Network Working Group                                        S. Krishnan
Request for Comments: 5722                                      Ericsson
Updates: 2460                                              December 2009
Category: Standards Track


                 Handling of Overlapping IPv6 Fragments

Abstract

   The fragmentation and reassembly algorithm specified in the base IPv6
   specification allows fragments to overlap.  This document
   demonstrates the security issues associated with allowing overlapping
   fragments and updates the IPv6 specification to explicitly forbid
   overlapping fragments.

Status of This Memo

   This document specifies an Internet standards track protocol for the
   Internet community, and requests discussion and suggestions for
   improvements.  Please refer to the current edition of the "Internet
   Official Protocol Standards" (STD 1) for the standardization state
   and status of this protocol.  Distribution of this memo is unlimited.

Copyright Notice

   Copyright (c) 2009 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 BSD License.

Table of Contents

   1. Introduction ....................................................2
      1.1. Conventions Used in This Document ..........................2
   2. Overlapping Fragments ...........................................2
   3. The Attack ......................................................3
   4. Node Behavior ...................................................5
   5. Security Considerations .........................................5
   6. Acknowledgements ................................................5
   7. References ......................................................6
      7.1. Normative References .......................................6
      7.2. Informative References .....................................6

1.  Introduction

   Fragmentation is used in IPv6 when the IPv6 packet will not fit
   inside the path MTU to its destination.  When fragmentation is
   performed, an IPv6 node uses a fragment header, as specified in
   Section 4.5 of the IPv6 base specification [RFC2460], to break down
   the datagram into smaller fragments that will fit in the path MTU.
   The destination node receives these fragments and reassembles them.
   The algorithm specified for fragmentation in [RFC2460] does not
   prevent the fragments from overlapping, and this can lead to some
   security issues with firewalls [RFC4942].  This document explores the
   issues that can be caused by overlapping fragments and updates the
   IPv6 specification to explicitly forbid overlapping fragments.

1.1.  Conventions Used in This Document

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in [RFC2119].

2.  Overlapping Fragments

   Commonly used firewalls use the algorithm specified in [RFC1858] to
   weed out malicious packets that try to overwrite parts of the
   transport-layer header in order to bypass inbound connection checks.
   [RFC1858] prevents an overlapping fragment attack on an upper-layer
   protocol (in this case, TCP) by recommending that packets with a
   fragment offset of 1 be dropped.  While this works well for IPv4
   fragments, it will not work for IPv6 fragments.  This is because the
   fragmentable part of the IPv6 packet can contain extension headers
   before the TCP header, making this check less effective.

3.  The Attack

   This attack describes how a malicious node can bypass a firewall
   using overlapping fragments.  Consider a sufficiently large IPv6
   packet that needs to be fragmented.

   +------------------+--------------------//-----------------------+
   |  Unfragmentable  |                 Fragmentable                |
   |       Part       |                     Part                    |
   +------------------+--------------------//-----------------------+

                     Figure 1: Large IPv6 Packet

   This packet is split into several fragments by the sender so that the
   packet can fit inside the path MTU.  Let's say the packet is split
   into two fragments.

   +------------------+--------+--------------------+
   |  Unfragmentable  |Fragment|       first        |
   |       Part       | Header |      fragment      |
   +------------------+--------+--------------------+

   +------------------+--------+--------------------+
   |  Unfragmentable  |Fragment|       second       |
   |       Part       | Header |      fragment      |
   +------------------+--------+--------------------+

           Figure 2: Fragmented IPv6 Packet

   Consider the first fragment.  Let's say it contains a destination
   options header (DOH) 80 octets long and is followed by a TCP header.

 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<==FH
 |NextHdr=DOH(60)|   Reserved    |   FragmentOffset = 0    |Res|1|
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 |                Identification=aaaabbbb                        |
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<==DOH
 |NextHdr=TCP(6) | HdrExtLen = 9 |                               |
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+                               +
 |                                                               |
 .                                                               .
 .                            Options                            .
 .                                                               .
 |                                                               |
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<==TCP
 |        Source Port            |       Destination Port        |
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 |                       Sequence Number                         |
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 |                    Acknowledgment Number                      |
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 | Offset| Reserved  |U|A|P|R|S|F|           Window              |
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

                      Figure 3: First Fragment

   The TCP header has the following values of the flags: S(YN)=1 and
   A(CK)=1.  This may make an inspecting stateful firewall think that it
   is a response packet for a connection request initiated from the
   trusted side of the firewall.  Hence, it will allow the fragment to
   pass.  It will also allow the following fragments with the same
   Fragment Identification value in the fragment header to pass through.

   A malicious node can form a second fragment with a TCP header that
   changes the flags and sets S(YN)=1 and A(CK)=0.  This can change the
   packet on the receiving end to consider the packet as a connection
   request instead of a response.  By doing this, the malicious node has
   bypassed the firewall's access control to initiate a connection
   request to a node protected by a firewall.

+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<==FH
|NextHdr=DOH(60)|   Reserved    |   FragmentOffset = 10   |Res|0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                Identification=aaaabbbb                        |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+<==TCP
|        Source Port            |       Destination Port        |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                       Sequence Number                         |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                    Acknowledgment Number                      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Offset| Reserved  |U|A|P|R|S|F|           Window              |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

                    Figure 4: Second Fragment

   Note that this attack is much more serious in IPv6 than in IPv4.  In
   IPv4, the overlapping part of the TCP header does not include the
   source and destination ports.  In IPv6, the attack can easily work to
   replace the source or destination port with an overlapping fragment.

4.  Node Behavior

      IPv6 nodes transmitting datagrams that need to be fragmented MUST NOT Expand
   create overlapping fragments.  When reassembling an IPv6 datagram, if
   one or more its constituent fragments is determined to be an
   overlapping fragment, the entire datagram (and any constituent
   fragments) MUST be silently discarded.

EID 3089 (Verified) is as follows:

Section: 4

Original Text:

   IPv6 nodes transmitting datagrams that need to be fragmented MUST NOT
   create overlapping fragments.  When reassembling an IPv6 datagram, if
   one or more its constituent fragments is determined to be an
   overlapping fragment, the entire datagram (and any constituent
   fragments, including those not yet received) MUST be silently
   discarded.

Corrected Text:

   IPv6 nodes transmitting datagrams that need to be fragmented MUST NOT
   create overlapping fragments.  When reassembling an IPv6 datagram, if
   one or more its constituent fragments is determined to be an
   overlapping fragment, the entire datagram (and any constituent
   fragments) MUST be silently discarded.

Notes:

Discarding fragments "including those not yet received" is not implementable. You'd have to keep state about the (source, destination, protocol, id) 4-tuple for MSL (120 seconds). If you do this you create two bugs:
- A new attack vector: an attacker could eat your resources. And if you just limit the number of such state entries then you fail to implement RFC 5722 correctly.
- It breaks at fairly low speeds. See draft-ietf-intarea-ipv4-id-update.

The proposal is simply to remove the "including those not yet received" bit. Normal host stacks do not keep state once a fragment has been reassembled. You reassemble the full packet and clear the fragment table. So this corrected text would align the RFC with actual practice.

This errata report results from an implementation attempt by OpenBSD.


   Nodes MAY also provide mechanisms to track the reception of such
   packets, for instance, by implementing counters or alarms relating to
   these events.

5.  Security Considerations

   This document discusses an attack that can be used to bypass IPv6
   firewalls using overlapping fragments.  It recommends disallowing
   overlapping fragments in order to prevent this attack.

6.  Acknowledgements

   The author would like to thank Thomas Narten, Doug Montgomery,
   Gabriel Montenegro, Remi Denis-Courmont, Marla Azinger, Arnaud
   Ebalard, Seiichi Kawamura, Behcet Sarikaya, Vishwas Manral, Christian
   Vogt, Bob Hinden, Carl Wallace, Jari Arkko, Pasi Eronen, Francis

   Dupont, Neville Brownlee, Dan Romascanu, Lars Eggert, Cullen
   Jennings, and Alfred Hoenes for their reviews and suggestions that
   made this document better.

7.  References

7.1.  Normative References

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

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

7.2.  Informative References

   [RFC1858]  Ziemba, G., Reed, D., and P. Traina, "Security
              Considerations for IP Fragment Filtering", RFC 1858,
              October 1995.

   [RFC4942]  Davies, E., Krishnan, S., and P. Savola, "IPv6
              Transition/Co-existence Security Considerations", RFC
              4942, September 2007.

Author's Address

   Suresh Krishnan
   Ericsson
   8400 Blvd Decarie
   Town of Mount Royal, Quebec
   Canada

   EMail: suresh.krishnan@ericsson.com