From owner-freebsd-isp Wed Aug 14 07:39:08 1996 Return-Path: owner-isp Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id HAA02404 for isp-outgoing; Wed, 14 Aug 1996 07:39:08 -0700 (PDT) Received: from brasil.moneng.mei.com (brasil.moneng.mei.com [151.186.109.160]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id HAA02395 for ; Wed, 14 Aug 1996 07:39:01 -0700 (PDT) Received: (from jgreco@localhost) by brasil.moneng.mei.com (8.7.Beta.1/8.7.Beta.1) id JAA11682; Wed, 14 Aug 1996 09:37:14 -0500 From: Joe Greco Message-Id: <199608141437.JAA11682@brasil.moneng.mei.com> Subject: Re: gated config for ospf and bgp To: jdd@vbc.net (Jim Dixon) Date: Wed, 14 Aug 1996 09:37:14 -0500 (CDT) Cc: mike@networx.ie, freebsd-isp@freebsd.org In-Reply-To: from "Jim Dixon" at Aug 13, 96 11:43:07 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-isp@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > On Tue, 13 Aug 1996, Michael Ryan wrote: > > > Has anybody got a sample gated config file which shows how > > to configure it for OSPF and BGP? > > > > If anybody replies, could you please also indicate your > > router topology, so I can understand the config file? > > OSPF and BGP4 can both get complex. What is your intended configuration? I don't know about Michael, but I'm looking to provide internal and external redundancy. Internally I'd like to provide dual backbone Ethernets and probably do some basic load balancing between them, but set it up so that in case of a hub or switch failure, or problem with an Ethernet NIC or something, the load automatically switches to the "good" network. >From an implementation point of view, customers here are generally connected via a single circuit, and connect to a border router. Our outbound connection to the Internet is done the same way. Our core routers either connect to other local equipment (DNS, mail, news servers, etc) or to core routers at other POP's. What I would like to see... -------- net1-> o-----|Core1 | -------------------------| -------- | | | |-------------------------------------------------------------------------| | | | | | | | | | | -------- -------- -------- -------- -------- -------- -------- -------- | |Border| |Border| |Border| |Border| |Border| |Border| |Border| |Border| | -------- -------- -------- -------- -------- -------- -------- -------- | | | | | | | | | | |-------------------------------------------------------------------------| | | | -------- -------------------------------|Core2 | -------- where some number "n" of the border routers connect to backbone service providers (and speak BGP4), others speak to customers, and independent core routers duplicate services (i.e. multiple nets, to which both core routers connect, such as net1 above). The strategy is obviously to attempt to provide continuous IP connectivity through a single point of failure. Externally, I would like to peer with multiple service providers, and I do not want to do so off of the same border router. That's _my_ goal. :-) ... JG