Tech-invite3GPPspaceIETFspace
96959493929190898887868584838281807978777675747372717069686766656463626160595857565554535251504948474645444342414039383736353433323130292827262524232221201918171615141312111009080706050403020100
in Index   Prev   Next

RFC 4953

Defending TCP Against Spoofing Attacks

Pages: 28
Informational

Top   ToC   RFC4953 - Page 1
Network Working Group                                           J. Touch
Request for Comments: 4953                                       USC/ISI
Category: Informational                                        July 2007


                 Defending TCP Against Spoofing Attacks

Status of This Memo

   This memo provides information for the Internet community.  It does
   not specify an Internet standard of any kind.  Distribution of this
   memo is unlimited.

Copyright Notice

   Copyright (C) The IETF Trust (2007).

Abstract

Recent analysis of potential attacks on core Internet infrastructure indicates an increased vulnerability of TCP connections to spurious resets (RSTs), sent with forged IP source addresses (spoofing). TCP has always been susceptible to such RST spoofing attacks, which were indirectly protected by checking that the RST sequence number was inside the current receive window, as well as via the obfuscation of TCP endpoint and port numbers. For pairs of well-known endpoints often over predictable port pairs, such as BGP or between web servers and well-known large-scale caches, increases in the path bandwidth- delay product of a connection have sufficiently increased the receive window space that off-path third parties can brute-force generate a viable RST sequence number. The susceptibility to attack increases with the square of the bandwidth, and thus presents a significant vulnerability for recent high-speed networks. This document addresses this vulnerability, discussing proposed solutions at the transport level and their inherent challenges, as well as existing network level solutions and the feasibility of their deployment. This document focuses on vulnerabilities due to spoofed TCP segments, and includes a discussion of related ICMP spoofing attacks on TCP connections.
Top   ToC   RFC4953 - Page 2

Table of Contents

1. Introduction ....................................................3 2. Background ......................................................4 2.1. Review of TCP Windows ......................................5 2.2. Recent BGP Attacks Using TCP RSTs ..........................6 2.3. TCP RST Vulnerability ......................................6 2.4. What Changed - the Ever-Opening Advertised Receive Window ..7 3. Proposed Solutions and Mitigations .............................10 3.1. Transport Layer Solutions .................................10 3.1.1. TCP MD5 Authentication .............................11 3.1.2. TCP RST Window Attenuation .........................11 3.1.3. TCP Timestamp Authentication .......................12 3.1.4. Other TCP Cookies ..................................13 3.1.5. Other TCP Considerations ...........................13 3.1.6. Other Transport Protocol Solutions .................14 3.2. Network Layer (IP) Solutions ..............................14 3.2.1. Address Filtering ..................................15 3.2.2. IPsec ..............................................16 4. ICMP ...........................................................17 5. Issues .........................................................18 5.1. Transport Layer (e.g., TCP) ...............................18 5.2. Network Layer (IP) ........................................19 5.3. Application Layer .........................................21 5.4. Link Layer ................................................21 5.5. Issues Discussion .........................................21 6. Security Considerations ........................................22 7. Conclusions ....................................................23 8. Acknowledgments ................................................23 9. Informative References .........................................24
Top   ToC   RFC4953 - Page 3

1. Introduction

Analysis of the Internet infrastructure has recently demonstrated a new version of a vulnerability in BGP connections between core routers using an attack based on RST spoofing from off-path attackers [9][10][48]. The attack itself is not new, having been documented nearly six years earlier [20]. Such connections, typically using TCP, can be susceptible to off-path third-party reset (RST) segments with forged source addresses (spoofed), which terminate the TCP connection. BGP routers react to a terminated TCP connection in various ways, which can amplify the impact of an attack, ranging from restarting the connection to deciding that the other router is unreachable and thus flushing the BGP routes [37]. This sort of attack affects other protocols besides BGP, involving any long-lived connection between well-known endpoints. The impact on the Internet infrastructure can be substantial (especially for the BGP case), and warrants immediate attention. TCP, like many other protocols, can be susceptible to these off-path third-party spoofing attacks. Such attacks rely on the increase of commodity platforms supporting public access to previously privileged resources, such as system-level (i.e., root) access. Given such access, it is trivial for anyone to generate a packet with any header desired. This, coupled with the lack of sufficient address filtering to drop such spoofed traffic, can increase the potential for off-path third- party spoofing attacks [9][10][48]. Proposed solutions include the deployment of existing Internet network and transport security as well as modifications to transport protocols that reduce its vulnerability to generated attacks [13][15][20][36][46]. One way to defeat spoofing is to validate the segments of a connection, either at the transport level or the network level. TCP with MD5 extensions provides this authentication at the transport level, and IPsec provides authentication at the network level [20][24][27]. In both cases, their deployment overhead may be prohibitive, e.g., it may not be feasible for public services, such as web servers, to be configured with the appropriate certificate authorities of large numbers of peers (for IPsec using the Internet Key Exchange Protocol (IKE)), or shared secrets (for IPsec in shared-secret mode, or TCP/MD5), because many clients may need to be configured rapidly without external assistance. Services located on public web servers connecting to large-scale caches or BGP with larger numbers of peers can fall into this category.
Top   ToC   RFC4953 - Page 4
   The remainder of this document outlines the recent attack scenario in
   detail and describes and compares a variety of solutions, including
   existing solutions based on TCP/MD5 and IPsec, as well as recently
   proposed solutions, including modifications to TCP's RST processing
   [36], modifications to TCP's timestamp processing [34], and
   modifications to IPsec and TCP/MD5 keying [45].  This document
   focuses on spoofing of TCP segments, although a discussion of related
   spoofing of ICMP packets based on spoofed TCP contents is also
   discussed.

   Note that the description of these attacks is not new; attacks using
   RSTs on BGP have been known since 1998, and were the reason for the
   development of TCP/MD5 [20].  The recent attack scenario was first
   documented by Convery at a NANOG (North American Network Operators'
   Group) meeting in 2003, but that analysis assumed the entire sequence
   space (2^32 packets) needed to be covered for an attack to succeed
   [10].  Watson's more detailed analysis discovered that a single
   packet anywhere in the current window could succeed at an attack
   [48].  This document adds the observation that susceptibility to
   attack is directly proportional to the square of bandwidth, due to
   the coupling between the linear increase in receive window size and
   linear increase in rate of a potential attack, as well as comparing
   the variety of more recent proposals, including modifications to TCP,
   use of IPsec, and use of TCP/MD5 to resist such attacks.

2. Background

The recent analysis of potential attacks on BGP has again raised the issue of TCP's vulnerability to off-path third-party spoofing attacks [9][10][48]. A variety of such attacks have been known for several years, including sending RSTs, SYNs, and even ACKs in an attempt to affect an existing connection or to load down servers. These attacks often combine external knowledge (e.g., to indicate the IP addresses to attack, the destination port number, and sometimes the Initial Sequence Number (ISN)) with brute-force capabilities enabled by modern computers and network bandwidths (e.g., to scan all source ports or an entire window space). Overall, such attacks are countered by the use of some form of authentication at the network (e.g., IPsec), transport (e.g., SYN cookies, TCP/MD5), or other layers. TCP already includes a weak form of such authentication in its check of segment sequence numbers against the current receiver window. Increases in the bandwidth-delay product for certain long connections have sufficiently weakened this type of weak authentication to make reliance on it inadvisable.
Top   ToC   RFC4953 - Page 5

2.1. Review of TCP Windows

Before proceeding, it is useful to review the terminology and components of TCP's windowing algorithm. TCP connections have three kinds of windows [1][35]: o Send window (SND.WND): the latest send window size. o Receive window (RCV.WND): the latest advertised receive window size. o Congestion window (CWND): the window determined by congestion feedback that limits how much of RCV.WND can be in-flight in a round-trip time. For TCP connections in most modern implementations, SND.WND and RCV.WND are the size of the corresponding send and receive socket buffers, and are configurable using socket buffer resizing commands. CWND determines how much data can be in transit in a round-trip time, SND.WND determines how much data the sender is willing to store on its side for possible retransmission due to loss, and RCV.WND determines the ability of the receiver to accommodate that loss and reorder received packets. CWND never grows beyond RCV.WND. High bandwidth-delay product networks need CWND to be sufficiently large to accommodate as much data as can be in transit in a round trip time; otherwise, their performance will suffer. As a result, it is recommended that users and various automatic programs increase RCV.WND to at least the size of bandwidth*delay (the bandwidth-delay product) [23][38]. As the bandwidth-delay product of the network increases, however, such increases in the advertised receive window can cause increased susceptibility to spoofing attacks, as the remainder of this document shows. This assumes, however, that the receive window size (e.g., via increased receive socket buffer configuration) is increased with the increased bandwidth-delay product; if not, then connection performance will degrade, but susceptibility to spoofing attacks will increase only linearly (with the rate at which the attacker can send spoofed packets), not as the square of the bandwidth. Note that either increase depends on the receive window itself, and is independent of the congestion state or amount of data transmitted.
Top   ToC   RFC4953 - Page 6

2.2. Recent BGP Attacks Using TCP RSTs

BGP represents a particular vulnerability to spoofing attacks because it uses TCP connectivity to infer routability, so losing a TCP connection with a BGP peer can result in the flushing of routes to that peer [37]. Until six years ago, such connections were assumed difficult to attack because they were described by a few comparatively obscure parameters [20]. Most TCP connections are protected by multiple levels of obfuscation except at the endpoints of the connection: o Both endpoint addresses are usually not well-known; although server addresses are advertised, clients are somewhat anonymous. o Both port numbers are usually not well-known; the server's is usually advertised (representing the service), but the client's is typically sufficiently unpredictable to an off-path third-party. o Valid sequence number space is not well-known. o Connections are relatively short-lived and valid sequence space changes, so any attempt to guess (e.g., by external knowledge or brute force) the above information is unlikely to be useful. BGP represents an exception to the above criteria (though not the only case). Both endpoints can be well-known, or guessed using hints from part of an AS path. The destination port is typically fixed to indicate the BGP service. The source port used by a BGP router is sometimes fixed and advertised to enable firewall configuration; even when not fixed, there are only approximately 65,000 valid source ports, which thus may be exhaustively attacked. Connections are long- lived, and, as noted before, some BGP implementations interpret successive TCP connection failures as routing failures, discarding the corresponding routing information. In addition, the valid sequence number space once thought to provide some protection has been significantly weakened by increasing advertised receive window sizes.

2.3. TCP RST Vulnerability

TCP has a known vulnerability to third-party spoofed segments. SYN flooding consumes server resources in half-open connections, affecting the server's ability to open new connections [4][11]. ACK spoofing can cause connections to transmit too much data too quickly, creating network congestion and segment loss, causing connections to slow to a crawl. In the most recent attacks on BGP, RSTs cause connections to be dropped. As noted earlier, some BGP
Top   ToC   RFC4953 - Page 7
   implementations interpret TCP connection termination, or a series of
   such failures, as a network failure [37].  This causes routers to
   drop the BGP routing information already exchanged, in addition to
   inhibiting their ongoing exchanges, thus amplifying the impact of the
   attack.  The result can affect routing paths throughout the Internet.

   The dangerous effects of RSTs on TCP have been known for many years,
   even when used by the legitimate endpoints of a connection.  TCP RSTs
   cause the receiver to drop all connection state; because the source
   is not required to maintain a TIME_WAIT state, such a RST can cause
   premature reuse of address/port pairs, potentially allowing segments
   from a previous connection to contaminate the data of a new
   connection, known as TIME_WAIT assassination [8].  In this case,
   assassination occurs inadvertently as the result of duplicate
   segments from a legitimate source, and can be avoided by blocking RST
   processing while in TIME_WAIT.  However, assassination can be useful
   to deliberately reduce the state held at servers; this requires that
   the source of the RSTs go into TIME_WAIT state to avoid such hazards,
   and that RSTs are not blocked in the TIME_WAIT state [12].

   Firewalls and load balancers, so-called 'middleboxes', sometimes emit
   RSTs on behalf of transited connections to optimize server
   performance, as noted in RFC 3360 [14].  This is effectively an on-
   path RST attack in which the RSTs are sent for benign or beneficial
   intent.  There are numerous hazards with such use of RSTs, outlined
   in that RFC.

2.4. What Changed - the Ever-Opening Advertised Receive Window

RSTs represent a hazard to TCP, especially when completely unvalidated. Fortunately, there are a number of obfuscation mechanisms that make it difficult for off-path third parties to forge (spoof) valid RSTs, as noted earlier. We have already shown it is easy to learn both endpoint addresses and ports for some protocols, notably BGP. The final obfuscation is the segment sequence number. TCP segments include a sequence number, which enables out-of-order receiver processing as well as duplicate detection. The sequence number space is also used to manage congestion, and indicates the index of the next byte to be transmitted or received. For RSTs, this is relevant because legitimate RSTs use the next sequence number in the transmitter window, and the receiver checks that incoming RSTs have a sequence number in the expected receive window. Such processing is intended to eliminate duplicate segments (somewhat moot for RSTs, though), and to drop RSTs that were part of previous connections.
Top   ToC   RFC4953 - Page 8
   TCP uses two window mechanisms, a primary mechanism for reordering
   and congestion control (which uses a space of 32 bits), and a
   secondary mechanism that scales this window [23][35].  The valid
   advertised receive window is a fraction, not to exceed approximately
   half, of this space, or ~2 billion (2 * 10^9, i.e., 2E9 or 2 U.S.
   billion).  Under typical configurations, the majority of TCP
   connections open to a very small fraction of this space, e.g.,
   10,000-60,000(approximately 5-100 segments).  This is because the
   advertised receive window typically matches the receive socket buffer
   size.  It is recommended that this buffer be tuned to match the needs
   of the connection, either manually or by automatic external means
   [38].

   On a low-loss path, the advertised receive window should be
   configured to match the path bandwidth-delay product, including
   buffering delays (assume 1 packet/hop) [38].  Many paths in the
   Internet have end-to-end bandwidths of under 1 Mbps, latencies under
   100 ms, and are under 15 hops, resulting in fairly small advertised
   receive windows as above (under 35,000 bytes).  Under these
   conditions, and further assuming that the initial sequence number is
   suitably (pseudo-randomly) chosen, a valid guessed sequence number
   would have odds of 1 in 57,000 of falling within the advertised
   receive window.  Put differently, a blind (i.e., off-path) attacker
   would need to send 57,000 RSTs with suitably spaced sequence number
   guesses within one round-trip time to successfully reset a
   connection.  At 1 Mbps, 57,000 (40 byte) RSTs would take only 20
   seconds to transmit, but this presumes that both IP addresses and
   both ports are known.  Absent knowledge of the source port, an off-
   path spoofer would need to try at least the entire range of 49152-
   65535, or 16,384 different ports, resulting in an attack that would
   take over 91 hours.  Because most TCP connections are comparatively
   short-lived, even this moderate variation in the source port is
   sufficient for such environments, although further port randomization
   may be recommended [29].

   Recent use of high bandwidth paths of 10 Gbps and higher results in
   bandwidth-delay products over 125 MB -- approximately 1/10 of TCP's
   overall maximum advertised receive window size (i.e., assuming the
   receive socket buffers are increased as much as possible) excluding
   scale, assuming the receiver allocates sufficient buffering (as
   discussed in Section 2).  Even under networks that are ten times
   slower (1 Gbps), the active advertised receive window covers 1/100th
   of the overall window size.  At these speeds, it takes only 10-100
   packets, or less than 32 microseconds, to correctly guess a valid
   sequence number and kill a connection.  A table of corresponding
   exposure to various amounts of RSTs is shown below, for various line
   rates, assuming the more conventional 100-ms latencies (though even
   100 ms is large for BGP cases):
Top   ToC   RFC4953 - Page 9
          BW       BW*delay     RSTs needed     Time needed
      ------------------------------------------------------------
       10 Gbps   125       MB          35     1 us (microsecond)
        1 Gbps    12.5     MB         344   110 us
      100 Mbps     1.25    MB       3,436    10 ms (millisecond)
       10 Mbps     0.125   MB      34,360     1 second
        1 Mbps     0.0125  MB     343,598     2 minutes
      100 Kbps     0.00125 MB   3,435,974     3 hours

                Figure 1: Time needed to kill a connection

   This table demonstrates that the effect of bandwidth on the
   vulnerability is squared; for every increase in bandwidth, there is a
   linear decrease in the number of sequence number guesses needed, as
   well as a linear decrease in the time needed to send a set of
   guesses.  Notably, as inter-router link bandwidths approach 1 Mbps,
   an 'exhaustive' attack becomes practical.  Checking that the RST
   sequence number is somewhere in the advertised receive window, out of
   the overall maximum receive window (2^32), is an insufficient
   obfuscation.

   Note that this table makes a number of assumptions:

   1. The overall bandwidth-delay product is relatively fixed.

   2. Traffic losses are negligible (insufficient to affect the
      congestion window over the duration of most of the connection).

   3. The advertised receive window is a large fraction of the overall
      maximum receive window size, e.g., because the receive socket
      buffers are set to match a large bandwidth-delay product.

   4. The attack bandwidth is similar to the end-to-end path bandwidth.

   Of these assumptions, the last two are more notable.  The issue of
   receive socket buffers was discussed in Section 2.  Figure 1
   summarized the time to a successful attack based on large advertised
   receive windows, but many current commercial routers have limits of
   128 KB for large devices, 32 KB for medium, and as little as 4 KB for
   modest ones.  Figure 2 shows the time and bandwidths needed to
   accomplish an attack on BGP sessions in the time shown for 100-ms
   latencies; for even short-range network latencies (10 ms), these
   sessions can be still be attacked over short timescales (minutes to
   hours).
Top   ToC   RFC4953 - Page 10
                   Receive
          BW     Buffer Size  RSTs needed     Time needed
      ------------------------------------------------------------
       10 Mbps     0.128 MB        33,555     1 second
        3 Mbps     0.032 MB       134,218    40 seconds
      300 Kbps     0.004 MB     1,073,742     1 hour

      Figure 2: Time needed to kill a connection with limited buffers

   The issue of the attack bandwidth is considered reasonable as
   follows:

   1. RSTs are substantially easier to send than data; they can be
      precomputed and they are smaller than data packets (40 bytes).

   2. Although susceptible connections use somewhat less ubiquitous
      high-bandwidth paths, the attack may be distributed, at which
      point only the ingress link of the attack is the primary
      limitation.

   3. For the purposes of the above table, we assume that the ingress at
      the attack has the same bandwidth as the path, as an
      approximation.

   The previous sections discussed the nature of the recent attacks on
   BGP due to the vulnerability of TCP to RST spoofing attacks, due
   largely to recent increases in the fraction of the TCP advertised
   receive window space in use for a single, long-lived connection.

3. Proposed Solutions and Mitigations

TCP currently authenticates received RSTs using the address and port pair numbers, and checks that the sequence number is inside the valid receiver window. The previous section demonstrated how TCP has become more vulnerable to RST spoofing attacks due to the increases in the receive window size. There are a number of current and proposed solutions to this vulnerability, all attempting to provide evidence that a received RST is legitimate.

3.1. Transport Layer Solutions

The transport layer represents the last place that segments can be authenticated before they affect connection management. TCP has a variety of current and proposed mechanisms to increase the authentication of segments, protecting against both off-path and on- path third-party spoofing attacks. Other transport protocols, such as SCTP and DCCP, also have limited antispoofing mechanisms.
Top   ToC   RFC4953 - Page 11

3.1.1. TCP MD5 Authentication

An extension to TCP supporting MD5 authentication was developed in 1998 specifically to authenticate BGP connections (although it can be used for any TCP connection) [20]. The extension relies on a pre- shared secret key to authenticate the entire TCP segment, including the data, TCP header, and TCP pseudo-header (certain fields of the IP header). All segments are protected, including RSTs, to be accepted only when their signature matches. This option, although widely deployed in Internet routers, is considered undeployable for widespread use because the need for pre-shared keys [3][30]. It further is considered computationally expensive for either hosts or routers due to the overhead of MD5 [43][44]. There are also concerns about the use of MD5 due to recent collision- based attacks [22]. Similar concerns exist for SHA-1, and the IETF is currently evaluating how these attacks impact the recommendation for using these hashes, both in TCP/MD5 and in the IPsec suite. For the purposes of this discussion, the particular algorithm used in either protocol suite is not the focus, and there is ongoing work to allow TCP/MD5 to evolve to a more general TCP security option [6][47].

3.1.2. TCP RST Window Attenuation

A recent proposal extends TCP to further constrain received RST to match the expected next sequence number [36]. This restores TCP's resistance to spurious RSTs, effectively limiting the receive window for RSTs to a single number. As a result, an attacker would need to send 2^32 different packets to brute-force guess the sequence number (worst case, the average would be half that); this makes TCP's vulnerability to attack independent of the size of the receive window (RCV.WND). The extension further modifies the RST receiver to react to incorrectly-numbered RSTs, by sending a zero-length ACK. If the RST source is legitimate, upon receipt of an ACK, the closed source would presumably emit a RST with the sequence number matching the ACK, correctly resetting the intended recipient. This modification changes TCP's control processing, adding to its complexity and thus potentially affecting its correctness (in contrast to adding MD5 signatures, which is orthogonal to TCP control processing altogether). For example, there may be complications between RSTs of different connections between the same pair of endpoints because RSTs flush the TIME-WAIT (as mentioned earlier). Further, this proposal modifies TCP so that, under some circumstances, a RST causes a reply (an ACK), in violation of generally accepted practice, if not gentle recommendation -- although this can be omitted, allowing timeouts to suffice. The advantage to this proposal is that it can be deployed incrementally and has benefit to the endpoint on which it is
Top   ToC   RFC4953 - Page 12
   deployed.  The other advantage to this proposal is that the window
   attenuation described here makes the vulnerability to spoofed RST
   packets independent of the size of the receive window.

   A variant of this proposal uses a different value to attenuate the
   window of viable RSTs.  It requires RSTs to carry the initial
   sequence number rather than the next expected sequence number, i.e.,
   the value negotiated on connection establishment [42][49].  This
   proposal has the advantage of using an explicitly negotiated value,
   but at the cost of changing the behavior of an unmodified endpoint to
   a currently valid RST.  It would thus be more difficult, without
   additional mechanism, to deploy incrementally.

   Another variant of this proposal involves increasing TCP's window
   space, rather than decreasing the valid range for RSTs, i.e.,
   increasing the sequence space from 32 bits to 64 bits.  This has the
   equivalent effect -- the ratio of the valid sequence numbers for any
   segment to the overall sequence number space is significantly
   reduced.  The use of the larger space, as with current schemes to
   establish weak authentication using initial sequence numbers (ISNs),
   is contingent on using suitably random values for the ISN.  Such
   randomness adds additional complexity to TCP both in specification
   and implementation, and provides only very weak authentication.  Such
   a modification is not obviously backward compatible, and would be
   thus difficult to deploy.

   A converse variant of increasing TCP's window space is to decrease
   the receive window (RCV.WND) explicitly, which would further reduce
   the effectiveness of spoofed RSTs with random sequence numbers.  This
   alternative may reduce the throughput of the connection, if the
   advertised receive window is smaller than the bandwidth-delay product
   of the connection.

3.1.3. TCP Timestamp Authentication

Another way to authenticate TCP segments is via its timestamp option, using the value as a sort of authentication [34]. This requires that the receiver TCP discard segments whose timestamp is outside the accepted window, which is derived from the timestamps of other packets from the same connection. This technique uses an existing TCP option, but also requires modified TCP control processing (with the same caveats) and may be difficult to deploy incrementally without further modifications. Additionally, the timestamp value may be easier to guess because it can be derived predictably, either assuming it represents actual time at the host, or by probing the host using unrelated benign traffic.
Top   ToC   RFC4953 - Page 13

3.1.4. Other TCP Cookies

All of the above techniques are variants of cookies, otherwise meaningless data whose value is used to validate the packet. In the case of MD5 checksums, the cookie is computed based on a shared secret. Note that even a signature can be guessed, and presents a 1 in 2^(signature length) probability of attack. The primary difference is that MD5 signatures are effectively one-time cookies, not predictable based on on-path snooping, because they are dependent on packet data and thus do not repeat. Window attenuation sequence numbers can be guessed by snooping the sequence number of current packets of an existing connection, and timestamps can be guessed even less directly, either by separate benign connections or by assuming they roughly correlate to local time. These variants of cookies are similar in spirit to TCP SYN cookies, again patching a vulnerability to off-path third-party spoofing attacks based on a (fairly weak, excepting MD5) form of authentication. Another form of cookie is the source port itself, which can be randomized but provides only 16 bits of protection (65,000 combinations), which may be exhaustively attacked. This can be combined with destination port randomization as well, but that would require a separate coordination mechanism (so both parties know which ports to use), which is equivalent to (and as infeasible for large-scale deployments as) exchanging a shared secret [39].

3.1.5. Other TCP Considerations

The analysis of the potential for RST spoofing above assumes that the advertised receive window is opened to the maximum extent suggested by the bandwidth-delay product of the end-to-end path, and that the window is opened to an appreciable fraction of the overall sequence number space. As noted earlier, for most common cases, connections are too brief or over bandwidths too low for such a large window to be useful. Expanding TCP's sequence number space is a direct way to further avoid such vulnerability, even for long connections over emerging bandwidths. If either manual tuning or automatic tuning of the advertised receive window (via receive buffer tuning) is not provided, this is not an issue (although connection performance will suffer) [38]. It may be sufficient for the endpoint to limit the advertised receive window by deliberately leaving it small. If the receive socket buffer is limited, e.g., to the ubiquitous default of 64 KB, the advertised receive window will not be as vulnerable even for very long connections over very high bandwidths. The vulnerability will grow linearly with the increased network speed, but not as the square. The consequence is lower sustained throughput, where only one window's worth of data per round-trip time (RTT) is exchanged.
Top   ToC   RFC4953 - Page 14
   This will keep the connection open longer; for long-lived connections
   with continuous sourced data, this may continue to present an attack
   opportunity, albeit a sparse and slow-moving target.  For the most
   recent case where BGP data is being exchanged between Internet
   routers, the data is bursty and the aggregate traffic may be small
   (i.e., unlikely to cover a substantial portion of the sequence space,
   even if long-lived), so smaller advertised receive windows (via small
   receiver buffers) may, in some cases, sufficiently address the
   immediate problem.  This assumes that the routing tables can be
   exchanged quickly enough with bandwidth reduced due to the smaller
   buffers, or perhaps that the advertised receive window is opened only
   during a large burst exchange (e.g., via some other signal between
   the two routers, or a time-based signal, though either would be
   nonstandard).

3.1.6. Other Transport Protocol Solutions

Segment authentication has been addressed at the transport layer in other protocols. Both SCTP and DCCP include cookies for connection establishment and use them to authenticate a variety of other control messages [28][41]. The inclusion of such mechanism at the transport protocol, although emerging as standard practice, complicates the design and implementation of new protocols [32]. As new attacks are discovered (SYN floods, RSTs, etc.), each protocol must be modified individually to compensate. A network solution may be more appropriate and efficient. It should be noted that RST attacks, which rely on brute-force, are relatively easy for intrusion detection software to detect at the TCP layer. Any connection that receives a large number of invalid -- outside-window -- RSTs might have subsequent RSTs blocked, to defeat such attacks. This would have the side-effect of blocking legitimate RSTs to that connection, which might then interfere with cleaning up the transport state between the endpoint peers. This side-effect, coupled with the increased monitoring load, might render such solutions undesirable in the general case, but they might usefully be applied to special cases, e.g., for BGP for routers.

3.2. Network Layer (IP) Solutions

There are two primary variants of network layer solutions to spoofing: address filtering and IPsec. Address filtering is an indirect system that relies on other parties to filter packets sent upstream of an attack, but does not necessarily require participation of the packet source. IPsec requires cooperation between the endpoints wanting to avoid attack on their connection, which currently involves preexisting shared knowledge of either a shared key or shared certificate authority.
Top   ToC   RFC4953 - Page 15

3.2.1. Address Filtering

Address filtering is often proposed as an alternative to protocol mechanisms to defeat IP source address spoofing [2][13]. Address filtering restricts traffic from downstream sources across transit networks based on the IP source address. A kind of filtering already occurs at the endpoints of a connection, because attack messages must match the socket pair to succeed; again, note that such attacks require knowing the entire socket pair, and are unlikely except in particular cases. This section discusses filtering based on address only, typically done at the borders of an AS. It can also restrict core-to-edge paths to reject traffic that should have originated further toward the edge. It cannot restrict traffic from edges lacking filtering through the core to a particular edge. As a result, each border router must perform the appropriate filtering for overall protection to result; failure of any border router to filter defeats the protection of all participants inside the border, and potentially those outside as well. Address filtering at the border can protect those inside the border from some kinds of spoofing, i.e., connections among those inside a border, because only interior addresses should originate inside the border. It cannot, however, protect connections including endpoints outside the border (i.e., those that traverse the AS boundary) except to restrict where the traffic enters from, e.g., if it expected from one AS and not another. As a result, address filtering is not a local solution that can be deployed to protect communicating pairs, but rather relies on a distributed infrastructure of trusted gateways filtering forged traffic where it enters the network. It is not feasible for local, incremental deployment, but may be applicable to connections among those inside the protected border in some scenarios. Applying filtering can also be useful to reduce the network load of spoofed traffic [31]. A more recent variant of address filtering checks the IP TTL (Time to Live) field, relying on the TTL set by the other end of the connection [15]. This technique has been used to provide filtering for BGP. It assumes the connection source TTL is set to 255; packets at the receiver are checked for TTL=255, and others are dropped. This restricts traffic to one hop upstream of the receiver (i.e., a BGP router), but those hops could include other user programs at those nodes (e.g., the BGP router's peer) or any traffic those nodes accept via tunnels -- because tunnels need not decrement TTLs, notably for "bump in the wire" (BITW) or BITW-equivalent scenarios [33] (see also Section 5.1 of [15] and [16]). TTL filtering works only where all traffic from the other end of the tunnel is trusted,
Top   ToC   RFC4953 - Page 16
   i.e., where it does not originate or transit spoofed traffic.  The
   use of TTL rather than link or network security also assumes an
   untampered point-to-point link, where no other traffic can be spoofed
   onto a link.

   This method of filtering works best where traffic originates one hop
   away, so that the address filtering is based on the trust of only
   directly-connected (tunneled or otherwise) nodes.  Like conventional
   address filtering, this reduces spoofing traffic in general, but is
   not considered a reliable security mechanism because it relies on
   distributed filtering (e.g., the fact that upstream nodes do not
   terminate tunnels arbitrarily).

3.2.2. IPsec

TCP is susceptible to RSTs, but also to other off-path and on-path spoofing attacks, including SYN attacks. Other transport protocols, such as UDP and RTP are equally susceptible. Although emerging transport protocols attempt to defeat such attacks at the transport layer, such attacks take advantage of network layer identity spoofing. The packet is coming from an endpoint that is spoofing another endpoint, either upstream or somewhere else in the Internet. IPsec was designed specifically to establish and enforce authentication of a packet's source and contents in order to most directly and explicitly address this security vulnerability. The larger problem with IPsec is that of key distribution and use. IPsec is often cumbersome, and has only recently been supported in many end-system operating systems. More importantly, it relies on preshared keys, signed X.509 certificates, or a trusted third-party (e.g., Kerberos) key infrastructure to establish and exchange keying information (e.g., via IKE). Each of these issues presents challenges when using IPsec to secure traffic to a well-known server, whose clients may not support IPsec or may not have registered with a previously-known certificate authority (CA). These keying challenges are being addressed in the IETF in ways that will enable servers secure associations with other parties without advance coordination [45][46]. This can be especially useful for publicly-available servers, or for protecting connections to servers that -- for whatever reason -- have not or will not deploy conventional IPsec certificates (i.e., core Internet BGP routers).
Top   ToC   RFC4953 - Page 17

4. ICMP

Just as spoofed TCP packets can terminate a connection, so too can spoofed ICMP packets. ICMP can be used to launch a variety of attacks on TCP including connection resets, path-MTU attacks, and can also be used to attack the host with non-TCP 'ping of death' and 'smurf attacks', etc. [40]. ICMP thus represents a substantial threat to TCP, but this is not the focus of this document, although a number of protections are discussed below because some are comparable to TCP anti-spoofing techniques. Note also that ICMP attacks on TCP assume that the socket pair is known by the attacker, which is unlikely except for a subset of services between pairs of widely- known endpoints. TCP headers can be included inside certain ICMP messages [7]. There have been recent suggestions to validate the sequence number of TCP headers when they occur inside ICMP messages [18]. This sequence checking is similar to checks that would occur for conventional data packets in TCP, but is being proposed in the spirit of the RST window attenuation described in Section 3.1.2. Some such checks may be reasonable, especially where they parallel the validations already performed by TCP processing, notably where they emulate the semantics of such processing. For example, the TCP checksum should be validated (if the entire TCP segment is contained in the ICMP message) before any fields of the TCP header are examined, to avoid reacting to corrupted packets. Similarly, if the TCP MD5 option is present, its signature should probably be validated before considering the contents of the message. Such validation can ensure that the packet was not corrupted prior to the ICMP generation (checksum), that the packet was one sent by the source (IPsec or TCP/MD5 authenticated), or that the packet was not in the network for an excess of 2*MSL (valid sequence number). ICMP presents a particular challenge because some messages can reset a connection more easily -- with less validation -- than even some spoofed TCP segments. One other proposed alternative is to change TCP's reaction to ICMPs after a connection is established; that may leave TCP susceptible during connection establishment and modifies TCP's reaction to certain valid network events [19]. This considers the context-sensitivity of ICMP messages, as does IPsec in some tunneled configurations, but the recommendations are ambiguous regarding such filtering [27]. Ultimately, requiring TCP ICMP messages to be 'in window' may be insufficient protection, as this document shows for spoofed data. ICMP packets can be authenticated when originating at known, trusted endpoints, such as endpoints of connections or routers in known
Top   ToC   RFC4953 - Page 18
   domains with preexisting IPsec associations.  Unfortunately, they
   also can originate at other places in the network.  In addition, some
   networks filter all ICMP packets because validation may not be
   possible, especially because they can be injected from anywhere in a
   network, and so cannot be easily and locally address filtered [27].
   As a result, they are not addressed separately in the issues or
   security considerations of this document further.

5. Issues

There are a number of existing and proposed solutions addressing the vulnerability of transport protocols in general (and TCP in specific) to off-path third-party spoofing attacks. As shown, these operate at the transport or network layer. Transport solutions require separate modification of each transport protocol, addressing network identity spoofing separately in the context of each transport association. Network solutions require distributed coordination (filtering) or can be computationally intensive and require pervasive registration of certificate authorities with every possible endpoint (authentication). This section explains these observations further.

5.1. Transport Layer (e.g., TCP)

Transport solutions rely on shared cookies to authenticate segments, including data, transport header, and even pseudo-header (e.g., fixed portions of the outer IP header in TCP). Because the Internet relies on stateless network protocols, it makes sense to rely on state establishment and maintenance available in some transport layers not only for the connection but for authentication state. Three-way handshakes and heartbeats can be used to negotiate authentication state in conjunction with connection parameters, which can be stored with connection state easily. As noted earlier, transport layer solutions require separate modification of all transport protocols to include authentication. Not all transport protocols support negotiated endpoint state (e.g., UDP), and legacy protocols have been notoriously difficult to safely augment. Not all authentication solutions are created equal, either, and relying on a variety of transport solutions exposes end-systems to increased potential for incorrectly specified or implemented solutions. Transport authentication has often been developed piece- wise, in response to specific attacks, e.g., SYN cookies and RST window attenuation [4][36]. Transport layer solutions are not only per-protocol, but often per- connection. This has both advantages and drawbacks. One advantage to transport layer solutions is that they can protect the transport protocol when lower layers have failed, e.g., due to bugs in
Top   ToC   RFC4953 - Page 19
   implementation.  TCP already includes a variety of packet validation
   mechanisms to protect in these cases, e.g., checking that RSTs are
   in-window.  More strict checks can increase the protections provided,
   e.g., to protect against misaddressed RSTs that end up in-window (via
   TCPsecure) or to protect against connection interruption due to RSTs,
   SYNs, or data injection from misaddressed packets (TCP/MD5) [36].

   Another advantage is that transport layer protections can be more
   specifically limited to a particular connection.  Because each
   connection negotiates its state separately, that state can be more
   specifically tied to that connection.  This is both an advantage and
   a drawback.  It can make it easier to tie security to an individual
   connection, although in practice a shared secret or certificate will
   generally be shared across multiple connections.

   As a drawback, each transport connection needs to negotiate and
   maintain authentication state separately.  Some overhead is not
   amortized over multiple connections, e.g., overheads in packet
   exchanges, whereas other overheads are not amortized over different
   transport protocols, e.g., design and implementation complexity --
   both as would be the case in a network layer solution.  Because the
   authentication happens later in packet processing than is required,
   additional endpoint resources may be needlessly consumed, e.g., in
   demultiplexing received packets, indexing connection identifiers, and
   continuing to buffer spoofed packets, etc., only to be dropped later
   at the transport layer.

5.2. Network Layer (IP)

A network layer solution avoids the hazards of multiple transport variants, using a single shared endpoint authentication mechanism early in receiver packet processing to discard unauthenticated packets at the network layer instead. This defeats spoofing entirely because spoofing involves masquerading as another endpoint, and network layer security validates the endpoint as the source of the packets it emits. Such a network level solution protects all transport protocols as a result, including both legacy and emerging protocols, and reduces the complexity of these protocols as well. A shared solution also reduces protocol overhead, and decouples the management (and refreshing) of authentication state from that of individual transport connections. Finally, a network layer solution protects not only the transport layer but the network layer as well, e.g., from IGMP, and some kinds of ICMP (Section 4), spoofing attacks. The IETF Proposed Standard protocol for network layer authentication is IPsec [27]. IPsec specifies the overall architecture, including header authentication (AH) [25] and encapsulation (ESP) modes [26].
Top   ToC   RFC4953 - Page 20
   AH authenticates both the IP header and IP data, whereas ESP
   authenticates only the IP data (e.g., transport header and payload).
   AH is being phased out since ESP is more efficient and the Security
   Parameters Index (SPI) includes sufficient information to verify the
   IP header anyway [27].  These two modes describe the security applied
   to individual packets within the IPsec system; key exchange and
   management is performed either out-of-band (via pre-shared keys) or
   by an automated key exchange protocol, e.g., IKE [24].

   IPsec already provides authentication of an IP header and its data
   contents sufficient to defeat both on-path and off-path third-party
   spoofing attacks.  IKE can configure authentication between two
   endpoints on a per-endpoint, per-protocol, or per-connection basis,
   as desired.  IKE also can perform automatic periodic re-keying,
   further defeating crypto-analysis based on snooping (clandestine data
   collection).  The use of IPsec is already commonly strongly
   recommended for protected infrastructure.

   Existing IPsec is not appropriate for many deployments.  It is
   computationally intensive both in key management and individual
   packet authentication [43].  This computational overhead can be
   prohibitive, and so often requires additional hardware, especially in
   commercial routers.  As importantly, IKE is not anonymous; keys can
   be exchanged between parties only if they trust each other's X.509
   certificates, trust some other third-party to help with key
   generation (e.g., Kerberos), or pre-share a key.  These certificates
   provide identification (the other party knows who you are) only where
   the certificates themselves are signed by certificate authorities
   (CAs) that both parties already trust.  To a large extent, the CAs
   themselves are the pre-shared keys that help IKE establish security
   association keys, which are then used in the authentication
   algorithms.

   Alternative mechanisms are under development to address this
   limitation, to allow publicly-accessible servers to secure
   connections to clients not known in advance, or to allow unilateral
   relaxation of identity validation so that the remaining protections
   of IPsec can be made available [45][46].  In particular, these
   mechanisms can prevent a client (but without knowing who that client
   is) from being affected by spoofing from other clients, even when the
   attackers are on the same communications path.

   IPsec, although widely available both in commercial routers and
   commodity end-systems, is not often used except between parties that
   already have a preexisting relationship (employee/employer, between
   two ISPs, etc.).  Servers to anonymous clients (e.g., customer/
   business) or more open services (e.g., BGP, where routers may have
   large numbers of peers) are unmanageable, due to the breadth and flux
Top   ToC   RFC4953 - Page 21
   of CAs.  New endpoints cannot establish IPsec associations with such
   servers unless their own certificate is signed by a CA already
   trusted by the server.  Different servers -- even within the same
   overall system (e.g., BGP) -- often cannot or will not trust
   overlapping subsets of CAs in general.

5.3. Application Layer

There are a number of application layer authentication mechanisms, often implicit within end-to-end encryption. Application layer security (e.g., TLS, SSH, or MD5 checksums within a BGP stream) provides the ultimate protection of application data from all intermediaries, including network routers as well as exposure at other layers in the end-systems. This is the only way to ultimately protect the application data. Application authentication cannot protect either the network or transport protocols from spoofing attacks, however. Spoofed packets interfere with network processing or reset transport connections before the application checks the data. Authentication needs to winnow these packets and drop them before they interfere at these lower layers. An alternate application layer solution would involve resilience to reset connections. If the application can recover from such connection interruptions, then such attacks have less impact. Unfortunately, attackers still affect the application, e.g., in the cost of restarting connections, delays until connections are restarted, or increased connection establishment messages on the network. Some applications -- notably BGP -- even interpret TCP connection reliability as an indicator of route path stability, which is why attacks on BGP have such substantial consequences.

5.4. Link Layer

Link layer security operates separately on each hop of an Internet. Such security can be critical in protecting link resources, such as bandwidth and link management protocols. Protection at this layer cannot suffice for network or transport layers, because it cannot authenticate the endpoint source of a packet. Link authentication ensures only the source of the current link hop where it is examined.

5.5. Issues Discussion

The issues raised in this section suggest that there are challenges with all solutions to transport protection from spoofing attacks. This raises the potential need for alternate security levels. While it is already widely recognized that security needs to occur
Top   ToC   RFC4953 - Page 22
   simultaneously at many protocol layers, there also may be utility in
   supporting a variety of strengths at a single layer.  For example,
   IPsec already supports a variety of algorithms (MD5, SHA1, etc., for
   authentication), but always assumes that:

   1. The entire body of the packet is secured.

   2. Security associations are established only where identity is
      authenticated by a known certificate authority or other pre-shared
      key.

   3. Both on-path and off-path third-party spoofing attacks must be
      defeated.

   These assumptions are prohibitive, especially in many cases of
   spoofing attacks.  For spoofing, the primary issue is whether packets
   are coming from the same party the server can reach.  Only the IP
   header is fundamentally in question, so securing the entire packet
   (1) is computational overkill.  It is sufficient to authenticate the
   other party as "a party you have exchanged packets with", rather than
   establishing their trusted identity ("Bill" vs. "Bob") as in (2).
   Finally, many cookie systems use clear-text (unencrypted), fixed
   cookie values, providing reasonable (1 in 2^{cookie-size}) protection
   against off-path third-party spoof attacks, but not addressing on-
   path attacks at all.  Such potential solutions are discussed in the
   Better Than Nothing Security (BTNS) documents [5][45][46].  Note also
   that NULL Encryption in IPsec applies a variant of this cookie, where
   the SPI is the cookie, and no further encryption is applied [17].

6. Security Considerations

This entire document focuses on increasing the security of transport protocols and their resistance to spoofing attacks. Security is addressed throughout. This document describes a number of techniques for defeating spoofing attacks. Those relying on clear-text cookies, either explicit or implicit (e.g., window sequence attenuation) do not protect from on- path spoofing attacks, since valid values can be learned from prior traffic. Those relying on true authentication algorithms are stronger, protecting even from on-path attacks, because the authentication hash in a single packet approaches the behavior of "one-time" cookies. The security of various levels of the protocol stack is addressed. Spoofing attacks are fundamentally identity masquerading, so we believe the most appropriate solutions defeat these at the network layer, where end-to-end identity lies. Some transport protocols
Top   ToC   RFC4953 - Page 23
   subsume endpoint identity information from the network layer (e.g.,
   TCP pseudo-headers), whereas others establish per-connection identity
   based on exchanged nonces (e.g., SCTP).  It is reasonable, if not
   recommended, to address security at all layers of the protocol stack.

   Note that Network Address Translators (NATs) and other middleboxes
   complicate the design and deployment of techniques to defeat spoofing
   attacks.  Devices such as these, that modify IP and/or TCP headers
   in-transit, generate traffic equivalent to a spoofing attack, and
   thus should be inhibited by antispoofing mechanisms.  Details of
   these middlebox-related problems are out of scope for this document,
   but issues thereof are addressed in RFCs and emerging documents that
   discuss the interactions between such devices and the Internet
   architecture, e.g., [21].  Fortunately, many of the most critical
   TCP-based connections -- in particular, those supporting routing
   protocols like BGP -- do not traverse such middleboxes, and are not
   affected by this limitation.

7. Conclusions

This document describes the details of the recent BGP spoofing attacks involving spurious RSTs, which could be used to shutdown TCP connections. It summarizes and discusses a variety of current and proposed solutions at various protocol layers.

8. Acknowledgments

This document was inspired by discussions in the TCPM WG <http://www.ietf.org/html.charters/tcpm-charter.html> about the recent spoofed RST attacks on BGP routers, including R. Stewart's document (whose author list has since evolved) [36][42]. The analysis of the attack issues, alternate solutions, and the anonymous security proposed solutions were the result of discussions on that list as well as with USC/ISI's T. Faber, A. Falk, G. Finn, and Y. Wang. R. Atkinson suggested the UDP variant of TCP/MD5, P. Goyette suggested using the ISN to seed TCP/MD5, and L. Wood suggested using the ISN to validate RSTs. Other improvements are due to the input of various members of the IETF's TCPM WG, notably detailed feedback from F. Gont, P. Savola, and A. Hoenes. This document was prepared using 2-Word-v2.0.template.dot.
Top   ToC   RFC4953 - Page 24

9. Informative References

[1] Allman, M., Paxson, V., and W. Stevens, "TCP Congestion Control", RFC 2581, April 1999. [2] Baker, F. and P. Savola, "Ingress Filtering for Multihomed Networks", BCP 84, RFC 3704, March 2004. [3] Bellovin, S. and A. Zinin, "Standards Maturity Variance Regarding the TCP MD5 Signature Option (RFC 2385) and the BGP-4 Specification", RFC 4278, January 2006. [4] Bernstein, D., "SYN cookies", 1997, <http://cr.yp.to/syncookies.html>. [5] Better Than Nothing Security [BTNS] WG web pages, <http://www.postel.org/anonsec>. [6] Bonica, R., Weis, B., Viswanathan, S., Lange, A., and O. Wheeler, "Authentication for TCP-based Routing and Management Protocols", Work in Progress, February 2007. [7] Braden, R., "Requirements for Internet Hosts - Communication Layers", STD 3, RFC 1122, October 1989. [8] Braden, R., "TIME-WAIT Assassination Hazards in TCP", RFC 1337, May 1992. [9] CERT alert: "Technical Cyber Security Alert TA04-111A: Vulnerabilities in TCP", April 20, 2004, <http://www.us-cert.gov/cas/techalerts/TA04-111A.html>. [10] Convery, S., and M. Franz, "BGP Vulnerability Testing: Separating Fact from FUD", 2003, <http://www.nanog.org/mtg-0306/pdf/franz.pdf>. [11] Eddy, W., "TCP SYN Flooding Attacks and Common Mitigations", Work in Progress, May 2007. [12] Faber, T., J. Touch, and W. Yue, "The TIME-WAIT state in TCP and Its Effect on Busy Servers", Proc. Infocom 1999, pp. 1573- 1583, Mar. 1999. [13] Ferguson, P. and D. Senie, "Network Ingress Filtering: Defeating Denial of Service Attacks which employ IP Source Address Spoofing", BCP 38, RFC 2827, May 2000.
Top   ToC   RFC4953 - Page 25
   [14]  Floyd, S., "Inappropriate TCP Resets Considered Harmful", BCP
         60, RFC 3360, August 2002.

   [15]   Gill, V., Heasley, J., and D. Meyer, "The Generalized TTL
         Security Mechanism (GTSM)", RFC 3682, February 2004.

   [16]  Gill, V., Heasley, J., Meyer, D., Savola, P., Ed., and C.
         Pignataro, "The Generalized TTL Security Mechanism (GTSM)",
         Work in Progress, June 2007.

   [17]  Glenn, R. and S. Kent, "The NULL Encryption Algorithm and Its
         Use With IPsec", RFC 2410, November 1998.

   [18]  Gont, F., "ICMP attacks against TCP", Work in Progress, May
         2007.

   [19]  Gont, F., "TCP's Reaction to Soft Errors", Work in Progress,
         June 2007.

   [20]  Heffernan, A., "Protection of BGP Sessions via the TCP MD5
         Signature Option", RFC 2385, August 1998.

   [21]  Holdrege, M. and P. Srisuresh, "Protocol Complications with the
         IP Network Address Translator", RFC 3027, January 2001.

   [22]  Housley, R., Post to IETF Discussion mailing list regarding his
         IETF 64 Security Area presentation,
         ID=7.0.0.10.2.20051124135914.00f50558@vigilsec.com, Nov. 24,
         2005, <http://www1.ietf.org/
         mail-archive/ietf/Current/maillist.html>.

   [23]  Jacobson, V., Braden, R., and D. Borman, "TCP Extensions for
         High Performance", RFC 1323, May 1992.

   [24]  Kaufman, C., Ed., "Internet Key Exchange (IKEv2) Protocol", RFC
         4306, December 2005.

   [25]  Kent, S., "IP Authentication Header", RFC 4302, December 2005.

   [26]  Kent, S., "IP Encapsulating Security Payload (ESP)", RFC 4303,
         December 2005.

   [27]  Kent, S. and K. Seo, "Security Architecture for the Internet
         Protocol", RFC 4301, December 2005.

   [28]  Kohler, E., Handley, M., and S. Floyd, "Datagram Congestion
         Control Protocol (DCCP)", RFC 4340, March 2006.
Top   ToC   RFC4953 - Page 26
   [29]  Larsen, M., and F. Gont, "Port Randomization", Work in
         Progress, February 2007.

   [30]  Leech, M., "Key Management Considerations for the TCP MD5
         Signature Option", RFC 3562, July 2003.

   [31]  Moore, D., G. Voelker, and S. Savage, "Inferring Internet
         Denial-of-Service Activity", Proc. Usenix Security Symposium,
         Aug. 2001.

   [32]  O'Malley, S. and L. Peterson, "TCP Extensions Considered
         Harmful", RFC 1263, October 1991.

   [33]  Perkins, C., "IP Encapsulation within IP", RFC 2003, October
         1996.

   [34]  Poon, K., "Use of TCP timestamp option to defend against blind
         spoofing attack", Work in Progress, October 2004.

   [35]  Postel, J., "Transmission Control Protocol", STD 7, RFC 793,
         September 1981.

   [36]  Ramaiah, A., Stewart, R., and M. Dalal, "Improving TCP's
         Robustness to Blind In-Window Attacks", Work in Progress, July
         2007.

   [37]  Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A Border
         Gateway Protocol 4 (BGP-4)", RFC 4271, January 2006.

   [38]  Semke, J., J. Mahdavi, and M. Mathis, "Automatic TCP Buffer
         Tuning", ACM SIGCOMM '98/ Computer Communication Review, volume
         28, number 4, Oct. 1998.

   [39]  Shepard, T., "Reassign Port Number option for TCP", Work in
         Progress, July 2004.

   [40]  Shirey, R., "Internet Security Glossary, Version 2", Work in
         Progress, November 2006.

   [41]  Stewart, R., Xie, Q., Morneault, K., Sharp, C., Schwarzbauer,
         H., Taylor, T., Rytina, I., Kalla, M., Zhang, L., and V.
         Paxson, "Stream Control Transmission Protocol", RFC 2960,
         October 2000.

   [42]  TCPM: IETF TCPM Working Group and mailing list,
         <http://www.ietf.org/html.charters/tcpm-charter.html>.

   [43]  Touch, J., "Report on MD5 Performance", RFC 1810, June 1995.
Top   ToC   RFC4953 - Page 27
   [44]  Touch, J., "Performance Analysis of MD5", Proc. Sigcomm 1995,
         pp. 77-86, Mar. 1999.

   [45]  Touch, J., "ANONsec: Anonymous Security to Defend Against
         Spoofing Attacks", Work in Progress, May 2004.

   [46]  Touch, J., Black, D., and Y. Wang, "Problem and Applicability
         Statement for Better Than Nothing Security (BTNS)", Work in
         Progress, February 2007.

   [47]  Touch, J. and A. Mankin, "The TCP Simple Authentication
         Option", Work in Progress, July 2007.

   [48]  Watson, P., "Slipping in the Window: TCP Reset attacks",
         Presentation at 2004 CanSecWest,
         <http://cansecwest.com/csw04archive.html>.

   [49]  Wood, L., Post to TCPM mailing list regarding use of ISN in
         RSTs, ID=Pine.GSO.4.50.0404232249570.5889-
         100000@argos.ee.surrey.ac.uk, Apr. 23, 2004,
         <http://www1.ietf.org/mail-archive/web/tcpm/current/
         msg00213.html>.

Author's Addresses

Joe Touch USC/ISI 4676 Admiralty Way Marina del Rey, CA 90292-6695 U.S.A. Phone: +1 (310) 448-9151 Fax: +1 (310) 448-9300 EMail: touch@isi.edu URI: http://www.isi.edu/touch
Top   ToC   RFC4953 - Page 28
Full Copyright Statement

   Copyright (C) The IETF Trust (2007).

   This document is subject to the rights, licenses and restrictions
   contained in BCP 78, and except as set forth therein, the authors
   retain all their rights.

   This document and the information contained herein are provided on an
   "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
   OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
   THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS
   OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
   THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
   WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Intellectual Property

   The IETF takes no position regarding the validity or scope of any
   Intellectual Property Rights or other rights that might be claimed to
   pertain to the implementation or use of the technology described in
   this document or the extent to which any license under such rights
   might or might not be available; nor does it represent that it has
   made any independent effort to identify any such rights.  Information
   on the procedures with respect to rights in RFC documents can be
   found in BCP 78 and BCP 79.

   Copies of IPR disclosures made to the IETF Secretariat and any
   assurances of licenses to be made available, or the result of an
   attempt made to obtain a general license or permission for the use of
   such proprietary rights by implementers or users of this
   specification can be obtained from the IETF on-line IPR repository at
   http://www.ietf.org/ipr.

   The IETF invites any interested party to bring to its attention any
   copyrights, patents or patent applications, or other proprietary
   rights that may cover technology that may be required to implement
   this standard.  Please address the information to the IETF at
   ietf-ipr@ietf.org.

Acknowledgement

   Funding for the RFC Editor function is currently provided by the
   Internet Society.