Re: make pg_ctl more friendly

From: Nathan Bossart <nathandbossart(at)gmail(dot)com>
To: Junwang Zhao <zhjwpku(at)gmail(dot)com>
Cc: Nazir Bilal Yavuz <byavuz81(at)gmail(dot)com>, Crisp Lee <litianxiang01(at)gmail(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, pgsql-hackers(at)lists(dot)postgresql(dot)org, Bruce Momjian <bruce(at)momjian(dot)us>
Subject: Re: make pg_ctl more friendly
Date: 2024-01-15 21:39:32
Message-ID: 20240115213932.GA40905@nathanxps13
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

+ POSTMASTER_RECOVERY_SHUTDOWN,

Perhaps this should be POSTMASTER_SHUTDOWN_IN_RECOVERY to match the state
in the control file?

+ case POSTMASTER_RECOVERY_SHUTDOWN:
+ print_msg(_("PITR shutdown\n"));
+ print_msg(_("update configuration for startup again if needed\n"));
+ break;

I'm not sure I agree that this is a substantially friendlier message. From
a quick skim of the thread, it seems like you want to avoid sending a scary
error message if Postgres was intentionally shut down while in recovery.
If I got this particular message, I think I would be worried that something
went wrong during my point-in-time restore, and I'd be scrambling to figure
out what configuration this message wants me to update.

If I'm correctly interpreting the intent here, it might be worth fleshing
out the messages a bit more. For example, instead of "PITR shutdown,"
perhaps we could say "shut down while in recovery." And maybe we should
point to the specific settings in the latter message.

--
Nathan Bossart
Amazon Web Services: https://aws.amazon.com

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Paquier 2024-01-15 23:28:25 Re: Add PQsendSyncMessage() to libpq
Previous Message Jeff Davis 2024-01-15 21:34:18 Re: [17] CREATE SUBSCRIPTION ... SERVER