Things Multihoming in IPv6 (MULTI6) Developers Should Think About
draft-ietf-multi6-things-to-think-about-01

MULTI6 Working Group                                             E. Lear
Internet-Draft                                             Cisco Systems
Expires: July 7, 2005                                    January 6, 2005

               Things MULTI6 Developers should think about
                draft-ietf-multi6-things-to-think-about-01

Status of this Memo

    This document is an Internet-Draft and is subject to all provisions
    of section 3 of RFC 3667.  By submitting this Internet-Draft, each
    author represents that any applicable patent or other IPR claims of
    which he or she is aware have been or will be disclosed, and any of
    which he or she become aware will be disclosed, in accordance with
    RFC 3668.

    Internet-Drafts are working documents of the Internet Engineering
    Task Force (IETF), its areas, and its working groups.  Note that
    other groups may also distribute working documents as
    Internet-Drafts.

    Internet-Drafts are draft documents valid for a maximum of six months
    and may be updated, replaced, or obsoleted by other documents at any
    time.  It is inappropriate to use Internet-Drafts as reference
    material or to cite them other than as "work in progress."

    The list of current Internet-Drafts can be accessed at
    http://www.ietf.org/ietf/1id-abstracts.txt.

    The list of Internet-Draft Shadow Directories can be accessed at
    http://www.ietf.org/shadow.html.

    This Internet-Draft will expire on July 7, 2005.

Copyright Notice

    Copyright (C) The Internet Society (2005).

Abstract

    This document specifies a set of questions that authors should be
    prepared to answer as part of a solution to multihoming with IPv6.
    The questions do not assume that multihoming is the only problem of
    interest, nor do they demand a more general solution either.

Lear                      Expires July 7, 2005                  [Page 1]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

Table of Contents

    1.   Introduction . . . . . . . . . . . . . . . . . . . . . . . .   4
      1.1  Reading this document  . . . . . . . . . . . . . . . . . .   4
    2.   On the wire behavior . . . . . . . . . . . . . . . . . . . .   5
      2.1  How will your solution solve the multihoming problem?  . .   5
      2.2  At what layer is your solution applied, and how? . . . . .   5
      2.3  Why is the layer you chose the correct one?  . . . . . . .   5
      2.4  Does your solution address mobility? . . . . . . . . . . .   5
      2.5  Does your solution expand the size of an IP packet?  . . .   5
      2.6  Will your solution add additional latency? . . . . . . . .   5
      2.7  Can multihoming capabilities be negotiated end to end
           during a connection? . . . . . . . . . . . . . . . . . . .   5
      2.8  Do you change the way fragmenting is handled?  . . . . . .   5
      2.9  Are there any layer 2 implications to your proposal? . . .   6
    3.   Identifiers and locators . . . . . . . . . . . . . . . . . .   7
      3.1  Uniqueness . . . . . . . . . . . . . . . . . . . . . . . .   7
      3.2  Does your solution provide for a split between
           identifiers and locators?  . . . . . . . . . . . . . . . .   7
      3.3  What is the lifetime of a binding from an identifier to
           a locator? . . . . . . . . . . . . . . . . . . . . . . . .   7
      3.4  How is the binding updated?  . . . . . . . . . . . . . . .   7
      3.5  How does a host know its identity? . . . . . . . . . . . .   7
      3.6  Can a host have multiple identifiers?  . . . . . . . . . .   7
      3.7  If you have separate locators and identifiers, how will
           they be mapped?  . . . . . . . . . . . . . . . . . . . . .   7
      3.8  Does your solution create an alternate "DNS-like"
           service? . . . . . . . . . . . . . . . . . . . . . . . . .   7
      3.9  Please describe authentication/authorization . . . . . . .   7
      3.10   Is your mechanism hierarchical?  . . . . . . . . . . . .   7
      3.11   Middlebox interactions . . . . . . . . . . . . . . . . .   7
      3.12   Are there any implications for scoped addressing?  . . .   8
    4.   Routing system interactions  . . . . . . . . . . . . . . . .   9
      4.1  Does your solution change existing aggregation methods?  .   9
      4.2  Does the solution solve traffic engineering
           requirements?  . . . . . . . . . . . . . . . . . . . . . .   9
      4.3  Does the solution offer ways for site the to manage its
           traffic flows? . . . . . . . . . . . . . . . . . . . . . .   9
      4.4  If you introduce any new name spaces, do they require
           aggregation? . . . . . . . . . . . . . . . . . . . . . . .   9
      4.5  Does your solution interact with Autonomous System
           numbering? . . . . . . . . . . . . . . . . . . . . . . . .   9
      4.6  Are there any changes to ICMP error semantics? . . . . . .   9
    5.   Name service interactions  . . . . . . . . . . . . . . . . .  10
      5.1  Please explain the relationship of your solution to DNS  .  10
      5.2  Please explain interactions with "2-faced" DNS . . . . . .  10
      5.3  Does your solution require centralized registration? . . .  10
      5.4  Have you checked for DNS circular dependencies?  . . . . .  10

Lear                      Expires July 7, 2005                  [Page 2]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

      5.5  What if a DNS server itself is multihomed? . . . . . . . .  10
      5.6  What additional load will be placed on DNS servers?  . . .  10
      5.7  Any upstream provider support required?  . . . . . . . . .  10
      5.8  How do you debug connectivity? . . . . . . . . . . . . . .  11
    6.   Application concerns and backward compatibility  . . . . . .  12
      6.1  What application/API changes are needed? . . . . . . . . .  12
      6.2  Is this solution backward compatible with "old" IP
           version 6? . . . . . . . . . . . . . . . . . . . . . . . .  12
      6.3  Is your solution backward compatible with IPv4?  . . . . .  12
      6.4  Can IPv4 devices take advantage of this solution?  . . . .  12
      6.5  What is the impact of your solution on different types
           of sites?  . . . . . . . . . . . . . . . . . . . . . . . .  13
      6.6  How will your solution interact with other middleboxes?  .  13
      6.7  Referrals  . . . . . . . . . . . . . . . . . . . . . . . .  13
      6.8  Demonstrate use with a real life complex application . . .  13
    7.   Legal concerns . . . . . . . . . . . . . . . . . . . . . . .  14
    8.   Security Considerations  . . . . . . . . . . . . . . . . . .  15
    9.   Acknowledgments  . . . . . . . . . . . . . . . . . . . . . .  16
    10.  References . . . . . . . . . . . . . . . . . . . . . . . . .  17
    10.1   Normative References . . . . . . . . . . . . . . . . . . .  17
    10.2   Informative References . . . . . . . . . . . . . . . . . .  17
         Author's Address . . . . . . . . . . . . . . . . . . . . . .  17
    A.   Differences between pre-WG version and this version  . . . .  18
         Intellectual Property and Copyright Statements . . . . . . .  19

Lear                      Expires July 7, 2005                  [Page 3]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

1.  Introduction

    At the time of this writing there are quite a number of proposed
    solutions to the problem of multihoming within IPv6, and related
    problems such as the locator/identifier split.

    This document contains several set of questions that attempt to focus
    these solutions on operational problems.  This document does not
    suggest methods to solve the problem.  Rather, we simply want to
    ensure that while solving a problem the medicine is not worse than
    the cure.  We focus on practical operational problems that both
    single-homed and multihomed deployments may face.

    It is the hope of the author perhaps others that the authors of the
    proposed solutions will use this document to identify gaps in their
    solutions, and cooperate to close those gaps.

1.1  Reading this document

    The questions are organized along the following lines:
    o  changes to on the wire behavior;
    o  routing system interactions;
    o  identifier/mapping split;
    o  application concerns and backward compatibility;
    o  name service interactions;
    o  legal concerns; and
    o  security considerations.

    In reality many questions cut across all of these concerns.  For
    instance, the identifier / locator split has substantial application
    implications, and every area has security considerations.

    Unless it is blatantly obvious, each question contains some reasoning
    as to why it is being asked.  It is envisioned that no solution will
    answer every question with completeness, but that there will be
    tradeoffs to be made.  The answers by the various designers of
    solutions will hopefully shed some light on which tradeoffs we as a
    community wish to make.

    It would seem silly for people who have written out detailed answers
    to these questions to have to repeat the exercise.  Therefore, a
    simple reference to existing documents will suffice, so long as the
    answer is complete.  If it is not complete, then feel free to
    reference it and add what text is necessary to make the answer
    complete.

    This document presumes a familiarity with RFC 3582 [2], and does not
    attempt to repeat the requirements work gathered there.

Lear                      Expires July 7, 2005                  [Page 4]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

2.  On the wire behavior

2.1  How will your solution solve the multihoming problem?

    Please scope the problem you are attempting to solve and what you are
    not attempting to solve.

2.2  At what layer is your solution applied, and how?

    Is it applied in every packet?  If so, what fields are used?

2.3  Why is the layer you chose the correct one?

    Each layer has its benefits and tradeoffs.  For instance, transport
    layer solutions would require that EVERY transport be modified, while
    IP layer solutions may entail expansion of the packet or a change to
    the pseudo-header (thus requiring changes to the transport layer).

2.4  Does your solution address mobility?

    If so, how are rendezvous handled?  Can your solution handle both
    locators changing at the same time?  If so, please explain.  Should
    it?  If not, how will your solution interact with MOBILEIP-V6 [3]
    (MIPv6)

2.5  Does your solution expand the size of an IP packet?

    Expanding the size of an IP packet may cause excessive fragmentation
    in some circumstances.

2.6  Will your solution add additional latency?

    Latency is an important factor in many applications, including voice.
    Any substantial amount of additional latency, including session
    initiation would be highly undesirable.

2.7  Can multihoming capabilities be negotiated end to end during a
     connection?

    If the proposal introduces additional overhead, can the information
    be somehow piggybacked on messages that are already used?  This would
    be useful in order to keep connection setup constant.  Please also
    indicate any drawbacks that might apply due to this piggybacking.

2.8  Do you change the way fragmenting is handled?

    If you use a shim approach, do you fragment above or below the shim?
    How are fragments identified, so that they can be reassembled?  If

Lear                      Expires July 7, 2005                  [Page 5]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

    you use any additional names, do they need to be associated with
    fragments?  If not, why not?  If so, how will that happen?

2.9  Are there any layer 2 implications to your proposal?

    While Ipv6 has a simplified approach to layer 2, perhaps you
    unsimplified it.  If so, please provide details.

Lear                      Expires July 7, 2005                  [Page 6]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

3.  Identifiers and locators

3.1  Uniqueness

3.2  Does your solution provide for a split between identifiers and
     locators?

3.3  What is the lifetime of a binding from an identifier to a locator?

3.4  How is the binding updated?

    Will transport connections remain up when new paths become available
    or when old ones become unavailable?  How does the end node discover
    these events?

3.5  How does a host know its identity?

    If you are establishing a new identity, how does the host learn it?

3.6  Can a host have multiple identifiers?

    If so, how does an application choose an identity?

3.7  If you have separate locators and identifiers, how will they be
     mapped?

    Does the mapping work in both directions?  How would someone
    debugging a network determine which end stations are involved?

3.8  Does your solution create an alternate "DNS-like" service?

    If you use mechanisms other than DNS, first, why was DNS not
    appropriate?  Also, how will this other mechanism interact with DNS?
    What are its scaling properties?

3.9  Please describe authentication/authorization

    How are bindings authenticated and authorized.  What technology do
    you build on for this mechanism?

3.10  Is your mechanism hierarchical?

    Please describe the hierarchical breakdown.

3.11  Middlebox interactions

    What are the implications for firewalls?  What are the interactions
    with NAT?  What are the interactions with web caches? What

Lear                      Expires July 7, 2005                  [Page 7]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

    complications are introduced with your solution?  For instance, are
    there implication for ingress filters?  If so, what are they?

    When considering this question there are really two issues.  First,
    will middleboxes impede your solution by rewriting headers in some
    way, as NATs do for IP addresses, and web caches do at higher layers?
    Second, is there a way in which middleboxes are actually part of your
    solution?  In particular, are they required? This would be the case,
    for example, with GSE (8+8).

3.12  Are there any implications for scoped addressing?

    Please see RFC 3513 [1].  How does your mechanism interact with
    multicast?

    How does your solution interact with link-local addressing

    How does your solution interact with Son-Of-Sitelocal (whatever that
    will be)?

Lear                      Expires July 7, 2005                  [Page 8]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

4.  Routing system interactions

4.1  Does your solution change existing aggregation methods?

    Routing on the Internet scales today because hosts and networks can
    be aggregated into a relatively small number of entries.  Does your
    solution change the way in which route aggregation occurs?

4.2  Does the solution solve traffic engineering requirements?

    One of the significant goals of IPv4 multihoming solutions has been
    to be able to perform traffic engineering based on appropriately
    adjusting the BGP advertisements.  If the prefixes used by such sites
    was be aggregated (particularly beyond the site"s border), the site"s
    ability to perform traffic engineering would be diminished.

4.3  Does the solution offer ways for site the to manage its traffic
     flows?

    If so, how?  Is this controllable on a per-host basis, or on a
    per-site basis?

4.4  If you introduce any new name spaces, do they require aggregation?

    Is it desirable or required that in order to scale distribution of
    any mapping information an aggregation method be introduced?

4.5  Does your solution interact with Autonomous System numbering?

    If your solution involves address prefixes distributed using BGP4+,
    does it interact with the use of AS numbers and if so how? Will it
    require additional AS numbers?

4.6  Are there any changes to ICMP error semantics?

    Do you create new codes?  If so, why and what do they mean? Will a
    host that is not aware of your scheme see them?

Lear                      Expires July 7, 2005                  [Page 9]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

5.  Name service interactions

5.1  Please explain the relationship of your solution to DNS

    If your solution uses new names for identifiers, please explain what
    mappings are defined, and how they are performed?

    If there are any additional administrative requirements, such as new
    zones or RR types to manage, please explain them as well.

5.2  Please explain interactions with "2-faced" DNS

    2-faced DNS is used so that hosts behind a NAT get one address for
    internal hosts, while hosts outside the NAT get another.  Similar
    mechanisms are used for application layer gateways, such as SOCKS
    [5].

5.3  Does your solution require centralized registration?

    For instance, if you are using the DNS, what will be the top level
    domain, and how will the name space distribute through it?

    Also, how will the centralized registration be managed?

5.4  Have you checked for DNS circular dependencies?

    If you are using the DNS in your solution, is it required for
    connectivity?  What happens if the DNS fails?  Can communication
    between the DNS resolver and the server make use of your solution?
    What about between the application and the resolver?

5.5  What if a DNS server itself is multihomed?

    If a link fails or a service is dropped, how will it impact DNS?
    Again are there any dependency loops?  Perhaps diagram out your
    dependencies to make sure.

5.6  What additional load will be placed on DNS servers?

    Can the load be distributed?  Remember that DNS is optimized for READ
    operations.

5.7  Any upstream provider support required?

    If so, please describe.  For instance, today reverse mappings are
    delegated down from upstream providers.  How would this work with
    your solution?

Lear                      Expires July 7, 2005                 [Page 10]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

5.8  How do you debug connectivity?

    How would tools like ping and traceroute need to be enhanced? What
    additional tools would prove useful or necessary?  For instance, if
    there is an id/locator split, can one ping an identifier?  If so,
    what gets returned?

Lear                      Expires July 7, 2005                 [Page 11]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

6.  Application concerns and backward compatibility

6.1  What application/API changes are needed?

    Will old code just work with the new mechanism? For instance, what
    about code that uses gethostbyname()?

    Will getaddrinfo() need to change?

    What about other API calls?

    There are several possible approaches.  For instance, a multihoming
    service could attempt to require no changes to the API, in which case
    it is possible that IP addresses might become opaque blobs that work
    with the API, but might break operational assumptions that
    applications make about addresses.  Consider the case of a web server
    that wants to log IP addresses.  How will it accomplish this task?

    Another approach is to have some sort of compatibility library for
    legacy applications, but also provide a richer calling interface for
    transparency.

    Yet another approach would be to only provide the new functionality
    to those applications that make use of a new calling interface.

    One useful exercise would be to provide code fragments that
    demonstrate any API changes.

6.2  Is this solution backward compatible with "old" IP version 6?

    Can it be deployed incrementally?  Please describe how.

    Does your solution impose requirements on non-multihomed/non-mobile
    hosts?

    What happens if someone plugs in a normal IPv6 node?

6.3  Is your solution backward compatible with IPv4?

    How will your mechanism interact with 6to4 gateways and IPv4 hosts?

6.4  Can IPv4 devices take advantage of this solution?

    Can the same mechanism somehow be used on the existing network?  N.B.
    this is NOT a primary consideration, but perhaps a side benefit of a
    particular solution.

Lear                      Expires July 7, 2005                 [Page 12]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

6.5  What is the impact of your solution on different types of sites?

    What will the impact of your solution be on the following types of
    systems?
       single homed sites
       small multihomed sites
       large multihomed sites
       ad-hoc sites
       short lived connections (think aggregator wireless ISPs)

    In particular, consider ongoing administration, renumbering events,
    and mobile work forces.

6.6  How will your solution interact with other middleboxes?

6.7  Referrals

    How will your solution handle referrals, such as those within FTP or
    various conferencing or other peer to peer systems?

    Referrals exist within various other protocols, such as so-called
    "peer to peer" applications.  Note that referrals might suffer three
    types of failure:
       firewall and NAT - just as FTP active mode experiences today with
       relatively simple firewalls?
       time-based - is there something ephemeral about the nature of the
       solution that might cause a referral (such as a URL) to fail over
       time, more so than what we have today?
       location-based - if the binding varies based on where the parties
       are in the network, if one moves will they no longer be able to
       find one another?

6.8  Demonstrate use with a real life complex application

    Provide a detailed walk-through of SIP+RTSP when one or several of
    the peers are multihomed.  How does your analysis change when
    encrypted rtsp is used or when SIP with S/MIME e2e signalling is
    used?

Lear                      Expires July 7, 2005                 [Page 13]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

7.  Legal concerns

    Are you introducing a namespace that might involve mnemonics? Doing
    so might introduce trademark concerns.  If so, how do you plan to
    address such concerns?

    Are there any organizations required to manage a new name space? If
    so, please describe what they are and how the method will scale.

Lear                      Expires July 7, 2005                 [Page 14]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

8.  Security Considerations

    How secure should a multi6 solution be?  This is a reasonable
    question for each solution to answer.  The author opines that the
    worst case should be no worse than what we have today.  For example,
    would a multi6 solution open up a host on either end of a
    communication to a time-based attack?  Any such risks should be
    clearly stated by the authors.  Considerable time should be spent on
    threat analysis.  Please see [4] for more details.

    As IP addresses can often be tied to individuals, are there any
    auditing or privacy concerns introduced by your solution?

Lear                      Expires July 7, 2005                 [Page 15]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

9.  Acknowledgments

    The author wishes to acknowledge everyone in the multi6 group and
    elsewhere that is putting forward proposals.  It is easy to ask
    questions like the ones found in this draft.  It is quite a bit
    harder to develop running code to answer them.  Marcelo Bagnulo, Kurt
    Erik Lindqvist, Joe Touch, Patrik Faltstrom, Brian Carpenter, and
    Iljitsch van Beijnum provided input to this document.

Lear                      Expires July 7, 2005                 [Page 16]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

10.  References

10.1  Normative References

    [1]  Hinden, R. and S. Deering, "Internet Protocol Version 6 (IPv6)
         Addressing Architecture", RFC 3513, April 2003.

    [2]  Abley, J., Black, B. and V. Gill, "Goals for IPv6
         Site-Multihoming Architectures", RFC 3582, August 2003.

    [3]  Johnson, D., Perkins, C. and J. Arkko, "Mobility Support in
         IPv6", draft-ietf-mobileip-ipv6-24 (work in progress), July
         2003.

    [4]  Nordmark, E. and T. Li, "Threats relating to IPv6 multihoming
         solutions", draft-nordmark-multi6-threats-02 (work in progress),
         June 2004.

10.2  Informative References

    [5]  Kitamura, H., "A SOCKS-based IPv6/IPv4 Gateway Mechanism", RFC
         3089, April 2001.

Author's Address

    Eliot Lear
    Cisco Systems
    170 W. Tasman Dr.
    San Jose, CA  95134-1706
    US

    EMail: lear@cisco.com

Lear                      Expires July 7, 2005                 [Page 17]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

Appendix A.  Differences between pre-WG version and this version

    This version has been substantially reorganized.  The questions from
    previous versions are largely the same.  We've gone into just a bit
    more depth on the application end.  The WG has indicated that it
    would like to continue along that path.  Please send your comments
    and questions to the working group in this regard.

    This version has also been spell checked.  Finally.

Lear                      Expires July 7, 2005                 [Page 18]

Internet-Draft       MULTI6 Solution Questionnaire          January 2005

Intellectual Property Statement

    The IETF takes no position regarding the validity or scope of any
    Intellectual Property Rights or other rights that might be claimed to
    pertain to the implementation or use of the technology described in
    this document or the extent to which any license under such rights
    might or might not be available; nor does it represent that it has
    made any independent effort to identify any such rights.  Information
    on the procedures with respect to rights in RFC documents can be
    found in BCP 78 and BCP 79.

    Copies of IPR disclosures made to the IETF Secretariat and any
    assurances of licenses to be made available, or the result of an
    attempt made to obtain a general license or permission for the use of
    such proprietary rights by implementers or users of this
    specification can be obtained from the IETF on-line IPR repository at
    http://www.ietf.org/ipr.

    The IETF invites any interested party to bring to its attention any
    copyrights, patents or patent applications, or other proprietary
    rights that may cover technology that may be required to implement
    this standard.  Please address the information to the IETF at
    ietf-ipr@ietf.org.

Disclaimer of Validity

    This document and the information contained herein are provided on an
    "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
    OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET
    ENGINEERING TASK FORCE DISCLAIM 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.

Copyright Statement

    Copyright (C) The Internet Society (2005).  This document is subject
    to the rights, licenses and restrictions contained in BCP 78, and
    except as set forth therein, the authors retain all their rights.

Acknowledgment

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

Lear                      Expires July 7, 2005                 [Page 19]