Re: Alter index rename concurrently to

Andres Freund <andres@anarazel.de>

From: Andres Freund <andres@anarazel.de>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Peter Eisentraut <peter.eisentraut@2ndquadrant.com>, Tom Lane <tgl@sss.pgh.pa.us>, Andrey Klychkov <aaklychkov@mail.ru>, Victor Yegorov <vyegorov@gmail.com>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2018-08-02T20:44:41Z
Lists: pgsql-hackers
Hi,

On 2018-08-02 16:30:42 -0400, Robert Haas wrote:
> With this change, we'd be able to say that new statements will
> definitely see the results of concurrent DDL.

I don't think it really gets you that far. Because you're suggesting to
do it at the parse stage, you suddenly have issues with prepared
statements. Some client libraries, and a lot more frameworks,
automatically use prepared statements when they see the same query text
multiple times.  And this'll not improve anything on that.

ISTM, if you want to increase consistency in this area, you've to go
further. E.g. processing invalidations in StartTransactionCommand() in
all states, which'd give you a lot more consistency.

Greetings,

Andres Freund


Commits

  1. Lower lock level for renaming indexes