Merge pull request #13 from kate-osborn/grafana

Add grafana dashboard and prometheus config

Author: kate-osborn

examples/grafana/README.md

@@ -0,0 +1,55 @@
+# Grafana Dashboard
+
+The `NGINX Mesh Top` Grafana dashboard provides a high-level picture
+of the request volume and success rate of all your applications in the mesh.
+NGINX Service Mesh deploys Grafana and adds `NGINX Mesh Top` as the
+default dashboard. If you prefer to use an existing Grafana deployment, you can import
+the `NGINX Mesh Top` dashboard included in this directory. 
+
+## Prerequisties
+- Grafana version >= 5.2.0
+- Prometheus datasource must be
+  [configured](https://prometheus.io/docs/visualization/grafana/#creating-a-prometheus-data-source).
+  
+  **Note:** If you are using the Prometheus server deployed by NGINX Service Mesh you can
+  find the address by running [`nginx-meshctl
+  config`](https://docs.nginx.com/nginx-service-mesh/reference/nginx-meshctl/#usage).
+  If you are using your own Prometheus server, make sure to add the NGINX
+  Service Mesh [scrape config](../prometheus/README.md) to your
+  Prometheus configuration.
+ 
+
+## Installing the NGINX Mesh Top Dashboard
+
+To install the NGINX Mesh Top Dashboard using the Grafana UI follow these steps:
+- Select the New Dashboard button and click Import.
+- Upload the [nginx-mesh-top.json](nginx-mesh-top.json) file or copy and paste the contents of the file in the textbox and click Load.
+- Select the Prometheus data source you configured previously from the dropdown
+  menu and click Import.
+
+![nginx-mesh-top](dashboard.png)
+
+
+### Features
+NGINX Mesh Top dashboard includes the following stats and graphs:
+
+- Stats:
+  - Global Success Rate: the percentage of requests that are successful over
+    the last 30 seconds (`nginxplus_upstream_server_responses` with response codes of `1xx` or `2xx`).
+  - Global Request Volume: the volume of requests over the last 30 seconds,
+    measured in requests per second (`nginxplus_http_total_requests`) . 
+  - Pods Monitored: The number of pods being scraped by Prometheus.
+- Graphs:
+  - Request Volume: this graph shows the volume of requests per Pod, in
+    requests per second (`nginxplus_http_total_requests`).
+  - Pod Success:  this graph shows the percentage of requests that are
+    successful per Pod (`nginxplus_upstream_server_responses` with response codes of
+    `1xx` or `2xx`).
+
+
+## Customizing your own Dashboard
+
+NGINX Service Mesh exposes NGINX Plus metrics in Prometheus format that can be used to
+create your own panels and custom dashboards. For a list of available metrics, labels,
+and example Prometheus queries, see the [Traffic Metrics
+guide](https://docs.nginx.com/nginx-service-mesh/guides/traffic-metrics/).