Horton

Re: Excessive PostmasterIsAlive calls slow down WAL redo

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es
From: Michael Paquier <michael@paquier.xyz>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andres Freund <andres@anarazel.de>, Heikki Linnakangas <hlinnaka@iki.fi>, pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2018-04-06T05:08:36Z
Lists: pgsql-hackers 
On Thu, Apr 05, 2018 at 02:39:27PM -0400, Tom Lane wrote:
> While it's not POSIX, at least some platforms are capable of delivering
> a separate signal on parent process death.  Perhaps using that where
> available would be enough of an answer.

Are you referring to prctl here?

+1 on improving performance of PostmasterIsAlive() if possible instead
of checking for it every N records.  You barely see records creating a
database from a large template close to each other but that could hurt
the response time of the redo process.
--
Michael

Commits

  1. Poll postmaster less frequently in recovery.

  2. Use signals for postmaster death on FreeBSD.

  3. Use signals for postmaster death on Linux.

  4. Introduce a pipe between postmaster and each backend, which can be used to