Date: Wed, 28 Oct 2020 16:43:44 +0900
From: Michael Paquier <michael@paquier.xyz>
To: Anastasia Lubennikova <a.lubennikova@postgrespro.ru>
Cc: Michael Banck <michael.banck@credativ.de>,
	PostgreSQL Development <pgsql-hackers@postgresql.org>
Subject: Re: [patch] Fix checksum verification in base backups for zero page
 headers
Message-ID: <20201028074344.GL28445@paquier.xyz>
References: <20201007081852.GC30037@paquier.xyz>
 <608f3476-0598-2514-2c03-e05c7d2b0cbd@postgrespro.ru>
 <20201020062432.GA30362@paquier.xyz>
 <69b38451-e91f-95fc-c3ce-06eb778d36e0@postgrespro.ru>
 <20201022012519.GF1475@paquier.xyz>
 <036235b4-2461-c4b2-a448-7875d943ce1e@postgrespro.ru>
 <20201022230008.GK1475@paquier.xyz>
 <20201026011315.GA2973@paquier.xyz>
 <ce25a60e-bd70-6554-1e8f-7e4dc9f6ce59@postgrespro.ru>
 <20201028071156.GK28445@paquier.xyz>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha512;
	protocol="application/pgp-signature"; boundary="2+jOUFLb7N+7fooZ"
Content-Disposition: inline
In-Reply-To: <20201028071156.GK28445@paquier.xyz>
Precedence: bulk


--2+jOUFLb7N+7fooZ
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Wed, Oct 28, 2020 at 04:11:56PM +0900, Michael Paquier wrote:
> Thanks.  The patch for v13 cannot use a macro, but one of the versions
> of upthread would do just fine.  I have been wondering about using the
> new CheckBuffer() for the purpose of the retry to make it
> concurrent-safe, but by looking at the code I think that we would run
> unto problems when trying to open through smgr.c any relation file in
> global/ as these require an invalid backend ID, and a WAL sender does
> not satisfy that (see the assertion in GetRelationPath()).

Actually, scratch that..  It looks like I am wrong here.  By using
smgropen() with InvalidBackendId and a RelFileNode built using the
path of the file being sent, similarly to what pg_rewind is doing in
isRelDataFile(), we should have everything that's needed.  This is
too complicated for a backpatch and we should have some consolidation
with pg_rewind, so using the sleep/retry for v13 sounds like a safer
path to take in the stable branch.
--
Michael

--2+jOUFLb7N+7fooZ
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
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=E/ix
-----END PGP SIGNATURE-----

--2+jOUFLb7N+7fooZ--