Content-Type: text/plain;
	charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3864.400.21\))
Subject: Re: Warn on missing replica identity in CREATE/ALTER PUBLICATION
From: Chao Li <li.evan.chao@gmail.com>
In-Reply-To: 
 <CAP6NMUJEvkD1BKm8YCVch5YhcwqaVeLQXWGQhqF-+gdDufgs3w@mail.gmail.com>
Date: Thu, 23 Apr 2026 14:50:22 +0800
Cc: shveta malik <shveta.malik@gmail.com>,
 pgsql-hackers@lists.postgresql.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <35B6A159-C2EF-4346-9394-7CB3E168B1A5@gmail.com>
References: 
 <CAP6NMUKYgRJ3Z5ThhtXpYYkRBffT=fAStwQ3_1Cab0tREZ=5NQ@mail.gmail.com>
 <CAJpy0uAd2BY_jinMWscfLr11WHra7sxrNc18jKbkhp=4CkAFig@mail.gmail.com>
 <CAP6NMUJEvkD1BKm8YCVch5YhcwqaVeLQXWGQhqF-+gdDufgs3w@mail.gmail.com>
To: =?utf-8?B?5Y2X5ouT5byl?= <minamitakuya@lifull.com>
Archived-At: 
 <https://www.postgresql.org/message-id/35B6A159-C2EF-4346-9394-7CB3E168B1A5%40gmail.com>
Precedence: bulk


> On Apr 23, 2026, at 13:46, =E5=8D=97=E6=8B=93=E5=BC=A5 =
<minamitakuya@lifull.com> wrote:
>=20
> # Reply draft v2 to Shveta
>=20
> ---
>=20
> Hi Shveta,
>=20
> Thanks for pointing out that thread. I've read through it carefully.
>=20
> I believe the two proposals address different aspects of the same
> problem:
>=20
> - The fallback RI approach changes runtime behavior so that tables
>  without a primary key can still replicate UPDATE/DELETE.
> - This proposal simply warns at DDL time that a publication contains
>  tables whose replica identity will cause UPDATE/DELETE to fail at
>  replication time.
>=20
> A WARNING at publication creation time is useful regardless of whether
> a fallback mechanism exists, because:
>=20
> - If a table has REPLICA IDENTITY DEFAULT with no primary key, it
>  silently falls back to NOTHING. Combining that with a publication
>  that publishes updates/deletes is guaranteed to fail at runtime.
>  A WARNING at DDL time closes this gap.
> - Even users who explicitly set REPLICA IDENTITY NOTHING and add the
>  table to an update/delete publication would benefit from a reminder,
>  since that combination cannot succeed.
> - The WARNING does not change any existing behavior =E2=80=94 it only =
makes
>  the misconfiguration visible earlier.
>=20
> Notably, Euler mentioned in that thread [1] that he would "suggest a
> way to disallow or add a warning message while creating the
> publication or adding new tables", which is exactly what this proposal
> does.
>=20
> That said, I see the two proposals as complementary. Should I continue
> this as a separate thread, or would it be better to join the existing
> discussion?
>=20
> I have a working patch covering all publication paths (FOR TABLE,
> FOR TABLES IN SCHEMA, FOR ALL TABLES, ALTER PUBLICATION). Happy to
> post it either way.
>=20
> [1] =
https://www.postgresql.org/message-id/a9da608f-24be-4213-a712-8592852d37f1=
%40app.fastmail.com
>=20

You are very welcome to join the thread, as the initiator of that =
thread.

I am not personally against your idea of adding such a warning message, =
but I think it would be better to consider the two features together as =
a whole solution from a system perspective.

In any case, new features will have to wait for v20 until July, so we =
still have time for more discussion and deeper consideration.

Best regards,
--
Chao Li (Evan)
HighGo Software Co., Ltd.
https://www.highgo.com/