diff options
Diffstat (limited to '')
-rw-r--r-- | changelog.d/13535.misc | 1 | ||||
-rw-r--r-- | synapse/handlers/federation.py | 56 | ||||
-rw-r--r-- | synapse/handlers/federation_event.py | 49 |
3 files changed, 90 insertions, 16 deletions
diff --git a/changelog.d/13535.misc b/changelog.d/13535.misc new file mode 100644 index 0000000000..6b190181c8 --- /dev/null +++ b/changelog.d/13535.misc @@ -0,0 +1 @@ +Add metrics to time how long it takes us to do backfill processing (`synapse_federation_backfill_processing_before_time_seconds`, `synapse_federation_backfill_processing_after_time_seconds`). diff --git a/synapse/handlers/federation.py b/synapse/handlers/federation.py index d13011d138..a09eaa4379 100644 --- a/synapse/handlers/federation.py +++ b/synapse/handlers/federation.py @@ -32,6 +32,7 @@ from typing import ( ) import attr +from prometheus_client import Histogram from signedjson.key import decode_verify_key_bytes from signedjson.sign import verify_signed_json from unpaddedbase64 import decode_base64 @@ -79,6 +80,24 @@ if TYPE_CHECKING: logger = logging.getLogger(__name__) +# Added to debug performance and track progress on optimizations +backfill_processing_before_timer = Histogram( + "synapse_federation_backfill_processing_before_time_seconds", + "sec", + [], + buckets=( + 1.0, + 5.0, + 10.0, + 20.0, + 30.0, + 40.0, + 60.0, + 80.0, + "+Inf", + ), +) + def get_domains_from_state(state: StateMap[EventBase]) -> List[Tuple[str, int]]: """Get joined domains from state @@ -138,6 +157,7 @@ class FederationHandler: def __init__(self, hs: "HomeServer"): self.hs = hs + self.clock = hs.get_clock() self.store = hs.get_datastores().main self._storage_controllers = hs.get_storage_controllers() self._state_storage_controller = self._storage_controllers.state @@ -197,12 +217,39 @@ class FederationHandler: return. This is used as part of the heuristic to decide if we should back paginate. """ + # Starting the processing time here so we can include the room backfill + # linearizer lock queue in the timing + processing_start_time = self.clock.time_msec() + async with self._room_backfill.queue(room_id): - return await self._maybe_backfill_inner(room_id, current_depth, limit) + return await self._maybe_backfill_inner( + room_id, + current_depth, + limit, + processing_start_time=processing_start_time, + ) async def _maybe_backfill_inner( - self, room_id: str, current_depth: int, limit: int + self, + room_id: str, + current_depth: int, + limit: int, + *, + processing_start_time: int, ) -> bool: + """ + Checks whether the `current_depth` is at or approaching any backfill + points in the room and if so, will backfill. We only care about + checking backfill points that happened before the `current_depth` + (meaning less than or equal to the `current_depth`). + + Args: + room_id: The room to backfill in. + current_depth: The depth to check at for any upcoming backfill points. + limit: The max number of events to request from the remote federated server. + processing_start_time: The time when `maybe_backfill` started + processing. Only used for timing. + """ backwards_extremities = [ _BackfillPoint(event_id, depth, _BackfillPointType.BACKWARDS_EXTREMITY) for event_id, depth in await self.store.get_oldest_event_ids_with_depth_in_room( @@ -433,6 +480,11 @@ class FederationHandler: return False + processing_end_time = self.clock.time_msec() + backfill_processing_before_timer.observe( + (processing_start_time - processing_end_time) / 1000 + ) + success = await try_backfill(likely_domains) if success: return True diff --git a/synapse/handlers/federation_event.py b/synapse/handlers/federation_event.py index dd0d610fe9..f40b071a74 100644 --- a/synapse/handlers/federation_event.py +++ b/synapse/handlers/federation_event.py @@ -29,7 +29,7 @@ from typing import ( Tuple, ) -from prometheus_client import Counter +from prometheus_client import Counter, Histogram from synapse import event_auth from synapse.api.constants import ( @@ -98,6 +98,26 @@ soft_failed_event_counter = Counter( "Events received over federation that we marked as soft_failed", ) +# Added to debug performance and track progress on optimizations +backfill_processing_after_timer = Histogram( + "synapse_federation_backfill_processing_after_time_seconds", + "sec", + [], + buckets=( + 1.0, + 5.0, + 10.0, + 20.0, + 30.0, + 40.0, + 60.0, + 80.0, + 120.0, + 180.0, + "+Inf", + ), +) + class FederationEventHandler: """Handles events that originated from federation. @@ -604,20 +624,21 @@ class FederationEventHandler: if not events: return - # if there are any events in the wrong room, the remote server is buggy and - # should not be trusted. - for ev in events: - if ev.room_id != room_id: - raise InvalidResponseError( - f"Remote server {dest} returned event {ev.event_id} which is in " - f"room {ev.room_id}, when we were backfilling in {room_id}" - ) + with backfill_processing_after_timer.time(): + # if there are any events in the wrong room, the remote server is buggy and + # should not be trusted. + for ev in events: + if ev.room_id != room_id: + raise InvalidResponseError( + f"Remote server {dest} returned event {ev.event_id} which is in " + f"room {ev.room_id}, when we were backfilling in {room_id}" + ) - await self._process_pulled_events( - dest, - events, - backfilled=True, - ) + await self._process_pulled_events( + dest, + events, + backfilled=True, + ) @trace async def _get_missing_events_for_pdu( |