Add missing break out seqscan loop in logical replication

alvherre · alvherre · commit 380bc8829594 · 2020-02-03T18:59:12.000-03:00
When replica identity is FULL (an admittedly unusual case), the loop that searches for tuples in execReplication.c didn't stop scanning the table when once a matching tuple was found. Add the missing 'break'. Note slight behavior change: we now return the first matching tuple rather than the last one. They are supposed to be indistinguishable anyway, so this shouldn't matter. Author: Konstantin Knizhnik Discussion: https://postgr.es/m/379743f6-ae91-b866-f7a2-5624e6d2b0a4@postgrespro.ru
diff --git a/src/backend/executor/execReplication.c b/src/backend/executor/execReplication.c
@@ -314,6 +314,9 @@ RelationFindReplTupleSeq(Relation rel, LockTupleMode lockmode,
 			XactLockTableWait(xwait, NULL, NULL, XLTW_None);
 			goto retry;
 		}
+
+		/* Found our tuple and it's not locked */
+		break;
 	}
 
 	/* Found tuple, try to lock it in the lockmode. */

Original file line number	Diff line number	Diff line change
`@@ -314,6 +314,9 @@ RelationFindReplTupleSeq(Relation rel, LockTupleMode lockmode,`
`314`	`314`	`XactLockTableWait(xwait, NULL, NULL, XLTW_None);`
`315`	`315`	`goto retry;`
`316`	`316`	`}`
	`317`	`+`
	`318`	`+ /* Found our tuple and it's not locked */`
	`319`	`+ break;`
`317`	`320`	`}`
`318`	`321`
`319`	`322`	`/* Found tuple, try to lock it in the lockmode. */`