Description of problem:
Run migrate playbook to migrate etcd2 to etcd3 against an upgraded v3.6 cluster, etcd migration failed at task [openshift_master_facts : Set master facts].

fatal: [x.x.x.x]: FAILED! => {
    "failed": true
}

MSG:

The task includes an option with an undefined variable. The error was: {{ hostvars | oo_select_keys(groups['oo_etcd_to_config'] | union(groups['oo_new_etcd_to_config'])) | oo_collect('openshift.common.hostname') | default(none, true) }}: 'dict object' has no attribute 'oo_new_etcd_to_config'

The error appears to have been in '/usr/share/ansible/openshift-ansible/roles/openshift_master_facts/tasks/main.yml': line 31, column 3, but may
be elsewhere in the file depending on the exact syntax problem.

The offending line appears to be:
- name: Set master facts
  ^ here

exception type: <class 'ansible.errors.AnsibleUndefinedVariable'>
exception: {{ hostvars | oo_select_keys(groups['oo_etcd_to_config'] | union(groups['oo_new_etcd_to_config'])) | oo_collect('openshift.common.hostname') | default(none, true) }}: 'dict object' has no attribute 'oo_new_etcd_to_config'
        to retry, use: --limit @/usr/share/ansible/openshift-ansible/playbooks/byo/openshift-etcd/migrate.retry


Version-Release number of the following components:
atomic-openshift-utils-3.6.173.0.45-1.git.0.dc70c99.el7.noarch
ansible-2.4.0.0-4.el7.noarch

How reproducible:
always

Steps to Reproduce:
1. Upgrade ocp v3.5 to ocp v3.6
2. Do etcd migration
# ansible-playbook -i hosts /usr/share/ansible/openshift-ansible/playbooks/byo/openshift-etcd/migrate.yml -v
3.

Actual results:
Migration failed.

Expected results:
Migration succeed.

Additional info:
Hosts and log in attachment.