Alert KubeVirtComponentExceedsRequestedMemory / KubeVirtComponentExceedsRequestedCPU

[rbo]

Fired for components:

• virt-controller
• virt-api
• virt-handler

Configured requested resources:

$ for i in virt-controller virt-api  ; do echo "======= $i ========"; oc get deployment $i -n
 openshift-cnv -o yaml | grep 'resources:' -A5; done;
======= virt-controller ========
        resources:
          requests:
            cpu: 10m
            memory: 250Mi
        securityContext:
          allowPrivilegeEscalation: false
======= virt-api ========
        resources:
          requests:
            cpu: 5m
            memory: 300Mi
        securityContext:
          allowPrivilegeEscalation: false

$ oc get daemonsets.apps virt-handler -o yaml | grep 'resources:' -A5
        resources:
          requests:
            cpu: 10m
            memory: 300Mi
        securityContext:
          privileged: true
--
        resources: {}
        securityContext:
          privileged: true
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
        volumeMounts:

Useage:

• virt-api ~ 350Mi
• virt-controller ~ 270Mi
• virt-handler ~ 380Mi

Documentation:

• https://docs.openshift.com/container-platform/4.12/virt/logging_events_monitoring/virt-runbooks.html#virt-runbook-kubevirtcomponentexceedsrequestedmemory_virt-runbooks
• Upstream: https://kubevirt.io/monitoring/runbooks/KubeVirtComponentExceedsRequestedMemory

[rbo]

There is currently no option to change the request:

• https://issues.redhat.com/browse/CNV-28429

[rbo]

We have the same problem but with CPU : KubeVirtComponentExceedsRequestedCPU