Add some metrics for inbound and outbound federation processing times (#7755)

changelog.d/7755.misc

@@ -0,0 +1 @@
+Add some metrics for inbound and outbound federation latencies: `synapse_federation_server_pdu_process_time` and `synapse_event_processing_lag_by_event`.

synapse/federation/federation_server.py

@@ -18,7 +18,7 @@ import logging
 from typing import Any, Callable, Dict, List, Match, Optional, Tuple, Union
 
 from canonicaljson import json
-from prometheus_client import Counter
+from prometheus_client import Counter, Histogram
 
 from twisted.internet import defer
 from twisted.internet.abstract import isIPAddress
@@ -70,6 +70,10 @@ received_queries_counter = Counter(
     "synapse_federation_server_received_queries", "", ["type"]
 )
 
+pdu_process_time = Histogram(
+    "synapse_federation_server_pdu_process_time", "Time taken to process an event",
+)
+
 
 class FederationServer(FederationBase):
     def __init__(self, hs):
@@ -271,6 +275,7 @@ class FederationServer(FederationBase):
 
             for pdu in pdus_by_room[room_id]:
                 event_id = pdu.event_id
+                with pdu_process_time.time():
                     with nested_logging_context(event_id):
                         try:
                             await self._handle_received_pdu(origin, pdu)

synapse/federation/sender/__init__.py

@@ -201,8 +201,16 @@ class FederationSender(object):
 
                     logger.debug("Sending %s to %r", event, destinations)
 
+                    if destinations:
                         self._send_pdu(event, destinations)
 
+                        now = self.clock.time_msec()
+                        ts = await self.store.get_received_ts(event.event_id)
+
+                        synapse.metrics.event_processing_lag_by_event.labels(
+                            "federation_sender"
+                        ).observe(now - ts)
+
                 async def handle_room_events(events: Iterable[EventBase]) -> None:
                     with Measure(self.clock, "handle_room_events"):
                         for event in events:

synapse/handlers/appservice.py

@@ -114,6 +114,12 @@ class ApplicationServicesHandler(object):
                         for service in services:
                             self.scheduler.submit_event_for_as(service, event)
 
+                        now = self.clock.time_msec()
+                        ts = yield self.store.get_received_ts(event.event_id)
+                        synapse.metrics.event_processing_lag_by_event.labels(
+                            "appservice_sender"
+                        ).observe(now - ts)
+
                     @defer.inlineCallbacks
                     def handle_room_events(events):
                         for event in events:

synapse/metrics/__init__.py

@@ -463,6 +463,12 @@ event_processing_last_ts = Gauge("synapse_event_processing_last_ts", "", ["name"
 # finished being processed.
 event_processing_lag = Gauge("synapse_event_processing_lag", "", ["name"])
 
+event_processing_lag_by_event = Histogram(
+    "synapse_event_processing_lag_by_event",
+    "Time between an event being persisted and it being queued up to be sent to the relevant remote servers",
+    ["name"],
+)
+
 # Build info of the running server.
 build_info = Gauge(
     "synapse_build_info", "Build information", ["pythonversion", "version", "osversion"]