summary refs log tree commit diff
path: root/synapse/storage
diff options
context:
space:
mode:
Diffstat (limited to 'synapse/storage')
-rw-r--r--synapse/storage/databases/main/events_worker.py20
1 files changed, 19 insertions, 1 deletions
diff --git a/synapse/storage/databases/main/events_worker.py b/synapse/storage/databases/main/events_worker.py
index 5914a35420..29c99c6357 100644
--- a/synapse/storage/databases/main/events_worker.py
+++ b/synapse/storage/databases/main/events_worker.py
@@ -2110,11 +2110,29 @@ class EventsWorkerStore(SQLBaseStore):
     def _get_partial_state_events_batch_txn(
         txn: LoggingTransaction, room_id: str
     ) -> List[str]:
+        # we want to work through the events from oldest to newest, so
+        # we only want events whose prev_events do *not* have partial state - hence
+        # the 'NOT EXISTS' clause in the below.
+        #
+        # This is necessary because ordering by stream ordering isn't quite enough
+        # to ensure that we work from oldest to newest event (in particular,
+        # if an event is initially persisted as an outlier and later de-outliered,
+        # it can end up with a lower stream_ordering than its prev_events).
+        #
+        # Typically this means we'll only return one event per batch, but that's
+        # hard to do much about.
+        #
+        # See also: https://github.com/matrix-org/synapse/issues/13001
         txn.execute(
             """
             SELECT event_id FROM partial_state_events AS pse
                 JOIN events USING (event_id)
-            WHERE pse.room_id = ?
+            WHERE pse.room_id = ? AND
+               NOT EXISTS(
+                  SELECT 1 FROM event_edges AS ee
+                     JOIN partial_state_events AS prev_pse ON (prev_pse.event_id=ee.prev_event_id)
+                     WHERE ee.event_id=pse.event_id
+               )
             ORDER BY events.stream_ordering
             LIMIT 100
             """,