Red Hat OpenShift Virtualization

Purpose

Red Hat OpenShift Virtualization, a feature included with Red Hat OpenShift, provides a platform for running and deploying virtual machine (VM) workloads.

Collecting capacity data

1. Configure Red Hat Open Shift cluster:
   
   
   • Create a ServiceAccount for License Metric Tool server by using the YAML configuration below. Apply it through the + button in the Open Shift Web Console or with the oc apply command:

     
     apiVersion: v1
     kind: ServiceAccount
     metadata:
       name: <SERVICE_ACCOUNT_NAME> 
       namespace: <SERVICE_ACCOUNT_NAMESPACE>
     

   • Create a token for the newly created ServiceAccount by using the YAML configuration below:

     
     apiVersion: v1
     kind: Secret
     metadata:
       name: ilmt-service-account-token
       namespace: <SERVICE_ACCOUNT_NAMESPACE>
       annotations:
         kubernetes.io/service-account.name: <SERVICE_ACCOUNT_NAME>
     type: kubernetes.io/service-account-token
     

   • Assign the necessary permissions to the previously created ServiceAccount by using the YAML configuration below:

     
     kind: ClusterRole
     apiVersion: rbac.authorization.k8s.io/v1
     metadata:
       name: ilmt-cluster-nodes-list
     rules:
       - verbs:
           - list
         apiGroups:
           - ''
         resources:
           - nodes
     ---
     kind: ClusterRoleBinding
     apiVersion: rbac.authorization.k8s.io/v1
     metadata:
       name: ilmt-cluster-nodes-list
     subjects:
       - kind: ServiceAccount
         name: <SERVICE_ACCOUNT_NAME>
         namespace: <SERVICE_ACCOUNT_NAMESPACE> 
     roleRef:
       apiGroup: rbac.authorization.k8s.io
       kind: ClusterRole
       name: ilmt-cluster-nodes-list
     ---
     kind: ClusterRole
     apiVersion: rbac.authorization.k8s.io/v1
     metadata:
       name: ilmt-cluster-virtualmachineinstance-list
     rules:
       - verbs:
           - list
         apiGroups:
           - kubevirt.io
         resources:
           - virtualmachineinstances
     ---
     kind: ClusterRoleBinding
     apiVersion: rbac.authorization.k8s.io/v1
     metadata:
       name: ilmt-cluster-virtualmachineinstance-list
     subjects:
       - kind: ServiceAccount
         name: <SERVICE_ACCOUNT_NAME> 
         namespace: <SERVICE_ACCOUNT_NAMESPACE>
     roleRef:
       apiGroup: rbac.authorization.k8s.io
       kind: ClusterRole
       name: ilmt-cluster-virtualmachineinstance-list
     

     Note: Replace <SERVICE_ACCOUNT_NAME> with the name of your service account and <SERVICE_ACCOUNT_NAMESPACE> with a namespace in which the service account is created.
   
   
   Since it is not possible to retrieve all the information required for subcapacity license calculation, such as processor type or the number of worker node sockets, it is necessary to either:

   • Define a pair of custom node labels in each and every node that is a virtual machine's hypervisor:
     • ibm.ilmt.sockets_count - containing the number of sockets
     • ibm.ilmt.cpu-model.vendor_id - with possible values of AMD/Intel.
   • Install and configure a Node Feature Discovery (Red Hat provided, and not Community Edition) add-on to your cluster to automatically add a set of the following nodes labels to all your nodes:
     • feature.node.kubernetes.io/cpu-model.vendor_id
     • feature.node.kubernetes.io/ibm.ilmt.socket_count
     
     
     The first label is added automatically without any additional configuration of this add-on, immediately after installing the 'NodeFeatureDiscoverie' operand from the 'Node Feature Discovery' operator. However, to add a second label, an additional custom resource of the NodeFeatureRule type needs to be added. To do so, use the following YAML configuration:

     apiVersion: nfd.openshift.io/v1alpha1
     kind: NodeFeatureRule
     metadata:
       name: ibm.ilmt.capacity-sockets
       namespace: openshift-nfd
     spec:
       rules:
         - name: "ibm.ilmt.capacity-sockets"
           labelsTemplate: |
             {{ range .cpu.topology }}ibm.ilmt.{{ .Name }}={{ .Value }}
             {{ end }}
           matchFeatures:
             - feature: cpu.topology
               matchExpressions:
                 socket_count: {op: Exists}
     

   
   

   If the processor type or the number of sockets on a worker node cannot be detected, a default value of 120 PVUs per core will be used to calculate the PVU license consumption for applications installed on virtual machines running on that node.

   To obtain the Service Account Token, follow these steps:

   • Go to Workload > Secrets and find ilmt-service-account-token.
   • In the Secret details, locate the token in the Data section and copy it from there.

   Note: Do not copy the token from the YAML tab, as this token is encoded in Base64 and it will not work.
   
   
   To obtain the Cluster API address:

   • Go to Home > Overview.
   • In the Details widget, find Cluster API address. For example: https://api.test.ibm.com:6443.
2. Configure the License Metric Tool server:

   
   

   For Central and Local VM Manager Tool

   • Create a VM Manager connection as described in the documentation: Adding VM managers in local and central mode

   For Distributed and Disconnected VM Manager Tool

   • Create a VM Manager connection as described in the documentation: Adding and editing VM managers in distributed and disconnected mode:

Supported versions

For information about supported versions, go to IBM® License Metric Tool 9.2 - Supported Operating Systems and click Supported Hypervisor.