Description of problem:
If we didn't use the default etcd storage location,  upgrade will abort in the task "Generate etcd backup".
In fact, the etcd storage location was recorded in master-config.yaml. it is better to get the etcd storage location from config file rather that use the default values.


Version-Release number of selected component (if applicable):
atomic-openshift-utils.noarch 0:3.0.7

How reproducible:
always

Steps to Reproduce:
1. change the openshift to use different etcd storage locations in master config file.
2. upgrade using ansible playbook


Actual results:
TASK: [Generate etcd backup] ************************************************** 
failed: [10.3.8.193] => {"changed": true, "cmd": ["etcdctl", "backup", "--data-dir=/var/lib/openshift/openshift.local.etcd", "--backup-dir=/var/lib/openshift/etcd-backup-20151107172308"], "delta": "0:00:00.015862", "end": "2015-11-07 04:23:07.176924", "rc": 1, "start": "2015-11-07 04:23:07.161062", "warnings": []}
FATAL: all hosts have already failed -- aborting

PLAY RECAP ******************************************************************** 
           to retry, use: --limit @/root/upgrade.retry

10.3.8.193                 : ok=15   changed=4    unreachable=0    failed=1   
localhost                  : ok=3    changed=0    unreachable=0    failed=0   

Expected results:

The upgrade success.
ansible could get etcd storage location from master configuration files.

Additional info: