Docs about metrics.

deploy/README.md

@@ -72,6 +72,21 @@ $ kubectl apply -f controller.yaml
 The ``image`` in the manifest must specify the controller (named
 ``varnish-ingress/controller`` above).
 
+The manifest specifies a ``containerPort`` for HTTP, at which the
+controller listens for the ``/metrics`` endpoint to publish
+[metrics](ref-metrics.md) suitable for integration with
+[Prometheus](https://prometheus.io/docs/introduction/overview/):
+
+```
+        ports:
+        - name: http
+          containerPort: 8080
+```
+
+The default value for the port number is 8080; to set a different
+value, use the ``-metricsport`` [command-line option](ref-cli-options.md)
+for the controller.
+
 It does *not* make sense to deploy more than one replica of the
 controller. If there are more controllers, all of them will connect to
 the Varnish instances and send them the same administrative

docs/README.md

@@ -8,6 +8,7 @@ The docs in this folder cover these topics:
   * [``BackendConfig`` Custom Resource](ref-backend-cfg.md)
   * [controller command-line options](ref-cli-options.md)
   * [customizing the Pod template](varnish-pod-template.md) for Varnish
+  * [metrics](ref-metrics.md) published by the controller
 
 * [Logging, Events and the Varnish Service monitor](monitor.md)
 

docs/monitor.md

@@ -8,6 +8,12 @@ cluster.
 * Ingress controller log
 * Kubernetes Events generated by the controller
 * Varnish Service monitor
+* [metrics](ref-metrics.md) published by the controller at the
+  ``/metrics`` endpoint, suitable for integration with
+  [Prometheus](https://prometheus.io/docs/introduction/overview/)
+
+See the [metrics reference](ref-metrics.md) for details about the
+metrics.  The other subjects are covered in the following.
 
 ## Ingress controller log
 

docs/ref-cli-options.md

@@ -23,6 +23,8 @@ Usage of ./k8s-ingress:
     	log to standard error instead of files
   -masterurl string
     	cluster master URL, for out-of-cluster runs
+  -metricsport uint
+	port at which to listen for the /metrics endpoint (default 8080)
   -monitorintvl duration
 	interval at which the monitor thread checks and updates
 	instances of Varnish that implement Ingress.
@@ -87,6 +89,14 @@ deactivated for values <= 0. The monitor sleeps this long between
 monitor runs for Varnish Services. See the documentation at the link
 for more details.
 
+``-metricsport`` (default 8080) sets the port number at which the
+controller listens for the HTTP endpoint ``/metrics`` to publish
+[metrics](/docs/ref-metrics.md) that are suitable for integration with
+[Prometheus](https://prometheus.io/docs/introduction/overview/). It
+must match the value of the ``containerPort`` configured for ``http``
+in the [Pod template](/deploy/controller.yaml) for the controller
+(cf. the [deplyoment instructions](/deploy#deploy-the-controller)).
+
 ``-log-level`` sets the log level for the main controller code,
 ``INFO`` by default.