From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Martin A. Brown" Date: Fri, 09 May 2003 15:55:39 +0000 Subject: Re: [LARTC] load balancing Message-Id: List-Id: References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: lartc@vger.kernel.org Hello V=EDctor, : I tried this and it works when an interface goes down, but how can I do : if it's not an interface but one of these gateways that goes down? I : would like that in this case, the apropriate "nexthop" drops too, so my : traffic goes through the other two, and when that fallen gateway is : once again alive, the "nexthop" comes alive too, so it's once again : used. How can I do this? You probably should try out the dead gateway detection patches by Julian Anastasov [1] documented here [2]. I have not yet used the DGD patches, but my understanding is that they'll detect unreachability of the nexthop and mark a route as dead until the nexthop becomes reachable again. As I'm sure you can imagine, it's much more difficult to identify problems further upstream--and that's what dynamic routing protocols are for, anyway. : >/sbin/ip route add equalize \ : > nexthop via $ADSL_GW1 dev $ADSL_IFACE1 weight 1\ : > nexthop via $ADSL_GW2 dev $ADSL_IFACE2 weight 1\ : > nexthop via $ADSL_GW3 dev $ADSL_IFACE3 weight 1 Best of luck, -Martin [1] http://www.ssi.bg/~ja/#routes [2] http://www.ssi.bg/~ja/dgd-usage.txt --=20 Martin A. Brown --- SecurePipe, Inc. --- mabrown@securepipe.com _______________________________________________ LARTC mailing list / LARTC@mailman.ds9a.nl http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/