Re: pg15b2: large objects lost on upgrade

On Mon, Aug 8, 2022 at 09:51:46PM -0400, Tom Lane wrote:
> Bruce Momjian <bruce(at)momjian(dot)us> writes:
> >> Hmmm ... now that you mention it, I see nothing in 002_pg_upgrade.pl
> >> that attempts to turn off autovacuum on either the source server or
> >> the destination. So one plausible theory is that autovac moved the
> >> numbers since we checked.
>
> > Uh, pg_upgrade assumes autovacuum is not running, and tries to enforce
> > this:
>
> The problems come from autovac running before or after pg_upgrade.
>
> > Perhaps the test script should do something similar,
>
> I'm not on board with that, for the reasons I gave upthread.

Uh, I assume it is this paragraph:

> If that is the explanation, then it leaves us with few good options.
> I am not in favor of disabling autovacuum in the test: ordinary
> users are not going to do that while pg_upgrade'ing, so it'd make
> the test less representative of real-world usage, which seems like
> a bad idea. We could either drop this particular check again, or
> weaken it to allow new relfrozenxid >= old relfrozenxid, likewise
> relminxid.

I thought the test was setting up a configuration that would never be
used by normal servers. Is that false?

--
Bruce Momjian <bruce(at)momjian(dot)us> https://momjian.us
EDB https://enterprisedb.com

Indecision is a decision. Inaction is an action. Mark Batterson