Network Working Group                                          A. Retana
Request for Comments: 3021                                      R. White
Category: Standards Track                                  Cisco Systems
                                                              V. Fuller
                                                    GTE Internetworking
                                                           D. McPherson
                                                         Amber Networks
                                                          December 2000


          Using 31-Bit Prefixes on IPv4 Point-to-Point Links

Status of this Memo

  This document specifies an Internet standards track protocol for the
  Internet community, and requests discussion and suggestions for
  improvements.  Please refer to the current edition of the "Internet
  Official Protocol Standards" (STD 1) for the standardization state
  and status of this protocol.  Distribution of this memo is unlimited.

Copyright Notice

  Copyright (C) The Internet Society (2000).  All Rights Reserved.

Abstract

  With ever-increasing pressure to conserve IP address space on the
  Internet, it makes sense to consider where relatively minor changes
  can be made to fielded practice to improve numbering efficiency.  One
  such change, proposed by this document, is to halve the amount of
  address space assigned to point-to-point links (common throughout the
  Internet infrastructure) by allowing the use of 31-bit subnet masks
  in a very limited way.

1. Introduction and Motivation

  The perceived problem of a lack of Internet addresses has driven a
  number of changes in address space usage and a number of different
  approaches to solving the problem:

  - More stringent address space allocation guidelines, enforced by the
    IANA and the regional address assignment authorities [RFC2050].

  - Use of Network Address Translators (NATs), where a small number of
    IANA-compliant addresses are shared by a larger pool of private,
    non-globally routed addresses topologically behind a NAT box
    [RFC1631].




Retana, et al.              Standards Track                     [Page 1]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


  - Deployment of a new Internet Protocol to increase the size of the
    address space.  One such protocol, IPv6 [RFC2460], has been through
    the IETF process but has yet to see production deployment.  Should
    it be, deployed, it will still face a many year transition period.

  Prior to the availability of a larger address space, it seems prudent
  to consider opportunities for making more efficient use of the
  existing address space.

  One such (small) opportunity is to change the way that point-to-point
  links are numbered.  One option, which is used today on some parts of
  the Internet, is to simply not number point-to-point links between
  routers.  While this practice may seem, at first, to handily resolve
  the problem, it causes a number of problems of its own, including the
  inability to consistently manage the unnumbered link or reach a
  router through it, difficulty in management and debugging of those
  links, and the lack of standardization [RFC1812].

  In current practice, numbered Internet subnets do not use longer than
  a 30-bit subnet mask (in most cases), which requires four addresses
  per link - two host addresses, one all-zeros network, and one all-
  ones broadcast.  This is unfortunate for point-to-point links, since
  they can only possibly have two identifying endpoints and don't
  support the notion of broadcast - any packet which is transmitted by
  one end of a link is always received by the other.

  A third option is to use host addresses on both ends of a point-to-
  point link.  This option provides the same address space savings as
  using a 31-bit subnet mask, but may only be used in links using PPP
  encapsulation [RFC1332].  The use of host addresses allows for the
  assignment of IP addresses belonging to different networks at each
  side of the link, causing link and network management not to be
  straight forward.

  This document is based on the idea that conserving IP addresses on
  point-to-point links (using longer than a 30-bit subnet mask) while
  maintaining manageability and standard interaction is possible.
  Existing documentation [RFC950] has already hinted at the possible
  use of a 1-bit wide host-number field.

  The savings in address space resulting from this change is easily
  seen--each point-to-point link in a large network would consume two
  addresses instead of four.  In a network with 500 point-to-point
  links, for example, this practice would amount to a savings of 1000
  addresses (the equivalent of four class C address spaces).






Retana, et al.              Standards Track                     [Page 2]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


2. Considerations of 31-Bit Prefixes

  This section discusses the possible effects, on Internet routing and
  operations, of using 31-bit prefixes on point-to-point links.  The
  considerations made here are also reflected in Section 3.

  For the length of this document, an IP address will be interpreted
  as:

       <Network-number><Host-number>

  where the <Host-number> represents the unmasked portion of the
  address and it SHOULD be at least 1 bit wide.  The "-1" notation is
  used to mean that the field has all 1 bits.  For purposes of this
  discussion, the routing system is considered capable of classless, or
  CIDR [RFC1519], routing.

2.1. Addressing

  If a 31-bit subnet mask is assigned to a point-to-point link, it
  leaves the <Host-number> with only 1 bit.  Consequently, only two
  possible addresses may result:

       {<Network-number>, 0} and {<Network-number>, -1}

  These addresses have historically been associated with network and
  broadcast addresses (see Section 2.2).  In a point-to-point link with
  a 31-bit subnet mask, the two addresses above MUST be interpreted as
  host addresses.

2.2. Broadcast and Network Addresses

  There are several historically recognized broadcast addresses
  [RFC1812] on IP segments:

     (a) the directed broadcast

          {<Network-number>, -1}

          {<Network-number>, 0}

        The network address itself {<Network-number>, 0} is an
        obsolete form of directed broadcast, but it may still be used
        by older hosts.







Retana, et al.              Standards Track                     [Page 3]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


     (b) the link local (or limited) broadcast

          {-1, -1}

          {0, 0}

        The {0, 0} form of a limited broadcast is obsolete, but may
        still be present in a network.

  Using a 31-bit prefix length leaves only two numbering possibilities
  (see Section 2.1), eliminating the use of a directed broadcast to the
  link (see Section 2.2.1).  The limited broadcast MUST be used for all
  broadcast traffic on a point-to-point link with a 31-bit subnet mask
  assigned to it.

  The <Network-number> is assigned by the network administrator as
  unique to the local routing domain.  The decision as to whether a
  destination IP address should be a directed broadcast or not is made
  by the router directly connected to the destination segment.  Current
  forwarding schemes and algorithms are not affected in remote routers.

  The intent of this document is to discuss the applicability and
  operation of 31-bit prefixes on point-to-point links.  The effects
  (if any) on other types of interfaces are not considered.

2.2.1. Directed Broadcast

  When a device wants to reach all the hosts on a given (remote, rather
  than directly connected) subnet, it may set the packet's destination
  address to the link's subnet broadcast address.  This operation is
  not possible for point-to-point links with a 31-bit prefix.

  As discussed in Section 6, the loss of functionality of a directed
  broadcast may actually be seen as a beneficial side effect, as it
  slightly enhances the network's resistance to a certain class of DoS
  Attacks [RFC2644, SMURF].

2.3. Impact on Current Routing Protocols

  Networks with 31-bit prefixes have no impact on current routing
  protocols.  Most of the currently deployed routing protocols have
  been designed to provide classless routing.  Furthermore, the
  communication between peers is done using multicast, limited
  broadcast or unicast addresses (all on the local network), none of
  which are affected with the use of 31-bit subnet masks.






Retana, et al.              Standards Track                     [Page 4]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


3. Recommendations

  The considerations presented in Section 2 affect other published
  work.  This section details the updates made to other documents.

3.1. "Requirements for Internet Hosts -- Communication Layers" [RFC1122]

  Section 3.2.1.3 (e) is replaced with:

     (e)  { <Network-number>, <Subnet-number>, -1 }

        Directed broadcast to the specified subnet.  It MUST NOT be
        used as a source address, except when the originator is one of
        the endpoints of a point-to-point link with a 31-bit mask.

  A new section (numbered 3.2.1.3 (h)) is added:

     (h)  { <Network-number>, <Subnet-number>, 0 }

        Subnetwork number.  SHOULD NOT be used as a source address,
        except when the originator is one of the endpoints of a point-
        to-point link with a 31-bit mask.  For other types of links, a
        packet with such a destination SHOULD be silently discarded.
        If these packets are not silently discarded, they MUST be
  treated
        as IP broadcasts [RFC1812].

3.2. "Assigned Numbers" [RFC1700]

  Sub-section (e) of the "Special Addresses" section in the
  "Introduction" is replaced with:

     (e)   {<Network-number>, <Subnet-number>, -1}

        Directed broadcast to specified subnet.  Can only be used as a
        destination address.  However, in the case where the originator
        is one of the endpoints of a point-to-point link with a 31-bit
        mask, it can also be used as a source address.

3.3. "Requirements for IP Version 4 Routers" [RFC1812]

  Section 4.2.2.11 (d) is replaced with:

     (d) { <Network-prefix>, -1 }

        Directed Broadcast - a broadcast directed to the specified
        network prefix.  It MUST NOT be used as a source address,
        except when the originator is one of the endpoints of a point-



Retana, et al.              Standards Track                     [Page 5]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


        to-point link with a 31-bit mask.  A router MAY originate
        Network Directed Broadcast packets.  A router MAY have a
        configuration option to allow it to receive directed broadcast
        packets, however this option MUST be disabled by default, and
        thus the router MUST NOT receive Network Directed Broadcast
        packets unless specifically configured by the end user.

  The text above includes the update made by [RFC2644].

  A new section (numbered 4.2.2.11 (f)) is added:

     (f)  { <Network-number>, <Subnet-number>, 0 }

        Subnetwork number.  SHOULD NOT be used as a source address,
        except when the originator is one of the endpoints of a point-
        to-point link with a 31-bit mask.  For other types of links, a
        packet with such a destination SHOULD be silently discarded.
        If these packets are not silently discarded, they MUST be
        treated as IP broadcasts.

  Sections 4.2.3.1 (1), (2) and (4) are replaced with:

     (1) MUST treat as IP broadcasts packets addressed to
     255.255.255.255 or { <Network-prefix>, -1 }.

     In a point-to-point link with a 31-bit mask, a packet addressed to
     { <Network-prefix>, -1 } corresponds to one of the endpoints of
     such link, it MUST be treated as directed to the router on which
     the address is applied.

     (2) SHOULD silently discard on receipt (i.e., do not even deliver
     to applications in the router) any packet addressed to 0.0.0.0 or
     { <Network-prefix>, 0 }.  If these packets are not silently
     discarded, they MUST be treated as IP broadcasts (see Section
     [5.3.5]).  There MAY be a configuration option to allow receipt of
     these packets.  This option SHOULD default to discarding them.

     In a point-to-point link with a 31-bit mask, a packet addressed to
     { <Network-prefix>, 0 } corresponds to one of the endpoints of
     such link, it MUST be treated as directed to the router on which
     the address is applied.

     (4) SHOULD NOT originate datagrams addressed to 0.0.0.0 or {
     <Network-prefix>, 0 }.  There MAY be a configuration option to
     allow generation of these packets (instead of using the relevant
     1s format broadcast).  This option SHOULD default to not
     generating them.




Retana, et al.              Standards Track                     [Page 6]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


     In a point-to-point link with a 31-bit mask, the configuration of
     such a mask SHOULD allow for the generation of datagrams addressed
     to { <Network-prefix>, 0 }.

  The following text is added to section 4.3.3.9:

     The 255.255.255.255 IP broadcast address MUST be used for
     broadcast Address Mask Replies in point-to-point links with 31-bit
     subnet masks

4. Operational Experience

  The recommendations presented in this document have been implemented
  by several router vendors in beta code.  The implementation has been
  tested by at least three ISPs with positive results (i.e., no
  problems have been found).  Among the routing protocols tested
  successfully are OSPF, IS-IS, BGP and EIGRP.

  It is expected that the implementation will be officially released
  within the next few months and that other vendors will adopt it.

5. Deployment Considerations

  The intent of this document is to discuss the applicability and
  operation of 31-bit prefixes on point-to-point links.  The effects
  (if any) on other types of interfaces are not considered.  Note that
  a point-to-point link in which only one end supports the use of 31-
  bit prefixes may not operate correctly.

6. Security Considerations

  In the light of various denial of service (DoS) attacks on various
  networks within the Internet, security has become a major concern.
  The use of 31-bit subnet masks within the core of the Internet will
  reduce the number of physical links against which a DoS attack
  relying on packet replication through the use of directed broadcasts
  can be launched [RFC2644, SMURF].

  Overall, implementation of this document recommendation will improve
  the Internet's resilience to these types of DoS attacks.

7. Acknowledgements

  The authors of this document do not make any claims on the
  originality of the ideas described.  Among other people, we would
  like to acknowledge Alex Zinin for his comments, and the many people
  who have tested 31 bit subnet masks in their labs and networks.




Retana, et al.              Standards Track                     [Page 7]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


8. References

  [RFC950]  Mogul, J. and J. Postel, "Internet Standard Subnetting
            Procedure", STD 5, RFC 950, August 1985.

  [RFC1122] Braden, R., "Requirements for Internet Hosts --
            Communication Layers", STD 3, RFC 1122, October 1989.

  [RFC1332] McGregor, G., "The PPP Internet Protocol Control Protocol
            (IPCP)", RFC 1332, May 1992.

  [RFC1519] Fuller, V., Li, T., Yu, J. and K. Varadhan, "Classless
            Inter-Domain Routing (CIDR): an Address Assignment and
            Aggregation Strategy", RFC 1519, September 1993.

  [RFC1631] Egevang, K. and P. Francis, "The IP Network Address
            Translator (NAT)", RFC 1631, May 1994.

  [RFC1700] Reynolds, J. and J. Postel, "Assigned Numbers", STD 2, RFC
            1700, October 1994.

  [RFC1812] Baker, F., "Requirements for IP Version 4 Routers", RFC
            1812, June 1995.

  [RFC2050] Hubbard, K., Kosters, M., Conrad, D., Karrenberg, D. and J.
            Postel, "Internet Registry IP Allocation Guidelines", BCP
            12, RFC 2050, November 1996.

  [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6
            (IPv6) Specification", RFC 2460, December 1998.

  [RFC2644] Senie, D., "Changing the Default for Directed Broadcasts in
            Routers", BCP 34, RFC 2644, August 1999.

  [SMURF]   Huegen, C., "The Latest in Denial of Service Attacks:
            'Smurfing':  Description and Information to Minimize
            Effects", URL:
            http://users.quadrunner.com/chuegen/smurf.cgi













Retana, et al.              Standards Track                     [Page 8]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


9. Authors' Addresses

  Alvaro Retana
  Cisco Systems, Inc.
  7025 Kit Creek Rd.
  Research Triangle Park, NC 27709

  EMail: [email protected]


  Russ White
  Cisco Systems, Inc.
  7025 Kit Creek Rd.
  Research Triangle Park, NC 27709

  EMail: [email protected]


  Vince Fuller
  GTE Internetworking
  3801 E. Bayshore Rd.
  Palo Alto, CA, 94303

  EMail: [email protected]


  Danny McPherson
  Amber Networks
  2465 Augustine Drive
  Santa Clara, CA  95054

  EMail: [email protected]



















Retana, et al.              Standards Track                     [Page 9]

RFC 3021             31-Bit Prefixes on IPv4 Links         December 2000


Full Copyright Statement

  Copyright (C) The Internet Society (2000).  All Rights Reserved.

  This document and translations of it may be copied and furnished to
  others, and derivative works that comment on or otherwise explain it
  or assist in its implementation may be prepared, copied, published
  and distributed, in whole or in part, without restriction of any
  kind, provided that the above copyright notice and this paragraph are
  included on all such copies and derivative works.  However, this
  document itself may not be modified in any way, such as by removing
  the copyright notice or references to the Internet Society or other
  Internet organizations, except as needed for the purpose of
  developing Internet standards in which case the procedures for
  copyrights defined in the Internet Standards process must be
  followed, or as required to translate it into languages other than
  English.

  The limited permissions granted above are perpetual and will not be
  revoked by the Internet Society or its successors or assigns.

  This document and the information contained herein is provided on an
  "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING
  TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
  BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
  HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
  MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Acknowledgement

  Funding for the RFC Editor function is currently provided by the
  Internet Society.



















Retana, et al.              Standards Track                    [Page 10]


You are viewing proxied material from gopher.rbfh.de. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.