Message-ID: <f80622aa9f68adb5e3434b1471a91b96f1fc3c8a.camel@cybertec.at>
Subject: Re: pg_upgrade, can you use the rsync method of upgrading a standby
 if you used clone mode?
From: Laurenz Albe <laurenz.albe@cybertec.at>
To: davidnbooth@gmail.com, pgsql-docs@lists.postgresql.org
Date: Wed, 03 Sep 2025 09:44:43 +0200
In-Reply-To: <175682776043.778.6556350471379069615@wrigleys.postgresql.org>
References: <175682776043.778.6556350471379069615@wrigleys.postgresql.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.56.2 (3.56.2-1.fc42) 
MIME-Version: 1.0
Archived-At: 
 <https://www.postgresql.org/message-id/f80622aa9f68adb5e3434b1471a91b96f1fc3c8a.camel%40cybertec.at>
Precedence: bulk

On Tue, 2025-09-02 at 15:42 +0000, PG Doc comments form wrote:
> In the documentation for pgupgrade
> (https://www.postgresql.org/docs/17/pgupgrade.html), step 13 gives
> instructions for quickly upgrading standby servers with rsync, and says t=
hat
> it only works if you used link mode.  However, step 12 gives the impressi=
on
> that clone mode is very similar to link mode.  I think the documentation =
for
> step 13 would be improved by explicitly stating whether clone mode is
> compatible with the rsync method or not.

That would be interesting to know, right.

But after the slightly frustrating experience that I had when I tried to
improve that section of the documentation [1], I am not sure if it would
be wise to recommend the rsync procedure in more cases.

The sentence from the thread that stuck in my memory was Robert Haas'

  Also, let me express my general terror at the idea of anyone actually
  using this procedure.

Yours,
Laurenz Albe


 [1]: https://www.postgresql.org/message-id/flat/22f129004bb66cd91e1dfd3345=
a9787f5039f3ae.camel%40cybertec.at