re: EBGP over unnumbered serial lines

toconnor@bbn.com Thu, 28 September 1995 21:34 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa18913; 28 Sep 95 17:34 EDT
Received: from CNRI.Reston.VA.US by IETF.CNRI.Reston.VA.US id aa18909; 28 Sep 95 17:34 EDT
Received: from interlock.ans.net by CNRI.Reston.VA.US id aa21650; 28 Sep 95 17:34 EDT
Received: by interlock.ans.net id AA60566 (InterLock SMTP Gateway 3.0 for iwg-out@ans.net); Thu, 28 Sep 1995 17:23:59 -0400
Message-Id: <199509282123.AA60566@interlock.ans.net>
Received: by interlock.ans.net (Protected-side Proxy Mail Agent-1); Thu, 28 Sep 1995 17:23:59 -0400
To: bgp@ans.net
Subject: re: EBGP over unnumbered serial lines
Date: Thu, 28 Sep 95 17:20:56 -0400
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: toconnor@bbn.com

Radha Gowda <rxg@proteon.com> said

> Here is Yakov's response to my question ...
>
>  >> I think that the "same subnet" restriction is mostly intended to
>  >> make sure that the external neighbors are one hop away from each 
>  >> other (on a common Data Link subnetwork).

BBN supports the idea of a "private" link between 2 routers. In this case
the Next Hop attribute is not the address of this link, but another of the
router's interfaces. This works with both external and internal neighbors.

Of course there are a number of special cases, but the basic idea is that
when a packet is sent out over the private link, the return address must
be one of the non-private interfaces of the sender. If the routers are
internal neighbors, then each has link-state info for the other, so the
return route exists. If they are external neighbors, then each advertises
their interfaces to the other, so again a return route exists.