From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Re: bonding and ifenslave version. Date: Thu, 4 Aug 2011 09:57:04 -0700 Message-ID: <20110804095704.46c453d0@nehalam.ftrdhcpuser.net> References: <1312315615-5739-1-git-send-email-nicolas.2p.debian@free.fr> <4E38EE23.8030703@gmail.com> <4E39A3A7.1080905@gmail.com> <1718.1312402077@death> <4E39B1AB.7050502@gmail.com> <20110803143338.03988aab@nehalam.ftrdhcpuser.net> <4E3A2EBA.4040003@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Jay Vosburgh , Andy Gospodarek , netdev@vger.kernel.org To: Nicolas de =?ISO-8859-1?B?UGVzbG/8YW4=?= Return-path: Received: from mail.vyatta.com ([76.74.103.46]:55962 "EHLO mail.vyatta.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755286Ab1HDQ4y convert rfc822-to-8bit (ORCPT ); Thu, 4 Aug 2011 12:56:54 -0400 In-Reply-To: <4E3A2EBA.4040003@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, 04 Aug 2011 07:31:38 +0200 Nicolas de Peslo=FCan wrote: > Le 03/08/2011 23:33, Stephen Hemminger a =E9crit : > > On Wed, 03 Aug 2011 22:38:03 +0200 > > Nicolas de Peslo=FCan wrote: > > >> At least, we can: > >> - update the bonding documentation to clarify that ifenslave is de= precated (and move most ifenslave > >> related stuffs at the end of the documentation); > >> - possibly issue a warning when the API is used, suggesting to use= sysfs instead. > >> > >> I can take care of the documentation update if appropriate. > >> > >> Nicolas. > > > > Or I could just put a shell script that does the same thing in ipro= ute. >=20 > You mean, to replace ifenslave (from ifenslave.c) by a shell script b= y the same name that gives=20 > users the same command-line interface but use sysfs instead of the io= ctl API internally? >=20 > I had this in mind two years ago, in a private conversation while cha= nging the ifenslave package on=20 > Debian. >=20 > But chance exist that those who expect ifenslave to work would like t= o disable sysfs support if not=20 > used. I assume that totally disabling sysfs at compile time would red= uce the size of the kernel far=20 > more that disabling the ioctl API. So, for those doing embedded thing= s, this may makes sens to have=20 > an ifenslave version that doesn't rely on sysfs. >=20 > Or do you mean, to replace ifenslave (.c) by a shell script that use = netlink internally? Are all the=20 > bonding options available through netlink? I remember many conversati= on about using netlink for=20 > bonding setup, but don't remember seeing that in the actual code. The sysfs interface is the least desirable. The ifenslave script would just call 'ip link'