RE: Added schema level support for publication.
От | tanghy.fnst@fujitsu.com |
---|---|
Тема | RE: Added schema level support for publication. |
Дата | |
Msg-id | OS0PR01MB6113E8695060774C51D8B378FBB79@OS0PR01MB6113.jpnprd01.prod.outlook.com обсуждение исходный текст |
Ответ на | Re: Added schema level support for publication. (Greg Nancarrow <gregn4422@gmail.com>) |
Список | pgsql-hackers |
On Wednesday, October 13, 2021 4:10 PM Greg Nancarrow <gregn4422@gmail.com> wrote: > Also, I found the following scenario where the data is double-published: > > (1) PUB: CREATE PUBLICATION pub FOR TABLE sch1.sale_201901, TABLE > sch1.sale_201902 WITH (publish_via_partition_root=true); > (2) SUB: CREATE SUBSCRIPTION sub CONNECTION 'dbname=postgres > host=localhost port=5432' PUBLICATION pub; > (3) PUB: INSERT INTO sch.sale VALUES('2019-01-01', 'AU', 'cpu', 5), > ('2019-01-02', 'AU', 'disk', 8); > (4) SUB: SELECT * FROM sch.sale; > (5) PUB: ALTER PUBLICATION pub ADD ALL TABLES IN SCHEMA sch; > (6) SUB: ALTER SUBSCRIPTION sub REFRESH PUBLICATION; > (7) SUB: SELECT * FROM sch.sale; > > sale_date | country_code | product_sku | units > ------------+--------------+-------------+------- > 2019-01-01 | AU | cpu | 5 > 2019-01-02 | AU | disk | 8 > 2019-01-01 | AU | cpu | 5 > 2019-01-02 | AU | disk | 8 > > I changed your test step in (5) and used "ADD TABLE" command as below: ALTER PUBLICATION pub ADD TABLE sch.sale; I could get the same result on HEAD. So I think it's not a problem related to this patch. Maybe we can post this issue in a new thread. Regards Tang
В списке pgsql-hackers по дате отправления: