Re: pg15b2: large objects lost on upgrade

From: Justin Pryzby <pryzby(at)telsasoft(dot)com>
To: Robert Haas <robertmhaas(at)gmail(dot)com>
Cc: pgsql-hackers(at)postgresql(dot)org, Bruce Momjian <bruce(at)momjian(dot)us>, Shruthi Gowda <gowdashru(at)gmail(dot)com>
Subject: Re: pg15b2: large objects lost on upgrade
Date: 2022-07-06 11:56:00
Message-ID: 20220706115600.GU13040@telsasoft.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Jul 05, 2022 at 02:40:21PM -0400, Robert Haas wrote:
> On Tue, Jul 5, 2022 at 12:56 PM Justin Pryzby <pryzby(at)telsasoft(dot)com> wrote:
> > My patch also leaves a 0 byte file around from initdb, which is harmless, but
> > dirty.
> >
> > I've seen before where a bunch of 0 byte files are abandoned in an
> > otherwise-empty tablespace, with no associated relation, and I have to "rm"
> > them to be able to drop the tablespace. Maybe that's a known issue, maybe it's
> > due to crashes or other edge case, maybe it's of no consequence, and maybe it's
> > already been fixed or being fixed already. But it'd be nice to avoid another
> > way to have a 0 byte files - especially ones named with system OIDs.
>
> Do you want to add something to the patch to have pg_upgrade remove
> the stray file?

I'm looking into it, but it'd help to hear suggestions about where to put it.
My current ideas aren't very good.

--
Justin

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Svetlana Derevyanko 2022-07-06 12:00:55 Re[2]: [PATCH] Optional OR REPLACE in CREATE OPERATOR statement
Previous Message Dilip Kumar 2022-07-06 11:54:43 Re: making relfilenodes 56 bits