Date: Fri, 8 Apr 2005 11:50:40 -0400
From: Andrew Sullivan <ajs@crankycanuck.ca>
To: pgsql-sql@postgresql.org
Subject: Re: Question on triggers and plpgsql
Message-ID: <20050408155040.GC27718@phlogiston.dyndns.org>
References: <4255AA15.8000508@mochima.com>
	<BE1E8DAA-A829-11D9-9125-000A95B03262@pgedit.com>
	<fef7e50cf33b53ca481ecb2554cd6403@mail.nih.gov>
	<14244.1112970986@sss.pgh.pa.us>
	<20050408145928.GA27718@phlogiston.dyndns.org>
	<14740.1112974547@sss.pgh.pa.us>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <14740.1112974547@sss.pgh.pa.us>
User-Agent: Mutt/1.5.6+20040907i

On Fri, Apr 08, 2005 at 11:35:47AM -0400, Tom Lane wrote:
> What this does do for you is replace the risk of phantom emails (mail
> sent but corresponding action inside the database never committed)
> with the risk of duplicate emails (mail-sender sends you another one
> after it restarts).  In most cases I think I'd prefer the latter.

Me too.  Besides, you already have this risk with SMTP, because a
message can be queued and accepted on the remote side when the local
side goes away, so that the session is completed improperly. 
Depending on configuration and a bunch of painful start-up
possibilities with the server, you might well get a duplicate copy of
a mail transmitted later.  (In the present age, given the remarkable
quality of networks and mail servers everyone has, you almost never
have this happen any more.  But it's still strictly speaking
possible.)

A

-- 
Andrew Sullivan  | ajs@crankycanuck.ca
The fact that technology doesn't work is no bar to success in the marketplace.
		--Philip Greenspun