* Update etcd deployment to use correct cert and key

* Update to use admin cert for etcdctl commands

* Update handler to use admin cert too

* Only download container/file if host is in defined group

* Set correct when clause

* Fix last entries

* Update download groups

Kubernetes makes this namespace automatically, so there is
no need for kubespray to manage it.

* Added retries in pre-upgrade.yml and retries while applying kube-dns.yml

* Removed trailing spaces

* Remove old docker packages

This removes docker packages that are obsolete if docker-ce packages are to be installed, which fixes some package conflict issues that can occur during upgrades.

* Add support for setting obsoletes=0 when installing docker with yum

install using Helm:
- Prometheus Operator
- metrics for k8s cluster including: grafana dashboard, alertmanager, node exporters

base project:
https://github.com/coreos/prometheus-operator

the issue:
https://github.com/kubernetes-incubator/kubespray/issues/2042

Previous PR, raw ansible without Helm:
https://github.com/kubernetes-incubator/kubespray/pull/2499

This file should only be updated during kubelet upgrade so that
master components are not accidentally restarted first during
preinstall stage.

- Rename template for /etc/cni/net.d/00-weave.conflist to 00-weave.conflist.j2
- Apply resources requests/limits to both container weave and weave-npc

https://coreos.com/etcd/docs/latest/tuning.html

etcd_snapshot_count
and
ionice priority

The default for kibana_base_url does not make sense an makes kibana unusable. The default path forces a 404 when you try to open kibana in the browser. Not setting kibana_base_url works just fine.

- Fix #2414, so namespace isolation should now works
- Update weave-net.yml.j2 as per latest https://cloud.weave.works/k8s/net
- Other minor fixup

can be set to one of
first-found
can-reach 
interface

- Updating to use calico-node v2.6.7
- A few updates to their manifests too