| 1 |
<pre>Network Working Group P. Savola |
| 2 |
Request for Comments: 3964 CSC/FUNET |
| 3 |
Category: Informational C. Patel |
| 4 |
All Play, No Work |
| 5 |
December 2004 |
| 6 |
|
| 7 |
|
| 8 |
<span class="h1">Security Considerations for 6to4</span> |
| 9 |
|
| 10 |
Status of this Memo |
| 11 |
|
| 12 |
This memo provides information for the Internet community. It does |
| 13 |
not specify an Internet standard of any kind. Distribution of this |
| 14 |
memo is unlimited. |
| 15 |
|
| 16 |
Copyright Notice |
| 17 |
|
| 18 |
Copyright (C) The Internet Society (2004). |
| 19 |
|
| 20 |
Abstract |
| 21 |
|
| 22 |
The IPv6 interim mechanism 6to4 (<a href="./rfc3056">RFC3056</a>) uses automatic |
| 23 |
IPv6-over-IPv4 tunneling to interconnect IPv6 networks. The |
| 24 |
architecture includes 6to4 routers and 6to4 relay routers, which |
| 25 |
accept and decapsulate IPv4 protocol-41 ("IPv6-in-IPv4") traffic from |
| 26 |
any node in the IPv4 internet. This characteristic enables a number |
| 27 |
of security threats, mainly Denial of Service. It also makes it |
| 28 |
easier for nodes to spoof IPv6 addresses. This document discusses |
| 29 |
these issues in more detail and suggests enhancements to alleviate |
| 30 |
the problems. |
| 31 |
|
| 32 |
|
| 33 |
|
| 34 |
|
| 35 |
|
| 36 |
|
| 37 |
|
| 38 |
|
| 39 |
|
| 40 |
|
| 41 |
|
| 42 |
|
| 43 |
|
| 44 |
|
| 45 |
|
| 46 |
|
| 47 |
|
| 48 |
|
| 49 |
|
| 50 |
|
| 51 |
|
| 52 |
<span class="grey">Savola & Patel Informational [Page 1]</span></pre> |
| 53 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-2" ></span> |
| 54 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 55 |
|
| 56 |
|
| 57 |
Table of Contents |
| 58 |
|
| 59 |
<a href="#section-1">1</a>. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . <a href="#page-3">3</a> |
| 60 |
<a href="#section-2">2</a>. Different 6to4 Forwarding Scenarios . . . . . . . . . . . . . <a href="#page-4">4</a> |
| 61 |
<a href="#section-2.1">2.1</a>. From 6to4 to 6to4 . . . . . . . . . . . . . . . . . . . <a href="#page-4">4</a> |
| 62 |
<a href="#section-2.2">2.2</a>. From Native to 6to4 . . . . . . . . . . . . . . . . . . <a href="#page-5">5</a> |
| 63 |
<a href="#section-2.3">2.3</a>. From 6to4 to Native . . . . . . . . . . . . . . . . . . <a href="#page-5">5</a> |
| 64 |
<a href="#section-2.4">2.4</a>. Other Models . . . . . . . . . . . . . . . . . . . . . . <a href="#page-6">6</a> |
| 65 |
<a href="#section-2.4.1">2.4.1</a>. BGP between 6to4 Routers and Relays . . . . . . <a href="#page-6">6</a> |
| 66 |
<a href="#section-2.4.2">2.4.2</a>. 6to4 as an Optimization Method . . . . . . . . . <a href="#page-7">7</a> |
| 67 |
<a href="#section-2.4.3">2.4.3</a>. 6to4 as Tunnel End-Point Addressing Mechanism . . <a href="#page-8">8</a> |
| 68 |
<a href="#section-3">3</a>. Functionalities of 6to4 Network Components . . . . . . . . . . <a href="#page-9">9</a> |
| 69 |
<a href="#section-3.1">3.1</a>. 6to4 Routers . . . . . . . . . . . . . . . . . . . . . . <a href="#page-9">9</a> |
| 70 |
<a href="#section-3.2">3.2</a>. 6to4 Relay Routers . . . . . . . . . . . . . . . . . . . <a href="#page-10">10</a> |
| 71 |
<a href="#section-4">4</a>. Threat Analysis . . . . . . . . . . . . . . . . . . . . . . . <a href="#page-11">11</a> |
| 72 |
<a href="#section-4.1">4.1</a>. Attacks on 6to4 Networks . . . . . . . . . . . . . . . . <a href="#page-12">12</a> |
| 73 |
<a href="#section-4.1.1">4.1.1</a>. Attacks with ND Messages . . . . . . . . . . . . <a href="#page-13">13</a> |
| 74 |
<a href="#section-4.1.2">4.1.2</a>. Spoofing Traffic to 6to4 Nodes . . . . . . . . . <a href="#page-14">14</a> |
| 75 |
<a href="#section-4.1.3">4.1.3</a>. Reflecting Traffic to 6to4 Nodes . . . . . . . . <a href="#page-17">17</a> |
| 76 |
<a href="#section-4.1.4">4.1.4</a>. Local IPv4 Broadcast Attack . . . . . . . . . . <a href="#page-19">19</a> |
| 77 |
<a href="#section-4.2">4.2</a>. Attacks on Native IPv6 Internet . . . . . . . . . . . . <a href="#page-20">20</a> |
| 78 |
<a href="#section-4.2.1">4.2.1</a>. Attacks with ND Messages . . . . . . . . . . . . <a href="#page-21">21</a> |
| 79 |
<a href="#section-4.2.2">4.2.2</a>. Spoofing Traffic to Native IPv6 Nodes. . . . . . <a href="#page-21">21</a> |
| 80 |
<a href="#section-4.2.3">4.2.3</a>. Reflecting Traffic to Native IPv6 Nodes . . . . <a href="#page-23">23</a> |
| 81 |
<a href="#section-4.2.4">4.2.4</a>. Local IPv4 Broadcast Attack . . . . . . . . . . <a href="#page-24">24</a> |
| 82 |
<a href="#section-4.2.5">4.2.5</a>. Theft of Service . . . . . . . . . . . . . . . . <a href="#page-25">25</a> |
| 83 |
<a href="#section-4.2.6">4.2.6</a>. Relay Operators Seen as Source of Abuse . . . . <a href="#page-26">26</a> |
| 84 |
<a href="#section-4.3">4.3</a>. Attacks on IPv4 Internet . . . . . . . . . . . . . . . . <a href="#page-28">28</a> |
| 85 |
<a href="#section-4.4">4.4</a>. Summary of the Attacks . . . . . . . . . . . . . . . . . <a href="#page-28">28</a> |
| 86 |
<a href="#section-5">5</a>. Implementing Proper Security Checks in 6to4 . . . . . . . . . <a href="#page-30">30</a> |
| 87 |
<a href="#section-5.1">5.1</a>. Encapsulating IPv6 into IPv4 . . . . . . . . . . . . . . <a href="#page-31">31</a> |
| 88 |
<a href="#section-5.2">5.2</a>. Decapsulating IPv4 into IPv6 . . . . . . . . . . . . . . <a href="#page-31">31</a> |
| 89 |
<a href="#section-5.3">5.3</a>. IPv4 and IPv6 Sanity Checks . . . . . . . . . . . . . . <a href="#page-32">32</a> |
| 90 |
<a href="#section-5.3.1">5.3.1</a>. IPv4 . . . . . . . . . . . . . . . . . . . . . . <a href="#page-32">32</a> |
| 91 |
<a href="#section-5.3.2">5.3.2</a>. IPv6 . . . . . . . . . . . . . . . . . . . . . . <a href="#page-33">33</a> |
| 92 |
<a href="#section-5.3.3">5.3.3</a>. Optional Ingress Filtering . . . . . . . . . . . <a href="#page-33">33</a> |
| 93 |
<a href="#section-5.3.4">5.3.4</a>. Notes about the Checks . . . . . . . . . . . . . <a href="#page-33">33</a> |
| 94 |
<a href="#section-6">6</a>. Issues in 6to4 Implementation and Use . . . . . . . . . . . . <a href="#page-34">34</a> |
| 95 |
<a href="#section-6.1">6.1</a>. Implementation Considerations with Automatic Tunnels . . <a href="#page-34">34</a> |
| 96 |
<a href="#section-6.2">6.2</a>. A Different Model for 6to4 Deployment . . . . . . . . . <a href="#page-35">35</a> |
| 97 |
<a href="#section-7">7</a>. Security Considerations . . . . . . . . . . . . . . . . . . . <a href="#page-36">36</a> |
| 98 |
<a href="#section-8">8</a>. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . <a href="#page-36">36</a> |
| 99 |
<a href="#section-9">9</a>. References . . . . . . . . . . . . . . . . . . . . . . . . . . <a href="#page-37">37</a> |
| 100 |
<a href="#appendix-A">A</a>. Some Trivial Attack Scenarios Outlined . . . . . . . . . . . . <a href="#page-39">39</a> |
| 101 |
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . <a href="#page-40">40</a> |
| 102 |
Full Copyright Statement . . . . . . . . . . . . . . . . . . . . . <a href="#page-41">41</a> |
| 103 |
|
| 104 |
|
| 105 |
|
| 106 |
|
| 107 |
|
| 108 |
<span class="grey">Savola & Patel Informational [Page 2]</span></pre> |
| 109 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-3" ></span> |
| 110 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 111 |
|
| 112 |
|
| 113 |
<span class="h2"><a class="selflink" id="section-1" href="#section-1">1</a>. Introduction</span> |
| 114 |
|
| 115 |
The IPv6 interim mechanism "6to4" [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>] specifies automatic |
| 116 |
IPv6-over-IPv4 tunneling to interconnect isolated IPv6 clouds by |
| 117 |
embedding the tunnel IPv4 address in the IPv6 6to4 prefix. |
| 118 |
|
| 119 |
Two characteristics of the 6to4 mechanism introduce most of the |
| 120 |
security considerations: |
| 121 |
|
| 122 |
1. All 6to4 routers must accept and decapsulate IPv4 packets from |
| 123 |
every other 6to4 router, and from 6to4 relays. |
| 124 |
|
| 125 |
2. 6to4 relay routers must accept traffic from any native IPv6 node. |
| 126 |
|
| 127 |
As the 6to4 router must accept traffic from any other 6to4 router or |
| 128 |
relay, a certain requirement for trust is implied, and there are no |
| 129 |
strict constraints on what the IPv6 packet may contain. Thus, |
| 130 |
addresses within the IPv4 and IPv6 headers may be spoofed, and this |
| 131 |
leads to various types of threats, including different flavors of |
| 132 |
Denial of Service attacks. |
| 133 |
|
| 134 |
The 6to4 specification outlined a few security considerations and |
| 135 |
rules but was ambiguous as to their exact requirement level. |
| 136 |
Moreover, the description of the considerations was rather short, and |
| 137 |
some of them have proven difficult to understand or impossible to |
| 138 |
implement. |
| 139 |
|
| 140 |
This document analyzes the 6to4 security issues in more detail and |
| 141 |
outlines some enhancements and caveats. |
| 142 |
|
| 143 |
Sections <a href="#section-2">2</a> and <a href="#section-3">3</a> are more or less introductory, rehashing how 6to4 is |
| 144 |
used today based on the 6to4 specification, so that it is easier to |
| 145 |
understand how security could be affected. <a href="#section-4">Section 4</a> provides a |
| 146 |
threat analysis for implementations that already implement most of |
| 147 |
the security checks. <a href="#section-5">Section 5</a> describes the optimal |
| 148 |
decapsulation/encapsulation rules for 6to4 implementations, and |
| 149 |
<a href="#section-6">Section 6</a> provides further discussion on a few issues that have |
| 150 |
proven difficult to implement. <a href="#appendix-A">Appendix A</a> outlines a few possible |
| 151 |
trivial attack scenarios in which very little or no security has been |
| 152 |
implemented. |
| 153 |
|
| 154 |
For the sake of simplicity, in this document, the native Internet is |
| 155 |
assumed to encompass IPv6 networks formed by using other transition |
| 156 |
mechanisms (e.g., <a href="./rfc2893">RFC 2893</a> [<a href="#ref-4" title=""Transition Mechanisms for IPv6 Hosts and Routers"">4</a>]), as these mechanisms cannot talk |
| 157 |
directly to the 6to4 network. |
| 158 |
|
| 159 |
|
| 160 |
|
| 161 |
|
| 162 |
|
| 163 |
|
| 164 |
<span class="grey">Savola & Patel Informational [Page 3]</span></pre> |
| 165 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-4" ></span> |
| 166 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 167 |
|
| 168 |
|
| 169 |
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", |
| 170 |
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this |
| 171 |
document are to be interpreted as described in <a href="https://www.rfc-editor.org/bcp/bcp14">BCP 14</a>, <a href="./rfc2119">RFC 2119</a> [<a href="#ref-2" title=""Key words for use in RFCs to Indicate Requirement Levels"">2</a>]. |
| 172 |
|
| 173 |
Throughout this memo, IPv4 addresses from blocks 7.0.0.0/24, |
| 174 |
8.0.0.0/24, and 9.0.0.0/24 are used for demonstrative purposes, to |
| 175 |
represent global IPv4 addresses that have no relation to each other. |
| 176 |
This approach was chosen instead of just using addresses from |
| 177 |
192.0.2.0/24 [<a href="#ref-5" title=""Special-Use IPv4 Addresses"">5</a>] for two reasons: to use addresses whose 6to4 mapping |
| 178 |
is glaringly obvious, and to make it obvious that the IPv4 addresses |
| 179 |
of different 6to4 gateways need not have any relation to each other. |
| 180 |
|
| 181 |
<span class="h2"><a class="selflink" id="section-2" href="#section-2">2</a>. Different 6to4 Forwarding Scenarios</span> |
| 182 |
|
| 183 |
Note that when one communicates between 6to4 and native domains, the |
| 184 |
6to4 relays that will be used in the two directions are very likely |
| 185 |
different; routing is highly asymmetric. Because of this, it is not |
| 186 |
feasible to limit relays from which 6to4 routers may accept traffic. |
| 187 |
|
| 188 |
The first three subsections introduce the most common forms of 6to4 |
| 189 |
operation. Other models are presented in the fourth subsection. |
| 190 |
|
| 191 |
<span class="h3"><a class="selflink" id="section-2.1" href="#section-2.1">2.1</a>. From 6to4 to 6to4</span> |
| 192 |
|
| 193 |
6to4 domains always exchange 6to4 traffic directly via IPv4 |
| 194 |
tunneling; the endpoint address V4ADDR is derived from 6to4 prefix |
| 195 |
2002:V4ADDR::/48 of the destination. |
| 196 |
|
| 197 |
.--------. _----_ .--------. |
| 198 |
| 6to4 | _( IPv4 )_ | 6to4 | |
| 199 |
| router | <====> ( Internet ) <===> | router | |
| 200 |
'--------' (_ _) '--------' |
| 201 |
^ '----' ^ |
| 202 |
| Direct tunneling over IPv4 | |
| 203 |
V V |
| 204 |
.--------. .-------. |
| 205 |
| 6to4 | | 6to4 | |
| 206 |
| host | | host | |
| 207 |
'--------' '--------' |
| 208 |
|
| 209 |
Figure 1 |
| 210 |
|
| 211 |
It is required that every 6to4 router consider every other 6to4 |
| 212 |
router it wants to talk to be "on-link" (with IPv4 as the |
| 213 |
link-layer). |
| 214 |
|
| 215 |
|
| 216 |
|
| 217 |
|
| 218 |
|
| 219 |
|
| 220 |
<span class="grey">Savola & Patel Informational [Page 4]</span></pre> |
| 221 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-5" ></span> |
| 222 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 223 |
|
| 224 |
|
| 225 |
<span class="h3"><a class="selflink" id="section-2.2" href="#section-2.2">2.2</a>. From Native to 6to4</span> |
| 226 |
|
| 227 |
When native domains send traffic to 6to4 prefix 2002:V4ADDR::/48, it |
| 228 |
will be routed to the topologically nearest advertising 6to4 relay |
| 229 |
(advertising route to 2002::/16). The 6to4 relay will tunnel the |
| 230 |
traffic over IPv4 to the corresponding IPv4 address V4ADDR. |
| 231 |
|
| 232 |
Note that IPv4 address 9.0.0.1 here is just an example of a global |
| 233 |
IPv4 address, and it is assigned to the 6to4 router's |
| 234 |
pseudo-interface. |
| 235 |
|
| 236 |
Closest to |
| 237 |
"Native IPv6 node" |
| 238 |
.--------. _----_ .------------. .--------. |
| 239 |
| Native | _( IPv6 )_ | 6to4 relay | Tunneled | 6to4 | |
| 240 |
| IPv6 | -> ( Internet ) --> | router | =========> | router | |
| 241 |
| node | (_ _) '------------' 9.0.0.1 '--------' |
| 242 |
'--------' '----' dst_v6=2002:0900:0001::1 | |
| 243 |
V |
| 244 |
.-------. |
| 245 |
| 6to4 | |
| 246 |
| host | |
| 247 |
'--------' |
| 248 |
|
| 249 |
Figure 2 |
| 250 |
|
| 251 |
<span class="h3"><a class="selflink" id="section-2.3" href="#section-2.3">2.3</a>. From 6to4 to Native</span> |
| 252 |
|
| 253 |
6to4 domains send traffic to native domains by tunneling it over IPv4 |
| 254 |
to their configured 6to4 relay router, or the closest one found by |
| 255 |
using 6to4 IPv4 Anycast [<a href="#ref-3" title=""An Anycast Prefix for 6to4 Relay Routers"">3</a>]. The relay will decapsulate the packet |
| 256 |
and forward it to native IPv6 Internet, as would any other IPv6 |
| 257 |
packet. |
| 258 |
|
| 259 |
Note that the destination IPv6 address in the packet is a non-6to4 |
| 260 |
address and is assumed to be 2001:db8::1 in the example. |
| 261 |
|
| 262 |
|
| 263 |
|
| 264 |
|
| 265 |
|
| 266 |
|
| 267 |
|
| 268 |
|
| 269 |
|
| 270 |
|
| 271 |
|
| 272 |
|
| 273 |
|
| 274 |
|
| 275 |
|
| 276 |
<span class="grey">Savola & Patel Informational [Page 5]</span></pre> |
| 277 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-6" ></span> |
| 278 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 279 |
|
| 280 |
|
| 281 |
Configured |
| 282 |
-or- |
| 283 |
found by IPv4 Anycast |
| 284 |
.--------. _----_ .------------. .--------. |
| 285 |
| Native | _( IPv6 )_ | 6to4 relay | Tunneled | 6to4 | |
| 286 |
| Client | <- ( Internet ) <-- | router | <========= | router | |
| 287 |
'--------' (_ _) '------------' 192.88.99.1'--------' |
| 288 |
2001:db8::1 '----' (or configured) ^ |
| 289 |
| |
| 290 |
.-------. |
| 291 |
| 6to4 | |
| 292 |
| client | |
| 293 |
'--------' |
| 294 |
|
| 295 |
Figure 3 |
| 296 |
|
| 297 |
<span class="h3"><a class="selflink" id="section-2.4" href="#section-2.4">2.4</a>. Other Models</span> |
| 298 |
|
| 299 |
These are more or less special cases of 6to4 operations. In later |
| 300 |
chapters, unless otherwise stated, only the most generally used |
| 301 |
models (above) will be considered. |
| 302 |
|
| 303 |
<span class="h4"><a class="selflink" id="section-2.4.1" href="#section-2.4.1">2.4.1</a>. BGP between 6to4 Routers and Relays</span> |
| 304 |
|
| 305 |
Section 5.2.2.2 in [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>] presents a model where, instead of static |
| 306 |
configuration, BGP [<a href="#ref-6" title=""A Border Gateway Protocol 4 (BGP-4)"">6</a>] is used between 6to4 relay routers and 6to4 |
| 307 |
routers (for outgoing relay selection only). |
| 308 |
|
| 309 |
Going further than [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>], if the 6to4 router established a BGP session |
| 310 |
between all the possible 6to4 relays and advertised its /48 prefix to |
| 311 |
them, the traffic from non-6to4 sites would always come from a |
| 312 |
"known" relay. Alternatively, the 6to4 relays might advertise the |
| 313 |
more specific 6to4 routes between 6to4 relays. |
| 314 |
|
| 315 |
Both of these approaches are obviously infeasible due to scalability |
| 316 |
issues. |
| 317 |
|
| 318 |
Neither of these models are known to be used at the time of writing; |
| 319 |
this is probably because parties that need 6to4 are not able to run |
| 320 |
BGP, and because setting up these sessions would be much more work |
| 321 |
for relay operators. |
| 322 |
|
| 323 |
|
| 324 |
|
| 325 |
|
| 326 |
|
| 327 |
|
| 328 |
|
| 329 |
|
| 330 |
|
| 331 |
|
| 332 |
<span class="grey">Savola & Patel Informational [Page 6]</span></pre> |
| 333 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-7" ></span> |
| 334 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 335 |
|
| 336 |
|
| 337 |
<span class="h4"><a class="selflink" id="section-2.4.2" href="#section-2.4.2">2.4.2</a>. 6to4 as an Optimization Method</span> |
| 338 |
|
| 339 |
Some sites seem to use 6to4 as an IPv6 connectivity "optimization |
| 340 |
method"; that is, they also have non-6to4 addresses on their nodes |
| 341 |
and border routers but also employ 6to4 to reach 6to4 sites. |
| 342 |
|
| 343 |
This is typically done to be able to reach 6to4 destinations by |
| 344 |
direct tunneling without using relays at all. |
| 345 |
|
| 346 |
These sites also publish both 6to4 and non-6to4 addresses in DNS to |
| 347 |
affect inbound connections. If the source host's default address |
| 348 |
selection [<a href="#ref-7" title=""Default Address Selection for Internet Protocol version 6 (IPv6)"">7</a>] works properly, 6to4 sources will use 6to4 addresses to |
| 349 |
reach the site and non-6to4 nodes use non-6to4 addresses. If this |
| 350 |
behavior of foreign nodes can be assumed, the security threats to |
| 351 |
such sites can be significantly simplified. |
| 352 |
|
| 353 |
|
| 354 |
|
| 355 |
|
| 356 |
|
| 357 |
|
| 358 |
|
| 359 |
|
| 360 |
|
| 361 |
|
| 362 |
|
| 363 |
|
| 364 |
|
| 365 |
|
| 366 |
|
| 367 |
|
| 368 |
|
| 369 |
|
| 370 |
|
| 371 |
|
| 372 |
|
| 373 |
|
| 374 |
|
| 375 |
|
| 376 |
|
| 377 |
|
| 378 |
|
| 379 |
|
| 380 |
|
| 381 |
|
| 382 |
|
| 383 |
|
| 384 |
|
| 385 |
|
| 386 |
|
| 387 |
|
| 388 |
<span class="grey">Savola & Patel Informational [Page 7]</span></pre> |
| 389 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-8" ></span> |
| 390 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 391 |
|
| 392 |
|
| 393 |
<span class="h4"><a class="selflink" id="section-2.4.3" href="#section-2.4.3">2.4.3</a>. 6to4 as Tunnel End-Point Addressing Mechanism</span> |
| 394 |
|
| 395 |
6to4 addresses can also be used only as an IPv6-in-IPv4 tunnel |
| 396 |
endpoint addressing and routing mechanism. |
| 397 |
|
| 398 |
An example of this is interconnecting 10 branch offices where nodes |
| 399 |
use non-6to4 addresses. Only the offices' border routers need to be |
| 400 |
aware of 6to4, and use 6to4 addresses solely for addressing the |
| 401 |
tunnels between different branch offices. An example is provided in |
| 402 |
the figure below. |
| 403 |
|
| 404 |
2001:db8:0:10::/60 2001:db8:0:20::/60 |
| 405 |
.--------. .--------. |
| 406 |
( Branch 1 ) ( Branch 2 ) |
| 407 |
'--------' '--------' |
| 408 |
| | |
| 409 |
.--------. _----_ .--------. |
| 410 |
| 6to4 | _( IPv4 )_ | 6to4 | |
| 411 |
| router | <====> ( Internet ) <===> | router | |
| 412 |
'--------' (_ _) '--------' |
| 413 |
9.0.0.1 '----' 8.0.0.2 |
| 414 |
^^ |
| 415 |
|| |
| 416 |
vv |
| 417 |
.--------. |
| 418 |
| 6to4 | 7.0.0.3 |
| 419 |
| router | |
| 420 |
'--------' |
| 421 |
| 2001:db8::/48 |
| 422 |
.-----------. |
| 423 |
( Main Office ) |
| 424 |
'-----------' |
| 425 |
^ |
| 426 |
| |
| 427 |
v |
| 428 |
_----_ |
| 429 |
_( IPv6 )_ |
| 430 |
( Internet ) |
| 431 |
(_ _) |
| 432 |
'----' |
| 433 |
|
| 434 |
Figure 4 |
| 435 |
|
| 436 |
In the figure, the main office sets up two routes: |
| 437 |
|
| 438 |
2001:db8:0:10::/60 -> 2002:0900:0001::1 |
| 439 |
|
| 440 |
2001:db8:0:20::/60 -> 2002:0800:0002::1 |
| 441 |
|
| 442 |
|
| 443 |
|
| 444 |
<span class="grey">Savola & Patel Informational [Page 8]</span></pre> |
| 445 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-9" ></span> |
| 446 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 447 |
|
| 448 |
|
| 449 |
And a branch office sets up two routes as well: |
| 450 |
|
| 451 |
2001:db8:0:20::/60 -> 2002:0800:0002::1 |
| 452 |
|
| 453 |
default -> 2002:0700:0003::1 |
| 454 |
|
| 455 |
Thus, the IPv4 Internet is treated as an NBMA link-layer for |
| 456 |
interconnecting 6to4-enabled sites; with explicit routes, 6to4 |
| 457 |
addressing need not be used in routers other than the 6to4 edge |
| 458 |
routers. However, note that if a branch office sends a packet to any |
| 459 |
6to4 destination, it will not go through the main office, as the 6to4 |
| 460 |
2002::/16 route overrides the default route. |
| 461 |
|
| 462 |
This approach may make addressing and routing slightly easier in some |
| 463 |
circumstances. |
| 464 |
|
| 465 |
<span class="h2"><a class="selflink" id="section-3" href="#section-3">3</a>. Functionalities of 6to4 Network Components</span> |
| 466 |
|
| 467 |
This section summarizes the main functionalities of the 6to4 network |
| 468 |
components (6to4 routers, and 6to4 relays) and the security checks |
| 469 |
they must do. The pseudo-code for the security checks is provided in |
| 470 |
<a href="#section-5">Section 5</a>. |
| 471 |
|
| 472 |
This section summarizes the main functions of the various components |
| 473 |
of a 6to4 network: 6to4 relay routers and 6to4 routers. Refer to |
| 474 |
<a href="./rfc3056#section-1.1">Section 1.1 of RFC 3056</a> [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>] for 6to4-related definitions. |
| 475 |
|
| 476 |
<span class="h3"><a class="selflink" id="section-3.1" href="#section-3.1">3.1</a>. 6to4 Routers</span> |
| 477 |
|
| 478 |
The 6to4 routers act as the border routers of a 6to4 domain. It does |
| 479 |
not have a native global IPv6 address except in certain special |
| 480 |
cases. Since the specification [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>] uses the term "6to4 router", this |
| 481 |
memo does the same; however, note that in this definition, we also |
| 482 |
include a single host with a 6to4 pseudo-interface, which doesn't |
| 483 |
otherwise act as a router. The main functions of the 6to4 router are |
| 484 |
as follows: |
| 485 |
|
| 486 |
o Provide IPv6 connectivity to local clients and routers. |
| 487 |
|
| 488 |
o Tunnel packets sent to foreign 6to4 addresses to the destination |
| 489 |
6to4 router using IPv4. |
| 490 |
|
| 491 |
o Forward packets sent to locally configured 6to4 addresses to the |
| 492 |
6to4 network. |
| 493 |
|
| 494 |
o Tunnel packets sent to non-6to4 addresses to the configured/ |
| 495 |
closest-by-anycast 6to4 relay router. |
| 496 |
|
| 497 |
|
| 498 |
|
| 499 |
|
| 500 |
<span class="grey">Savola & Patel Informational [Page 9]</span></pre> |
| 501 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-10" ></span> |
| 502 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 503 |
|
| 504 |
|
| 505 |
o Decapsulate directly received IPv4 packets from foreign 6to4 |
| 506 |
addresses. |
| 507 |
|
| 508 |
o Decapsulate IPv4 packets received via the relay closest to the |
| 509 |
native IPv6 sources. Note that it is not easily distinguishable |
| 510 |
whether the packet was received from a 6to4 relay router or from a |
| 511 |
spoofing third party. |
| 512 |
|
| 513 |
The 6to4 router should also perform security checks on traffic that |
| 514 |
it receives from other 6to4 relays, or 6to4 routers, or from within |
| 515 |
the 6to4 site. These checks include the following: |
| 516 |
|
| 517 |
o Disallow traffic that has private, broadcast or certain specific |
| 518 |
reserved IPv4 addresses (see <a href="#section-5.3.1">Section 5.3.1</a> for details) in |
| 519 |
tunnels, or the matching 6to4 prefixes. |
| 520 |
|
| 521 |
o Disallow traffic from 6to4 routers in which the IPv4 tunnel source |
| 522 |
address does not match the 6to4 prefix. (Note that the |
| 523 |
pseudo-interface must pick the IPv4 address corresponding to the |
| 524 |
prefix when encapsulating, or problems may ensue, e.g., on |
| 525 |
multi-interface routers.) |
| 526 |
|
| 527 |
o Disallow traffic in which the destination IPv6 address is not a |
| 528 |
global address; in particular, link-local addresses, mapped |
| 529 |
addresses, and such should not be used. |
| 530 |
|
| 531 |
o Disallow traffic transmission to other 6to4 domains through 6to4 |
| 532 |
relay router or via some third party 6to4 router. (To avoid |
| 533 |
transmission to the relay router, the pseudo-interface prefix |
| 534 |
length must be configured correctly to /16. Further, to avoid the |
| 535 |
traffic being discarded, 6to4 source addresses must always |
| 536 |
correspond to the IPv4 address encapsulating the traffic.) |
| 537 |
|
| 538 |
o Discard traffic received from other 6to4 domains via a 6to4 relay |
| 539 |
router. |
| 540 |
|
| 541 |
o Discard traffic received for prefixes other than one's own 6to4 |
| 542 |
prefix(es). |
| 543 |
|
| 544 |
<span class="h3"><a class="selflink" id="section-3.2" href="#section-3.2">3.2</a>. 6to4 Relay Routers</span> |
| 545 |
|
| 546 |
The 6to4 relay router acts as a relay between all 6to4 domains and |
| 547 |
native IPv6 networks; more specifically, it |
| 548 |
|
| 549 |
o advertises the reachability of the 2002::/16 prefix to native IPv6 |
| 550 |
routing, thus receiving traffic to all 6to4 addresses from the |
| 551 |
closest native IPv6 nodes, |
| 552 |
|
| 553 |
|
| 554 |
|
| 555 |
|
| 556 |
<span class="grey">Savola & Patel Informational [Page 10]</span></pre> |
| 557 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-11" ></span> |
| 558 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 559 |
|
| 560 |
|
| 561 |
o advertises (if <a href="./rfc3068">RFC 3068</a> [<a href="#ref-3" title=""An Anycast Prefix for 6to4 Relay Routers"">3</a>] is implemented) the reachability of |
| 562 |
IPv4 "6to4 relay anycast prefix" (192.88.99.0/24) to IPv4 routing, |
| 563 |
thus receiving some tunneled traffic to native IPv6 nodes from |
| 564 |
6to4 routers. |
| 565 |
|
| 566 |
o decapsulates and forwards packets received from 6to4 addresses |
| 567 |
through tunneling, by using normal IPv6 routing, and |
| 568 |
|
| 569 |
o tunnels packets received through normal IPv6 routing from native |
| 570 |
addresses that are destined for 2002::/16 to the corresponding |
| 571 |
6to4 router. |
| 572 |
|
| 573 |
The 6to4 relay should also perform security checks on traffic that it |
| 574 |
receives from 6to4 routers, or from native IPv6 nodes. These checks |
| 575 |
are as follows: |
| 576 |
|
| 577 |
o Disallow traffic that has private, broadcast, or certain specific |
| 578 |
reserved IPv4 addresses in tunnels, or in the matching 6to4 |
| 579 |
prefixes. |
| 580 |
|
| 581 |
o Disallow traffic from 6to4 routers in which the IPv4 tunnel source |
| 582 |
address does not match the 6to4 prefix. (Note that the |
| 583 |
pseudo-interface must pick the IPv4 address corresponding to the |
| 584 |
prefix when encapsulating, or problems may ensue, e.g., on |
| 585 |
multi-interface routers.) |
| 586 |
|
| 587 |
o Disallow traffic in which the destination IPv6 address is not a |
| 588 |
global address; in particular, link-local addresses, mapped |
| 589 |
addresses, and such should not be used. |
| 590 |
|
| 591 |
o Discard traffic received from 6to4 routers with the destination as |
| 592 |
a 6to4 prefix. |
| 593 |
|
| 594 |
<span class="h2"><a class="selflink" id="section-4" href="#section-4">4</a>. Threat Analysis</span> |
| 595 |
|
| 596 |
This section discusses attacks against the 6to4 network or attacks |
| 597 |
caused by the 6to4 network. The threats are discussed in light of |
| 598 |
the 6to4 deployment models defined in <a href="#section-2">Section 2</a>. |
| 599 |
|
| 600 |
There are three general types of threats: |
| 601 |
|
| 602 |
1. Denial-of-Service (DoS) attacks, in which a malicious node |
| 603 |
prevents communication between the node under attack and other |
| 604 |
nodes. |
| 605 |
|
| 606 |
|
| 607 |
|
| 608 |
|
| 609 |
|
| 610 |
|
| 611 |
|
| 612 |
<span class="grey">Savola & Patel Informational [Page 11]</span></pre> |
| 613 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-12" ></span> |
| 614 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 615 |
|
| 616 |
|
| 617 |
2. Reflection Denial-of-Service (DoS) attacks, in which a malicious |
| 618 |
node reflects the traffic off unsuspecting nodes to a particular |
| 619 |
node (node under attack) in order to prevent communication |
| 620 |
between the node under attack and other nodes. |
| 621 |
|
| 622 |
3. Service theft, in which a malicious node/site/operator may make |
| 623 |
unauthorized use of service. |
| 624 |
|
| 625 |
6to4 also provides a means for a "meta-threat", traffic laundering, |
| 626 |
in which some other attack is channeled through the third parties to |
| 627 |
make tracing the real origin of the attack more difficult. This is |
| 628 |
used in conjunction with other threats, whether specific to 6to4 or |
| 629 |
not. |
| 630 |
|
| 631 |
At this point it is important to reiterate that the attacks are |
| 632 |
possible because |
| 633 |
|
| 634 |
1. 6to4 routers have to consider all 6to4 relays, and other 6to4 |
| 635 |
routers, as "on-link", |
| 636 |
|
| 637 |
2. 6to4 relays have to consider all 6to4 routers as "on-link", and |
| 638 |
|
| 639 |
3. it has been discovered that at least a couple of major 6to4 |
| 640 |
implementations do not implement all the security checks. |
| 641 |
|
| 642 |
The attacks' descriptions are classified based on the target of the |
| 643 |
attack: |
| 644 |
|
| 645 |
1. Attacks on 6to4 networks. |
| 646 |
|
| 647 |
2. Attacks on IPv6 networks. |
| 648 |
|
| 649 |
3. Attacks on IPv4 networks. |
| 650 |
|
| 651 |
Note that one of the mitigation methods listed for various attacks is |
| 652 |
based on the premise that 6to4 relays could have a feature limiting |
| 653 |
traffic to/from specific 6to4 sites. At the time of this writing, |
| 654 |
this feature is speculative, and more work needs to be done to |
| 655 |
determine the logistics. |
| 656 |
|
| 657 |
<span class="h3"><a class="selflink" id="section-4.1" href="#section-4.1">4.1</a>. Attacks on 6to4 Networks</span> |
| 658 |
|
| 659 |
This section describes attacks against 6to4 networks. Attacks that |
| 660 |
leverage 6to4 networks, but for which the ultimate victim is |
| 661 |
elsewhere (e.g., a native IPv6 user, an IPv4 user), are described |
| 662 |
later in the memo. |
| 663 |
|
| 664 |
|
| 665 |
|
| 666 |
|
| 667 |
|
| 668 |
<span class="grey">Savola & Patel Informational [Page 12]</span></pre> |
| 669 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-13" ></span> |
| 670 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 671 |
|
| 672 |
|
| 673 |
6to4 relays and routers are IPv4 nodes, and there is no way for any |
| 674 |
6to4 router to confirm the identity of the IPv4 node from which it |
| 675 |
receives traffic -- whether from a legitimate 6to4 relay or some |
| 676 |
other node. A 6to4 router has to process traffic from all IPv4 |
| 677 |
nodes. Malicious IPv4 nodes can exploit this property and attack |
| 678 |
nodes within the 6to4 network. |
| 679 |
|
| 680 |
It is possible to conduct a variety of attacks on the 6to4 nodes. |
| 681 |
These attacks are as follows: |
| 682 |
|
| 683 |
1. Attacks with Neighbor Discovery (ND) Messages |
| 684 |
|
| 685 |
2. Spoofing traffic to 6to4 nodes |
| 686 |
|
| 687 |
3. Reflecting traffic from 6to4 nodes |
| 688 |
|
| 689 |
4. Local IPv4 broadcast attack |
| 690 |
|
| 691 |
<span class="h4"><a class="selflink" id="section-4.1.1" href="#section-4.1.1">4.1.1</a>. Attacks with ND Messages</span> |
| 692 |
|
| 693 |
ATTACK DESCRIPTION |
| 694 |
|
| 695 |
Since the 6to4 router assumes that all the other 6to4 routers and |
| 696 |
6to4 relays are "on-link", it is possible to attack the 6to4 router |
| 697 |
by using ND messages from any node in the IPv4 network, unless a |
| 698 |
prior trust relationship has been established. |
| 699 |
|
| 700 |
The attacks target the 6to4 pseudo-interface. As long as the 6to4 |
| 701 |
addresses are not used in the source or destination address, the |
| 702 |
security checks specified by 6to4 take no stance on these packets. |
| 703 |
Typically they use link-local addresses. |
| 704 |
|
| 705 |
For example, an attack could be a Route Advertisement or Neighbor |
| 706 |
Advertisement message crafted specifically to cause havoc; the |
| 707 |
addresses in such a packet could resemble to the following: |
| 708 |
|
| 709 |
src_v6 = fe80::2 (forged address) |
| 710 |
dst_v6 = fe80::1 (valid or invalid address) |
| 711 |
src_v4 = 8.0.0.1 (valid or forged address) |
| 712 |
dst_v4 = 9.0.0.2 (valid address, matches dst_v6) |
| 713 |
|
| 714 |
These attacks are exacerbated if the implementation supports more |
| 715 |
tunneling mechanisms than 6to4 (or configured tunneling) because it |
| 716 |
is impossible to disambiguate such mechanisms, making it difficult to |
| 717 |
enable strict security checks (see <a href="#section-6.1">Section 6.1</a>). |
| 718 |
|
| 719 |
The Neighbor Discovery threats (Redirect DoS, or DoS) are described |
| 720 |
in [<a href="#ref-8" title=""IPv6 Neighbor Discovery (ND) Trust Models and Threats"">8</a>]. Note that all attacks may not be applicable, as the 6to4 |
| 721 |
|
| 722 |
|
| 723 |
|
| 724 |
<span class="grey">Savola & Patel Informational [Page 13]</span></pre> |
| 725 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-14" ></span> |
| 726 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 727 |
|
| 728 |
|
| 729 |
pseudo-interface is assumed not to have a link-layer address (<a href="#section-3.8">Section</a> |
| 730 |
<a href="#section-3.8">3.8</a> <a href="./rfc2893">RFC 2893</a> [<a href="#ref-4" title=""Transition Mechanisms for IPv6 Hosts and Routers"">4</a>]). However, note that the 6to4 router can be either |
| 731 |
a router or host from the Neighbor Discovery perspective. |
| 732 |
|
| 733 |
THREAT ANALYSIS AND MITIGATION METHODS |
| 734 |
|
| 735 |
The attacks can be mitigated by using any of the following methods: |
| 736 |
|
| 737 |
o The usage of ND messages could be prohibited. This implies that |
| 738 |
all packets using addresses of scope link-local will be silently |
| 739 |
discarded. <a href="./rfc3056#section-3.1">Section 3.1 of RFC 3056</a> [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>] leaves scope for future |
| 740 |
uses of link-local address. This method has its pitfalls: It |
| 741 |
would prohibit any sort of ND message and thus close the doors on |
| 742 |
development and use of other ND options. Whether this is a |
| 743 |
significant problem is another thing. |
| 744 |
|
| 745 |
o The 6to4 pseudo-interface could be insulated from the other |
| 746 |
interfaces, particularly the other tunnel interfaces (if any), for |
| 747 |
example by using a separate neighbor cache. |
| 748 |
|
| 749 |
o If ND messages are needed, either IPsec [<a href="#ref-4" title=""Transition Mechanisms for IPv6 Hosts and Routers"">4</a>] or an extension of |
| 750 |
SEND could be used [<a href="#ref-9" title=""SEcure Neighbor Discovery (SEND)"">9</a>] to secure packet exchange using the |
| 751 |
link-local address; vanilla SEND would not work, as the link-layer |
| 752 |
does not have an address -- and IPsec would be rather complex. |
| 753 |
|
| 754 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 755 |
|
| 756 |
Even though rather simply fixed, this attack is not new as such; the |
| 757 |
same is possible by using automatic tunneling [<a href="#ref-4" title=""Transition Mechanisms for IPv6 Hosts and Routers"">4</a>] or configured |
| 758 |
tunneling (if one is able to spoof source IPv4 address to that of the |
| 759 |
tunnel end-point). |
| 760 |
|
| 761 |
However, as 6to4 provides open decapsulation, and automatic tunneling |
| 762 |
is being deprecated [<a href="#ref-10" title=""Basic Transition Mechanisms for IPv6 Hosts and Routers"">10</a>], 6to4 provides an easy means, which would |
| 763 |
not exist without it. |
| 764 |
|
| 765 |
<span class="h4"><a class="selflink" id="section-4.1.2" href="#section-4.1.2">4.1.2</a>. Spoofing Traffic to 6to4 Nodes</span> |
| 766 |
|
| 767 |
ATTACK DESCRIPTION |
| 768 |
|
| 769 |
The attacker - a malicious IPv4 or IPv6 node - can send packets that |
| 770 |
are difficult to trace (e.g., due to spoofing or going through a |
| 771 |
relay) to a 6to4 node. This can be used e.g., to accomplish a DoS |
| 772 |
attack. |
| 773 |
|
| 774 |
The IPv6 and IPv4 addresses of the packets will be similar to the |
| 775 |
following: |
| 776 |
|
| 777 |
|
| 778 |
|
| 779 |
|
| 780 |
<span class="grey">Savola & Patel Informational [Page 14]</span></pre> |
| 781 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-15" ></span> |
| 782 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 783 |
|
| 784 |
|
| 785 |
src_v6 = 2001:db8::1 (forged address) |
| 786 |
dst_v6 = 2002:0900:0002::1 (valid address) |
| 787 |
src_v4 = 8.0.0.1 (valid or forged address) |
| 788 |
dst_v4 = 9.0.0.2 (valid address, matches dst_v6) |
| 789 |
|
| 790 |
For attacks launched from a native IPv6 node, the src_v4 will be the |
| 791 |
address of the relay through which the traffic will reach the 6to4 |
| 792 |
node. From IPv4 nodes, src_v4 can be either a spoofed source address |
| 793 |
or the real one. |
| 794 |
|
| 795 |
The 6to4 router receives these packets from 8.0.0.1, decapsulates |
| 796 |
them, discards the IPv4 header containing the source address 8.0.0.1, |
| 797 |
and processes them as normal (the attacker has guessed or obtained |
| 798 |
"dst_v6" by using one of a number of techniques). |
| 799 |
|
| 800 |
This is a DoS attack on 6to4 nodes. |
| 801 |
|
| 802 |
This attack is similar to those shown in [<a href="#ref-11" title=""Security of IPv6 Routing Header and Home Address Options"">11</a>]. |
| 803 |
|
| 804 |
EXTENSIONS |
| 805 |
|
| 806 |
Replies to the traffic will be directed to the src_v6 address, |
| 807 |
resulting in 6to4 nodes participating in a reflection DoS. This |
| 808 |
attack is described in more detail in <a href="#section-4.2.3">Section 4.2.3</a>. The replies |
| 809 |
(e.g., TCP SYN ACK, TCP RST, ICMPv6 Echo Reply, input sent to UDP |
| 810 |
echo service, ICMPv6 Destination Unreachable) are sent to the victim |
| 811 |
(src_v6), above. All the traces from the original attacker (src_v4) |
| 812 |
have been discarded. These return packets will go through a relay. |
| 813 |
|
| 814 |
Certain 6to4 networks may have a trivial ACL (Access Control List) |
| 815 |
based firewall that allows traffic to pass through if it comes from |
| 816 |
particular source(s). Such a firewalling mechanism can be bypassed |
| 817 |
by address spoofing. This attack can therefore be used for trivial |
| 818 |
ACL avoidance as well. These attacks might be hampered because the |
| 819 |
replies from the 6to4 node to the spoofed address will be lost. |
| 820 |
|
| 821 |
THREAT ANALYSIS AND SOLUTIONS/MITIGATION METHODS |
| 822 |
|
| 823 |
The Denial-of-Service attack based on traffic spoofing is not new; |
| 824 |
the only twists come from the fact that traces of an attack are more |
| 825 |
easily lost, and that spoofing the IPv6 address is possible even to |
| 826 |
those who are unable to do so in their current networks. The 6to4 |
| 827 |
router typically does not log IPv4 addresses (as they would be |
| 828 |
treated as L2 addresses), and thus the source of the attack (if |
| 829 |
launched from an IPv4 node) is lost. Because traces to the src_v4 |
| 830 |
address are easily lost, these attacks can also be launched from IPv4 |
| 831 |
nodes whose connections are ingress-filtered. |
| 832 |
|
| 833 |
|
| 834 |
|
| 835 |
|
| 836 |
<span class="grey">Savola & Patel Informational [Page 15]</span></pre> |
| 837 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-16" ></span> |
| 838 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 839 |
|
| 840 |
|
| 841 |
However, often this is not a real factor, as usually the attackers |
| 842 |
are just zombies and real attackers may not even care whether the |
| 843 |
unspoofed source address is discovered. |
| 844 |
|
| 845 |
Malicious native IPv6 nodes could be caught easily if ingress |
| 846 |
filtering was enabled everywhere in the IPv6 Internet. |
| 847 |
|
| 848 |
These attacks are easy to perform, but the extent of harm is limited: |
| 849 |
|
| 850 |
o For every packet sent, at most one reply packet is generated: |
| 851 |
there is no amplification factor. |
| 852 |
|
| 853 |
o Attack packets, if initiated from an IPv6 node, will pass through |
| 854 |
choke point(s), namely a 6to4 relay; in addition to physical |
| 855 |
limitations, these could implement some form of 6to4-site-specific |
| 856 |
traffic limiting. |
| 857 |
|
| 858 |
On the other hand, a variety of factors can make the attacks serious: |
| 859 |
|
| 860 |
o The attacker may have the ability to choose the relay, and he |
| 861 |
might employ the ones best suited for the attacks. Also, many |
| 862 |
relays use 192.88.99.1 [<a href="#ref-3" title=""An Anycast Prefix for 6to4 Relay Routers"">3</a>] as the source address, making tracing |
| 863 |
even more difficult (also see <a href="#section-4.2.6">Section 4.2.6</a>). |
| 864 |
|
| 865 |
o The relay's IPv4 address may be used as a source address for these |
| 866 |
attacks, potentially causing a lot of complaints or other actions, |
| 867 |
as the relay might seem to be the source of the attack (see |
| 868 |
<a href="#section-4.2.6">Section 4.2.6</a> for more). |
| 869 |
|
| 870 |
Some of the mitigation methods for such attacks are as follows: |
| 871 |
|
| 872 |
1. Ingress filtering in the native IPv6 networks to prevent packets |
| 873 |
with spoofed IPv6 sources from being transmitted. This would, |
| 874 |
thus, make it easy to identify the source of the attack. |
| 875 |
Unfortunately, it would depend on significant (or even complete) |
| 876 |
ingress filtering everywhere in other networks; while this is |
| 877 |
highly desirable, it may not be feasible. |
| 878 |
|
| 879 |
2. Security checks in the 6to4 relay. The 6to4 relay must drop |
| 880 |
traffic (from the IPv6 Internet) that has 6to4 addresses as |
| 881 |
source address; see <a href="#section-5">Section 5</a> for more detail. This has very |
| 882 |
little cost. |
| 883 |
|
| 884 |
However, these mitigation methods do not address the case of an IPv4 |
| 885 |
node sending encapsulated IPv6 packets. |
| 886 |
|
| 887 |
No simple way to prevent such attacks exists, and longer-term |
| 888 |
solutions, such as ingress filtering [<a href="#ref-12" title=""Network Ingress Filtering: Defeating Denial of Service Attacks which employ IP Source Address Spoofing"">12</a>] or itrace [<a href="#ref-13" title=""ICMP Traceback Messages"">13</a>], would have |
| 889 |
|
| 890 |
|
| 891 |
|
| 892 |
<span class="grey">Savola & Patel Informational [Page 16]</span></pre> |
| 893 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-17" ></span> |
| 894 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 895 |
|
| 896 |
|
| 897 |
to be deployed in both IPv6 and IPv4 networks to help identify the |
| 898 |
source of the attacks. A total penetration is likely impossible. |
| 899 |
(Note that itrace work has been discontinued, as of this writing in |
| 900 |
July 2004.) |
| 901 |
|
| 902 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 903 |
|
| 904 |
Traffic spoofing is not a new phenomenon in IPv4 or IPv6. 6to4 just |
| 905 |
makes it easier: Anyone can, regardless of ingress filtering, spoof a |
| 906 |
native IPv6 address to a 6to4 node, even if "maximal security" would |
| 907 |
be implemented and deployed. Losing trails is also easier. |
| 908 |
|
| 909 |
Therefore, depending on how much one assumes ingress filtering is |
| 910 |
deployed for IPv4 and IPv6, this could be considered either a very |
| 911 |
serious issue or close to irrelevant compared to the IP spoofing |
| 912 |
capabilities without 6to4. |
| 913 |
|
| 914 |
<span class="h4"><a class="selflink" id="section-4.1.3" href="#section-4.1.3">4.1.3</a>. Reflecting Traffic to 6to4 Nodes</span> |
| 915 |
|
| 916 |
ATTACK DESCRIPTION |
| 917 |
|
| 918 |
Spoofed traffic (as described in <a href="#section-4.2.2">Section 4.2.2</a>) may be sent to native |
| 919 |
IPv6 nodes to perform a reflection attack against 6to4 nodes. |
| 920 |
|
| 921 |
The spoofed traffic is sent to a native IPv6 node, either from an |
| 922 |
IPv4 node (through a 6to4 relay) or from a native IPv6 node (unless |
| 923 |
ingress filtering has been deployed). With the former, the sent |
| 924 |
packets would resemble the following: |
| 925 |
|
| 926 |
src_v6 = 2002:1234:1234::1 (forged address of the target 6to4 node) |
| 927 |
dst_v6 = 2002:0900:0002::1 (valid address) |
| 928 |
src_v4 = 8.0.0.1 (valid or invalid address) |
| 929 |
dst_v4 = 9.0.0.2 (valid address, matches dst_v6) |
| 930 |
|
| 931 |
Note that an attack through the relay is prevented if the relay |
| 932 |
implements proper decapsulation security checks (see <a href="#section-5">Section 5</a> for |
| 933 |
details) unless the IPv4 node can spoof the source address to match |
| 934 |
src_v6. Similarly, the attack from native IPv6 nodes could be |
| 935 |
prevented by global ingress filtering deployment. |
| 936 |
|
| 937 |
These attacks can be initiated by native IPv6, IPv4, or 6to4 nodes. |
| 938 |
|
| 939 |
EXTENSIONS |
| 940 |
|
| 941 |
A distributed Reflection DoS can be performed if a large number of |
| 942 |
nodes are involved in sending spoofed traffic with the same src_v6. |
| 943 |
|
| 944 |
|
| 945 |
|
| 946 |
|
| 947 |
|
| 948 |
<span class="grey">Savola & Patel Informational [Page 17]</span></pre> |
| 949 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-18" ></span> |
| 950 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 951 |
|
| 952 |
|
| 953 |
Malicious 6to4 nodes can also (try to) initiate this attack by |
| 954 |
bouncing traffic off 6to4 nodes in other 6to4 sites. However, this |
| 955 |
attack may not be possible, as the 6to4 router (in the site from |
| 956 |
which the attack is launched) will filter packets with forged source |
| 957 |
addresses (with security checks mentioned in <a href="#section-5">Section 5</a>). |
| 958 |
|
| 959 |
THREAT ANALYSIS AND SOLUTIONS/MITIGATION METHODS |
| 960 |
|
| 961 |
In this case, the reverse traffic comprises replies to the messages |
| 962 |
received by the 6to4 nodes. The attacker has less control on the |
| 963 |
packet type, and this would inhibit certain types of attacks. For |
| 964 |
example, flooding a 6to4 node with TCP SYN packets will not be |
| 965 |
possible (but e.g., a SYN-ACK or RST would be). |
| 966 |
|
| 967 |
These attacks may be mitigated in various ways: |
| 968 |
|
| 969 |
o Implementation of ingress filtering by the IPv4 service providers. |
| 970 |
This would prevent forging of the src_v4 address and help in |
| 971 |
closing down on the culprit IPv4 nodes. Note that it will be |
| 972 |
difficult to shut down the attack if a large number of IPv4 nodes |
| 973 |
are involved. |
| 974 |
|
| 975 |
These attacks may be also be stopped at the 6to4 sites if the |
| 976 |
culprit src_v4 address is identified, and if it is constant, by |
| 977 |
filtering traffic from this address. Note that it would be |
| 978 |
difficult to implement this method if appropriate logging were not |
| 979 |
done by the 6to4 router or if a large number of 6to4 nodes, and/or |
| 980 |
a large number of IPv4 nodes were participating in the attack. |
| 981 |
|
| 982 |
Unfortunately, because many IPv4 service providers don't implement |
| 983 |
ingress filtering, for whatever reasons, this may not be a |
| 984 |
satisfactory solution. |
| 985 |
|
| 986 |
o Implementation of ingress filtering by all IPv6 service providers |
| 987 |
would eliminate this attack, because src_v6 could not be spoofed |
| 988 |
as a 6to4 address. However, expecting this to happen may not be |
| 989 |
practical. |
| 990 |
|
| 991 |
o Proper implementation of security checks (see <a href="#section-5">Section 5</a>) both at |
| 992 |
the 6to4 relays and routers would eliminate an attack launched |
| 993 |
from an IPv4 node, except when the IPv4 source address was also |
| 994 |
spoofed -- but then the attacker would have been able to attack |
| 995 |
the ultimate destination directly. |
| 996 |
|
| 997 |
o Rate limiting traffic at the 6to4 relays. In a scenario where |
| 998 |
most of the traffic is passing through few 6to4 relays, these |
| 999 |
relays can implement traffic rate-limiting features and rate-limit |
| 1000 |
the traffic from 6to4 sites. |
| 1001 |
|
| 1002 |
|
| 1003 |
|
| 1004 |
<span class="grey">Savola & Patel Informational [Page 18]</span></pre> |
| 1005 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-19" ></span> |
| 1006 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1007 |
|
| 1008 |
|
| 1009 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 1010 |
|
| 1011 |
This particular attack can be mitigated by proper implementation of |
| 1012 |
security checks (which is quite straightforward) and ingress |
| 1013 |
filtering; when ingress filtering is not implemented, it is typically |
| 1014 |
easier to attack directly than through reflection -- unless "traffic |
| 1015 |
laundering" is an explicit goal of the attack. Therefore, this |
| 1016 |
attack does not seem very serious. |
| 1017 |
|
| 1018 |
<span class="h4"><a class="selflink" id="section-4.1.4" href="#section-4.1.4">4.1.4</a>. Local IPv4 Broadcast Attack</span> |
| 1019 |
|
| 1020 |
ATTACK DESCRIPTION |
| 1021 |
|
| 1022 |
This threat is applicable if the 6to4 router does not check whether |
| 1023 |
the IPv4 address to which it tries to send encapsulated IPv6 packets |
| 1024 |
is a local broadcast address or a multicast address. |
| 1025 |
|
| 1026 |
This threat is described in the specification [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>], and implementing |
| 1027 |
the checks eliminates this threat. However, as checks have not been |
| 1028 |
widely implemented, the threat is included here for completeness. |
| 1029 |
|
| 1030 |
There practically two kinds of attacks: when a local 6to4 user tries |
| 1031 |
to send packets to the address corresponding to the broadcast |
| 1032 |
address, and when someone is able to do so remotely. |
| 1033 |
|
| 1034 |
In the first option, assume that 9.0.0.255 is the 6to4 router's |
| 1035 |
broadcast address. After receiving the packet with a destination |
| 1036 |
address like "2002:0900:00ff::bbbb" from a local 6to4 node, if the |
| 1037 |
router doesn't check the destination address for subnet broadcast, it |
| 1038 |
would send the encapsulated protocol-41 packet to 9.0.0.255. This |
| 1039 |
would be received by all nodes in the subnet, and the responses would |
| 1040 |
be directed to the 6to4 router. |
| 1041 |
|
| 1042 |
Malicious sites may also embed forged 6to4 addresses in the DNS, use |
| 1043 |
of which by a 6to4 node would result in a local broadcast by the 6to4 |
| 1044 |
router. One way to perform this attack would be to send an HTML mail |
| 1045 |
containing a link to an invalid URL (for example, |
| 1046 |
http://[2002:0900:00ff::bbbb]/index.html) to all users in a 6to4 |
| 1047 |
technology based network. Opening of the mail simultaneously would |
| 1048 |
result in a broadcast storm. |
| 1049 |
|
| 1050 |
The second kind of attack is more complex: The attack can be |
| 1051 |
initiated by IPv4 nodes not belonging to the local network as long as |
| 1052 |
they can send traffic with invalid (for example 2002:0900:00ff::bbbb) |
| 1053 |
source address. The 6to4 router has to respond to the traffic by |
| 1054 |
sending ICMPv6 packets back to the source, (e.g., Hop Limit Exceeded |
| 1055 |
or Destination Unreachable). The packet would be as follows: |
| 1056 |
|
| 1057 |
|
| 1058 |
|
| 1059 |
|
| 1060 |
<span class="grey">Savola & Patel Informational [Page 19]</span></pre> |
| 1061 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-20" ></span> |
| 1062 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1063 |
|
| 1064 |
|
| 1065 |
src_v6 = 2002:0800:00ff::bbbb (broadcast address of the router) |
| 1066 |
dst_v6 = 2002:0800:0001::0001 (valid non-existent address) |
| 1067 |
|
| 1068 |
This is a DoS attack. |
| 1069 |
|
| 1070 |
EXTENSIONS |
| 1071 |
|
| 1072 |
The attacks could also be directed at non-local broadcast addresses, |
| 1073 |
but these would be so-called "IPv4 directed broadcasts", which have |
| 1074 |
(luckily enough) already been extensively blocked in the Internet. |
| 1075 |
|
| 1076 |
THREAT ANALYSIS AND SOLUTIONS/MITIGATION METHODS |
| 1077 |
|
| 1078 |
The attack is based on the premise that the 6to4 router has to send a |
| 1079 |
packet that embeds an invalid IPv4 address to an IPv6 address. Such |
| 1080 |
an attack is easily thwarted by ensuring that the 6to4 router does |
| 1081 |
not transmit packets to invalid IPv4 addresses. Specifically, |
| 1082 |
traffic should not be sent to broadcast or multicast IPv4 addresses. |
| 1083 |
|
| 1084 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 1085 |
|
| 1086 |
The first threat is similar to what is already possible with IPv4, |
| 1087 |
but IPv6 does not have broadcast addresses. |
| 1088 |
|
| 1089 |
The second, a more complex threat, is, similarly, also available in |
| 1090 |
IPv4. |
| 1091 |
|
| 1092 |
In consequence, the security does not seem to be significantly worse |
| 1093 |
than with IPv4, and even that is restricted to the site(s) with 6to4 |
| 1094 |
implementations that haven't been secured as described in <a href="#section-5">Section 5</a>. |
| 1095 |
|
| 1096 |
<span class="h3"><a class="selflink" id="section-4.2" href="#section-4.2">4.2</a>. Attacks on Native IPv6 Internet</span> |
| 1097 |
|
| 1098 |
This section describes attacks against native IPv6 Internet that |
| 1099 |
somehow leverage 6to4 architecture. Attacks against 6to4 nodes were |
| 1100 |
described in the previous section. |
| 1101 |
|
| 1102 |
6to4 and IPv4 nodes can access native IPv6 nodes through the 6to4 |
| 1103 |
relay routers. Thus, the 6to4 relays play a crucial role in any |
| 1104 |
attack on native IPv6 nodes by IPv4 nodes or 6to4 nodes. |
| 1105 |
|
| 1106 |
6to4 relays have only one significant security check they must |
| 1107 |
perform for general safety: When decapsulating IPv4 packets, they |
| 1108 |
check that 2002:V4ADDR::/48 and V4ADDR match in the source address. |
| 1109 |
If this is not done, several threats become more serious; in the |
| 1110 |
following analysis, it is assumed that such checks are implemented. |
| 1111 |
|
| 1112 |
|
| 1113 |
|
| 1114 |
|
| 1115 |
|
| 1116 |
<span class="grey">Savola & Patel Informational [Page 20]</span></pre> |
| 1117 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-21" ></span> |
| 1118 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1119 |
|
| 1120 |
|
| 1121 |
6to4 relay should not relay packets between 6to4 addresses. In |
| 1122 |
particular, packets decapsulated from 6to4 routers should not be |
| 1123 |
encapsulated toward 6to4 routers, as described in <a href="#section-5">Section 5</a>. |
| 1124 |
Similarly, packets with 6to4 source and destination addresses sent |
| 1125 |
from IPv6 nodes should not be relayed. It is not clear whether this |
| 1126 |
kind of check is typically implemented. The attacks described below |
| 1127 |
assume that such checks are not implemented. |
| 1128 |
|
| 1129 |
<span class="h4"><a class="selflink" id="section-4.2.1" href="#section-4.2.1">4.2.1</a>. Attacks with ND Messages</span> |
| 1130 |
|
| 1131 |
These attacks are the same as those employed against 6to4 routers, as |
| 1132 |
described in <a href="#section-4.1.1">Section 4.1.1</a>. |
| 1133 |
|
| 1134 |
<span class="h4"><a class="selflink" id="section-4.2.2" href="#section-4.2.2">4.2.2</a>. Spoofing Traffic to Native IPv6 Node</span> |
| 1135 |
|
| 1136 |
ATTACK DESCRIPTION |
| 1137 |
|
| 1138 |
The attacker - a malicious IPv4 or 6to4 node - can send packets with |
| 1139 |
a spoofed (or not spoofed) 6to4 source address to a native IPv6 node |
| 1140 |
to accomplish a DoS attack. |
| 1141 |
|
| 1142 |
The threat is similar to that involving 6to4 routers, as described in |
| 1143 |
<a href="#section-4.1.2">Section 4.1.2</a>. |
| 1144 |
|
| 1145 |
The difference here is that the attack is initiated by IPv4 or 6to4 |
| 1146 |
nodes. The source IPv6 address may or may not be spoofed. Note |
| 1147 |
that, as mentioned above, the relay is assumed to correlate the |
| 1148 |
source IPv4 address with the address embedded in the source IPv6 |
| 1149 |
address during decapsulation. A side effect is that all spoofed |
| 1150 |
traffic will have a 6to4 source address. |
| 1151 |
|
| 1152 |
EXTENSIONS |
| 1153 |
|
| 1154 |
Spoofed traffic may also be sent to native IPv6 nodes either by other |
| 1155 |
native IPv6 nodes, by 6to4 nodes, or by malicious IPv4 nodes to |
| 1156 |
conduct Reflection DoS on either native IPv6 nodes or 6to4 nodes. |
| 1157 |
|
| 1158 |
Certain native IPv6 networks may have a trivial ACL (Access Control |
| 1159 |
List) based firewall that allows traffic to pass through if it comes |
| 1160 |
from particular source(s). Such a firewalling mechanism can be |
| 1161 |
bypassed by address spoofing. This attack can therefore be used for |
| 1162 |
trivial ACL avoidance as well. These attacks might be hampered by |
| 1163 |
lost replies from the 6to4 node to the spoofed address. |
| 1164 |
|
| 1165 |
|
| 1166 |
|
| 1167 |
|
| 1168 |
|
| 1169 |
|
| 1170 |
|
| 1171 |
|
| 1172 |
<span class="grey">Savola & Patel Informational [Page 21]</span></pre> |
| 1173 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-22" ></span> |
| 1174 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1175 |
|
| 1176 |
|
| 1177 |
THREAT ANALYSIS AND SOLUTIONS/MITIGATION METHODS |
| 1178 |
|
| 1179 |
The Denial-of-Service attack based on traffic spoofing is not new; |
| 1180 |
the only twist is that traces of an attack are more easily lost. The |
| 1181 |
6to4 relay typically does not log IPv4 addresses (as they would be |
| 1182 |
treated as L2 addresses), and thus the source of the attack (if |
| 1183 |
launched from an IPv4 node) is lost. Because traces to the src_v4 |
| 1184 |
address are easily lost, these attacks can also be launched from IPv4 |
| 1185 |
nodes whose connections are ingress-filtered. |
| 1186 |
|
| 1187 |
These attacks might not be easy to perform and might be hampered |
| 1188 |
because of the following: |
| 1189 |
|
| 1190 |
o It might be difficult to launch such attacks from 6to4 nodes |
| 1191 |
because even if the 6to4 routers allow spoofing of the source IPv6 |
| 1192 |
address, the 6to4 relay would check whether the source V4ADDR is |
| 1193 |
the same as the one embedded in the source IPv6 address. Thus, |
| 1194 |
6to4 nodes will be forced to use the correct IPv6 prefix while |
| 1195 |
launching an attack, making it easy to close such attacks. |
| 1196 |
|
| 1197 |
o Packets may pass through choke point(s), namely a 6to4 relay. In |
| 1198 |
addition to physical limitations, there could be some sort of |
| 1199 |
traffic rate limiting mechanisms that may be implemented, and |
| 1200 |
these could tone down the attack. |
| 1201 |
|
| 1202 |
o For every packet sent, at most one reply packet is generated: |
| 1203 |
There is no amplification factor. |
| 1204 |
|
| 1205 |
Some of the mitigation methods for such attacks are as follows: |
| 1206 |
|
| 1207 |
1. Ingress filtering in the IPv4 Internet to prevent packets with a |
| 1208 |
spoofed IPv4 source from being transmitted. As the relay checks |
| 1209 |
that the 6to4 address embeds the IPv4 address, no spoofing can be |
| 1210 |
achieved unless IPv4 addresses can be spoofed. However, this |
| 1211 |
would probably be an unfeasible requirement. |
| 1212 |
|
| 1213 |
2. Security checks in the 6to4 relay. The 6to4 relay must drop |
| 1214 |
traffic (from 6to4 nodes, or IPv4 nodes) with non-6to4 addresses |
| 1215 |
as the source address, or for which the source IPv4 address does |
| 1216 |
not match the address embedded in the source IPv6 address. |
| 1217 |
|
| 1218 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 1219 |
|
| 1220 |
Compared to <a href="#section-4.1.2">Section 4.1.2</a>, which describes more serious threats, this |
| 1221 |
threat appears to be slightly more manageable. If the relays perform |
| 1222 |
proper decapsulation checks, the spoofing can only be achieved, to a |
| 1223 |
6to4 source address, when the IPv4 address is spoofable as well. |
| 1224 |
|
| 1225 |
|
| 1226 |
|
| 1227 |
|
| 1228 |
<span class="grey">Savola & Patel Informational [Page 22]</span></pre> |
| 1229 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-23" ></span> |
| 1230 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1231 |
|
| 1232 |
|
| 1233 |
<span class="h4"><a class="selflink" id="section-4.2.3" href="#section-4.2.3">4.2.3</a>. Reflecting Traffic to Native IPv6 Nodes</span> |
| 1234 |
|
| 1235 |
ATTACK DESCRIPTION |
| 1236 |
|
| 1237 |
These reflection attacks are similar to that involving 6to4 routers, |
| 1238 |
as described in <a href="#section-4.1.3">Section 4.1.3</a>. Traffic may be reflected off native |
| 1239 |
IPv6 nodes, or off 6to4 nodes. The attack can be initiated by one of |
| 1240 |
the following: |
| 1241 |
|
| 1242 |
o Native IPv6 nodes. These nodes can send invalid traffic with |
| 1243 |
spoofed native IPv6 addresses to valid 6to4 nodes. Replies from |
| 1244 |
the 6to4 nodes are part of a reflection attack. |
| 1245 |
|
| 1246 |
o IPv4 nodes. These nodes can send traffic with native IPv6 source |
| 1247 |
addresses (encapsulated by the IPv4 node itself into a protocol-41 |
| 1248 |
packet) to 6to4 nodes. Replies from the 6to4 nodes are part of a |
| 1249 |
reflection attack. |
| 1250 |
|
| 1251 |
o 6to4 nodes. These nodes can perform attacks similar to those by |
| 1252 |
IPv4 nodes, but this would require spoofing of the source address |
| 1253 |
at the 6to4 site before encapsulation, which is likely to be |
| 1254 |
difficult. |
| 1255 |
|
| 1256 |
When launched from a native IPv6 node, the traffic goes through 6to4 |
| 1257 |
relays twice, both before and after the reflection; when launched |
| 1258 |
from a 6to4/IPv4 node, the traffic goes through a relay only after |
| 1259 |
the reflection. |
| 1260 |
|
| 1261 |
EXTENSIONS |
| 1262 |
|
| 1263 |
A distributed reflection DoS can be performed if a large number of |
| 1264 |
native IPv6 nodes or IPv4/6to4 nodes are involved in sending spoofed |
| 1265 |
traffic with the same source IPv6 address. |
| 1266 |
|
| 1267 |
THREAT ANALYSIS AND SOLUTIONS/MITIGATION METHODS |
| 1268 |
|
| 1269 |
Some of the mitigation methods for such attacks are as follows: |
| 1270 |
|
| 1271 |
1. Attacks from the native IPv6 nodes could be stopped by |
| 1272 |
implementing ingress filtering in the IPv6 Internet; hopefully |
| 1273 |
this will become commonplace, but past experience of IPv4 ingress |
| 1274 |
filtering deployment (or lack thereof) does not promise much. |
| 1275 |
|
| 1276 |
2. Two measures are needed to stop or mitigate the attacks from IPv4 |
| 1277 |
nodes: 1) Implementing ingress filtering in the IPv4 internet, |
| 1278 |
and 2) logging IPv4 source addresses in the 6to4 router. |
| 1279 |
|
| 1280 |
|
| 1281 |
|
| 1282 |
|
| 1283 |
|
| 1284 |
<span class="grey">Savola & Patel Informational [Page 23]</span></pre> |
| 1285 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-24" ></span> |
| 1286 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1287 |
|
| 1288 |
|
| 1289 |
3. Attacks from 6to4 nodes in other sites can be stopped if the 6to4 |
| 1290 |
routers in those sites implement egress filtering. This could be |
| 1291 |
done by those sites, but the sites that are most likely to be |
| 1292 |
abused are typically also those most likely to neglect installing |
| 1293 |
appropriate filtering at their edges. |
| 1294 |
|
| 1295 |
4. The traffic passes through one or two relays, and traffic rate |
| 1296 |
limiting in the 6to4 relays might help tone down the reflection |
| 1297 |
attack. |
| 1298 |
|
| 1299 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 1300 |
|
| 1301 |
Even though there are means to mitigate it, the attack is still |
| 1302 |
rather efficient, especially when used by native IPv6 nodes with |
| 1303 |
spoofed addresses. Using 6to4 relays and routers could easily take |
| 1304 |
down the 6to4 relay system and/or provide an easy means for traffic |
| 1305 |
laundering. However, if the attack is intended to DoS the victim, |
| 1306 |
this can be achieved more smoothly by doing it directly (as the |
| 1307 |
source address spoofing was available as well). |
| 1308 |
|
| 1309 |
Therefore, the threat to the availability and stability of the 6to4 |
| 1310 |
relay system itself seems to be higher than to the native IPv6 |
| 1311 |
Internet. |
| 1312 |
|
| 1313 |
<span class="h4"><a class="selflink" id="section-4.2.4" href="#section-4.2.4">4.2.4</a>. Local IPv4 Broadcast Attack</span> |
| 1314 |
|
| 1315 |
This attack is similar to the ones employed against 6to4 routers, as |
| 1316 |
described in <a href="#section-4.1.4">Section 4.1.4</a>. There are slight differences with regard |
| 1317 |
to the source of the attacks. This attack can be initiated by: |
| 1318 |
|
| 1319 |
o native IPv6 nodes that may send traffic to the relay's subnet |
| 1320 |
broadcast address, and |
| 1321 |
|
| 1322 |
o IPv4 nodes that may send traffic with a spoofed source IP address |
| 1323 |
(to be the relay's broadcast address) to elicit replies (e.g., |
| 1324 |
ICMPv6 Hop Limit Exceeded) from the 6to4 relay to its local nodes. |
| 1325 |
|
| 1326 |
The first approach is more dangerous than those in <a href="#section-4.1.4">Section 4.1.4</a> |
| 1327 |
because it can be initiated by any IPv6 node (allowed to use the |
| 1328 |
relay); the approach is not limited to local users. |
| 1329 |
|
| 1330 |
The second approach is trickier and not really useful. For it to |
| 1331 |
succeed, the relay would have to accept native source addresses over |
| 1332 |
the 6to4 pseudo-interface (we did not assume this check was |
| 1333 |
implemented), as if coming from another relay, triggering an ICMPv6 |
| 1334 |
message to the relay's local IPv4 subnet. The former method is more |
| 1335 |
lucrative. |
| 1336 |
|
| 1337 |
|
| 1338 |
|
| 1339 |
|
| 1340 |
<span class="grey">Savola & Patel Informational [Page 24]</span></pre> |
| 1341 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-25" ></span> |
| 1342 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1343 |
|
| 1344 |
|
| 1345 |
EXTENSIONS |
| 1346 |
|
| 1347 |
None. |
| 1348 |
|
| 1349 |
THREAT ANALYSIS AND SOLUTIONS/MITIGATION METHODS |
| 1350 |
|
| 1351 |
The threat is restricted to the relay's local subnet and is fixed by |
| 1352 |
tightening the 6to4 security checks. |
| 1353 |
|
| 1354 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 1355 |
|
| 1356 |
This scenario is caused by 6to4, but fortunately the issue is not |
| 1357 |
serious. |
| 1358 |
|
| 1359 |
<span class="h4"><a class="selflink" id="section-4.2.5" href="#section-4.2.5">4.2.5</a>. Theft of Service</span> |
| 1360 |
|
| 1361 |
ATTACK DESCRIPTION |
| 1362 |
|
| 1363 |
The 6to4 relay administrators would often want to use some policy to |
| 1364 |
limit the use of the relay to specific 6to4 sites and/or specific |
| 1365 |
IPv6 sites. |
| 1366 |
|
| 1367 |
The policy control is usually enacted by applying restrictions to |
| 1368 |
where the routing information for 2002::/16 and/or 192.188.99.0/24 |
| 1369 |
(if the anycast address used [<a href="#ref-3" title=""An Anycast Prefix for 6to4 Relay Routers"">3</a>]) will spread. |
| 1370 |
|
| 1371 |
Some users may be able to use the service regardless of these |
| 1372 |
controls, by |
| 1373 |
|
| 1374 |
o configuring the address of the relay using its IPv4 address |
| 1375 |
instead of 192.88.99.1, or |
| 1376 |
|
| 1377 |
o using the routing header to route IPv6 packets to reach specific |
| 1378 |
6to4 relays. (Other routing tricks, such as using static routes, |
| 1379 |
may also be used.) |
| 1380 |
|
| 1381 |
EXTENSIONS |
| 1382 |
|
| 1383 |
None. |
| 1384 |
|
| 1385 |
THREAT ANALYSIS AND SOLUTIONS/MITIGATION METHODS |
| 1386 |
|
| 1387 |
Attempts to use the relay's IPv4 address instead of 192.88.99.1 can |
| 1388 |
be mitigated in the following ways: |
| 1389 |
|
| 1390 |
1. IPv4 domains should prevent use of the actual IPv4 address of the |
| 1391 |
relay instead of 192.88.99.1. |
| 1392 |
|
| 1393 |
|
| 1394 |
|
| 1395 |
|
| 1396 |
<span class="grey">Savola & Patel Informational [Page 25]</span></pre> |
| 1397 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-26" ></span> |
| 1398 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1399 |
|
| 1400 |
|
| 1401 |
2. Usage of access lists in the 6to4 relay to limit access. This is |
| 1402 |
only feasible if the number of IP networks the relay is supposed |
| 1403 |
to serve is relatively low. |
| 1404 |
|
| 1405 |
3. The 6to4 relay should filter out arriving tunneled packets with |
| 1406 |
protocol 41 (IPv6) that do not have 192.88.99.1 as the |
| 1407 |
destination address. |
| 1408 |
|
| 1409 |
The other threat, of using routing tricks in the IPv6 networks to |
| 1410 |
reach the 6to4 relay, has similar solutions: |
| 1411 |
|
| 1412 |
1. Usage of access lists in the relay to limit access. |
| 1413 |
|
| 1414 |
2. Filtering out the packets with a routing header (although this |
| 1415 |
may have other implications). |
| 1416 |
|
| 1417 |
3. Monitoring the source addresses going through the relay to |
| 1418 |
detect, e.g., peers setting up static routes. |
| 1419 |
|
| 1420 |
Routing Header is not specific to 6to4. The main thing one could do |
| 1421 |
with it here would be to select the relay. Some generic threats |
| 1422 |
about routing header use are described in [<a href="#ref-11" title=""Security of IPv6 Routing Header and Home Address Options"">11</a>]. |
| 1423 |
|
| 1424 |
As this threat does not have implications for anything other than the |
| 1425 |
organization providing 6to4 relay, it is not analyzed any further. |
| 1426 |
|
| 1427 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 1428 |
|
| 1429 |
These threats are specific to 6to4 relays (or in general anycast |
| 1430 |
services) and do not exist in networks without 6to4. |
| 1431 |
|
| 1432 |
<span class="h4"><a class="selflink" id="section-4.2.6" href="#section-4.2.6">4.2.6</a>. Relay Operators Seen as Source of Abuse</span> |
| 1433 |
|
| 1434 |
ATTACK DESCRIPTION |
| 1435 |
|
| 1436 |
Several attacks use 6to4 relays to anonymize the traffic; this often |
| 1437 |
results in packets being tunneled from the relay to a supposedly-6to4 |
| 1438 |
site. |
| 1439 |
|
| 1440 |
However, as was pointed out in <a href="#section-4.2">Section 4.2</a>, the IPv4 source address |
| 1441 |
used by the relay could, on a cursory look, be seen as the source of |
| 1442 |
these "protocol-41" attacks. |
| 1443 |
|
| 1444 |
This could cause a number of concerns for the operators deploying |
| 1445 |
6to4 relay service, including the following: |
| 1446 |
|
| 1447 |
o being contacted a lot (via email, phone, fax, or lawyers) on |
| 1448 |
suspected "abuse", |
| 1449 |
|
| 1450 |
|
| 1451 |
|
| 1452 |
<span class="grey">Savola & Patel Informational [Page 26]</span></pre> |
| 1453 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-27" ></span> |
| 1454 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1455 |
|
| 1456 |
|
| 1457 |
o having the whole IPv4 address range rejected as a source of abuse |
| 1458 |
or spam, causing outage to other operations as well, or |
| 1459 |
|
| 1460 |
o causing the whole IPv4 address range to be blacklisted in some |
| 1461 |
"spammer databases", if the relay were used for those purposes. |
| 1462 |
|
| 1463 |
This threat seems slightly similar to the outburst of SMTP abuse |
| 1464 |
caused by open relays but is more generic. |
| 1465 |
|
| 1466 |
EXTENSIONS |
| 1467 |
|
| 1468 |
None. |
| 1469 |
|
| 1470 |
THREAT ANALYSIS AND SOLUTIONS/MITIGATION METHODS |
| 1471 |
|
| 1472 |
This problem can be avoided (or, really, "made someone else's |
| 1473 |
problem") by using the 6to4 anycast address in 192.88.99.0/24 as the |
| 1474 |
source address. Blacklisting or rejecting this should not cause |
| 1475 |
problems to the other operations. |
| 1476 |
|
| 1477 |
Further, when someone files complaints to the owner of |
| 1478 |
192.88.99.0/24, depending on which registry they are querying, they |
| 1479 |
might get, for example: |
| 1480 |
|
| 1481 |
o knowledge that this is a special IANA address block, with no real |
| 1482 |
contact person, |
| 1483 |
|
| 1484 |
o knowledge that this is a special address block for <a href="./rfc3068">RFC 3068</a>, or |
| 1485 |
|
| 1486 |
o knowledge that this is a special address block for <a href="./rfc3068">RFC 3068</a>, and |
| 1487 |
that there are multiple entries by relay operators in the |
| 1488 |
database. |
| 1489 |
|
| 1490 |
Any of these, at least when processed by a human, should show that |
| 1491 |
the 6to4 relay is in fact innocent. Of course, this could result in |
| 1492 |
reports going to the closest anycast 6to4 relay as well, which had |
| 1493 |
nothing to do with the incident. |
| 1494 |
|
| 1495 |
However, the widespread usage of 192.88.99.1 as the source address |
| 1496 |
may make it more difficult to disambiguate the relays, which might be |
| 1497 |
a useful feature for debugging purposes. |
| 1498 |
|
| 1499 |
COMPARISON TO SITUATION WITHOUT 6to4 |
| 1500 |
|
| 1501 |
This threat is caused by 6to4 deployment but can be avoided, at least |
| 1502 |
in the short-term, by using 192.88.99.1 as the source address. |
| 1503 |
|
| 1504 |
|
| 1505 |
|
| 1506 |
|
| 1507 |
|
| 1508 |
<span class="grey">Savola & Patel Informational [Page 27]</span></pre> |
| 1509 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-28" ></span> |
| 1510 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1511 |
|
| 1512 |
|
| 1513 |
<span class="h3"><a class="selflink" id="section-4.3" href="#section-4.3">4.3</a>. Attacks on IPv4 Internet</span> |
| 1514 |
|
| 1515 |
There are two types of attacks on the IPv4 internet - spoofed |
| 1516 |
traffic, and reflection. These can be initiated by native IPv6 |
| 1517 |
nodes, 6to4 nodes, and IPv4 nodes. |
| 1518 |
|
| 1519 |
Attacks initiated by IPv4 nodes that send spoofed traffic, which |
| 1520 |
would not use the 6to4 infrastructure, are considered out of the |
| 1521 |
scope of this document. 6to4 infrastructure may be used in |
| 1522 |
reflection attacks initiated by IPv4 nodes. |
| 1523 |
|
| 1524 |
It is difficult for these attacks to be effective, as the traffic |
| 1525 |
sent out will be IPv6-in-IPv4. Such traffic will be rejected by most |
| 1526 |
IPv4 nodes unless they have implemented some sort of IPv6-in-IPv4 |
| 1527 |
tunneling. |
| 1528 |
|
| 1529 |
<span class="h3"><a class="selflink" id="section-4.4" href="#section-4.4">4.4</a>. Summary of the Attacks</span> |
| 1530 |
|
| 1531 |
Columns: |
| 1532 |
|
| 1533 |
o Section number. The section that describes the attack. |
| 1534 |
|
| 1535 |
o Attack name. |
| 1536 |
|
| 1537 |
o Initiator. The node that initiates the attack. |
| 1538 |
|
| 1539 |
* I_4 - IPv4 node |
| 1540 |
|
| 1541 |
* I_6 - native IPv6 node |
| 1542 |
|
| 1543 |
* 6to4 - 6to4 node |
| 1544 |
|
| 1545 |
* * - All of the above |
| 1546 |
|
| 1547 |
o Victim. The victim node |
| 1548 |
|
| 1549 |
* I_4 - IPv4 node |
| 1550 |
|
| 1551 |
* I_6 - native IPv6 node |
| 1552 |
|
| 1553 |
* 6to4 - 6to4 node |
| 1554 |
|
| 1555 |
* Relay - 6to4 relay |
| 1556 |
|
| 1557 |
* Router - 6to4 router |
| 1558 |
|
| 1559 |
|
| 1560 |
|
| 1561 |
|
| 1562 |
|
| 1563 |
|
| 1564 |
<span class="grey">Savola & Patel Informational [Page 28]</span></pre> |
| 1565 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-29" ></span> |
| 1566 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1567 |
|
| 1568 |
|
| 1569 |
o ToA. Type of Attack |
| 1570 |
|
| 1571 |
* D - DoS |
| 1572 |
|
| 1573 |
* R - Reflection DoS |
| 1574 |
|
| 1575 |
* T - Theft of Service |
| 1576 |
|
| 1577 |
o Fix. Specified who is responsible for fixing the attack. |
| 1578 |
|
| 1579 |
* 6 - The 6to4 developer and/or operator can completely mitigate |
| 1580 |
this attack. |
| 1581 |
|
| 1582 |
* 6* - The 6to4 developer and/or operator can partially mitigate |
| 1583 |
this attack. |
| 1584 |
|
| 1585 |
* E - This threat cannot be fixed by the 6to4 developer or the |
| 1586 |
6to4 operator. |
| 1587 |
|
| 1588 |
Summary of attacks on a 6to4 network: |
| 1589 |
|
| 1590 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1591 |
| Sec | Attack name |Initiator| Victim | ToA | Fix | |
| 1592 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1593 |
| 4.1.1 | Attacks with ND | I_4 | Router | D | 6 | |
| 1594 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1595 |
| 4.1.2 | Spoofing Traffic | I_4,I_6 | 6to4 | D | E | |
| 1596 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1597 |
| 4.1.3 | Reflection Attacks | * | 6to4 | R | 6* | |
| 1598 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1599 |
| 4.1.4 | Local IPv4 Broadcast | * | Router | D | 6 | |
| 1600 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1601 |
|
| 1602 |
Figure 9 |
| 1603 |
|
| 1604 |
|
| 1605 |
|
| 1606 |
|
| 1607 |
|
| 1608 |
|
| 1609 |
|
| 1610 |
|
| 1611 |
|
| 1612 |
|
| 1613 |
|
| 1614 |
|
| 1615 |
|
| 1616 |
|
| 1617 |
|
| 1618 |
|
| 1619 |
|
| 1620 |
<span class="grey">Savola & Patel Informational [Page 29]</span></pre> |
| 1621 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-30" ></span> |
| 1622 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1623 |
|
| 1624 |
|
| 1625 |
Summary of attacks on the native IPv6 internet: |
| 1626 |
|
| 1627 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1628 |
| Sec | Attack name |Initiator| Victim | ToA | Fix | |
| 1629 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1630 |
| 4.2.1 | Attacks with ND | I_4 | Relay | D | 6 | |
| 1631 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1632 |
| 4.2.2 | Spoofing Traffic | I_4,6to4| I_6 | D | 6* | |
| 1633 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1634 |
| 4.2.3 | Reflection Attacks | * | I_6 | R | 6* | |
| 1635 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1636 |
| 4.2.4 | Local IPv4 Broadcast | * | Relay | D | 6 | |
| 1637 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1638 |
| 4.2.5 | Theft of Service | 6to4 | Relay | T | 6 | |
| 1639 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1640 |
| 4.2.6 | Relay Operators ... | - | - | D | 1) | |
| 1641 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1642 |
|
| 1643 |
Figure 10 |
| 1644 |
|
| 1645 |
Notes: |
| 1646 |
|
| 1647 |
1) This attack is a side-effect of the other attacks and thus does |
| 1648 |
not have any Initiator, Victim, and Fix. It is a Denial of Service |
| 1649 |
attack not on the network but on the organization in-charge of the |
| 1650 |
relay. |
| 1651 |
|
| 1652 |
Summary of attacks on IPv4 internet: |
| 1653 |
|
| 1654 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1655 |
| Sec | Attack name |Initiator| Victim | ToA | Fix | |
| 1656 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1657 |
| 4.3 | Spoofing Traffic | * | I_4 | D | 6* | |
| 1658 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1659 |
| 4.3 | Reflection Attacks | * | I_4 | R | 6* | |
| 1660 |
+-------+----------------------+---------+----------+-----+-----+ |
| 1661 |
|
| 1662 |
Figure 11 |
| 1663 |
|
| 1664 |
<span class="h2"><a class="selflink" id="section-5" href="#section-5">5</a>. Implementing Proper Security Checks in 6to4</span> |
| 1665 |
|
| 1666 |
This section describes several ways to implement the security checks |
| 1667 |
required or implied by the specification [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>] or augmented by this |
| 1668 |
memo. These do not, in general, protect against most of the threats |
| 1669 |
listed above in the "Threat Analysis" section. They are only |
| 1670 |
prerequisites for a relatively safe and simple 6to4 implementation. |
| 1671 |
|
| 1672 |
|
| 1673 |
|
| 1674 |
|
| 1675 |
|
| 1676 |
<span class="grey">Savola & Patel Informational [Page 30]</span></pre> |
| 1677 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-31" ></span> |
| 1678 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1679 |
|
| 1680 |
|
| 1681 |
Note that, in general, the 6to4 router or relay does not know whether |
| 1682 |
it is acting as a router or relay. It would be possible to include a |
| 1683 |
toggle to specify the behaviour, to be used when, e.g., the interface |
| 1684 |
is brought up, but as of February 2004, no implementations were known |
| 1685 |
to do that. Therefore, the checks are described as that which works |
| 1686 |
independently of whether the node is a router or relay. |
| 1687 |
|
| 1688 |
<span class="h3"><a class="selflink" id="section-5.1" href="#section-5.1">5.1</a>. Encapsulating IPv6 into IPv4</span> |
| 1689 |
|
| 1690 |
The checks described in this section are to be performed when |
| 1691 |
encapsulating IPv6 into IPv4. |
| 1692 |
|
| 1693 |
The encapsulation rules are mainly designed to keep implementors from |
| 1694 |
"shooting themselves in the foot." For example, the source address |
| 1695 |
check would verify that the packet will be acceptable to the |
| 1696 |
decapsulator, or the sanity checks would ensure that addresses |
| 1697 |
derived from private addresses are not used (which would be equally |
| 1698 |
unacceptable). |
| 1699 |
|
| 1700 |
src_v6 and dst_v6 MUST pass ipv6-sanity checks (see below) else drop |
| 1701 |
if prefix (src_v6) == 2002::/16 |
| 1702 |
ipv4 address embedded in src_v6 MUST match src_v4 |
| 1703 |
else if prefix (dst_v6) == 2002::/16 |
| 1704 |
dst_v4 SHOULD NOT be assigned to the router |
| 1705 |
else |
| 1706 |
drop |
| 1707 |
/* we somehow got a native-native ipv6 packet */ |
| 1708 |
fi |
| 1709 |
accept |
| 1710 |
|
| 1711 |
<span class="h3"><a class="selflink" id="section-5.2" href="#section-5.2">5.2</a>. Decapsulating IPv4 into IPv6</span> |
| 1712 |
|
| 1713 |
The checks described in this section are to be performed when |
| 1714 |
decapsulating IPv4 into IPv6. They will be performed in both the |
| 1715 |
6to4 router and relay. |
| 1716 |
|
| 1717 |
src_v4 and dst_v4 MUST pass ipv4-sanity checks, else drop |
| 1718 |
src_v6 and dst_v6 MUST pass ipv6-sanity checks, else drop |
| 1719 |
if prefix (dst_v6) == 2002::/16 |
| 1720 |
ipv4 address embedded in dst_v6 MUST match dst_v4 |
| 1721 |
if prefix (src_v6) == 2002::/16 |
| 1722 |
ipv4 address embedded in src_v6 MUST match src_v4 |
| 1723 |
dst_v4 SHOULD be assigned to the router |
| 1724 |
fi |
| 1725 |
elif prefix (src_v6) == 2002::/16 |
| 1726 |
ipv4 address embedded in src_v6 MUST match src_v4 |
| 1727 |
dst_v4 SHOULD be assigned to the router (see notes below) |
| 1728 |
|
| 1729 |
|
| 1730 |
|
| 1731 |
|
| 1732 |
<span class="grey">Savola & Patel Informational [Page 31]</span></pre> |
| 1733 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-32" ></span> |
| 1734 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1735 |
|
| 1736 |
|
| 1737 |
else |
| 1738 |
drop |
| 1739 |
/* the we somehow got a native-native ipv6 packet */ |
| 1740 |
fi |
| 1741 |
accept |
| 1742 |
|
| 1743 |
<span class="h3"><a class="selflink" id="section-5.3" href="#section-5.3">5.3</a>. IPv4 and IPv6 Sanity Checks</span> |
| 1744 |
|
| 1745 |
The encapsulation and decapsulation checks include certain sanity |
| 1746 |
checks for both IPv4 and IPv6. These are described here in detail. |
| 1747 |
|
| 1748 |
<span class="h4"><a class="selflink" id="section-5.3.1" href="#section-5.3.1">5.3.1</a>. IPv4</span> |
| 1749 |
|
| 1750 |
IPv4 address MUST be a global unicast address, as required by the |
| 1751 |
6to4 specification. The disallowed addresses include those defined |
| 1752 |
in [<a href="#ref-14" title=""Requirements for IP Version 4 Routers"">14</a>], and others widely used and known not to be global. These |
| 1753 |
are |
| 1754 |
|
| 1755 |
o 0.0.0.0/8 (the system has no address assigned yet) |
| 1756 |
|
| 1757 |
o 10.0.0.0/8 (private) |
| 1758 |
|
| 1759 |
o 127.0.0.0/8 (loopback) |
| 1760 |
|
| 1761 |
o 172.16.0.0/12 (private) |
| 1762 |
|
| 1763 |
o 192.168.0.0/16 (private) |
| 1764 |
|
| 1765 |
o 169.254.0.0/16 (IANA Assigned DHCP link-local) |
| 1766 |
|
| 1767 |
o 224.0.0.0/4 (multicast) |
| 1768 |
|
| 1769 |
o 240.0.0.0/4 (reserved and broadcast) |
| 1770 |
|
| 1771 |
In addition, the address MUST NOT be any of the system's broadcast |
| 1772 |
addresses. This is especially important if the implementation is |
| 1773 |
made so that it can |
| 1774 |
|
| 1775 |
o receive and process encapsulated IPv4 packets arriving at its |
| 1776 |
broadcast addresses, or |
| 1777 |
|
| 1778 |
o send encapsulated IPv4 packets to one of its broadcast addresses. |
| 1779 |
|
| 1780 |
|
| 1781 |
|
| 1782 |
|
| 1783 |
|
| 1784 |
|
| 1785 |
|
| 1786 |
|
| 1787 |
|
| 1788 |
<span class="grey">Savola & Patel Informational [Page 32]</span></pre> |
| 1789 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-33" ></span> |
| 1790 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1791 |
|
| 1792 |
|
| 1793 |
<span class="h4"><a class="selflink" id="section-5.3.2" href="#section-5.3.2">5.3.2</a>. IPv6</span> |
| 1794 |
|
| 1795 |
IPv6 address MUST NOT be |
| 1796 |
|
| 1797 |
o 0::/16 (compatible, mapped addresses, loopback, unspecified, ...) |
| 1798 |
|
| 1799 |
o fe80::/10 (link-local) |
| 1800 |
|
| 1801 |
o fec0::/10 (site-local) |
| 1802 |
|
| 1803 |
o ff00::/8 (any multicast) |
| 1804 |
|
| 1805 |
Note: Only link-local multicast would be strictly required, but it is |
| 1806 |
believed that multicast with 6to4 will not be feasible, so it has |
| 1807 |
been disallowed as well. |
| 1808 |
|
| 1809 |
In addition, it MUST be checked that equivalent 2002:V4ADDR::/48 |
| 1810 |
checks, where V4ADDR is any of the above IPv4 addresses, will not be |
| 1811 |
passed. |
| 1812 |
|
| 1813 |
<span class="h4"><a class="selflink" id="section-5.3.3" href="#section-5.3.3">5.3.3</a>. Optional Ingress Filtering</span> |
| 1814 |
|
| 1815 |
In addition, the implementation in the 6to4 router may perform some |
| 1816 |
form of ingress filtering (e.g., Unicast Reverse Path Forwarding |
| 1817 |
checks). For example, if the 6to4 router has multiple interfaces, of |
| 1818 |
which some are "internal", receiving either IPv4 or IPv6 packets with |
| 1819 |
source address belonging to any of these internal networks from the |
| 1820 |
Internet might be disallowed. |
| 1821 |
|
| 1822 |
If these checks are implemented and enabled by default, it's |
| 1823 |
recommended that there be a toggle to disable them if needed. |
| 1824 |
|
| 1825 |
<span class="h4"><a class="selflink" id="section-5.3.4" href="#section-5.3.4">5.3.4</a>. Notes about the Checks</span> |
| 1826 |
|
| 1827 |
The rule "dst_v4 SHOULD be assigned to the router" is not needed if |
| 1828 |
the 6to4 router implementation only accepts and processes |
| 1829 |
encapsulated IPv4 packets arriving to its unicast IPv4 addresses, and |
| 1830 |
when the destination address is known to be a local broadcast |
| 1831 |
address, it does not try to encapsulate and send packets to it. (See |
| 1832 |
Sections <a href="#section-4.1.4">4.1.4</a> and 4.2.4 about this threat.) |
| 1833 |
|
| 1834 |
Some checks, especially the IPv4/IPv6 Sanity Checks, could be at |
| 1835 |
least partially implementable with system-level access lists, if one |
| 1836 |
would like to avoid placing too many restrictions in the 6to4 |
| 1837 |
implementation itself. This depends on how many hooks are in place |
| 1838 |
for the access lists. In practice, it seems that this could not be |
| 1839 |
done effectively enough unless the access list mechanism is able to |
| 1840 |
parse the encapsulated packets. |
| 1841 |
|
| 1842 |
|
| 1843 |
|
| 1844 |
<span class="grey">Savola & Patel Informational [Page 33]</span></pre> |
| 1845 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-34" ></span> |
| 1846 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1847 |
|
| 1848 |
|
| 1849 |
<span class="h2"><a class="selflink" id="section-6" href="#section-6">6</a>. Issues in 6to4 Implementation and Use</span> |
| 1850 |
|
| 1851 |
This section tries to give an overview of some of the problems 6to4 |
| 1852 |
implementations face, and the kind of generic problems the 6to4 users |
| 1853 |
could come up with. |
| 1854 |
|
| 1855 |
<span class="h3"><a class="selflink" id="section-6.1" href="#section-6.1">6.1</a>. Implementation Considerations with Automatic Tunnels</span> |
| 1856 |
|
| 1857 |
There is a problem with multiple transition mechanisms if strict |
| 1858 |
security checks are implemented. This may vary a bit from |
| 1859 |
implementation to implementation. |
| 1860 |
|
| 1861 |
Consider three mechanisms using automatic tunneling: 6to4, ISATAP |
| 1862 |
[<a href="#ref-15" title=""Intra-Site Automatic Tunnel Addressing Protocol (ISATAP)"">15</a>], and Automatic Tunneling using Compatible Addresses [<a href="#ref-4" title=""Transition Mechanisms for IPv6 Hosts and Routers"">4</a>] |
| 1863 |
(currently removed [<a href="#ref-10" title=""Basic Transition Mechanisms for IPv6 Hosts and Routers"">10</a>] but typically still supported). All of these |
| 1864 |
use IP-IP (protocol 41) [<a href="#ref-16" title=""IP in IP Tunneling"">16</a>] IPv4 encapsulation with, more or less, a |
| 1865 |
pseudo-interface. |
| 1866 |
|
| 1867 |
When a router, which has any two of these enabled, receives an IPv4 |
| 1868 |
encapsulated IPv6 packet |
| 1869 |
|
| 1870 |
src_v6 = 2001:db8::1 |
| 1871 |
dst_v6 = 2002:1010:1010::2 |
| 1872 |
src_v4 = 10.0.0.1 |
| 1873 |
dst_v4 = 20.20.20.20 |
| 1874 |
|
| 1875 |
What can it do? How should it decide which transition mechanism this |
| 1876 |
belongs to; there is no "transition mechanism number" in the IPv6 or |
| 1877 |
IPv4 header to signify this. (This can also be viewed as a |
| 1878 |
flexibility benefit.) |
| 1879 |
|
| 1880 |
Without any kind of security checks (in any of the implemented |
| 1881 |
methods), these often just "work", as the mechanisms aren't |
| 1882 |
differentiated but handled in "one big lump". |
| 1883 |
|
| 1884 |
Configured tunneling [<a href="#ref-4" title=""Transition Mechanisms for IPv6 Hosts and Routers"">4</a>] does not suffer from this, as it is |
| 1885 |
point-to-point and based on src_v6/dst_v6 pairs of both IPv4 and IPv6 |
| 1886 |
addresses, so the tunnel interface can be logically deduced. |
| 1887 |
|
| 1888 |
Solutions for this include 1) not using more than one automatic |
| 1889 |
tunneling mechanism in a node and 2) binding different mechanisms to |
| 1890 |
different IPv4 addresses. |
| 1891 |
|
| 1892 |
|
| 1893 |
|
| 1894 |
|
| 1895 |
|
| 1896 |
|
| 1897 |
|
| 1898 |
|
| 1899 |
|
| 1900 |
<span class="grey">Savola & Patel Informational [Page 34]</span></pre> |
| 1901 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-35" ></span> |
| 1902 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1903 |
|
| 1904 |
|
| 1905 |
<span class="h3"><a class="selflink" id="section-6.2" href="#section-6.2">6.2</a>. A Different Model for 6to4 Deployment</span> |
| 1906 |
|
| 1907 |
Even though this was already discussed in <a href="#section-4.1.2">Section 4.1.2</a>, it bears |
| 1908 |
some additional elaboration, as it was the only problem that cannot |
| 1909 |
be even partially solved using the current deployment model. There |
| 1910 |
are some mitigation methods. |
| 1911 |
|
| 1912 |
6to4 routers receive traffic from non-6to4 ("native") sources via |
| 1913 |
6to4 relays. 6to4 routers have no way of matching the IPv4 source |
| 1914 |
address of the relay with the non-6to4 IPv6 address of the source. |
| 1915 |
Consequently, anyone can spoof any non-6to4 IPv6 address by sending |
| 1916 |
traffic, encapsulated, directly to 6to4 routers. |
| 1917 |
|
| 1918 |
It could be possible to turn the deployment assumptions of 6to4 |
| 1919 |
around a bit to eliminate some threats caused by untrusted 6to4 |
| 1920 |
relays: |
| 1921 |
|
| 1922 |
o Every dual-stack site (or even ISP) would be required to have its |
| 1923 |
own 6to4 relay. (This assumes that IPv6-only is so far away that |
| 1924 |
6to4 would be retired by that point.) That is, there would not be |
| 1925 |
third-party relays, and 2002::/16 and 192.88.99.0/24 routes would |
| 1926 |
not need to be advertised globally. |
| 1927 |
|
| 1928 |
o The security implications of 6to4 use could be pushed back to the |
| 1929 |
level of trust inside the site or ISP (or their acceptable use |
| 1930 |
policies). This is something that the sites and ISPs should |
| 1931 |
already be familiar with already. |
| 1932 |
|
| 1933 |
However, this presents a number of problems: |
| 1934 |
|
| 1935 |
This model would shift most of the burden of supporting 6to4 to IPv6 |
| 1936 |
sites that don't employ or use 6to4 at all, i.e., "those who deploy |
| 1937 |
proper native dual-stack." It could be argued that the deployment |
| 1938 |
pain should be borne by 6to4 users, not by the others. |
| 1939 |
|
| 1940 |
The main advantage of 6to4 is easy deployment and free relays. This |
| 1941 |
would require that everyone the 6to4 sites wish to communicate with |
| 1942 |
implement these measures. |
| 1943 |
|
| 1944 |
The model would not fix the "relay spoofing problem", unless |
| 1945 |
everybody also deployed 6to4 addresses on the nodes (alongside with |
| 1946 |
native addresses, if necessary), which would in turn change 6to4 to |
| 1947 |
operate without relays completely. |
| 1948 |
|
| 1949 |
|
| 1950 |
|
| 1951 |
|
| 1952 |
|
| 1953 |
|
| 1954 |
|
| 1955 |
|
| 1956 |
<span class="grey">Savola & Patel Informational [Page 35]</span></pre> |
| 1957 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-36" ></span> |
| 1958 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 1959 |
|
| 1960 |
|
| 1961 |
<span class="h2"><a class="selflink" id="section-7" href="#section-7">7</a>. Security Considerations</span> |
| 1962 |
|
| 1963 |
This document discusses security considerations of 6to4. |
| 1964 |
|
| 1965 |
Even if proper checks are implemented, there are a large number of |
| 1966 |
different security threats; these threats are analyzed in <a href="#section-4">Section 4</a>. |
| 1967 |
|
| 1968 |
There are mainly four classes of potential problem sources: |
| 1969 |
|
| 1970 |
1. 6to4 routers not being able to identify whether relays are |
| 1971 |
legitimate |
| 1972 |
|
| 1973 |
2. Wrong or impartially implemented 6to4 router or relay security |
| 1974 |
checks |
| 1975 |
|
| 1976 |
3. 6to4 architecture used to participate in DoS or reflected DoS |
| 1977 |
attacks or made to participate in "packet laundering", i.e., |
| 1978 |
making another attack harder to trace |
| 1979 |
|
| 1980 |
4. 6to4 relays being subject to "administrative abuse" e.g., theft |
| 1981 |
of service or being seen as a source of abuse. |
| 1982 |
|
| 1983 |
The first is the toughest problem, still under research. The second |
| 1984 |
can be fixed by ensuring the correctness of implementations; this is |
| 1985 |
important. The third is also a very difficult problem, impossible to |
| 1986 |
solve completely; therefore it is important to be able to analyze |
| 1987 |
whether this results in a significant increase of threats. The |
| 1988 |
fourth problem seems to have feasible solutions. |
| 1989 |
|
| 1990 |
These are analyzed in detail in "Threat Analysis", in <a href="#section-4">Section 4</a>. |
| 1991 |
|
| 1992 |
<span class="h2"><a class="selflink" id="section-8" href="#section-8">8</a>. Acknowledgments</span> |
| 1993 |
|
| 1994 |
Some issues were first brought up by Itojun Hagino in [<a href="#ref-17" title=""Possible abuse against IPv6 transition technologies"">17</a>], and Alain |
| 1995 |
Durand introduced one specific problem at IETF51 in August 2001 |
| 1996 |
(though there was some discussion on the list prior to that); these |
| 1997 |
two gave the authors the push to start looking into the details of |
| 1998 |
securing 6to4. |
| 1999 |
|
| 2000 |
Alexey Kuznetsov brought up the implementation problem with IPv6 |
| 2001 |
martian checks. Christian Huitema formulated the rules that rely on |
| 2002 |
6to4 relays using only anycast. Keith Moore brought up the point |
| 2003 |
about reduced flexibility. Brian Carpenter, Tony Hain, and Vladislav |
| 2004 |
Yasevich are acknowledged for lengthy discussions. Alain Durand |
| 2005 |
reminded the authors about relay spoofing problems. Brian Carpenter |
| 2006 |
reminded the authors about the BGP-based 6to4 router model. |
| 2007 |
Christian Huitema gave a push for a more complete threat analysis. |
| 2008 |
Itojun Hagino spelled out the operators' fears about 6to4 relay |
| 2009 |
|
| 2010 |
|
| 2011 |
|
| 2012 |
<span class="grey">Savola & Patel Informational [Page 36]</span></pre> |
| 2013 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-37" ></span> |
| 2014 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 2015 |
|
| 2016 |
|
| 2017 |
abuse. Rob Austein brought up the idea of a different 6to4 |
| 2018 |
deployment model. |
| 2019 |
|
| 2020 |
In the latter phase, discussions with Christian Huitema, Brian |
| 2021 |
Carpenter, and Alain Durand were helpful when improving the document. |
| 2022 |
|
| 2023 |
David Malone, Iljitsch van Beijnum, and Tim Chown gave feedback on |
| 2024 |
the document. |
| 2025 |
|
| 2026 |
<span class="h2"><a class="selflink" id="section-9" href="#section-9">9</a>. References</span> |
| 2027 |
|
| 2028 |
<span class="h3"><a class="selflink" id="section-9.1" href="#section-9.1">9.1</a>. Normative References</span> |
| 2029 |
|
| 2030 |
[<a id="ref-1">1</a>] Carpenter, B. and K. Moore, "Connection of IPv6 Domains via IPv4 |
| 2031 |
Clouds", <a href="./rfc3056">RFC 3056</a>, February 2001. |
| 2032 |
|
| 2033 |
[<a id="ref-2">2</a>] Bradner, S., "Key words for use in RFCs to Indicate Requirement |
| 2034 |
Levels", <a href="https://www.rfc-editor.org/bcp/bcp14">BCP 14</a>, <a href="./rfc2119">RFC 2119</a>, March 1997. |
| 2035 |
|
| 2036 |
[<a id="ref-3">3</a>] Huitema, C., "An Anycast Prefix for 6to4 Relay Routers", <a href="./rfc3068">RFC</a> |
| 2037 |
<a href="./rfc3068">3068</a>, June 2001. |
| 2038 |
|
| 2039 |
<span class="h3"><a class="selflink" id="section-9.2" href="#section-9.2">9.2</a>. Informative References</span> |
| 2040 |
|
| 2041 |
[<a id="ref-4">4</a>] Gilligan, R. and E. Nordmark, "Transition Mechanisms for IPv6 |
| 2042 |
Hosts and Routers", <a href="./rfc2893">RFC 2893</a>, August 2000. |
| 2043 |
|
| 2044 |
[<a id="ref-5">5</a>] IANA, "Special-Use IPv4 Addresses", <a href="./rfc3330">RFC 3330</a>, September 2002. |
| 2045 |
|
| 2046 |
[<a id="ref-6">6</a>] Rekhter, Y. and T. Li, "A Border Gateway Protocol 4 (BGP-4)", |
| 2047 |
<a href="./rfc1771">RFC 1771</a>, March 1995. |
| 2048 |
|
| 2049 |
[<a id="ref-7">7</a>] Draves, R., "Default Address Selection for Internet Protocol |
| 2050 |
version 6 (IPv6)", <a href="./rfc3484">RFC 3484</a>, February 2003. |
| 2051 |
|
| 2052 |
[<a id="ref-8">8</a>] Nikander, P., Kempf, J., and E. Nordmark, "IPv6 Neighbor |
| 2053 |
Discovery (ND) Trust Models and Threats", <a href="./rfc3756">RFC 3756</a>, May 2004. |
| 2054 |
|
| 2055 |
[<a id="ref-9">9</a>] Arkko, J., Kempf, J., Sommerfeld, B., Zill, B., and P. Nikander, |
| 2056 |
"SEcure Neighbor Discovery (SEND)", Work in Progress, July 2004. |
| 2057 |
|
| 2058 |
[<a id="ref-10">10</a>] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms for |
| 2059 |
IPv6 Hosts and Routers", Work in Progress, September 2004. |
| 2060 |
|
| 2061 |
[<a id="ref-11">11</a>] Savola, P., "Security of IPv6 Routing Header and Home Address |
| 2062 |
Options", Work in Progress, March 2002. |
| 2063 |
|
| 2064 |
|
| 2065 |
|
| 2066 |
|
| 2067 |
|
| 2068 |
<span class="grey">Savola & Patel Informational [Page 37]</span></pre> |
| 2069 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-38" ></span> |
| 2070 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 2071 |
|
| 2072 |
|
| 2073 |
[<a id="ref-12">12</a>] Ferguson, P. and D. Senie, "Network Ingress Filtering: Defeating |
| 2074 |
Denial of Service Attacks which employ IP Source Address |
| 2075 |
Spoofing", <a href="https://www.rfc-editor.org/bcp/bcp38">BCP 38</a>, <a href="./rfc2827">RFC 2827</a>, May 2000. |
| 2076 |
|
| 2077 |
[<a id="ref-13">13</a>] Bellovin, S., Leech, M. and T. Taylor, "ICMP Traceback |
| 2078 |
Messages", Work in Progress, February 2003. |
| 2079 |
|
| 2080 |
[<a id="ref-14">14</a>] Baker, F., "Requirements for IP Version 4 Routers", <a href="./rfc1812">RFC 1812</a>, |
| 2081 |
June 1995. |
| 2082 |
|
| 2083 |
[<a id="ref-15">15</a>] Templin, F., Gleeson, T., Talwar, M. and D. Thaler, "Intra-Site |
| 2084 |
Automatic Tunnel Addressing Protocol (ISATAP)", Work in |
| 2085 |
Progress, May 2004. |
| 2086 |
|
| 2087 |
[<a id="ref-16">16</a>] Simpson, W., "IP in IP Tunneling", <a href="./rfc1853">RFC 1853</a>, October 1995. |
| 2088 |
|
| 2089 |
[<a id="ref-17">17</a>] Hagino, J., "Possible abuse against IPv6 transition |
| 2090 |
technologies", Work in Progress, July 2000. |
| 2091 |
|
| 2092 |
|
| 2093 |
|
| 2094 |
|
| 2095 |
|
| 2096 |
|
| 2097 |
|
| 2098 |
|
| 2099 |
|
| 2100 |
|
| 2101 |
|
| 2102 |
|
| 2103 |
|
| 2104 |
|
| 2105 |
|
| 2106 |
|
| 2107 |
|
| 2108 |
|
| 2109 |
|
| 2110 |
|
| 2111 |
|
| 2112 |
|
| 2113 |
|
| 2114 |
|
| 2115 |
|
| 2116 |
|
| 2117 |
|
| 2118 |
|
| 2119 |
|
| 2120 |
|
| 2121 |
|
| 2122 |
|
| 2123 |
|
| 2124 |
<span class="grey">Savola & Patel Informational [Page 38]</span></pre> |
| 2125 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-39" ></span> |
| 2126 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 2127 |
|
| 2128 |
|
| 2129 |
<span class="h2"><a class="selflink" id="appendix-A" href="#appendix-A">Appendix A</a>. Some Trivial Attack Scenarios Outlined</span> |
| 2130 |
|
| 2131 |
Here, a few trivial attack scenarios are outlined -- ones that are |
| 2132 |
prevented by implementing checks listed in [<a href="#ref-1" title=""Connection of IPv6 Domains via IPv4 Clouds"">1</a>] or in <a href="#section-6">section 6</a>. |
| 2133 |
|
| 2134 |
When two 6to4 routers send traffic to each others' domains, the |
| 2135 |
packet sent by RA to RB resembles the following: |
| 2136 |
|
| 2137 |
src_v6 = 2002:0800:0001::aaaa |
| 2138 |
dst_v6 = 2002:0800:0002::bbbb |
| 2139 |
src_v4 = 8.0.0.1 (added when encapsulated to IPv4) |
| 2140 |
dst_v4 = 8.0.0.2 (added when encapsulated to IPv4) |
| 2141 |
|
| 2142 |
When the packet is received by IPv4 stack on RB, it will be |
| 2143 |
decapsulated so that only src_v6 and dst_v6 remain, as originally |
| 2144 |
sent by RA: |
| 2145 |
|
| 2146 |
src_v6 = 2002:0800:0001::aaaa |
| 2147 |
dst_v6 = 2002:0800:0002::bbbb |
| 2148 |
|
| 2149 |
As every other node is just one hop away (IPv6-wise) and the |
| 2150 |
link-layer (IPv4) addresses are lost, this may open many |
| 2151 |
possibilities for misuse. |
| 2152 |
|
| 2153 |
As an example, unidirectional IPv6 spoofing is made trivial because |
| 2154 |
nobody can check (without delving into IP-IP packets) whether the |
| 2155 |
encapsulated IPv6 addresses were authentic. (With native IPv6, this |
| 2156 |
can be done by, e.g., RPF-like mechanisms or access lists in upstream |
| 2157 |
routers.) |
| 2158 |
|
| 2159 |
src_v6 = 2002:1234:5678::aaaa (forged) |
| 2160 |
dst_v6 = 2002:0800:0002::bbbb |
| 2161 |
src_v4 = 8.0.0.1 (added when encapsulated to IPv4) |
| 2162 |
dst_v4 = 8.0.0.2 (added when encapsulated to IPv4) |
| 2163 |
|
| 2164 |
A similar attack with "src" being the native address is made |
| 2165 |
possible, even with the security checks, by having the sender node |
| 2166 |
pretend to be a 6to4 relay router. |
| 2167 |
|
| 2168 |
More worries come into the picture if, e.g., |
| 2169 |
|
| 2170 |
src_v6 = ::ffff:[some trusted IPv4 in a private network] |
| 2171 |
src_v6/dst_v6 = ::ffff:127.0.0.1 |
| 2172 |
src_v6/dst_v6 = ::1 |
| 2173 |
src_v6/dst_v6 = ... |
| 2174 |
|
| 2175 |
|
| 2176 |
|
| 2177 |
|
| 2178 |
|
| 2179 |
|
| 2180 |
<span class="grey">Savola & Patel Informational [Page 39]</span></pre> |
| 2181 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-40" ></span> |
| 2182 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 2183 |
|
| 2184 |
|
| 2185 |
Some implementations might have been careful enough to design the |
| 2186 |
stack so as to avoid the incoming (or reply) packets going to IPv4 |
| 2187 |
packet processing through special addresses (e.g., IPv4-mapped |
| 2188 |
addresses), but who can say for all ... |
| 2189 |
|
| 2190 |
Authors' Addresses |
| 2191 |
|
| 2192 |
Pekka Savola |
| 2193 |
CSC/FUNET |
| 2194 |
Espoo |
| 2195 |
Finland |
| 2196 |
|
| 2197 |
EMail: [email protected] |
| 2198 |
|
| 2199 |
|
| 2200 |
Chirayu Patel |
| 2201 |
All Play, No Work |
| 2202 |
185, Defence Colony |
| 2203 |
Bangalore, Karnataka 560038 |
| 2204 |
India |
| 2205 |
|
| 2206 |
Phone: +91-98452-88078 |
| 2207 |
EMail: [email protected] |
| 2208 |
URI: <a href="http://www.chirayu.org">http://www.chirayu.org</a> |
| 2209 |
|
| 2210 |
|
| 2211 |
|
| 2212 |
|
| 2213 |
|
| 2214 |
|
| 2215 |
|
| 2216 |
|
| 2217 |
|
| 2218 |
|
| 2219 |
|
| 2220 |
|
| 2221 |
|
| 2222 |
|
| 2223 |
|
| 2224 |
|
| 2225 |
|
| 2226 |
|
| 2227 |
|
| 2228 |
|
| 2229 |
|
| 2230 |
|
| 2231 |
|
| 2232 |
|
| 2233 |
|
| 2234 |
|
| 2235 |
|
| 2236 |
<span class="grey">Savola & Patel Informational [Page 40]</span></pre> |
| 2237 |
<hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-41" ></span> |
| 2238 |
<span class="grey"><a href="./rfc3964">RFC 3964</a> Security Considerations for 6to4 December 2004</span> |
| 2239 |
|
| 2240 |
|
| 2241 |
Full Copyright Statement |
| 2242 |
|
| 2243 |
Copyright (C) The Internet Society (2004). |
| 2244 |
|
| 2245 |
This document is subject to the rights, licenses and restrictions |
| 2246 |
contained in <a href="https://www.rfc-editor.org/bcp/bcp78">BCP 78</a>, and except as set forth therein, the authors |
| 2247 |
retain all their rights. |
| 2248 |
|
| 2249 |
This document and the information contained herein are provided on an |
| 2250 |
"AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS |
| 2251 |
OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE INTERNET |
| 2252 |
ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, |
| 2253 |
INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE |
| 2254 |
INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED |
| 2255 |
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. |
| 2256 |
|
| 2257 |
Intellectual Property |
| 2258 |
|
| 2259 |
The IETF takes no position regarding the validity or scope of any |
| 2260 |
Intellectual Property Rights or other rights that might be claimed to |
| 2261 |
pertain to the implementation or use of the technology described in |
| 2262 |
this document or the extent to which any license under such rights |
| 2263 |
might or might not be available; nor does it represent that it has |
| 2264 |
made any independent effort to identify any such rights. Information |
| 2265 |
on the IETF's procedures with respect to rights in IETF Documents can |
| 2266 |
be found in <a href="https://www.rfc-editor.org/bcp/bcp78">BCP 78</a> and <a href="https://www.rfc-editor.org/bcp/bcp79">BCP 79</a>. |
| 2267 |
|
| 2268 |
Copies of IPR disclosures made to the IETF Secretariat and any |
| 2269 |
assurances of licenses to be made available, or the result of an |
| 2270 |
attempt made to obtain a general license or permission for the use of |
| 2271 |
such proprietary rights by implementers or users of this |
| 2272 |
specification can be obtained from the IETF on-line IPR repository at |
| 2273 |
<a href="http://www.ietf.org/ipr">http://www.ietf.org/ipr</a>. |
| 2274 |
|
| 2275 |
The IETF invites any interested party to bring to its attention any |
| 2276 |
copyrights, patents or patent applications, or other proprietary |
| 2277 |
rights that may cover technology that may be required to implement |
| 2278 |
this standard. Please address the information to the IETF at ietf- |
| 2279 |
[email protected]. |
| 2280 |
|
| 2281 |
Acknowledgement |
| 2282 |
|
| 2283 |
Funding for the RFC Editor function is currently provided by the |
| 2284 |
Internet Society. |
| 2285 |
|
| 2286 |
|
| 2287 |
|
| 2288 |
|
| 2289 |
|
| 2290 |
|
| 2291 |
|
| 2292 |
Savola & Patel Informational [Page 41] |
| 2293 |
</pre> |