Content-Type: text/plain;
	charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3864.100.1.1.5\))
Subject: Re: The ability of postgres to determine loss of files of the main
 fork
From: Frits Hoogland <frits.hoogland@gmail.com>
In-Reply-To: <cbnna46p73gtyzbw22x67vuihcqmqwfbjgbzkdmrynovatvjlf@kwks4cei2jgh>
Date: Wed, 1 Oct 2025 17:25:00 +0200
Cc: Laurenz Albe <laurenz.albe@cybertec.at>,
 Aleksander Alekseev <aleksander@tigerdata.com>,
 pgsql-hackers@lists.postgresql.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <BED39015-2035-4565-80C1-3B9C40EA01CB@gmail.com>
References: <013D63E2-5D75-492E-85FF-1D5CC0148C82@gmail.com>
 <CAJ7c6TMYvYBQpgv8k+4QnajqPcroOzrj7d3snBBTKX4tHQfVYw@mail.gmail.com>
 <45b3a1dd6be28dd409850a5503c029a97ca0754e.camel@cybertec.at>
 <483AD9AA-39C1-4F4B-B0D6-4CBA0014E877@gmail.com>
 <083a4d75ab127a603743058bce26e79ba793494e.camel@cybertec.at>
 <cbnna46p73gtyzbw22x67vuihcqmqwfbjgbzkdmrynovatvjlf@kwks4cei2jgh>
To: Andres Freund <andres@anarazel.de>
Archived-At: 
 <https://www.postgresql.org/message-id/BED39015-2035-4565-80C1-3B9C40EA01CB%40gmail.com>
Precedence: bulk


> On 1 Oct 2025, at 15:49, Andres Freund <andres@anarazel.de> wrote:
>=20
> Hi,
>=20
> On 2025-10-01 15:39:04 +0200, Laurenz Albe wrote:
>> On Wed, 2025-10-01 at 13:58 +0200, Frits Hoogland wrote:
>>> I am proposing the database to have the ability to detect when it =
has missing segments.
>>=20
>> Just a random idea: one solution would be if each segment has a flag =
that indicates
>> if that is the last segment or not.  But that would break the on-disk =
storage format,
>> unless there is room left for an extra flag somewhere in the current =
layout.
>=20
> It'd also make extensions / truncations more complicated. I rather =
doubt we're
> going there. Right now relation extension aren't WAL logged. While =
there might
> be reasons to change that, I don't think this is enough justification =
for
> doing so.
>=20

What would be a achievable way of making postgres under the relation =
size?
How about a field in pg_class that keeps the final data page, so that =
the catalog
keeps the size, which then allows utilities and the database itself to =
understand how
many segments should exist?

> Greetings,
>=20
> Andres