Skip site navigation (1) Skip section navigation (2)

Re: Postgres8.2 - turning off BINGLOG/PARSELOG

From: Mario Splivalo <mario(dot)splivalo(at)mobart(dot)hr>
To: Alvaro Herrera <alvherre(at)commandprompt(dot)com>
Cc: pgsql-admin(at)postgresql(dot)org
Subject: Re: Postgres8.2 - turning off BINGLOG/PARSELOG
Date: 2007-06-19 15:17:21
Message-ID: 1182266241.13561.52.camel@localhost.localdomain (view raw, whole thread or download thread mbox)
Lists: pgsql-admin
On Tue, 2007-06-19 at 10:52 -0400, Alvaro Herrera wrote:
> Mario Splivalo wrote:
> > I'm finaly switching to 8.2. I notice that ugly 'feature' from pg8.1 has
> > been fixed: logging the parametars of prepared statements. It's a bit
> > ugly, but it's there :)
> > 2007-06-19 16:22:15.555 CEST [4596] <octopussy2> BINDLOG:  duration:
> > 0.027 ms  bind S_2: COMMIT
> > 2007-06-19 16:22:15.556 CEST [4596] <octopussy2> COMMITLOG:  duration:
> > 0.029 ms  execute S_2: COMMIT
> >
> > log_line_prefix = '%m [%p] <%d> %i'
> One thing you should be aware of is that PARSELOG etc are really "PARSE"
> and "LOG", but there is no intervening space since you put no space at
> the end of log_line_prefix.  So I suggest changing it to
> log_line_prefix = '%m [%p] <%d> %i '

Huh, thnx! :) Silly me, I tought that %i actually stands for 'SELECT *
FROM ...' :)

> Now, regarding your question,
> > Is there a way to turn off these PARSELOG and BINDLOG messages? I realy
> > need only the SELECTLOG (and SELECTDETAIL because there are the actuall
> > parametars)?
> I don't think so.  You can increase log_min_duration_statement a bit so
> that only statements that take "long" are logged.

Why not? I need to have log_min_duration_statement set to 0 because I do
want every SQL sent to the postgres loged. It's essential in debugging.
But since for every SELECT i have at least 3 lines in my log file,
debugging is much harder (ok, not that harder, but still, it would be
much easier without it). It would be realy nice if one could turn that

When I switch to JDBC driver for pg (pg74.215.jdbc3.jar), I have single
SELECT in a log for a SELECT sent from the client. I guess that's
because there is no prepared statements in pg74.215.jdbc3.jar, but that
is for pg-jdbc mailinglist.

10x again!


In response to

pgsql-admin by date

Next:From: Karl WrightDate: 2007-06-19 15:18:42
Subject: Export/import issue/question
Previous:From: Joshua_KramerDate: 2007-06-19 14:54:06
Subject: Re: [pgsql-advocacy] [PERFORM] Postgres VS Oracle

Privacy Policy | About PostgreSQL
Copyright © 1996-2018 The PostgreSQL Global Development Group