Subject: Re: Sync Rep Design
From: Simon Riggs <simon@2ndQuadrant.com>
To: Marti Raudsepp <marti@juffo.org>
Cc: Stefan Kaltenbrunner <stefan@kaltenbrunner.cc>,
	pgsql-hackers@postgresql.org
In-Reply-To: <AANLkTikcji+2VuE95y0rgwD3bTjK5wSLAJuPwRUFfE81@mail.gmail.com>
References: <1293729972.1892.23804.camel@ebony>
	<4D1CC47E.7050207@kaltenbrunner.cc> <1293735853.1892.24886.camel@ebony>
	<AANLkTikcji+2VuE95y0rgwD3bTjK5wSLAJuPwRUFfE81@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"
Date: Thu, 30 Dec 2010 21:10:00 +0000
Message-ID: <1293743400.1892.26310.camel@ebony>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit

On Thu, 2010-12-30 at 22:11 +0200, Marti Raudsepp wrote:

> I think a comment about the "head-of-line blocking" nature of
> streaming repliaction is in order. If you execute massive writes in
> async mode and then run a transaction in sync mode, its commit will be
> delayed until all the async transactions before it have been applied
> on the slave.

Not really sure I understand what you want me to add there. The case you
mention is identical whether we use the word "async" or "sync" where you
mention "in async mode".

Replication doesn't wait until a sync commit is requested, it is
continuously active.

Sync rep's only addition are the reply messages.

-- 
 Simon Riggs           http://www.2ndQuadrant.com/books/
 PostgreSQL Development, 24x7 Support, Training and Services