Bug 2114922

Summary:	Can run with host-Model cpuModel even if it is in ObsoleteCPUModels
Product:	Container Native Virtualization (CNV)	Reporter:	Barak <bmordeha>
Component:	Virtualization	Assignee:	Barak <bmordeha>
Status:	CLOSED ERRATA	QA Contact:	zhe peng <zpeng>
Severity:	high	Docs Contact:
Priority:	high
Version:	4.12.0	CC:	acardace, fdeutsch, sgott
Target Milestone:	---
Target Release:	4.13.0
Hardware:	All
OS:	Unspecified
Whiteboard:
Fixed In Version:	hco-bundle-registry-container-v4.13.0.rhel9-848	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2023-05-18 02:55:41 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Barak 2022-08-03 14:11:32 UTC

Description of problem:
Currently if a node has host-model cpu that is included in ObsoleteCPUModels
and we create vmi with host-model cpu it can still be scheduled to that node,
than the vmi will run with host-model cpu and won't be able to migrate.

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


How reproducible:
Add a node's host-model cpu model to the ObsoleteCPUModels
can find out which model is that through the node's label:
"host-model-cpu.node.kubevirt.io/<HostModelOfTheNode>": "true"  

Create a vm to start on the node with host-model(the default cpuModel)

Actual results:
The vm will be scheduled and run with the obsolete cpuModel

Expected results:
Virt launcher shouldn't be scheduled

Additional info:

Comment 1 sgott 2022-08-10 12:29:53 UTC

In this scenario, we allow a VM to be created that cannot be migrated, which will block upgrade. Because of this we're increasing the severity.

Comment 5 zhe peng 2023-03-07 10:40:27 UTC

verify with build: CNV-v4.13.0.rhel9-1639

step:
1: check cluster node cpu model(in the test is "EPYC-Rome")
2: add  cpu mode to ObsoleteCPUModels
$ oc get HyperConverged kubevirt-hyperconverged -n openshift-cnv -o yaml
...
  obsoleteCPUs:
    cpuModels:
    - EPYC-Rome
...
3: check node's label
...
host-model-cpu.node.kubevirt.io/EPYC-Rome: "true"
...

4: create a vm with default template
$ oc get vm
NAME        AGE     STATUS               READY
vm-fedora   6m24s   ErrorUnschedulable   False

$ oc get pods
NAME                            READY   STATUS    RESTARTS   AGE
virt-launcher-vm-fedora-dqnh6   0/1     Pending   0          6m39s

error msg is:
Events:
  Type     Reason            Age   From               Message
  ----     ------            ----  ----               -------
  Warning  FailedScheduling  7m5s  default-scheduler  0/6 nodes are available: 3 Insufficient devices.kubevirt.io/kvm, 3 Insufficient devices.kubevirt.io/tun, 3 Insufficient devices.kubevirt.io/vhost-net, 3 node(s) had untolerated taint {node-role.kubernetes.io/master: }, 6 node(s) didn't match Pod's node affinity/selector. preemption: 0/6 nodes are available: 6 Preemption is not helpful for scheduling..

move to verified.

Comment 8 errata-xmlrpc 2023-05-18 02:55:41 UTC

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 (Moderate: OpenShift Virtualization 4.13.0 Images security, bug fix, and enhancement update), 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/RHSA-2023:3205