Collapse documents from Knative Serving to docs repo (#803)

This collapses changes from application-deubgging-guide and
performance-investigation-guide in the Knative serving repo into the
debugging-application-issuses and debugging-performance-issues documents
in the documents repo.

serving/accessing-logs.md

@@ -35,6 +35,23 @@ up the necessary components first.
 
 <!-- TODO: create a video walkthrough of the Kibana UI -->
 
+### Accessing stdout/stderr logs
+
+To find the logs sent to `stdout/stderr` from your application in the Kibana UI:
+
+1. Click `Discover` on the left side bar.
+1. Choose `logstash-*` index pattern on the left top.
+1. Input `tag: kubernetes*` in the top search bar then search.
+
+### Accessing request logs
+
+To find the request logs of your application in the Kibana UI :
+
+1. Click `Discover` on the left side bar.
+1. Choose `logstash-*` index pattern on the left top.
+1. Input `tag: "requestlog.logentry.istio-system"` in the top search bar then
+   search.
+
 ### Accessing configuration and revision logs
 
 To access the logs for a configuration:

serving/debugging-application-issues.md

@@ -41,6 +41,72 @@ details, see Knative
 [Error Conditions and Reporting](https://github.com/knative/serving/blob/master/docs/spec/errors.md)(currently
 some of them are not implemented yet).
 
+### Check ClusterIngress/Istio routing
+
+Run the following command to list all the cluster ingress, with their labels
+
+```shell
+kubectl get clusteringress -o=custom-columns='NAME:.metadata.name,LABELS:.metadata.labels'
+NAME                   LABELS
+helloworld-go-h5kd4    map[serving.knative.dev/route:helloworld-go serving.knative.dev/routeNamespace:default]
+```
+
+The labels `serving.knative.dev/route` and `serving.knative.dev/routeNamespace`
+will tell exactly which Route a ClusterIngress is a child resource of. Find the
+one corresponding to your Route. If a ClusterIngress does not exist, the route
+controller believes that the Revisions targeted by your Route/Service isn't
+ready. Please proceed to later sections to diagnose Revision readiness status.
+
+Otherwise, run the following command to look at the ClusterIngress created for
+your Route
+
+```
+kubectl get clusteringress <CLUSTERINGRESS_NAME> -o yaml
+```
+
+particularly, look at the `status:` section. If the ClusterIngress is working
+correctly, we should see the condition with `type=Ready` to have `status=True`.
+Otherwise, there will be error messages.
+
+Now, if ClusterIngress shows status Ready, there must be a corresponding
+VirtualService. Run the following command:
+
+```shell
+kubectl get virtualservice <CLUSTERINGRESS_NAME> -n knative-serving -o yaml
+```
+
+the network configuration in VirtualService must match that of ClusterIngress
+and Route. VirtualService currently doesn't expose a Status field, so if one
+exists and have matching configurations with ClusterIngress and Route, you may
+want to wait a little bit for those settings to propagate.
+
+If you are familar with Istio and `istioctl`, you may try using `istioctl` to
+look deeper using Istio
+[guide](https://istio.io/help/ops/traffic-management/proxy-cmd/).
+
+### Check Ingress status
+
+Before Knative 0.3 we use a LoadBalancer service call `knative-ingressgateway`
+to handle ingress. Since Knative 0.3 we now use `istio-ingressgateway` Service.
+
+To check the IP address of your Ingress, use
+
+```shell
+kubectl get svc -n istio-system istio-ingressgateway
+```
+
+Or replace that with `knative-ingressgateway` if you are using Knative release
+older than 0.3.
+
+If there is no external IP address, use
+
+```shell
+kubectl describe svc istio-ingressgateway -n istio-system
+```
+
+to see a reason why IP addresses weren't provisioned. Most likely it is due to a
+quota issue.
+
 ## Check Revision status
 
 If you configure your `Route` with `Configuration`, run the following command to
@@ -73,6 +139,7 @@ If you see this condition, check the following to continue debugging:
 
 - [Check Pod status](#check-pod-status)
 - [Check application logs](#check-application-logs)
+- [Check Istio routing](#check-clusteringressistio-routing)
 
 If you see other conditions, to debug further: