From owner-freebsd-stable@FreeBSD.ORG Wed Apr 2 00:26:57 2003 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id EA3EB37B401 for ; Wed, 2 Apr 2003 00:26:57 -0800 (PST) Received: from CRWdog.demon.co.uk (12-228-200-39.client.attbi.com [12.228.200.39]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6B51243F93 for ; Wed, 2 Apr 2003 00:26:57 -0800 (PST) (envelope-from spadger@best.com) Received: from localhost (localhost [127.0.0.1]) by CRWdog.demon.co.uk (Postfix) with ESMTP id 7A80A9E; Wed, 2 Apr 2003 00:26:21 -0800 (PST) X-Mailer: exmh version 2.5 07/13/2001 with nmh-1.0.4 To: Gunnar Flygt In-Reply-To: Message from Gunnar Flygt of "Wed, 02 Apr 2003 09:09:52 +0200." <20030402070952.GD37762@sr.se> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 2003 00:26:21 -0800 From: Andy Sparrow Message-Id: <20030402082621.7A80A9E@CRWdog.demon.co.uk> cc: stable@freebsd.org Subject: Re: 4.8-RELEASE vs SA-03:07 X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 02 Apr 2003 08:26:58 -0000 > On Tue, Apr 01, 2003 at 06:21:10PM -0800, Kris Kennaway wrote: > > On Tue, Apr 01, 2003 at 06:12:44PM -0800, Marcel Moolenaar wrote: > > > > > I think this does not automaticly follow. If you use the -RC label for > > > identifying the release when it's in a state of final QA and not to > > > identify the release when it's in -ALPHA or -BETA state, then you avoid > > > using the -RELEASE label when it's still possible that tags slide. > > > > And what about the last-minute (but easily fixed) bug that is > > discovered after the -RELEASE tag goes down, whenever that happens in > > the process? Either you slide the tag to fix the bug or you don't. > > That's what the original poster was talking about. > > > > Kris > > Isn't this much fuzz for nothing, really? All there is to it is > wait for the "Release Officer" to announce that 4.8 i RELEASED. > > Then it is Released. Yes. I don't really understand why the Release Engineering is done the way it is. When I was releasing commercial software, I'd simply lock and tag the tree with a unique, temporary, label, check out a clean tree against that label, unlock the tree and build and package the software. If the build subsequently passed QA, I laid down another label based on the one used to check out the passing build. All the temporary build tags would get removed automatically anyway when they were more than a few days old. You just had to lay down additional tags for "milestones" that you wanted to permanently get back to (e.g. pre-releases, various betas, code forks etc.) If you're only building releases from source trees checked out against labels, there's little need to keep the tree frozen for weeks at a time, it seems to me. Cheers, AS