RE: BUG #18644: ALTER PUBLICATION ... SET (publish_via_partition_root) wrong/undocumented behavior.
От | Hayato Kuroda (Fujitsu) |
---|---|
Тема | RE: BUG #18644: ALTER PUBLICATION ... SET (publish_via_partition_root) wrong/undocumented behavior. |
Дата | |
Msg-id | TYAPR01MB569256840EA1354B4F8BD9E2F54D2@TYAPR01MB5692.jpnprd01.prod.outlook.com обсуждение исходный текст |
Ответ на | Re: BUG #18644: ALTER PUBLICATION ... SET (publish_via_partition_root) wrong/undocumented behavior. (Maxim Boguk <maxim.boguk@gmail.com>) |
Ответы |
Re: BUG #18644: ALTER PUBLICATION ... SET (publish_via_partition_root) wrong/undocumented behavior.
|
Список | pgsql-bugs |
Dear Maxim, > Problem also happens if publication is created via FOR ALL TABLES / > FOR TABLES IN SCHEMA (without directly including partition head into > publication). Thanks for pointing out. I also confirmed this issue can happen for FOR ALL TABLES/FOR TABLES IN SCHEMA publications. I modified a documentation. > What is expected behavior in case when only partitions are included > into publication/subscription set, but partition head isn't included > and publish_via_partition_root='true' executed? Just to confirm - you meant like below definitions, right? ``` create table tab (a int) partition by range (a); create table tab_1 partition of tab for values from (-10) to (0); create table tab_2 partition of tab for values from (0) to (10); create publication pub for table tab_1, tab_2 with (publish_via_partition_root = true); ``` For now, changes are replicated as leaf table's one in above case. And I think it is the expected behavior because users expressly specifies them. Best regards, Hayato Kuroda FUJITSU LIMITED
В списке pgsql-bugs по дате отправления: