Re: pg_upgrade fails if privileges have been granted on pg_start_backup()/pg_stop_backup() in the old cluster

From: Noah Misch <noah(at)leadboat(dot)com>
To: Helena Zetterström <helena(dot)zetterstrom(at)chalmers(dot)se>
Cc: "pgsql-bugs(at)lists(dot)postgresql(dot)org" <pgsql-bugs(at)lists(dot)postgresql(dot)org>
Subject: Re: pg_upgrade fails if privileges have been granted on pg_start_backup()/pg_stop_backup() in the old cluster
Date: 2022-11-03 02:30:05
Message-ID: 20221103023005.GA3400565@rfd.leadboat.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

On Wed, Nov 02, 2022 at 02:47:27PM +0000, Helena Zetterström wrote:
> In version 15 the functions pg_start_backup()/pg_stop_backup() have been renamed to pg_backup_start()/pg_backup_stop().
>
> If privileges on these functions have been granted in the old cluster the pg_upgrade fails.
> Our barman user has been granted execute priviliges on these function in PostgreSQL 14 and to avoid the pg_upgrade to fail we had to revoke these privileges in the old cluster before running the pg_upgrade.
> No warning/notification is given about this when running pg_upgrade --check.
> I don´t know if this is a bug or not but it is really inconvenient to have the pg_upgrade fail/crash (child process exited abnormally: status 256) due to it being unable to grant execute on these no longer existing functions.
> It would have been good if a notification about this could have been shown when running the pg_upgrade with the --check option before performing the actual upgrade.

This thread was an unsuccessful attempt to fix that class of problem:
https://www.postgresql.org/message-id/flat/f85991ad-bbd4-ad57-fde4-e12f0661dbf0%40postgrespro.ru

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Julien Rouhaud 2022-11-03 06:03:11 Re: BUG #17636: terminating connection because of crash of another server process
Previous Message Helena Zetterström 2022-11-02 14:47:27 pg_upgrade fails if privileges have been granted on pg_start_backup()/pg_stop_backup() in the old cluster