Routing Loop Attack Using IPv6 Automatic Tunnels: Problem Statement and Proposed Mitigations

Internet Engineering Task Force (IETF)                        G. Nakibly
Request for Comments: 6324                                        NEWRSC
Category: Informational                                       F. Templin
ISSN: 2070-1721                             Boeing Research & Technology
                                                             August 2011


           Routing Loop Attack Using IPv6 Automatic Tunnels:
               Problem Statement and Proposed Mitigations

Abstract

   This document is concerned with security vulnerabilities in IPv6-in-
   IPv4 automatic tunnels.  These vulnerabilities allow an attacker to
   take advantage of inconsistencies between the IPv4 routing state and
   the IPv6 routing state.  The attack forms a routing loop that can be
   abused as a vehicle for traffic amplification to facilitate denial-
   of-service (DoS) attacks.  The first aim of this document is to
   inform on this attack and its root causes.  The second aim is to
   present some possible mitigation measures.  It should be noted that
   at the time of this writing there are no known reports of malicious
   attacks exploiting these vulnerabilities.  Nonetheless, these
   vulnerabilities can be activated by accidental misconfiguration.

Status of This Memo

   This document is not an Internet Standards Track specification; it is
   published for informational purposes.

   This document is a product of the Internet Engineering Task Force
   (IETF).  It represents the consensus of the IETF community.  It has
   received public review and has been approved for publication by the
   Internet Engineering Steering Group (IESG).  Not all documents
   approved by the IESG are a candidate for any level of Internet
   Standard; see Section 2 of RFC 5741.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at
   http://www.rfc-editor.org/info/rfc6324.

Copyright Notice

   Copyright (c) 2011 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
   (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 ....................................................2
   2. A Detailed Description of the Attack ............................4
   3. Proposed Mitigation Measures ....................................6
      3.1. Verification of Endpoint Existence .........................6
           3.1.1. Neighbor Cache Check ................................6
           3.1.2. Known IPv4 Address Check ............................7
      3.2. Operational Measures .......................................7
           3.2.1. Avoiding a Shared IPv4 Link .........................7
           3.2.2. A Single Border Router ..............................8
           3.2.3. A Comprehensive List of Tunnel Routers ..............9
           3.2.4. Avoidance of On-Link Prefixes .......................9
      3.3. Destination and Source Address Checks .....................15
           3.3.1. Known IPv6 Prefix Check ............................16
   4. Recommendations ................................................17
   5. Security Considerations ........................................17
   6. Acknowledgments ................................................18
   7. References .....................................................18
      7.1. Normative References ......................................18
      7.2. Informative References ....................................19

1.  Introduction

   IPv6-in-IPv4 tunnels are an essential part of many migration plans
   for IPv6.  They allow two IPv6 nodes to communicate over an IPv4-only
   network.  Automatic tunnels that assign IPv6 prefixes with stateless
   address mapping properties (hereafter called "automatic tunnels") are
   a category of tunnels in which a tunneled packet's egress IPv4
   address is embedded within the destination IPv6 address of the
   packet.  An automatic tunnel's router is a router that respectively
   encapsulates and decapsulates the IPv6 packets into and out of the
   tunnel.

   Reference [USENIX09] pointed out the existence of a vulnerability in
   the design of IPv6 automatic tunnels.  Tunnel routers operate on the
   implicit assumption that the destination address of an incoming IPv6
   packet is always an address of a valid node that can be reached via
   the tunnel.  The assumption of path validity can introduce routing
   loops as the inconsistency between the IPv4 routing state and the
   IPv6 routing state allows a routing loop to be formed.  Although
   those loops will not trap normal data, they will catch traffic
   targeted at addresses that have become unavailable, and misconfigured
   traffic can enter the loop.

   The looping vulnerability can be triggered accidentally, or exploited
   maliciously by an attacker crafting a packet that is routed over a
   tunnel to a node that is not associated with the packet's
   destination.  This node may forward the packet out of the tunnel to
   the native IPv6 network.  There, the packet is routed back to the
   ingress point, which forwards it back into the tunnel.  Consequently,
   the packet loops in and out of the tunnel.  The loop terminates only
   when the Hop Limit field in the IPv6 header of the packet is
   decremented to zero.  This vulnerability can be abused as a vehicle
   for traffic amplification to facilitate DoS attacks [RFC4732].

   Without compensating security measures in place, all IPv6 automatic
   tunnels that are based on protocol-41 encapsulation [RFC4213] are
   vulnerable to such an attack, including the Intra-Site Automatic
   Tunnel Addressing Protocol (ISATAP) [RFC5214], 6to4 [RFC3056], and
   6rd (IPv6 Rapid Deployment on IPv4 Infrastructures) [RFC5969].  It
   should be noted that this document does not consider non-protocol-41
   encapsulation attacks.  In particular, we do not address the Teredo
   [RFC4380] attacks described in [USENIX09].  These attacks are
   considered in [TEREDO-LOOPS].

   The aim of this document is to shed light on the routing loop attack
   and describe possible mitigation measures that should be considered
   by operators of current IPv6 automatic tunnels and by designers of
   future ones.  We note that tunnels may be deployed in various
   operational environments, e.g., service provider networks, enterprise
   networks, etc.  Specific issues related to the attack that are
   derived from the operational environment are not considered in this
   document.

   Routing loops pose a risk to the stability of a network.
   Furthermore, they provide an opening for denial-of-service attacks
   that exploit the existence of the loop to increase the traffic load
   in the network.  Section 3 of this document discusses a number of
   mitigation measures.  The most desirable mitigation, however, is to
   operate the network in such a way that routing loops cannot take
   place (see Section 3.2).

2.  A Detailed Description of the Attack

   In this section, we shall denote an IPv6 address of a node by an IPv6
   prefix assigned to the tunnel and an IPv4 address of the tunnel
   endpoint, i.e., Addr(Prefix, IPv4).  Note that the IPv4 address may
   or may not be part of the prefix (depending on the specification of
   the tunnel's protocol).  The IPv6 address may be dependent on
   additional bits in the interface ID; however, for our discussion
   their exact value is not important.

   The two victims of this attack are routers -- R1 and R2 -- that
   service two different tunnel prefixes -- Prf1 and Prf2.  Both routers
   have the capability to forward IPv6 packets in and out of their
   respective tunnels.  The two tunnels need not be based on the same
   tunnel protocol.  The only condition is that the two tunnel protocols
   be based on protocol-41 encapsulation.  The IPv4 address of R1 is
   IP1, while the prefix of its tunnel is Prf1.  IP2 and Prf2 are the
   respective values for R2.  We assume that IP1 and IP2 belong to the
   same address realm, i.e., they are either both public, or both
   private and belong to the same internal network.  The following
   network diagram depicts the locations of the two routers.  The
   numbers indicate the packets of the attack and the path they
   traverse, as described below.

         [ Packet 1 ]
   v6src = Addr(Prf1, IP2)                     [ Packet 2 ]
   v6dst = Addr(Prf2, IP1)                v6src = Addr(Prf1, IP2)
   v4src = IP2; v4dst = IP1 +----------+  v6dst = Addr(Prf2, IP1)
              //===========>|  Router  |-----------------\
             ||             |    R1    |                 |
             ||             +----------+                 v
            .-.                                         .-.
         ,-(  _)-.                                   ,-(  _)-.
      .-(_ IPv4  )-.                              .-(_ IPv6  )-.
    (__   Network   )                           (__   Network   )
       `-(______)-'                                `-(______)-'
             ^^                                          |
             ||             +----------+                 |
              \\============|  Router  |<----------------/
         [ Packet 1 ]       |    R2    |    [ Packets 0 and 2 ]
   v6src = Addr(Prf1, IP2)  +----------+  v6src = Addr(Prf1, IP2)
   v6dst = Addr(Prf2, IP1)                v6dst = Addr(Prf2, IP1)
   v4src = IP2; v4dst = IP1

              Legend: ====> - tunneled IPv6, ---> - native IPv6

                Figure 1: The Network Setting of the Attack

   The attack is initiated by an accidentally or maliciously produced
   IPv6 packet (packet 0 in Figure 1) destined to a fictitious endpoint
   that appears to be reached via Prf2 and has IP1 as its IPv4 address,
   i.e., Addr(Prf2, IP1).  The source address of the packet is an
   address with Prf1 as the prefix and IP2 as the embedded IPv4 address,
   i.e., Addr(Prf1, IP2).  As the prefix of the destination address is
   Prf2, the packet will be routed over the IPv6 network to R2.

   R2 receives the packet through its IPv6 interface and forwards it
   into the tunnel with an IPv4 header having a destination address
   derived from the IPv6 destination, i.e., IP1.  The source address is
   the address of R2, i.e., IP2.  The packet (packet 1 in Figure 1) is
   routed over the IPv4 network to R1, which receives the packet on its
   IPv4 interface.  It processes the packet as a packet that originates
   from one of the end nodes of Prf1.

   Since the IPv4 source address corresponds to the IPv6 source address,
   R1 will decapsulate the packet.  Since the packet's IPv6 destination
   is outside of Prf1, R1 will forward the packet onto a native IPv6
   interface.  The forwarded packet (packet 2 in Figure 1) is identical
   to the original attack packet.  Hence, it is routed back to R2, in
   which the loop starts again.  Note that the packet may not
   necessarily be transported from R1 over the native IPv6 network.  R1
   may be connected to the IPv6 network through another tunnel.

   The crux of the attack is as follows.  The attacker exploits the fact
   that R2 does not know that R1 does not configure addresses from Prf2
   and that R1 does not know that R2 does not configure addresses from
   Prf1.  The IPv4 network acts as a shared link layer for the two
   tunnels.  Hence, the packet is repeatedly forwarded by both routers.
   It is noted that the attack will fail when the IPv4 network cannot
   transport packets between the tunnels, for example, when the two
   routers belong to different IPv4 address realms or when ingress/
   egress filtering is exercised between the routers.

   The loop will stop when the Hop Limit field of the packet reaches
   zero.  After a single loop, the Hop Limit field is decreased by the
   number of IPv6 routers on the path from R1 to R2.  Therefore, the
   number of loops is inversely proportional to the number of IPv6 hops
   between R1 and R2.

   The tunnels used by R1 and R2 may be any combination of automatic
   tunnel types, e.g., ISATAP, 6to4, and 6rd.  This has the exception
   that both tunnels cannot be of type 6to4, since two 6to4 routers
   share the same IPv6 prefix, i.e., there is only one 6to4 prefix
   (2002::/16) in the Internet.  For example, if the attack were to be
   launched on an ISATAP router (R1) and 6to4 relay (R2), then the
   destination and source addresses of the attack packet would be
   2002:IP1:* and Prf1::0200:5efe:IP2, respectively.

3.  Proposed Mitigation Measures

   This section presents some possible mitigation measures for the
   attack described above.  We shall discuss the advantages and
   disadvantages of each measure.

   The proposed measures fall under the following three categories:

   o  Verification of endpoint existence

   o  Operational measures

   o  Destination and source address checks

3.1.  Verification of Endpoint Existence

   The routing loop attack relies on the fact that a router does not
   know whether there is an endpoint that can be reached via its tunnel
   that has the source or destination address of the packet.  This
   category includes mitigation measures that aim to verify that there
   is a node that participates in the tunnel and that its address
   corresponds to the packet's destination or source addresses, as
   appropriate.

3.1.1.  Neighbor Cache Check

   One way that the router can verify that an end host exists and can be
   reached via the tunnel is by checking whether a valid entry exists
   for it in the neighbor cache of the corresponding tunnel interface.
   The neighbor cache entry can be populated through, e.g., an initial
   reachability check, receipt of neighbor discovery messages,
   administrative configuration, etc.

   When the router has a packet to send to a potential tunnel host for
   which there is no neighbor cache entry, it can perform an initial
   reachability check on the packet's destination address, e.g., as
   specified in the second paragraph of Section 8.4 of [RFC5214].  (The
   router can similarly perform a "reverse reachability" check on the
   packet's source address when it receives a packet from a potential
   tunnel host for which there is no neighbor cache entry.)  This
   reachability check parallels the address resolution specifications in
   Section 7.2 of [RFC4861], i.e., the router maintains a small queue of
   packets waiting for reachability confirmation to complete.  If
   confirmation succeeds, the router discovers that a legitimate tunnel
   host responds to the address.  Otherwise, the router discards
   subsequent packets and returns ICMP destination unreachable
   indications as specified in Section 7.2.2 of [RFC4861].

   Note that this approach assumes that the neighbor cache will remain
   coherent and not be subject to malicious attack, which must be
   confirmed based on specific deployment scenarios.  One possible way
   for an attacker to subvert the neighbor cache is to send false
   neighbor discovery messages with a spoofed source address.

3.1.2.  Known IPv4 Address Check

   Another approach that enables a router to verify that an end host
   exists and can be reached via the tunnel is simply by pre-configuring
   the router with the set of IPv4 addresses and prefixes that are
   authorized to use the tunnel.  Upon this configuration, the router
   can perform the following simple checks:

   o  When the router forwards an IPv6 packet into the tunnel interface
      with a destination address that matches an on-link prefix and that
      embeds the IPv4 address IP1, it discards the packet if IP1 does
      not belong to the configured list of IPv4 addresses.

   o  When the router receives an IPv6 packet on the tunnel's interface
      with a source address that matches an on-link prefix and that
      embeds the IPv4 address IP2, it discards the packet if IP2 does
      not belong to the configured list of IPv4 addresses.

3.2.  Operational Measures

   The following measures can be taken by the network operator.  Their
   aim is to configure the network in such a way that the attacks cannot
   take place.

3.2.1.  Avoiding a Shared IPv4 Link

   As noted above, the attack relies on having an IPv4 network as a
   shared link layer between more than one tunnel.  From this, the
   following two mitigation measures arise:

3.2.1.1.  Filtering IPv4 Protocol-41 Packets

   In this measure, a tunnel router may drop all IPv4 protocol-41
   packets received or sent over interfaces that are attached to an
   untrusted IPv4 network.  This will cut off any IPv4 network as a
   shared link.  This measure has the advantage of simplicity.  However,
   such a measure may not always be suitable for scenarios where IPv4
   connectivity is essential on all interfaces.  Most notably, filtering
   of IPv4 protocol-41 packets that belong to a 6to4 tunnel can have
   adverse effects on unsuspecting users [RFC6343].

3.2.1.2.  Operational Avoidance of Multiple Tunnels

   This measure mitigates the attack by simply allowing for a single
   IPv6 tunnel to operate in a bounded IPv4 network.  For example, the
   attack cannot take place in broadband home networks.  In such cases,
   there is a small home network having a single residential gateway
   that serves as a tunnel router.  A tunnel router is vulnerable to the
   attack only if it has at least two interfaces with a path to the
   Internet: a tunnel interface and a native IPv6 interface (as depicted
   in Figure 1).  However, a residential gateway usually has only a
   single interface to the Internet; therefore, the attack cannot take
   place.  Moreover, if there are only one or a few tunnel routers in
   the IPv4 network and all participate in the same tunnel, then there
   is no opportunity for perpetuating the loop.

   This approach has the advantage that it avoids the attack profile
   altogether without need for explicit mitigations.  However, it
   requires careful configuration management, which may not be tenable
   in large and/or unbounded IPv4 networks.

3.2.2.  A Single Border Router

   It is reasonable to assume that a tunnel router shall accept or
   forward tunneled packets only over its tunnel interface.  It is also
   reasonable to assume that a tunnel router shall accept or forward
   IPv6 packets only over its IPv6 interface.  If these two interfaces
   are physically different, then the network operator can mitigate the
   attack by ensuring that the following condition holds: there is no
   path between these two interfaces that does not go through the tunnel
   router.

   The above condition ensures that an encapsulated packet that is
   transmitted over the tunnel interface will not get to another tunnel
   router and from there to the IPv6 interface of the first router.  The
   condition also ensures the reverse direction, i.e., an IPv6 packet
   that is transmitted over the IPv6 interface will not get to another
   tunnel router and from there to the tunnel interface of the first
   router.  This condition is essentially translated to a scenario in
   which the tunnel router is the only border router between the IPv6
   network and the IPv4 network to which it is attached (as in the
   broadband home network scenario mentioned above).

3.2.3.  A Comprehensive List of Tunnel Routers

   If a tunnel router can be configured with a comprehensive list of
   IPv4 addresses of all other tunnel routers in the network, then the
   router can use the list as a filter to discard any tunneled packets
   coming from or destined to other routers.  For example, a tunnel
   router can use the network's ISATAP Potential Router List (PRL)
   [RFC5214] as a filter as long as there is operational assurance that
   all ISATAP routers are listed and that no other types of tunnel
   routers are present in the network.

   This measure parallels the one proposed for 6rd in [RFC5969] where
   the 6rd Border Relay filters all known relay addresses of other
   tunnels inside the ISP's network.

   This measure is especially useful for intra-site tunneling
   mechanisms, such as ISATAP and 6rd, since filtering can be exercised
   on well-defined site borders.  A specific ISATAP operational scenario
   for which this mitigation applies is described in Section 3 of
   [ISATAP-OPS].

3.2.4.  Avoidance of On-Link Prefixes

   The looping attack exploits the fact that a router is permitted to
   assign non-link-local IPv6 prefixes on its tunnel interfaces, which
   could cause it to send tunneled packets to other routers that do not
   configure an address from the prefix.  Therefore, if the router does
   not assign non-link-local IPv6 prefixes on its tunnel interfaces,
   there is no opportunity for it to initiate the loop.  If the router
   further ensures that the routing state is consistent for the packets
   it receives on its tunnel interfaces, there is no opportunity for it
   to propagate a loop initiated by a different router.

   This mitigation measure is available only to ISATAP routers, since
   the ISATAP stateless address mapping operates only on the Interface
   Identifier portion of the IPv6 address, and not on the IPv6 prefix.
   This measure is also only applicable on ISATAP links on which IPv4
   source address spoofing is disabled.  Finally, the measure is only
   applicable on ISATAP links on which nodes support the Dynamic Host
   Configuration Protocol for IPv6 (DHCPv6) [RFC3315].  The following
   sections discuss the operational configurations necessary to
   implement the measure.

3.2.4.1.  ISATAP Router Interface Types

   ISATAP provides a Potential Router List (PRL) to further ensure a
   loop-free topology.  Routers that are members of the PRL for the site
   configure their site-facing ISATAP interfaces as advertising router
   interfaces (see [RFC4861], Section 6.2.2), and therefore may send
   Router Advertisement (RA) messages that include non-zero Router
   Lifetimes.  Routers that are not members of the PRL for the site
   configure their site-facing ISATAP interfaces as non-advertising
   router interfaces.

3.2.4.2.  ISATAP Source Address Verification

   ISATAP nodes employ the source address verification checks specified
   in Section 7.3 of [RFC5214] as a prerequisite for decapsulation of
   packets received on an ISATAP interface.  To enable the on-link
   prefix avoidance procedures outlined in this section, ISATAP nodes
   must employ an additional source address verification check; namely,
   the node also considers the outer IPv4 source address correct for the
   inner IPv6 source address if:

   o  a forwarding table entry exists that lists the packet's IPv4
      source address as the link-layer address corresponding to the
      inner IPv6 source address via the ISATAP interface.

3.2.4.3.  ISATAP Host Behavior

   ISATAP hosts send Router Solicitation (RS) messages to obtain RA
   messages from an advertising ISATAP router as specified in [RFC4861]
   and [RFC5214].  When stateful address autoconfiguration services are
   available, the host can acquire IPv6 addresses using DHCPv6
   [RFC3315].

   To acquire addresses, the host performs standard DHCPv6 exchanges
   while mapping the IPv6 "All_DHCP_Relay_Agents_and_Servers" link-
   scoped multicast address to the IPv4 address of the advertising
   router.  The host should also use DHCPv6 Authentication in
   environments where authentication of the DHCPv6 exchanges is
   required.

   After the host receives IPv6 addresses, it assigns them to its ISATAP
   interface and forwards any of its outbound IPv6 packets via the
   advertising router as a default router.  The advertising router in
   turn maintains IPv6 forwarding table entries that list the IPv4
   address of the host as the link-layer address of the delegated IPv6
   addresses.

3.2.4.4.  ISATAP Router Behavior

   In many use case scenarios (e.g., enterprise networks, Mobile Ad Hoc
   Networks (MANETs), etc.), advertising and non-advertising ISATAP
   routers can engage in a proactive dynamic IPv6 routing protocol
   (e.g., OSPFv3, the Routing Information Protocol Next Generation
   (RIPng), etc.) over their ISATAP interfaces so that IPv6 routing/
   forwarding tables can be populated and standard IPv6 forwarding
   between ISATAP routers can be used.  In other scenarios (e.g., large
   enterprise networks, etc.), this might be impractical due to scaling
   issues.  When a proactive dynamic routing protocol cannot be used,
   non-advertising ISATAP routers send RS messages to obtain RA messages
   from an advertising ISATAP router; i.e., they act as "hosts" on their
   non-advertising ISATAP interfaces.

   Non-advertising ISATAP routers can also acquire IPv6 prefixes, e.g.,
   through the use of DHCPv6 Prefix Delegation [RFC3633] via an
   advertising router in the same fashion as described above for host-
   based DHCPv6 stateful address autoconfiguration.  The advertising
   router in turn maintains IPv6 forwarding table entries that list the
   IPv4 address of the non-advertising router as the link-layer address
   of the next hop toward the delegated IPv6 prefixes.

   After the non-advertising router acquires IPv6 prefixes, it can
   sub-delegate them to routers and links within its attached IPv6 edge
   networks, then can forward any outbound IPv6 packets coming from its
   edge networks via other ISATAP nodes on the link.

3.2.4.5.  Reference Operational Scenario

   Figure 2 depicts a reference ISATAP network topology for operational
   avoidance of on-link non-link-local IPv6 prefixes.  The scenario
   shows two advertising ISATAP routers ('A', 'B'), two non-advertising
   ISATAP routers ('C', 'E'), an ISATAP host ('G'), and three ordinary
   IPv6 hosts ('D', 'F', 'H') in a typical deployment configuration:

                       .-(::::::::)      2001:db8:3::1
                    .-(::: IPv6 :::)-.  +-------------+
                   (:::: Internet ::::) | IPv6 Host H |
                    `-(::::::::::::)-'  +-------------+
                       `-(::::::)-'
                   ,~~~~~~~~~~~~~~~~~,
              ,----|companion gateway|--.
             /     '~~~~~~~~~~~~~~~~~'  :
            /                           |.
         ,-'                              `.
        ;  +------------+   +------------+  )
        :  |  Router A  |   |  Router B  |  /    fe80::*192.0.2.5
         : |  (ISATAP)  |   |  (ISATAP)  | ;       2001:db8:2::1
         + +------------+   +------------+  \    +--------------+
        ; fe80::*192.0.2.1  fe80::*192.0.2.2 :   |   (ISATAP)   |
        |                                   ;    |    Host G    |
        :              IPv4 Site         -+-'    +--------------+
         `-. (PRL: 192.0.2.1, 192.0.2.2)  .)
            \                           _)
             `-----+--------)----+'----'
        fe80::*192.0.2.3         fe80::*192.0.2.4          .-.
        +--------------+         +--------------+       ,-(  _)-.
        |   (ISATAP)   |         |   (ISATAP)   |    .-(_ IPv6  )-.
        |   Router C   |         |   Router E   |--(__Edge Network )
        +--------------+         +--------------+     `-(______)-'
         2001:db8:0::/48          2001:db8:1::/48           |
                |                                     2001:db8:1::1
               .-.                                   +-------------+
            ,-(  _)-.       2001:db8:0::1            | IPv6 Host F |
         .-(_ IPv6  )-.   +-------------+            +-------------+
       (__Edge Network )--| IPv6 Host D |
          `-(______)-'    +-------------+

      (* == "5efe:")

                Figure 2: Reference ISATAP Network Topology

   In Figure 2, advertising ISATAP routers 'A' and 'B' within the IPv4
   site connect to the IPv6 Internet, either directly or via a companion
   gateway.  'A' configures a provider network IPv4 interface with
   address 192.0.2.1 and arranges to add the address to the provider
   network PRL.  'A' next configures an advertising ISATAP router
   interface with link-local IPv6 address fe80::5efe:192.0.2.1 over the
   IPv4 interface.  In the same fashion, 'B' configures the IPv4
   interface address 192.0.2.2, adds the address to the PRL, then
   configures the IPv6 ISATAP interface link-local address
   fe80::5efe:192.0.2.2.

   Non-advertising ISATAP router 'C' connects to one or more IPv6 edge
   networks and also connects to the site via an IPv4 interface with
   address 192.0.2.3, but it does not add the IPv4 address to the site's
   PRL.  'C' next configures a non-advertising ISATAP router interface
   with link-local address fe80::5efe:192.0.2.3, then receives the IPv6
   prefix 2001:db8:0::/48 through a DHCPv6 prefix delegation exchange
   via one of 'A' or 'B'.  'C' then engages in an IPv6 routing protocol
   over its ISATAP interface and announces the delegated IPv6 prefix.
   'C' finally sub-delegates the prefix to its attached edge networks,
   where IPv6 host 'D' autoconfigures the address 2001:db8:0::1.

   Non-advertising ISATAP router 'E' connects to the site, configures
   its ISATAP interface, receives a DHCPv6 prefix delegation, and
   engages in the IPv6 routing protocol the same as for router 'C'.  In
   particular, 'E' configures the IPv4 address 192.0.2.4, the ISATAP
   link-local address fe80::5efe:192.0.2.4, and the delegated IPv6
   prefix 2001:db8:1::/48.  'E' finally sub-delegates the prefix to its
   attached edge networks, where IPv6 host 'F' autoconfigures IPv6
   address 2001:db8:1::1.

   ISATAP host 'G' connects to the site via an IPv4 interface with
   address 192.0.2.5, and also configures an ISATAP host interface with
   link-local address fe80::5efe:192.0.2.5 over the IPv4 interface.  'G'
   next configures a default IPv6 route with next-hop address
   fe80::5efe:192.0.2.2 via the ISATAP interface, then receives the IPv6
   address 2001:db8:2::1 from a DHCPv6 address configuration exchange
   via 'B'.  When 'G' receives the IPv6 address, it assigns the address
   to the ISATAP interface but does not assign a non-link-local IPv6
   prefix to the interface.

   Finally, IPv6 host 'H' connects to an IPv6 network outside of the
   ISATAP domain.  'H' configures its IPv6 interface in a manner
   specific to its attached IPv6 link, and autoconfigures the IPv6
   address 2001:db8:3::1.

   Following this autoconfiguration, when host 'D' has an IPv6 packet to
   send to host 'F', it prepares the packet with source address
   2001:db8:0::1 and destination address 2001:db8:1::1, then sends the
   packet into the edge network where it will eventually be forwarded to
   router 'C'.  'C' then uses ISATAP encapsulation to forward the packet
   to router 'E', since it has discovered a route to 2001:db8:1::/48
   with next hop 'E' via dynamic routing over the ISATAP interface.
   Router 'E' finally forwards the packet to host 'F'.

   In a second scenario, when 'D' has a packet to send to ISATAP host
   'G', it prepares the packet with source address 2001:db8:0::1 and
   destination address 2001:db8:2::1, then sends the packet into the
   edge network where it will eventually be forwarded to router 'C' the
   same as above.  'C' then uses ISATAP encapsulation to forward the
   packet to router 'A' (i.e., a router that advertises "default"),
   which in turn forwards the packet to 'G'.  Note that this operation
   entails two hops across the ISATAP link (i.e., one from 'C' to 'A',
   and a second from 'A' to 'G').  If 'G' also participates in the
   dynamic IPv6 routing protocol, however, 'C' could instead forward the
   packet directly to 'G' without involving 'A'.

   In a third scenario, when 'D' has a packet to send to host 'H' in the
   IPv6 Internet, the packet is forwarded to 'C' the same as above.  'C'
   then forwards the packet to 'A', which forwards the packet into the
   IPv6 Internet.

   In a final scenario, when 'G' has a packet to send to host 'H' in the
   IPv6 Internet, the packet is forwarded directly to 'B', which
   forwards the packet into the IPv6 Internet.

3.2.4.6.  Scaling Considerations

   Figure 2 depicts an ISATAP network topology with only two advertising
   ISATAP routers within the provider network.  In order to support
   larger numbers of non-advertising ISATAP routers and ISATAP hosts,
   the provider network can deploy more advertising ISATAP routers to
   support load balancing and generally shortest-path routing.

   Such an arrangement requires that the advertising ISATAP routers
   participate in an IPv6 routing protocol instance so that IPv6
   address/prefix delegations can be mapped to the correct router.  The
   routing protocol instance can be configured as either a full mesh
   topology involving all advertising ISATAP routers, or as a partial
   mesh topology with each advertising ISATAP router associating with
   one or more companion gateways.  Each such companion gateway would in
   turn participate in a full mesh between all companion gateways.

3.2.4.7.  On-Demand Dynamic Routing

   With respect to the reference operational scenario depicted in
   Figure 2, there will be many use cases in which a proactive dynamic
   IPv6 routing protocol cannot be used.  For example, in large
   enterprise network deployments it would be impractical for all
   routers to engage in a common routing protocol instance, due to
   scaling considerations.

   In those cases, an on-demand routing capability can be enabled in
   which ISATAP nodes send initial packets via an advertising ISATAP
   router and receive redirection messages back.  For example, when a
   non-advertising ISATAP router 'B' has a packet to send to a host
   located behind non-advertising ISATAP router 'D', it can send the
   initial packets via advertising router 'A', which will return
   redirection messages to inform 'B' that 'D' is a better first hop.
   Protocol details for this ISATAP redirection are specified in [AERO].

3.3.  Destination and Source Address Checks

   Tunnel routers can use a source address check mitigation measure when
   they forward an IPv6 packet into a tunnel interface with an IPv6
   source address that embeds one of the router's configured IPv4
   addresses.  Similarly, tunnel routers can use a destination address
   check mitigation measure when they receive an IPv6 packet on a tunnel
   interface with an IPv6 destination address that embeds one of the
   router's configured IPv4 addresses.  These checks should correspond
   to both tunnels' IPv6 address formats, regardless of the type of
   tunnel the router employs.

   For example, if tunnel router R1 (of any tunnel protocol) forwards a
   packet into a tunnel interface with an IPv6 source address that
   matches the 6to4 prefix 2002:IP1::/48, the router discards the packet
   if IP1 is one of its own IPv4 addresses.  In a second example, if
   tunnel router R2 receives an IPv6 packet on a tunnel interface with
   an IPv6 destination address with an off-link prefix but with an
   interface identifier that matches the ISATAP address suffix
   ::0200:5efe:IP2, the router discards the packet if IP2 is one of its
   own IPv4 addresses.

   Hence, a tunnel router can avoid the attack by performing the
   following checks:

   o  When the router forwards an IPv6 packet into a tunnel interface,
      it discards the packet if the IPv6 source address has an off-link
      prefix but embeds one of the router's configured IPv4 addresses.

   o  When the router receives an IPv6 packet on a tunnel interface, it
      discards the packet if the IPv6 destination address has an off-
      link prefix but embeds one of the router's configured IPv4
      addresses.

   This approach has the advantage that no ancillary state is required,
   since checking is through static lookup in the lists of IPv4 and IPv6
   addresses belonging to the router.  However, this approach has some
   inherent limitations:

   o  The checks incur an overhead that is proportional to the number of
      IPv4 addresses assigned to the router.  If a router is assigned
      many addresses, the additional processing overhead for each packet
      may be considerable.  Note that an unmitigated attack packet would
      be repetitively processed by the router until the Hop Limit
      expires, which may require as many as 255 iterations.  Hence, an
      unmitigated attack will consume far more aggregate processing
      overhead than per-packet address checks even if the router assigns
      a large number of addresses.

   o  The checks should be performed for the IPv6 address formats of
      every existing automatic IPv6 tunnel protocol (that uses
      protocol-41 encapsulation).  Hence, the checks must be updated as
      new protocols are defined.

   o  Before the checks can be performed, the format of the address must
      be recognized.  There is no guarantee that this can be generally
      done.  For example, one cannot determine if an IPv6 address is a
      6rd one; hence, the router would need to be configured with a list
      of all applicable 6rd prefixes (which may be prohibitively large)
      in order to unambiguously apply the checks.

   o  The checks cannot be performed if the embedded IPv4 address is a
      private one [RFC1918], since it is ambiguous in scope.  Namely,
      the private address may be legitimately allocated to another node
      in another routing region.

   The last limitation may be relieved if the router has some
   information that allows it to unambiguously determine the scope of
   the address.  The check in the following subsection is one example
   for this.

3.3.1.  Known IPv6 Prefix Check

   A router may be configured with the full list of IPv6 subnet prefixes
   assigned to the tunnels attached to its current IPv4 routing region.
   In such a case, it can use the list to determine when static
   destination and source address checks are possible.  By keeping track
   of the list of IPv6 prefixes assigned to the tunnels in the IPv4
   routing region, a router can perform the following checks on an
   address that embeds a private IPv4 address:

   o  When the router forwards an IPv6 packet into its tunnel with a
      source address that embeds a private IPv4 address and matches an
      IPv6 prefix in the prefix list, it determines whether the packet
      should be discarded or forwarded by performing the source address
      check specified in Section 3.3.

   o  When the router receives an IPv6 packet on its tunnel interface
      with a destination address that embeds a private IPv4 address and
      matches an IPv6 prefix in the prefix list, it determines whether
      the packet should be discarded or forwarded by performing the
      destination address check specified in Section 3.3.

   The disadvantage of this approach is that the administrative overhead
   for maintaining the list of IPv6 subnet prefixes associated with an
   IPv4 routing region may become unwieldy should that list be long
   and/or frequently updated.

4.  Recommendations

   In light of the mitigation measures proposed above, we make the
   following recommendations in decreasing order of importance:

   1.  When possible, it is recommended that the attacks be
       operationally eliminated (as per the measures proposed in
       Section 3.2).

   2.  For tunnel routers that keep a coherent and trusted neighbor
       cache that includes all legitimate endpoints of the tunnel, we
       recommend exercising the neighbor cache check.

   3.  For tunnel routers that can implement the Neighbor Reachability
       Check, we recommend exercising it.

   4.  For tunnels having a small and static list of endpoints, we
       recommend exercising the known IPv4 address check.

   5.  We generally do not recommend using the destination and source
       address checks, since they cannot mitigate routing loops with 6rd
       routers.  Therefore, these checks should not be used alone unless
       there is operational assurance that other measures are exercised
       to prevent routing loops with 6rd routers.

   As noted earlier, tunnels may be deployed in various operational
   environments.  There is a possibility that other mitigation measures
   may be feasible in specific deployment scenarios.  The above
   recommendations are general and do not attempt to cover such
   scenarios.

5.  Security Considerations

   This document aims at presenting possible solutions to the routing
   loop attack that involves automatic tunnels' routers.  It contains
   various checks that aim to recognize and drop specific packets that
   have strong potential to cause a routing loop.  These checks do not
   introduce new security threats.

6.  Acknowledgments

   This work has benefited from discussions on the V6OPS, 6MAN, and
   SECDIR mailing lists.  The document has further benefited from
   comments received from members of the IESG during their review.
   Dmitry Anipko, Fred Baker, Stewart Bryant, Remi Despres, Adrian
   Farrell, Fernando Gont, Christian Huitema, Joel Jaeggli, and Dave
   Thaler are acknowledged for their contributions.

7.  References

7.1.  Normative References

   [RFC1918]  Rekhter, Y., Moskowitz, B., Karrenberg, D., de Groot, G.,
              and E. Lear, "Address Allocation for Private Internets",
              BCP 5, RFC 1918, February 1996.

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

   [RFC3315]  Droms, R., Ed., Bound, J., Volz, B., Lemon, T., Perkins,
              C., and M. Carney, "Dynamic Host Configuration Protocol
              for IPv6 (DHCPv6)", RFC 3315, July 2003.

   [RFC3633]  Troan, O. and R. Droms, "IPv6 Prefix Options for Dynamic
              Host Configuration Protocol (DHCP) version 6", RFC 3633,
              December 2003.

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

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

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

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

7.2.  Informative References

   [AERO]     Templin, F., Ed., "Asymmetric Extended Route Optimization
              (AERO)", Work in Progress, June 2011.

   [ISATAP-OPS]
              Templin, F., "Operational Guidance for IPv6 Deployment in
              IPv4 Sites using ISATAP", Work in Progress, July 2011.

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

   [RFC4732]  Handley, M., Ed., Rescorla, E., Ed., and IAB, "Internet
              Denial-of-Service Considerations", RFC 4732,
              December 2006.

   [RFC6343]  Carpenter, B., "Advisory Guidelines for 6to4 Deployment",
              RFC 6343, August 2011.

   [TEREDO-LOOPS]
              Gont, F., "Mitigating Teredo Rooting Loop Attacks", Work
              in Progress, September 2010.

   [USENIX09] Nakibly, G. and M. Arov, "Routing Loop Attacks using IPv6
              Tunnels", USENIX WOOT, August 2009.

Authors' Addresses

   Gabi Nakibly
   National EW Research & Simulation Center
   Rafael - Advanced Defense Systems
   P.O. Box 2250 (630)
   Haifa  31021
   Israel

   EMail: gnakibly@yahoo.com


   Fred L. Templin
   Boeing Research & Technology
   P.O. Box 3707 MC 7L-49
   Seattle, WA  98124
   USA

   EMail: fltemplin@acm.org