Re: Add a GUC variable that control logical replication
От | Euler Taveira |
---|---|
Тема | Re: Add a GUC variable that control logical replication |
Дата | |
Msg-id | CAHE3wgij+1=UBPFgtSwJAFGsqJf89jw2oziM+pUgkK-zGx=hwQ@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: Add a GUC variable that control logical replication (Peter Eisentraut <peter.eisentraut@2ndquadrant.com>) |
Ответы |
Re: Add a GUC variable that control logical replication
|
Список | pgsql-hackers |
Em sáb, 19 de out de 2019 às 14:11, Peter Eisentraut <peter.eisentraut@2ndquadrant.com> escreveu: > > On 2019-09-18 11:33, Quan Zongliang wrote: > > On 2019/9/18 17:11, Peter Eisentraut wrote: > >> Why do you need to turn off replication when there is "maintenance" on a > >> table? What does that even mean? > >> > > In a table, the user only keep data for a period of time and delete > > expired records every day, involving about 10 million to 20 million > > records at a time. They want to not pass similar bulk operations in > > logical replication. > > You can probably achieve that using ALTER PUBLICATION to disable > publication of deletes or truncates, as the case may be, either > permanently or just for the duration of the operations you want to skip. > ... then you are skipping all tables in the publication. I think this feature is not essential for unidirectional logical replication. However, it is important for multi-master replication. Data synchronization tool will generate transactions with rows that are already in the other node(s) so those transactions can't be replicated to avoid (expensive) conflicts. -- Euler Taveira Timbira - http://www.timbira.com.br/ PostgreSQL: Consultoria, Desenvolvimento, Suporte 24x7 e Treinamento
В списке pgsql-hackers по дате отправления: