Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261727AbVEPQQm (ORCPT ); Mon, 16 May 2005 12:16:42 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261731AbVEPQQl (ORCPT ); Mon, 16 May 2005 12:16:41 -0400 Received: from w240.dkm.cz ([62.24.88.240]:15370 "EHLO mail.spitalnik.net") by vger.kernel.org with ESMTP id S261727AbVEPQQj convert rfc822-to-8bit (ORCPT ); Mon, 16 May 2005 12:16:39 -0400 From: Jan Spitalnik To: Linux Kernel Mailing List Subject: Re: RR for route Date: Mon, 16 May 2005 18:16:30 +0200 User-Agent: KMail/1.8.50 References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-2" Content-Transfer-Encoding: 8BIT Content-Disposition: inline Message-Id: <200505161816.31171.jan@spitalnik.net> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1267 Lines: 37 Dne po 16. kv?tna 2005 18:04 Jan Engelhardt napsal(a): > Hello, > > > does anybody know of a way to round-robin over multiple default gateways > with the same metric? Currently, it looks like it's always the first GW > which is taken. > > > Jan Engelhardt Hi, this not round robin but the traffic is spread over more GW's config IP_ROUTE_MULTIPATH bool "IP: equal cost multipath" depends on IP_ADVANCED_ROUTER help Normally, the routing tables specify a single action to be taken in a deterministic manner for a given packet. If you say Y here however, it becomes possible to attach several actions to a packet pattern, in effect specifying several alternative paths to travel for those packets. The router considers all these paths to be of equal "cost" and chooses one of them in a non-deterministic fashion if a matching packet arrives. -- Jan Spitalnik jan@spitalnik.net If everyone used log base e, we'd all be happy! - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/