Internet Engineering Task Force                                  G. Chen
Internet-Draft                                                    Z. Cao
Intended status: Informational                              China Mobile
Expires: April 05, 2014                                           C. Xie
                                                           China Telecom
                                                                D. Binet
                                                   France Telecom-Orange
                                                        October 02, 2013

                     NAT64 Operational Experiences


   This document summarizes NAT64 function deployment scenarios and
   operational experience.  Both NAT64 Carrier Grade NAT (NAT64-CGN) and
   NAT64 server Front End (NAT64-FE) are considered in this document.

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 April 05, 2014.

Copyright Notice

   Copyright (c) 2013 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

Chen, et al.             Expires April 05, 2014                 [Page 1]

Internet-Draft              NAT64 Experiences               October 2013

   the Trust Legal Provisions and are provided without warranty as
   described in the Simplified BSD License.

Table of Contents

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   2
   2.  Terminology . . . . . . . . . . . . . . . . . . . . . . . . .   3
   3.  NAT64 Networking Experiences  . . . . . . . . . . . . . . . .   4
     3.1.  NAT64-CGN Consideration . . . . . . . . . . . . . . . . .   4
       3.1.1.  NAT64-CGN Usages  . . . . . . . . . . . . . . . . . .   4
       3.1.2.  DNS64 Deployment  . . . . . . . . . . . . . . . . . .   4
       3.1.3.  NAT64 Placement . . . . . . . . . . . . . . . . . . .   4
       3.1.4.  Co-existence of NAT64 and NAT44 . . . . . . . . . . .   5
     3.2.  NAT64-FE Consideration  . . . . . . . . . . . . . . . . .   6
   4.  High Availability . . . . . . . . . . . . . . . . . . . . . .   7
     4.1.  Redundancy Design . . . . . . . . . . . . . . . . . . . .   7
     4.2.  Load Balancing  . . . . . . . . . . . . . . . . . . . . .   8
   5.  Source Address Transparency . . . . . . . . . . . . . . . . .   9
     5.1.  Traceability  . . . . . . . . . . . . . . . . . . . . . .   9
     5.2.  Geo-location  . . . . . . . . . . . . . . . . . . . . . .  10
   6.  Quality of Experience . . . . . . . . . . . . . . . . . . . .  10
     6.1.  Service Reachability  . . . . . . . . . . . . . . . . . .  11
     6.2.  Resource Reservation  . . . . . . . . . . . . . . . . . .  11
   7.  MTU Considerations  . . . . . . . . . . . . . . . . . . . . .  12
   8.  ULA Usages  . . . . . . . . . . . . . . . . . . . . . . . . .  12
   9.  Security Considerations . . . . . . . . . . . . . . . . . . .  13
   10. IANA Considerations . . . . . . . . . . . . . . . . . . . . .  14
   11. Acknowledgements  . . . . . . . . . . . . . . . . . . . . . .  14
   12. Additional Author List  . . . . . . . . . . . . . . . . . . .  14
   13. References  . . . . . . . . . . . . . . . . . . . . . . . . .  15
     13.1.  Normative References . . . . . . . . . . . . . . . . . .  15
     13.2.  Informative References . . . . . . . . . . . . . . . . .  16
   Appendix A.  Testing Results of Application Behavior  . . . . . .  18
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .  19

1.  Introduction

   IPv6 is the only sustainable solution for numbering nodes on Internet
   due to the IPv4 depletion.  Network operators have to deploy
   IPv6-only networks in order to meet the needs of the expanding
   internet without available IPv4 addresses.

   Single stack IPv6 network deployment can simplify network's
   provisioning.  Some justifications have been described in 464xlat
   [RFC6877].  As an example, IPv6-only connectivity confers some
   benefits to mobile operators.  In such mobile context, it enables the
   use of a single IPv6 Packet Data Protocol(PDP) context or Evolved
   Packet System (EPS) bearer if Long Term Evolution (LTE) network is

Chen, et al.             Expires April 05, 2014                 [Page 2]

Internet-Draft              NAT64 Experiences               October 2013

   considered, which eliminates significant network costs caused by
   doubling the number of PDP contexts in some cases and the need of
   IPv4 addresses to be assigned to customers.  In broadband networks
   overall, it can allow for the scaling of edge-network growth
   decoupled from IPv4 numbering limitations.

   In a transition scenario, some existing networks are likely to be
   IPv4-only configured for quite a long time.  IPv6 networks and hosts
   will need to coexist with IPv4 numbered resources.  Widespread dual-
   stack deployments have not materialized at the anticipated rate over
   the last 10 years, one possible conclusion being that legacy networks
   will not make the jump quickly.  The Internet will include nodes that
   are dual-stack, nodes that remain IPv4-only, and nodes that can be
   deployed as IPv6-only nodes.  A translation mechanism based on a
   NAT64[RFC6146] [RFC6145]function is likely to be a key element of the
   Internet for IPv6-IPv4 interoperability.

   [RFC6036] reports at least 30% of operators plan to run some kind of
   translator (presumably NAT64/DNS64).  Advice on NAT64 deployment and
   operations are therefore of some importance.  [RFC6586] documents the
   implications for IPv6 only networks.  This document intends to be
   specific to NAT64 network planning.

2.  Terminology

   In regards to IPv4/IPv6 translation, [RFC6144] has described a
   framework of enabling networks to make interworking possible between
   IPv4 and IPv6 networks.  This document has further categorized
   different NAT64 function locations and use cases.  The principle
   distinction of location is if the NAT64 is located in a Carrier Grade
   NAT or server Front End. The terms of NAT-CGN/FE are understood to be
   a topological distinction indicating different features employed in a
   NAT64 deployment.

   NAT64-CGN:  A NAT64-CGN is placed in an ISP network.  IPv6
      subscribers leverage the NAT64-CGN to access existing IPv4
      internet services.  The ISP as an administrative entity takes full
      control on the IPv6 side, but has limited or no control on the
      IPv4 side.  NAT64-CGN may have to consider the IPv4 Internet
      environment and services to make appropriate configurations.

   NAT64-FE:  A NAT64-FE is generally a device with NAT64 functionality
      in a content provider or data center network.  It could be for
      example a traffic load balancer or a firewall.  The operator of
      the NAT64-FE has full control over the IPv4 network within the
      data center, but only limited influence or control over the
      external IPv6 network.

Chen, et al.             Expires April 05, 2014                 [Page 3]

Internet-Draft              NAT64 Experiences               October 2013

3.  NAT64 Networking Experiences

3.1.  NAT64-CGN Consideration

3.1.1.  NAT64-CGN Usages

   Fixed network operators and mobile operators may locate NAT64 in
   access networks or in mobile core networks.  It can be built into
   various devices, including routers, gateways or firewalls in order to
   connect IPv6 users to the IPv4 Internet.  With regard to the numbers
   of users and the shortage of public IPv4 addresses, stateful
   NAT64[RFC6146] is more adapted to perform some maximal sharing of
   public IPv4 addresses.  The usage of stateless NAT64 can be seen with
   better transparency features
   [I-D.ietf-softwire-stateless-4v6-motivation], while it has to be
   coordinated with A+P[RFC6346] processes as specified in
   [I-D.ietf-softwire-map-t]and [I-D.ietf-softwire-4rd] in order to cope
   with IPv4 shortage.

3.1.2.  DNS64 Deployment

   DNS64[RFC6147] is recommended for use in combination with stateful
   NAT64, and will likely be an essential part of an IPv6 single-stack
   network that couples to the IPv4 Internet. 464xlat[RFC6877] is
   proposed to enable access of IPv4 only applications or applications
   that call IPv4 literal addresses.  Using DNS64 will help 464xlat to
   automatically discover NAT64 prefix through
   [I-D.ietf-behave-nat64-discovery-heuristic].  Berkeley Internet Name
   Daemon (BIND) software supports the function.  It's important to note
   that DNS64 generates the synthetic AAAA reply when services only
   register A records.  Operators should not expect to access IPv4 parts
   of a dual-stack server using NAT64/DNS64.  The traffic is forwarded
   on IPv6 paths if dual-stack servers are targeted.  IPv6 traffic may
   be routed not going through NAT64.  Only the traffic going to
   IPv4-only service would traverse NAT64.  In some sense, it encourages
   IPv6 transmission and restrains NAT uses compared to NAT44(if used),
   on which all traffic flows have to be traversed and translated.  In
   some cases, NAT64-CGN may serve double roles, i.e. a translator and
   IPv6 forwarder.  Some failure cases may be occurred once NAT64 serves
   a IPv6 gateway while is configured only with IPv4 on WAN links.  We
   tested on Top100 websites (referring to [Alexa] statistics) in such
   condition. 43% of websites are failed to be connected since those
   websites have both AAAA and A records.  Therefore, it's recommended
   to enable NAT64 WAN links with dual-stack connections in such case.

3.1.3.  NAT64 Placement

Chen, et al.             Expires April 05, 2014                 [Page 4]

Internet-Draft              NAT64 Experiences               October 2013

   All connections to IPv4 services from IPv6-only clients must traverse
   the NAT64-CGN.  It can be advantageous from the vantage-point of
   troubleshooting and traffic engineering to carry the IPv6 traffic
   natively for as long as possible within an access network and
   translate packets only at or near the network egress.  NAT64 can be
   considered as a feature of the Autonomous System (AS) border in fixed
   networks.  And, it is likely to be deployed in an IP node beyond the
   Gateway GPRS Support Node (GGSN) or Public Data Network- Gateway
   (PDN-GW) in mobile networks or directly in the gateway itself in some
   situations.  This allows consistent attribution and traceability
   within the service provider network.  It has been observed that the
   process of correlating log information is problematic from multiple-
   vendor's equipment due to inconsistent formats of log records.
   Placing NAT64 in a centralized location may reduce diversity of log
   format and simplify the network provisioning.  Moreover, since NAT64
   is only targeted at serving traffic flows from IPv6 to IPv4-only
   services, the user traffic volume should not be as high as in a NAT44
   scenario, and therefore, the gateway's capacity in such location may
   not be as much of a concern or a hurdle to deployment.  On the other
   hand, the placement in a centralized way would require more strict
   high availability (HA) design.  It would also make geo-location based
   on IPv4 addresses rather inaccurate as it is currently the case for
   NAT44 CGN already deployed in ISP networks.  More considerations or
   workarounds on HA and traceability could be found at Section 4 and
   Section 5.

3.1.4.  Co-existence of NAT64 and NAT44

   NAT64 could likely co-exist with NAT44 in a dual-stack network mostly
   because IPv4 private addresses are allocated to customers.  The
   coexistence has already appeared in mobile networks, in which dual
   stack mobile phones normally initiate some dual-stack PDN/PDP
   Type[RFC6459] to query both IPv4/IPv6 address and IPv4 allocated
   addresses are very often private ones.  [RFC6724] always prioritizes
   IPv6 connections regardless of whether the end-to-end path is native
   IPv6 or IPv6 translated to IPv4 via NAT64/DNS64.  Conversely, Happy
   Eyeballs[RFC6555] will direct some IP flows across IPv4 paths.  The
   selection of IPv4/IPv6 paths may depend on particular implementation
   choices or settings on a host-by-host basis, and may differ from an
   operator's deterministic scheme.  Our tests verified that hosts may
   find themselves switching between IPv4 and IPv6 paths as they access
   identical service, but at different times
   [I-D.kaliwoda-sunset4-dual-ipv6-coexist].  Since the topology on each
   path is different, it may cause unstable user experiences and some
   degradation of Quality of Experience (QoE) when fallback to the other
   protocol is not powerful enough for example.  It's also difficult for
   operators to debug the issue and make optimal resource usages on both
   NAT44 and NAT64.  It's desirable to find the solutions that will

Chen, et al.             Expires April 05, 2014                 [Page 5]

Internet-Draft              NAT64 Experiences               October 2013

   allow introducing IPv6/IPv4 translation service to IPv6-only hosts
   while keeping dual-stack hosts unaffected and IPv4 service unchanged.

3.2.  NAT64-FE Consideration

   Some Internet Content Providers (ICPs) may locate NAT64 in front of
   an Internet Data Center (IDC), for example co-located with load
   balancing function.  Load balancers are employed to connect different
   IP family domains, meanwhile distribute workloads across multiple
   domains or internal servers actually.  In some cases, IPv4 addresses
   exhaustion may not be a problem in some IDC's networks.  IPv6 support
   for some applications may require some investments and workloads so
   IPv6 support may not be a priority.  The use of NAT64 may be served
   to support widespread IPv6 adoption on the Internet while maintaining
   IPv4-only applications access.

   Different strategy has been described in [RFC6883]referred to as
   "inside out" and "outside in".  An IDC operator may implement the
   following practices in the NAT64-FE networking.

   o  Some ICPs who already have satisfactory operational experiences
      would adopt single stack IPv6 operations to build up their data
      center network, servers and applications since it allows new
      services delivery without having to integrate consideration of
      IPv4 NAT and address limitations of IPv4 networks.  Stateless
      NAT64[RFC6145]is used to provide services for IPv4-only
      subscribers.  [I-D.anderson-siit-dc]has provided further
      descriptions and guidelines.

   o  ICPs who attempt to offer customers IPv6 support in their
      application farms at an early stage may likely run some proxies,
      which are configured to handle incoming IPv6 flows and proxy them
      to IPv4 back-end systems.  Many load balancers have already
      integrated some proxy functionality.  IPv4 addresses configured in
      the proxy can be multiplexed like a stateful NAT64 performs.  A
      similar challenge exists once increasingly numerous users in IPv6
      Internet access an IPv4 network.  High loads on load-balancers may
      be apt to cause additional latency, IPv4 pool exhaustion, etc.
      Therefore, this approach is only reasonable at an early stage.
      ICPs may learn from the experiences and move on to dual-stack or
      IPv6 single stack in a further stage, since the native IPv6 is
      always more desirable than transition solutions.

   [RFC6144] recommends that AAAA records of load-balancers or
   application servers can be directly registered in the authoritative
   DNS servers requiring to populate these servers with corresponding
   AAAA records.  In this case, there is no need to deploy DNS64
   servers.  Those AAAA records can be some native IPv6 addresses or

Chen, et al.             Expires April 05, 2014                 [Page 6]

Internet-Draft              NAT64 Experiences               October 2013

   some IPv4-converted IPv6 addresses[RFC6052].  The type of IPv6
   address does not give the possibility to nodes to get any information
   about NAT64 presence on communication path and the possibility to
   prefer IPv4 path or the IPv6 path in dual-stack networks.  Using an
   independent sub domain e.g. may help to identify
   experimental ipv6 services to users.  How to design the FQDN for the
   IPv6 service is out-of-scope of this document.

4.  High Availability

4.1.  Redundancy Design

   High Availability (HA) is a major requirement for every service and
   network services.  The deployment of redundancy mechanism is an
   essential approach to avoid single-point failure and significantly
   increase the network reliability.  It's not only useful to stateful
   NAT64 cases, but also to stateless NAT64 gateways.

   Three redundancy modes are mainly used hereafter: cold standby, warm
   standby and hot standby.

   o  Cold standby can't replicate the NAT64 states from the primary
      equipment to the backup.  Administrators switch on the backup
      NAT64 only if the primary NAT64 fails.  As the results, all the
      existing established sessions will be disconnected.  The internal
      hosts are required to re-establish sessions to the external hosts.
      Since the backup NAT64 is manually configured to switch over to
      active NAT64, it may have unpredictable impacts to the ongoing
      services.  Normally, the handover would take several minutes so as
      to wait for the whole process of NAT64 bootstrap loader.

   o  Warm standby is a flavor of the cold standby mode.  Backup NAT64
      would keep running once the primary NAT64 is working.  This makes
      warm standby less time consuming during the traffic failover.
      Virtual Router Redundancy Protocol (VRRP)[RFC5798] can be a
      solution to enable automatic handover in the warm standby.  It was
      tested that the handover takes as maximum as 1 minute if the
      backup NAT64 needs to take over routing and re-construct the
      Binding Information Bases (BIBs) for 30 million sessions.  In
      deployment phase, operators could balance loads on distinct NAT64s
      devices.  Those NAT64s make a warm backup of each other.

   o  Hot standby must synchronize the BIBs between the primary NAT64
      and backup.  When the primary NAT64 fails, backup NAT64 would take
      over and maintain the state of all existing sessions.  The
      internal hosts don't have to re-connect the external hosts.  The
      handover time has been extremely reduced.  Thanks to Bidirectional
      Forwarding Detection (BFD) [RFC5880] combining with VRRP, a delay

Chen, et al.             Expires April 05, 2014                 [Page 7]

Internet-Draft              NAT64 Experiences               October 2013

      of only 35ms for 30 million sessions handover was observed during
      testing.  In some sense, it could guarantee the session continuity
      for every service.  In order to timely transmit states
      information, operators may have to deploy extra transport links
      between primary NAT64 and distant backup.

   In general, cold-standby and warm-standby is simpler and less
   resource intensive, but it requires clients to re-establish sessions
   when a fail-over occurs.  Hot standby doubles resource's consumption
   to synchronize the states, but it achieve seamless handover.  The
   consideration of redundancy mode for stateless NAT64 is simple,
   because it doesn't have to consider time consuming for states
   maintenance.  The warm standby is sufficient for stateless NAT64.  In
   regards to stateful NAT64, it maybe useful to investigate performance
   tolerance of applications and the traffic characteristics in a
   particular network.  Some testing results are shown in the
   Appendix A.

   Our statistics in a mobile network shown that almost 91.21% of amount
   of traffic is accounted by browsing services.  Those services don't
   require session continuity.  The handover time of warm standby is
   qualified to the delay tolerance.  Hot-standby does not offer much
   benefit for those sessions on this point.  In a fixed network, HTTP
   streaming, p2p and online games would be the major
   traffic[Cisco-VNI].  Consideration should be given to the importance
   of maintaining bindings for those sessions across failover.
   Operators may also consider the Average Revenue Per User (ARPU)
   factors to deploy suitable redundancy mode.  Warm standby may still
   be adopted to cover most services while hot standby could be used to
   upgrade Quality of Experience (QoE) using DNS64 with different
   synthetic responses for limited traffic.  Further considerations are
   discussed at Section 6.

4.2.  Load Balancing

   Load balancing is used to accompany redundancy design so that better
   scalability and resiliency could be achieved.  Stateless NAT64s allow
   asymmetric routing while anycast-based solutions are recommended in
   [I-D.ietf-softwire-map-deployment].  The deployment of load balancing
   may make more sense to stateful NAT64s for the sake of single-point
   failure avoidance.  Since the NAT64-CGN and NAT64-FE have distinct
   facilities, the following lists the considerations for each case.

   o  NAT64-CGN equipment doesn't implement load balancer functions on a
      board card.  Therefore, the gateways have to resort to DNS64 or
      internal host's behavior.  Once DNS64 is deployed, the load
      balancing can be performed by synthesizing AAAA response with
      different IPv6 prefixes.  For the applications not requiring DNS

Chen, et al.             Expires April 05, 2014                 [Page 8]

Internet-Draft              NAT64 Experiences               October 2013

      resolver, internal hosts could learn multiple IPv6 prefixes
      through the approaches defined
      in[I-D.ietf-behave-nat64-discovery-heuristic] and then select one
      based on a given prefix selection policy.

   o  A dedicated Load Balancer could be deployed at front of a NAT64-FE
      farm.  Load Balancer uses proxy mode to redirect the flows to the
      appropriate NAT64 instance.  Stateful NAT64s require a
      deterministic pattern to arrange the traffic in order to ensure
      outbound/inbound flows traverse the identical NAT64.  Therefore,
      static scheduling algorithms, for example source-address based
      policy, is preferred.  A dynamic algorithm, for example Round-
      Robin, may have impacts on applications seeking session
      continuity, which described in the Table 1.

5.  Source Address Transparency

5.1.  Traceability

   Traceability is required in many cases such as identifying malicious
   attacks sources and accounting requirements.  Operators are asked to
   record the NAT64 log information for specific periods of time.  In
   our lab testing, the log information from 200,000 subscribers have
   been collected from a stateful NAT64 gateway for 60 days.
   Syslog[RFC5424] has been adopted to transmit log message from NAT64
   to a log station.  Each log message contains transport protocol,
   source IPv6 address:port, translated IPv4 address: port and
   timestamp.  It takes almost 125 bytes long in ASCII format.  It has
   been verified that the volume of recorded information reach up to
   42.5 terabytes in the raw format and 29.07 terabytes in a compact
   format.  Operators have to build up dedicated transport links,
   storage system and servers for the purpose.  There are also several
   implementations to mitigate the issue.  For example, stateful NAT64
   could configure with bulk port allocation method.  Once a subscriber
   creates the first session, a number of ports are pre-allocated.  A
   bulk allocation message is logged indicating this allocation.
   Subsequent session creations will use one of the pre-allocated port
   and hence does not require logging.  The log volume in this case may
   be only one thousandth of dynamic port allocation.  Some
   implementations may adopt static port-range allocations
   [I-D.donley-behave-deterministic-cgn] which eliminates the need for
   per-subscriber logging.  As a side effect, the IPv4 multiplexing
   efficiency is decreased regarding to those methods.  For example, the
   utilization ratio of public IPv4 address is dropped approximately to
   75% when NAT64 gateway is configured with bulk port allocation (The
   lab testing allocates each subscriber with 400 ports).  In addition,
   port-range based allocation should also consider port randomization
   described in [RFC6056] . A trade-off among address multiplexing

Chen, et al.             Expires April 05, 2014                 [Page 9]

Internet-Draft              NAT64 Experiences               October 2013

   efficiency, logging storage compression and port allocation
   complexity should be considered.  More discussions could be found in
   [I-D.chen-sunset4-cgn-port-allocation].Basically, the decision
   depends on usable IPv4 resource and investments of log systems.

5.2.  Geo-location

   IP addresses are usually used as inputs to geo-location services.
   The use of address sharing will prevent these systems from resolving
   the location of a host based on IP address alone.  Applications that
   assume such geographic information may not work as intended.  The
   possible solutions listed in [RFC6967] are intended to bridge the
   gap.  However, those solutions can only provide a sub-optimal
   substitution to solve the problem of host identification, in
   particular it may not today solve problems with source identification
   through translation.  The following lists current practices to
   mitigate the issue.

   o  Operators who adopt NAT64-FE may leverage the application layer
      proxies, e.g. X-Forwarded-For (XFF)
      [I-D.ietf-appsawg-http-forwarded], to convey the IPv6 source
      address in HTTP headers.  Those messages would be passed on to
      web-servers.  The queried server can lookup Radius servers for the
      target subscribers based on IPv6 addresses included in XFF HTTP
      headers.  XFF is the de facto standard which has been integrated
      in most Load Balancers.  Therefore, it may be superior to use in a
      NAT-FE environment.  In the downsides, XFF is specific to HTTP.
      It restricts the usages so that the solution can't be applied to
      requests made over HTTPs.  This makes geo-location problematic for
      HTTPs based services.

   o  The NAT64-CGN equipment may not implement XFF.  Geo-location based
      on shared IPv4 address is rather inaccurate in that case.
      Operators could subdivide the outside IPv4 address pool so an IPv6
      address can be translated depending on their geographical
      locations.  As consequence, location information can be identified
      from a certain IPv4 address range.  [RFC6967] also enumerates
      several options to reveal the host identifier.  Each solution
      likely has their-own specific usage.  For the geo-location systems
      relying on a Radius database[RFC5580], we have investigated to
      deliver NAT64 BIBs and Session Table Entrys (STEs) to a Radius
      server[I-D.chen-behave-nat64-radius-extension].  This method could
      provide geo-location system with an internal IPv6 address to
      identify each user.  It can get along with [RFC5580] to convey
      original source address through same message bus.

6.  Quality of Experience

Chen, et al.             Expires April 05, 2014                [Page 10]

Internet-Draft              NAT64 Experiences               October 2013

6.1.  Service Reachability

   NAT64 is providing a translation capability between IPv6 and IPv4
   end-nodes.  In order to provide the reachability between two IP
   address families, NAT64-CGN has to implement appropriate application
   aware functions, i.e. Application Layer Gateway (ALG), where address
   translation is not itself sufficient and security mechanisms do not
   render it infeasible.  Most NAT64-CGNs mainly provide FTP-
   ALG[RFC6384].  NAT64-FEs may have functional richness on Load
   Balancer, for example HTTP-ALG, HTTPs-ALG, RTSP-ALG and SMTP-ALG have
   been supported.  It should be noted that ALGs may impact the
   performance on a NAT64 box to some extent.  ISPs as well as content
   providers might choose to avoid situations where the imposition of an
   ALG might be required.  At the same time, it is also important to
   remind customers and application developers that IPv6 end-to-end
   usage does not require ALG imposition and therefore results in a
   better overall user experience.

6.2.  Resource Reservation

   Session status normally is managed by a static timer.  For example,
   the value of the "established connection idle-timeout" for TCP
   sessions must not be less than 2 hours 4 minutes[RFC5382] and 5
   minutes for UDP sessions[RFC4787].  In some cases, NAT resource maybe
   significantly consumed by largely inactive users.  The NAT translator
   and other customers would suffer from service degradation due to port
   consummation by other subscribers using the same NAT64 device.  A
   flexible NAT session control is desirable to resolve the issues.
   PCP[RFC6887] could be a candidate to provide such capability.  A
   NAT64-CGN should integrate with a PCP server, to allocate available
   IPv4 address/port resources.  Resources could be assigned to PCP
   clients through PCP MAP/PEER mode.  Such ability can be considered to
   upgrade user experiences, for example assigning different sizes of
   port ranges for different subscribers.  Those mechanisms are also
   helpful to minimize terminal battery consumption and reduce the
   number of keep-alive messages to be sent by mobile terminal devices.

   Subscribers can also benefit from network reliability.  It has been
   discussed that hot-standby offers satisfactory experience once outage
   of primary NAT64 is occurred.  Operators may rightly be concerned
   about the considerable investment required for NAT64 equipment
   relative to low ARPU income.  For example, transport links may cost
   much, because primary NAT64 and backup are normally located at
   different locations, separated by a relatively large distance.
   Additional maintenance has to be spent to ensure the connectivity
   quality.  However, that may be necessary to some applications, which
   are delay-sensitive and seek session continuity, for example on-line
   games and live-streaming.  Operators may be able to get added-values

Chen, et al.             Expires April 05, 2014                [Page 11]

Internet-Draft              NAT64 Experiences               October 2013

   from those services by offering first-class services.  It can be pre-
   configured on the gateway to hot-standby modes depending on
   subscriber's profile.  The rest of other sessions can be covered by
   cold/warm standby.

7.  MTU Considerations

   IPv6 requires that every link in the internet have an Maximum
   Transmission Unit (MTU) of 1280 octets or greater[RFC2460].  However,
   in case of NAT64 translation deployment, some IPv4 MTU constrained
   link will be used in some communication path and originating IPv6
   nodes may therefore receive an ICMP Packet Too Big (PTB) message,
   reporting a Next-Hop MTU less than 1280 bytes.  The result would be
   that IPv6 allows packets to contain a fragmentation header, without
   the packet being fragmented into multiple pieces.  A NAT64 would
   receive IPv6 packets with fragmentation header in which "M" flag
   equal to 0 and "Fragment Offset" equal to 0.  Those packets likely
   impact other fragments already queued with the same set of {IPv6
   Source Address, IPv6 Destination Address, Fragment Identification}.
   If the NAT64 box is compliant with [RFC5722], there is risk that all
   the fragments have to be dropped.

   [RFC6946] discusses how this situation could be exploited by an
   attacker to perform fragmentation-based attacks, and also proposes an
   improved handling of such packets.  It required enhancements on NAT64
   gateway implementations to isolate packet's processing.  NAT64 should
   follow the recommendation and take steps to prevent the risks of

   Another approach that potentially avoids this issue is to configure
   IPv4 MTU more than 1260 bytes.  It would forbid the occurrence of PTB
   smaller than 1280 bytes.  Such an operational consideration is hard
   to universally apply to the legacy "IPv4 Internet" NAT64-CGN bridged.
   However, it's a feasible approach in NAT64-FE cases, since a IPv4
   network NAT64-FE connected is rather well-organized and operated by a
   IDC operator or content provider.  Therefore, the MTU of IPv4 network
   in NAT64-FE case are strongly recommended to set to more than 1260

8.  ULA Usages

Chen, et al.             Expires April 05, 2014                [Page 12]

Internet-Draft              NAT64 Experiences               October 2013

   Unique Local Addresses (ULAs) are defined in [RFC4193] to be
   renumbered within a network site for local communications.  Operators
   may use ULAs as NAT64 prefixes to provide site-local IPv6
   connectivity.  Those ULA prefixes are stripped when the packets going
   to the IPv4 Internet, therefore ULAs are only valid in the IPv6 site.
   The use of ULAs could help in identifying the translation
   traffic.[I-D.ietf-v6ops-ula-usage-recommendations] has provided
   further guidance for the ULAs usages.

   We configure ULAs as NAT64 prefixes on a NAT64-CGN.  If a host is
   only assigned with an IPv6 address and connected to NAT64-CGN, when
   connect to an IPv4 service, it would receive AAAA record generated by
   the DNS64 with the ULA prefix.  A Global Unicast Address (GUA) will
   be selected as the source address to the ULA destination address.
   When the host has both IPv4 and IPv6 address, it would initiate both
   A and AAAA record lookup may due to Happy Eyeballs [RFC6555], then
   both original A record and DNS64-generated AAAA record would be
   received.  The host would prefer the AAAA record as the destination,
   if the address selection process still follows the recommendation of
   [RFC3484].  While, if the host implements the default policy table as
   defined in [RFC6724], since the IPv4 takes the precedence over ULA
   (which is difference than [RFC3484]), an IPv4 path will be always
   preferred.  It may be undesirable because the traffic path is
   unexpected due to vary implementations on hosts.  Operators have to
   make site-specific rules to gauge the host's behaviors, which may
   likely add the operational complexity.

   ULAs can't work when hosts transit the Internet to connect with
   NAT64.  Therefore, ULAs are inapplicable to the case of NAT64-FE.

9.  Security Considerations

   his document presents the deployment experiences of NAT64 in CGN and
   FE scenarios.  In general, RFC 6146[RFC6146] provides TCP-tracking,
   address-dependent filtering mechanisms to protect NAT64 from
   Distributed Denial of Service (DDoS).  In NAT64-CGN cases, operators
   also could adopt unicast Reverse Path Forwarding (uRPF)[RFC3704] and
   black/white-list to enhance the security by specifying access
   policies.  For example, NAT64-CGN should forbid establish NAT64 BIB
   for incoming IPv6 packets if uRPF in Strict or Loose mode check does
   not pass or whose source IPv6 address is associated to black-lists.

   The stateful NAT64-FE creates state and maps that connection to an
   internally-facing IPv4 address and port.  An attacker can consume the
   resources of the NAT64-FE device by sending an excessive number of
   connection attempts.  Without a DDoS limitation mechanism, the
   NAT64-FE is exposed to attacks.  Load Balancer is recommended to
   enable the capabilities of line rate DDOS defense, such as the

Chen, et al.             Expires April 05, 2014                [Page 13]

Internet-Draft              NAT64 Experiences               October 2013

   employment of SYN PROXY-COOKIE.  Security domain division is
   necessary as well in this case.  Therefore, Load Balancers could not
   only serve for optimization of traffic distribution, but also prevent
   service from quality deterioration due to security attacks.

10.  IANA Considerations

   This memo includes no request to IANA.

11.  Acknowledgements

   The authors would like to thank Jari Arkko, Dan Wing, Remi Despres,
   Fred Baker, Hui Deng, Lee Howard, Iljitsch van Beijnum, Philip
   Matthews, Randy Bush, Mikael Abrahamsson, Lorenzo Colitti and Sheng
   Jiang for their helpful comments.

   Many thanks to Wesley George and Satoru Matsushima for their detailed

   The authors especially thank Joel Jaeggli and Ray Hunter for his
   efforts and contributions on editing which substantially improves the
   legibility of the document.

   Thanks to Cameron Byrne who was an active co-author of some earlier
   versions of this draft.

12.  Additional Author List

   The following are extended authors who contributed to the effort:

   Qiong Sun
   China Telecom
   Room 708, No.118, Xizhimennei Street
   Beijing 100035
   Phone: +86-10-58552936

   QiBo Niu
   50,RuanJian Road.
   YuHua District,
   Nan Jing  210012

Chen, et al.             Expires April 05, 2014                [Page 14]

Internet-Draft              NAT64 Experiences               October 2013

13.  References

13.1.  Normative References

              Petersson, A. and M. Nilsson, "Forwarded HTTP Extension",
              draft-ietf-appsawg-http-forwarded-10 (work in progress),
              October 2012.

              Savolainen, T., Korhonen, J., and D. Wing, "Discovery of
              the IPv6 Prefix Used for IPv6 Address Synthesis", draft-
              ietf-behave-nat64-discovery-heuristic-17 (work in
              progress), April 2013.

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

   [RFC3704]  Baker, F. and P. Savola, "Ingress Filtering for Multihomed
              Networks", BCP 84, RFC 3704, March 2004.

   [RFC4193]  Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast
              Addresses", RFC 4193, October 2005.

   [RFC4787]  Audet, F. and C. Jennings, "Network Address Translation
              (NAT) Behavioral Requirements for Unicast UDP", BCP 127,
              RFC 4787, January 2007.

   [RFC5382]  Guha, S., Biswas, K., Ford, B., Sivakumar, S., and P.
              Srisuresh, "NAT Behavioral Requirements for TCP", BCP 142,
              RFC 5382, October 2008.

   [RFC5424]  Gerhards, R., "The Syslog Protocol", RFC 5424, March 2009.

   [RFC5580]  Tschofenig, H., Adrangi, F., Jones, M., Lior, A., and B.
              Aboba, "Carrying Location Objects in RADIUS and Diameter",
              RFC 5580, August 2009.

   [RFC5722]  Krishnan, S., "Handling of Overlapping IPv6 Fragments",
              RFC 5722, December 2009.

   [RFC5798]  Nadas, S., "Virtual Router Redundancy Protocol (VRRP)
              Version 3 for IPv4 and IPv6", RFC 5798, March 2010.

   [RFC5880]  Katz, D. and D. Ward, "Bidirectional Forwarding Detection
              (BFD)", RFC 5880, June 2010.

Chen, et al.             Expires April 05, 2014                [Page 15]

Internet-Draft              NAT64 Experiences               October 2013

   [RFC6052]  Bao, C., Huitema, C., Bagnulo, M., Boucadair, M., and X.
              Li, "IPv6 Addressing of IPv4/IPv6 Translators", RFC 6052,
              October 2010.

   [RFC6145]  Li, X., Bao, C., and F. Baker, "IP/ICMP Translation
              Algorithm", RFC 6145, April 2011.

   [RFC6146]  Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful
              NAT64: Network Address and Protocol Translation from IPv6
              Clients to IPv4 Servers", RFC 6146, April 2011.

   [RFC6147]  Bagnulo, M., Sullivan, A., Matthews, P., and I. van
              Beijnum, "DNS64: DNS Extensions for Network Address
              Translation from IPv6 Clients to IPv4 Servers", RFC 6147,
              April 2011.

   [RFC6384]  van Beijnum, I., "An FTP Application Layer Gateway (ALG)
              for IPv6-to-IPv4 Translation", RFC 6384, October 2011.

   [RFC6555]  Wing, D. and A. Yourtchenko, "Happy Eyeballs: Success with
              Dual-Stack Hosts", RFC 6555, April 2012.

   [RFC6724]  Thaler, D., Draves, R., Matsumoto, A., and T. Chown,
              "Default Address Selection for Internet Protocol Version 6
              (IPv6)", RFC 6724, September 2012.

   [RFC6887]  Wing, D., Cheshire, S., Boucadair, M., Penno, R., and P.
              Selkirk, "Port Control Protocol (PCP)", RFC 6887, April

   [RFC6946]  Gont, F., "Processing of IPv6 "Atomic" Fragments", RFC
              6946, May 2013.

13.2.  Informative References

   [Alexa]    Alexa, "", April 2013.

              Cisco, "Cisco Visual Networking Index: Forecast and
              Methodology, 2012-2017,
    ", May 2013.

              Anderson, T., "Stateless IP/ICMP Translation in IPv6 Data
              Centre Environments", draft-anderson-siit-dc-00 (work in
              progress), November 2012.


Chen, et al.             Expires April 05, 2014                [Page 16]

Internet-Draft              NAT64 Experiences               October 2013

              Chen, G. and D. Binet, "Radius Attributes for Stateful
              NAT64", draft-chen-behave-nat64-radius-extension-00 (work
              in progress), July 2013.

              Chen, G., "Analysis of NAT64 Port Allocation Method",
              draft-chen-sunset4-cgn-port-allocation-02 (work in
              progress), July 2013.

              Donley, C., Grundemann, C., Sarawat, V., Sundaresan, K.,
              and O. Vautrin, "Deterministic Address Mapping to Reduce
              Logging in Carrier Grade NAT Deployments", draft-donley-
              behave-deterministic-cgn-06 (work in progress), July 2013.

              Despres, R., Jiang, S., Penno, R., Lee, Y., Chen, G., and
              M. Chen, "IPv4 Residual Deployment via IPv6 - a Stateless
              Solution (4rd)", draft-ietf-softwire-4rd-06 (work in
              progress), July 2013.

              Sun, Q., Chen, M., Chen, G., Tsou, T., and S. Perreault,
              "Mapping of Address and Port (MAP) - Deployment
              Considerations", draft-ietf-softwire-map-deployment-02
              (work in progress), July 2013.

              Li, X., Bao, C., Dec, W., Troan, O., Matsushima, S., and
              T. Murakami, "Mapping of Address and Port using
              Translation (MAP-T)", draft-ietf-softwire-map-t-04 (work
              in progress), September 2013.

              Boucadair, M., Matsushima, S., Lee, Y., Bonness, O.,
              Borges, I., and G. Chen, "Motivations for Carrier-side
              Stateless IPv4 over IPv6 Migration Solutions", draft-ietf-
              softwire-stateless-4v6-motivation-05 (work in progress),
              November 2012.

              Liu, B., Jiang, S., and C. Byrne, "Recommendations of
              Using Unique Local Addresses", draft-ietf-v6ops-ula-usage-
              recommendations-00 (work in progress), May 2013.


Chen, et al.             Expires April 05, 2014                [Page 17]

Internet-Draft              NAT64 Experiences               October 2013

              Kaliwoda, A. and D. Binet, "Co-existence of both dual-
              stack and IPv6-only hosts", draft-kaliwoda-sunset4-dual-
              ipv6-coexist-01 (work in progress), October 2012.

   [RFC3484]  Draves, R., "Default Address Selection for Internet
              Protocol version 6 (IPv6)", RFC 3484, February 2003.

   [RFC6036]  Carpenter, B. and S. Jiang, "Emerging Service Provider
              Scenarios for IPv6 Deployment", RFC 6036, October 2010.

   [RFC6056]  Larsen, M. and F. Gont, "Recommendations for Transport-
              Protocol Port Randomization", BCP 156, RFC 6056, January

   [RFC6144]  Baker, F., Li, X., Bao, C., and K. Yin, "Framework for
              IPv4/IPv6 Translation", RFC 6144, April 2011.

   [RFC6269]  Ford, M., Boucadair, M., Durand, A., Levis, P., and P.
              Roberts, "Issues with IP Address Sharing", RFC 6269, June

   [RFC6346]  Bush, R., "The Address plus Port (A+P) Approach to the
              IPv4 Address Shortage", RFC 6346, August 2011.

   [RFC6459]  Korhonen, J., Soininen, J., Patil, B., Savolainen, T.,
              Bajko, G., and K. Iisakkila, "IPv6 in 3rd Generation
              Partnership Project (3GPP) Evolved Packet System (EPS)",
              RFC 6459, January 2012.

   [RFC6586]  Arkko, J. and A. Keranen, "Experiences from an IPv6-Only
              Network", RFC 6586, April 2012.

   [RFC6877]  Mawatari, M., Kawashima, M., and C. Byrne, "464XLAT:
              Combination of Stateful and Stateless Translation", RFC
              6877, April 2013.

   [RFC6883]  Carpenter, B. and S. Jiang, "IPv6 Guidance for Internet
              Content Providers and Application Service Providers", RFC
              6883, March 2013.

   [RFC6967]  Boucadair, M., Touch, J., Levis, P., and R. Penno,
              "Analysis of Potential Solutions for Revealing a Host
              Identifier (HOST_ID) in Shared Address Deployments", RFC
              6967, June 2013.

Appendix A.  Testing Results of Application Behavior

Chen, et al.             Expires April 05, 2014                [Page 18]

Internet-Draft              NAT64 Experiences               October 2013

   We test several application behaviors in a lab environment to
   evaluate the impact when a primary NAT64 is out of service.  In this
   testing, participants are asked to connect a IPv6-only WiFi network
   using laptops, tablets or mobile phones.  NAT64 is deployed as the
   gateway to connect Internet service.  The tested applications are
   shown in the below table.  Cold standby, warm standby and hot standby
   are taken truns to be tested.  The partipants may experience service
   interruption due to the NAT64 handover.  Different interruption
   intervals are tested to gauge application behaviors.  The results are
   illuminated as below.

                  Table 1: The acceptable delay of applications
   |     APPs       | Acceptable Interrupt   |   Session Continuity    |
   |                |        Recovery        |                         |
   | Web Browse     |As maximum as 6s        |  No                     |
   |Http streaming  |As maximum as 10s(cache)|  Yes                    |
   | Gaming         | 200ms~400ms            |  Yes                    |
   | P2P            | 10~16s                 |  Yes                    |
   |Instant Message |1 minute                |  Yes                    |
   |Mail            |30 seconds              |  No                     |
   |Downloading     |1 minutes               |  No                     |

Authors' Addresses

   Gang Chen
   China Mobile
   53A,Xibianmennei Ave.,
   Xuanwu District,
   Beijing  100053


Chen, et al.             Expires April 05, 2014                [Page 19]

Internet-Draft              NAT64 Experiences               October 2013

   Zhen Cao
   China Mobile
   53A,Xibianmennei Ave.,
   Xuanwu District,
   Beijing  100053


   Chongfeng Xie
   China Telecom
   Room 708 No.118, Xizhimenneidajie
   Beijing  100035


   David Binet
   France Telecom-Orange


Chen, et al.             Expires April 05, 2014                [Page 20]