Content-Type: text/plain;
	charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3826.700.81\))
Subject: Re: Wrong comment for ReplicationSlotCreate
From: Chao Li <li.evan.chao@gmail.com>
In-Reply-To: <CAJDiXgj-UHGy9gKeqO=sQnXq1p88ULcBgpf62_p2n2+Ay84uMw@mail.gmail.com>
Date: Sun, 4 Jan 2026 07:39:13 +0800
Cc: Postgres hackers <pgsql-hackers@lists.postgresql.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <A474F4A5-EE98-40D9-95F7-543BE81EEB1C@gmail.com>
References: <CAJDiXggZXQZ7bD0QcTizDt6us9aX6ZKK4dWxzgb5x3+TsVHjqQ@mail.gmail.com>
 <00DB978B-828D-44FE-8A0B-1643CE03878E@gmail.com>
 <CAJDiXghEf_BviOC9Ci-it2p1v0DOa+0yqfCyBV89rXSkcd-5JA@mail.gmail.com>
 <AC737D02-4021-4800-9F2B-5D93239BD043@gmail.com>
 <CAJDiXgj-UHGy9gKeqO=sQnXq1p88ULcBgpf62_p2n2+Ay84uMw@mail.gmail.com>
To: Daniil Davydov <3danissimo@gmail.com>
Archived-At: <https://www.postgresql.org/message-id/A474F4A5-EE98-40D9-95F7-543BE81EEB1C%40gmail.com>
Precedence: bulk


> On Jan 2, 2026, at 00:31, Daniil Davydov <3danissimo@gmail.com> wrote:
>=20
> Hi,
>=20
> On Wed, Dec 31, 2025 at 9:32=E2=80=AFAM Chao Li =
<li.evan.chao@gmail.com> wrote:
>>=20
>> You=E2=80=99re right that during CREATE_REPLICATION_SLOT, =
SnapBuildFindSnapshot() ensures we don=E2=80=99t miss PREPARE records =
for prepared transactions that exist at creation time.
>>=20
>> 1462aad2e introduced support for altering logical replication slot =
options, including two_phase, after the slot has been created. The =
commit comment says:
>> ```
>>    Changing the 'two_phase' option for a subscription from 'true' to =
'false'
>>    is permitted only when there are no pending prepared transactions
>>    corresponding to that subscription. Otherwise, the changes of =
already
>>    prepared transactions can be replicated again along with their =
corresponding
>>    commit leading to duplicate data or errors.
>> ```
>>=20
>> true->false is not allowed, however false->true is permitted. So, I =
think the old comment is still possible today:
>> ```
>> *     Note that enabling this option after decoding has already =
advanced
>> *     may result in missing PREPARE records for transactions that =
were
>> *     prepared before the option was enabled.
>> ```
>>=20
>=20
> Hm, I still can't understand why the comment that you provided is =
correct.
>=20
> How can we "miss PREPARE records" if slot creation requires all  =
prepared
> transactions to finish? The commit message says about risks during the
> change of the parameter "on the fly". But we are dealing with slot =
creation.
>=20
> --
> Best regards,
> Daniil Davydov

No problem, maybe I am wrong. Then please ignore my comment and wait for =
other review comments.

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