From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jarek Poplawski Subject: Re: [PATCH] Improve cached ECMP documentation Date: Wed, 7 Mar 2007 14:54:15 +0100 Message-ID: <20070307135415.GB2268@ff.dom.local> References: <20070307073932.GA1675@ff.dom.local> <45EEB218.8000804@linpro.no> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "David S\. Miller" , netdev@vger.kernel.org To: Tore Anderson Return-path: Received: from mx10.go2.pl ([193.17.41.74]:57935 "EHLO poczta.o2.pl" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S965336AbXCGNuS (ORCPT ); Wed, 7 Mar 2007 08:50:18 -0500 Content-Disposition: inline In-Reply-To: <45EEB218.8000804@linpro.no> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Wed, Mar 07, 2007 at 01:37:44PM +0100, Tore Anderson wrote: > Document that equal-cost multipath routing with caching does not work > for forwarded packets. > > Signed-Off-By: Tore Anderson > --- > > * Jarek Poplawski > > >It is probably one of the most often asked questions > >on the LARTC, so I'd suggest to look at its archives. > > > >[...] > > > >Multipath with caching doesn't work with forwarding. > > Well, I did look at LARTC's archives (and again after reading your > mail), still couldn't find anything. Sorry for bothering you, though. I don't feel bothered at all. Your letter could make an impression there is a lack of help and google not working, but it's not always true. So, sometimes people probably wait to see, they will only help and not do all your work. > When I disabled the support it worked just fine (no need for iptables > CONNMARK trickery or any of Julian's patches), so thanks for your help! Without this there is a possibility the route could be changed during connection. > Anyway, it strikes me as a bit odd that this feature is found under > CONFIG_IP_ADVANCED_ROUTER when it doesn't work for routers. That > could at least be better documented - here's a patch to do just that. Maybe it's intended for something even more advanced, I don't know. I checked this long time ago and later seen someone was trying to fix something around this - so I don't know what is the current state. But you are right - this should be better documented. Cheers, Jarek P.