From: Josh Berkus <josh@agliodbs.com>
Reply-To: josh@agliodbs.com
Organization: Aglio Database Solutions
To: Bruce Momjian <pgman@candle.pha.pa.us>
Subject: Re: log_min_duration_statement versus log_statement
Date: Wed, 29 Jun 2005 14:32:55 -0700
User-Agent: KMail/1.8
Cc: Christian Robottom Reis <kiko@async.com.br>,
	pgsql-docs@postgresql.org, Guilherme Salgado <salgado@async.com.br>
References: <200506290122.j5T1Mt721527@candle.pha.pa.us>
In-Reply-To: <200506290122.j5T1Mt721527@candle.pha.pa.us>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <200506291432.56196.josh@agliodbs.com>

Bruce,

> OK, what if we change the documentaiton to be:
>
> 	log_min_duration_statement = 1000 # Log all statements whose
> 	                 # execution time exceeds the value, in
> 	                 # milliseconds.  -1 disables. Zero logs
> 	                 # all statements and their durations.
>
> I think you are confused because log_min_duration_statement is really
> about _when_ to log (duration > ? ms), and what to log (the statement).

No, I see what Christian is complaining about.   If one sets 
log_statement="All", then all statements will be logged regardless of 
duration or the setting in log_min_duration_statement.  So it should 
actually read:

 	log_min_duration_statement = 1000 # Log all statements whose
 	                 # execution time exceeds the value, in
 	                 # milliseconds.  -1 disables. This is in addition to
                         # any statements logged according to 
log_statement.

I'm with Christian on wanting to change the behavior so that it does what 
the comments says it does currently, but I don't think we'll get that done 
today.

--Josh

-- 
--Josh

Josh Berkus
Aglio Database Solutions
San Francisco