Description of problem:
While enabling vsphere cloudprovider, task failed at:

TASK [openshift_cloud_provider : Check to see if the vsphere cluster role already exists] ***
Wednesday 06 June 2018  22:54:28 -0400 (0:00:00.071)       0:07:51.431 ******** 

fatal: [qe-ghuang-node-registry-router-1.0606-zoy.qe.rhcloud.com]: FAILED! => {"changed": false, "cmd": "oc get clusterrole", "failed": true, "msg": "[Errno 2] No such file or directory", "rc": 2}
fatal: [openshift-132.lab.eng.nay.redhat.com]: FAILED! => {"changed": false, "cmd": "oc get clusterrole", "failed": true, "msg": "[Errno 2] No such file or directory", "rc": 2}

Version-Release number of the following components:
openshift-ansible-3.10.0-0.63.0.git.0.961c60d.el7.noarch.rpm

How reproducible:
always

Steps to Reproduce:
1. Trigger installation against vsphere with cloudprovider enabled


Actual results:
see above

Expected results:

Additional info:
Two issues:

1. No any openshift packages installed at that moment, hence no oc client installed.

2. The task should be executed against first master, instead of the nodes.