Network Working Group                                           R. Asati
Internet-Draft                                                  H. Singh
Updates: 4862 (if approved)                                    W. Beebee
Intended status: Standards Track                     Cisco Systems, Inc.
Expires: October 8, 2012 March 10, 2013                                          E. Dart
                                              Lawrence Berkeley National
                                                               W. George
                                                       Time Warner Cable
                                                            C. Pignataro
                                                     Cisco Systems, Inc.
                                                       September 6, 2012

                  Enhanced Duplicate Address Detection


   Appendix A of the IPv6 Duplicate Address Detection (DAD) document in
   RFC 4862 discusses Loopback Suppression and DAD.  However, RFC 4862
   does not settle on one specific automated means to detect loopback of
   Neighbor Discovery (ND of RFC 4861) messages used by DAD.  Several
   service provider communities have expressed a need for automated
   detection of looped backed ND messages used by DAD.  This document
   includes mitigation techniques and then outlines the Enhanced DAD
   algorithm to automate detection of looped back IPv6 ND messages used
   by DAD.  For network loopback tests, the Enhanced DAD algorithm
   allows IPv6 to self-heal after a loopback is placed and removed.
   Further, for certain access networks the document automates resolving
   a specific duplicate address conflict.

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

   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 October 8, 2012. March 10, 2013.

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
   ( 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.  Terminology  . . . . . . . . . . . . . . . . . . . . . . . . .  3
   2.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  3
   3.  Operational Mitigation Options . . . . . . . . . . . . . . . .  4
     3.1.  Disable DAD on Interface . . . . . . . . . . . . . . . . .  4
     3.2.  Dynamic Disable/Enable of DAD Using Layer 2 Protocol . . .  5
     3.3.  Operational Considerations . . . . . . . . . . . . . . . .  5
   4.  The Enhanced DAD Algorithm . . . . . . . . . . . . . . . . . .  6
     4.1.  General Rules  . . . . . . . . . . . . . . . . . . . . . .  7
     4.2.  Processing Rules for Senders . . . . . . . . . . . . . . .  7
     4.3.  Processing Rules for Receivers . . . . . . . . . . . . . .  7
     4.4.  Impact on SEND . . . . . . . . . . . . . . . . . . . . . .  7
     4.5.  Changes to RFC 4862  . . . . . . . . . . . . . . . . . . .  8
   5.  Actions to Perform on Detecting a Genuine Duplicate  . . . . .  8
   6.  Security Considerations  . . . . . . . . . . . . . . . . . . .  9
   7.  IANA Considerations  . . . . . . . . . . . . . . . . . . . . .  9
   8.  Acknowledgements . . . . . . . . . . . . . . . . . . . . . . .  9
   9.  Normative References . . . . . . . . . . . . . . . . . . . . .  9
   Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 10

1.  Terminology

   o  DAD-failed state - Duplication Address Detection failure as
      specified in [RFC4862].  Failure also includes if the Target
      Address is optimistic.  Optimistic DAD is specified in [RFC4429].

   o  Looped back message - also referred to as a reflected message.
      The message sent by the sender is received by the sender due to
      the network or a Upper Layer Protocol on the sender looping the
      message back.

   o  Loopback - A function in which the router's interface (or the
      circuit to which the router's interface is connected) is looped
      back or connected to itself.  Loopback causes packets sent by the
      interface to be received by the interface, and results in
      interface unavailability for regular data traffic forwarding.  See
      more details in section 9.1 of [RFC1247].  The Loopback function
      is commonly used in an interface context to gain information on
      the quality of the interface, by employing mechanisms such as
      ICMPv6 pings, bit-error tests, etc.  In a circuit context, it is
      used in wide area environments including optical dense wave
      division multiplexing (DWDM) and SONET/SDH for fault isolation
      (e.g. by placing a loopback at different geographic locations
      along the path of a wide area circuit to help locate a circuit
      fault).  The Loopback function may be employed locally or

   o  NS(DAD) - shorthand notation to denote an NS with unspecified IPv6
      source-address issued during DAD.

2.  Introduction

   Appendix A of [RFC4862] discusses Loopback Suppression and Duplicate
   Address Detection (DAD).  However, [RFC4862] does not settle on one
   specific automated means to detect loopback of ND messages used by
   DAD.  One specific DAD message is a Neighbor Solicitation (NS),
   specified in [RFC4861].  The NS is issued by the network interface of
   an IPv6 node for DAD.  Another message involved in DAD is a Neighbor
   Advertisement (NA).  The Enhanced DAD algorithm proposed in this
   document focuses on detecting an NS looped back to the transmitting
   interface during the DAD operation.  Detecting a looped back NA is of
   no use because no problems with DAD will occur if a node receives a
   looped back NA.  Detecting of any other looped back ND messages
   outside of the DAD operation is not critical and thus this document
   does not cover such detection.  The document also includes a
   Mitigation section that discusses means already available to mitigate
   the loopback problem.

   Recently service providers have reported a DAD loopback problem.
   Loopback testing is underway on a circuit connected to an interface
   on a router.  The interface on the router is enabled for IPv6.  The
   interface issues a NS for the IPv6 link-local address DAD.  The NS is
   reflected back to the router interface due to the loopback condition
   of the circuit, and the router interface enters a DAD-failed state.
   In contrast to IPv4, IPv6 will not return to operation on the
   interface when the loopback condition is cleared without manual

   There are other conditions which will also trigger similar problems
   with DAD Loopback.  While the following example is not a common
   configuration, it has occurred in a large service provider network.
   It is necessary to address it in the proposed solution because the
   trigger scenario has the potential to cause significant IPv6 service
   outages when it does occur.  Two broadband modems in the same
   location are served by the same service provider and both modems are
   served by one access concentrator and one layer 3 interface on the
   access concentrator.  The two modems have the Ethernet ports of each
   modem connected to a network hub.  The access concentrator serving
   the modems is the first-hop IPv6 router for the modems.  The access
   concentrator also supports proxying of DAD messages.  Each modem is
   enabled for at least data services.  The network interface of the
   access concentrator serving the two broadband modems is enabled for
   IPv6 and the interface issues a NS(DAD) message for the IPv6 link-
   local address.  The NS message reaches one modem first and this modem
   sends the message to the network hub which sends the message to the
   second modem which forwards the message back to the access
   concentrator.  The looped back NS message causes the network
   interface on the access concentrator to be in a DAD-failed state.
   Such a network interface typically serves up to 100 thousand
   broadband modems causing all the modems (and hosts behind the modems)
   to fail to get IPv6 online on the access network.  Additionally, it
   may be tedious for the access concentrator to find out which of the
   six thousand or more homes looped back the DAD message.  Clearly
   there is a need for automated detection of looped back NS messages
   during DAD operations by a node.

3.  Operational Mitigation Options

   Two mitigation options are described below.  The mechanisms do not
   require any change to existing implementations.

3.1.  Disable DAD on Interface

   One can disable DAD on an interface and then there is no NS(DAD)
   issued to be looped back.  DAD is disabled by setting the interface's
   DupAddrDetectTransmits variable to zero.  While this mitigation may
   be the simplest the mitigation has three drawbacks.

   It would likely require careful analysis of configuration on such
   point-to-point interfaces, a one-time manual configuration on each of
   such interfaces, and more importantly, genuine duplicates in the link
   will not be detected.

   A Service Provider router such as an access concentrator or network operator MAY use
   core router SHOULD support this mitigation. mitigation strategy.

3.2.  Dynamic Disable/Enable of DAD Using Layer 2 Protocol

   It is possible that one or more layer 2 protocols include provisions
   to detect the existence of a loopback on an interface circuit,
   usually by comparing protocol data sent and received.  For example,
   PPP uses magic number (section 6.4 of [RFC1661]) to detect a loopback
   on an interface.

   When a layer 2 protocol detects that a loopback is present on an
   interface circuit, the device MUST temporarily disable DAD on the
   interface, and when the protocol detects that a loopback is no longer
   present (or the interface state has changed), the device MUST
   (re-)enable DAD on that interface.

   This solution requires no protocol changes.  This solution SHOULD be
   enabled by default, and MUST be a configurable option.

   This mitigation has several benefits.  They are

   1.  It leverages layer 2 protocol's built-in loopback detection
       capability, if available.

   2.  It scales better (since it relies on an event-driven), requires
       no additional state, timer etc.  This may be a significant
       scaling consideration on devices with hundreds or thousands of
       interfaces that may be in loopback for long periods of time (such
       as while awaiting turn-up or during long-duration intrusive bit
       error rate tests).

3.3.  Operational Considerations

   The mitigation options discussed in the document do not require the
   devices on both ends of the circuit to support the mitigation
   functionality simultaneously, and do not propose any capability
   negotiation.  Suffice to say that the mitigation options are well
   effective for the unidirectional loopback.

   The mitigation options may not be effective for the bidirectional
   loopback (i.e. the loopback is placed in both directions of the
   circuit interface, so as to identify the faulty segment) if only one
   device followed a mitigation option specified in this document, since
   the other device would follow current behavior and disable IPv6 on
   that interface due to DAD until manual intervention restores it.

   This is nothing different from what happens today (without the
   solutions proposed by this document) in case of unidirectional
   loopback.  Hence, it is expected that an operator would resort to
   manual intervention for the devices not compliant with this document,
   as usual.

4.  The Enhanced DAD Algorithm

   The Enhanced DAD algorithm covers detection of a looped back NS(DAD)
   message.  The document proposes use of the Nonce Option specified in
   the SEND document of [RFC3971].  The nonce is a random number as
   specified in [RFC3971].  If SEND is enabled on the router and the
   router also supports the Enhanced DAD algorithm (specified in this
   document), there is integration with the Enhanced DAD algorithm and
   SEND.  See more details in the Impact on SEND section.

   When the IPv6 network interface issues a NS(DAD) message, the
   interface includes the Nonce Option in the NS(DAD) message and saves
   the nonce in local store.  Subsequently if the interface receives an
   identical NS(DAD) message, the interface logs a system management
   message, updates any statistics counter, and drops the looped back
   NS(DAD).  If the DupAddrDetectTransmits variable for the interface is
   greater than one, subsequent NS(DAD) messages for the same Target
   Address should be suppressed.  If the interface receives a NS(DAD)
   message with a different nonce but TargetAddress matches a tentative
   or optimistic address on the interface, the interface logs a DAD-
   failed system management message, updates any statistics, and behaves
   identical to the behavior specified in [RFC4862] for DAD failure.

   Six bytes of random nonce is sufficiently large for nonce collisions.
   However if there is a collision because two nodes generated the same
   random nonce (that are using the same Target address in their
   NS(DAD)), then the algorithm will incorrectly detect a looped back
   NS(DAD) when the NS(DAD) was issued to signal a genuine duplicate.
   Since each looped back NS(DAD) event is logged to system management,
   the administrator of the network will have to intervene manually.

   The algorithm is capable of detecting any ND solicitation (NS and
   Router Solicitation) or advertisement (NA and Router Advertisement)
   that is looped back.  However, saving a nonce and nonce related data
   for all ND messages has impact on memory of the node and also adds
   the algorithm state to a substantially larger number of ND messages.
   Therefore this document does not recommend using the algorithm
   outside of the DAD operation by an interface on a node.

4.1.  General Rules

   If an IPv6 node implements the Enhanced DAD algorithm, the node MUST
   implement detection of looped back NS(DAD) messages during DAD for an
   interface address.

4.2.  Processing Rules for Senders

   If a node has been configured to use the Enhanced DAD algorithm, when
   sending a NS(DAD) for a tentative or optimistic interface address the
   sender MUST generate a random nonce associated with the interface
   address, MUST save the nonce, and MUST include the nonce in the Nonce
   Option included in the NS(DAD).  If a looped back NS(DAD) is detected
   by the interface, and if the DupAddrDetectTransmits variable for the
   interface is greater than one, subsequent NS(DAD) messages for the
   same Target Address SHOULD be suppressed.

4.3.  Processing Rules for Receivers

   If the node has been configured to use the Enhanced DAD algorithm and
   an interface on the node receives any NS(DAD) message that matches
   the interface address (in tentative or optimistic state), the
   receiver compares the nonce in the message with the saved nonce.  If
   a match is found, the node SHOULD log a system management message,
   SHOULD update any statistics counter, and MUST drop the received
   message.  If the received NS(DAD) message includes a nonce and no
   match is found with the saved nonce, the node SHOULD log a system
   management message for DAD-failed and SHOULD update any statistics

4.4.  Impact on SEND

   The SEND document uses the Nonce Option in the context of matching an
   NA with an NS.  However, no text in SEND has an explicit mention of
   detecting looped back ND messages.  If this document updates
   [RFC4862], SEND should be updated to integrate with the Enhanced DAD
   algorithm.  A minor update to SEND would be to explicitly mention
   that the nonce in SEND is also used by SEND to detect looped back NS
   messages during DAD operations by the node.  In a mixed SEND
   environment with SEND and unsecured nodes, the lengths of the nonce
   used by SEND and unsecured nodes MUST be identical.

4.5.  Changes to RFC 4862

   The following text is added to [RFC4862].

   A network interface of an IPv6 node SHOULD implement the Enhanced DAD
   algorithm.  For example, if the interface on an IPv6 node is
   connected to a circuit that supports loopback testing, then the node
   should implement the Enhanced DAD algorithm that allows the IPv6
   interface to self-heal after loopback testing is ended on the
   circuit.  Another example is when the IPv6 interface resides on an
   access concentrator running DAD Proxy.  The interface supports up to
   100 thousand IPv6 clients (broadband modems) connected to the
   interface.  If the interface performs DAD for its IPv6 link-local
   address and if the DAD probe is reflected back to the interface, the
   interface is stuck in DAD failed state and IPv6 services to the 100
   thousand clients is denied.  Disabling DAD for such an IPv6 interface
   on an access concentrator is not an option because the network also
   needs to detect genuine duplicates in the interface downstream
   network.  The Enhanced DAD algorithm also facilitates detecting a
   genuine duplicate for the interface on the access concentrator.  See
   the Actions to Perform on Detecting a Genuine Duplicate section of
   the Enhanced DAD document.

5.  Actions to Perform on Detecting a Genuine Duplicate

   As described in paragraphs above the nonce can also serve to detect
   genuine duplicates even when the network has potential for looping
   back ND messages.  When a genuine duplicate is detected, the node
   follows the manual intervention specified in section 5.4.5 of
   [RFC4862].  However, in certain networks such as an access network if
   the genuine duplicate matches the tentative or optimistic IPv6
   address of a network interface of the access concentrator, automated
   actions are proposed.

   One access network is a cable broadband deployment where the access
   concentrator is the first-hop IPv6 router to several thousand
   broadband modems.  The router also supports proxying of DAD messages.
   The network interface on the access concentrator initiates DAD for an
   IPv6 address and detects a genuine duplicate due to receiving an
   NS(DAD) or an NA message.  On detecting such a duplicate the access
   concentrator logs a system management message, drops the received ND
   message, and blocks the modem on whose layer 2 service identifier the
   NS(DAD) or NA message was received on.

   The network described above follows a trust model where a trusted
   router serves un-trusted IPv6 host nodes.  Operators of such networks
   have a desire to take automated action if a network interface of the
   trusted router has a tentative or optimistic address duplicate with a
   host served by trusted router interface.  Any other network that
   follows the same trust model MAY use the automated actions proposed
   in this section.

6.  Security Considerations

   The nonce can be exploited by a rogue deliberately changing the nonce
   to fail the looped back detection specified by the Enhanced DAD
   algorithm.  SEND is recommended for this exploit.  For any mitigation
   suggested in the document such as disabling DAD has an obvious
   security issue before a remote node on the link can issue reflected
   NS(DAD) messages.  Again, SEND is recommended for this exploit.

7.  IANA Considerations


8.  Acknowledgements

   Thanks (in alphabetical order by first name) to Dmitry Anipko, Eric
   Levy-Abegnoli, Erik Nordmark, Fred Templin, Suresh Krishnan, and
   Tassos Chatzithomaoglou for their guidance and review of the
   document.  Thanks to Thomas Narten for encouraging this work.  Thanks
   to Steinar Haug and Scott Beuker for describing the use cases.

9.  Normative References

   [RFC1247]  Moy, J., "OSPF Version 2", RFC 1247, July 1991.

   [RFC1661]  Simpson, W., "The Point-to-Point Protocol (PPP)", STD 51,
              RFC 1661, July 1994.

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119, March 1997.

   [RFC3971]  Arkko, J., Kempf, J., Zill, B., and P. Nikander, "SEcure
              Neighbor Discovery (SEND)", RFC 3971, March 2005.

   [RFC4429]  Moore, N., "Optimistic Duplicate Address Detection (DAD)
              for IPv6", RFC 4429, April 2006.

   [RFC4861]  Narten, T., Nordmark, E., Simpson, W., and H. Soliman,
              "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861,
              September 2007.

   [RFC4862]  Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless
              Address Autoconfiguration", RFC 4862, September 2007.

Authors' Addresses

   Rajiv Asati
   Cisco Systems, Inc.
   7025 Kit Creek road
   Research Triangle Park, NC  27709-4987


   Hemant Singh
   Cisco Systems, Inc.
   1414 Massachusetts Ave.
   Boxborough, MA  01719

   Phone: +1 978 936 1622

   Wes Beebee
   Cisco Systems, Inc.
   1414 Massachusetts Ave.
   Boxborough, MA  01719

   Phone: +1 978 936 2030

   Eli Dart
   Lawrence Berkeley National Laboratory
   1 Cyclotron Road, Berkeley, CA 94720

   Wes George
   Time Warner Cable
   13820 Sunrise Valley Drive
   Herndon, VA  20171


   Carlos Pignataro
   Cisco Systems, Inc.
   7200-12 Kit Creek Road
   Research Triangle Park, NC  27709