metrics: added prometheus metrics for fleet

Documentation/metrics.md

@@ -0,0 +1,39 @@
+# Metrics
+
+**NOTE: The metrics feature is considered experimental. We may add/change/remove metrics without warning in future releases.**
+
+fleet uses [Prometheus][prometheus] for metrics reporting in the server. The metrics can be used for real-time monitoring and debugging.
+
+See the [etcd metrics.md][etcd-metrics] for more information about metrics used in etcd and Prometheus.
+
+The naming of metrics follows the suggested best practice of Prometheus.  A metric name has an fleet prefix as its namespace and a subsystem prefix (for example engine and registry).
+
+To get metrics data you have to request data from fleetd Unix domain socket:
+
+```sh
+curl --unix-socket /var/run/fleet.sock http:/metrics
+```
+
+If you've configured fleetd to [listen TCP socket][tcp-api], please use this command:
+
+```sh
+curl http://127.0.0.1:49153/metrics
+```
+
+fleetd exposes the following metrics:
+
+| Name                                    | Description                                      | Type      |
+|-----------------------------------------|--------------------------------------------------|-----------|
+| engine_leader_start_time                | Timestamp when this fleetd became leader         | Gauge     |
+| engine_task_count_total                 | The total number of executed tasks               | Counter   |
+| engine_task_failure_count_total         | The total number of failed tasks                 | Counter   |
+| engine_reconcile_count_total            | The total number of reconcile rounds             | Counter   |
+| engine_reconcile_duration_second        | The latency distribution of reconcile rounds     | Histogram |
+| engine_reconcile_failure_count_total    | The total number of reconcile failures           | Counter   |
+| registry_operation_count_total          | The total number of registry operations          | Counter   |
+| registry_operation_failed_count_total   | The total number of failed registry operations   | Counter   |
+| registry_operation_duration_second      | The latency distribution of registry operations  | Histogram |
+
+[etcd-metrics]: https://github.com/coreos/etcd/blob/master/Documentation/metrics.md
+[prometheus]: http://prometheus.io/
+[tcp-api]: deployment-and-configuration.md#api