Horton

Re: pg15b2: large objects lost on upgrade

Robert Haas <robertmhaas@gmail.com>

View thread
From: Robert Haas <robertmhaas@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Noah Misch <noah@leadboat.com>, Andrew Dunstan <andrew@dunslane.net>, Bruce Momjian <bruce@momjian.us>, Michael Paquier <michael@paquier.xyz>, Justin Pryzby <pryzby@telsasoft.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>, Shruthi Gowda <gowdashru@gmail.com>
Date: 2022-07-30T00:08:02Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Revert recent changes to 002_pg_upgrade.pl.

  2. Revise test case added in 43746996399541ecb5c7b188725a5f097c15ceae.

  3. Use TRUNCATE to preserve relfilenode for pg_largeobject + index.

  4. Preserve relfilenode of pg_largeobject and its index across pg_upgrade.

  5. Have VACUUM warn on relfrozenxid "in the future".

  6. Set relfrozenxid to oldest extant XID seen by VACUUM.

  7. pg_upgrade: Preserve relfilenodes and tablespace OIDs.

  8. Perform a lot more sanity checks when freezing tuples.

Attachments

On Fri, Jul 29, 2022 at 8:02 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Personally I'd try to replace the two horizon-collection steps with
> $newnode->psql calls, using extra_params to inject the '-o' and target
> filename command line words.  But if you want to try adding -X as
> a quicker answer, maybe that will be enough.

Here's a patch that uses a variant of that approach: it just runs
safe_psql straight up and gets the output, then writes it out to temp
files if the output doesn't match and we need to run diff. Let me know
what you think of this.

While working on this, I noticed a few other problems. One is that the
query doesn't have an ORDER BY clause, which it really should, or the
output won't be stable. And the other is that I think we should be
testing against the regression database, not the postgres database,
because it's got a bunch of user tables in it, not just
pg_largeobject.

-- 
Robert Haas
EDB: http://www.enterprisedb.com