Message-ID: <4473512E.7020805@ca.sophos.com>
Date: Tue, 23 May 2006 11:15:10 -0700
From: Mischa Sandberg <mischa@ca.sophos.com>
Reply-To: mischa@ca.sophos.com
User-Agent: Thunderbird 1.5.0.2 (X11/20060420)
MIME-Version: 1.0
To: pgsql-performance@postgresql.org
Subject: Re: Selects query stats?
References: <9C4E40CF-6C06-4297-AF32-A7EF2A3F3C0F@hi5.com>
In-Reply-To: <9C4E40CF-6C06-4297-AF32-A7EF2A3F3C0F@hi5.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Dan Gorman wrote:
> All,
> 
> I might be completely crazy here, but it seems every other database 
> exposes select query stats. Postgres only exposes 
> updates/deletes/inserts. Is there something I am missing here?

Perhaps.

You can EXPLAIN ANALYZE a SELECT, just like i/u/d -- but then you
don't get the normal result set back. Is that what you mean?

You can turn on log_min_duration_statement and get total SELECT duration
logged.

There's a thread in pgsql-hackers ("Re: Porting MSSQL to PGSQL: trace and 
profile") about server-side logging of query plans and stats (for all four of 
s/i/u/d), which is indeed not there in PG.

-- 
Engineers think that equations approximate reality.
Physicists think that reality approximates the equations.
Mathematicians never make the connection.