluster gets stopped in few minutes and the status gets updated accordingly on the console as shown in Figure 5.

    CPS-StoppedInstance

    Figure 5

    Similarly, you can also start a cluster by using the Start cluster option in the Actions menu as shown in Figure 6.

    CPS-StartAction

    Figure 6

    It takes few minutes for the virtual machines to come up and services to restart. Confirm whether the cluster is started as shown in Figure 7.

    CPS-RestartedInstance

    Figure 7

    Bringing your cluster under maintenance on IBM Cloud Pak System 2.3.3.3

    If you want to perform any maintenance operations on the cluster, click the Maintenance mode in the Actions menu to bring the cluster under maintenance mode. See Figure 3 for reference.

    If you want to stop/start virtual machine from the console individually, follow the mentioned step to first bring the cluster under maintenance mode. After the maintenance tasks are completed, cluster can be brought back to running state by turning off Maintenance mode in the Actions menu.

     
    Marking your cluster important on IBM Cloud Pak System 2.3.3.3

    IBM Cloud Pak System allows you to tag your clusters as important. Use the Mark cluster switch in the Actions menu to mark or unmark a cluster. Marking prevents the cluster to be accidently deleted or stopped.

     
    Deleting your cluster on IBM Cloud Pak System 2.3.3.0

    To delete a cluster, click Delete cluster in the Actions menu as shown in Figure 8.

    CPS-DeleteOCPCluster

    Figure 8

    A message box appears prompting you for confirmation as shown in Figure 9. Type DELETE to confirm.

    CPS-ConfirmDelete_Small

    Figure 9

    Cluster gets deleted in few minutes and all resources get added back to the available resource pools of the cloud group.

  4. Scale up worker nodes after deployment for your cluster on IBM Cloud Pak System 2.3.3.3

    While sizing is performed before cluster deployment, there might be a need to add more workers to support additional workloads post cluster deployment. IBM Cloud Pak System has a simplified process of adding more workers to OpenShift Container Platform cluster post deployment with just one click.

    Before you add worker nodes to your cluster, review the available resources in your system to ensure that enough vCPUs, RAM, IPs, and disk storage are available in the cloud group. Each worker that gets added to the cluster will have the same hardware configuration as the worker node that got configured during deployment time.

    To add worker nodes to cluster, navigate to Managed accelerator instances>Your cluster page. Go to Nodes tab and click Add workers as shown in Figure 10.

    ClickAddWorker

    Figure 10

    Provide the number of worker nodes you want to add to the cluster as shown in Figure 11 and click Submit.

    AddWorker

    Figure 11

    In IBM Cloud Pak System 2.3.3.3 there is provision to add (scale-up) different type of worker nodes post deployment. You can choose to scale a worker node or storage worker node as shown above in Figure 11. 

    A successful operation submission message appears on top of the page as shown in Figure 12.

    CPS-WorkerScaleupSubmitted

    Figure 12

    After few minutes, a new worker node is added to cluster. To confirm, go to Managed accelerator instances >Your cluster page and see the History section as shown in Figure 13.

    NewNodeAdded

    Figure 13

  5. Scale down worker nodes after deployment for your cluster on IBM Cloud Pak System 2.3.3.3

    Worker nodes can also be removed from OpenShift Container Platform cluster after deployment. IBM Cloud Pak System has a simplified process of removing worker nodes from OpenShift Container Platform cluster post deployment with just one click.

    To remove worker nodes from cluster, navigate to Managed accelerator instances >Your cluster page. Go to Nodes tab and for specific that you wish to remove, open context menu. Select Remove option as shown in Figure 14.

    Note: Ensure to schedule pods to different nodes in cluster as required before deleting a node.

    RemoveNode

    Figure 14

    A message box appears prompting you for confirmation as shown in Figure 15. Type REMOVE to confirm.

    RemoveConfirm

    Figure 15

    A successful operation submission message appears on top of the page as shown in Figure 16.

    RemoveSubmitted

    Figure 16

    After few minutes, worker node is removed from the cluster. To confirm, go to Managed accelerator instances >Your cluster page and see the History section and Nodes section as shown in Figure 17.

    RemovalCompleted

    Figure 17

    Note: You can only scale down worker nodes. Control plane nodes scaling up/down is not allowed and recommended by Red Hat.

  6. Conclusion

    IBM Cloud Pak System 2.3.3.3 enables you to quickly roll out one or more Red Hat OpenShift 4.6 clusters. It not only simplifies the roll out and management of OpenShift Container Platform cluster processes, but also provides one click day 2 operations for your cluster.

Join The Discussion