RE: speed up a logical replica setup - Mailing list pgsql-hackers

From Hayato Kuroda (Fujitsu)
Subject RE: speed up a logical replica setup
Date
Msg-id OSBPR01MB2552487ED013E710117DEADBF5D62@OSBPR01MB2552.jpnprd01.prod.outlook.com
Whole thread Raw
In response to Re: speed up a logical replica setup  (Amit Kapila <amit.kapila16@gmail.com>)
Responses Re: speed up a logical replica setup
List pgsql-hackers
Dear Amit, Euler,

Thanks for giving comment! 0001 was modified per suggestions.

> Yeah, it is a good idea to add a new option for two_phase but that
> should be done in the next version. For now, I suggest updating the
> docs and probably raising a warning (if max_prepared_transactions !=
> 0) as suggested by Noah. This WARNING is useful because one could
> expect that setting max_prepared_transactions != 0 means apply will
> happen at prepare time after the subscriber is created by this tool.
> The WARNING will be useful even if we support two_phase option as the
> user may have set the non-zero value of max_prepared_transactions but
> didn't use two_phase option.

I also think it should be tunable, in PG18+. 0002 adds a description in doc.
Also, this executable warns if the max_prepared_transactions != 0 on the
publisher. Subscriber-side is not checked now because I don't think it is helpful,
but it can be easily added. I did not add test for that, because current test code
does not check outputs.

Best Regards,
Hayato Kuroda
FUJITSU LIMITED
https://www.fujitsu.com/ 


Attachment

pgsql-hackers by date:

Previous
From: "David G. Johnston"
Date:
Subject: Re: Should we document how column DEFAULT expressions work?
Next
From: Daniel Gustafsson
Date:
Subject: Re: Improve the connection failure error messages