Received: from localhost (unknown [200.46.204.187]) by postgresql.org (Postfix) with ESMTP id 96D3C2E24FB for ; Wed, 14 Nov 2007 09:42:18 -0400 (AST) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.187]) (amavisd-maia, port 10024) with ESMTP id 38308-04 for ; Wed, 14 Nov 2007 09:42:06 -0400 (AST) Received: from gauss.credativ.com (gauss.credativ.com [212.112.227.254]) by postgresql.org (Postfix) with ESMTP id 1CCFA2E0F8F for ; Wed, 14 Nov 2007 09:42:18 -0400 (AST) Received: from linuxhotel125.linuxhotel.de (port-212-202-236-178.static.qsc.de [212.202.236.178]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by gauss.credativ.com (Postfix) with ESMTP id 855A077511; Wed, 14 Nov 2007 14:42:15 +0100 (CET) From: Peter Eisentraut To: pgsql-advocacy@postgresql.org Subject: Re: Avoiding upgrade backlash Date: Wed, 14 Nov 2007 14:42:08 +0100 User-Agent: KMail/1.9.7 Cc: "Leif B. Kristensen" References: <200711121442.31233.josh@agliodbs.com> <200711141228.33802.leif@solumslekt.org> <200711141354.49400.leif@solumslekt.org> In-Reply-To: <200711141354.49400.leif@solumslekt.org> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200711141442.11913.peter_e@gmx.net> X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200711/86 X-Sequence-Number: 12903 Am Mittwoch, 14. November 2007 schrieb Leif B. Kristensen: > >from 8.2 to 8.3, it's imperative that Gentoo implements an adequate > transition procedure. Blocking the old version will create > a 'chicken-and-egg' situation where you are left incapable of upgrading > your existing databases. According to the developers, it's important > that a dump of the old database should be done with the new 8.3 version > of pg_dump. Note that this is actually not true on both accounts. Firstly, no developer has ever claimed that, and second it's factually false. It has always been "better" to use the newer pg_dump, but never "important" or imperative. -- Peter Eisentraut http://developer.postgresql.org/~petere/