Add prometheus metrics to router

[akkoma] / lib / pleroma / web / telemetry.ex

diff --git a/lib/pleroma/web/telemetry.ex b/lib/pleroma/web/telemetry.ex
index 73ce6a85cdda2ce425994afc4f536d0602cae80d..435f557992c233fbdfa0bcb4ff8db3032e705bae 100644 (file)
--- a/lib/pleroma/web/telemetry.ex
+++ b/lib/pleroma/web/telemetry.ex
@@ -10,27 +10,19 @@ defmodule Pleroma.Web.Telemetry do
   @impl true
   def init(_arg) do
     children = [
-      # Telemetry poller will execute the given period measurements
-      # every 10_000ms. Learn more here: https://hexdocs.pm/telemetry_metrics
       {:telemetry_poller, measurements: periodic_measurements(), period: 10_000},
-      # Add reporters as children of your supervision tree.
-      # {Telemetry.Metrics.ConsoleReporter, metrics: metrics()},
-      {TelemetryMetricsPrometheus, metrics: metrics(), plug_cowboy_opts: [ip: {127, 0, 0, 1}]}
+      {TelemetryMetricsPrometheus, metrics: prometheus_metrics(), plug_cowboy_opts: [ip: {127, 0, 0, 1}]}
     ]
 
     Supervisor.init(children, strategy: :one_for_one)
   end
 
-  def metrics do
+  @doc """
+  A seperate set of metrics for distributions because phoenix dashboard does NOT handle
+  them well
+  """
+  defp distribution_metrics do
     [
-      # Phoenix Metrics
-      summary("phoenix.endpoint.stop.duration",
-        unit: {:native, :millisecond}
-      ),
-      summary("phoenix.router_dispatch.stop.duration",
-        tags: [:route],
-        unit: {:native, :millisecond}
-      ),
       distribution(
         "phoenix.router_dispatch.stop.duration",
         # event_name: [:pleroma, :repo, :query, :total_time],
@@ -61,20 +53,9 @@ defmodule Pleroma.Web.Telemetry do
           buckets: [0.01, 0.025, 0.05, 0.1, 0.2, 0.5, 1, 2.5, 5, 10]
         ]
       ),
-      summary("pleroma.repo.query.total_time", unit: {:native, :millisecond}),
-      summary("pleroma.repo.query.decode_time", unit: {:native, :millisecond}),
-      summary("pleroma.repo.query.query_time", unit: {:native, :millisecond}),
-      summary("pleroma.repo.query.queue_time", unit: {:native, :millisecond}),
-      summary("pleroma.repo.query.idle_time", unit: {:native, :millisecond}),
-
-      # VM Metrics
-      summary("vm.memory.total", unit: {:byte, :kilobyte}),
-      summary("vm.total_run_queue_lengths.total"),
-      summary("vm.total_run_queue_lengths.cpu"),
-      summary("vm.total_run_queue_lengths.io"),
       distribution(
-        "oban_job_completion",
-        event_name: [:oban, :job, :stop],
+        "oban_job_exception",
+        event_name: [:oban, :job, :exception],
         measurement: :duration,
         tags: [:worker],
         tag_values: fn tags -> Map.put(tags, :worker, tags.job.worker) end,
@@ -84,33 +65,62 @@ defmodule Pleroma.Web.Telemetry do
         ]
       ),
       distribution(
-        "oban_job_exception",
-        event_name: [:oban, :job, :exception],
+        "tesla_request_completed",
+        event_name: [:tesla, :request, :stop],
         measurement: :duration,
-        tags: [:worker],
-        tag_values: fn tags -> Map.put(tags, :worker, tags.job.worker) end,
+        tags: [:response_code],
+        tag_values: fn tags -> Map.put(tags, :response_code, tags.env.status) end,
         unit: {:native, :second},
         reporter_options: [
           buckets: [0.01, 0.025, 0.05, 0.1, 0.2, 0.5, 1, 2.5, 5, 10]
         ]
       ),
       distribution(
-        "tesla_request_completed",
-        event_name: [:tesla, :request, :stop],
+        "oban_job_completion",
+        event_name: [:oban, :job, :stop],
         measurement: :duration,
-        tags: [:response_code],
-        tag_values: fn tags -> Map.put(tags, :response_code, tags.env.status) end,
+        tags: [:worker],
+        tag_values: fn tags -> Map.put(tags, :worker, tags.job.worker) end,
         unit: {:native, :second},
         reporter_options: [
           buckets: [0.01, 0.025, 0.05, 0.1, 0.2, 0.5, 1, 2.5, 5, 10]
         ]
+      )
+    ]
+  end
+
+  defp summary_metrics do
+    [
+      # Phoenix Metrics
+      summary("phoenix.endpoint.stop.duration",
+        unit: {:native, :millisecond}
+      ),
+      summary("phoenix.router_dispatch.stop.duration",
+        tags: [:route],
+        unit: {:native, :millisecond}
       ),
+      summary("pleroma.repo.query.total_time", unit: {:native, :millisecond}),
+      summary("pleroma.repo.query.decode_time", unit: {:native, :millisecond}),
+      summary("pleroma.repo.query.query_time", unit: {:native, :millisecond}),
+      summary("pleroma.repo.query.queue_time", unit: {:native, :millisecond}),
+      summary("pleroma.repo.query.idle_time", unit: {:native, :millisecond}),
+
+      # VM Metrics
+      summary("vm.memory.total", unit: {:byte, :kilobyte}),
+      summary("vm.total_run_queue_lengths.total"),
+      summary("vm.total_run_queue_lengths.cpu"),
+      summary("vm.total_run_queue_lengths.io"),
+
+
       last_value("pleroma.local_users.total"),
       last_value("pleroma.domains.total"),
       last_value("pleroma.local_statuses.total")
     ]
   end
 
+  def prometheus_metrics, do: summary_metrics() ++ distribution_metrics()
+  def live_dashboard_metrics, do: summary_metrics()
+
   defp periodic_measurements do
     [
       {__MODULE__, :instance_stats, []}