chore(gke): install and dns instructions for gke

Author: slack

src/quickstart/provider/gke/dns.md

@@ -1,5 +1,47 @@
-    $ kubectl --namespace=deis describe svc deis-router | grep LoadBalancer
+## Find Your Load Balancer Address
 
-should exit this doc with `DEIS_HOSTNAME` set
+On Google Container Engine, Deis Workflow will automatically provision and
+attach a Google Cloud Loadbalancer to the router copmonent. This component is
+responsible for routing HTTP and HTTPS requests from the public internet to
+applications that are deployed and managed by Deis Workflow.
 
-[next: deploy your first app](../../deploy-an-app.md)
+By describing the `deis-router` service, you can see what IP address has been
+allocated by Google Cloud for your Deis Workflow cluster:
+
+```
+$ kubectl --namespace=deis describe svc deis-router | grep LoadBalancer
+Type:                   LoadBalancer
+LoadBalancer Ingress:   104.197.125.75
+```
+
+## Prepare the Hostname
+
+Now that you have the ip address of your load balancer we can use the `nip.io`
+DNS service to route arbitrary hostnames to the Deis Workflow edge router. This
+lets us point the Workflow CLI at your cluster without having to either use
+your own domain or update DNS!
+
+To verify the Workflow API server and nip.io, construct your hostname by taking
+the ip address for your load balancer and adding `nip.io`. For our example
+above, the address would be: `104.197.125.75.nip.io`.
+
+Nip answers with the ip address no matter the hostname:
+```
+$ host 104.197.125.75.nip.io
+104.197.125.75.nip.io has address 104.197.125.75
+$ host something-random.104.197.125.75.nip.io
+something-random.104.197.125.75.nip.io has address 104.197.125.75
+```
+
+By default, any HTTP traffic for the hostname `deis` will be sent to the Workflow API service. To test that everything is connected properly you may validate connectivity using `curl`:
+
+```
+$ curl http://deis.104.197.125.75.nip.io/v2/ && echo
+{"detail":"Authentication credentials were not provided."}
+```
+
+You should see a failed request because we provided no credentials to the API server.
+
+Remember the hostname, we will use it in the next step.
+
+You are now ready to [register an admin user and deploy your first app](../../deploy-an-app.md).

src/quickstart/provider/gke/install-gke.md

@@ -1,4 +1,4 @@
-# Installing Deis Workflow
+# Install Deis Workflow on Google Compute Engine
 
 ## Check Your Setup
 
@@ -67,6 +67,24 @@ If you would like `kubectl` to automatically update as the pod states change, ru
 $ kubectl get pods --namespace=deis -w
 ```
 
+Depending on the order in which the Workflow components start, you may see a few components restart. This is common during the installation process, if a component's dependencies are not yet available the component will exit and Kubernetes will automatically restart the containers.
+
+Here, you can see that controller, builder and registry all took a few loops before there were able to star:
+```
+workflow [refactor-docs*]$ kubectl get pods --namespace=deis
+NAME                          READY     STATUS    RESTARTS   AGE
+deis-builder-miekp            1/1       Running   1          2m
+deis-controller-egu7x         1/1       Running   3          2m
+deis-database-ok3ev           1/1       Running   0          2m
+deis-logger-fluentd-d5cb9     1/1       Running   0          2m
+deis-logger-fluentd-u6azj     1/1       Running   0          2m
+deis-logger-rf3z9             1/1       Running   0          2m
+deis-minio-sdfyz              1/1       Running   0          2m
+deis-registry-f534k           1/1       Running   4          2m
+deis-router-t3qb2             1/1       Running   0          2m
+deis-workflow-manager-kbpw3   1/1       Running   0          2m
+```
+
 Once you see all of the pods in the `READY` state, Deis Workflow is up and running!
 
-Next, [configure dns](dns) so you can register your first user.
+Next, [configure dns](dns.md) so you can register your first user and deploy an application.