From: Tom Lane <tgl@sss.pgh.pa.us>
To: Alvaro Herrera <alvherre@2ndquadrant.com>
cc: Andres Freund <andres@anarazel.de>, Robert Haas <robertmhaas@gmail.com>,
        David Rowley <david.rowley@2ndquadrant.com>,
        PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Subject: Re: pg_dump is broken for partition tablespaces
In-reply-to: <20190422190813.GA30369@alvherre.pgsql>
References: <20190422190813.GA30369@alvherre.pgsql>
Comments: In-reply-to Alvaro Herrera <alvherre@2ndquadrant.com>
	message dated "Mon, 22 Apr 2019 15:08:13 -0400"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <22941.1555961042.1@sss.pgh.pa.us>
Date: Mon, 22 Apr 2019 15:24:02 -0400
Message-ID: <22942.1555961042@sss.pgh.pa.us>
Precedence: bulk

Alvaro Herrera <alvherre@2ndquadrant.com> writes:
> On 2019-Apr-22, Andres Freund wrote:
>> Why is the obvious answer is to not just remove the whole tablespace
>> inheritance behaviour?

> Because it was requested by many, and there were plenty of people
> surprised that things didn't work that way.

There are lots of things in SQL that people find surprising.
In this particular case, "we can't do it because it conflicts with
ancient decisions about how PG tablespaces work" seems like a
defensible answer, even without getting into the question of
whether "partitions inherit their tablespace from the parent"
is really any less surprising than "partitions work exactly like
normal tables as far as tablespace selection goes".

			regards, tom lane