From owner-freebsd-wireless@freebsd.org Wed Oct 12 15:17:14 2016 Return-Path: Delivered-To: freebsd-wireless@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 023B8C0F774 for ; Wed, 12 Oct 2016 15:17:14 +0000 (UTC) (envelope-from fbsd@opal.com) Received: from homobox.opal.com (opalcom-1-pt.tunnel.tserv4.nyc4.ipv6.he.net [IPv6:2001:470:1f06:113d::2]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "opal.com", Issuer "OpalCA" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id A9410C5A for ; Wed, 12 Oct 2016 15:17:13 +0000 (UTC) (envelope-from fbsd@opal.com) Received: from shibato (ANice-654-1-24-64.w83-197.abo.wanadoo.fr [83.197.219.64]) (authenticated bits=0) by homobox.opal.com (8.15.2/8.15.2) with ESMTPSA id u9CFH7MD065083 (version=TLSv1.2 cipher=AES128-GCM-SHA256 bits=128 verify=NO) for ; Wed, 12 Oct 2016 11:17:10 -0400 (EDT) (envelope-from fbsd@opal.com) Date: Wed, 12 Oct 2016 17:17:00 +0200 From: "J.R. Oldroyd" To: freebsd-wireless@freebsd.org Subject: Regression: ethernet + wireless/ath under lagg Message-ID: <20161012171700.5b4f9ea8@shibato> X-Mailer: Claws Mail 3.13.2 (GTK+ 2.24.29; amd64-portbld-freebsd10.1) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.4.3 (homobox.opal.com [100.64.0.6]); Wed, 12 Oct 2016 11:17:10 -0400 (EDT) X-Spam-Status: No, score=1.9 required=5.0 tests=AWL,BAYES_00, FSL_HELO_NON_FQDN_1,RCVD_IN_PBL,RCVD_IN_SORBS_DUL,RDNS_DYNAMIC shortcircuit=no autolearn=no autolearn_force=no version=3.4.1 X-Spam-Level: * X-Spam-Checker-Version: SpamAssassin 3.4.1 (2015-04-28) on homobox.opal.com X-BeenThere: freebsd-wireless@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Discussions of 802.11 stack, tools device driver development." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 12 Oct 2016 15:17:14 -0000 Since something like 9.0-release, I have used the following to config an re and an ath interface together in a lagg. The intent is to have both ethernet and wireless interfaces share the lagg and be seamlessly pluggable between the two networks. It was necessary to set the ath's MAC address to be the same as the re's, else the wlan created from the ath would not associate. (The wlan's MAC address must be the same as the underlying device's MAC, else the wlan fails WPA authentication. Changing the underlying device's MAC before the wlan is created works OK.) This config has worked just fine on both 9.x and 10.x: In /etc/rc.conf: ifconfig_re0="up" ifconfig_re0_ipv6="inet6 accept_rtadv" ifconfig_ath0="`ifconfig re0 ether`" ifconfig_ath0="ether ${ifconfig_ath0##*ether }" wlans_ath0=wlan0 create_args_wlan0="regdomain FCC country US" ifconfig_wlan0="WPA" ifconfig_wlan0_ipv6="inet6 accept_rtadv" cloned_interfaces="lagg0" ifconfig_lagg0="laggproto failover laggport re0 laggport wlan0 DHCP" # WAN_IF ifconfig_lagg0_ipv6="inet6 accept_rtadv" Now, with 11.0-release, this no longer works because the ath interface doesn't exist any more, so ifconfig cannot pre-change its MAC. And, I have been unable to find a way of either determining or setting the ath's MAC address using sysctls. Now, the following does work, hard-coding the ath's MAC, and changing the re's address to that value, instead of the other way around: ether_ath0="70:1a:11:22:33:44" # actual ath0 MAC address ifconfig_re0="ether $ether_ath0 up" ifconfig_re0_ipv6="inet6 accept_rtadv" wlans_ath0=wlan0 create_args_wlan0="regdomain FCC country US" ifconfig_wlan0="WPA" ifconfig_wlan0_ipv6="inet6 accept_rtadv" cloned_interfaces="lagg0" ifconfig_lagg0="laggproto failover laggport re0 laggport wlan0 DHCP" # WAN_IF ifconfig_lagg0_ipv6="inet6 accept_rtadv" Obviously, hard-coding a MAC address isn't a good solution. Am I missing something? Is there a way to config this without a hard-coded address? Or, do we need to fix the bug that is preventing setting the wlan's MAC address to be something different from the underlying ath's address? -jr