Faster joins: Update room stats and the user directory on workers whe…

…n finishing join (#14874) * Faster joins: Update room stats and user directory on workers when done When finishing a partial state join to a room, we update the current state of the room without persisting additional events. Workers receive notice of the current state update over replication, but neglect to wake the room stats and user directory updaters, which then get incidentally triggered the next time an event is persisted or an unrelated event persister sends out a stream position update. We wake the room stats and user directory updaters at the appropriate time in this commit. Part of #12814 and #12815. Signed-off-by: Sean Quah <[email protected]> * fixup comment Signed-off-by: Sean Quah <[email protected]>
matrix-org · Jan 23, 2023 · 2ec9c58 · 2ec9c58
1 parent 22cc93a
commit 2ec9c58
Show file tree

Hide file tree

Showing 4 changed files with 11 additions and 5 deletions.
diff --git a/changelog.d/14874.bugfix b/changelog.d/14874.bugfix
@@ -0,0 +1 @@
+Faster joins: Fix a bug in worker deployments where the room stats and user directory would not get updated when finishing a fast join until another event is sent or received.
diff --git a/synapse/handlers/federation.py b/synapse/handlers/federation.py
@@ -1880,9 +1880,10 @@ async def _sync_partial_state_room(
                         room_id
                     )
 
-                    # TODO(faster_joins) update room stats and user directory?
-                    #   https://github.com/matrix-org/synapse/issues/12814
-                    #   https://github.com/matrix-org/synapse/issues/12815
+                    # Poke the notifier so that other workers see the write to
+                    # the un-partial-stated rooms stream.
+                    self._notifier.notify_replication()
+
                     return
 
                 # we raced against more events arriving with partial state. Go round

diff --git a/synapse/replication/tcp/client.py b/synapse/replication/tcp/client.py
@@ -207,6 +207,12 @@ async def on_rdata(
             # we don't need to optimise this for multiple rows.
             for row in rows:
                 if row.type != EventsStreamEventRow.TypeId:
+                    # The row's data is an `EventsStreamCurrentStateRow`.
+                    # When we recompute the current state of a room based on forward
+                    # extremities (see `update_current_state`), no new events are
+                    # persisted, so we must poke the replication callbacks ourselves.
+                    # This functionality is used when finishing up a partial state join.
+                    self.notifier.notify_replication()
                     continue
                 assert isinstance(row, EventsStreamRow)
                 assert isinstance(row.data, EventsStreamEventRow)

diff --git a/synapse/storage/controllers/state.py b/synapse/storage/controllers/state.py
@@ -493,8 +493,6 @@ async def get_current_state_deltas(
                  up to date.
         """
         # FIXME(faster_joins): what do we do here?
-        #   https://github.com/matrix-org/synapse/issues/12814
-        #   https://github.com/matrix-org/synapse/issues/12815
         #   https://github.com/matrix-org/synapse/issues/13008
 
         return await self.stores.main.get_partial_current_state_deltas(