From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nicolas Dichtel Subject: Re: [PATCH net-next 0/3 v4] changes to make ipv4 routing table aware of next-hop link status Date: Wed, 17 Jun 2015 21:52:11 +0200 Message-ID: <5581CFEB.3010000@6wind.com> References: <1434386001-1157-1-git-send-email-gospo@cumulusnetworks.com> <55819ACC.3010702@6wind.com> <20150617191012.GJ588@gospo.home.greyhouse.net> Reply-To: nicolas.dichtel@6wind.com Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org, davem@davemloft.net, ddutt@cumulusnetworks.com, sfeldma@gmail.com, alexander.duyck@gmail.com, hannes@stressinduktion.org, stephen@networkplumber.org To: Andy Gospodarek Return-path: Received: from mail-wg0-f45.google.com ([74.125.82.45]:33467 "EHLO mail-wg0-f45.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751067AbbFQTwO (ORCPT ); Wed, 17 Jun 2015 15:52:14 -0400 Received: by wgez8 with SMTP id z8so45652987wge.0 for ; Wed, 17 Jun 2015 12:52:13 -0700 (PDT) In-Reply-To: <20150617191012.GJ588@gospo.home.greyhouse.net> Sender: netdev-owner@vger.kernel.org List-ID: Le 17/06/2015 21:10, Andy Gospodarek a =E9crit : > On Wed, Jun 17, 2015 at 06:05:32PM +0200, Nicolas Dichtel wrote: >> Le 15/06/2015 18:33, Andy Gospodarek a =E9crit : >>> This series adds the ability to have the Linux kernel track whether= or >>> not a particular route should be used based on the link-status of t= he >>> interface associated with the next-hop. >>> >>> Before this patch any link-failure on an interface that was serving= as a >>> gateway for some systems could result in those systems being isolat= ed >> >from the rest of the network as the stack would continue to attempt= to >>> send frames out of an interface that is actually linked-down. When= the >>> kernel is responsible for all forwarding, it should also be respons= ible >>> for taking action when the traffic can no longer be forwarded -- th= ere >>> is no real need to outsource link-monitoring to userspace anymore. >>> >>> This feature is only enabled with the new per-interface or ipv4 glo= bal >>> sysctls called 'ignore_routes_with_linkdown'. >>> >>> net.ipv4.conf.all.ignore_routes_with_linkdown =3D 0 >>> net.ipv4.conf.default.ignore_routes_with_linkdown =3D 0 >>> net.ipv4.conf.lo.ignore_routes_with_linkdown =3D 0 >>> ... >> Sorry for my late reply, but is it possible to advertise this sysctl= via the >> netconf infra (grep NETCONFA_)? > I would be happy to do that. Do you see it as a requirement for this > set to be accepted? No, but it would be great to do it before net-next closes. > > I've got to add ipv6 support, so netconf support could also easily be > added in the kernel and iproute2. Ok.