From de0d088adc0cf3d5bbd80238b88143426cd6eaca Mon Sep 17 00:00:00 2001 From: Erik Johnston Date: Tue, 20 Apr 2021 14:11:24 +0100 Subject: [PATCH] Add presence federation stream (#9819) --- changelog.d/9819.feature | 1 + synapse/handlers/presence.py | 243 ++++++++++++++++++-- synapse/replication/tcp/client.py | 7 +- synapse/replication/tcp/streams/__init__.py | 3 + synapse/replication/tcp/streams/_base.py | 24 ++ tests/handlers/test_presence.py | 179 +++++++++++++- 6 files changed, 426 insertions(+), 31 deletions(-) create mode 100644 changelog.d/9819.feature diff --git a/changelog.d/9819.feature b/changelog.d/9819.feature new file mode 100644 index 000000000..f56b0bb3b --- /dev/null +++ b/changelog.d/9819.feature @@ -0,0 +1 @@ +Add experimental support for handling presence on a worker. diff --git a/synapse/handlers/presence.py b/synapse/handlers/presence.py index bd2382193..598466c9b 100644 --- a/synapse/handlers/presence.py +++ b/synapse/handlers/presence.py @@ -24,6 +24,7 @@ The methods that define policy are: import abc import contextlib import logging +from bisect import bisect from contextlib import contextmanager from typing import ( TYPE_CHECKING, @@ -53,7 +54,9 @@ from synapse.replication.http.presence import ( ReplicationBumpPresenceActiveTime, ReplicationPresenceSetState, ) +from synapse.replication.http.streams import ReplicationGetStreamUpdates from synapse.replication.tcp.commands import ClearUserSyncsCommand +from synapse.replication.tcp.streams import PresenceFederationStream, PresenceStream from synapse.state import StateHandler from synapse.storage.databases.main import DataStore from synapse.types import Collection, JsonDict, UserID, get_domain_from_id @@ -128,10 +131,10 @@ class BasePresenceHandler(abc.ABC): self.is_mine_id = hs.is_mine_id self._federation = None - if hs.should_send_federation() or not hs.config.worker_app: + if hs.should_send_federation(): self._federation = hs.get_federation_sender() - self._send_federation = hs.should_send_federation() + self._federation_queue = PresenceFederationQueue(hs, self) self._busy_presence_enabled = hs.config.experimental.msc3026_enabled @@ -254,9 +257,17 @@ class BasePresenceHandler(abc.ABC): """ pass - async def process_replication_rows(self, token, rows): - """Process presence stream rows received over replication.""" - pass + async def process_replication_rows( + self, stream_name: str, instance_name: str, token: int, rows: list + ): + """Process streams received over replication.""" + await self._federation_queue.process_replication_rows( + stream_name, instance_name, token, rows + ) + + def get_federation_queue(self) -> "PresenceFederationQueue": + """Get the presence federation queue.""" + return self._federation_queue async def maybe_send_presence_to_interested_destinations( self, states: List[UserPresenceState] @@ -266,12 +277,9 @@ class BasePresenceHandler(abc.ABC): users. """ - if not self._send_federation: + if not self._federation: return - # If this worker sends federation we must have a FederationSender. - assert self._federation - states = [s for s in states if self.is_mine_id(s.user_id)] if not states: @@ -427,7 +435,14 @@ class WorkerPresenceHandler(BasePresenceHandler): # If this is a federation sender, notify about presence updates. await self.maybe_send_presence_to_interested_destinations(states) - async def process_replication_rows(self, token, rows): + async def process_replication_rows( + self, stream_name: str, instance_name: str, token: int, rows: list + ): + await super().process_replication_rows(stream_name, instance_name, token, rows) + + if stream_name != PresenceStream.NAME: + return + states = [ UserPresenceState( row.user_id, @@ -729,12 +744,10 @@ class PresenceHandler(BasePresenceHandler): self.state, ) - # Since this is master we know that we have a federation sender or - # queue, and so this will be defined. - assert self._federation - for destinations, states in hosts_and_states: - self._federation.send_presence_to_destinations(states, destinations) + self._federation_queue.send_presence_to_destinations( + states, destinations + ) async def _handle_timeouts(self): """Checks the presence of users that have timed out and updates as @@ -1213,13 +1226,9 @@ class PresenceHandler(BasePresenceHandler): user_presence_states ) - # Since this is master we know that we have a federation sender or - # queue, and so this will be defined. - assert self._federation - # Send out user presence updates for each destination for destination, user_state_set in presence_destinations.items(): - self._federation.send_presence_to_destinations( + self._federation_queue.send_presence_to_destinations( destinations=[destination], states=user_state_set ) @@ -1864,3 +1873,197 @@ async def get_interested_remotes( hosts_and_states.append(([host], states)) return hosts_and_states + + +class PresenceFederationQueue: + """Handles sending ad hoc presence updates over federation, which are *not* + due to state updates (that get handled via the presence stream), e.g. + federation pings and sending existing present states to newly joined hosts. + + Only the last N minutes will be queued, so if a federation sender instance + is down for longer then some updates will be dropped. This is OK as presence + is ephemeral, and so it will self correct eventually. + + On workers the class tracks the last received position of the stream from + replication, and handles querying for missed updates over HTTP replication, + c.f. `get_current_token` and `get_replication_rows`. + """ + + # How long to keep entries in the queue for. Workers that are down for + # longer than this duration will miss out on older updates. + _KEEP_ITEMS_IN_QUEUE_FOR_MS = 5 * 60 * 1000 + + # How often to check if we can expire entries from the queue. + _CLEAR_ITEMS_EVERY_MS = 60 * 1000 + + def __init__(self, hs: "HomeServer", presence_handler: BasePresenceHandler): + self._clock = hs.get_clock() + self._notifier = hs.get_notifier() + self._instance_name = hs.get_instance_name() + self._presence_handler = presence_handler + self._repl_client = ReplicationGetStreamUpdates.make_client(hs) + + # Should we keep a queue of recent presence updates? We only bother if + # another process may be handling federation sending. + self._queue_presence_updates = True + + # Whether this instance is a presence writer. + self._presence_writer = hs.config.worker.worker_app is None + + # The FederationSender instance, if this process sends federation traffic directly. + self._federation = None + + if hs.should_send_federation(): + self._federation = hs.get_federation_sender() + + # We don't bother queuing up presence states if only this instance + # is sending federation. + if hs.config.worker.federation_shard_config.instances == [ + self._instance_name + ]: + self._queue_presence_updates = False + + # The queue of recently queued updates as tuples of: `(timestamp, + # stream_id, destinations, user_ids)`. We don't store the full states + # for efficiency, and remote workers will already have the full states + # cached. + self._queue = [] # type: List[Tuple[int, int, Collection[str], Set[str]]] + + self._next_id = 1 + + # Map from instance name to current token + self._current_tokens = {} # type: Dict[str, int] + + if self._queue_presence_updates: + self._clock.looping_call(self._clear_queue, self._CLEAR_ITEMS_EVERY_MS) + + def _clear_queue(self): + """Clear out older entries from the queue.""" + clear_before = self._clock.time_msec() - self._KEEP_ITEMS_IN_QUEUE_FOR_MS + + # The queue is sorted by timestamp, so we can bisect to find the right + # place to purge before. Note that we are searching using a 1-tuple with + # the time, which does The Right Thing since the queue is a tuple where + # the first item is a timestamp. + index = bisect(self._queue, (clear_before,)) + self._queue = self._queue[index:] + + def send_presence_to_destinations( + self, states: Collection[UserPresenceState], destinations: Collection[str] + ) -> None: + """Send the presence states to the given destinations. + + Will forward to the local federation sender (if there is one) and queue + to send over replication (if there are other federation sender instances.). + + Must only be called on the master process. + """ + + # This should only be called on a presence writer. + assert self._presence_writer + + if self._federation: + self._federation.send_presence_to_destinations( + states=states, + destinations=destinations, + ) + + if not self._queue_presence_updates: + return + + now = self._clock.time_msec() + + stream_id = self._next_id + self._next_id += 1 + + self._queue.append((now, stream_id, destinations, {s.user_id for s in states})) + + self._notifier.notify_replication() + + def get_current_token(self, instance_name: str) -> int: + """Get the current position of the stream. + + On workers this returns the last stream ID received from replication. + """ + if instance_name == self._instance_name: + return self._next_id - 1 + else: + return self._current_tokens.get(instance_name, 0) + + async def get_replication_rows( + self, + instance_name: str, + from_token: int, + upto_token: int, + target_row_count: int, + ) -> Tuple[List[Tuple[int, Tuple[str, str]]], int, bool]: + """Get all the updates between the two tokens. + + We return rows in the form of `(destination, user_id)` to keep the size + of each row bounded (rather than returning the sets in a row). + + On workers this will query the master process via HTTP replication. + """ + if instance_name != self._instance_name: + # If not local we query over http replication from the master + result = await self._repl_client( + instance_name=instance_name, + stream_name=PresenceFederationStream.NAME, + from_token=from_token, + upto_token=upto_token, + ) + return result["updates"], result["upto_token"], result["limited"] + + # We can find the correct position in the queue by noting that there is + # exactly one entry per stream ID, and that the last entry has an ID of + # `self._next_id - 1`, so we can count backwards from the end. + # + # Since the start of the queue is periodically truncated we need to + # handle the case where `from_token` stream ID has already been dropped. + start_idx = max(from_token - self._next_id, -len(self._queue)) + + to_send = [] # type: List[Tuple[int, Tuple[str, str]]] + limited = False + new_id = upto_token + for _, stream_id, destinations, user_ids in self._queue[start_idx:]: + if stream_id > upto_token: + break + + new_id = stream_id + + to_send.extend( + (stream_id, (destination, user_id)) + for destination in destinations + for user_id in user_ids + ) + + if len(to_send) > target_row_count: + limited = True + break + + return to_send, new_id, limited + + async def process_replication_rows( + self, stream_name: str, instance_name: str, token: int, rows: list + ): + if stream_name != PresenceFederationStream.NAME: + return + + # We keep track of the current tokens (so that we can catch up with anything we missed after a disconnect) + self._current_tokens[instance_name] = token + + # If we're a federation sender we pull out the presence states to send + # and forward them on. + if not self._federation: + return + + hosts_to_users = {} # type: Dict[str, Set[str]] + for row in rows: + hosts_to_users.setdefault(row.destination, set()).add(row.user_id) + + for host, user_ids in hosts_to_users.items(): + states = await self._presence_handler.current_state_for_users(user_ids) + self._federation.send_presence_to_destinations( + states=states.values(), + destinations=[host], + ) diff --git a/synapse/replication/tcp/client.py b/synapse/replication/tcp/client.py index ce5d651cb..4f3c6a18b 100644 --- a/synapse/replication/tcp/client.py +++ b/synapse/replication/tcp/client.py @@ -29,7 +29,6 @@ from synapse.replication.tcp.streams import ( AccountDataStream, DeviceListsStream, GroupServerStream, - PresenceStream, PushersStream, PushRulesStream, ReceiptsStream, @@ -191,8 +190,6 @@ class ReplicationDataHandler: self.stop_pusher(row.user_id, row.app_id, row.pushkey) else: await self.start_pusher(row.user_id, row.app_id, row.pushkey) - elif stream_name == PresenceStream.NAME: - await self._presence_handler.process_replication_rows(token, rows) elif stream_name == EventsStream.NAME: # We shouldn't get multiple rows per token for events stream, so # we don't need to optimise this for multiple rows. @@ -221,6 +218,10 @@ class ReplicationDataHandler: membership=row.data.membership, ) + await self._presence_handler.process_replication_rows( + stream_name, instance_name, token, rows + ) + # Notify any waiting deferreds. The list is ordered by position so we # just iterate through the list until we reach a position that is # greater than the received row position. diff --git a/synapse/replication/tcp/streams/__init__.py b/synapse/replication/tcp/streams/__init__.py index fb74ac4e9..4c0023c68 100644 --- a/synapse/replication/tcp/streams/__init__.py +++ b/synapse/replication/tcp/streams/__init__.py @@ -30,6 +30,7 @@ from synapse.replication.tcp.streams._base import ( CachesStream, DeviceListsStream, GroupServerStream, + PresenceFederationStream, PresenceStream, PublicRoomsStream, PushersStream, @@ -50,6 +51,7 @@ STREAMS_MAP = { EventsStream, BackfillStream, PresenceStream, + PresenceFederationStream, TypingStream, ReceiptsStream, PushRulesStream, @@ -71,6 +73,7 @@ __all__ = [ "Stream", "BackfillStream", "PresenceStream", + "PresenceFederationStream", "TypingStream", "ReceiptsStream", "PushRulesStream", diff --git a/synapse/replication/tcp/streams/_base.py b/synapse/replication/tcp/streams/_base.py index 520c45f15..9d75a89f1 100644 --- a/synapse/replication/tcp/streams/_base.py +++ b/synapse/replication/tcp/streams/_base.py @@ -290,6 +290,30 @@ class PresenceStream(Stream): ) +class PresenceFederationStream(Stream): + """A stream used to send ad hoc presence updates over federation. + + Streams the remote destination and the user ID of the presence state to + send. + """ + + @attr.s(slots=True, auto_attribs=True) + class PresenceFederationStreamRow: + destination: str + user_id: str + + NAME = "presence_federation" + ROW_TYPE = PresenceFederationStreamRow + + def __init__(self, hs: "HomeServer"): + federation_queue = hs.get_presence_handler().get_federation_queue() + super().__init__( + hs.get_instance_name(), + federation_queue.get_current_token, + federation_queue.get_replication_rows, + ) + + class TypingStream(Stream): TypingStreamRow = namedtuple( "TypingStreamRow", ("room_id", "user_ids") # str # list(str) diff --git a/tests/handlers/test_presence.py b/tests/handlers/test_presence.py index 2d12e8289..61271cd08 100644 --- a/tests/handlers/test_presence.py +++ b/tests/handlers/test_presence.py @@ -21,6 +21,7 @@ from synapse.api.constants import EventTypes, Membership, PresenceState from synapse.api.presence import UserPresenceState from synapse.api.room_versions import KNOWN_ROOM_VERSIONS from synapse.events.builder import EventBuilder +from synapse.federation.sender import FederationSender from synapse.handlers.presence import ( EXTERNAL_PROCESS_EXPIRY, FEDERATION_PING_INTERVAL, @@ -471,6 +472,168 @@ class PresenceHandlerTestCase(unittest.HomeserverTestCase): self.assertEqual(state.state, PresenceState.OFFLINE) +class PresenceFederationQueueTestCase(unittest.HomeserverTestCase): + def prepare(self, reactor, clock, hs): + self.presence_handler = hs.get_presence_handler() + self.clock = hs.get_clock() + self.instance_name = hs.get_instance_name() + + self.queue = self.presence_handler.get_federation_queue() + + def test_send_and_get(self): + state1 = UserPresenceState.default("@user1:test") + state2 = UserPresenceState.default("@user2:test") + state3 = UserPresenceState.default("@user3:test") + + prev_token = self.queue.get_current_token(self.instance_name) + + self.queue.send_presence_to_destinations((state1, state2), ("dest1", "dest2")) + self.queue.send_presence_to_destinations((state3,), ("dest3",)) + + now_token = self.queue.get_current_token(self.instance_name) + + rows, upto_token, limited = self.get_success( + self.queue.get_replication_rows("master", prev_token, now_token, 10) + ) + + self.assertEqual(upto_token, now_token) + self.assertFalse(limited) + + expected_rows = [ + (1, ("dest1", "@user1:test")), + (1, ("dest2", "@user1:test")), + (1, ("dest1", "@user2:test")), + (1, ("dest2", "@user2:test")), + (2, ("dest3", "@user3:test")), + ] + + self.assertCountEqual(rows, expected_rows) + + def test_send_and_get_split(self): + state1 = UserPresenceState.default("@user1:test") + state2 = UserPresenceState.default("@user2:test") + state3 = UserPresenceState.default("@user3:test") + + prev_token = self.queue.get_current_token(self.instance_name) + + self.queue.send_presence_to_destinations((state1, state2), ("dest1", "dest2")) + + now_token = self.queue.get_current_token(self.instance_name) + + self.queue.send_presence_to_destinations((state3,), ("dest3",)) + + rows, upto_token, limited = self.get_success( + self.queue.get_replication_rows("master", prev_token, now_token, 10) + ) + + self.assertEqual(upto_token, now_token) + self.assertFalse(limited) + + expected_rows = [ + (1, ("dest1", "@user1:test")), + (1, ("dest2", "@user1:test")), + (1, ("dest1", "@user2:test")), + (1, ("dest2", "@user2:test")), + ] + + self.assertCountEqual(rows, expected_rows) + + def test_clear_queue_all(self): + state1 = UserPresenceState.default("@user1:test") + state2 = UserPresenceState.default("@user2:test") + state3 = UserPresenceState.default("@user3:test") + + prev_token = self.queue.get_current_token(self.instance_name) + + self.queue.send_presence_to_destinations((state1, state2), ("dest1", "dest2")) + self.queue.send_presence_to_destinations((state3,), ("dest3",)) + + self.reactor.advance(10 * 60 * 1000) + + now_token = self.queue.get_current_token(self.instance_name) + + rows, upto_token, limited = self.get_success( + self.queue.get_replication_rows("master", prev_token, now_token, 10) + ) + self.assertEqual(upto_token, now_token) + self.assertFalse(limited) + self.assertCountEqual(rows, []) + + prev_token = self.queue.get_current_token(self.instance_name) + + self.queue.send_presence_to_destinations((state1, state2), ("dest1", "dest2")) + self.queue.send_presence_to_destinations((state3,), ("dest3",)) + + now_token = self.queue.get_current_token(self.instance_name) + + rows, upto_token, limited = self.get_success( + self.queue.get_replication_rows("master", prev_token, now_token, 10) + ) + self.assertEqual(upto_token, now_token) + self.assertFalse(limited) + + expected_rows = [ + (3, ("dest1", "@user1:test")), + (3, ("dest2", "@user1:test")), + (3, ("dest1", "@user2:test")), + (3, ("dest2", "@user2:test")), + (4, ("dest3", "@user3:test")), + ] + + self.assertCountEqual(rows, expected_rows) + + def test_partially_clear_queue(self): + state1 = UserPresenceState.default("@user1:test") + state2 = UserPresenceState.default("@user2:test") + state3 = UserPresenceState.default("@user3:test") + + prev_token = self.queue.get_current_token(self.instance_name) + + self.queue.send_presence_to_destinations((state1, state2), ("dest1", "dest2")) + + self.reactor.advance(2 * 60 * 1000) + + self.queue.send_presence_to_destinations((state3,), ("dest3",)) + + self.reactor.advance(4 * 60 * 1000) + + now_token = self.queue.get_current_token(self.instance_name) + + rows, upto_token, limited = self.get_success( + self.queue.get_replication_rows("master", prev_token, now_token, 10) + ) + self.assertEqual(upto_token, now_token) + self.assertFalse(limited) + + expected_rows = [ + (2, ("dest3", "@user3:test")), + ] + self.assertCountEqual(rows, []) + + prev_token = self.queue.get_current_token(self.instance_name) + + self.queue.send_presence_to_destinations((state1, state2), ("dest1", "dest2")) + self.queue.send_presence_to_destinations((state3,), ("dest3",)) + + now_token = self.queue.get_current_token(self.instance_name) + + rows, upto_token, limited = self.get_success( + self.queue.get_replication_rows("master", prev_token, now_token, 10) + ) + self.assertEqual(upto_token, now_token) + self.assertFalse(limited) + + expected_rows = [ + (3, ("dest1", "@user1:test")), + (3, ("dest2", "@user1:test")), + (3, ("dest1", "@user2:test")), + (3, ("dest2", "@user2:test")), + (4, ("dest3", "@user3:test")), + ] + + self.assertCountEqual(rows, expected_rows) + + class PresenceJoinTestCase(unittest.HomeserverTestCase): """Tests remote servers get told about presence of users in the room when they join and when new local users join. @@ -482,10 +645,17 @@ class PresenceJoinTestCase(unittest.HomeserverTestCase): def make_homeserver(self, reactor, clock): hs = self.setup_test_homeserver( - "server", federation_http_client=None, federation_sender=Mock() + "server", + federation_http_client=None, + federation_sender=Mock(spec=FederationSender), ) return hs + def default_config(self): + config = super().default_config() + config["send_federation"] = True + return config + def prepare(self, reactor, clock, hs): self.federation_sender = hs.get_federation_sender() self.event_builder_factory = hs.get_event_builder_factory() @@ -529,9 +699,6 @@ class PresenceJoinTestCase(unittest.HomeserverTestCase): # Add a new remote server to the room self._add_new_user(room_id, "@alice:server2") - # We shouldn't have sent out any local presence *updates* - self.federation_sender.send_presence.assert_not_called() - # When new server is joined we send it the local users presence states. # We expect to only see user @test2:server, as @test:server is offline # and has a zero last_active_ts @@ -550,7 +717,6 @@ class PresenceJoinTestCase(unittest.HomeserverTestCase): self.federation_sender.reset_mock() self._add_new_user(room_id, "@bob:server3") - self.federation_sender.send_presence.assert_not_called() self.federation_sender.send_presence_to_destinations.assert_called_once_with( destinations=["server3"], states={expected_state} ) @@ -595,9 +761,6 @@ class PresenceJoinTestCase(unittest.HomeserverTestCase): self.reactor.pump([0]) # Wait for presence updates to be handled - # We shouldn't have sent out any local presence *updates* - self.federation_sender.send_presence.assert_not_called() - # We expect to only send test2 presence to server2 and server3 expected_state = self.get_success( self.presence_handler.current_state_for_user("@test2:server")