Chapter 3. Updating Red Hat OpenShift Data Foundation 4.17 to 4.18

• For Internal and Internal-attached deployments, upgrading OpenShift Data Foundation upgrades all OpenShift Data Foundation services including the backend Red Hat Ceph Storage (RHCS) cluster.

• For External mode deployments, upgrading OpenShift Data Foundation only upgrades the OpenShift Data Foundation service while the backend Ceph storage cluster remains untouched and needs to be upgraded separately.

  You must upgrade Red Hat Ceph Storage along with OpenShift Data Foundation to get new feature support, security fixes, and other bug fixes. As there is no dependency on RHCS upgrade, you can upgrade the OpenShift Data Foundation operator first followed by RHCS upgrade or vice-versa. For more information about RHCS releases, see the knowledgebase solution, solution.

Prerequisites

• Ensure that the OpenShift Container Platform cluster has been updated to the latest stable release of version 4.18.X, see Updating Clusters.

• Ensure that the OpenShift Data Foundation cluster is healthy and data is resilient.

  • Navigate to Storage Data Foundation Storage Systems tab and then click on the storage system name.
  • Check for the green tick on the status card of both Overview - Block and File and Object tabs. Green tick indicates that the storage cluster, object service and data resiliency are all healthy.

• Ensure that all OpenShift Data Foundation Pods, including the operator pods, are in Running state in the openshift-storage namespace.

  To view the state of the pods, on the OpenShift Web Console, click Workloads Pods. Select openshift-storage from the Project drop-down list.

  Note

  If the Show default projects option is disabled, use the toggle button to list all the default projects.

• Ensure that you have sufficient time to complete the OpenShift Data Foundation update process, as the update time varies depending on the number of OSDs that run in the cluster.

• Optional: To reduce the upgrade time for large clusters that are using CSI plugins, make sure to tune the following parameters in the rook-ceph-operator-config configmap to a higher count or percentage.

  • CSI_RBD_PLUGIN_UPDATE_STRATEGY_MAX_UNAVAILABLE

  • CSI_CEPHFS_PLUGIN_UPDATE_STRATEGY_MAX_UNAVAILABLE

    Note

    By default, the rook-ceph-operator-config configmap is empty and you need to add the data key.

    This affects CephFS and CephRBD daemonsets and allows the pods to restart simultaneously or be unavailable and reduce the upgrade time. For an optimal value, you can set the parameter values to 20%. However, if the value is too high, disruption for new volumes might be observed during the upgrade.

• Prerequisite relevant only for OpenShift Data Foundation deployments on AWS using AWS Security Token Service (STS)

  Add another entry in the trust policy for noobaa-core account as follows:

  1. Log into AWS web console where the AWS role resides using http://console.aws.amazon.com/.
  2. Enter the IAM management tool and click Roles.

  3. Find the name of the role created for AWS STS to support Multicloud Object Gateway (MCG) authentication using the following command in OpenShift CLI:

     $ oc get deployment noobaa-operator -o yaml -n openshift-storage | grep ROLEARN -A1
               value: arn:aws:iam::123456789101:role/your-role-name-here

     Copy to Clipboard Toggle word wrap
  4. Search for the role name that you obtained from the previous step in the tool and click on the role name.
  5. Under the role summary, click Trust relationships.
  6. In the Trusted entities tab, click Edit trust policy on the right.
  7. Under the “Action”: “sts:AssumeRoleWithWebIdentity” field, there are two fields to enable access for two NooBaa service accounts noobaa and noobaa-endpoint. Add another entry for the core pod’s new service account name, system:serviceaccount:openshift-storage:noobaa-core.

  8. Click Update policy at the bottom right of the page.

     The update might take about 5 minutes to get in place.

Verification steps

• Check the Version after the OpenShift Data Foundation name and check the operator status.

  • Navigate to Operators Installed Operators and select the openshift-storage project.
  • When the upgrade completes, the version updates to a new version number for OpenShift Data Foundation and status changes to Succeeded with a green tick.

• Verify that the OpenShift Data Foundation cluster is healthy and data is resilient.

  • Navigate to Storage Data Foundation Storage Systems tab and then click on the storage system name.
  • Check for the green tick on the status card of Overview- Block and File and Object tabs. Green tick indicates that the storage cluster, object service and data resiliency is healthy.
• If verification steps fail, contact Red Hat Support.