Make federation catchup send last event from any server. (#9640)

Currently federation catchup will send the last *local* event that we
failed to send to the remote. This can cause issues for large rooms
where lots of servers have sent events while the remote server was down,
as when it comes back up again it'll be flooded with events from various
points in the DAG.

Instead, let's make it so that all the servers send the most recent
events, even if its not theirs. The remote should deduplicate the
events, so there shouldn't be much overhead in doing this.
Alternatively, the servers could only send local events if they were
also extremities and hope that the other server will send the event
over, but that is a bit risky.

1 files changed, 49 insertions, 0 deletions

diff --git a/tests/federation/test_federation_catch_up.py b/tests/federation/test_federation_catch_up.py
index 6f96cd7940..95eac6a5a3 100644
--- a/tests/federation/test_federation_catch_up.py
+++ b/tests/federation/test_federation_catch_up.py
@@ -2,6 +2,7 @@ from typing import List, Tuple
 
 from mock import Mock
 
+from synapse.api.constants import EventTypes
 from synapse.events import EventBase
 from synapse.federation.sender import PerDestinationQueue, TransactionManager
 from synapse.federation.units import Edu
@@ -421,3 +422,51 @@ class FederationCatchUpTestCases(FederatingHomeserverTestCase):
         self.assertNotIn("zzzerver", woken)
         # - all destinations are woken exactly once; they appear once in woken.
         self.assertCountEqual(woken, server_names[:-1])
+
+    @override_config({"send_federation": True})
+    def test_not_latest_event(self):
+        """Test that we send the latest event in the room even if its not ours."""
+
+        per_dest_queue, sent_pdus = self.make_fake_destination_queue()
+
+        # Make a room with a local user, and two servers. One will go offline
+        # and one will send some events.
+        self.register_user("u1", "you the one")
+        u1_token = self.login("u1", "you the one")
+        room_1 = self.helper.create_room_as("u1", tok=u1_token)
+
+        self.get_success(
+            event_injection.inject_member_event(self.hs, room_1, "@user:host2", "join")
+        )
+        event_1 = self.get_success(
+            event_injection.inject_member_event(self.hs, room_1, "@user:host3", "join")
+        )
+
+        # First we send something from the local server, so that we notice the
+        # remote is down and go into catchup mode.
+        self.helper.send(room_1, "you hear me!!", tok=u1_token)
+
+        # Now simulate us receiving an event from the still online remote.
+        event_2 = self.get_success(
+            event_injection.inject_event(
+                self.hs,
+                type=EventTypes.Message,
+                sender="@user:host3",
+                room_id=room_1,
+                content={"msgtype": "m.text", "body": "Hello"},
+            )
+        )
+
+        self.get_success(
+            self.hs.get_datastore().set_destination_last_successful_stream_ordering(
+                "host2", event_1.internal_metadata.stream_ordering
+            )
+        )
+
+        self.get_success(per_dest_queue._catch_up_transmission_loop())
+
+        # We expect only the last message from the remote, event_2, to have been
+        # sent, rather than the last *local* event that was sent.
+        self.assertEqual(len(sent_pdus), 1)
+        self.assertEqual(sent_pdus[0].event_id, event_2.event_id)
+        self.assertFalse(per_dest_queue._catching_up)