Router-to-router operation !!

Atsushi Iwata <iwata@nwk.cl.nec.co.jp> Mon, 06 March 1995 07:30 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa20595; 6 Mar 95 2:30 EST
Received: from maelstrom.acton.timeplex.com by IETF.CNRI.Reston.VA.US id aa20591; 6 Mar 95 2:30 EST
Received: from research.gate.nec.co.jp (research.gate.nec.co.jp [202.32.8.49]) by maelstrom.acton.timeplex.com (8.6.9/ACTON-MAIN-1.2) with ESMTP id CAA13554 for <rolc@maelstrom.timeplex.com>; Mon, 6 Mar 1995 02:06:37 -0500
Received: from tomato.nwk.cl.nec.co.jp by research.gate.nec.co.jp (8.6.10+2.4W/941202) with SMTP id QAA25854; Mon, 6 Mar 1995 16:07:16 +0900
Received: by nwk.cl.nec.co.jp (5.65c/6.4JAIN-nwk-gw+2.1) id AA16633; Mon, 6 Mar 1995 16:07:12 +0900
Received: by localhost in nwk (4.1/6.4J.6-nwk_slave_1.5) id AA18938; Mon, 6 Mar 95 16:07:11 JST
Message-Id: <9503060707.AA18938@ninjo.nwk.cl.nec.co.jp>
To: rolc@maelstrom.timeplex.com
Cc: dkatz@cisco.com
Subject: Router-to-router operation !!
Date: Mon, 06 Mar 95 16:07:11 +0900
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Atsushi Iwata <iwata@nwk.cl.nec.co.jp>

ROLC working group members,

Though I thought that there have been the same kind of discussions,
unfortunately I cannot understand clearly why and when
router-to-router problems occur, in the "draft-ietf-rolc-nhrp-03.txt". 
Could anyone please show me an abstract example with figures ?

>>8.1 Router-to-Router Operation
>>
>>   In practice, the initiating and responding stations may be either
>>   hosts or routers.  However, there is a possibility under certain
>>   conditions that a stable routing loop may occur if NHRP is used
>>   between two routers.  This situation can be avoided if there are no
>>   "back door" paths between the entry and egress router outside of the
>>   NBMA network, and can be ameliorated by periodically reissuing the
>>   NHRP request.  If these conditions cannot be satisfied, the use of
>>   NHRP between routers is not recommended.
>>
>>   One approach to the router-to-router case that is being considered is
>>   to run a limited instance of a routing protocol between the two
>>   routers.  Any routing protocol that provides loop detection may be
>>   used.  This routing protocol instance will likely only carry a subset
>>   of the total routing information, and is unlikely to be closely
>>   integrated into the routing in which each of the routers is otherwise
>>   participating (due to the abitrary connectivity possible in such
>>   situations and its impact on the stability and quality of overall
>>   routing).  This approach is for further study.

Thank you.

  ***------------------------------------------------------------***
     Atsushi Iwata  Network Research Lab., C&C Research Labs.,NEC
	4-1-1 Miyazaki Miyamae-ku Kawasaki Kanagawa, JAPAN 216
Tel:+81-44-856-2123, Fax:+81-44-856-2230,E-Mail:iwata@nwk.cl.nec.co.jp