From owner-freebsd-questions@FreeBSD.ORG Mon Apr 3 12:58:29 2006 Return-Path: X-Original-To: freebsd-questions@freebsd.org Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0BB2916A422 for ; Mon, 3 Apr 2006 12:58:29 +0000 (UTC) (envelope-from freebsd@meijome.net) Received: from sigma.octantis.com.au (ns2.octantis.com.au [207.44.189.124]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8FA2343D45 for ; Mon, 3 Apr 2006 12:58:28 +0000 (GMT) (envelope-from freebsd@meijome.net) Received: (qmail 469 invoked from network); 3 Apr 2006 22:58:28 +1000 Received: from 210-84-32-140.dyn.iinet.net.au (HELO localhost) (210.84.32.140) by bravurasolutions.co.uk with (DHE-RSA-AES256-SHA encrypted) SMTP; 3 Apr 2006 22:58:27 +1000 Date: Mon, 3 Apr 2006 22:58:21 +1000 From: Norberto Meijome To: Miguel Message-ID: <20060403225821.45fcc26e@localhost> In-Reply-To: <442B2FC6.9040001@123.com.sv> References: <442B2FC6.9040001@123.com.sv> X-Mailer: Sylpheed-Claws 2.0.0 (GTK+ 2.8.16; i386-portbld-freebsd6.1) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: freebsd-questions@freebsd.org Subject: Re: terrible performance in 6.1beta4 X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Apr 2006 12:58:29 -0000 On Wed, 29 Mar 2006 19:09:26 -0600 Miguel wrote: > Hi, im loading a lot of information to a postgresql 8.1.3 database, > im using the copy command, using the same file and version of > postgres in gentoo it spend a few minutes (~4-5), in freebsd > 6.1-beta4 it has already spent 25 minutes and the server apparently > is doing nothing, [...] Hey Miguel, I would agree with all the comments about memory and buffers - tweak them till you allocate the right amount (i.e., make sure you don't OVER allocate...though i'm sure you'd see those errors in the pgsql log. what does ktrace show when you attach it to the process that is importing the data (client and server processes,actually). - have you tried doing a binary export and import? I've found them to be faster, though they may not work for what you need (I think I'm using them for backups... ). my hardware is pretty much the same as yours, though 4 GB RAM and 4 drives., but my DB is somewhat larger > and the reponse time of remote logins or running > simple commands like ls -l /etc takes a lot of time (35 secs ort so), > cpu usage is very low: ... and all these commands become responsive as usual the minute you kill the import process? btw, you definitely want to kill the indices / FKs in that table until you're finished with the initial import - it'll speed things up a lot. Beto