Fix partition table's REPLICA IDENTITY checking on the subscriber.

In logical replication, we will check if the target table on the
subscriber is updatable by comparing the replica identity of the table on
the publisher with the table on the subscriber. When the target table is a
partitioned table, we only check its replica identity but not for the
partition tables. This leads to assertion failure while applying changes
for update/delete as we expect those to succeed only when the
corresponding partition table has a primary key or has a replica
identity defined.

Fix it by checking the replica identity of the partition table while
applying changes.

Reported-by: Shi Yu
Author: Shi Yu, Hou Zhijie
Reviewed-by: Amit Langote, Amit Kapila
Backpatch-through: 13, where it was introduced
Discussion: https://postgr.es/m/OSZPR01MB6310F46CD425A967E4AEF736FDA49@OSZPR01MB6310.jpnprd01.prod.outlook.com

src/backend/replication/logical/relation.c

@@ -249,6 +249,67 @@ logicalrep_report_missing_attrs(LogicalRepRelation *remoterel,
 	}
 }
 
+/*
+ * Check if replica identity matches and mark the updatable flag.
+ *
+ * We allow for stricter replica identity (fewer columns) on subscriber as
+ * that will not stop us from finding unique tuple. IE, if publisher has
+ * identity (id,timestamp) and subscriber just (id) this will not be a
+ * problem, but in the opposite scenario it will.
+ *
+ * We just mark the relation entry as not updatable here if the local
+ * replica identity is found to be insufficient for applying
+ * updates/deletes (inserts don't care!) and leave it to
+ * check_relation_updatable() to throw the actual error if needed.
+ */
+static void
+logicalrep_rel_mark_updatable(LogicalRepRelMapEntry *entry)
+{
+	Bitmapset  *idkey;
+	LogicalRepRelation *remoterel = &entry->remoterel;
+	int			i;
+
+	entry->updatable = true;
+
+	idkey = RelationGetIndexAttrBitmap(entry->localrel,
+									   INDEX_ATTR_BITMAP_IDENTITY_KEY);
+	/* fallback to PK if no replica identity */
+	if (idkey == NULL)
+	{
+		idkey = RelationGetIndexAttrBitmap(entry->localrel,
+										   INDEX_ATTR_BITMAP_PRIMARY_KEY);
+
+		/*
+		 * If no replica identity index and no PK, the published table must
+		 * have replica identity FULL.
+		 */
+		if (idkey == NULL && remoterel->replident != REPLICA_IDENTITY_FULL)
+			entry->updatable = false;
+	}
+
+	i = -1;
+	while ((i = bms_next_member(idkey, i)) >= 0)
+	{
+		int			attnum = i + FirstLowInvalidHeapAttributeNumber;
+
+		if (!AttrNumberIsForUserDefinedAttr(attnum))
+			ereport(ERROR,
+					(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
+					 errmsg("logical replication target relation \"%s.%s\" uses "
+							"system columns in REPLICA IDENTITY index",
+							remoterel->nspname, remoterel->relname)));
+
+		attnum = AttrNumberGetAttrOffset(attnum);
+
+		if (entry->attrmap->attnums[attnum] < 0 ||
+			!bms_is_member(entry->attrmap->attnums[attnum], remoterel->attkeys))
+		{
+			entry->updatable = false;
+			break;
+		}
+	}
+}
+
 /*
  * Open the local relation associated with the remote one.
  *
@@ -307,7 +368,6 @@ logicalrep_rel_open(LogicalRepRelId remoteid, LOCKMODE lockmode)
 	if (!entry->localrelvalid)
 	{
 		Oid			relid;
-		Bitmapset  *idkey;
 		TupleDesc	desc;
 		MemoryContext oldctx;
 		int			i;
@@ -366,54 +426,10 @@ logicalrep_rel_open(LogicalRepRelId remoteid, LOCKMODE lockmode)
 		bms_free(missingatts);
 
 		/*
-		 * Check that replica identity matches. We allow for stricter replica
+		 * Set if the table's replica identity is enough to apply
-		 * identity (fewer columns) on subscriber as that will not stop us
+		 * update/delete.
-		 * from finding unique tuple. IE, if publisher has identity
-		 * (id,timestamp) and subscriber just (id) this will not be a problem,
-		 * but in the opposite scenario it will.
-		 *
-		 * Don't throw any error here just mark the relation entry as not
-		 * updatable, as replica identity is only for updates and deletes but
-		 * inserts can be replicated even without it.
 		 */
-		entry->updatable = true;
+		logicalrep_rel_mark_updatable(entry);
-		idkey = RelationGetIndexAttrBitmap(entry->localrel,
-										   INDEX_ATTR_BITMAP_IDENTITY_KEY);
-		/* fallback to PK if no replica identity */
-		if (idkey == NULL)
-		{
-			idkey = RelationGetIndexAttrBitmap(entry->localrel,
-											   INDEX_ATTR_BITMAP_PRIMARY_KEY);
-
-			/*
-			 * If no replica identity index and no PK, the published table
-			 * must have replica identity FULL.
-			 */
-			if (idkey == NULL && remoterel->replident != REPLICA_IDENTITY_FULL)
-				entry->updatable = false;
-		}
-
-		i = -1;
-		while ((i = bms_next_member(idkey, i)) >= 0)
-		{
-			int			attnum = i + FirstLowInvalidHeapAttributeNumber;
-
-			if (!AttrNumberIsForUserDefinedAttr(attnum))
-				ereport(ERROR,
-						(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
-						 errmsg("logical replication target relation \"%s.%s\" uses "
-								"system columns in REPLICA IDENTITY index",
-								remoterel->nspname, remoterel->relname)));
-
-			attnum = AttrNumberGetAttrOffset(attnum);
-
-			if (entry->attrmap->attnums[attnum] < 0 ||
-				!bms_is_member(entry->attrmap->attnums[attnum], remoterel->attkeys))
-			{
-				entry->updatable = false;
-				break;
-			}
-		}
 
 		entry->localrelvalid = true;
 	}
@@ -651,7 +667,8 @@ logicalrep_partition_open(LogicalRepRelMapEntry *root,
 			   attrmap->maplen * sizeof(AttrNumber));
 	}
 
-	entry->updatable = root->updatable;
+	/* Set if the table's replica identity is enough to apply update/delete. */
+	logicalrep_rel_mark_updatable(entry);
 
 	entry->localrelvalid = true;
 

src/backend/replication/logical/worker.c

@@ -1738,6 +1738,13 @@ apply_handle_insert_internal(ApplyExecutionData *edata,
 static void
 check_relation_updatable(LogicalRepRelMapEntry *rel)
 {
+	/*
+	 * For partitioned tables, we only need to care if the target partition is
+	 * updatable (aka has PK or RI defined for it).
+	 */
+	if (rel->localrel->rd_rel->relkind == RELKIND_PARTITIONED_TABLE)
+		return;
+
 	/* Updatable, no error. */
 	if (rel->updatable)
 		return;
@@ -2121,6 +2128,8 @@ apply_handle_tuple_routing(ApplyExecutionData *edata,
 	TupleTableSlot *remoteslot_part;
 	TupleConversionMap *map;
 	MemoryContext oldctx;
+	LogicalRepRelMapEntry *part_entry = NULL;
+	AttrMap    *attrmap = NULL;
 
 	/* ModifyTableState is needed for ExecFindPartition(). */
 	edata->mtstate = mtstate = makeNode(ModifyTableState);
@@ -2152,8 +2161,11 @@ apply_handle_tuple_routing(ApplyExecutionData *edata,
 		remoteslot_part = table_slot_create(partrel, &estate->es_tupleTable);
 	map = partrelinfo->ri_RootToPartitionMap;
 	if (map != NULL)
-		remoteslot_part = execute_attr_map_slot(map->attrMap, remoteslot,
+	{
+		attrmap = map->attrMap;
+		remoteslot_part = execute_attr_map_slot(attrmap, remoteslot,
 												remoteslot_part);
+	}
 	else
 	{
 		remoteslot_part = ExecCopySlot(remoteslot_part, remoteslot);
@@ -2161,6 +2173,14 @@ apply_handle_tuple_routing(ApplyExecutionData *edata,
 	}
 	MemoryContextSwitchTo(oldctx);
 
+	/* Check if we can do the update or delete on the leaf partition. */
+	if (operation == CMD_UPDATE || operation == CMD_DELETE)
+	{
+		part_entry = logicalrep_partition_open(relmapentry, partrel,
+											   attrmap);
+		check_relation_updatable(part_entry);
+	}
+
 	switch (operation)
 	{
 		case CMD_INSERT:
@@ -2182,15 +2202,10 @@ apply_handle_tuple_routing(ApplyExecutionData *edata,
 			 * suitable partition.
 			 */
 			{
-				AttrMap    *attrmap = map ? map->attrMap : NULL;
-				LogicalRepRelMapEntry *part_entry;
 				TupleTableSlot *localslot;
 				ResultRelInfo *partrelinfo_new;
 				bool		found;
 
-				part_entry = logicalrep_partition_open(relmapentry, partrel,
-													   attrmap);
-
 				/* Get the matching local tuple from the partition. */
 				found = FindReplTupleInLocalRel(estate, partrel,
 												&part_entry->remoterel,

src/test/subscription/t/013_partition.pl

@@ -868,4 +868,18 @@ $result = $node_subscriber2->safe_psql('postgres',
 	"SELECT a, b, c FROM tab5 ORDER BY 1");
 is($result, qq(3||1), 'updates of tab5 replicated correctly after altering table on publisher');
 
+# Test that replication works correctly as long as the leaf partition
+# has the necessary REPLICA IDENTITY, even though the actual target
+# partitioned table does not.
+$node_subscriber2->safe_psql('postgres',
+	"ALTER TABLE tab5 REPLICA IDENTITY NOTHING");
+
+$node_publisher->safe_psql('postgres', "UPDATE tab5 SET a = 4 WHERE a = 3");
+
+$node_publisher->wait_for_catchup('sub2');
+
+$result = $node_subscriber2->safe_psql('postgres',
+	"SELECT a, b, c FROM tab5_1 ORDER BY 1");
+is($result, qq(4||1), 'updates of tab5 replicated correctly');
+
 done_testing();