Uploaded image for project: 'OpenShift Virtualization'
  1. OpenShift Virtualization
  2. CNV-34138

[2244869] MTQ does not work with Auto Memory Limits

XMLWordPrintable

    • CNV Virtualization Sprint 245, CNV Virtualization Sprint 246
    • High
    • No

      Description of problem:
      Managed-Tenant-Quota does not help with increasing quotas for VM that set the memory lmits automatically (using Kubevirt featureGate: `AutoResourceLimitsGate`)

      The POD has memory requests and limits:

      > $ oc get pod virt-launcher-vm-fedora-guest-memory-96ck7 -o json | jq .spec.containers[0].resources
      > {
      > "limits":

      { > "memory": "2536Mi" > }

      ,
      > "requests":

      { > "memory": "1268Mi" > }

      > }

      And VMMRQ is big enough:

      > $ oc get vmmrq my-vmmrq-big -o json | jq .spec
      > {
      > "additionalMigrationResources":

      { > "limits.memory": "5Gi", > "requests.memory": "5Gi" > }

      > }

      But migration is still in Pending state:

      > $ oc get vmim
      > NAME PHASE VMI
      > kubevirt-migrate-vm-dcrvp Pending vm-fedora-guest-memory

      Version-Release number of selected component (if applicable):
      4.14

      How reproducible:
      100%

      Steps to Reproduce:
      1. enable FG in kubevirt: AutoResourceLimitsGate
      2. create VM with memory guest (without limits)
      3. create resourcequota
      4. create vmmrq
      5. migrate VM

      Actual results:
      Migration is in Pending state

      Expected results:
      Migration completed

      Additional info:

      Perhaps, this issue related to bug 2241953

            sgott@redhat.com Stuart Gott
            dshchedr@redhat.com Denys Shchedrivyi
            Kedar Bidarkar Kedar Bidarkar
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

              Created:
              Updated:
              Resolved: