Content-Type: text/plain;
	charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 16.0 \(3774.600.62\))
Subject: Re: Logical replication with temporary tables
From: Christophe Pettus <xof@thebuild.com>
In-Reply-To: <CAAZ6SnzRU=qo2nedBa6ESOHfV=WaRJk+UW5DUrxkVMu6Gjpwng@mail.gmail.com>
Date: Tue, 2 Jul 2024 17:49:57 -0700
Cc: pgsql-general@lists.postgresql.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <0FE55AC1-25C2-4B5B-A312-5F8EB4756B4B@thebuild.com>
References: <CAAZ6SnzRU=qo2nedBa6ESOHfV=WaRJk+UW5DUrxkVMu6Gjpwng@mail.gmail.com>
To: Stuart Campbell <stuart.campbell@ridewithvia.com>
Archived-At: <https://www.postgresql.org/message-id/0FE55AC1-25C2-4B5B-A312-5F8EB4756B4B%40thebuild.com>
Precedence: bulk


> On Jul 2, 2024, at 17:47, Stuart Campbell =
<stuart.campbell@ridewithvia.com> wrote:
> My question is whether there is some workaround that will let me =
create indexes on a temporary table, and also let me drop the temporary =
table, in a way that doesn't end up "degrading" replication? (Presumably =
that means avoiding writing to the WAL?)

This is a question for AWS.  Community PostgreSQL doesn't have any of =
these concepts, and this is all proprietary modifications to PostgreSQL =
by Amazon.=