Date: Sun, 31 Dec 2000 18:51:02 -0800 (PST) From: Tom Samplonius <tom@sdf.com> To: Vincent Poy <vince@oahu.WURLDLINK.NET> Cc: "Seo Boon, NG" <sbng@employees.org>, freebsd-isp@FreeBSD.ORG Subject: Re: Dynamic routing reference sites Message-ID: <Pine.BSF.4.05.10012311841020.16414-100000@misery.sdf.com> In-Reply-To: <Pine.BSF.4.31.0012311401010.2211-100000@oahu.WURLDLINK.NET>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 31 Dec 2000, Vincent Poy wrote: > > you typically do not get anything close to a full view when you peer. A peer > > only announce itself and it's customer. Hence, u can have many peers but it's > > rarely that u'll see 'gigs' of routes for all the peers. > > Yes but aren;t you supposed to get routes from each peer to build > your own routing table? Different meanings of the word "peer". It can mean any BGP peer, or it can mean a network peer, as opposed to a provider-customer relationship. If Sprint and MCI were to establish a new interconnection, they will peer, and usually offer each other routes for their own AS only so they don't transit traffic for each other. BGP route table sizes are likely different everywhere. There can differences in filters (either on what is being let in, or being let out). Some carriers have rather restrictive route acceptance policies. Typically route filters are designed to prevent tiny network routes from filling their tables. Also, routes for private local ASes are cause differences. Can we please let this thread die now? If you want to learn about BGP one fact at time, you'll be here a _long_ time. Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.05.10012311841020.16414-100000>