Deploying storage client in the Hosted Control Plane clusters

Steps to deploy storage client in the Hosted Control Plane clusters.

Before you begin

To enable replica -1:

If replica -1 needs to be enabled on the external storage client cluster, ensure that the cluster uses the same topology and that the nodes are labeled in the same way as the hub (provider) nodes. For instructions about enabling replica-1 for external storage clients, see Steps to enable replica -1.

Procedure

Create a storage client in the Hosted Control Plane clusters.
  1. Deploy the IBM Fusion in the Hosted Control Plane. For procedure to deploy, see Installing the IBM Fusion base.
  2. Install the Data Foundation service and use the following YAML file to create a FusionServiceInstance CR. For steps to install Data Foundation, see Installing Fusion Data Foundation and Configuring Fusion Data Foundation in provider mode.
    apiVersion: service.isf.ibm.com/v1
kind: FusionServiceInstance
metadata:
  name: odfmanager
  namespace: ibm-spectrum-fusion-ns
spec:
  creator: User
  doInstall: true
  parameters:
    - name: namespace
      provided: false
      value: openshift-storage
    - name: creator
      provided: false
      value: Fusion
    - name: backingStorageType
      provided: false
      value: Client
    - name: autoUpgrade
      provided: true
      value: 'false'
    - name: enableLVMStorage
      provided: false
      value: 'false'
  serviceDefinition: data-foundation-service
  triggerUpdate: false
  updateServiceCRSpec: false
  3. Create NetworkPolicy to allow the storage client to connect Data Foundation provider in the host cluster.
    Important: This step needs to be done in the host cluster.
    1. Run the following command to get the Hosted Control Plane pods namespace.
      oc get hostedcontrolplanes.hypershift.openshift.io -A | grep <your-hcp-name>
    2. Create the following YAML and replace the namespace with your Hosted Control Plane cluster. The default value is clusters-<hcp-cluster-name>.
      apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
  name: openshift-storage-egress
  namespace: <your-hcp-namespace>
spec:
  egress:
  - to:
    - namespaceSelector:
        matchLabels:
          kubernetes.io/metadata.name: openshift-storage
  podSelector:
    matchLabels:
      kubevirt.io: virt-launcher
  policyTypes:
  - Egress
  4. Generate the client onboarding token from the user interface as follows:
    Important: The onboarding token is valid for 48 hours and can be used only once.
    1. Log in the OpenShift® Container Platform web console.
    2. Go to Storage > Storage Clients.
    3. Click Generating client onboarding token to generate the onboarding token.

      As soon as you click the Generating client onboarding token, the Client onboarding token page is displayed.

    4. Click Copy to clipboard to copy the token to the clipboard.
  5. Get the storageProviderEndpoint as follows:
    1. Use the storageProviderEndpoint cluster IP.
    2. Run the following command to get the storageProviderEndpoint cluster IP.
      oc get svc -n openshift-storage ocs-provider-server
      Example output:
      NAME                  TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)     AGE
ocs-provider-server   ClusterIP   172.30.3.161   <none>        50051/TCP   53d
  6. Create Data Foundation cluster CR as follows:
    1. Create a YAML for Data Foundation cluster CR and replace the ONBOARDING_TOKEN and PROVIDER_SERVICE_ENDPOINT with the correct value.
      For example:
      apiVersion: odf.isf.ibm.com/v1
kind: OdfCluster
metadata:
  name: odfcluster
  namespace: ibm-spectrum-fusion-ns
spec:
  storageClient:
    enable: true
    onboardingTicket: $ONBOARDING_TOKEN
    storageProviderEndpoint: '$PROVIDER_SERVICE_ENDPOINT:50051'
  creator: CreatedByFusion
    2. Apply the Data Foundation cluster CR to create the storageclient.
  7. Verify the storageclient status from the IBM Fusion user interface.
  8. Verify the Storage Client and StorageClasses details from the OCP user interface.

What to do next

Steps to enable replica -1:

To enable replica -1 for external storage clients, do as follows:

  1. Enable the replica -1.

    Command example:
    oc patch storagecluster ocs-storagecluster -n openshift-storage --type json --patch '[{ "op": "replace", "path": "/spec/managedResources/cephNonResilientPools/enable", "value": true }]'
  2. Add the following annotation to the StorageConsumer.
    failureDomain=$(oc get storagecluster -n openshift-storage -o jsonpath='{.items[0].status.failureDomainKey}' )
oc annotate storageconsumer <consumer-name> \ -n openshift-storage \ ocs.openshift.io/non-resilient-pools-topology-key="$failureDomain"