Received: from localhost (unknown [200.46.204.191]) by postgresql.org (Postfix) with ESMTP id CF8A69F9D0C for ; Wed, 12 Sep 2007 16:58:46 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.191]) (amavisd-maia, port 10024) with ESMTP id 99835-03 for ; Wed, 12 Sep 2007 16:58:43 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from crestone.coronasolutions.com (crestone.coronasolutions.com [66.45.104.24]) by postgresql.org (Postfix) with ESMTP id 112819F9C3D for ; Wed, 12 Sep 2007 16:58:44 -0300 (ADT) Received: from localhost (localhost.localdomain [127.0.0.1]) by crestone.coronasolutions.com (Postfix) with ESMTP id C319C644C4A; Wed, 12 Sep 2007 13:58:38 -0600 (MDT) Received: from crestone.coronasolutions.com ([127.0.0.1]) by localhost (crestone.coronasolutions.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 09566-02; Wed, 12 Sep 2007 13:58:37 -0600 (MDT) Received: from mac.local (c-76-25-77-161.hsd1.co.comcast.net [76.25.77.161]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by crestone.coronasolutions.com (Postfix) with ESMTP id DA5F7644C48 for ; Wed, 12 Sep 2007 13:58:36 -0600 (MDT) Message-ID: <46E844EC.5050802@drivefaster.net> Date: Wed, 12 Sep 2007 13:58:36 -0600 From: Dan Harris User-Agent: Thunderbird 2.0.0.6 (Macintosh/20070728) MIME-Version: 1.0 To: PostgreSQL Performance Subject: pg_dump blocking create database? Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new-20030616-p10 (Debian) at coronasolutions.com X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200709/233 X-Sequence-Number: 26778 My PG server came to a screeching halt yesterday. Looking at top saw a very large number of "startup waiting" tasks. A pg_dump was running and one of my scripts had issued a CREATE DATABASE command. It looks like the CREATE DATABASE was exclusive but was having to wait for the pg_dump to finish, causing a massive traffic jam of locks behind it. Once I killed the pg_dump process, things returned to normal. Version 8.0.12. Is this a bug? It also concerns me because my workload is likely to expose this problem again.