diff --git a/src/backend/distributed/transaction/relation_access_tracking.c b/src/backend/distributed/transaction/relation_access_tracking.c
index c4335c250..b8ef8ba7b 100644
--- a/src/backend/distributed/transaction/relation_access_tracking.c
+++ b/src/backend/distributed/transaction/relation_access_tracking.c
@@ -713,14 +713,37 @@ CheckConflictingRelationAccesses(Oid relationId, ShardPlacementAccessType access
 		char *conflictingAccessTypeText =
 			PlacementAccessTypeToText(conflictingAccessType);
 
-		ereport(ERROR, (errmsg("cannot execute %s on reference relation \"%s\" because "
-							   "there was a parallel %s access to distributed relation "
-							   "\"%s\" in the same transaction",
-							   accessTypeText, relationName, conflictingAccessTypeText,
-							   conflictingRelationName),
-						errhint("Try re-running the transaction with "
-								"\"SET LOCAL citus.multi_shard_modify_mode TO "
-								"\'sequential\';\"")));
+		/*
+		 * Relation could already be dropped if the accessType is DDL and the
+		 * command that we were executing were a DROP command. In that case,
+		 * as this function is executed via DROP trigger, standard_ProcessUtility
+		 * had already dropped the table from PostgreSQL's perspective. Hence, it
+		 * returns NULL pointer for the name of the relation.
+		 */
+		if (relationName == NULL)
+		{
+			ereport(ERROR, (errmsg("cannot execute %s on reference relation because "
+								   "there was a parallel %s access to distributed relation "
+								   "\"%s\" in the same transaction",
+								   accessTypeText, conflictingAccessTypeText,
+								   conflictingRelationName),
+							errhint("Try re-running the transaction with "
+									"\"SET LOCAL citus.multi_shard_modify_mode TO "
+									"\'sequential\';\"")));
+		}
+		else
+		{
+			ereport(ERROR, (errmsg(
+								"cannot execute %s on reference relation \"%s\" because "
+								"there was a parallel %s access to distributed relation "
+								"\"%s\" in the same transaction",
+								accessTypeText, relationName,
+								conflictingAccessTypeText,
+								conflictingRelationName),
+							errhint("Try re-running the transaction with "
+									"\"SET LOCAL citus.multi_shard_modify_mode TO "
+									"\'sequential\';\"")));
+		}
 	}
 	else if (cacheEntry->referencingRelationsViaForeignKey != NIL &&
 			 accessType > PLACEMENT_ACCESS_SELECT)
diff --git a/src/test/regress/expected/foreign_key_restriction_enforcement.out b/src/test/regress/expected/foreign_key_restriction_enforcement.out
index 0fcade1c5..04209af5c 100644
--- a/src/test/regress/expected/foreign_key_restriction_enforcement.out
+++ b/src/test/regress/expected/foreign_key_restriction_enforcement.out
@@ -857,6 +857,35 @@ BEGIN;
 NOTICE:  truncate cascades to table "reference_table"
 NOTICE:  truncate cascades to table "on_update_fkey_table"
 ROLLBACK;
+-- case 4.7: SELECT to a dist table is followed by a DROP
+-- DROP following SELECT is important as we error out after
+-- the standart process utility hook drops the table.
+-- That could cause SIGSEGV before the patch.
+-- Below block should "successfully" error out
+BEGIN;
+	SELECT count(*) FROM on_update_fkey_table;
+ count
+---------------------------------------------------------------------
+  1001
+(1 row)
+
+	DROP TABLE reference_table CASCADE;
+NOTICE:  drop cascades to constraint fkey on table on_update_fkey_table
+ERROR:  cannot execute DDL on reference relation because there was a parallel SELECT access to distributed relation "on_update_fkey_table" in the same transaction
+ROLLBACK;
+-- case 4.8: Router SELECT to a dist table is followed by a TRUNCATE
+-- No errors expected from below block as SELECT there is a router
+-- query
+BEGIN;
+	SELECT count(*) FROM on_update_fkey_table WHERE id = 9;
+ count
+---------------------------------------------------------------------
+     1
+(1 row)
+
+	DROP TABLE reference_table CASCADE;
+NOTICE:  drop cascades to constraint fkey on table on_update_fkey_table
+ROLLBACK;
 RESET client_min_messages;
 \set VERBOSITY default
 -- case 5.1: Parallel UPDATE on distributed table follow by a SELECT
diff --git a/src/test/regress/sql/foreign_key_restriction_enforcement.sql b/src/test/regress/sql/foreign_key_restriction_enforcement.sql
index 8e482c857..3c2f42ba5 100644
--- a/src/test/regress/sql/foreign_key_restriction_enforcement.sql
+++ b/src/test/regress/sql/foreign_key_restriction_enforcement.sql
@@ -489,6 +489,24 @@ BEGIN;
 	TRUNCATE transitive_reference_table CASCADE;
 ROLLBACK;
 
+-- case 4.7: SELECT to a dist table is followed by a DROP
+-- DROP following SELECT is important as we error out after
+-- the standart process utility hook drops the table.
+-- That could cause SIGSEGV before the patch.
+-- Below block should "successfully" error out
+BEGIN;
+	SELECT count(*) FROM on_update_fkey_table;
+	DROP TABLE reference_table CASCADE;
+ROLLBACK;
+
+-- case 4.8: Router SELECT to a dist table is followed by a TRUNCATE
+-- No errors expected from below block as SELECT there is a router
+-- query
+BEGIN;
+	SELECT count(*) FROM on_update_fkey_table WHERE id = 9;
+	DROP TABLE reference_table CASCADE;
+ROLLBACK;
+
 RESET client_min_messages;
 \set VERBOSITY default