From owner-freebsd-current@FreeBSD.ORG Sat Apr 17 20:21:26 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 354B016A4CE for ; Sat, 17 Apr 2004 20:21:26 -0700 (PDT) Received: from sccrmhc12.comcast.net (sccrmhc12.comcast.net [204.127.202.56]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6EBB743D39 for ; Sat, 17 Apr 2004 20:21:25 -0700 (PDT) (envelope-from adam@migus.org) Received: from ganyopa.migus.org ([68.55.142.251]) by comcast.net (sccrmhc12) with ESMTP id <200404180321240120019rube>; Sun, 18 Apr 2004 03:21:24 +0000 Received: from localhost (localhost [127.0.0.1]) by ganyopa.migus.org (Postfix) with ESMTP id 475AD4D469; Sat, 17 Apr 2004 23:21:24 -0400 (EDT) Received: from ganyopa.migus.org ([127.0.0.1]) by localhost (ganyopa.migus.org [127.0.0.1]) (amavisd-new, port 10025) with ESMTP id 18638-09; Sat, 17 Apr 2004 23:21:23 -0400 (EDT) Received: by ganyopa.migus.org (Postfix, from userid 72) id 8A51B4C1C4; Sat, 17 Apr 2004 23:21:23 -0400 (EDT) Received: from ludo.migus.org (ludo.migus.org [192.168.4.4]) by webservices.migus.org (IMP) with HTTP for ; Sat, 17 Apr 2004 23:21:23 -0400 Message-ID: <1082258483.4081f43366454@webservices.migus.org> Date: Sat, 17 Apr 2004 23:21:23 -0400 From: "Adam C. Migus" To: Brooks Davis References: <1082233506.408192a2c476a@webservices.migus.org> <20040417235302.GC3534@Odin.AC.HMC.Edu> <20040417201913.P34536@sasami.jurai.net> <20040418005436.GC14645@Odin.AC.HMC.Edu> In-Reply-To: <20040418005436.GC14645@Odin.AC.HMC.Edu> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit User-Agent: Internet Messaging Program (IMP) 3.2.2 X-Originating-IP: 192.168.4.4 X-Virus-Scanned: by amavisd-new at migus.org cc: current@freebsd.org Subject: Re: VLAN support in rcNG [PATCH] X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 18 Apr 2004 03:21:26 -0000 Quoting Brooks Davis : > On Sat, Apr 17, 2004 at 08:20:17PM -0400, Matthew N. Dodd wrote: > > On Sat, 17 Apr 2004, Brooks Davis wrote: > > > It works by doing a major overhaul of the device cloning support > so you > > > can create interfaces of the form . and get an > vlan > > > interface with the appropriate parent and tag. > > > > Please make this behavior optional. > > The entire patch has exactly one change which is visiable to the > userland unless the user changes their behavior. That change is > that > when you do "ifconfig stf create" ifconfig does not output "stf0" > and > the interface created is named "sft" instead of "stf0". This is a > result of a minor change that allows you to create "6to4" a more > aptly named interface. This change is mostly an example of how to > write > new style interface cloners that's easier to understand then the > vlan > version which was the motivation for this overhaul. > > The vlan change allows vlan interfaces to work exactly like they did > before. It just adds some sugar so that the following sets of > commands > are equivalent: > > ifconfig xl0.2 > > vlanif=`ifconfig vlan create` > ifconfig $vlanif name xl0.2 > ifconfig xl0.2 vlandev xl0 vlantag 2 > > -- Brooks > > -- > Any statement of the form "X is the one, true Y" is FALSE. > PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4 > So it sounds like I say: cloned_interfaces="xl0.2" and get myself the interface I want, on the right vlan but what happens when I want to say: ifconfig_xl0.2="DHCP" That `.' is going to cause some problems isn't it? I asked earlier for reasons like this and compatability in general if there would be a way to create vlan interfaces named "vlan2" for example. The trouble with this naming convention is it doesn't imply it's physical device in it's name but the name doesn't contain a `.' which in things like sh(1) can cause problems. My patch may be a little more cumbersome, requiring explicit naming of the device but it does eliminiate ambiquity and permit that naming scheme with the ability to do DHCP and not trip over some other things in rcNG. -- Adam C. Migus -- http://people.migus.org/~adam/