Description of problem:
Deployment of OSE with one node on RHV engine + 2 hypervisors.
Ansible:

"2016-09-09 15:21:33,549 p=28375 u=foreman | TASK: [Create Hello Openshift application] ************************************
2016-09-09 15:21:49,349 p=28375 u=foreman | failed: [rhv-ose-ose-master1.example.com] => {"changed": true, "cmd": ["oc", "new-app", "http://www.github.com/dymurray/php-helloworld.git"], "delta": "0:00:15.648516", "end": "2016-09-09 19:21:49.311747", "rc": 1, "start": "2016-09-09 19:21:33.663231", "warnings": []}
2016-09-09 15:21:49,349 p=28375 u=foreman | stderr: error: unable to load template file "http://www.github.com/dymurray/php-helloworld.git": yaml: line 6: mapping values are not allowed in this context
error: no match for "http://www.github.com/dymurray/php-helloworld.git"
The 'oc new-app' command will match arguments to the following types:
1. Images tagged into image streams in the current project or the 'openshift' project
- if you don't specify a tag, we'll add ':latest'
2. Images in the Docker Hub, on remote registries, or on the local Docker engine
3. Templates in the current project or the 'openshift' project
4. Git repository URLs or local paths that point to Git repositories
--allow-missing-images can be used to point to an image that does not exist yet.
See 'oc new-app -h' for examples.
2016-09-09 15:21:49,349 p=28375 u=foreman | FATAL: all hosts have already failed -- aborting"

Might be related to bug 1373843, but since ansible reported a different issue I'm opening new bug. Possible case could be network fluctuation on bare metal environment.

Version-Release number of selected component (if applicable):
QCI-1.0-RHEL-7-20160908.1

How reproducible:
Happened to me once

Steps to Reproduce:
1. Deploy OSE on RHV engine + 2 hypervisors on bare metal environment
2. OSE fails at 95%
3.

Actual results:
Deployment failed; OSE failed at 95%

Expected results:
Deployment successful

Additional info:
I'll update this bug with any new information.