Feedback-ID: ia2694551:Fastmail
Date: Tue, 26 Mar 2024 16:39:55 +0100
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Nathan Bossart <nathandbossart@gmail.com>
Cc: Amit Kapila <amit.kapila16@gmail.com>,
	Robert Haas <robertmhaas@gmail.com>,
 Bertrand Drouvot <bertranddrouvot.pg@gmail.com>,
	shveta malik <shveta.malik@gmail.com>,
 Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>,
	Amit Kapila <akapila@postgresql.org>,
 "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Subject: Re: pgsql: Track last_inactive_time in pg_replication_slots.
Message-ID: <202403261539.xcjfle7sksz7@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20240326150918.GB3181099@nathanxps13>
Archived-At: 
 <https://www.postgresql.org/message-id/202403261539.xcjfle7sksz7%40alvherre.pgsql>
Precedence: bulk

On 2024-Mar-26, Nathan Bossart wrote:

> FWIW I'd really prefer to have something like max_slot_xid_age for this.  A
> time-based parameter would likely help with most cases, but transaction ID
> usage will vary widely from server to server, so it'd be nice to have
> something to protect against wraparound more directly.

Yeah, I tend to agree that an XID-based limit makes more sense than a
time-based one.

> I don't object to a
> time-based setting as well, but that won't always work as well for this
> particular use-case, especially if we are relying on users to set a
> slot-level parameter.

I think slot-level parameters are mostly useless, because it takes just
one slot where you forget to set it for disaster to strike.

-- 
Álvaro Herrera        Breisgau, Deutschland  —  https://www.EnterpriseDB.com/