Re: CREATE SUBSCRIPTION ... SERVER vs. pg_dump, etc.

From: Jeff Davis <pgsql(at)j-davis(dot)com>
To: Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>
Cc: Noah Misch <noah(at)leadboat(dot)com>, pgsql-hackers(at)postgresql(dot)org, "Hayato Kuroda (Fujitsu)" <kuroda(dot)hayato(at)fujitsu(dot)com>, Shlok Kyal <shlok(dot)kyal(dot)oss(at)gmail(dot)com>, Fujii Masao <masao(dot)fujii(at)gmail(dot)com>, yuanchao zhang <145zhangyc(at)gmail(dot)com>
Subject: Re: CREATE SUBSCRIPTION ... SERVER vs. pg_dump, etc.
Date: 2026-08-03 17:17:07
Message-ID: 1edc76f0011087ba8da69e1310a142d5876ceeed.camel@j-davis.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Mon, 2026-08-03 at 11:07 +0530, Amit Kapila wrote:

> OTOH, I am also fine if you prefer to remove the retain_dead_tuples
> check entirely from the ENABLE path and keep it in other existing
> paths as in attached. Actually, that will slightly simplify the code
> as well.

The patch looks good to me.

For approach 1 versus 2, I'll defer to you and authors/reviewers of the
feature who understand the use cases better. I slightly prefer the
above approach (don't connect at ENABLE time) because side effects at
DDL time are hard to reason about.

Regards,
Jeff Davis

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Masahiko Sawada 2026-08-03 18:36:11 Re: Proposal: Conflict log history table for Logical Replication
Previous Message Fujii Masao 2026-08-03 15:24:53 Re: 030_pg_recvlogical fails because the same PID is assigned