Description of problem:
--
- When users need to update cert, they want to make the impact minimal. 
- For example, when master.etcd-ca.crt has been expired, it seems we need to use redeploy-openshift-ca.yml. It means that all services including etcd, masters, nodes and docker have to be restarted. That's not good UX and cannot use for critical system.

Version-Release number of selected component (if applicable):
---
- 3.x

How reproducible:
---
- 100%

Steps to Reproduce:
---
1. Update master.etcd-ca.crt by using redeploy-openshift-ca.yml.

Actual results:
---
- All service are restarted.

Expected results:
---
- Not all services are rebooted.

- There are several options:
 a) playbook updates only master.etcd-ca.crt (or all of each certs).
 b) use current redeploy-openshift-ca.yml but restart operations can be conducted manually after the cert update.