Date: Tue, 22 Sep 2026 10:35:06 -0700
From: Noah Misch <noah@leadboat.com>
To: Heikki Linnakangas <hlinnaka@iki.fi>
Cc: Nathan Bossart <nathandbossart@gmail.com>,
	Melanie Plageman <melanieplageman@gmail.com>,
	Michael Paquier <michael@paquier.xyz>,
	Sami Imseih <samimseih.pg@gmail.com>,
	Andres Freund <andres@anarazel.de>, Naga Appani <nagnrik@gmail.com>,
	Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>,
	Tomas Vondra <tomas@vondra.me>, Xuneng Zhou <xunengzhou@gmail.com>,
	torikoshia <torikoshia@oss.nttdata.com>,
	Kirill Reshke <reshkekirill@gmail.com>,
	pgsql-hackers@postgresql.org
Subject: Re: Catversion bumps during beta (was Re: [Proposal] Expose internal
 MultiXact member count function for efficient monitoring)
Message-ID: <20260922173506.dd.noahmisch@microsoft.com>
References: <aVH6u99shKNCoo0Y@paquier.xyz>
 <CA+QeY+Bjbe8GuNYaDDHMAYJMoR9F+R7xLKWsUTjJrcxfzs_cZA@mail.gmail.com>
 <aVOCw5ANG4Tk2C55@paquier.xyz>
 <yhtrliiq662vmhqy2xnbg67ezrucmwtfuuus2zcjvmwgmracux@2cqaioxnntwz>
 <arG3RW2R3qsIj5zJ@paquier.xyz>
 <CAN12+YKc278ayOBzF4Rbosrz=5QzRJ7t7bHff-FKiJbbvgwFzg@mail.gmail.com>
 <arI1rEHcZKBIiKAo@paquier.xyz>
 <CAAKRu_bE11PUHDVthMoKXownefkzB8H31WLTvq2XXy5WgJsTxg@mail.gmail.com>
 <arKi8Jdv8EwASlgZ@nathan>
 <b18df008-65b9-4276-ab61-b99a24ad9599@iki.fi>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <b18df008-65b9-4276-ab61-b99a24ad9599@iki.fi>
User-Agent: Mutt/2.3.0 (2026-01-25)
Archived-At: 
 <https://www.postgresql.org/message-id/20260922173506.dd.noahmisch%40microsoft.com>
Precedence: bulk

On Tue, Sep 22, 2026 at 07:01:14PM +0300, Heikki Linnakangas wrote:
> I don't really mind catversion bumps during the beta period in general, I'm
> in the camp that if there's a change that would make the release better and
> it's not particularly risky, it's better to still sneak it in.
> 
> I feel I might be in the minority with that view though. Historically, we've
> avoided fixes that need catversion bumps late in the beta period, so that
> beta-testers don't need to pg_upgrade. I think that's a poor tradeoff, but
> if we're changing long-standing policy, the RMT probably isn't the right
> group for that. So, how do others feel?

I personally make no attempt to avoid late catversion bumps.