Re: [PATCH] improve the pg_upgrade error message

From: Daniel Gustafsson <daniel(at)yesql(dot)se>
To: Suraj Kharage <suraj(dot)kharage(at)enterprisedb(dot)com>
Cc: Jeevan Ladhe <jeevan(dot)ladhe(at)enterprisedb(dot)com>, Justin Pryzby <pryzby(at)telsasoft(dot)com>, Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: [PATCH] improve the pg_upgrade error message
Date: 2021-10-21 10:21:47
Message-ID: C8162D69-C6A4-496E-B565-6CC01D74A4DC@yesql.se
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> On 14 Jul 2021, at 07:27, Suraj Kharage <suraj(dot)kharage(at)enterprisedb(dot)com> wrote:

> Overall patch looks good to me.

Agreed, I think this is a good change and in line with how the check functions
work in general.

> Instead of giving suggestion about updating the pg_database catalog, can we give "ALTER DATABASE <datname> ALLOW_CONNECTIONS true;" command?

I would actually prefer to not give any suggestions at all, we typically don't
in these error messages. Since there are many ways to do it (dropping the
database being one) I think leaving that to the user is per application style.

> Also, it would be good if we give 2 spaces after full stop in an error message.

Correct, fixed in the attached which also tweaks the language slightly to match
other errors.

I propose to commit the attached, which also adds a function comment while
there, unless there are objections.

--
Daniel Gustafsson https://vmware.com/

Attachment Content-Type Size
0001-List-offending-databases-in-pg_upgrade-datallowconn-.patch application/octet-stream 3.1 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message hubert depesz lubaczewski 2021-10-21 10:52:53 Re: Can we get rid of repeated queries from pg_dump?
Previous Message Greg Nancarrow 2021-10-21 09:59:02 Re: Added schema level support for publication.