Bug 1645674

Please provide kubectl get -o yaml vmi …

# kubectl get -o yaml vmi vmi-migratable
apiVersion: kubevirt.io/v1alpha2
kind: VirtualMachineInstance
metadata:
  creationTimestamp: 2018-11-05T16:20:16Z
  finalizers:
  - foregroundDeleteVirtualMachine
  generation: 1
  labels:
    kubevirt.io/migrationTargetNodeName: cnv-executor-ipinto-node2.example.com
    kubevirt.io/nodeName: cnv-executor-ipinto-node1.example.com
    special: vmi-migratable
  name: vmi-migratable
  namespace: default
  resourceVersion: "4079601"
  selfLink: /apis/kubevirt.io/v1alpha2/namespaces/default/virtualmachineinstances/vmi-migratable
  uid: ae839663-e116-11e8-99f9-fa163ecc89c7
spec:
  domain:
    devices:
      disks:
      - disk:
          bus: virtio
        name: registrydisk
        volumeName: registryvolume
      interfaces:
      - bridge: {}
        name: default
    features:
      acpi:
        enabled: true
    firmware:
      uuid: 6e3b0e5b-4c9d-4dcb-a21a-89f05824b316
    machine:
      type: q35
    resources:
      requests:
        memory: 64M
  networks:
  - name: default
    pod: {}
  terminationGracePeriodSeconds: 0
  volumes:
  - name: registryvolume
    registryDisk:
      image: kubevirt/cirros-registry-disk-demo:latest
status:
  interfaces:
  - ipAddress: 10.129.0.185
  migrationState:
    migrationUid: e858c716-e117-11e8-99f9-fa163ecc89c7
    sourceNode: cnv-executor-ipinto-node1.example.com
    startTimestamp: 2018-11-05T16:29:18Z
    targetNode: cnv-executor-ipinto-node2.example.com
    targetNodeAddress: 10.130.0.48:43208
  nodeName: cnv-executor-ipinto-node1.example.com
  phase: Running

Can you please attach the domxml of the VM too? Just so we know what CPU model was set by kubevirt as it might have defaulted to host-model for example.

# cat /etc/libvirt/qemu/default_vmi-migratable.xml
<domain type='kvm'>
  <name>default_vmi-migratable</name>
  <uuid>0bcbe651-2a5f-4c91-8a74-11c037c23840</uuid>
  <metadata>
    <kubevirt xmlns="http://kubevirt.io">
      <uid>e81913cb-f32b-11e8-8e67-fa163e46d324</uid>
      <graceperiod>
        <deletionGracePeriodSeconds>0</deletionGracePeriodSeconds>
      </graceperiod>
    </kubevirt>
  </metadata>
  <memory unit='KiB'>62500</memory>
  <currentMemory unit='KiB'>62500</currentMemory>
  <vcpu placement='static'>1</vcpu>
  <iothreads>1</iothreads>
  <sysinfo type='smbios'>
    <system>
      <entry name='uuid'>0bcbe651-2a5f-4c91-8a74-11c037c23840</entry>
    </system>
  </sysinfo>
  <os>
    <type arch='x86_64' machine='pc-q35-2.12'>hvm</type>
    <boot dev='hd'/>
  </os>
  <features>
    <acpi/>
  </features>
  <cpu mode='host-model' check='partial'>
    <model fallback='allow'/>
  </cpu>
  <clock offset='utc'/>
  <on_poweroff>destroy</on_poweroff>
  <on_reboot>restart</on_reboot>
  <on_crash>destroy</on_crash>
  <devices>
    <emulator>/usr/bin/qemu-kvm</emulator>
    <disk type='file' device='disk'>
      <driver name='qemu' type='raw' cache='none'/>
      <source file='/var/run/kubevirt-ephemeral-disks/registry-disk-data/default/vmi-migratable/disk_registryvolume/disk-image.raw'/>
      <target dev='vda' bus='virtio'/>
      <alias name='ua-registrydisk'/>
      <address type='pci' domain='0x0000' bus='0x03' slot='0x00' function='0x0'/>
    </disk>
    <controller type='usb' index='0' model='none'/>
    <controller type='sata' index='0'>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x1f' function='0x2'/>
    </controller>
    <controller type='pci' index='0' model='pcie-root'/>
    <controller type='virtio-serial' index='0'>
      <address type='pci' domain='0x0000' bus='0x02' slot='0x00' function='0x0'/>
    </controller>
    <controller type='pci' index='1' model='pcie-root-port'>
      <model name='pcie-root-port'/>
      <target chassis='1' port='0x10'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0' multifunction='on'/>
    </controller>
    <controller type='pci' index='2' model='pcie-root-port'>
      <model name='pcie-root-port'/>
      <target chassis='2' port='0x11'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x1'/>
    </controller>
    <controller type='pci' index='3' model='pcie-root-port'>
      <model name='pcie-root-port'/>
      <target chassis='3' port='0x12'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x2'/>
    </controller>
    <controller type='pci' index='4' model='pcie-root-port'>
      <model name='pcie-root-port'/>
      <target chassis='4' port='0x13'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x3'/>
    </controller>
    <interface type='bridge'>
      <mac address='0a:58:0a:81:00:2f'/>
      <source bridge='k6t-eth0'/>
      <model type='virtio'/>
      <alias name='ua-default'/>
      <address type='pci' domain='0x0000' bus='0x01' slot='0x00' function='0x0'/>
    </interface>
    <serial type='unix'>
      <source mode='bind' path='/var/run/kubevirt-private/e81913cb-f32b-11e8-8e67-fa163e46d324/virt-serial0'/>
      <target type='isa-serial' port='0'>
        <model name='isa-serial'/>
      </target>
    </serial>
    <console type='unix'>
      <source mode='bind' path='/var/run/kubevirt-private/e81913cb-f32b-11e8-8e67-fa163e46d324/virt-serial0'/>
      <target type='serial' port='0'/>
    </console>
    <channel type='unix'>
      <target type='virtio' name='org.qemu.guest_agent.0'/>
      <address type='virtio-serial' controller='0' bus='0' port='1'/>
    </channel>
    <input type='mouse' bus='ps2'/>
    <input type='keyboard' bus='ps2'/>
    <graphics type='vnc' socket='/var/run/kubevirt-private/e81913cb-f32b-11e8-8e67-fa163e46d324/virt-vnc'>
      <listen type='socket' socket='/var/run/kubevirt-private/e81913cb-f32b-11e8-8e67-fa163e46d324/virt-vnc'/>
    </graphics>
    <video>
      <model type='vga' vram='16384' heads='1' primary='yes'/>
      <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x0'/>
    </video>
    <memballoon model='none'/>
  </devices>
</domain>

Quick research showed, that adding CPU model "Nehalem" to yaml file resolved the problem:

spec:
  domain:
    cpu:
      model: Nehalem

So the default cpu mode='host-model' means the VM uses most of the features of the node it was started at. Migrating the VM to any lesser host will end up like in your report.

Model Nehalem solves that for you only because it is compatible with both of your hosts.

We are working on adding support for CPU model compatibility based scheduling and expect to have it soon: https://github.com/kubevirt/kubevirt/pull/1721

Moving this to MODIFIED as PR 1721 was merged upstream before Releases 0.12 and 0.13 were branched.

Verified on latest d/s build, fixed

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHEA-2019:0417