From: Ronan Dunklau <ronan.dunklau@aiven.io>
To: Noah Misch <noah@leadboat.com>, Michael Paquier <michael@paquier.xyz>,
 Patrick =?ISO-8859-1?Q?St=E4hlin?= <patrick.staehlin@aiven.io>
Cc: pgsql-bugs <pgsql-bugs@lists.postgresql.org>
Subject: Re: FSM Corruption (was: Could not read block at end of the relation)
Date: Wed, 06 Mar 2024 12:08:38 +0100
Message-ID: <1979116.usQuhbGJ8B@aivenlaptop>
In-Reply-To: <aede09f6-be0c-e10b-6a4f-c8e021a014c9@aiven.io>
References: 
 <VtPLBUerbMM8BSDLQwIMV9Co_38iHS1TXE4V09x2ZVFhPkZMNE8LwgPSIzB8PHkVFXykGx04QNAzZVjJFAnU1w==@protonmail.internalid>
 <1958756.PYKUYFuaPT@aivenlaptop>
 <aede09f6-be0c-e10b-6a4f-c8e021a014c9@aiven.io>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="nextPart3470556.QJadu78ljV"
Content-Transfer-Encoding: 7Bit
Archived-At: 
 <https://www.postgresql.org/message-id/1979116.usQuhbGJ8B%40aivenlaptop>
Precedence: bulk

This is a multi-part message in MIME format.

--nextPart3470556.QJadu78ljV
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="iso-8859-1"

Le mercredi 6 mars 2024, 11:59:43 CET Patrick St=E4hlin a =E9crit :
> On 3/6/24 10:31, Ronan Dunklau wrote:
> > Le mardi 5 mars 2024, 00:05:03 CET Noah Misch a =E9crit :
> >> I would guess this one is more risky from a performance perspective,
> >> since
> >> we'd be adding to a hotter path under RelationGetBufferForTuple().=20
> >> Still,
> >> it's likely fine.
> >=20
> > I ended up implementing this in the attached patch. The idea is that we
> > detect if the FSM returns a page past the end of the relation, and igno=
re
> > it. In that case we will fallback through the extension mechanism.
>=20
> @@ -582,7 +583,17 @@ RelationGetBufferForTuple(Relation relation, Size le=
n,
>   		 * We have no cached target page, so ask the FSM for an=20
initial
>   		 * target.
>   		 */
> +		nblocks =3D RelationGetNumberOfBlocks(relation);
>   		targetBlock =3D GetPageWithFreeSpace(relation,=20
targetFreeSpace);
>=20
> I'd move the fetching of nblocks to after getting the targetBlock. This
> avoids extending the relation in the unlikely event of a FSM/relation
> extension in-between those two calls.

Good catch. Please find attached v2 of the patch.=20

=2D-
Ronan Dunklau
--nextPart3470556.QJadu78ljV
Content-Disposition: attachment;
 filename="0002-Detect-invalid-FSM-when-finding-a-block.patch"
Content-Transfer-Encoding: 7Bit
Content-Type: text/x-patch; charset="x-UTF_8J";
 name="0002-Detect-invalid-FSM-when-finding-a-block.patch"