Generic Tunnel MTU Determination
draft-generic-v6ops-tunmtu-00

Network Working Group                                    F. Templin, Ed.
Internet-Draft                              Boeing Research & Technology
Intended status: Informational                              May 22, 2012
Expires: November 23, 2012

                    Generic Tunnel MTU Determination
                   draft-generic-v6ops-tunmtu-00.txt

Abstract

   The IPv6-over-IPv4 tunnel MTU as defined by "Basic Transition
   Mechanisms for IPv6" is currently recommended to be set to 1480 or
   less when static MTU determination is used.  This is to avoid IPv4
   fragmentation within the tunnel, but requires the tunnel ingress to
   drop any IPv6 packet larger than 1480 bytes and return an ICMPv6
   Packet Too Big (PTB) message.  Concerns for operational issues with
   both IPv4 and IPv6 Path MTU Discovery point to the possibility of
   MTU-related black holes when a packet is dropped due to an MTU
   restriction.  Fortunately, the "Internet cell size" is 1500 bytes,
   i.e., the minimum MTU configured by the vast majority of links in the
   Internet.  This document therefore presents a method to boost the
   tunnel MTU to 1500 bytes.

Status of this Memo

   This Internet-Draft is submitted in full conformance with the
   provisions of BCP 78 and BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF).  Note that other groups may also distribute
   working documents as Internet-Drafts.  The list of current Internet-
   Drafts is at http://datatracker.ietf.org/drafts/current/.

   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."

   This Internet-Draft will expire on November 23, 2012.

Copyright Notice

   Copyright (c) 2012 IETF Trust and the persons identified as the
   document authors.  All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents

Templin                 Expires November 23, 2012               [Page 1]
Internet-Draft             Generic Tunnel MTU                   May 2012

   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document.  Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.  Code Components extracted from this document must
   include Simplified BSD License text as described in Section 4.e of
   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . . . 3
   2.  Problem Statement . . . . . . . . . . . . . . . . . . . . . . . 3
   3.  Tunnel MTU and Fragmentation  . . . . . . . . . . . . . . . . . 4
   4.  Setting the Identification Field  . . . . . . . . . . . . . . . 5
   5.  Applicability . . . . . . . . . . . . . . . . . . . . . . . . . 6
   6.  IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 6
   7.  Security Considerations . . . . . . . . . . . . . . . . . . . . 6
   8.  Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . 6
   9.  References  . . . . . . . . . . . . . . . . . . . . . . . . . . 7
     9.1.  Normative References  . . . . . . . . . . . . . . . . . . . 7
     9.2.  Informative References  . . . . . . . . . . . . . . . . . . 7
   Author's Address  . . . . . . . . . . . . . . . . . . . . . . . . . 8

Templin                 Expires November 23, 2012               [Page 2]
Internet-Draft             Generic Tunnel MTU                   May 2012

1.  Introduction

   The IPv6-over-IPv4 tunnel MTU as defined by "Basic Transition
   Mechanisms for IPv6" is currently recommended to be set to 1480 or
   less when static MTU determination is used [RFC4213].  This is to
   avoid IPv4 fragmentation within the tunnel [RFC0791], but requires
   the tunnel ingress to drop any IPv6 packet larger than 1480 bytes and
   return an ICMPv6 Packet Too Big (PTB) message [RFC2460].  Concerns
   for operational issues with both IPv4 and IPv6 Path MTU Discovery
   (PMTUD) [RFC1191][RFC1981] point to the possibility of MTU-related
   black holes when a packet is dropped due to an MTU restriction.
   Fortunately, the "Internet cell size" is 1500 bytes, i.e., the
   minimum MTU configured by the vast majority of links in the Internet.
   This document therefore presents a method to boost the tunnel MTU to
   1500 bytes.

   Pushing the tunnel MTU to 1500 bytes is met with the challenge that
   the addition of the IPv4 encapsulation header would cause a 1500 byte
   IPv6 packet to appear as a 1520 byte IPv4 packet on the wire.  This
   can result in the packet being either fragmented or dropped by an
   IPv4 router that configures a 1500 byte link, depending on the
   setting of the "Don't Fragment" (DF) bit in the IPv4 header.  Using
   the approach outlined in this document, the tunnel ingress avoids
   this issue by performing IPv6 fragmentation on the inner IPv6 packet
   before outer IPv4 encapsulation.  The approach is outlined in the
   following sections.

2.  Problem Statement

   When an IPv6 tunnel configures a smaller MTU than 1500 bytes, packets
   that are small enough to traverse earlier links in the path toward
   the final destination can suddenly be dropped at the tunnel ingress
   with an ICMPv6 PTB message returned to the original source.  However,
   operational experience has shown that the PTB messages can be lost in
   the network due to filtering in which case the source does not
   receive notification of the loss.  It is therefore highly desirable
   that the tunnel configure an MTU of at least 1500 bytes.

   One possibility is to use IP fragmentation of the outer IP layer
   protocol so that inner packets up to 1500 bytes are delivered even if
   the tunnel encapsulation causes the outer packet to be larger than
   1500 bytes.  However, IPv4 fragmentation has been shown to be
   dangerous at high data rates due to the 16-bit Identification field
   wrapping while reassemblies are still active.  Also, if outer IPv4
   fragmentation were used the tunnel egress would need to do the
   reassembly which can be an onerous burden when the egress is located
   on a router.

Templin                 Expires November 23, 2012               [Page 3]
Internet-Draft             Generic Tunnel MTU                   May 2012

   A second possibility is to enable IP PMTUD on the outer packet.
   However, the PTB messages that may result could either be lost on the
   return path to the tunnel ingress or may not contain enough
   information for translation into an inner packet PTB for delivery to
   the original source.  Still another possibility is for the tunnel
   ingress to maintain state about MTU sizes for various tunnel
   egresses, but this becomes unwieldy when the number of egresses is
   large.

   In short, PMTUD for both IPv4 and IPv6 is a mess and new approaches
   are needed.  Preferably, PMTUD can be avoided through operational
   arrangements, as described in the following sections.

3.  Tunnel MTU and Fragmentation

   Section 3.2 of [RFC4213] presents both static and dynamic MTU
   determination algorithms.  These algorithms have been shown to be
   problematic in many instances, as discussed in Section 2.  This
   document therefore proposes a new tunnel MTU and fragmentation method
   via the following algorithm:

      1. set the tunnel ingress MTU to 1500
      2. for IPv6 packets to be admitted into the tunnel,
         do the following:
         a) if the packet is 1501 or more, drop it and
            send an ICMPv6 PTB with MTU size 1500
         b) if the packet is between 1281 - 1500:
            - admit the packet into the tunnel, subject to
              rate limiting (see Section 3).
            - break the packet into 2 pieces, where the first
              piece is a random length between 512-1024 bytes
            - insert a fragment header on both pieces and set
              the Identification field to a 32-bit value as
              specified in Section 4. (If the original packet
              already included a fragment header, however the
              original fragment header is used.)
            - encapsulate both pieces in an IPv4 header and
              send them to the tunnel far end.
         c) if the packet is 1280 or less:
            - admit the packet into the tunnel
            - encapsulate the packet in an IPv4 header and
              send it to the tunnel far end
     3. the IPv6 destination host gets to reassemble if
        necessary

   In the above algorithm, the tunnel MTU can be set to 1500 bytes

Templin                 Expires November 23, 2012               [Page 4]
Internet-Draft             Generic Tunnel MTU                   May 2012

   independently of other nodes, i.e., the approach is incrementally
   deployable.

4.  Setting the Identification Field

   The algorithm in Section 3 ignores the IPv6 requirement that routers
   in the network must not fragment IPv6 packets.  However, we observe
   that the 32-bit Identification field provides sufficient protection
   against accidental reassembly of fragments from different IPv6
   packets if operational considerations are observed.

   Specifically, the tunnel ingress must ensure that there will be no
   IPv6 fragments alive in the system with duplicate Identification
   values.  Since [RFC2460] specifies that the maximum time a node may
   retain an incomplete fragmented packet is 60 seconds, this means that
   the tunnel ingress must not allow the Identification values to be
   repeated within this timeframe.  The tunnel ingress can therefore
   calculate a maximum data rate for admission of fragmented packets
   into the tunnel.

   To avoid Identification value duplication, the tunnel ingress must
   admit no more than (2^32 / 60) = 71582788 IPv6 packets requiring
   fragmentation into the tunnel per second.  In the worst case,
   consider that each packet is 1281 bytes (i.e., 10248 bits) in length.
   The tunnel ingress can then calculate the maximum data rate as
   (71582788 * 10248) = 733580411424 bits/sec, or approximately 733
   Gbps.  It is therefore essential that the tunnel ingress set a rate
   limit to no more than 733 Gbps for those packets that will require
   fragmentation.  This restriction can be relaxed if the tunnel ingress
   maintains a per-egress Identification value instead of a single
   Identification value for all tunnel egresses.

   The algorithm in Section 3 requires that the tunnel ingress perform
   IPv6 fragmentation (when necessary), but the tunnel egress does not
   perform reassembly; hence, the tunnel endpoints remain stateless.
   Instead, the final destination IPv6 host may be obliged to reassemble
   IPv6 packets up to 1500 bytes in length, but hosts are already
   required to reassemble at least that much by the IPv6 standard
   [RFC2460].

   Note that a possible conflict exists when IPv6 fragmentation has
   already been performed by a source host before the fragments arrive
   at the tunnel ingress.  In that case, there is a small possibility
   that the Identification values used by the source host will
   temporarily be in close correlation with those used by the tunnel
   ingress, where a "collision" may occur in which the fragments
   produced by the source host have the same Identification value as the

Templin                 Expires November 23, 2012               [Page 5]
Internet-Draft             Generic Tunnel MTU                   May 2012

   fragments produced by the tunnel ingress.  Two factors that mitigate
   such conflicts are the random length of the first fragment used by
   the tunnel ingress (i.e., to cause a length mismatch for colliding
   reassemblies) and, in even rarer instance, the use of the TCP/UDP
   checksum.

   For that reason, the Identification values chosen for the fragmented
   tunneled packets are *not* assigned sequentially to avoid corrupting
   a block of reassemblies at a single destination host.  Instead, the
   Identification values must be spaced so that the value begins
   counting at a random variable N (between 0 - (2^32 - 1)), then (N +
   C), then (N + 2C), then (N + 3C) etc. modulo 2^32 until the count
   returns again to N. The sequence then begins at (N+1), then ((N+1) +
   C), then ((N+1) + 2C), etc. until all integer values within 2^32 are
   covered.  Using this method, large numbers of outstanding
   reassemblies can be accommodated with only very rare instances of
   packet loss.

5.  Applicability

   This approach applies to all tunneling methods that use the basic
   transition mechanisms, including configured tunnels [RFC4213], 6to4
   [RFC3056], ISATAP [RFC5214], DSMIP [RFC5555], 6rd [RFC5969], etc.

   Note that this same approach can also be applied to tunneling methods
   that use other than the basic transition mechanisms.  These can
   include Teredo [RFC4380], LISP [I-D.ietf-lisp], SEAL
   [I-D.templin-intarea-seal], and even IPv6-over-IPv6 encapsulation
   [RFC2473]

6.  IANA Considerations

   There are no IANA considerations for this document.

7.  Security Considerations

   The security considerations for basic transition mechanisms apply
   also to this document.

8.  Acknowledgments

   This method was inspired through discussion on the IETF v6ops list in
   the May 2012 timeframe.

Templin                 Expires November 23, 2012               [Page 6]
Internet-Draft             Generic Tunnel MTU                   May 2012

9.  References

9.1.  Normative References

   [RFC0791]  Postel, J., "Internet Protocol", STD 5, RFC 791,
              September 1981.

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

   [RFC4213]  Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms
              for IPv6 Hosts and Routers", RFC 4213, October 2005.

9.2.  Informative References

   [I-D.ietf-lisp]
              Farinacci, D., Fuller, V., Meyer, D., and D. Lewis,
              "Locator/ID Separation Protocol (LISP)",
              draft-ietf-lisp-23 (work in progress), May 2012.

   [I-D.templin-intarea-seal]
              Templin, F., "The Subnetwork Encapsulation and Adaptation
              Layer (SEAL)", draft-templin-intarea-seal-42 (work in
              progress), December 2011.

   [RFC1191]  Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191,
              November 1990.

   [RFC1981]  McCann, J., Deering, S., and J. Mogul, "Path MTU Discovery
              for IP version 6", RFC 1981, August 1996.

   [RFC2473]  Conta, A. and S. Deering, "Generic Packet Tunneling in
              IPv6 Specification", RFC 2473, December 1998.

   [RFC3056]  Carpenter, B. and K. Moore, "Connection of IPv6 Domains
              via IPv4 Clouds", RFC 3056, February 2001.

   [RFC4380]  Huitema, C., "Teredo: Tunneling IPv6 over UDP through
              Network Address Translations (NATs)", RFC 4380,
              February 2006.

   [RFC5214]  Templin, F., Gleeson, T., and D. Thaler, "Intra-Site
              Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214,
              March 2008.

   [RFC5555]  Soliman, H., "Mobile IPv6 Support for Dual Stack Hosts and
              Routers", RFC 5555, June 2009.

Templin                 Expires November 23, 2012               [Page 7]
Internet-Draft             Generic Tunnel MTU                   May 2012

   [RFC5969]  Townsley, W. and O. Troan, "IPv6 Rapid Deployment on IPv4
              Infrastructures (6rd) -- Protocol Specification",
              RFC 5969, August 2010.

Author's Address

   Fred L. Templin (editor)
   Boeing Research & Technology
   P.O. Box 3707
   Seattle, WA  98124
   USA

   Email: fltemplin@acm.org

Templin                 Expires November 23, 2012               [Page 8]