Description of problem:
Run upgrade against container ocp. Upgrade fail at task [openshift_node : Wait for node to be ready]. Checked that node can not start. One of reasons should be that old Node service file was not updated to unwants openvwitch because ovs service file was removed before the task.

# cat /etc/systemd/system/atomic-openshift-node.service|grep openv
After=openvswitch.service
Wants=openvswitch.service
PartOf=openvswitch.service
  -v /lib/modules:/lib/modules -v /etc/origin/openvswitch:/etc/openvswitch \


# systemctl status openvswitch.service 
Unit openvswitch.service could not be found.

# systemctl status atomic-openshift-node -l
● atomic-openshift-node.service
   Loaded: loaded (/etc/systemd/system/atomic-openshift-node.service; enabled; vendor preset: disabled)
   Active: failed (Result: exit-code) since Wed 2018-04-25 04:17:03 EDT; 1h 8min ago
 Main PID: 22580 (code=exited, status=1/FAILURE)

Apr 25 04:17:02 qe-jliu-c39-master-etcd-1 atomic-openshift-node[22580]: I0425 04:17:02.536408   22637 factory.go:116] Factory "docker" was unable to handle container "/system.slice/var-lib-origin-openshift.local.volumes-pods-047fedd5\\x2d4861\\x2d11e8\\x2da006\\x2d42010af00014-volumes-kubernetes.io\\x7esecret-sdn\\x2dtoken\\x2d82s7t.mount"
Apr 25 04:17:02 qe-jliu-c39-master-etcd-1 atomic-openshift-node[22580]: I0425 04:17:02.536418   22637 factory.go:109] Factory "systemd" can handle container "/system.slice/var-lib-origin-openshift.local.volumes-pods-047fedd5\\x2d4861\\x2d11e8\\x2da006\\x2d42010af00014-volumes-kubernetes.io\\x7esecret-sdn\\x2dtoken\\x2d82s7t.mount", but ignoring.
Apr 25 04:17:02 qe-jliu-c39-master-etcd-1 atomic-openshift-node[22580]: I0425 04:17:02.536431   22637 manager.go:930] ignoring container "/system.slice/var-lib-origin-openshift.local.volumes-pods-047fedd5\\x2d4861\\x2d11e8\\x2da006\\x2d42010af00014-volumes-kubernetes.io\\x7esecret-sdn\\x2dtoken\\x2d82s7t.mount"
Apr 25 04:17:02 qe-jliu-c39-master-etcd-1 atomic-openshift-node[22580]: I0425 04:17:02.718150   22637 docker_server.go:73] Stop docker server
Apr 25 04:17:02 qe-jliu-c39-master-etcd-1 atomic-openshift-node[23611]: atomic-openshift-node
Apr 25 04:17:02 qe-jliu-c39-master-etcd-1 systemd[1]: atomic-openshift-node.service: main process exited, code=exited, status=1/FAILURE
Apr 25 04:17:03 qe-jliu-c39-master-etcd-1 systemd[1]: Stopped atomic-openshift-node.service.
Apr 25 04:17:03 qe-jliu-c39-master-etcd-1 systemd[1]: Unit atomic-openshift-node.service entered failed state.
Apr 25 04:17:03 qe-jliu-c39-master-etcd-1 systemd[1]: atomic-openshift-node.service failed.
Apr 25 05:00:36 qe-jliu-c39-master-etcd-1 systemd[1]: Cannot add dependency job for unit atomic-openshift-node.service, ignoring: Unit not found.


Version-Release number of the following components:
openshift-ansible-3.10.0-0.28.0.git.0.439cb5c.el7.noarch

How reproducible:
always

Steps to Reproduce:
1. Run upgrade against container ocp without setting openshift_use_system_containers

2.
3.

Actual results:
Upgrade failed.

Expected results:
Upgrade succeed.

Additional info:
Please attach logs from ansible-playbook with the -vvv flag