This reverts commit 4209f1cb.

* Security fixes for etcd

* Use certs when querying etcd

* Clear admin kubeconfig when rotating certs

* Update main.yml

* Rename dns_server to dnsmasq_dns_server so that it includes role prefix
as the var name is generic and conflicts when integrating with existing ansible automation.
*  Enable selinux state to be configurable with new var preinstall_selinux_state

* Set no_proxy to all local ips

* Use proxy settings on all necessary tasks

Fixes #1741

PID namespace sharing is disabled only in Kubernetes 1.7.
Explicitily enabling it by default could help reduce unexpected
results when upgrading to or downgrading from 1.7.

In case ssh.config is set to use bastion, synchronize needs to use it too.

* Upgrade to kubernetes v1.8.0

hyperkube no longer contains rsync, so now use cp

* Enable node authorization mode

* change kube-proxy cert group name

These facts can be generated in defaults with a performance
boost.

Also cleaned up duplicate etcd var names.

* Test if tokens are expired from host instead of inside container

* Update main.yml

Fixes #1716

Fix system namespace creation
Only rotate tokens when necessary

Also fix yaml lint issues

Fixes #1703

* Enable upgrade to kubeadm

* fix kubedns upgrade

* try upgrade route

* use init/upgrade strategy for kubeadm and ignore kubedns svc

* Use bin_dir for kubeadm

* delete more secrets

* fix waiting for terminating pods

* Manually enforce kube-proxy for kubeadm deploy

* remove proxy. update to kubeadm 1.8.0rc1

* feature: install project atomic CSS on RedHat family

* missing patch for this feature

* sub-role refactor

* Yamllint fix

Fixes #1666

Fixes #1660

The value cannot be determined properly via local facts, so
checking k8s api is the most reliable way to look up what hostname
is used when using a cloudprovider.