Debian Codegraph @

rfc5389.html

raw
1 <pre>Network Working Group J. Rosenberg
2 Request for Comments: 5389 Cisco
3 Obsoletes: <a href="./rfc3489">3489</a> R. Mahy
4 Category: Standards Track P. Matthews
5 Unaffiliated
6 D. Wing
7 Cisco
8 October 2008
9
10
11 <span class="h1">Session Traversal Utilities for NAT (STUN)</span>
12
13 Status of This Memo
14
15 This document specifies an Internet standards track protocol for the
16 Internet community, and requests discussion and suggestions for
17 improvements. Please refer to the current edition of the "Internet
18 Official Protocol Standards" (STD 1) for the standardization state
19 and status of this protocol. Distribution of this memo is unlimited.
20
21 Abstract
22
23 Session Traversal Utilities for NAT (STUN) is a protocol that serves
24 as a tool for other protocols in dealing with Network Address
25 Translator (NAT) traversal. It can be used by an endpoint to
26 determine the IP address and port allocated to it by a NAT. It can
27 also be used to check connectivity between two endpoints, and as a
28 keep-alive protocol to maintain NAT bindings. STUN works with many
29 existing NATs, and does not require any special behavior from them.
30
31 STUN is not a NAT traversal solution by itself. Rather, it is a tool
32 to be used in the context of a NAT traversal solution. This is an
33 important change from the previous version of this specification (<a href="./rfc3489">RFC</a>
34 <a href="./rfc3489">3489</a>), which presented STUN as a complete solution.
35
36 This document obsoletes <a href="./rfc3489">RFC 3489</a>.
37
38 Table of Contents
39
40 <a href="#section-1">1</a>. Introduction ....................................................<a href="#page-4">4</a>
41 <a href="#section-2">2</a>. Evolution from <a href="./rfc3489">RFC 3489</a> .........................................<a href="#page-4">4</a>
42 <a href="#section-3">3</a>. Overview of Operation ...........................................<a href="#page-5">5</a>
43 <a href="#section-4">4</a>. Terminology .....................................................<a href="#page-8">8</a>
44 <a href="#section-5">5</a>. Definitions .....................................................<a href="#page-8">8</a>
45 <a href="#section-6">6</a>. STUN Message Structure .........................................<a href="#page-10">10</a>
46 <a href="#section-7">7</a>. Base Protocol Procedures .......................................<a href="#page-12">12</a>
47 <a href="#section-7.1">7.1</a>. Forming a Request or an Indication ........................<a href="#page-12">12</a>
48 <a href="#section-7.2">7.2</a>. Sending the Request or Indication .........................<a href="#page-13">13</a>
49
50
51
52 <span class="grey">Rosenberg, et al. Standards Track [Page 1]</span></pre>
53 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-2" ></span>
54 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
55
56
57 <a href="#section-7.2.1">7.2.1</a>. Sending over UDP ...................................<a href="#page-13">13</a>
58 <a href="#section-7.2.2">7.2.2</a>. Sending over TCP or TLS-over-TCP ...................<a href="#page-14">14</a>
59 <a href="#section-7.3">7.3</a>. Receiving a STUN Message ..................................<a href="#page-16">16</a>
60 <a href="#section-7.3.1">7.3.1</a>. Processing a Request ...............................<a href="#page-17">17</a>
61 <a href="#section-7.3.1.1">7.3.1.1</a>. Forming a Success or Error Response .......<a href="#page-18">18</a>
62 <a href="#section-7.3.1.2">7.3.1.2</a>. Sending the Success or Error Response .....<a href="#page-19">19</a>
63 <a href="#section-7.3.2">7.3.2</a>. Processing an Indication ...........................<a href="#page-19">19</a>
64 <a href="#section-7.3.3">7.3.3</a>. Processing a Success Response ......................<a href="#page-19">19</a>
65 <a href="#section-7.3.4">7.3.4</a>. Processing an Error Response .......................<a href="#page-20">20</a>
66 <a href="#section-8">8</a>. FINGERPRINT Mechanism ..........................................<a href="#page-20">20</a>
67 <a href="#section-9">9</a>. DNS Discovery of a Server ......................................<a href="#page-21">21</a>
68 <a href="#section-10">10</a>. Authentication and Message-Integrity Mechanisms ...............<a href="#page-22">22</a>
69 <a href="#section-10.1">10.1</a>. Short-Term Credential Mechanism ..........................<a href="#page-22">22</a>
70 <a href="#section-10.1.1">10.1.1</a>. Forming a Request or Indication ...................<a href="#page-23">23</a>
71 <a href="#section-10.1.2">10.1.2</a>. Receiving a Request or Indication .................<a href="#page-23">23</a>
72 <a href="#section-10.1.3">10.1.3</a>. Receiving a Response ..............................<a href="#page-24">24</a>
73 <a href="#section-10.2">10.2</a>. Long-Term Credential Mechanism ...........................<a href="#page-24">24</a>
74 <a href="#section-10.2.1">10.2.1</a>. Forming a Request .................................<a href="#page-25">25</a>
75 <a href="#section-10.2.1.1">10.2.1.1</a>. First Request ............................<a href="#page-25">25</a>
76 <a href="#section-10.2.1.2">10.2.1.2</a>. Subsequent Requests ......................<a href="#page-26">26</a>
77 <a href="#section-10.2.2">10.2.2</a>. Receiving a Request ...............................<a href="#page-26">26</a>
78 <a href="#section-10.2.3">10.2.3</a>. Receiving a Response ..............................<a href="#page-27">27</a>
79 <a href="#section-11">11</a>. ALTERNATE-SERVER Mechanism ....................................<a href="#page-28">28</a>
80 <a href="#section-12">12</a>. Backwards Compatibility with <a href="./rfc3489">RFC 3489</a> .........................<a href="#page-28">28</a>
81 <a href="#section-12.1">12.1</a>. Changes to Client Processing .............................<a href="#page-29">29</a>
82 <a href="#section-12.2">12.2</a>. Changes to Server Processing .............................<a href="#page-29">29</a>
83 <a href="#section-13">13</a>. Basic Server Behavior .........................................<a href="#page-30">30</a>
84 <a href="#section-14">14</a>. STUN Usages ...................................................<a href="#page-30">30</a>
85 <a href="#section-15">15</a>. STUN Attributes ...............................................<a href="#page-31">31</a>
86 <a href="#section-15.1">15.1</a>. MAPPED-ADDRESS ...........................................<a href="#page-32">32</a>
87 <a href="#section-15.2">15.2</a>. XOR-MAPPED-ADDRESS .......................................<a href="#page-33">33</a>
88 <a href="#section-15.3">15.3</a>. USERNAME .................................................<a href="#page-34">34</a>
89 <a href="#section-15.4">15.4</a>. MESSAGE-INTEGRITY ........................................<a href="#page-34">34</a>
90 <a href="#section-15.5">15.5</a>. FINGERPRINT ..............................................<a href="#page-36">36</a>
91 <a href="#section-15.6">15.6</a>. ERROR-CODE ...............................................<a href="#page-36">36</a>
92 <a href="#section-15.7">15.7</a>. REALM ....................................................<a href="#page-38">38</a>
93 <a href="#section-15.8">15.8</a>. NONCE ....................................................<a href="#page-38">38</a>
94 <a href="#section-15.9">15.9</a>. UNKNOWN-ATTRIBUTES .......................................<a href="#page-38">38</a>
95 <a href="#section-15.10">15.10</a>. SOFTWARE ................................................<a href="#page-39">39</a>
96 <a href="#section-15.11">15.11</a>. ALTERNATE-SERVER ........................................<a href="#page-39">39</a>
97 <a href="#section-16">16</a>. Security Considerations .......................................<a href="#page-39">39</a>
98 <a href="#section-16.1">16.1</a>. Attacks against the Protocol .............................<a href="#page-39">39</a>
99 <a href="#section-16.1.1">16.1.1</a>. Outside Attacks ...................................<a href="#page-39">39</a>
100 <a href="#section-16.1.2">16.1.2</a>. Inside Attacks ....................................<a href="#page-40">40</a>
101 <a href="#section-16.2">16.2</a>. Attacks Affecting the Usage ..............................<a href="#page-40">40</a>
102 16.2.1. Attack I: Distributed DoS (DDoS) against a
103 Target ............................................<a href="#page-41">41</a>
104 <a href="#section-16.2.2">16.2.2</a>. Attack II: Silencing a Client .....................<a href="#page-41">41</a>
105
106
107
108 <span class="grey">Rosenberg, et al. Standards Track [Page 2]</span></pre>
109 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-3" ></span>
110 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
111
112
113 <a href="#section-16.2.3">16.2.3</a>. Attack III: Assuming the Identity of a Client .....<a href="#page-42">42</a>
114 <a href="#section-16.2.4">16.2.4</a>. Attack IV: Eavesdropping ..........................<a href="#page-42">42</a>
115 <a href="#section-16.3">16.3</a>. Hash Agility Plan ........................................<a href="#page-42">42</a>
116 <a href="#section-17">17</a>. IAB Considerations ............................................<a href="#page-42">42</a>
117 <a href="#section-18">18</a>. IANA Considerations ...........................................<a href="#page-43">43</a>
118 <a href="#section-18.1">18.1</a>. STUN Methods Registry ....................................<a href="#page-43">43</a>
119 <a href="#section-18.2">18.2</a>. STUN Attribute Registry ..................................<a href="#page-43">43</a>
120 <a href="#section-18.3">18.3</a>. STUN Error Code Registry .................................<a href="#page-44">44</a>
121 <a href="#section-18.4">18.4</a>. STUN UDP and TCP Port Numbers ............................<a href="#page-45">45</a>
122 <a href="#section-19">19</a>. Changes since <a href="./rfc3489">RFC 3489</a> ........................................<a href="#page-45">45</a>
123 <a href="#section-20">20</a>. Contributors ..................................................<a href="#page-47">47</a>
124 <a href="#section-21">21</a>. Acknowledgements ..............................................<a href="#page-47">47</a>
125 <a href="#section-22">22</a>. References ....................................................<a href="#page-47">47</a>
126 <a href="#section-22.1">22.1</a>. Normative References .....................................<a href="#page-47">47</a>
127 <a href="#section-22.2">22.2</a>. Informative References ...................................<a href="#page-48">48</a>
128 <a href="#appendix-A">Appendix A</a>. C Snippet to Determine STUN Message Types .............<a href="#page-50">50</a>
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164 <span class="grey">Rosenberg, et al. Standards Track [Page 3]</span></pre>
165 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-4" ></span>
166 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
167
168
169 <span class="h2"><a class="selflink" id="section-1" href="#section-1">1</a>. Introduction</span>
170
171 The protocol defined in this specification, Session Traversal
172 Utilities for NAT, provides a tool for dealing with NATs. It
173 provides a means for an endpoint to determine the IP address and port
174 allocated by a NAT that corresponds to its private IP address and
175 port. It also provides a way for an endpoint to keep a NAT binding
176 alive. With some extensions, the protocol can be used to do
177 connectivity checks between two endpoints [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>], or to relay
178 packets between two endpoints [<a href="#ref-BEHAVE-TURN" title="&quot;Traversal Using Relays around NAT (TURN): Relay Extensions to Session Traversal Utilities for NAT (STUN)&quot;">BEHAVE-TURN</a>].
179
180 In keeping with its tool nature, this specification defines an
181 extensible packet format, defines operation over several transport
182 protocols, and provides for two forms of authentication.
183
184 STUN is intended to be used in context of one or more NAT traversal
185 solutions. These solutions are known as STUN usages. Each usage
186 describes how STUN is utilized to achieve the NAT traversal solution.
187 Typically, a usage indicates when STUN messages get sent, which
188 optional attributes to include, what server is used, and what
189 authentication mechanism is to be used. Interactive Connectivity
190 Establishment (ICE) [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>] is one usage of STUN. SIP Outbound
191 [<a href="#ref-SIP-OUTBOUND" title="&quot;Managing Client Initiated Connections in the Session Initiation Protocol (SIP)&quot;">SIP-OUTBOUND</a>] is another usage of STUN. In some cases, a usage will
192 require extensions to STUN. A STUN extension can be in the form of
193 new methods, attributes, or error response codes. More information
194 on STUN usages can be found in <a href="#section-14">Section 14</a>.
195
196 <span class="h2"><a class="selflink" id="section-2" href="#section-2">2</a>. Evolution from <a href="./rfc3489">RFC 3489</a></span>
197
198 STUN was originally defined in <a href="./rfc3489">RFC 3489</a> [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>]. That
199 specification, sometimes referred to as "classic STUN", represented
200 itself as a complete solution to the NAT traversal problem. In that
201 solution, a client would discover whether it was behind a NAT,
202 determine its NAT type, discover its IP address and port on the
203 public side of the outermost NAT, and then utilize that IP address
204 and port within the body of protocols, such as the Session Initiation
205 Protocol (SIP) [<a href="./rfc3261" title="&quot;SIP: Session Initiation Protocol&quot;">RFC3261</a>]. However, experience since the publication
206 of <a href="./rfc3489">RFC 3489</a> has found that classic STUN simply does not work
207 sufficiently well to be a deployable solution. The address and port
208 learned through classic STUN are sometimes usable for communications
209 with a peer, and sometimes not. Classic STUN provided no way to
210 discover whether it would, in fact, work or not, and it provided no
211 remedy in cases where it did not. Furthermore, classic STUN's
212 algorithm for classification of NAT types was found to be faulty, as
213 many NATs did not fit cleanly into the types defined there.
214
215
216
217
218
219
220 <span class="grey">Rosenberg, et al. Standards Track [Page 4]</span></pre>
221 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-5" ></span>
222 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
223
224
225 Classic STUN also had a security vulnerability -- attackers could
226 provide the client with incorrect mapped addresses under certain
227 topologies and constraints, and this was fundamentally not solvable
228 through any cryptographic means. Though this problem remains with
229 this specification, those attacks are now mitigated through the use
230 of more complete solutions that make use of STUN.
231
232 For these reasons, this specification obsoletes <a href="./rfc3489">RFC 3489</a>, and instead
233 describes STUN as a tool that is utilized as part of a complete NAT
234 traversal solution. ICE [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>] is a complete NAT traversal
235 solution for protocols based on the offer/answer [<a href="./rfc3264" title="&quot;An Offer/Answer Model with Session Description Protocol (SDP)&quot;">RFC3264</a>]
236 methodology, such as SIP. SIP Outbound [<a href="#ref-SIP-OUTBOUND" title="&quot;Managing Client Initiated Connections in the Session Initiation Protocol (SIP)&quot;">SIP-OUTBOUND</a>] is a complete
237 solution for traversal of SIP signaling, and it uses STUN in a very
238 different way. Though it is possible that a protocol may be able to
239 use STUN by itself (classic STUN) as a traversal solution, such usage
240 is not described here and is strongly discouraged for the reasons
241 described above.
242
243 The on-the-wire protocol described here is changed only slightly from
244 classic STUN. The protocol now runs over TCP in addition to UDP.
245 Extensibility was added to the protocol in a more structured way. A
246 magic cookie mechanism for demultiplexing STUN with application
247 protocols was added by stealing 32 bits from the 128-bit transaction
248 ID defined in <a href="./rfc3489">RFC 3489</a>, allowing the change to be backwards
249 compatible. Mapped addresses are encoded using a new exclusive-or
250 format. There are other, more minor changes. See <a href="#section-19">Section 19</a> for a
251 more complete listing.
252
253 Due to the change in scope, STUN has also been renamed from "Simple
254 Traversal of UDP through NAT" to "Session Traversal Utilities for
255 NAT". The acronym remains STUN, which is all anyone ever remembers
256 anyway.
257
258 <span class="h2"><a class="selflink" id="section-3" href="#section-3">3</a>. Overview of Operation</span>
259
260 This section is descriptive only.
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276 <span class="grey">Rosenberg, et al. Standards Track [Page 5]</span></pre>
277 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-6" ></span>
278 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
279
280
281 /-----\
282 // STUN \\
283 | Server |
284 \\ //
285 \-----/
286
287
288
289
290 +--------------+ Public Internet
291 ................| NAT 2 |.......................
292 +--------------+
293
294
295
296 +--------------+ Private NET 2
297 ................| NAT 1 |.......................
298 +--------------+
299
300
301
302
303 /-----\
304 // STUN \\
305 | Client |
306 \\ // Private NET 1
307 \-----/
308
309
310 Figure 1: One Possible STUN Configuration
311
312 One possible STUN configuration is shown in Figure 1. In this
313 configuration, there are two entities (called STUN agents) that
314 implement the STUN protocol. The lower agent in the figure is the
315 client, and is connected to private network 1. This network connects
316 to private network 2 through NAT 1. Private network 2 connects to
317 the public Internet through NAT 2. The upper agent in the figure is
318 the server, and resides on the public Internet.
319
320 STUN is a client-server protocol. It supports two types of
321 transactions. One is a request/response transaction in which a
322 client sends a request to a server, and the server returns a
323 response. The second is an indication transaction in which either
324 agent -- client or server -- sends an indication that generates no
325 response. Both types of transactions include a transaction ID, which
326 is a randomly selected 96-bit number. For request/response
327
328
329
330
331
332 <span class="grey">Rosenberg, et al. Standards Track [Page 6]</span></pre>
333 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-7" ></span>
334 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
335
336
337 transactions, this transaction ID allows the client to associate the
338 response with the request that generated it; for indications, the
339 transaction ID serves as a debugging aid.
340
341 All STUN messages start with a fixed header that includes a method, a
342 class, and the transaction ID. The method indicates which of the
343 various requests or indications this is; this specification defines
344 just one method, Binding, but other methods are expected to be
345 defined in other documents. The class indicates whether this is a
346 request, a success response, an error response, or an indication.
347 Following the fixed header comes zero or more attributes, which are
348 Type-Length-Value extensions that convey additional information for
349 the specific message.
350
351 This document defines a single method called Binding. The Binding
352 method can be used either in request/response transactions or in
353 indication transactions. When used in request/response transactions,
354 the Binding method can be used to determine the particular "binding"
355 a NAT has allocated to a STUN client. When used in either request/
356 response or in indication transactions, the Binding method can also
357 be used to keep these "bindings" alive.
358
359 In the Binding request/response transaction, a Binding request is
360 sent from a STUN client to a STUN server. When the Binding request
361 arrives at the STUN server, it may have passed through one or more
362 NATs between the STUN client and the STUN server (in Figure 1, there
363 were two such NATs). As the Binding request message passes through a
364 NAT, the NAT will modify the source transport address (that is, the
365 source IP address and the source port) of the packet. As a result,
366 the source transport address of the request received by the server
367 will be the public IP address and port created by the NAT closest to
368 the server. This is called a reflexive transport address. The STUN
369 server copies that source transport address into an XOR-MAPPED-
370 ADDRESS attribute in the STUN Binding response and sends the Binding
371 response back to the STUN client. As this packet passes back through
372 a NAT, the NAT will modify the destination transport address in the
373 IP header, but the transport address in the XOR-MAPPED-ADDRESS
374 attribute within the body of the STUN response will remain untouched.
375 In this way, the client can learn its reflexive transport address
376 allocated by the outermost NAT with respect to the STUN server.
377
378 In some usages, STUN must be multiplexed with other protocols (e.g.,
379 [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>], [<a href="#ref-SIP-OUTBOUND" title="&quot;Managing Client Initiated Connections in the Session Initiation Protocol (SIP)&quot;">SIP-OUTBOUND</a>]). In these usages, there must be a way
380 to inspect a packet and determine if it is a STUN packet or not.
381 STUN provides three fields in the STUN header with fixed values that
382 can be used for this purpose. If this is not sufficient, then STUN
383 packets can also contain a FINGERPRINT value, which can further be
384 used to distinguish the packets.
385
386
387
388 <span class="grey">Rosenberg, et al. Standards Track [Page 7]</span></pre>
389 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-8" ></span>
390 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
391
392
393 STUN defines a set of optional procedures that a usage can decide to
394 use, called mechanisms. These mechanisms include DNS discovery, a
395 redirection technique to an alternate server, a fingerprint attribute
396 for demultiplexing, and two authentication and message-integrity
397 exchanges. The authentication mechanisms revolve around the use of a
398 username, password, and message-integrity value. Two authentication
399 mechanisms, the long-term credential mechanism and the short-term
400 credential mechanism, are defined in this specification. Each usage
401 specifies the mechanisms allowed with that usage.
402
403 In the long-term credential mechanism, the client and server share a
404 pre-provisioned username and password and perform a digest challenge/
405 response exchange inspired by (but differing in details) to the one
406 defined for HTTP [<a href="./rfc2617" title="&quot;HTTP Authentication: Basic and Digest Access Authentication&quot;">RFC2617</a>]. In the short-term credential mechanism,
407 the client and the server exchange a username and password through
408 some out-of-band method prior to the STUN exchange. For example, in
409 the ICE usage [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>] the two endpoints use out-of-band
410 signaling to exchange a username and password. These are used to
411 integrity protect and authenticate the request and response. There
412 is no challenge or nonce used.
413
414 <span class="h2"><a class="selflink" id="section-4" href="#section-4">4</a>. Terminology</span>
415
416 In this document, the key words "MUST", "MUST NOT", "REQUIRED",
417 "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY",
418 and "OPTIONAL" 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>
419 [<a href="./rfc2119" title="&quot;Key words for use in RFCs to Indicate Requirement Levels&quot;">RFC2119</a>] and indicate requirement levels for compliant STUN
420 implementations.
421
422 <span class="h2"><a class="selflink" id="section-5" href="#section-5">5</a>. Definitions</span>
423
424 STUN Agent: A STUN agent is an entity that implements the STUN
425 protocol. The entity can be either a STUN client or a STUN
426 server.
427
428 STUN Client: A STUN client is an entity that sends STUN requests and
429 receives STUN responses. A STUN client can also send indications.
430 In this specification, the terms STUN client and client are
431 synonymous.
432
433 STUN Server: A STUN server is an entity that receives STUN requests
434 and sends STUN responses. A STUN server can also send
435 indications. In this specification, the terms STUN server and
436 server are synonymous.
437
438 Transport Address: The combination of an IP address and port number
439 (such as a UDP or TCP port number).
440
441
442
443
444 <span class="grey">Rosenberg, et al. Standards Track [Page 8]</span></pre>
445 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-9" ></span>
446 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
447
448
449 Reflexive Transport Address: A transport address learned by a client
450 that identifies that client as seen by another host on an IP
451 network, typically a STUN server. When there is an intervening
452 NAT between the client and the other host, the reflexive transport
453 address represents the mapped address allocated to the client on
454 the public side of the NAT. Reflexive transport addresses are
455 learned from the mapped address attribute (MAPPED-ADDRESS or XOR-
456 MAPPED-ADDRESS) in STUN responses.
457
458 Mapped Address: Same meaning as reflexive address. This term is
459 retained only for historic reasons and due to the naming of the
460 MAPPED-ADDRESS and XOR-MAPPED-ADDRESS attributes.
461
462 Long-Term Credential: A username and associated password that
463 represent a shared secret between client and server. Long-term
464 credentials are generally granted to the client when a subscriber
465 enrolls in a service and persist until the subscriber leaves the
466 service or explicitly changes the credential.
467
468 Long-Term Password: The password from a long-term credential.
469
470 Short-Term Credential: A temporary username and associated password
471 that represent a shared secret between client and server. Short-
472 term credentials are obtained through some kind of protocol
473 mechanism between the client and server, preceding the STUN
474 exchange. A short-term credential has an explicit temporal scope,
475 which may be based on a specific amount of time (such as 5
476 minutes) or on an event (such as termination of a SIP dialog).
477 The specific scope of a short-term credential is defined by the
478 application usage.
479
480 Short-Term Password: The password component of a short-term
481 credential.
482
483 STUN Indication: A STUN message that does not receive a response.
484
485 Attribute: The STUN term for a Type-Length-Value (TLV) object that
486 can be added to a STUN message. Attributes are divided into two
487 types: comprehension-required and comprehension-optional. STUN
488 agents can safely ignore comprehension-optional attributes they
489 don't understand, but cannot successfully process a message if it
490 contains comprehension-required attributes that are not
491 understood.
492
493 RTO: Retransmission TimeOut, which defines the initial period of
494 time between transmission of a request and the first retransmit of
495 that request.
496
497
498
499
500 <span class="grey">Rosenberg, et al. Standards Track [Page 9]</span></pre>
501 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-10" ></span>
502 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
503
504
505 <span class="h2"><a class="selflink" id="section-6" href="#section-6">6</a>. STUN Message Structure</span>
506
507 STUN messages are encoded in binary using network-oriented format
508 (most significant byte or octet first, also commonly known as big-
509 endian). The transmission order is described in detail in <a href="./rfc791#appendix-B">Appendix&nbsp;B
510 of RFC 791</a> [<a href="./rfc0791" title="&quot;Internet Protocol&quot;">RFC0791</a>]. Unless otherwise noted, numeric constants are
511 in decimal (base 10).
512
513 All STUN messages MUST start with a 20-byte header followed by zero
514 or more Attributes. The STUN header contains a STUN message type,
515 magic cookie, transaction ID, and message length.
516
517 0 1 2 3
518 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
519 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
520 |0 0| STUN Message Type | Message Length |
521 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
522 | Magic Cookie |
523 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
524 | |
525 | Transaction ID (96 bits) |
526 | |
527 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
528
529 Figure 2: Format of STUN Message Header
530
531 The most significant 2 bits of every STUN message MUST be zeroes.
532 This can be used to differentiate STUN packets from other protocols
533 when STUN is multiplexed with other protocols on the same port.
534
535 The message type defines the message class (request, success
536 response, failure response, or indication) and the message method
537 (the primary function) of the STUN message. Although there are four
538 message classes, there are only two types of transactions in STUN:
539 request/response transactions (which consist of a request message and
540 a response message) and indication transactions (which consist of a
541 single indication message). Response classes are split into error
542 and success responses to aid in quickly processing the STUN message.
543
544
545
546
547
548
549
550
551
552
553
554
555
556 <span class="grey">Rosenberg, et al. Standards Track [Page 10]</span></pre>
557 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-11" ></span>
558 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
559
560
561 The message type field is decomposed further into the following
562 structure:
563
564 0 1
565 2 3 4 5 6 7 8 9 0 1 2 3 4 5
566
567 +--+--+-+-+-+-+-+-+-+-+-+-+-+-+
568 |M |M |M|M|M|C|M|M|M|C|M|M|M|M|
569 |11|10|9|8|7|1|6|5|4|0|3|2|1|0|
570 +--+--+-+-+-+-+-+-+-+-+-+-+-+-+
571
572 Figure 3: Format of STUN Message Type Field
573
574 Here the bits in the message type field are shown as most significant
575 (M11) through least significant (M0). M11 through M0 represent a 12-
576 bit encoding of the method. C1 and C0 represent a 2-bit encoding of
577 the class. A class of 0b00 is a request, a class of 0b01 is an
578 indication, a class of 0b10 is a success response, and a class of
579 0b11 is an error response. This specification defines a single
580 method, Binding. The method and class are orthogonal, so that for
581 each method, a request, success response, error response, and
582 indication are possible for that method. Extensions defining new
583 methods MUST indicate which classes are permitted for that method.
584
585 For example, a Binding request has class=0b00 (request) and
586 method=0b000000000001 (Binding) and is encoded into the first 16 bits
587 as 0x0001. A Binding response has class=0b10 (success response) and
588 method=0b000000000001, and is encoded into the first 16 bits as
589 0x0101.
590
591 Note: This unfortunate encoding is due to assignment of values in
592 [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>] that did not consider encoding Indications, Success, and
593 Errors using bit fields.
594
595 The magic cookie field MUST contain the fixed value 0x2112A442 in
596 network byte order. In <a href="./rfc3489">RFC 3489</a> [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>], this field was part of
597 the transaction ID; placing the magic cookie in this location allows
598 a server to detect if the client will understand certain attributes
599 that were added in this revised specification. In addition, it aids
600 in distinguishing STUN packets from packets of other protocols when
601 STUN is multiplexed with those other protocols on the same port.
602
603 The transaction ID is a 96-bit identifier, used to uniquely identify
604 STUN transactions. For request/response transactions, the
605 transaction ID is chosen by the STUN client for the request and
606 echoed by the server in the response. For indications, it is chosen
607 by the agent sending the indication. It primarily serves to
608 correlate requests with responses, though it also plays a small role
609
610
611
612 <span class="grey">Rosenberg, et al. Standards Track [Page 11]</span></pre>
613 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-12" ></span>
614 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
615
616
617 in helping to prevent certain types of attacks. The server also uses
618 the transaction ID as a key to identify each transaction uniquely
619 across all clients. As such, the transaction ID MUST be uniformly
620 and randomly chosen from the interval 0 .. 2**96-1, and SHOULD be
621 cryptographically random. Resends of the same request reuse the same
622 transaction ID, but the client MUST choose a new transaction ID for
623 new transactions unless the new request is bit-wise identical to the
624 previous request and sent from the same transport address to the same
625 IP address. Success and error responses MUST carry the same
626 transaction ID as their corresponding request. When an agent is
627 acting as a STUN server and STUN client on the same port, the
628 transaction IDs in requests sent by the agent have no relationship to
629 the transaction IDs in requests received by the agent.
630
631 The message length MUST contain the size, in bytes, of the message
632 not including the 20-byte STUN header. Since all STUN attributes are
633 padded to a multiple of 4 bytes, the last 2 bits of this field are
634 always zero. This provides another way to distinguish STUN packets
635 from packets of other protocols.
636
637 Following the STUN fixed portion of the header are zero or more
638 attributes. Each attribute is TLV (Type-Length-Value) encoded. The
639 details of the encoding, and of the attributes themselves are given
640 in <a href="#section-15">Section 15</a>.
641
642 <span class="h2"><a class="selflink" id="section-7" href="#section-7">7</a>. Base Protocol Procedures</span>
643
644 This section defines the base procedures of the STUN protocol. It
645 describes how messages are formed, how they are sent, and how they
646 are processed when they are received. It also defines the detailed
647 processing of the Binding method. Other sections in this document
648 describe optional procedures that a usage may elect to use in certain
649 situations. Other documents may define other extensions to STUN, by
650 adding new methods, new attributes, or new error response codes.
651
652 <span class="h3"><a class="selflink" id="section-7.1" href="#section-7.1">7.1</a>. Forming a Request or an Indication</span>
653
654 When formulating a request or indication message, the agent MUST
655 follow the rules in <a href="#section-6">Section 6</a> when creating the header. In addition,
656 the message class MUST be either "Request" or "Indication" (as
657 appropriate), and the method must be either Binding or some method
658 defined in another document.
659
660 The agent then adds any attributes specified by the method or the
661 usage. For example, some usages may specify that the agent use an
662 authentication method (<a href="#section-10">Section 10</a>) or the FINGERPRINT attribute
663 (<a href="#section-8">Section 8</a>).
664
665
666
667
668 <span class="grey">Rosenberg, et al. Standards Track [Page 12]</span></pre>
669 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-13" ></span>
670 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
671
672
673 If the agent is sending a request, it SHOULD add a SOFTWARE attribute
674 to the request. Agents MAY include a SOFTWARE attribute in
675 indications, depending on the method. Extensions to STUN should
676 discuss whether SOFTWARE is useful in new indications.
677
678 For the Binding method with no authentication, no attributes are
679 required unless the usage specifies otherwise.
680
681 All STUN messages sent over UDP SHOULD be less than the path MTU, if
682 known. If the path MTU is unknown, messages SHOULD be the smaller of
683 576 bytes and the first-hop MTU for IPv4 [<a href="./rfc1122" title="&quot;Requirements for Internet Hosts - Communication Layers&quot;">RFC1122</a>] and 1280 bytes for
684 IPv6 [<a href="./rfc2460" title="&quot;Internet Protocol, Version 6 (IPv6) Specification&quot;">RFC2460</a>]. This value corresponds to the overall size of the IP
685 packet. Consequently, for IPv4, the actual STUN message would need
686 to be less than 548 bytes (576 minus 20-byte IP header, minus 8-byte
687 UDP header, assuming no IP options are used). STUN provides no
688 ability to handle the case where the request is under the MTU but the
689 response would be larger than the MTU. It is not envisioned that
690 this limitation will be an issue for STUN. The MTU limitation is a
691 SHOULD, and not a MUST, to account for cases where STUN itself is
692 being used to probe for MTU characteristics [<a href="#ref-BEHAVE-NAT" title="&quot;NAT Behavior Discovery Using STUN&quot;">BEHAVE-NAT</a>]. Outside of
693 this or similar applications, the MTU constraint MUST be followed.
694
695 <span class="h3"><a class="selflink" id="section-7.2" href="#section-7.2">7.2</a>. Sending the Request or Indication</span>
696
697 The agent then sends the request or indication. This document
698 specifies how to send STUN messages over UDP, TCP, or TLS-over-TCP;
699 other transport protocols may be added in the future. The STUN usage
700 must specify which transport protocol is used, and how the agent
701 determines the IP address and port of the recipient. <a href="#section-9">Section 9</a>
702 describes a DNS-based method of determining the IP address and port
703 of a server that a usage may elect to use. STUN may be used with
704 anycast addresses, but only with UDP and in usages where
705 authentication is not used.
706
707 At any time, a client MAY have multiple outstanding STUN requests
708 with the same STUN server (that is, multiple transactions in
709 progress, with different transaction IDs). Absent other limits to
710 the rate of new transactions (such as those specified by ICE for
711 connectivity checks or when STUN is run over TCP), a client SHOULD
712 space new transactions to a server by RTO and SHOULD limit itself to
713 ten outstanding transactions to the same server.
714
715 <span class="h4"><a class="selflink" id="section-7.2.1" href="#section-7.2.1">7.2.1</a>. Sending over UDP</span>
716
717 When running STUN over UDP, it is possible that the STUN message
718 might be dropped by the network. Reliability of STUN request/
719 response transactions is accomplished through retransmissions of the
720
721
722
723
724 <span class="grey">Rosenberg, et al. Standards Track [Page 13]</span></pre>
725 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-14" ></span>
726 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
727
728
729 request message by the client application itself. STUN indications
730 are not retransmitted; thus, indication transactions over UDP are not
731 reliable.
732
733 A client SHOULD retransmit a STUN request message starting with an
734 interval of RTO ("Retransmission TimeOut"), doubling after each
735 retransmission. The RTO is an estimate of the round-trip time (RTT),
736 and is computed as described in <a href="./rfc2988">RFC 2988</a> [<a href="./rfc2988" title="&quot;Computing TCP's Retransmission Timer&quot;">RFC2988</a>], with two
737 exceptions. First, the initial value for RTO SHOULD be configurable
738 (rather than the 3 s recommended in <a href="./rfc2988">RFC 2988</a>) and SHOULD be greater
739 than 500 ms. The exception cases for this "SHOULD" are when other
740 mechanisms are used to derive congestion thresholds (such as the ones
741 defined in ICE for fixed rate streams), or when STUN is used in non-
742 Internet environments with known network capacities. In fixed-line
743 access links, a value of 500 ms is RECOMMENDED. Second, the value of
744 RTO SHOULD NOT be rounded up to the nearest second. Rather, a 1 ms
745 accuracy SHOULD be maintained. As with TCP, the usage of Karn's
746 algorithm is RECOMMENDED [<a href="#ref-KARN87" title="&quot;Improving Round-Trip Time Estimates in Reliable Transport Protocols&quot;">KARN87</a>]. When applied to STUN, it means
747 that RTT estimates SHOULD NOT be computed from STUN transactions that
748 result in the retransmission of a request.
749
750 The value for RTO SHOULD be cached by a client after the completion
751 of the transaction, and used as the starting value for RTO for the
752 next transaction to the same server (based on equality of IP
753 address). The value SHOULD be considered stale and discarded after
754 10 minutes.
755
756 Retransmissions continue until a response is received, or until a
757 total of Rc requests have been sent. Rc SHOULD be configurable and
758 SHOULD have a default of 7. If, after the last request, a duration
759 equal to Rm times the RTO has passed without a response (providing
760 ample time to get a response if only this final request actually
761 succeeds), the client SHOULD consider the transaction to have failed.
762 Rm SHOULD be configurable and SHOULD have a default of 16. A STUN
763 transaction over UDP is also considered failed if there has been a
764 hard ICMP error [<a href="./rfc1122" title="&quot;Requirements for Internet Hosts - Communication Layers&quot;">RFC1122</a>]. For example, assuming an RTO of 500 ms,
765 requests would be sent at times 0 ms, 500 ms, 1500 ms, 3500 ms, 7500
766 ms, 15500 ms, and 31500 ms. If the client has not received a
767 response after 39500 ms, the client will consider the transaction to
768 have timed out.
769
770 <span class="h4"><a class="selflink" id="section-7.2.2" href="#section-7.2.2">7.2.2</a>. Sending over TCP or TLS-over-TCP</span>
771
772 For TCP and TLS-over-TCP, the client opens a TCP connection to the
773 server.
774
775
776
777
778
779
780 <span class="grey">Rosenberg, et al. Standards Track [Page 14]</span></pre>
781 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-15" ></span>
782 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
783
784
785 In some usages of STUN, STUN is sent as the only protocol over the
786 TCP connection. In this case, it can be sent without the aid of any
787 additional framing or demultiplexing. In other usages, or with other
788 extensions, it may be multiplexed with other data over a TCP
789 connection. In that case, STUN MUST be run on top of some kind of
790 framing protocol, specified by the usage or extension, which allows
791 for the agent to extract complete STUN messages and complete
792 application layer messages. The STUN service running on the well-
793 known port or ports discovered through the DNS procedures in
794 <a href="#section-9">Section 9</a> is for STUN alone, and not for STUN multiplexed with other
795 data. Consequently, no framing protocols are used in connections to
796 those servers. When additional framing is utilized, the usage will
797 specify how the client knows to apply it and what port to connect to.
798 For example, in the case of ICE connectivity checks, this information
799 is learned through out-of-band negotiation between client and server.
800
801 When STUN is run by itself over TLS-over-TCP, the
802 TLS_RSA_WITH_AES_128_CBC_SHA ciphersuite MUST be implemented at a
803 minimum. Implementations MAY also support any other ciphersuite.
804 When it receives the TLS Certificate message, the client SHOULD
805 verify the certificate and inspect the site identified by the
806 certificate. If the certificate is invalid or revoked, or if it does
807 not identify the appropriate party, the client MUST NOT send the STUN
808 message or otherwise proceed with the STUN transaction. The client
809 MUST verify the identity of the server. To do that, it follows the
810 identification procedures defined in <a href="./rfc2818#section-3.1">Section&nbsp;3.1 of RFC 2818</a>
811 [<a href="./rfc2818" title="&quot;HTTP Over TLS&quot;">RFC2818</a>]. Those procedures assume the client is dereferencing a
812 URI. For purposes of usage with this specification, the client
813 treats the domain name or IP address used in <a href="#section-8.1">Section 8.1</a> as the host
814 portion of the URI that has been dereferenced. Alternatively, a
815 client MAY be configured with a set of domains or IP addresses that
816 are trusted; if a certificate is received that identifies one of
817 those domains or IP addresses, the client considers the identity of
818 the server to be verified.
819
820 When STUN is run multiplexed with other protocols over a TLS-over-TCP
821 connection, the mandatory ciphersuites and TLS handling procedures
822 operate as defined by those protocols.
823
824 Reliability of STUN over TCP and TLS-over-TCP is handled by TCP
825 itself, and there are no retransmissions at the STUN protocol level.
826 However, for a request/response transaction, if the client has not
827 received a response by Ti seconds after it sent the SYN to establish
828 the connection, it considers the transaction to have timed out. Ti
829 SHOULD be configurable and SHOULD have a default of 39.5s. This
830 value has been chosen to equalize the TCP and UDP timeouts for the
831 default initial RTO.
832
833
834
835
836 <span class="grey">Rosenberg, et al. Standards Track [Page 15]</span></pre>
837 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-16" ></span>
838 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
839
840
841 In addition, if the client is unable to establish the TCP connection,
842 or the TCP connection is reset or fails before a response is
843 received, any request/response transaction in progress is considered
844 to have failed.
845
846 The client MAY send multiple transactions over a single TCP (or TLS-
847 over-TCP) connection, and it MAY send another request before
848 receiving a response to the previous. The client SHOULD keep the
849 connection open until it:
850
851 o has no further STUN requests or indications to send over that
852 connection, and
853
854 o has no plans to use any resources (such as a mapped address
855 (MAPPED-ADDRESS or XOR-MAPPED-ADDRESS) or relayed address
856 [<a href="#ref-BEHAVE-TURN" title="&quot;Traversal Using Relays around NAT (TURN): Relay Extensions to Session Traversal Utilities for NAT (STUN)&quot;">BEHAVE-TURN</a>]) that were learned though STUN requests sent over
857 that connection, and
858
859 o if multiplexing other application protocols over that port, has
860 finished using that other application, and
861
862 o if using that learned port with a remote peer, has established
863 communications with that remote peer, as is required by some TCP
864 NAT traversal techniques (e.g., [<a href="#ref-MMUSIC-ICE-TCP" title="&quot;TCP Candidates with Interactive Connectivity Establishment (ICE)&quot;">MMUSIC-ICE-TCP</a>]).
865
866 At the server end, the server SHOULD keep the connection open, and
867 let the client close it, unless the server has determined that the
868 connection has timed out (for example, due to the client
869 disconnecting from the network). Bindings learned by the client will
870 remain valid in intervening NATs only while the connection remains
871 open. Only the client knows how long it needs the binding. The
872 server SHOULD NOT close a connection if a request was received over
873 that connection for which a response was not sent. A server MUST NOT
874 ever open a connection back towards the client in order to send a
875 response. Servers SHOULD follow best practices regarding connection
876 management in cases of overload.
877
878 <span class="h3"><a class="selflink" id="section-7.3" href="#section-7.3">7.3</a>. Receiving a STUN Message</span>
879
880 This section specifies the processing of a STUN message. The
881 processing specified here is for STUN messages as defined in this
882 specification; additional rules for backwards compatibility are
883 defined in <a href="#section-12">Section 12</a>. Those additional procedures are optional, and
884 usages can elect to utilize them. First, a set of processing
885 operations is applied that is independent of the class. This is
886 followed by class-specific processing, described in the subsections
887 that follow.
888
889
890
891
892 <span class="grey">Rosenberg, et al. Standards Track [Page 16]</span></pre>
893 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-17" ></span>
894 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
895
896
897 When a STUN agent receives a STUN message, it first checks that the
898 message obeys the rules of <a href="#section-6">Section 6</a>. It checks that the first two
899 bits are 0, that the magic cookie field has the correct value, that
900 the message length is sensible, and that the method value is a
901 supported method. It checks that the message class is allowed for
902 the particular method. If the message class is "Success Response" or
903 "Error Response", the agent checks that the transaction ID matches a
904 transaction that is still in progress. If the FINGERPRINT extension
905 is being used, the agent checks that the FINGERPRINT attribute is
906 present and contains the correct value. If any errors are detected,
907 the message is silently discarded. In the case when STUN is being
908 multiplexed with another protocol, an error may indicate that this is
909 not really a STUN message; in this case, the agent should try to
910 parse the message as a different protocol.
911
912 The STUN agent then does any checks that are required by a
913 authentication mechanism that the usage has specified (see
914 <a href="#section-10">Section 10</a>).
915
916 Once the authentication checks are done, the STUN agent checks for
917 unknown attributes and known-but-unexpected attributes in the
918 message. Unknown comprehension-optional attributes MUST be ignored
919 by the agent. Known-but-unexpected attributes SHOULD be ignored by
920 the agent. Unknown comprehension-required attributes cause
921 processing that depends on the message class and is described below.
922
923 At this point, further processing depends on the message class of the
924 request.
925
926 <span class="h4"><a class="selflink" id="section-7.3.1" href="#section-7.3.1">7.3.1</a>. Processing a Request</span>
927
928 If the request contains one or more unknown comprehension-required
929 attributes, the server replies with an error response with an error
930 code of 420 (Unknown Attribute), and includes an UNKNOWN-ATTRIBUTES
931 attribute in the response that lists the unknown comprehension-
932 required attributes.
933
934 The server then does any additional checking that the method or the
935 specific usage requires. If all the checks succeed, the server
936 formulates a success response as described below.
937
938 When run over UDP, a request received by the server could be the
939 first request of a transaction, or a retransmission. The server MUST
940 respond to retransmissions such that the following property is
941 preserved: if the client receives the response to the retransmission
942 and not the response that was sent to the original request, the
943 overall state on the client and server is identical to the case where
944 only the response to the original retransmission is received, or
945
946
947
948 <span class="grey">Rosenberg, et al. Standards Track [Page 17]</span></pre>
949 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-18" ></span>
950 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
951
952
953 where both responses are received (in which case the client will use
954 the first). The easiest way to meet this requirement is for the
955 server to remember all transaction IDs received over UDP and their
956 corresponding responses in the last 40 seconds. However, this
957 requires the server to hold state, and will be inappropriate for any
958 requests which are not authenticated. Another way is to reprocess
959 the request and recompute the response. The latter technique MUST
960 only be applied to requests that are idempotent (a request is
961 considered idempotent when the same request can be safely repeated
962 without impacting the overall state of the system) and result in the
963 same success response for the same request. The Binding method is
964 considered to be idempotent. Note that there are certain rare
965 network events that could cause the reflexive transport address value
966 to change, resulting in a different mapped address in different
967 success responses. Extensions to STUN MUST discuss the implications
968 of request retransmissions on servers that do not store transaction
969 state.
970
971 <span class="h5"><a class="selflink" id="section-7.3.1.1" href="#section-7.3.1.1">7.3.1.1</a>. Forming a Success or Error Response</span>
972
973 When forming the response (success or error), the server follows the
974 rules of <a href="#section-6">Section 6</a>. The method of the response is the same as that
975 of the request, and the message class is either "Success Response" or
976 "Error Response".
977
978 For an error response, the server MUST add an ERROR-CODE attribute
979 containing the error code specified in the processing above. The
980 reason phrase is not fixed, but SHOULD be something suitable for the
981 error code. For certain errors, additional attributes are added to
982 the message. These attributes are spelled out in the description
983 where the error code is specified. For example, for an error code of
984 420 (Unknown Attribute), the server MUST include an UNKNOWN-
985 ATTRIBUTES attribute. Certain authentication errors also cause
986 attributes to be added (see <a href="#section-10">Section 10</a>). Extensions may define other
987 errors and/or additional attributes to add in error cases.
988
989 If the server authenticated the request using an authentication
990 mechanism, then the server SHOULD add the appropriate authentication
991 attributes to the response (see <a href="#section-10">Section 10</a>).
992
993 The server also adds any attributes required by the specific method
994 or usage. In addition, the server SHOULD add a SOFTWARE attribute to
995 the message.
996
997 For the Binding method, no additional checking is required unless the
998 usage specifies otherwise. When forming the success response, the
999 server adds a XOR-MAPPED-ADDRESS attribute to the response, where the
1000 contents of the attribute are the source transport address of the
1001
1002
1003
1004 <span class="grey">Rosenberg, et al. Standards Track [Page 18]</span></pre>
1005 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-19" ></span>
1006 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1007
1008
1009 request message. For UDP, this is the source IP address and source
1010 UDP port of the request message. For TCP and TLS-over-TCP, this is
1011 the source IP address and source TCP port of the TCP connection as
1012 seen by the server.
1013
1014 <span class="h5"><a class="selflink" id="section-7.3.1.2" href="#section-7.3.1.2">7.3.1.2</a>. Sending the Success or Error Response</span>
1015
1016 The response (success or error) is sent over the same transport as
1017 the request was received on. If the request was received over UDP,
1018 the destination IP address and port of the response are the source IP
1019 address and port of the received request message, and the source IP
1020 address and port of the response are equal to the destination IP
1021 address and port of the received request message. If the request was
1022 received over TCP or TLS-over-TCP, the response is sent back on the
1023 same TCP connection as the request was received on.
1024
1025 <span class="h4"><a class="selflink" id="section-7.3.2" href="#section-7.3.2">7.3.2</a>. Processing an Indication</span>
1026
1027 If the indication contains unknown comprehension-required attributes,
1028 the indication is discarded and processing ceases.
1029
1030 The agent then does any additional checking that the method or the
1031 specific usage requires. If all the checks succeed, the agent then
1032 processes the indication. No response is generated for an
1033 indication.
1034
1035 For the Binding method, no additional checking or processing is
1036 required, unless the usage specifies otherwise. The mere receipt of
1037 the message by the agent has refreshed the "bindings" in the
1038 intervening NATs.
1039
1040 Since indications are not re-transmitted over UDP (unlike requests),
1041 there is no need to handle re-transmissions of indications at the
1042 sending agent.
1043
1044 <span class="h4"><a class="selflink" id="section-7.3.3" href="#section-7.3.3">7.3.3</a>. Processing a Success Response</span>
1045
1046 If the success response contains unknown comprehension-required
1047 attributes, the response is discarded and the transaction is
1048 considered to have failed.
1049
1050 The client then does any additional checking that the method or the
1051 specific usage requires. If all the checks succeed, the client then
1052 processes the success response.
1053
1054 For the Binding method, the client checks that the XOR-MAPPED-ADDRESS
1055 attribute is present in the response. The client checks the address
1056 family specified. If it is an unsupported address family, the
1057
1058
1059
1060 <span class="grey">Rosenberg, et al. Standards Track [Page 19]</span></pre>
1061 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-20" ></span>
1062 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1063
1064
1065 attribute SHOULD be ignored. If it is an unexpected but supported
1066 address family (for example, the Binding transaction was sent over
1067 IPv4, but the address family specified is IPv6), then the client MAY
1068 accept and use the value.
1069
1070 <span class="h4"><a class="selflink" id="section-7.3.4" href="#section-7.3.4">7.3.4</a>. Processing an Error Response</span>
1071
1072 If the error response contains unknown comprehension-required
1073 attributes, or if the error response does not contain an ERROR-CODE
1074 attribute, then the transaction is simply considered to have failed.
1075
1076 The client then does any processing specified by the authentication
1077 mechanism (see <a href="#section-10">Section 10</a>). This may result in a new transaction
1078 attempt.
1079
1080 The processing at this point depends on the error code, the method,
1081 and the usage; the following are the default rules:
1082
1083 o If the error code is 300 through 399, the client SHOULD consider
1084 the transaction as failed unless the ALTERNATE-SERVER extension is
1085 being used. See <a href="#section-11">Section 11</a>.
1086
1087 o If the error code is 400 through 499, the client declares the
1088 transaction failed; in the case of 420 (Unknown Attribute), the
1089 response should contain a UNKNOWN-ATTRIBUTES attribute that gives
1090 additional information.
1091
1092 o If the error code is 500 through 599, the client MAY resend the
1093 request; clients that do so MUST limit the number of times they do
1094 this.
1095
1096 Any other error code causes the client to consider the transaction
1097 failed.
1098
1099 <span class="h2"><a class="selflink" id="section-8" href="#section-8">8</a>. FINGERPRINT Mechanism</span>
1100
1101 This section describes an optional mechanism for STUN that aids in
1102 distinguishing STUN messages from packets of other protocols when the
1103 two are multiplexed on the same transport address. This mechanism is
1104 optional, and a STUN usage must describe if and when it is used. The
1105 FINGERPRINT mechanism is not backwards compatible with <a href="./rfc3489">RFC 3489</a>, and
1106 cannot be used in environments where such compatibility is required.
1107
1108 In some usages, STUN messages are multiplexed on the same transport
1109 address as other protocols, such as the Real Time Transport Protocol
1110 (RTP). In order to apply the processing described in <a href="#section-7">Section 7</a>, STUN
1111 messages must first be separated from the application packets.
1112
1113
1114
1115
1116 <span class="grey">Rosenberg, et al. Standards Track [Page 20]</span></pre>
1117 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-21" ></span>
1118 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1119
1120
1121 <a href="#section-6">Section 6</a> describes three fixed fields in the STUN header that can be
1122 used for this purpose. However, in some cases, these three fixed
1123 fields may not be sufficient.
1124
1125 When the FINGERPRINT extension is used, an agent includes the
1126 FINGERPRINT attribute in messages it sends to another agent.
1127 <a href="#section-15.5">Section 15.5</a> describes the placement and value of this attribute.
1128 When the agent receives what it believes is a STUN message, then, in
1129 addition to other basic checks, the agent also checks that the
1130 message contains a FINGERPRINT attribute and that the attribute
1131 contains the correct value. <a href="#section-7.3">Section 7.3</a> describes when in the
1132 overall processing of a STUN message the FINGERPRINT check is
1133 performed. This additional check helps the agent detect messages of
1134 other protocols that might otherwise seem to be STUN messages.
1135
1136 <span class="h2"><a class="selflink" id="section-9" href="#section-9">9</a>. DNS Discovery of a Server</span>
1137
1138 This section describes an optional procedure for STUN that allows a
1139 client to use DNS to determine the IP address and port of a server.
1140 A STUN usage must describe if and when this extension is used. To
1141 use this procedure, the client must know a server's domain name and a
1142 service name; the usage must also describe how the client obtains
1143 these. Hard-coding the domain name of the server into software is
1144 NOT RECOMMENDED in case the domain name is lost or needs to change
1145 for legal or other reasons.
1146
1147 When a client wishes to locate a STUN server in the public Internet
1148 that accepts Binding request/response transactions, the SRV service
1149 name is "stun". When it wishes to locate a STUN server that accepts
1150 Binding request/response transactions over a TLS session, the SRV
1151 service name is "stuns". STUN usages MAY define additional DNS SRV
1152 service names.
1153
1154 The domain name is resolved to a transport address using the SRV
1155 procedures specified in [<a href="./rfc2782" title="&quot;A DNS RR for specifying the location of services (DNS SRV)&quot;">RFC2782</a>]. The DNS SRV service name is the
1156 service name provided as input to this procedure. The protocol in
1157 the SRV lookup is the transport protocol the client will run STUN
1158 over: "udp" for UDP and "tcp" for TCP. Note that only "tcp" is
1159 defined with "stuns" at this time.
1160
1161 The procedures of <a href="./rfc2782">RFC 2782</a> are followed to determine the server to
1162 contact. <a href="./rfc2782">RFC 2782</a> spells out the details of how a set of SRV records
1163 is sorted and then tried. However, <a href="./rfc2782">RFC 2782</a> only states that the
1164 client should "try to connect to the (protocol, address, service)"
1165 without giving any details on what happens in the event of failure.
1166 When following these procedures, if the STUN transaction times out
1167 without receipt of a response, the client SHOULD retry the request to
1168
1169
1170
1171
1172 <span class="grey">Rosenberg, et al. Standards Track [Page 21]</span></pre>
1173 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-22" ></span>
1174 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1175
1176
1177 the next server in the ordered defined by <a href="./rfc2782">RFC 2782</a>. Such a retry is
1178 only possible for request/response transmissions, since indication
1179 transactions generate no response or timeout.
1180
1181 The default port for STUN requests is 3478, for both TCP and UDP.
1182
1183 Administrators of STUN servers SHOULD use this port in their SRV
1184 records for UDP and TCP. In all cases, the port in DNS MUST reflect
1185 the one on which the server is listening. The default port for STUN
1186 over TLS is 5349. Servers can run STUN over TLS on the same port as
1187 STUN over TCP if the server software supports determining whether the
1188 initial message is a TLS or STUN message.
1189
1190 If no SRV records were found, the client performs an A or AAAA record
1191 lookup of the domain name. The result will be a list of IP
1192 addresses, each of which can be contacted at the default port using
1193 UDP or TCP, independent of the STUN usage. For usages that require
1194 TLS, the client connects to one of the IP addresses using the default
1195 STUN over TLS port.
1196
1197 <span class="h2"><a class="selflink" id="section-10" href="#section-10">10</a>. Authentication and Message-Integrity Mechanisms</span>
1198
1199 This section defines two mechanisms for STUN that a client and server
1200 can use to provide authentication and message integrity; these two
1201 mechanisms are known as the short-term credential mechanism and the
1202 long-term credential mechanism. These two mechanisms are optional,
1203 and each usage must specify if and when these mechanisms are used.
1204 Consequently, both clients and servers will know which mechanism (if
1205 any) to follow based on knowledge of which usage applies. For
1206 example, a STUN server on the public Internet supporting ICE would
1207 have no authentication, whereas the STUN server functionality in an
1208 agent supporting connectivity checks would utilize short-term
1209 credentials. An overview of these two mechanisms is given in
1210 <a href="#section-3">Section 3</a>.
1211
1212 Each mechanism specifies the additional processing required to use
1213 that mechanism, extending the processing specified in <a href="#section-7">Section 7</a>. The
1214 additional processing occurs in three different places: when forming
1215 a message, when receiving a message immediately after the basic
1216 checks have been performed, and when doing the detailed processing of
1217 error responses.
1218
1219 <span class="h3"><a class="selflink" id="section-10.1" href="#section-10.1">10.1</a>. Short-Term Credential Mechanism</span>
1220
1221 The short-term credential mechanism assumes that, prior to the STUN
1222 transaction, the client and server have used some other protocol to
1223 exchange a credential in the form of a username and password. This
1224 credential is time-limited. The time limit is defined by the usage.
1225
1226
1227
1228 <span class="grey">Rosenberg, et al. Standards Track [Page 22]</span></pre>
1229 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-23" ></span>
1230 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1231
1232
1233 As an example, in the ICE usage [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>], the two endpoints use
1234 out-of-band signaling to agree on a username and password, and this
1235 username and password are applicable for the duration of the media
1236 session.
1237
1238 This credential is used to form a message-integrity check in each
1239 request and in many responses. There is no challenge and response as
1240 in the long-term mechanism; consequently, replay is prevented by
1241 virtue of the time-limited nature of the credential.
1242
1243 <span class="h4"><a class="selflink" id="section-10.1.1" href="#section-10.1.1">10.1.1</a>. Forming a Request or Indication</span>
1244
1245 For a request or indication message, the agent MUST include the
1246 USERNAME and MESSAGE-INTEGRITY attributes in the message. The HMAC
1247 for the MESSAGE-INTEGRITY attribute is computed as described in
1248 <a href="#section-15.4">Section 15.4</a>. Note that the password is never included in the
1249 request or indication.
1250
1251 <span class="h4"><a class="selflink" id="section-10.1.2" href="#section-10.1.2">10.1.2</a>. Receiving a Request or Indication</span>
1252
1253 After the agent has done the basic processing of a message, the agent
1254 performs the checks listed below in order specified:
1255
1256 o If the message does not contain both a MESSAGE-INTEGRITY and a
1257 USERNAME attribute:
1258
1259 * If the message is a request, the server MUST reject the request
1260 with an error response. This response MUST use an error code
1261 of 400 (Bad Request).
1262
1263 * If the message is an indication, the agent MUST silently
1264 discard the indication.
1265
1266 o If the USERNAME does not contain a username value currently valid
1267 within the server:
1268
1269 * If the message is a request, the server MUST reject the request
1270 with an error response. This response MUST use an error code
1271 of 401 (Unauthorized).
1272
1273 * If the message is an indication, the agent MUST silently
1274 discard the indication.
1275
1276 o Using the password associated with the username, compute the value
1277 for the message integrity as described in <a href="#section-15.4">Section 15.4</a>. If the
1278 resulting value does not match the contents of the MESSAGE-
1279 INTEGRITY attribute:
1280
1281
1282
1283
1284 <span class="grey">Rosenberg, et al. Standards Track [Page 23]</span></pre>
1285 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-24" ></span>
1286 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1287
1288
1289 * If the message is a request, the server MUST reject the request
1290 with an error response. This response MUST use an error code
1291 of 401 (Unauthorized).
1292
1293 * If the message is an indication, the agent MUST silently
1294 discard the indication.
1295
1296 If these checks pass, the agent continues to process the request or
1297 indication. Any response generated by a server MUST include the
1298 MESSAGE-INTEGRITY attribute, computed using the password utilized to
1299 authenticate the request. The response MUST NOT contain the USERNAME
1300 attribute.
1301
1302 If any of the checks fail, a server MUST NOT include a MESSAGE-
1303 INTEGRITY or USERNAME attribute in the error response. This is
1304 because, in these failure cases, the server cannot determine the
1305 shared secret necessary to compute MESSAGE-INTEGRITY.
1306
1307 <span class="h4"><a class="selflink" id="section-10.1.3" href="#section-10.1.3">10.1.3</a>. Receiving a Response</span>
1308
1309 The client looks for the MESSAGE-INTEGRITY attribute in the response.
1310 If present, the client computes the message integrity over the
1311 response as defined in <a href="#section-15.4">Section 15.4</a>, using the same password it
1312 utilized for the request. If the resulting value matches the
1313 contents of the MESSAGE-INTEGRITY attribute, the response is
1314 considered authenticated. If the value does not match, or if
1315 MESSAGE-INTEGRITY was absent, the response MUST be discarded, as if
1316 it was never received. This means that retransmits, if applicable,
1317 will continue.
1318
1319 <span class="h3"><a class="selflink" id="section-10.2" href="#section-10.2">10.2</a>. Long-Term Credential Mechanism</span>
1320
1321 The long-term credential mechanism relies on a long-term credential,
1322 in the form of a username and password that are shared between client
1323 and server. The credential is considered long-term since it is
1324 assumed that it is provisioned for a user, and remains in effect
1325 until the user is no longer a subscriber of the system, or is
1326 changed. This is basically a traditional "log-in" username and
1327 password given to users.
1328
1329 Because these usernames and passwords are expected to be valid for
1330 extended periods of time, replay prevention is provided in the form
1331 of a digest challenge. In this mechanism, the client initially sends
1332 a request, without offering any credentials or any integrity checks.
1333 The server rejects this request, providing the user a realm (used to
1334 guide the user or agent in selection of a username and password) and
1335 a nonce. The nonce provides the replay protection. It is a cookie,
1336 selected by the server, and encoded in such a way as to indicate a
1337
1338
1339
1340 <span class="grey">Rosenberg, et al. Standards Track [Page 24]</span></pre>
1341 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-25" ></span>
1342 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1343
1344
1345 duration of validity or client identity from which it is valid. The
1346 client retries the request, this time including its username and the
1347 realm, and echoing the nonce provided by the server. The client also
1348 includes a message-integrity, which provides an HMAC over the entire
1349 request, including the nonce. The server validates the nonce and
1350 checks the message integrity. If they match, the request is
1351 authenticated. If the nonce is no longer valid, it is considered
1352 "stale", and the server rejects the request, providing a new nonce.
1353
1354 In subsequent requests to the same server, the client reuses the
1355 nonce, username, realm, and password it used previously. In this
1356 way, subsequent requests are not rejected until the nonce becomes
1357 invalid by the server, in which case the rejection provides a new
1358 nonce to the client.
1359
1360 Note that the long-term credential mechanism cannot be used to
1361 protect indications, since indications cannot be challenged. Usages
1362 utilizing indications must either use a short-term credential or omit
1363 authentication and message integrity for them.
1364
1365 Since the long-term credential mechanism is susceptible to offline
1366 dictionary attacks, deployments SHOULD utilize passwords that are
1367 difficult to guess. In cases where the credentials are not entered
1368 by the user, but are rather placed on a client device during device
1369 provisioning, the password SHOULD have at least 128 bits of
1370 randomness. In cases where the credentials are entered by the user,
1371 they should follow best current practices around password structure.
1372
1373 <span class="h4"><a class="selflink" id="section-10.2.1" href="#section-10.2.1">10.2.1</a>. Forming a Request</span>
1374
1375 There are two cases when forming a request. In the first case, this
1376 is the first request from the client to the server (as identified by
1377 its IP address and port). In the second case, the client is
1378 submitting a subsequent request once a previous request/response
1379 transaction has completed successfully. Forming a request as a
1380 consequence of a 401 or 438 error response is covered in
1381 <a href="#section-10.2.3">Section 10.2.3</a> and is not considered a "subsequent request" and thus
1382 does not utilize the rules described in <a href="#section-10.2.1.2">Section 10.2.1.2</a>.
1383
1384 <span class="h5"><a class="selflink" id="section-10.2.1.1" href="#section-10.2.1.1">10.2.1.1</a>. First Request</span>
1385
1386 If the client has not completed a successful request/response
1387 transaction with the server (as identified by hostname, if the DNS
1388 procedures of <a href="#section-9">Section 9</a> are used, else IP address if not), it SHOULD
1389 omit the USERNAME, MESSAGE-INTEGRITY, REALM, and NONCE attributes.
1390 In other words, the very first request is sent as if there were no
1391 authentication or message integrity applied.
1392
1393
1394
1395
1396 <span class="grey">Rosenberg, et al. Standards Track [Page 25]</span></pre>
1397 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-26" ></span>
1398 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1399
1400
1401 <span class="h5"><a class="selflink" id="section-10.2.1.2" href="#section-10.2.1.2">10.2.1.2</a>. Subsequent Requests</span>
1402
1403 Once a request/response transaction has completed successfully, the
1404 client will have been presented a realm and nonce by the server, and
1405 selected a username and password with which it authenticated. The
1406 client SHOULD cache the username, password, realm, and nonce for
1407 subsequent communications with the server. When the client sends a
1408 subsequent request, it SHOULD include the USERNAME, REALM, and NONCE
1409 attributes with these cached values. It SHOULD include a MESSAGE-
1410 INTEGRITY attribute, computed as described in <a href="#section-15.4">Section 15.4</a> using the
1411 cached password.
1412
1413 <span class="h4"><a class="selflink" id="section-10.2.2" href="#section-10.2.2">10.2.2</a>. Receiving a Request</span>
1414
1415 After the server has done the basic processing of a request, it
1416 performs the checks listed below in the order specified:
1417
1418 o If the message does not contain a MESSAGE-INTEGRITY attribute, the
1419 server MUST generate an error response with an error code of 401
1420 (Unauthorized). This response MUST include a REALM value. It is
1421 RECOMMENDED that the REALM value be the domain name of the
1422 provider of the STUN server. The response MUST include a NONCE,
1423 selected by the server. The response SHOULD NOT contain a
1424 USERNAME or MESSAGE-INTEGRITY attribute.
1425
1426 o If the message contains a MESSAGE-INTEGRITY attribute, but is
1427 missing the USERNAME, REALM, or NONCE attribute, the server MUST
1428 generate an error response with an error code of 400 (Bad
1429 Request). This response SHOULD NOT include a USERNAME, NONCE,
1430 REALM, or MESSAGE-INTEGRITY attribute.
1431
1432 o If the NONCE is no longer valid, the server MUST generate an error
1433 response with an error code of 438 (Stale Nonce). This response
1434 MUST include NONCE and REALM attributes and SHOULD NOT include the
1435 USERNAME or MESSAGE-INTEGRITY attribute. Servers can invalidate
1436 nonces in order to provide additional security. See <a href="./rfc2617#section-4.3">Section&nbsp;4.3
1437 of [RFC2617]</a> for guidelines.
1438
1439 o If the username in the USERNAME attribute is not valid, the server
1440 MUST generate an error response with an error code of 401
1441 (Unauthorized). This response MUST include a REALM value. It is
1442 RECOMMENDED that the REALM value be the domain name of the
1443 provider of the STUN server. The response MUST include a NONCE,
1444 selected by the server. The response SHOULD NOT contain a
1445 USERNAME or MESSAGE-INTEGRITY attribute.
1446
1447
1448
1449
1450
1451
1452 <span class="grey">Rosenberg, et al. Standards Track [Page 26]</span></pre>
1453 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-27" ></span>
1454 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1455
1456
1457 o Using the password associated with the username in the USERNAME
1458 attribute, compute the value for the message integrity as
1459 described in <a href="#section-15.4">Section 15.4</a>. If the resulting value does not match
1460 the contents of the MESSAGE-INTEGRITY attribute, the server MUST
1461 reject the request with an error response. This response MUST use
1462 an error code of 401 (Unauthorized). It MUST include REALM and
1463 NONCE attributes and SHOULD NOT include the USERNAME or MESSAGE-
1464 INTEGRITY attribute.
1465
1466 If these checks pass, the server continues to process the request.
1467 Any response generated by the server (excepting the cases described
1468 above) MUST include the MESSAGE-INTEGRITY attribute, computed using
1469 the username and password utilized to authenticate the request. The
1470 REALM, NONCE, and USERNAME attributes SHOULD NOT be included.
1471
1472 <span class="h4"><a class="selflink" id="section-10.2.3" href="#section-10.2.3">10.2.3</a>. Receiving a Response</span>
1473
1474 If the response is an error response with an error code of 401
1475 (Unauthorized), the client SHOULD retry the request with a new
1476 transaction. This request MUST contain a USERNAME, determined by the
1477 client as the appropriate username for the REALM from the error
1478 response. The request MUST contain the REALM, copied from the error
1479 response. The request MUST contain the NONCE, copied from the error
1480 response. The request MUST contain the MESSAGE-INTEGRITY attribute,
1481 computed using the password associated with the username in the
1482 USERNAME attribute. The client MUST NOT perform this retry if it is
1483 not changing the USERNAME or REALM or its associated password, from
1484 the previous attempt.
1485
1486 If the response is an error response with an error code of 438 (Stale
1487 Nonce), the client MUST retry the request, using the new NONCE
1488 supplied in the 438 (Stale Nonce) response. This retry MUST also
1489 include the USERNAME, REALM, and MESSAGE-INTEGRITY.
1490
1491 The client looks for the MESSAGE-INTEGRITY attribute in the response
1492 (either success or failure). If present, the client computes the
1493 message integrity over the response as defined in <a href="#section-15.4">Section 15.4</a>, using
1494 the same password it utilized for the request. If the resulting
1495 value matches the contents of the MESSAGE-INTEGRITY attribute, the
1496 response is considered authenticated. If the value does not match,
1497 or if MESSAGE-INTEGRITY was absent, the response MUST be discarded,
1498 as if it was never received. This means that retransmits, if
1499 applicable, will continue.
1500
1501
1502
1503
1504
1505
1506
1507
1508 <span class="grey">Rosenberg, et al. Standards Track [Page 27]</span></pre>
1509 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-28" ></span>
1510 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1511
1512
1513 <span class="h2"><a class="selflink" id="section-11" href="#section-11">11</a>. ALTERNATE-SERVER Mechanism</span>
1514
1515 This section describes a mechanism in STUN that allows a server to
1516 redirect a client to another server. This extension is optional, and
1517 a usage must define if and when this extension is used.
1518
1519 A server using this extension redirects a client to another server by
1520 replying to a request message with an error response message with an
1521 error code of 300 (Try Alternate). The server MUST include an
1522 ALTERNATE-SERVER attribute in the error response. The error response
1523 message MAY be authenticated; however, there are uses cases for
1524 ALTERNATE-SERVER where authentication of the response is not possible
1525 or practical.
1526
1527 A client using this extension handles a 300 (Try Alternate) error
1528 code as follows. The client looks for an ALTERNATE-SERVER attribute
1529 in the error response. If one is found, then the client considers
1530 the current transaction as failed, and reattempts the request with
1531 the server specified in the attribute, using the same transport
1532 protocol used for the previous request. That request, if
1533 authenticated, MUST utilize the same credentials that the client
1534 would have used in the request to the server that performed the
1535 redirection. If the client has been redirected to a server on which
1536 it has already tried this request within the last five minutes, it
1537 MUST ignore the redirection and consider the transaction to have
1538 failed. This prevents infinite ping-ponging between servers in case
1539 of redirection loops.
1540
1541 <span class="h2"><a class="selflink" id="section-12" href="#section-12">12</a>. Backwards Compatibility with <a href="./rfc3489">RFC 3489</a></span>
1542
1543 This section defines procedures that allow a degree of backwards
1544 compatibility with the original protocol defined in <a href="./rfc3489">RFC 3489</a>
1545 [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>]. This mechanism is optional, meant to be utilized only in
1546 cases where a new client can connect to an old server, or vice versa.
1547 A usage must define if and when this procedure is used.
1548
1549 <a href="#section-19">Section 19</a> lists all the changes between this specification and <a href="./rfc3489">RFC</a>
1550 <a href="./rfc3489">3489</a> [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>]. However, not all of these differences are important,
1551 because "classic STUN" was only used in a few specific ways. For the
1552 purposes of this extension, the important changes are the following.
1553 In <a href="./rfc3489">RFC 3489</a>:
1554
1555 o UDP was the only supported transport.
1556
1557 o The field that is now the magic cookie field was a part of the
1558 transaction ID field, and transaction IDs were 128 bits long.
1559
1560
1561
1562
1563
1564 <span class="grey">Rosenberg, et al. Standards Track [Page 28]</span></pre>
1565 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-29" ></span>
1566 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1567
1568
1569 o The XOR-MAPPED-ADDRESS attribute did not exist, and the Binding
1570 method used the MAPPED-ADDRESS attribute instead.
1571
1572 o There were three comprehension-required attributes, RESPONSE-
1573 ADDRESS, CHANGE-REQUEST, and CHANGED-ADDRESS, that have been
1574 removed from this specification.
1575
1576 * CHANGE-REQUEST and CHANGED-ADDRESS are now part of the NAT
1577 Behavior Discovery usage [<a href="#ref-BEHAVE-NAT" title="&quot;NAT Behavior Discovery Using STUN&quot;">BEHAVE-NAT</a>], and the other is
1578 deprecated.
1579
1580 <span class="h3"><a class="selflink" id="section-12.1" href="#section-12.1">12.1</a>. Changes to Client Processing</span>
1581
1582 A client that wants to interoperate with an [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>] server SHOULD
1583 send a request message that uses the Binding method, contains no
1584 attributes, and uses UDP as the transport protocol to the server. If
1585 successful, the success response received from the server will
1586 contain a MAPPED-ADDRESS attribute rather than an XOR-MAPPED-ADDRESS
1587 attribute. A client seeking to interoperate with an older server
1588 MUST be prepared to receive either. Furthermore, the client MUST
1589 ignore any Reserved comprehension-required attributes that might
1590 appear in the response. Of the Reserved attributes in <a href="#section-18.2">Section 18.2</a>,
1591 0x0002, 0x0004, 0x0005, and 0x000B may appear in Binding responses
1592 from a server compliant to <a href="./rfc3489">RFC 3489</a>. Other than this change, the
1593 processing of the response is identical to the procedures described
1594 above.
1595
1596 <span class="h3"><a class="selflink" id="section-12.2" href="#section-12.2">12.2</a>. Changes to Server Processing</span>
1597
1598 A STUN server can detect when a given Binding request message was
1599 sent from an <a href="./rfc3489">RFC 3489</a> [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>] client by the absence of the correct
1600 value in the magic cookie field. When the server detects an <a href="./rfc3489">RFC 3489</a>
1601 client, it SHOULD copy the value seen in the magic cookie field in
1602 the Binding request to the magic cookie field in the Binding response
1603 message, and insert a MAPPED-ADDRESS attribute instead of an XOR-
1604 MAPPED-ADDRESS attribute.
1605
1606 The client might, in rare situations, include either the RESPONSE-
1607 ADDRESS or CHANGE-REQUEST attributes. In these situations, the
1608 server will view these as unknown comprehension-required attributes
1609 and reply with an error response. Since the mechanisms utilizing
1610 those attributes are no longer supported, this behavior is
1611 acceptable.
1612
1613 The <a href="./rfc3489">RFC 3489</a> version of STUN lacks both the magic cookie and the
1614 FINGERPRINT attribute that allows for a very high probability of
1615 correctly identifying STUN messages when multiplexed with other
1616 protocols. Therefore, STUN implementations that are backwards
1617
1618
1619
1620 <span class="grey">Rosenberg, et al. Standards Track [Page 29]</span></pre>
1621 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-30" ></span>
1622 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1623
1624
1625 compatible with <a href="./rfc3489">RFC 3489</a> SHOULD NOT be used in cases where STUN will
1626 be multiplexed with another protocol. However, that should not be an
1627 issue as such multiplexing was not available in <a href="./rfc3489">RFC 3489</a>.
1628
1629 <span class="h2"><a class="selflink" id="section-13" href="#section-13">13</a>. Basic Server Behavior</span>
1630
1631 This section defines the behavior of a basic, stand-alone STUN
1632 server. A basic STUN server provides clients with server reflexive
1633 transport addresses by receiving and replying to STUN Binding
1634 requests.
1635
1636 The STUN server MUST support the Binding method. It SHOULD NOT
1637 utilize the short-term or long-term credential mechanism. This is
1638 because the work involved in authenticating the request is more than
1639 the work in simply processing it. It SHOULD NOT utilize the
1640 ALTERNATE-SERVER mechanism for the same reason. It MUST support UDP
1641 and TCP. It MAY support STUN over TCP/TLS; however, TLS provides
1642 minimal security benefits in this basic mode of operation. It MAY
1643 utilize the FINGERPRINT mechanism but MUST NOT require it. Since the
1644 stand-alone server only runs STUN, FINGERPRINT provides no benefit.
1645 Requiring it would break compatibility with <a href="./rfc3489">RFC 3489</a>, and such
1646 compatibility is desirable in a stand-alone server. Stand-alone STUN
1647 servers SHOULD support backwards compatibility with [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>]
1648 clients, as described in <a href="#section-12">Section 12</a>.
1649
1650 It is RECOMMENDED that administrators of STUN servers provide DNS
1651 entries for those servers as described in <a href="#section-9">Section 9</a>.
1652
1653 A basic STUN server is not a solution for NAT traversal by itself.
1654 However, it can be utilized as part of a solution through STUN
1655 usages. This is discussed further in <a href="#section-14">Section 14</a>.
1656
1657 <span class="h2"><a class="selflink" id="section-14" href="#section-14">14</a>. STUN Usages</span>
1658
1659 STUN by itself is not a solution to the NAT traversal problem.
1660 Rather, STUN defines a tool that can be used inside a larger
1661 solution. The term "STUN usage" is used for any solution that uses
1662 STUN as a component.
1663
1664 At the time of writing, three STUN usages are defined: Interactive
1665 Connectivity Establishment (ICE) [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>], Client-initiated
1666 connections for SIP [<a href="#ref-SIP-OUTBOUND" title="&quot;Managing Client Initiated Connections in the Session Initiation Protocol (SIP)&quot;">SIP-OUTBOUND</a>], and NAT Behavior Discovery
1667 [<a href="#ref-BEHAVE-NAT" title="&quot;NAT Behavior Discovery Using STUN&quot;">BEHAVE-NAT</a>]. Other STUN usages may be defined in the future.
1668
1669 A STUN usage defines how STUN is actually utilized -- when to send
1670 requests, what to do with the responses, and which optional
1671 procedures defined here (or in an extension to STUN) are to be used.
1672 A usage would also define:
1673
1674
1675
1676 <span class="grey">Rosenberg, et al. Standards Track [Page 30]</span></pre>
1677 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-31" ></span>
1678 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1679
1680
1681 o Which STUN methods are used.
1682
1683 o What authentication and message-integrity mechanisms are used.
1684
1685 o The considerations around manual vs. automatic key derivation for
1686 the integrity mechanism, as discussed in [<a href="./rfc4107" title="&quot;Guidelines for Cryptographic Key Management&quot;">RFC4107</a>].
1687
1688 o What mechanisms are used to distinguish STUN messages from other
1689 messages. When STUN is run over TCP, a framing mechanism may be
1690 required.
1691
1692 o How a STUN client determines the IP address and port of the STUN
1693 server.
1694
1695 o Whether backwards compatibility to <a href="./rfc3489">RFC 3489</a> is required.
1696
1697 o What optional attributes defined here (such as FINGERPRINT and
1698 ALTERNATE-SERVER) or in other extensions are required.
1699
1700 In addition, any STUN usage must consider the security implications
1701 of using STUN in that usage. A number of attacks against STUN are
1702 known (see the Security Considerations section in this document), and
1703 any usage must consider how these attacks can be thwarted or
1704 mitigated.
1705
1706 Finally, a usage must consider whether its usage of STUN is an
1707 example of the Unilateral Self-Address Fixing approach to NAT
1708 traversal, and if so, address the questions raised in <a href="./rfc3424">RFC 3424</a>
1709 [<a href="./rfc3424" title="&quot;IAB Considerations for UNilateral Self-Address Fixing (UNSAF) Across Network Address Translation&quot;">RFC3424</a>].
1710
1711 <span class="h2"><a class="selflink" id="section-15" href="#section-15">15</a>. STUN Attributes</span>
1712
1713 After the STUN header are zero or more attributes. Each attribute
1714 MUST be TLV encoded, with a 16-bit type, 16-bit length, and value.
1715 Each STUN attribute MUST end on a 32-bit boundary. As mentioned
1716 above, all fields in an attribute are transmitted most significant
1717 bit first.
1718
1719 0 1 2 3
1720 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
1721 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1722 | Type | Length |
1723 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1724 | Value (variable) ....
1725 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1726
1727 Figure 4: Format of STUN Attributes
1728
1729
1730
1731
1732 <span class="grey">Rosenberg, et al. Standards Track [Page 31]</span></pre>
1733 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-32" ></span>
1734 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1735
1736
1737 The value in the length field MUST contain the length of the Value
1738 part of the attribute, prior to padding, measured in bytes. Since
1739 STUN aligns attributes on 32-bit boundaries, attributes whose content
1740 is not a multiple of 4 bytes are padded with 1, 2, or 3 bytes of
1741 padding so that its value contains a multiple of 4 bytes. The
1742 padding bits are ignored, and may be any value.
1743
1744 Any attribute type MAY appear more than once in a STUN message.
1745 Unless specified otherwise, the order of appearance is significant:
1746 only the first occurrence needs to be processed by a receiver, and
1747 any duplicates MAY be ignored by a receiver.
1748
1749 To allow future revisions of this specification to add new attributes
1750 if needed, the attribute space is divided into two ranges.
1751 Attributes with type values between 0x0000 and 0x7FFF are
1752 comprehension-required attributes, which means that the STUN agent
1753 cannot successfully process the message unless it understands the
1754 attribute. Attributes with type values between 0x8000 and 0xFFFF are
1755 comprehension-optional attributes, which means that those attributes
1756 can be ignored by the STUN agent if it does not understand them.
1757
1758 The set of STUN attribute types is maintained by IANA. The initial
1759 set defined by this specification is found in <a href="#section-18.2">Section 18.2</a>.
1760
1761 The rest of this section describes the format of the various
1762 attributes defined in this specification.
1763
1764 <span class="h3"><a class="selflink" id="section-15.1" href="#section-15.1">15.1</a>. MAPPED-ADDRESS</span>
1765
1766 The MAPPED-ADDRESS attribute indicates a reflexive transport address
1767 of the client. It consists of an 8-bit address family and a 16-bit
1768 port, followed by a fixed-length value representing the IP address.
1769 If the address family is IPv4, the address MUST be 32 bits. If the
1770 address family is IPv6, the address MUST be 128 bits. All fields
1771 must be in network byte order.
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788 <span class="grey">Rosenberg, et al. Standards Track [Page 32]</span></pre>
1789 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-33" ></span>
1790 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1791
1792
1793 The format of the MAPPED-ADDRESS attribute is:
1794
1795 0 1 2 3
1796 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
1797 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1798 |0 0 0 0 0 0 0 0| Family | Port |
1799 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1800 | |
1801 | Address (32 bits or 128 bits) |
1802 | |
1803 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1804
1805 Figure 5: Format of MAPPED-ADDRESS Attribute
1806
1807 The address family can take on the following values:
1808
1809 0x01:IPv4
1810 0x02:IPv6
1811
1812 The first 8 bits of the MAPPED-ADDRESS MUST be set to 0 and MUST be
1813 ignored by receivers. These bits are present for aligning parameters
1814 on natural 32-bit boundaries.
1815
1816 This attribute is used only by servers for achieving backwards
1817 compatibility with <a href="./rfc3489">RFC 3489</a> [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>] clients.
1818
1819 <span class="h3"><a class="selflink" id="section-15.2" href="#section-15.2">15.2</a>. XOR-MAPPED-ADDRESS</span>
1820
1821 The XOR-MAPPED-ADDRESS attribute is identical to the MAPPED-ADDRESS
1822 attribute, except that the reflexive transport address is obfuscated
1823 through the XOR function.
1824
1825 The format of the XOR-MAPPED-ADDRESS is:
1826
1827 0 1 2 3
1828 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
1829 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1830 |x x x x x x x x| Family | X-Port |
1831 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1832 | X-Address (Variable)
1833 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
1834
1835 Figure 6: Format of XOR-MAPPED-ADDRESS Attribute
1836
1837 The Family represents the IP address family, and is encoded
1838 identically to the Family in MAPPED-ADDRESS.
1839
1840
1841
1842
1843
1844 <span class="grey">Rosenberg, et al. Standards Track [Page 33]</span></pre>
1845 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-34" ></span>
1846 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1847
1848
1849 X-Port is computed by taking the mapped port in host byte order,
1850 XOR'ing it with the most significant 16 bits of the magic cookie, and
1851 then the converting the result to network byte order. If the IP
1852 address family is IPv4, X-Address is computed by taking the mapped IP
1853 address in host byte order, XOR'ing it with the magic cookie, and
1854 converting the result to network byte order. If the IP address
1855 family is IPv6, X-Address is computed by taking the mapped IP address
1856 in host byte order, XOR'ing it with the concatenation of the magic
1857 cookie and the 96-bit transaction ID, and converting the result to
1858 network byte order.
1859
1860 The rules for encoding and processing the first 8 bits of the
1861 attribute's value, the rules for handling multiple occurrences of the
1862 attribute, and the rules for processing address families are the same
1863 as for MAPPED-ADDRESS.
1864
1865 Note: XOR-MAPPED-ADDRESS and MAPPED-ADDRESS differ only in their
1866 encoding of the transport address. The former encodes the transport
1867 address by exclusive-or'ing it with the magic cookie. The latter
1868 encodes it directly in binary. <a href="./rfc3489">RFC 3489</a> originally specified only
1869 MAPPED-ADDRESS. However, deployment experience found that some NATs
1870 rewrite the 32-bit binary payloads containing the NAT's public IP
1871 address, such as STUN's MAPPED-ADDRESS attribute, in the well-meaning
1872 but misguided attempt at providing a generic ALG function. Such
1873 behavior interferes with the operation of STUN and also causes
1874 failure of STUN's message-integrity checking.
1875
1876 <span class="h3"><a class="selflink" id="section-15.3" href="#section-15.3">15.3</a>. USERNAME</span>
1877
1878 The USERNAME attribute is used for message integrity. It identifies
1879 the username and password combination used in the message-integrity
1880 check.
1881
1882 The value of USERNAME is a variable-length value. It MUST contain a
1883 UTF-8 [<a href="./rfc3629" title="&quot;UTF-8, a transformation format of ISO 10646&quot;">RFC3629</a>] encoded sequence of less than 513 bytes, and MUST
1884 have been processed using SASLprep [<a href="./rfc4013" title="&quot;SASLprep: Stringprep Profile for User Names and Passwords&quot;">RFC4013</a>].
1885
1886 <span class="h3"><a class="selflink" id="section-15.4" href="#section-15.4">15.4</a>. MESSAGE-INTEGRITY</span>
1887
1888 The MESSAGE-INTEGRITY attribute contains an HMAC-SHA1 [<a href="./rfc2104" title="&quot;HMAC: Keyed-Hashing for Message Authentication&quot;">RFC2104</a>] of
1889 the STUN message. The MESSAGE-INTEGRITY attribute can be present in
1890 any STUN message type. Since it uses the SHA1 hash, the HMAC will be
1891 20 bytes. The text used as input to HMAC is the STUN message,
1892 including the header, up to and including the attribute preceding the
1893 MESSAGE-INTEGRITY attribute. With the exception of the FINGERPRINT
1894 attribute, which appears after MESSAGE-INTEGRITY, agents MUST ignore
1895 all other attributes that follow MESSAGE-INTEGRITY.
1896
1897
1898
1899
1900 <span class="grey">Rosenberg, et al. Standards Track [Page 34]</span></pre>
1901 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-35" ></span>
1902 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1903
1904
1905 The key for the HMAC depends on whether long-term or short-term
1906 credentials are in use. For long-term credentials, the key is 16
1907 bytes:
1908
1909 key = MD5(username ":" realm ":" SASLprep(password))
1910
1911 That is, the 16-byte key is formed by taking the MD5 hash of the
1912 result of concatenating the following five fields: (1) the username,
1913 with any quotes and trailing nulls removed, as taken from the
1914 USERNAME attribute (in which case SASLprep has already been applied);
1915 (2) a single colon; (3) the realm, with any quotes and trailing nulls
1916 removed; (4) a single colon; and (5) the password, with any trailing
1917 nulls removed and after processing using SASLprep. For example, if
1918 the username was 'user', the realm was 'realm', and the password was
1919 'pass', then the 16-byte HMAC key would be the result of performing
1920 an MD5 hash on the string 'user:realm:pass', the resulting hash being
1921 0x8493fbc53ba582fb4c044c456bdc40eb.
1922
1923 For short-term credentials:
1924
1925 key = SASLprep(password)
1926
1927 where MD5 is defined in <a href="./rfc1321">RFC 1321</a> [<a href="./rfc1321" title="&quot;The MD5 Message-Digest Algorithm&quot;">RFC1321</a>] and SASLprep() is defined
1928 in <a href="./rfc4013">RFC 4013</a> [<a href="./rfc4013" title="&quot;SASLprep: Stringprep Profile for User Names and Passwords&quot;">RFC4013</a>].
1929
1930 The structure of the key when used with long-term credentials
1931 facilitates deployment in systems that also utilize SIP. Typically,
1932 SIP systems utilizing SIP's digest authentication mechanism do not
1933 actually store the password in the database. Rather, they store a
1934 value called H(A1), which is equal to the key defined above.
1935
1936 Based on the rules above, the hash used to construct MESSAGE-
1937 INTEGRITY includes the length field from the STUN message header.
1938 Prior to performing the hash, the MESSAGE-INTEGRITY attribute MUST be
1939 inserted into the message (with dummy content). The length MUST then
1940 be set to point to the length of the message up to, and including,
1941 the MESSAGE-INTEGRITY attribute itself, but excluding any attributes
1942 after it. Once the computation is performed, the value of the
1943 MESSAGE-INTEGRITY attribute can be filled in, and the value of the
1944 length in the STUN header can be set to its correct value -- the
1945 length of the entire message. Similarly, when validating the
1946 MESSAGE-INTEGRITY, the length field should be adjusted to point to
1947 the end of the MESSAGE-INTEGRITY attribute prior to calculating the
1948 HMAC. Such adjustment is necessary when attributes, such as
1949 FINGERPRINT, appear after MESSAGE-INTEGRITY.
1950
1951
1952
1953
1954
1955
1956 <span class="grey">Rosenberg, et al. Standards Track [Page 35]</span></pre>
1957 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-36" ></span>
1958 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
1959
1960
1961 <span class="h3"><a class="selflink" id="section-15.5" href="#section-15.5">15.5</a>. FINGERPRINT</span>
1962
1963 The FINGERPRINT attribute MAY be present in all STUN messages. The
1964 value of the attribute is computed as the CRC-32 of the STUN message
1965 up to (but excluding) the FINGERPRINT attribute itself, XOR'ed with
1966 the 32-bit value 0x5354554e (the XOR helps in cases where an
1967 application packet is also using CRC-32 in it). The 32-bit CRC is
1968 the one defined in ITU V.42 [<a href="#ref-ITU.V42.2002" title="&quot;Error- correcting Procedures for DCEs Using Asynchronous- to-Synchronous Conversion&quot;">ITU.V42.2002</a>], which has a generator
1969 polynomial of x32+x26+x23+x22+x16+x12+x11+x10+x8+x7+x5+x4+x2+x+1.
1970 When present, the FINGERPRINT attribute MUST be the last attribute in
1971 the message, and thus will appear after MESSAGE-INTEGRITY.
1972
1973 The FINGERPRINT attribute can aid in distinguishing STUN packets from
1974 packets of other protocols. See <a href="#section-8">Section 8</a>.
1975
1976 As with MESSAGE-INTEGRITY, the CRC used in the FINGERPRINT attribute
1977 covers the length field from the STUN message header. Therefore,
1978 this value must be correct and include the CRC attribute as part of
1979 the message length, prior to computation of the CRC. When using the
1980 FINGERPRINT attribute in a message, the attribute is first placed
1981 into the message with a dummy value, then the CRC is computed, and
1982 then the value of the attribute is updated. If the MESSAGE-INTEGRITY
1983 attribute is also present, then it must be present with the correct
1984 message-integrity value before the CRC is computed, since the CRC is
1985 done over the value of the MESSAGE-INTEGRITY attribute as well.
1986
1987 <span class="h3"><a class="selflink" id="section-15.6" href="#section-15.6">15.6</a>. ERROR-CODE</span>
1988
1989 The ERROR-CODE attribute is used in error response messages. It
1990 contains a numeric error code value in the range of 300 to 699 plus a
1991 textual reason phrase encoded in UTF-8 [<a href="./rfc3629" title="&quot;UTF-8, a transformation format of ISO 10646&quot;">RFC3629</a>], and is consistent
1992 in its code assignments and semantics with SIP [<a href="./rfc3261" title="&quot;SIP: Session Initiation Protocol&quot;">RFC3261</a>] and HTTP
1993 [<a href="./rfc2616" title="&quot;Hypertext Transfer Protocol -- HTTP/1.1&quot;">RFC2616</a>]. The reason phrase is meant for user consumption, and can
1994 be anything appropriate for the error code. Recommended reason
1995 phrases for the defined error codes are included in the IANA registry
1996 for error codes. The reason phrase MUST be a UTF-8 [<a href="./rfc3629" title="&quot;UTF-8, a transformation format of ISO 10646&quot;">RFC3629</a>] encoded
1997 sequence of less than 128 characters (which can be as long as 763
1998 bytes).
1999
2000 0 1 2 3
2001 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
2002 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2003 | Reserved, should be 0 |Class| Number |
2004 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2005 | Reason Phrase (variable) ..
2006 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2007
2008 Figure 7: ERROR-CODE Attribute
2009
2010
2011
2012 <span class="grey">Rosenberg, et al. Standards Track [Page 36]</span></pre>
2013 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-37" ></span>
2014 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2015
2016
2017 To facilitate processing, the class of the error code (the hundreds
2018 digit) is encoded separately from the rest of the code, as shown in
2019 Figure 7.
2020
2021 The Reserved bits SHOULD be 0, and are for alignment on 32-bit
2022 boundaries. Receivers MUST ignore these bits. The Class represents
2023 the hundreds digit of the error code. The value MUST be between 3
2024 and 6. The Number represents the error code modulo 100, and its
2025 value MUST be between 0 and 99.
2026
2027 The following error codes, along with their recommended reason
2028 phrases, are defined:
2029
2030 300 Try Alternate: The client should contact an alternate server for
2031 this request. This error response MUST only be sent if the
2032 request included a USERNAME attribute and a valid MESSAGE-
2033 INTEGRITY attribute; otherwise, it MUST NOT be sent and error
2034 code 400 (Bad Request) is suggested. This error response MUST
2035 be protected with the MESSAGE-INTEGRITY attribute, and receivers
2036 MUST validate the MESSAGE-INTEGRITY of this response before
2037 redirecting themselves to an alternate server.
2038
2039 Note: Failure to generate and validate message integrity
2040 for a 300 response allows an on-path attacker to falsify a
2041 300 response thus causing subsequent STUN messages to be
2042 sent to a victim.
2043
2044 400 Bad Request: The request was malformed. The client SHOULD NOT
2045 retry the request without modification from the previous
2046 attempt. The server may not be able to generate a valid
2047 MESSAGE-INTEGRITY for this error, so the client MUST NOT expect
2048 a valid MESSAGE-INTEGRITY attribute on this response.
2049
2050 401 Unauthorized: The request did not contain the correct
2051 credentials to proceed. The client should retry the request
2052 with proper credentials.
2053
2054 420 Unknown Attribute: The server received a STUN packet containing
2055 a comprehension-required attribute that it did not understand.
2056 The server MUST put this unknown attribute in the UNKNOWN-
2057 ATTRIBUTE attribute of its error response.
2058
2059 438 Stale Nonce: The NONCE used by the client was no longer valid.
2060 The client should retry, using the NONCE provided in the
2061 response.
2062
2063 500 Server Error: The server has suffered a temporary error. The
2064 client should try again.
2065
2066
2067
2068 <span class="grey">Rosenberg, et al. Standards Track [Page 37]</span></pre>
2069 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-38" ></span>
2070 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2071
2072
2073 <span class="h3"><a class="selflink" id="section-15.7" href="#section-15.7">15.7</a>. REALM</span>
2074
2075 The REALM attribute may be present in requests and responses. It
2076 contains text that meets the grammar for "realm-value" as described
2077 in <a href="./rfc3261">RFC 3261</a> [<a href="./rfc3261" title="&quot;SIP: Session Initiation Protocol&quot;">RFC3261</a>] but without the double quotes and their
2078 surrounding whitespace. That is, it is an unquoted realm-value (and
2079 is therefore a sequence of qdtext or quoted-pair). It MUST be a
2080 UTF-8 [<a href="./rfc3629" title="&quot;UTF-8, a transformation format of ISO 10646&quot;">RFC3629</a>] encoded sequence of less than 128 characters (which
2081 can be as long as 763 bytes), and MUST have been processed using
2082 SASLprep [<a href="./rfc4013" title="&quot;SASLprep: Stringprep Profile for User Names and Passwords&quot;">RFC4013</a>].
2083
2084 Presence of the REALM attribute in a request indicates that long-term
2085 credentials are being used for authentication. Presence in certain
2086 error responses indicates that the server wishes the client to use a
2087 long-term credential for authentication.
2088
2089 <span class="h3"><a class="selflink" id="section-15.8" href="#section-15.8">15.8</a>. NONCE</span>
2090
2091 The NONCE attribute may be present in requests and responses. It
2092 contains a sequence of qdtext or quoted-pair, which are defined in
2093 <a href="./rfc3261">RFC 3261</a> [<a href="./rfc3261" title="&quot;SIP: Session Initiation Protocol&quot;">RFC3261</a>]. Note that this means that the NONCE attribute
2094 will not contain actual quote characters. See <a href="./rfc2617">RFC 2617</a> <a href="./rfc2617#section-4.3">[RFC2617],
2095 Section&nbsp;4.3</a>, for guidance on selection of nonce values in a server.
2096
2097 It MUST be less than 128 characters (which can be as long as 763
2098 bytes).
2099
2100 <span class="h3"><a class="selflink" id="section-15.9" href="#section-15.9">15.9</a>. UNKNOWN-ATTRIBUTES</span>
2101
2102 The UNKNOWN-ATTRIBUTES attribute is present only in an error response
2103 when the response code in the ERROR-CODE attribute is 420.
2104
2105 The attribute contains a list of 16-bit values, each of which
2106 represents an attribute type that was not understood by the server.
2107
2108 0 1 2 3
2109 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
2110 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2111 | Attribute 1 Type | Attribute 2 Type |
2112 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2113 | Attribute 3 Type | Attribute 4 Type ...
2114 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
2115
2116
2117 Figure 8: Format of UNKNOWN-ATTRIBUTES Attribute
2118
2119
2120
2121
2122
2123
2124 <span class="grey">Rosenberg, et al. Standards Track [Page 38]</span></pre>
2125 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-39" ></span>
2126 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2127
2128
2129 Note: In [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>], this field was padded to 32 by duplicating the
2130 last attribute. In this version of the specification, the normal
2131 padding rules for attributes are used instead.
2132
2133 <span class="h3"><a class="selflink" id="section-15.10" href="#section-15.10">15.10</a>. SOFTWARE</span>
2134
2135 The SOFTWARE attribute contains a textual description of the software
2136 being used by the agent sending the message. It is used by clients
2137 and servers. Its value SHOULD include manufacturer and version
2138 number. The attribute has no impact on operation of the protocol,
2139 and serves only as a tool for diagnostic and debugging purposes. The
2140 value of SOFTWARE is variable length. It MUST be a UTF-8 [<a href="./rfc3629" title="&quot;UTF-8, a transformation format of ISO 10646&quot;">RFC3629</a>]
2141 encoded sequence of less than 128 characters (which can be as long as
2142 763 bytes).
2143
2144 <span class="h3"><a class="selflink" id="section-15.11" href="#section-15.11">15.11</a>. ALTERNATE-SERVER</span>
2145
2146 The alternate server represents an alternate transport address
2147 identifying a different STUN server that the STUN client should try.
2148
2149 It is encoded in the same way as MAPPED-ADDRESS, and thus refers to a
2150 single server by IP address. The IP address family MUST be identical
2151 to that of the source IP address of the request.
2152
2153 <span class="h2"><a class="selflink" id="section-16" href="#section-16">16</a>. Security Considerations</span>
2154
2155 <span class="h3"><a class="selflink" id="section-16.1" href="#section-16.1">16.1</a>. Attacks against the Protocol</span>
2156
2157 <span class="h4"><a class="selflink" id="section-16.1.1" href="#section-16.1.1">16.1.1</a>. Outside Attacks</span>
2158
2159 An attacker can try to modify STUN messages in transit, in order to
2160 cause a failure in STUN operation. These attacks are detected for
2161 both requests and responses through the message-integrity mechanism,
2162 using either a short-term or long-term credential. Of course, once
2163 detected, the manipulated packets will be dropped, causing the STUN
2164 transaction to effectively fail. This attack is possible only by an
2165 on-path attacker.
2166
2167 An attacker that can observe, but not modify, STUN messages in-
2168 transit (for example, an attacker present on a shared access medium,
2169 such as Wi-Fi), can see a STUN request, and then immediately send a
2170 STUN response, typically an error response, in order to disrupt STUN
2171 processing. This attack is also prevented for messages that utilize
2172 MESSAGE-INTEGRITY. However, some error responses, those related to
2173 authentication in particular, cannot be protected by MESSAGE-
2174 INTEGRITY. When STUN itself is run over a secure transport protocol
2175 (e.g., TLS), these attacks are completely mitigated.
2176
2177
2178
2179
2180 <span class="grey">Rosenberg, et al. Standards Track [Page 39]</span></pre>
2181 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-40" ></span>
2182 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2183
2184
2185 Depending on the STUN usage, these attacks may be of minimal
2186 consequence and thus do not require message integrity to mitigate.
2187 For example, when STUN is used to a basic STUN server to discover a
2188 server reflexive candidate for usage with ICE, authentication and
2189 message integrity are not required since these attacks are detected
2190 during the connectivity check phase. The connectivity checks
2191 themselves, however, require protection for proper operation of ICE
2192 overall. As described in <a href="#section-14">Section 14</a>, STUN usages describe when
2193 authentication and message integrity are needed.
2194
2195 Since STUN uses the HMAC of a shared secret for authentication and
2196 integrity protection, it is subject to offline dictionary attacks.
2197 When authentication is utilized, it SHOULD be with a strong password
2198 that is not readily subject to offline dictionary attacks.
2199 Protection of the channel itself, using TLS, mitigates these attacks.
2200 However, STUN is most often run over UDP and in those cases, strong
2201 passwords are the only way to protect against these attacks.
2202
2203 <span class="h4"><a class="selflink" id="section-16.1.2" href="#section-16.1.2">16.1.2</a>. Inside Attacks</span>
2204
2205 A rogue client may try to launch a DoS attack against a server by
2206 sending it a large number of STUN requests. Fortunately, STUN
2207 requests can be processed statelessly by a server, making such
2208 attacks hard to launch.
2209
2210 A rogue client may use a STUN server as a reflector, sending it
2211 requests with a falsified source IP address and port. In such a
2212 case, the response would be delivered to that source IP and port.
2213 There is no amplification of the number of packets with this attack
2214 (the STUN server sends one packet for each packet sent by the
2215 client), though there is a small increase in the amount of data,
2216 since STUN responses are typically larger than requests. This attack
2217 is mitigated by ingress source address filtering.
2218
2219 Revealing the specific software version of the agent through the
2220 SOFTWARE attribute might allow them to become more vulnerable to
2221 attacks against software that is known to contain security holes.
2222 Implementers SHOULD make usage of the SOFTWARE attribute a
2223 configurable option.
2224
2225 <span class="h3"><a class="selflink" id="section-16.2" href="#section-16.2">16.2</a>. Attacks Affecting the Usage</span>
2226
2227 This section lists attacks that might be launched against a usage of
2228 STUN. Each STUN usage must consider whether these attacks are
2229 applicable to it, and if so, discuss counter-measures.
2230
2231 Most of the attacks in this section revolve around an attacker
2232 modifying the reflexive address learned by a STUN client through a
2233
2234
2235
2236 <span class="grey">Rosenberg, et al. Standards Track [Page 40]</span></pre>
2237 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-41" ></span>
2238 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2239
2240
2241 Binding request/response transaction. Since the usage of the
2242 reflexive address is a function of the usage, the applicability and
2243 remediation of these attacks are usage-specific. In common
2244 situations, modification of the reflexive address by an on-path
2245 attacker is easy to do. Consider, for example, the common situation
2246 where STUN is run directly over UDP. In this case, an on-path
2247 attacker can modify the source IP address of the Binding request
2248 before it arrives at the STUN server. The STUN server will then
2249 return this IP address in the XOR-MAPPED-ADDRESS attribute to the
2250 client, and send the response back to that (falsified) IP address and
2251 port. If the attacker can also intercept this response, it can
2252 direct it back towards the client. Protecting against this attack by
2253 using a message-integrity check is impossible, since a message-
2254 integrity value cannot cover the source IP address, since the
2255 intervening NAT must be able to modify this value. Instead, one
2256 solution to preventing the attacks listed below is for the client to
2257 verify the reflexive address learned, as is done in ICE [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>].
2258 Other usages may use other means to prevent these attacks.
2259
2260 <span class="h4"><a class="selflink" id="section-16.2.1" href="#section-16.2.1">16.2.1</a>. Attack I: Distributed DoS (DDoS) against a Target</span>
2261
2262 In this attack, the attacker provides one or more clients with the
2263 same faked reflexive address that points to the intended target.
2264 This will trick the STUN clients into thinking that their reflexive
2265 addresses are equal to that of the target. If the clients hand out
2266 that reflexive address in order to receive traffic on it (for
2267 example, in SIP messages), the traffic will instead be sent to the
2268 target. This attack can provide substantial amplification,
2269 especially when used with clients that are using STUN to enable
2270 multimedia applications. However, it can only be launched against
2271 targets for which packets from the STUN server to the target pass
2272 through the attacker, limiting the cases in which it is possible.
2273
2274 <span class="h4"><a class="selflink" id="section-16.2.2" href="#section-16.2.2">16.2.2</a>. Attack II: Silencing a Client</span>
2275
2276 In this attack, the attacker provides a STUN client with a faked
2277 reflexive address. The reflexive address it provides is a transport
2278 address that routes to nowhere. As a result, the client won't
2279 receive any of the packets it expects to receive when it hands out
2280 the reflexive address. This exploitation is not very interesting for
2281 the attacker. It impacts a single client, which is frequently not
2282 the desired target. Moreover, any attacker that can mount the attack
2283 could also deny service to the client by other means, such as
2284 preventing the client from receiving any response from the STUN
2285 server, or even a DHCP server. As with the attack in <a href="#section-16.2.1">Section 16.2.1</a>,
2286 this attack is only possible when the attacker is on path for packets
2287 sent from the STUN server towards this unused IP address.
2288
2289
2290
2291
2292 <span class="grey">Rosenberg, et al. Standards Track [Page 41]</span></pre>
2293 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-42" ></span>
2294 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2295
2296
2297 <span class="h4"><a class="selflink" id="section-16.2.3" href="#section-16.2.3">16.2.3</a>. Attack III: Assuming the Identity of a Client</span>
2298
2299 This attack is similar to attack II. However, the faked reflexive
2300 address points to the attacker itself. This allows the attacker to
2301 receive traffic that was destined for the client.
2302
2303 <span class="h4"><a class="selflink" id="section-16.2.4" href="#section-16.2.4">16.2.4</a>. Attack IV: Eavesdropping</span>
2304
2305 In this attack, the attacker forces the client to use a reflexive
2306 address that routes to itself. It then forwards any packets it
2307 receives to the client. This attack would allow the attacker to
2308 observe all packets sent to the client. However, in order to launch
2309 the attack, the attacker must have already been able to observe
2310 packets from the client to the STUN server. In most cases (such as
2311 when the attack is launched from an access network), this means that
2312 the attacker could already observe packets sent to the client. This
2313 attack is, as a result, only useful for observing traffic by
2314 attackers on the path from the client to the STUN server, but not
2315 generally on the path of packets being routed towards the client.
2316
2317 <span class="h3"><a class="selflink" id="section-16.3" href="#section-16.3">16.3</a>. Hash Agility Plan</span>
2318
2319 This specification uses HMAC-SHA-1 for computation of the message
2320 integrity. If, at a later time, HMAC-SHA-1 is found to be
2321 compromised, the following is the remedy that will be applied.
2322
2323 We will define a STUN extension that introduces a new message-
2324 integrity attribute, computed using a new hash. Clients would be
2325 required to include both the new and old message-integrity attributes
2326 in their requests or indications. A new server will utilize the new
2327 message-integrity attribute, and an old one, the old. After a
2328 transition period where mixed implementations are in deployment, the
2329 old message-integrity attribute will be deprecated by another
2330 specification, and clients will cease including it in requests.
2331
2332 It is also important to note that the HMAC is done using a key that
2333 is itself computed using an MD5 of the user's password. The choice
2334 of the MD5 hash was made because of the existence of legacy databases
2335 that store passwords in that form. If future work finds that an HMAC
2336 of an MD5 input is not secure, and a different hash is needed, it can
2337 also be changed using this plan. However, this would require
2338 administrators to repopulate their databases.
2339
2340 <span class="h2"><a class="selflink" id="section-17" href="#section-17">17</a>. IAB Considerations</span>
2341
2342 The IAB has studied the problem of Unilateral Self-Address Fixing
2343 (UNSAF), which is the general process by which a client attempts to
2344 determine its address in another realm on the other side of a NAT
2345
2346
2347
2348 <span class="grey">Rosenberg, et al. Standards Track [Page 42]</span></pre>
2349 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-43" ></span>
2350 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2351
2352
2353 through a collaborative protocol reflection mechanism (<a href="./rfc3424">RFC3424</a>
2354 [<a href="./rfc3424" title="&quot;IAB Considerations for UNilateral Self-Address Fixing (UNSAF) Across Network Address Translation&quot;">RFC3424</a>]). STUN can be used to perform this function using a
2355 Binding request/response transaction if one agent is behind a NAT and
2356 the other is on the public side of the NAT.
2357
2358 The IAB has mandated that protocols developed for this purpose
2359 document a specific set of considerations. Because some STUN usages
2360 provide UNSAF functions (such as ICE [<a href="#ref-MMUSIC-ICE" title="&quot;Interactive Connectivity Establishment (ICE): A Protocol for Network Address Translator (NAT) Traversal for Offer/Answer Protocols&quot;">MMUSIC-ICE</a>] ), and others do
2361 not (such as SIP Outbound [<a href="#ref-SIP-OUTBOUND" title="&quot;Managing Client Initiated Connections in the Session Initiation Protocol (SIP)&quot;">SIP-OUTBOUND</a>]), answers to these
2362 considerations need to be addressed by the usages themselves.
2363
2364 <span class="h2"><a class="selflink" id="section-18" href="#section-18">18</a>. IANA Considerations</span>
2365
2366 IANA has created three new registries: a "STUN Methods Registry", a
2367 "STUN Attributes Registry", and a "STUN Error Codes Registry". IANA
2368 has also changed the name of the assigned IANA port for STUN from
2369 "nat-stun-port" to "stun".
2370
2371 <span class="h3"><a class="selflink" id="section-18.1" href="#section-18.1">18.1</a>. STUN Methods Registry</span>
2372
2373 A STUN method is a hex number in the range 0x000 - 0xFFF. The
2374 encoding of STUN method into a STUN message is described in
2375 <a href="#section-6">Section 6</a>.
2376
2377 The initial STUN methods are:
2378
2379 0x000: (Reserved)
2380 0x001: Binding
2381 0x002: (Reserved; was SharedSecret)
2382
2383 STUN methods in the range 0x000 - 0x7FF are assigned by IETF Review
2384 [<a href="./rfc5226" title="">RFC5226</a>]. STUN methods in the range 0x800 - 0xFFF are assigned by
2385 Designated Expert [<a href="./rfc5226" title="">RFC5226</a>]. The responsibility of the expert is to
2386 verify that the selected codepoint(s) are not in use and that the
2387 request is not for an abnormally large number of codepoints.
2388 Technical review of the extension itself is outside the scope of the
2389 designated expert responsibility.
2390
2391 <span class="h3"><a class="selflink" id="section-18.2" href="#section-18.2">18.2</a>. STUN Attribute Registry</span>
2392
2393 A STUN Attribute type is a hex number in the range 0x0000 - 0xFFFF.
2394 STUN attribute types in the range 0x0000 - 0x7FFF are considered
2395 comprehension-required; STUN attribute types in the range 0x8000 -
2396 0xFFFF are considered comprehension-optional. A STUN agent handles
2397 unknown comprehension-required and comprehension-optional attributes
2398 differently.
2399
2400 The initial STUN Attributes types are:
2401
2402
2403
2404 <span class="grey">Rosenberg, et al. Standards Track [Page 43]</span></pre>
2405 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-44" ></span>
2406 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2407
2408
2409 Comprehension-required range (0x0000-0x7FFF):
2410 0x0000: (Reserved)
2411 0x0001: MAPPED-ADDRESS
2412 0x0002: (Reserved; was RESPONSE-ADDRESS)
2413 0x0003: (Reserved; was CHANGE-ADDRESS)
2414 0x0004: (Reserved; was SOURCE-ADDRESS)
2415 0x0005: (Reserved; was CHANGED-ADDRESS)
2416 0x0006: USERNAME
2417 0x0007: (Reserved; was PASSWORD)
2418 0x0008: MESSAGE-INTEGRITY
2419 0x0009: ERROR-CODE
2420 0x000A: UNKNOWN-ATTRIBUTES
2421 0x000B: (Reserved; was REFLECTED-FROM)
2422 0x0014: REALM
2423 0x0015: NONCE
2424 0x0020: XOR-MAPPED-ADDRESS
2425
2426 Comprehension-optional range (0x8000-0xFFFF)
2427 0x8022: SOFTWARE
2428 0x8023: ALTERNATE-SERVER
2429 0x8028: FINGERPRINT
2430
2431 STUN Attribute types in the first half of the comprehension-required
2432 range (0x0000 - 0x3FFF) and in the first half of the comprehension-
2433 optional range (0x8000 - 0xBFFF) are assigned by IETF Review
2434 [<a href="./rfc5226" title="">RFC5226</a>]. STUN Attribute types in the second half of the
2435 comprehension-required range (0x4000 - 0x7FFF) and in the second half
2436 of the comprehension-optional range (0xC000 - 0xFFFF) are assigned by
2437 Designated Expert [<a href="./rfc5226" title="">RFC5226</a>]. The responsibility of the expert is to
2438 verify that the selected codepoint(s) are not in use, and that the
2439 request is not for an abnormally large number of codepoints.
2440 Technical review of the extension itself is outside the scope of the
2441 designated expert responsibility.
2442
2443 <span class="h3"><a class="selflink" id="section-18.3" href="#section-18.3">18.3</a>. STUN Error Code Registry</span>
2444
2445 A STUN error code is a number in the range 0 - 699. STUN error codes
2446 are accompanied by a textual reason phrase in UTF-8 [<a href="./rfc3629" title="&quot;UTF-8, a transformation format of ISO 10646&quot;">RFC3629</a>] that is
2447 intended only for human consumption and can be anything appropriate;
2448 this document proposes only suggested values.
2449
2450 STUN error codes are consistent in codepoint assignments and
2451 semantics with SIP [<a href="./rfc3261" title="&quot;SIP: Session Initiation Protocol&quot;">RFC3261</a>] and HTTP [<a href="./rfc2616" title="&quot;Hypertext Transfer Protocol -- HTTP/1.1&quot;">RFC2616</a>].
2452
2453 The initial values in this registry are given in <a href="#section-15.6">Section 15.6</a>.
2454
2455
2456
2457
2458
2459
2460 <span class="grey">Rosenberg, et al. Standards Track [Page 44]</span></pre>
2461 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-45" ></span>
2462 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2463
2464
2465 New STUN error codes are assigned based on IETF Review [<a href="./rfc5226" title="">RFC5226</a>].
2466 The specification must carefully consider how clients that do not
2467 understand this error code will process it before granting the
2468 request. See the rules in <a href="#section-7.3.4">Section 7.3.4</a>.
2469
2470 <span class="h3"><a class="selflink" id="section-18.4" href="#section-18.4">18.4</a>. STUN UDP and TCP Port Numbers</span>
2471
2472 IANA has previously assigned port 3478 for STUN. This port appears
2473 in the IANA registry under the moniker "nat-stun-port". In order to
2474 align the DNS SRV procedures with the registered protocol service,
2475 IANA is requested to change the name of protocol assigned to port
2476 3478 from "nat-stun-port" to "stun", and the textual name from
2477 "Simple Traversal of UDP Through NAT (STUN)" to "Session Traversal
2478 Utilities for NAT", so that the IANA port registry would read:
2479
2480 stun 3478/tcp Session Traversal Utilities for NAT (STUN) port
2481 stun 3478/udp Session Traversal Utilities for NAT (STUN) port
2482
2483 In addition, IANA has assigned port number 5349 for the "stuns"
2484 service, defined over TCP and UDP. The UDP port is not currently
2485 defined; however, it is reserved for future use.
2486
2487 <span class="h2"><a class="selflink" id="section-19" href="#section-19">19</a>. Changes since <a href="./rfc3489">RFC 3489</a></span>
2488
2489 This specification obsoletes <a href="./rfc3489">RFC 3489</a> [<a href="./rfc3489" title="&quot;STUN - Simple Traversal of User Datagram Protocol (UDP) Through Network Address Translators (NATs)&quot;">RFC3489</a>]. This specification
2490 differs from <a href="./rfc3489">RFC 3489</a> in the following ways:
2491
2492 o Removed the notion that STUN is a complete NAT traversal solution.
2493 STUN is now a tool that can be used to produce a NAT traversal
2494 solution. As a consequence, changed the name of the protocol to
2495 Session Traversal Utilities for NAT.
2496
2497 o Introduced the concept of STUN usages, and described what a usage
2498 of STUN must document.
2499
2500 o Removed the usage of STUN for NAT type detection and binding
2501 lifetime discovery. These techniques have proven overly brittle
2502 due to wider variations in the types of NAT devices than described
2503 in this document. Removed the RESPONSE-ADDRESS, CHANGED-ADDRESS,
2504 CHANGE-REQUEST, SOURCE-ADDRESS, and REFLECTED-FROM attributes.
2505
2506 o Added a fixed 32-bit magic cookie and reduced length of
2507 transaction ID by 32 bits. The magic cookie begins at the same
2508 offset as the original transaction ID.
2509
2510
2511
2512
2513
2514
2515
2516 <span class="grey">Rosenberg, et al. Standards Track [Page 45]</span></pre>
2517 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-46" ></span>
2518 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2519
2520
2521 o Added the XOR-MAPPED-ADDRESS attribute, which is included in
2522 Binding responses if the magic cookie is present in the request.
2523 Otherwise, the <a href="./rfc3489">RFC 3489</a> behavior is retained (that is, Binding
2524 response includes MAPPED-ADDRESS). See discussion in XOR-MAPPED-
2525 ADDRESS regarding this change.
2526
2527 o Introduced formal structure into the message type header field,
2528 with an explicit pair of bits for indication of request, response,
2529 error response, or indication. Consequently, the message type
2530 field is split into the class (one of the previous four) and
2531 method.
2532
2533 o Explicitly point out that the most significant 2 bits of STUN are
2534 0b00, allowing easy differentiation with RTP packets when used
2535 with ICE.
2536
2537 o Added the FINGERPRINT attribute to provide a method of definitely
2538 detecting the difference between STUN and another protocol when
2539 the two protocols are multiplexed together.
2540
2541 o Added support for IPv6. Made it clear that an IPv4 client could
2542 get a v6 mapped address, and vice versa.
2543
2544 o Added long-term-credential-based authentication.
2545
2546 o Added the SOFTWARE, REALM, NONCE, and ALTERNATE-SERVER attributes.
2547
2548 o Removed the SharedSecret method, and thus the PASSWORD attribute.
2549 This method was almost never implemented and is not needed with
2550 current usages.
2551
2552 o Removed recommendation to continue listening for STUN responses
2553 for 10 seconds in an attempt to recognize an attack.
2554
2555 o Changed transaction timers to be more TCP friendly.
2556
2557 o Removed the STUN example that centered around the separation of
2558 the control and media planes. Instead, provided more information
2559 on using STUN with protocols.
2560
2561 o Defined a generic padding mechanism that changes the
2562 interpretation of the length attribute. This would, in theory,
2563 break backwards compatibility. However, the mechanism in <a href="./rfc3489">RFC 3489</a>
2564 never worked for the few attributes that weren't aligned naturally
2565 on 32-bit boundaries.
2566
2567 o REALM, SERVER, reason phrases, and NONCE limited to 127
2568 characters. USERNAME to 513 bytes.
2569
2570
2571
2572 <span class="grey">Rosenberg, et al. Standards Track [Page 46]</span></pre>
2573 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-47" ></span>
2574 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2575
2576
2577 o Changed the DNS SRV procedures for TCP and TLS. UDP remains the
2578 same as before.
2579
2580 <span class="h2"><a class="selflink" id="section-20" href="#section-20">20</a>. Contributors</span>
2581
2582 Christian Huitema and Joel Weinberger were original co-authors of <a href="./rfc3489">RFC</a>
2583 <a href="./rfc3489">3489</a>.
2584
2585 <span class="h2"><a class="selflink" id="section-21" href="#section-21">21</a>. Acknowledgements</span>
2586
2587 The authors would like to thank Cedric Aoun, Pete Cordell, Cullen
2588 Jennings, Bob Penfield, Xavier Marjou, Magnus Westerlund, Miguel
2589 Garcia, Bruce Lowekamp, and Chris Sullivan for their comments, and
2590 Baruch Sterman and Alan Hawrylyshen for initial implementations.
2591 Thanks for Leslie Daigle, Allison Mankin, Eric Rescorla, and Henning
2592 Schulzrinne for IESG and IAB input on this work.
2593
2594 <span class="h2"><a class="selflink" id="section-22" href="#section-22">22</a>. References</span>
2595
2596 <span class="h3"><a class="selflink" id="section-22.1" href="#section-22.1">22.1</a>. Normative References</span>
2597
2598 [<a id="ref-ITU.V42.2002">ITU.V42.2002</a>] International Telecommunications Union, "Error-
2599 correcting Procedures for DCEs Using Asynchronous-
2600 to-Synchronous Conversion", ITU-T Recommendation
2601 V.42, March 2002.
2602
2603 [<a id="ref-RFC0791">RFC0791</a>] Postel, J., "Internet Protocol", STD 5, <a href="./rfc791">RFC 791</a>,
2604 September 1981.
2605
2606 [<a id="ref-RFC1122">RFC1122</a>] Braden, R., "Requirements for Internet Hosts -
2607 Communication Layers", STD 3, <a href="./rfc1122">RFC 1122</a>,
2608 October 1989.
2609
2610 [<a id="ref-RFC1321">RFC1321</a>] Rivest, R., "The MD5 Message-Digest Algorithm",
2611 <a href="./rfc1321">RFC 1321</a>, April 1992.
2612
2613 [<a id="ref-RFC2104">RFC2104</a>] Krawczyk, H., Bellare, M., and R. Canetti, "HMAC:
2614 Keyed-Hashing for Message Authentication",
2615 <a href="./rfc2104">RFC 2104</a>, February 1997.
2616
2617 [<a id="ref-RFC2119">RFC2119</a>] Bradner, S., "Key words for use in RFCs to Indicate
2618 Requirement Levels", <a href="https://www.rfc-editor.org/bcp/bcp14">BCP 14</a>, <a href="./rfc2119">RFC 2119</a>, March 1997.
2619
2620 [<a id="ref-RFC2460">RFC2460</a>] Deering, S. and R. Hinden, "Internet Protocol,
2621 Version 6 (IPv6) Specification", <a href="./rfc2460">RFC 2460</a>,
2622 December 1998.
2623
2624
2625
2626
2627
2628 <span class="grey">Rosenberg, et al. Standards Track [Page 47]</span></pre>
2629 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-48" ></span>
2630 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2631
2632
2633 [<a id="ref-RFC2617">RFC2617</a>] Franks, J., Hallam-Baker, P., Hostetler, J.,
2634 Lawrence, S., Leach, P., Luotonen, A., and L.
2635 Stewart, "HTTP Authentication: Basic and Digest
2636 Access Authentication", <a href="./rfc2617">RFC 2617</a>, June 1999.
2637
2638 [<a id="ref-RFC2782">RFC2782</a>] Gulbrandsen, A., Vixie, P., and L. Esibov, "A DNS
2639 RR for specifying the location of services (DNS
2640 SRV)", <a href="./rfc2782">RFC 2782</a>, February 2000.
2641
2642 [<a id="ref-RFC2818">RFC2818</a>] Rescorla, E., "HTTP Over TLS", <a href="./rfc2818">RFC 2818</a>, May 2000.
2643
2644 [<a id="ref-RFC2988">RFC2988</a>] Paxson, V. and M. Allman, "Computing TCP's
2645 Retransmission Timer", <a href="./rfc2988">RFC 2988</a>, November 2000.
2646
2647 [<a id="ref-RFC3629">RFC3629</a>] Yergeau, F., "UTF-8, a transformation format of ISO
2648 10646", STD 63, <a href="./rfc3629">RFC 3629</a>, November 2003.
2649
2650 [<a id="ref-RFC4013">RFC4013</a>] Zeilenga, K., "SASLprep: Stringprep Profile for
2651 User Names and Passwords", <a href="./rfc4013">RFC 4013</a>, February 2005.
2652
2653 <span class="h3"><a class="selflink" id="section-22.2" href="#section-22.2">22.2</a>. Informative References</span>
2654
2655 [<a id="ref-BEHAVE-NAT">BEHAVE-NAT</a>] MacDonald, D. and B. Lowekamp, "NAT Behavior
2656 Discovery Using STUN", Work in Progress, July 2008.
2657
2658 [<a id="ref-BEHAVE-TURN">BEHAVE-TURN</a>] Rosenberg, J., Mahy, R., and P. Matthews,
2659 "Traversal Using Relays around NAT (TURN): Relay
2660 Extensions to Session Traversal Utilities for NAT
2661 (STUN)", Work in Progress, July 2008.
2662
2663 [<a id="ref-KARN87">KARN87</a>] Karn, P. and C. Partridge, "Improving Round-Trip
2664 Time Estimates in Reliable Transport Protocols",
2665 SIGCOMM 1987, August 1987.
2666
2667 [<a id="ref-MMUSIC-ICE">MMUSIC-ICE</a>] Rosenberg, J., "Interactive Connectivity
2668 Establishment (ICE): A Protocol for Network Address
2669 Translator (NAT) Traversal for Offer/Answer
2670 Protocols", Work in Progress, October 2007.
2671
2672 [<a id="ref-MMUSIC-ICE-TCP">MMUSIC-ICE-TCP</a>] Rosenberg, J., "TCP Candidates with Interactive
2673 Connectivity Establishment (ICE)", Work
2674 in Progress, July 2008.
2675
2676 [<a id="ref-RFC2616">RFC2616</a>] Fielding, R., Gettys, J., Mogul, J., Frystyk, H.,
2677 Masinter, L., Leach, P., and T. Berners-Lee,
2678 "Hypertext Transfer Protocol -- HTTP/1.1",
2679 <a href="./rfc2616">RFC 2616</a>, June 1999.
2680
2681
2682
2683
2684 <span class="grey">Rosenberg, et al. Standards Track [Page 48]</span></pre>
2685 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-49" ></span>
2686 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2687
2688
2689 [<a id="ref-RFC3261">RFC3261</a>] Rosenberg, J., Schulzrinne, H., Camarillo, G.,
2690 Johnston, A., Peterson, J., Sparks, R., Handley,
2691 M., and E. Schooler, "SIP: Session Initiation
2692 Protocol", <a href="./rfc3261">RFC 3261</a>, June 2002.
2693
2694 [<a id="ref-RFC3264">RFC3264</a>] Rosenberg, J. and H. Schulzrinne, "An Offer/Answer
2695 Model with Session Description Protocol (SDP)",
2696 <a href="./rfc3264">RFC 3264</a>, June 2002.
2697
2698 [<a id="ref-RFC3424">RFC3424</a>] Daigle, L. and IAB, "IAB Considerations for
2699 UNilateral Self-Address Fixing (UNSAF) Across
2700 Network Address Translation", <a href="./rfc3424">RFC 3424</a>,
2701 November 2002.
2702
2703 [<a id="ref-RFC3489">RFC3489</a>] Rosenberg, J., Weinberger, J., Huitema, C., and R.
2704 Mahy, "STUN - Simple Traversal of User Datagram
2705 Protocol (UDP) Through Network Address Translators
2706 (NATs)", <a href="./rfc3489">RFC 3489</a>, March 2003.
2707
2708 [<a id="ref-RFC4107">RFC4107</a>] Bellovin, S. and R. Housley, "Guidelines for
2709 Cryptographic Key Management", <a href="https://www.rfc-editor.org/bcp/bcp107">BCP 107</a>, <a href="./rfc4107">RFC 4107</a>,
2710 June 2005.
2711
2712 [<a id="ref-RFC5226">RFC5226</a>] Narten, T. and H. Alvestrand, "Guidelines for
2713 Writing an IANA Considerations Section in RFCs",
2714 <a href="https://www.rfc-editor.org/bcp/bcp26">BCP 26</a>, <a href="./rfc5226">RFC 5226</a>, May 2008.
2715
2716 [<a id="ref-SIP-OUTBOUND">SIP-OUTBOUND</a>] Jennings, C. and R. Mahy, "Managing Client
2717 Initiated Connections in the Session Initiation
2718 Protocol (SIP)", Work in Progress, June 2008.
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740 <span class="grey">Rosenberg, et al. Standards Track [Page 49]</span></pre>
2741 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-50" ></span>
2742 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2743
2744
2745 <span class="h2"><a class="selflink" id="appendix-A" href="#appendix-A">Appendix A</a>. C Snippet to Determine STUN Message Types</span>
2746
2747 Given a 16-bit STUN message type value in host byte order in msg_type
2748 parameter, below are C macros to determine the STUN message types:
2749
2750 #define IS_REQUEST(msg_type) (((msg_type) &amp; 0x0110) == 0x0000)
2751 #define IS_INDICATION(msg_type) (((msg_type) &amp; 0x0110) == 0x0010)
2752 #define IS_SUCCESS_RESP(msg_type) (((msg_type) &amp; 0x0110) == 0x0100)
2753 #define IS_ERR_RESP(msg_type) (((msg_type) &amp; 0x0110) == 0x0110)
2754
2755
2756 Authors' Addresses
2757
2758 Jonathan Rosenberg
2759 Cisco
2760 Edison, NJ
2761 US
2762
2763 EMail: [email protected]
2764 URI: <a href="http://www.jdrosen.net">http://www.jdrosen.net</a>
2765
2766
2767 Rohan Mahy
2768 Unaffiliated
2769
2770 EMail: [email protected]
2771
2772
2773 Philip Matthews
2774 Unaffiliated
2775
2776 EMail: [email protected]
2777
2778
2779 Dan Wing
2780 Cisco
2781 771 Alder Drive
2782 San Jose, CA 95035
2783 US
2784
2785 EMail: [email protected]
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796 <span class="grey">Rosenberg, et al. Standards Track [Page 50]</span></pre>
2797 <hr class='noprint'/><!--NewPage--><pre class='newpage'><span id="page-51" ></span>
2798 <span class="grey"><a href="./rfc5389">RFC 5389</a> STUN October 2008</span>
2799
2800
2801 Full Copyright Statement
2802
2803 Copyright (C) The IETF Trust (2008).
2804
2805 This document is subject to the rights, licenses and restrictions
2806 contained in <a href="https://www.rfc-editor.org/bcp/bcp78">BCP 78</a>, and except as set forth therein, the authors
2807 retain all their rights.
2808
2809 This document and the information contained herein are provided on an
2810 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
2811 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
2812 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS
2813 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
2814 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
2815 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
2816
2817 Intellectual Property
2818
2819 The IETF takes no position regarding the validity or scope of any
2820 Intellectual Property Rights or other rights that might be claimed to
2821 pertain to the implementation or use of the technology described in
2822 this document or the extent to which any license under such rights
2823 might or might not be available; nor does it represent that it has
2824 made any independent effort to identify any such rights. Information
2825 on the procedures with respect to rights in RFC documents can be
2826 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>.
2827
2828 Copies of IPR disclosures made to the IETF Secretariat and any
2829 assurances of licenses to be made available, or the result of an
2830 attempt made to obtain a general license or permission for the use of
2831 such proprietary rights by implementers or users of this
2832 specification can be obtained from the IETF on-line IPR repository at
2833 <a href="http://www.ietf.org/ipr">http://www.ietf.org/ipr</a>.
2834
2835 The IETF invites any interested party to bring to its attention any
2836 copyrights, patents or patent applications, or other proprietary
2837 rights that may cover technology that may be required to implement
2838 this standard. Please address the information to the IETF at
2839 [email protected].
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852 Rosenberg, et al. Standards Track [Page 51]
2853 </pre>