Faster room joins: fix race in recalculation of current room state (#13151)

Bounce recalculation of current state to the correct event persister and
move recalculation of current state into the event persistence queue, to
avoid concurrent updates to a room's current state.

Also give recalculation of a room's current state a real stream
ordering.

Signed-off-by: Sean Quah <seanq@matrix.org>

1 files changed, 2 insertions, 7 deletions

diff --git a/synapse/handlers/federation.py b/synapse/handlers/federation.py

index 3c44b4bf86..e2564e9340 100644
--- a/synapse/handlers/federation.py
+++ b/synapse/handlers/federation.py
@@ -1559,14 +1559,9 @@ class FederationHandler:
                 # all the events are updated, so we can update current state and
                 # clear the lazy-loading flag.
                 logger.info("Updating current state for %s", room_id)
-                # TODO(faster_joins): support workers
+                # TODO(faster_joins): notify workers in notify_room_un_partial_stated
                 #   https://github.com/matrix-org/synapse/issues/12994
-                assert (
-                    self._storage_controllers.persistence is not None
-                ), "worker-mode deployments not currently supported here"
-                await self._storage_controllers.persistence.update_current_state(
-                    room_id
-                )
+                await self.state_handler.update_current_state(room_id)
 
                 logger.info("Clearing partial-state flag for %s", room_id)
                 success = await self.store.clear_partial_state_room(room_id)