Re: Remove remnants of "snapshot too old"

From: Andres Freund <andres(at)anarazel(dot)de>
To: Heikki Linnakangas <hlinnaka(at)iki(dot)fi>
Cc: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: Remove remnants of "snapshot too old"
Date: 2024-12-04 01:20:19
Message-ID: 5cewdgo6jo4vblpqet4eukdszom5vevdfdebhnyazb5bbfvib2@u2e4ke5or6cl
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

On 2024-12-03 22:06:59 +0200, Heikki Linnakangas wrote:
> I spotted some more remnants of the "snapshot too old" feature that was
> removed in v17. Barring objections, I will commit the attached patch to tidy
> up.

Most of this I agree with. But I'm not sure just removing the toast snapshot
stuff is good - we've had a bunch of bugs where we don't hold a snapshot for
long enough to actually ensure that toast tuples stay alive. It's not legal to
fetch a toast id in one snapshot, release that, and then fetch the toast tuple
with a fresh snapshot. I think the removal of

> - if (snapshot == NULL)
> - elog(ERROR, "cannot fetch toast data without an active snapshot");

will make it easier to introduce further such bugs?

Greetings,

Andres Freund

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2024-12-04 01:24:39 Re: Remove remnants of "snapshot too old"
Previous Message Anton A. Melnikov 2024-12-04 01:00:53 Re: shared-memory based stats collector - v70