Rewrote AWS Terraform deployment for AWS Kargo. It supports now
multiple Availability Zones, AWS Loadbalancer for Kubernetes API,
Bastion Host, ...

For more information see README

Add upgrade-cluster and reset playbooks to syntax check

Added host_vars to gitignore

Use docker-engine 1.13.1

The default version of Docker was switched to 1.13 in #1059.  This
change also bumped ubuntu from installing docker-engine 1.13.0 to
1.13.1.  This PR updates os families which had 1.13 defined, but
were using 1.13.0.

The impetus for this change is an issue running tiller 1.2.3 on
docker 1.13.0.  See discussion [1][2].

[1] https://github.com/kubernetes/helm/issues/1838
[2] https://github.com/kubernetes-incubator/kargo/pull/1100

Since inventory ships with kargo, the ability to change functionality
without having a dirty git index is nice.  An example, we wish to change
is the version of docker deployed to our CentOS systems.  Due to an issue
with tiller and docker 1.13, we wish to deploy docker 1.12.  Since this
change does not belong in Kargo, we wish to locally override the docker
version, until the issue is sorted.

Restart kube-controller for host_resolvconf mode

Fixed counter in ETCD Openssl.conf

Allow connections from pods to local endpoints

Fix vault role in upgrade-cluster.yml

By default Calico blocks traffic from endpoints
to the host itself by using an iptables DROP
action. It could lead to a situation when service
has one alive endpoint, but pods which run on
the same node can not access it. Changed the action
to RETURN.

Ensure vagrant uses flannel

The Vagrantfile is setup to use flannel.  The default network
was changed to Calico (#1031).  However, the Vagrantfile was
not updated to reflect this.  Ensuring the Vagrantfile remains
functional on master, until someone decides to make it work
with Calico.

Adding KUBELET_CLOUDPROVIDER to kubelet.rkt.service

Remove upgrade from the ubuntu-rkt-sep CI job

Change kube-api default port from 443 to 6443

Operator can specify any port for kube-api (6443 default) This helps in
case where some pods such as Ingress require 443 exclusively.

Closes: 820
Signed-off-by: Sergii Golovatiuk <sgolovatiuk@mirantis.com>

Lower default memory requests

Make etcd data dir configurable.

Closes: #1073
Signed-off-by: Sergii Golovatiuk <sgolovatiuk@mirantis.com>

This is to address out of memory issues on CI as well as help
fit deployments for people starting out with kargo on smaller
machines

When a apiserver_loadbalancer_domain_name is added to the Openssl.conf
the counter gets not increased correctly. This didnt seem to have an
effect at the current kargo version.

Using the command module instead of raw

Align LB defaults with the HA docs

Fix cluster.yml file extension in docs

Increase SSL TTL to 3650 days

In real scenarios 365 days is short period of time. 3650 days is good
enough for long running k8s environments

Uncomment one key/value in all.yml

all.yaml shouldn't be empty otherwise ansible won't be able to merge 2
dicts.

Related bug: ansible/issues/21889

Uncommented group_vars variables

Add default var role