From owner-freebsd-questions@FreeBSD.ORG Thu Sep 17 17:21:01 2009 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 63A40106566C for ; Thu, 17 Sep 2009 17:21:01 +0000 (UTC) (envelope-from mel.flynn+fbsd.questions@mailing.thruhere.net) Received: from mailhub.rachie.is-a-geek.net (rachie.is-a-geek.net [66.230.99.27]) by mx1.freebsd.org (Postfix) with ESMTP id 32D8F8FC13 for ; Thu, 17 Sep 2009 17:21:00 +0000 (UTC) Received: from smoochies.rachie.is-a-geek.net (mailhub.lan.rachie.is-a-geek.net [192.168.2.11]) by mailhub.rachie.is-a-geek.net (Postfix) with ESMTP id 93E277E818; Thu, 17 Sep 2009 09:21:13 -0800 (AKDT) From: Mel Flynn To: freebsd-questions@freebsd.org Date: Thu, 17 Sep 2009 19:20:58 +0200 User-Agent: KMail/1.12.1 (FreeBSD/8.0-BETA4; KDE/4.3.1; i386; ; ) References: <7B9397B189EB6E46A5EE7B4C8A4BB7CB3042DC5F@MBX03.exg5.exghost.com> <200909161941.04441.mel.flynn+fbsd.questions@mailing.thruhere.net> <7B9397B189EB6E46A5EE7B4C8A4BB7CB3042DDC4@MBX03.exg5.exghost.com> In-Reply-To: <7B9397B189EB6E46A5EE7B4C8A4BB7CB3042DDC4@MBX03.exg5.exghost.com> MIME-Version: 1.0 Content-Type: Text/Plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Message-Id: <200909171920.58874.mel.flynn+fbsd.questions@mailing.thruhere.net> Cc: Peter Steele Subject: Re: Can lagg0 failback be prevented? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 17 Sep 2009 17:21:01 -0000 On Wednesday 16 September 2009 20:58:45 Peter Steele wrote: > >Not really, unless you manually change master. However I believe this also > > causes a slight or even bigger network outage. Any reason you're not > > using loadbalance algorithm, since it seems to suit you better? > > Our resident network guru is quite opposed to using the loadbalancing > option since it comes with a lot of potentially undesirable baggage of its > own... Then your best option is to patch lagg(4) with an "avail" algorithm, that prefers $master and sticks with an interface till it's detected down. When done properly the chances are good to get this into base. Another approach would be to change the failover with a 'fader' algorithm, that gradually fades from one nic to the other, kind of like an audio mixer, though I'm not sure if that's possible and would work satisfactory. -- Mel