Directed ARP
RFC 1433

Document Type RFC - Experimental (March 1993; No errata)
Authors Jeff Wong , John Garrett  , John Hagan 
Last updated 2013-03-02
Stream IETF
Formats plain text html pdf htmlized bibtex
Stream WG state (None)
Document shepherd No shepherd assigned
IESG IESG state RFC 1433 (Experimental)
Consensus Boilerplate Unknown
Telechat date
Responsible AD (None)
Send notices to (None)
Network Working Group                                        J. Garrett
Request for Comments: 1433                       AT&T Bell Laboratories
                                                               J. Hagan
                                             University of Pennsylvania
                                                                J. Wong
                                                 AT&T Bell Laboratories
                                                             March 1993

                              Directed ARP

Status of this Memo

   This memo defines an Experimental Protocol for the Internet
   community.  Discussion and suggestions for improvement are requested.
   Please refer to the current edition of the "IAB Official Protocol
   Standards" for the standardization state and status of this protocol.
   Distribution of this memo is unlimited.


   A router with an interface to two IP networks via the same link level
   interface could observe that the two IP networks share the same link
   level network, and could advertise that information to hosts (via
   ICMP Redirects) and routers (via dynamic routing protocols).
   However, a host or router on only one of the IP networks could not
   use that information to communicate directly with hosts and routers
   on the other IP network unless it could resolve IP addresses on the
   "foreign" IP network to their corresponding link level addresses.
   Directed ARP is a dynamic address resolution procedure that enables
   hosts and routers to resolve advertised potential next-hop IP
   addresses on foreign IP networks to their associated link level


   The authors are indebted to Joel Halpern of Network Systems
   Corporation and David O'Leary who provided valuable comments and
   insight to the authors, as well as ongoing moral support as the
   presentation of this material evolved through many drafts.  Members
   of the IPLPDN working group also provided valuable comments during
   presentations and through the IPLPDN mailing list.  Chuck Hedrick of
   Rutgers University, Paul Tsuchiya of Bell Communications Research,
   and Doris Tillman of AT&T Bell Laboratories provided early insight as
   well as comments on early drafts.

Garrett, Hagan & Wong                                           [Page 1]
RFC 1433                      Directed ARP                    March 1993

1.  Terminology

   A "link level network" is the upper layer of what is sometimes
   referred to (e.g., OSI parlance) as the "subnetwork", i.e., the
   layers below IP.  The term "link level" is used to avoid potential
   confusion with the term "IP sub-network", and to identify addresses
   (i.e., "link level address") associated with the network used to
   transport IP datagrams.

   From the perspective of a host or router, an IP network is "foreign"
   if the host or router does not have an address on the IP network.

2.  Introduction

   Multiple IP networks may be administered on the same link level
   network (e.g., on a large public data network).  A router with a
   single interface on two IP networks could use existing routing update
   procedures to advertise that the two IP networks shared the same link
   level network.  Cost/performance benefits could be achieved if hosts
   and routers that were not on the same IP network could use that
   advertised information, and exchange packets directly, rather than
   through the dual addressed router.  But a host or router can not send
   packets directly to an IP address without first resolving the IP
   address to its link level address.

   IP address resolution procedures are established independently for
   each IP network.  For example, on an SMDS network [1], address
   resolution may be achieved using the Address Resolution Protocol
   (ARP) [2], with a separate SMDS ARP Request Address (e.g., an SMDS
   Multicast Group Address) associated with each IP network.  A host or
   router that was not configured with the appropriate ARP Request
   Address would have no way to learn the ARP Request Address associated
   with an IP network, and would not send an ARP Request to the
   appropriate ARP Request Address.  On an Ethernet network a host or
   router might guess that an IP address could be resolved by sending an
   ARP Request to the broadcast address.  But if the IP network used a
   different address resolution procedure (e.g., administered address
   resolution tables), the ARP Request might go unanswered.

   Directed ARP is a procedure that enables a router advertising that an
   IP address is on a shared link level network to also aid in resolving
   the IP address to its associated link level address.  By removing
   address resolution constraints, Directed ARP enables dynamic routing
   protocols such as BGP [3] and OSPF [4] to advertise and use routing
   information that leads to next-hop addresses on "foreign" IP
   networks.  In addition, Directed ARP enables routers to advertise
Show full document text