1.  vi codeflare-stack-kfdef.yaml
and change the tarball to point to my branch
    #uri: https://github.com/opendatahub-io/distributed-workloads/tarball/main
    uri: https://github.com/jbusche/distributed-workloads/tarball/kuberay-100

2.  Deploy ODH 1.11.0 from rolling channel
Deploy the CodeFlare from alpha channel 1.0.0
Deploy the KFDEFs
oc apply -f https://raw.githubusercontent.com/opendatahub-io/odh-manifests/master/kfdef/odh-core.yaml -n opendatahub
# Use the modified kfdef that uses my branch:
oc apply -f codeflare-stack-kfdef.yaml

3.  Check that the KubeRay is using 1.0.0
oc describe deploy kuberay-operator  |grep Image:
    Image:      quay.io/kuberay/operator:v1.0.0

4. Deploy the 0_basic_ray.ipynb and 2_basic_jobs.ipynb notebook items for testing

5.  Check that the Ray cluster runs right
oc get raycluster -n default
NAME      DESIRED WORKERS   AVAILABLE WORKERS   STATUS   AGE
jobtest   2                 2                   ready    11m

6. Check that the sample job completes:
AppStatus:
  msg: <NONE>
  num_restarts: -1
  roles:
  - replicas:
    - hostname: ''
      id: 0
      role: mnist
      state: !!python/object/apply:torchx.specs.api.AppState
      - 2
      structured_error_msg: <NONE>
    role: mnist
  state: RUNNING (3)
  structured_error_msg: <NONE>
  ui_url: null