Horton

Re: Question regarding "Make archiver process an auxiliary process. commit"

Nathan Bossart <nathandbossart@gmail.com>

View thread View on postgr.es
From: Nathan Bossart <nathandbossart@gmail.com>
To: Sravan Kumar <sravanvcybage@gmail.com>
Cc: Kyotaro Horiguchi <horikyota.ntt@gmail.com>, bharath.rupireddyforpostgres@gmail.com, pgsql-hackers@lists.postgresql.org
Date: 2023-01-20T19:39:56Z
Lists: pgsql-hackers

Attachments

On Wed, Jan 04, 2023 at 11:35:33AM +0530, Sravan Kumar wrote:
> I have added the thread to the commitfest: https://commitfest.postgresql.org/42/
> Did you get a chance to review the patch? Please let me know if you
> need anything from my end.

This seems like worthwhile simplification to me.  Ultimately, your patch
shouldn't result in any sort of signficant behavior change, and I don't see
any reason to further complicate the timeout calculation.  The copy loop
will run any time the archiver's latch is set, and it'll wait up to 60
seconds otherwise.  As discussed upthread, it might be possible to remove
the timeout completely, but that probably deserves its own thread.

I noticed that time.h is no longer needed by the archiver, so I removed
that and fixed an indentation nitpick in the attached v2.  I'm going to set
the commitfest entry to ready-for-committer shortly after sending this
message.

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

Commits

  1. Simplify main waiting loop of the archiver process

  2. Remove promote_trigger_file.

  3. Suppress useless wakeups in walreceiver.

  4. Make archiver process an auxiliary process.