From: | "Hayato Kuroda (Fujitsu)" <kuroda(dot)hayato(at)fujitsu(dot)com> |
---|---|
To: | 'Ashutosh Bapat' <ashutosh(dot)bapat(dot)oss(at)gmail(dot)com> |
Cc: | Shubham Khanna <khannashubham1197(at)gmail(dot)com>, Nisha Moond <nisha(dot)moond412(at)gmail(dot)com>, Ajin Cherian <itsajin(at)gmail(dot)com>, Peter Smith <smithpb2250(at)gmail(dot)com>, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>, vignesh C <vignesh21(at)gmail(dot)com>, Shlok Kyal <shlok(dot)kyal(dot)oss(at)gmail(dot)com> |
Subject: | RE: Enhance 'pg_createsubscriber' to retrieve databases automatically when no database is provided. |
Date: | 2025-03-25 12:07:59 |
Message-ID: | OSCPR01MB14966D4FCABD9902D0F15FBB2F5A72@OSCPR01MB14966.jpnprd01.prod.outlook.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Dear Ashutosh,
> The new description doesn't mention the link between the source and
> target database.
Yes, I intentionally removed.
> And I think it's incorrect. Not all databases on the
> target server will receive a subscription. Only those which have the
> same name as a database on the source server. Am I correct?
I assumed the point that the target is the streaming standby and all changes
between them would be eventually resolved. IIUC any differences between instances
would be resolved or the command would raise an ERROR. Thus, users do not have to
consider the differences of nodes so that we can simplify the description.
Analysis
=======
If some databases are missing on the source, it would not be listed as target.
pg_createsubscriber can wait until all changes are replicated before the promotion
so these databases would be removed even on target. OK, it could work.
If some databases exist only on the source, it would be listed. This command may
fail if database written in dbinfo[0].subconninfo does not exist on the target.
Otherwise they would be eventually replicated. Looks OK.
> I prefer
> the previous wording, even if it's a bit complex, over a simpler but
> incorrect description.
Definitely, and I know description in v20 is correct. Let's keep current
one if my idea above is wrong.
Best regards,
Hayato Kuroda
FUJITSU LIMITED
From | Date | Subject | |
---|---|---|---|
Next Message | Ashutosh Bapat | 2025-03-25 12:09:19 | Re: Reducing memory consumed by RestrictInfo list translations in partitionwise join planning |
Previous Message | Ashutosh Bapat | 2025-03-25 12:00:12 | Re: Enhance 'pg_createsubscriber' to retrieve databases automatically when no database is provided. |