Fix unread counts on large servers

changelog.d/13140.bugfix

@@ -0,0 +1 @@
+Fix unread counts for users on large servers. Introduced in v1.62.0rc1.

synapse/storage/databases/main/event_push_actions.py

@@ -861,11 +861,13 @@ def _handle_new_receipts_for_notifs_txn(self, txn: LoggingTransaction) -> bool:
             retcol="stream_id",
         )
 
+        max_receipts_stream_id = self._receipts_id_gen.get_current_token()
+
         sql = """
             SELECT r.stream_id, r.room_id, r.user_id, e.stream_ordering
             FROM receipts_linearized AS r
             INNER JOIN events AS e USING (event_id)
-            WHERE r.stream_id > ? AND user_id LIKE ?
+            WHERE ? < r.stream_id AND r.stream_id <= ? AND user_id LIKE ?
             ORDER BY r.stream_id ASC
             LIMIT ?
         """
@@ -884,6 +886,13 @@ def _handle_new_receipts_for_notifs_txn(self, txn: LoggingTransaction) -> bool:
         )
         rows = txn.fetchall()
 
+        old_rotate_stream_ordering = self.db_pool.simple_select_one_onecol_txn(
+            txn,
+            table="event_push_summary_stream_ordering",
+            keyvalues={},
+            retcol="stream_ordering",
+        )
+
         # For each new read receipt we delete push actions from before it and
         # recalculate the summary.
         for _, room_id, user_id, stream_ordering in rows:
@@ -902,13 +911,6 @@ def _handle_new_receipts_for_notifs_txn(self, txn: LoggingTransaction) -> bool:
                 (room_id, user_id, stream_ordering),
             )
 
-            old_rotate_stream_ordering = self.db_pool.simple_select_one_onecol_txn(
-                txn,
-                table="event_push_summary_stream_ordering",
-                keyvalues={},
-                retcol="stream_ordering",
-            )
-
             notif_count, unread_count = self._get_notif_unread_count_for_user_room(
                 txn, room_id, user_id, stream_ordering, old_rotate_stream_ordering
             )
@@ -927,18 +929,16 @@ def _handle_new_receipts_for_notifs_txn(self, txn: LoggingTransaction) -> bool:
 
         # We always update `event_push_summary_last_receipt_stream_id` to
         # ensure that we don't rescan the same receipts for remote users.
-        #
-        # This requires repeatable read to be safe, as we need the
-        # `MAX(stream_id)` to not include any new rows that have been committed
-        # since the start of the transaction (since those rows won't have been
-        # returned by the query above). Alternatively we could query the max
-        # stream ID at the start of the transaction and bound everything by
-        # that.
-        txn.execute(
-            """
-            UPDATE event_push_summary_last_receipt_stream_id
-            SET stream_id = (SELECT COALESCE(MAX(stream_id), 0) FROM receipts_linearized)
-            """
+
+        upper_limit = max_receipts_stream_id
+        if len(rows) >= limit:
+            upper_limit = rows[-1][0]
+
+        self.db_pool.simple_update_txn(
+            txn,
+            table="event_push_summary_last_receipt_stream_id",
+            keyvalues={},
+            updatevalues={"stream_id": upper_limit},
         )
 
         return len(rows) < limit