From owner-freebsd-doc@FreeBSD.ORG Sun Jan 27 07:44:30 2013 Return-Path: Delivered-To: freebsd-doc@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by hub.freebsd.org (Postfix) with ESMTP id 9188142B for ; Sun, 27 Jan 2013 07:44:30 +0000 (UTC) (envelope-from ike@blackskyresearch.net) Received: from rs149.luxsci.com (rs149.luxsci.com [64.49.224.181]) by mx1.freebsd.org (Postfix) with ESMTP id 3BFEE391 for ; Sun, 27 Jan 2013 07:44:30 +0000 (UTC) Received: from rs149.luxsci.com (localhost.localdomain [127.0.0.1]) by rs149.luxsci.com (8.14.4/8.13.8) with ESMTP id r0R7iRBN018593; Sun, 27 Jan 2013 02:44:27 -0500 Received: (from root@localhost) by rs149.luxsci.com (8.14.4/8.13.8/Submit) id r0R7i4HY018494; Sun, 27 Jan 2013 07:44:04 GMT Received: (from sender 74627) (rs149.luxsci.com [127.0.0.1]) by LuxSci SP; Sun, 27 Jan 2013 07:44:04 +0000 Subject: Re: removing CVS in Handbook Updating and Upgrading chapter Content-Type: text/plain; charset=us-ascii From: "Isaac (.ike) Levy" In-Reply-To: Date: Sun, 27 Jan 2013 02:43:06 -0500 Content-Transfer-Encoding: quoted-printable References: <1359241802-3572135.75152325.fr0QN9mrI032137@rs149.luxsci.com> To: Eitan Adler X-Lux-Comment: Message r0R7h69h018255 sent by user #74627 Message-Id: <1359272644-7225342.63753895.fr0R7h69h018255@rs149.luxsci.com> X-Comment: LuxSci SP Message ID - 1359272644-7225342.63753895 Cc: freebsd-doc@freebsd.org X-BeenThere: freebsd-doc@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Documentation project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 27 Jan 2013 07:44:30 -0000 On Jan 26, 2013, at 9:26 PM, Eitan Adler wrote: > On 26 January 2013 21:05, Warren Block >=20 >>> Not sure if I need to explain this, but: >>> For a large number of system integrators, building userland/kernel = from >>> source is critical. >>> Most of these builds happen before ports/pkg get installed, (if they = even >>> do). The current state of SVN, binary packages, ports mechanism = changes, >>> and otherwise- all make for some nasty chicken/egg problems for many = systems >>> integrators. >=20 > We should certainly be striving to remove references to CVS from as > many places as possible. =20 That, I agree with, excepting fetching src/base for the = buildworld;makeworld dance. cvsup for base/src does not even have an end-of-life date yet, and the = replacements are either: - not clear - not finished - complex, confused, and difficult for systems integrators (people who buildworld/buildkernel as the stable base underneath = everything) > I have no objections to leaving one > (obviously marked) chapter about the old way of doing things. For developers, I can agree that cvsup is totally the wrong tool. Yet, that information should maybe be in the developers handbook, = perhaps in the "tools" section: = http://www.freebsd.org/doc/en_US.ISO8859-1/books/developers-handbook/tools= .html (If yall' think that's a good idea, I'll happily write up a page to = start with) >> This part of the Handbook refers to fetching source for -CURRENT or = -STABLE. >> We should not be suggesting CVS to new users who want to run = development >> versions of FreeBSD. =20 Oh, I see your perspective, but FreeBSD systems administrators are well = accustomed to running -CURRENT and -STABLE, along with -REL and -RELENG, = in production. Here's some circumstances admins may build STABLE: - Prepping an environment for upcoming releases - Providing QA/bugs back to FreeBSD *before* REL goes out the door - STABLE may contain a necessary bugfix, security fix, or other = necessary feature in a given environment Here's some circumstances admins may build a given moment of CURRENT: - All the same issues from STABLE could warrant some point of CURRENT - some existing code may need to be manually backported quickly in an = emergency (I've done my fair share of "grep hacking" to solve crisis situations = as an SA) I hope that's understood? Best, .ike