Re: Is it correct to update db state in control file as "shutting down" during end-of-recovery checkpoint? - Mailing list pgsql-hackers

From Michael Paquier
Subject Re: Is it correct to update db state in control file as "shutting down" during end-of-recovery checkpoint?
Date
Msg-id YfCcl5yvktiVD1fA@paquier.xyz
Whole thread Raw
In response to Re: Is it correct to update db state in control file as "shutting down" during end-of-recovery checkpoint?  ("Bossart, Nathan" <bossartn@amazon.com>)
Responses Re: Is it correct to update db state in control file as "shutting down" during end-of-recovery checkpoint?
Re: Is it correct to update db state in control file as "shutting down" during end-of-recovery checkpoint?
List pgsql-hackers
On Tue, Jan 25, 2022 at 07:20:05PM +0000, Bossart, Nathan wrote:
> I looked into removing the "shutdown" variable in favor of using
> "flags" everywhere, but the patch was quite messy and repetitive.  I
> think another way to make things less confusing is to replace
> "shutdown" with an inverse variable called "online."  The attached
> patch does it this way.

Yeah, that sounds like a good compromise.  At least, I find the whole
a bit easier to follow.

Heikki was planning to commit a large refactoring of xlog.c, so we'd
better wait for that to happen before concluding here.
--
Michael

Attachment

pgsql-hackers by date:

Previous
From: Thomas Munro
Date:
Subject: Re: ssl_passphrase_callback failure in REL_14_STABLE
Next
From: Amit Kapila
Date:
Subject: Re: Skipping logical replication transactions on subscriber side