When we deploy a MC with the kerneltype functionality and we deploy any custom osImage, the nodes become degraded.

$ oc get clusterversion
NAME      VERSION                              AVAILABLE   PROGRESSING   SINCE   STATUS
version   4.12.0-0.nightly-2022-10-05-053337   True        False         3h53m   Cluster version is 4.12.0-0.nightly-2022-10-05-053337

Always

1. Create a MC in order to user a realtime kernel

cat << EOF | oc create -f -
apiVersion: machineconfiguration.openshift.io/v1
kind: MachineConfig
metadata:
  labels:
    machineconfiguration.openshift.io/role: "worker"
  name: change-worker-kernelarg-realtime
spec:
  kernelType: realtime
EOF


2. Wait for the pools to be updated

$oc get mcp

This is the status in the worker nodes after the MC is applied
$ rpm-ostree status
State: idle
Deployments:
* ostree-unverified-registry:quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:0daf5c4a35424410e88dde102022fc3581302bc8a98e09e2e4748502c59b3661
                   Digest: sha256:0daf5c4a35424410e88dde102022fc3581302bc8a98e09e2e4748502c59b3661
                Timestamp: 2022-10-11T08:04:15Z
      RemovedBasePackages: kernel-core kernel-modules kernel kernel-modules-extra 4.18.0-372.26.1.el8_6
          LayeredPackages: kernel-rt-core kernel-rt-kvm kernel-rt-modules
                           kernel-rt-modules-extra


And the new kernel is deployed properly
$ uname -a
Linux ip-10-0-159-135 4.18.0-372.26.1.rt7.183.el8_6.x86_64 #1 SMP PREEMPT_RT Sat Aug 27 22:04:33 EDT 2022 x86_64 x86_64 x86_64 GNU/Linux 

3. Create any custom osImage, for example one using this dockerfile.

Use the right base image:
# Get base image
$ oc adm release info --image-for "rhel-coreos-8"
quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:0daf5c4a35424410e88dde102022fc3581302bc8a98e09e2e4748502c59b3661 

# Build and push to your repo this Dockerfile
FROM quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:0daf5c4a35424410e88dde102022fc3581302bc8a98e09e2e4748502c59b3661 
RUN printf '[baseos]\nname=CentOS-$releasever - Base\nbaseurl=http://mirror.centos.org/centos/$releasever-stream/BaseOS/$basearch/os/\ngpgcheck=0\nenabled=1\n\n[appstream]\nname=CentOS-$releasever - AppStream\nbaseurl=http://mirror.centos.org/centos/$releasever-stream/AppStream/$basearch/os/\ngpgcheck=0\nenabled=1\n\n' > /etc/yum.repos.d/centos.repo && \
    rpm-ostree install zsh && \
    rpm-ostree cleanup -m && \
    ostree container commit 

4. Create a MC to deploy the new custom osImage

cat << EOF | oc create -f -
kind: MachineConfig
apiVersion: machineconfiguration.openshift.io/v1
metadata:
  labels:
    machineconfiguration.openshift.io/role: "worker"
  name: "tc-54915-layering-kerneltype-worker"
spec:
  osImageURL: "quay.io/examplerepo/layering@sha256:879c8f770a580b03bcf32c710f13cdc868156c50abda279b8e2d977d3b40f3f8"
EOF

The worker pool becomes degraded, reporting this error:

  - lastTransitionTime: "2022-10-11T12:18:39Z"
    message: 'Node ip-10-0-159-135.us-east-2.compute.internal is reporting: "failed
      to update OS to quay.io/examplerepo/layering@sha256:879c8f770a580b03bcf32c710f13cdc868156c50abda279b8e2d977d3b40f3f8
      : error running rpm-ostree rebase --experimental ostree-unverified-registry:quay.io/examplerepo/layering@sha256:879c8f770a580b03bcf32c710f13cdc868156c50abda279b8e2d977d3b40f3f8:
      \x1b[0m\x1b[31merror: \x1b[0mNo enabled repositories\n: exit status 1"'
    reason: 1 nodes are reporting degraded status on sync
    status: "True"
    type: NodeDegraded

The worker pool should not be degraded. The realtime kernel should be applied properly.