idnits 2.17.00 (12 Aug 2021) /tmp/idnits60333/draft-tschofenig-mip6-ice-01.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 16. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 1071. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 1082. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 1089. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 1095. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) == There are 1 instance of lines with private range IPv4 addresses in the document. If these are generic example addresses, they should be changed to use any of the ranges defined in RFC 6890 (or successor): 192.0.2.x, 198.51.100.x or 203.0.113.x. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust Copyright Line does not match the current year -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (July 9, 2007) is 5430 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC3775' is defined on line 973, but no explicit reference was found in the text == Unused Reference: 'RFC4474' is defined on line 1010, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-monami6-multiplecoa' is defined on line 1014, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-mip4-dsmipv4' is defined on line 1024, but no explicit reference was found in the text == Outdated reference: draft-ietf-mmusic-ice has been published as RFC 5245 ** Obsolete normative reference: RFC 3775 (Obsoleted by RFC 6275) == Outdated reference: draft-ietf-behave-turn has been published as RFC 5766 == Outdated reference: draft-ietf-shim6-failure-detection has been published as RFC 5534 -- Obsolete informational reference (is this intentional?): RFC 3489 (Obsoleted by RFC 5389) == Outdated reference: draft-ietf-behave-rfc3489bis has been published as RFC 5389 == Outdated reference: A later version (-05) exists of draft-wing-behave-nat-control-stun-usage-02 -- Obsolete informational reference (is this intentional?): RFC 4474 (Obsoleted by RFC 8224) == Outdated reference: draft-ietf-monami6-multiplecoa has been published as RFC 5648 == Outdated reference: A later version (-06) exists of draft-ietf-mip6-nemo-v4traversal-04 == Outdated reference: draft-ietf-mip4-dsmipv4 has been published as RFC 5454 == Outdated reference: A later version (-03) exists of draft-bajko-mip6-rrtfw-01 Summary: 3 errors (**), 0 flaws (~~), 15 warnings (==), 10 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 MIP6 H. Tschofenig 3 Internet-Draft Nokia Siemens Networks 4 Intended status: Standards Track G. Bajko 5 Expires: January 10, 2008 Nokia 6 July 9, 2007 8 Mobile IP Interactive Connectivity Establishment (M-ICE) 9 draft-tschofenig-mip6-ice-01.txt 11 Status of this Memo 13 By submitting this Internet-Draft, each author represents that any 14 applicable patent or other IPR claims of which he or she is aware 15 have been or will be disclosed, and any of which he or she becomes 16 aware will be disclosed, in accordance with Section 6 of BCP 79. 18 Internet-Drafts are working documents of the Internet Engineering 19 Task Force (IETF), its areas, and its working groups. Note that 20 other groups may also distribute working documents as Internet- 21 Drafts. 23 Internet-Drafts are draft documents valid for a maximum of six months 24 and may be updated, replaced, or obsoleted by other documents at any 25 time. It is inappropriate to use Internet-Drafts as reference 26 material or to cite them other than as "work in progress." 28 The list of current Internet-Drafts can be accessed at 29 http://www.ietf.org/ietf/1id-abstracts.txt. 31 The list of Internet-Draft Shadow Directories can be accessed at 32 http://www.ietf.org/shadow.html. 34 This Internet-Draft will expire on January 10, 2008. 36 Copyright Notice 38 Copyright (C) The IETF Trust (2007). 40 Abstract 42 This document describes how the Interactive Connectivity 43 Establishment (ICE) methodology can be used for Mobile IP to 44 determine whether end-to-end communication is possible. ICE makes 45 use of the Session Traversal Utilities for NAT (STUN) protocol in 46 addition to mechanisms for checking connectivity between peers. 47 After running the ICE the two MIP end points will be able to 48 communicate directly or through a relay via Network Address 49 Translators (NATs), Network Address and Port Translators (NAPTs) and 50 firewalls. 52 This document addresses also the problems raised in RFC 4487 "Mobile 53 IPv6 and Firewalls: Problem Statement". 55 Table of Contents 57 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 58 1.1. Gathering Candidate Addresses . . . . . . . . . . . . . . 5 59 1.2. Connectivity Checks . . . . . . . . . . . . . . . . . . . 5 60 1.3. Sorting Candidates . . . . . . . . . . . . . . . . . . . . 6 61 1.4. Frozen Candidates . . . . . . . . . . . . . . . . . . . . 6 62 1.5. Security for Checks . . . . . . . . . . . . . . . . . . . 6 63 1.6. Concluding M-ICE . . . . . . . . . . . . . . . . . . . . . 6 64 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 6 65 3. Design Choices . . . . . . . . . . . . . . . . . . . . . . . . 7 66 4. Sending the Initial Offer . . . . . . . . . . . . . . . . . . 8 67 5. Receiving the Initial Offer . . . . . . . . . . . . . . . . . 8 68 6. Receipt of the Initial Answer . . . . . . . . . . . . . . . . 9 69 7. Performing Connectivity Checks . . . . . . . . . . . . . . . . 9 70 8. Concluding ICE Processing . . . . . . . . . . . . . . . . . . 9 71 9. Subsequent Offer/Answer Exchanges . . . . . . . . . . . . . . 9 72 10. Keepalives . . . . . . . . . . . . . . . . . . . . . . . . . . 9 73 11. Attribute Encoding . . . . . . . . . . . . . . . . . . . . . . 10 74 12. Demultiplexing MIP and STUN messages . . . . . . . . . . . . . 12 75 13. Example . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 76 14. Security Considerations . . . . . . . . . . . . . . . . . . . 13 77 14.1. Attacks on Connectivity Checks . . . . . . . . . . . . . . 13 78 14.2. Attacks on Address Gathering . . . . . . . . . . . . . . . 16 79 14.3. Attacks on the Offer/Answer Exchanges . . . . . . . . . . 17 80 14.4. Insider Attacks . . . . . . . . . . . . . . . . . . . . . 17 81 14.4.1. MIP Amplification Attack . . . . . . . . . . . . . . 17 82 14.4.2. STUN Amplification Attack . . . . . . . . . . . . . . 18 83 15. IAB Considerations . . . . . . . . . . . . . . . . . . . . . . 18 84 15.1. Problem Definition . . . . . . . . . . . . . . . . . . . . 18 85 15.2. Exit Strategy . . . . . . . . . . . . . . . . . . . . . . 19 86 15.3. Brittleness Introduced by M-ICE . . . . . . . . . . . . . 19 87 15.4. Requirements for a Long Term Solution . . . . . . . . . . 20 88 15.5. Issues with Existing NAPT Boxes . . . . . . . . . . . . . 21 89 16. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 21 90 17. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 21 91 18. References . . . . . . . . . . . . . . . . . . . . . . . . . . 21 92 18.1. Normative References . . . . . . . . . . . . . . . . . . . 21 93 18.2. Informative References . . . . . . . . . . . . . . . . . . 22 94 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 23 95 Intellectual Property and Copyright Statements . . . . . . . . . . 24 97 1. Introduction 99 In a typical Mobile IP deployment, there are two endpoints, mobile 100 node and correspondent nodes, which want to communicate. They are 101 able to communicate indirectly via a combination of Mobile IP 102 signaling and reverse tunneling. A couple of different extensions 103 are available for Mobile IP that allow multiple care-of addresses, 104 IPv4/IPv6 interworking and different routes to be used through the 105 network. 107 Unfortunately, it is likely that some of the available paths do not 108 work due to connectivity problems caused by firewalling behavior. 109 The VoIP community has investigated these problems extensively and 110 developed a protocols and a methodology to systematically perform 111 connectivity checks in order to determine a working path between the 112 two end points. The Interactive Connectivity Establishment (ICE) 113 specification describes how the Session Traversal Utilities for NAT 114 (STUN) protocol can be used to execute these checks. This document 115 suggests to utilize the ICE methodology and if possible STUN for 116 Mobile IP, both Mobile IPv4 and Mobile IPv6. We call this usage 117 Mobile IP - ICE, M-ICE for short. 119 This document, however, concentrates on Mobile IPv6 as a starting 120 point. A future version of this document will also describe the 121 operation using Mobile IPv4. The ideal outcome is that the best 122 available path through the network can be chosen when Mobile IP is 123 used regardless of the MIP version and the environmental problems 124 the two end points are facing. 126 At the beginning of the M-ICE process, the end points are ignorant of 127 their own topologies. They might or might not be behind a NAT (or 128 multiple tiers of NATs) and might be behind firewalls that limit the 129 ability to communicate in different ways between the end points. 130 M-ICE allows these end points to discover enough information about 131 their topologies to potentially find one or more paths by which they 132 can communicate. 134 Figure 1 shows a typical environment for M-ICE deployment. The two 135 end points are labelled L and R (for left and right, which helps 136 visualize call flows). Both L and R are behind their own respective 137 NATs or firewalls though they may not be aware of it. The type of 138 NAT or firewall and their properties are also unknown. L and R are 139 capable of engaging in an end-to-end mobility protocol exchange. 140 This exchange will occur through mobility anchor points, such as Home 141 Agents. 143 In this architecture the ICE functionality of TURN servers is 144 provided by the Home Agent via reverse tunneling. In this document 145 we assume that the STUN server is co-located with the Home Agent 146 since it is convenient from a security and configuration point of 147 view even though it is, from a solution point of view, not necessary. 149 +--------+ Mobility +--------+ 150 | Home | Signalling | Home | 151 | Agent/ |----------------------------| Agent/ | 152 | STUN | | STUN | 153 | Server | | Server | 154 +--------+ +--------+ 155 ^ ^ 156 | | 157 | | 158 Mobility | |Mobility 159 Signalling| |Signalling 160 | | 161 | | 162 +---v----+ +---v----+ 163 | FW/NAT | | FW/NAT | 164 +---^----+ +---^----+ 165 | | 166 | | 167 v v 168 +--------+ +--------+ 169 | Agent | | Agent | 170 | L | | R | 171 +--------+ +--------+ 173 Figure 1: Overview 175 The basic idea behind M-ICE is as follows: each end point has a 176 variety of candidate TRANSPORT ADDRESSES (combination of IP address, 177 transport protocol (UDP), and port) it could use to communicate with 178 the other end point. 180 To avoid unnecessary UDP encapsulation of end-to-end traffic in 181 case there is no need todo so, it is also possible to consider 182 using IP addresses rather than focusing exclusively on TRANSPORT 183 ADDRESSES. For example, two MIP hosts behind the same NAT do not 184 need to use UDP encapsulation. If there is no NAT or firewall 185 between the two communicating nodes then there is again no need to 186 provide support for UDP encapsulation. A future version of this 187 document will provide support for this functionality. 189 Potentially, any of L's candidate transport addresses can be used to 190 communicate with any of R's candidate transport addresses. In 191 practice, however, many combinations do not work. For instance, if L 192 and R are both behind NATs, their directly attached interface 193 addresses (e.g., 192.168.1.100) are unlikely to be able to 194 communicate. The purpose of M-ICE is to discover which pairs of 195 addresses will work. The way that M-ICE does this is to 196 systematically try all possible pairs (in a carefully sorted order) 197 until it finds one or more that works. Once found, the best pair is 198 used for subsequent communication between the hosts. 200 1.1. Gathering Candidate Addresses 202 In order to execute ICE, an agent has to identify all of its address 203 candidates. A CANDIDATE is a transport address - a combination of IP 204 address and port for a particular transport protocol. 206 This document uses three types of candidates: 208 1. One viable candidate is a transport address obtained directly 209 from a local interface. Such a candidate is called a HOST 210 CANDIDATE. 211 2. Translated addresses on the public side of a NAT (called SERVER 212 REFLEXIVE CANDIDATES). This address is obtained via STUN. 213 3. Addresses obtained via relaying traffic through a Home Agent, 214 called RELAYED CANDIDATES. 216 1.2. Connectivity Checks 218 Once L has gathered all of its candidates, it orders them in highest 219 to lowest priority and sends them to R over the signalling channel. 220 We refer to the signaling channel to the end-to-end MIP exchange. 221 The extension to exchange candidates can be found in Section 11. 223 When R receives the L's MIP message, R performs the same candidate 224 gathering process and responds with its own list of candidates. At 225 the end of this process, each agent has a complete list of both its 226 candidates and its peer's candidates. It pairs them up, resulting in 227 CANDIDATE PAIRS. To see which pairs work, each agent schedules a 228 series of connectivity CHECKS. Each check is a STUN transaction that 229 the client will perform on a particular candidate pair by sending a 230 STUN request from the local candidate to the remote candidate; a 231 response indicates there is connectivity to the peer using that 232 candidate address. 234 It is important to note that the STUN requests are sent to and from 235 the exact same IP addresses and ports that will be used for 236 subsequent data traffic. 238 1.3. Sorting Candidates 240 Because the algorithm above searches all candidate pairs, if a 241 working pair exists it will eventually find it no matter what order 242 the candidates are tried in. In order to produce faster (and better) 243 results, the candidates are sorted in a specified order. The 244 resulting list of sorted candidate pairs is called the CHECK LIST. 246 1.4. Frozen Candidates 248 The concept of frozen candidates is not applied when ICE is applied 249 to MIP. [Editor's Note: More investigations are needed to evaluate 250 whether this is indeed true and the concept of frozen candidates can 251 be ignored.] 253 1.5. Security for Checks 255 Because the ICE algorithm is used to discover which addresses can be 256 used to send traffic between two end points, it is important to 257 ensure that the process cannot be hijacked to send traffic to the 258 wrong location. Each STUN connectivity check is covered by a message 259 authentication code (MAC). There are two ways to generate the keying 260 material for this MAC. Either keying material is derived from the 261 keying material generated by the return routability procedure or new 262 keying material is distributed separately as excercised in ICE. 263 This document currently uses the latter technique without a strong 264 preference. 265 In any case, this MAC provides message integrity and data origin 266 authentication, thus stopping an attacker from forging or modifying 267 connectivity check messages. 269 1.6. Concluding M-ICE 271 ICE checks are performed in a specific sequence, so that high 272 priority candidate pairs are checked first, followed by lower 273 priority ones. 275 2. Terminology 277 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 278 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 279 document are to be interpreted as described in RFC 2119 [RFC2119]. 281 This document heavily relies on the terminology introduced in 282 [I-D.ietf-mmusic-ice]. 284 3. Design Choices 286 The work in this document is guided by the following design choices, 287 namely: 289 o The offer/answer exchange described in ICE [I-D.ietf-mmusic-ice] 290 is mapped to the end-to-end MIP signaling exchange. For end-to- 291 end communication this document assumes that MIPv6 signaling are 292 allowed to be exchanged between the mobile node and the 293 correspondent node. When Network Address Translators and 294 firewalls are located along the path then direct end-to-end 295 communication between the two end points is typically not possible 296 and hence this protocol interaction is provided via MIP Home 297 Agents. The functionality described in [I-D.bajko-mip6-rrtfw] is 298 used. 299 o We assume that MIP initiators and MIP responders implement and use 300 STUN. For performing connectivity checks a couple of other 301 alternatives are, however, possible: 303 It would be possible to utilize the SHIM6 REAchability Protocol 304 (REAP) [I-D.ietf-shim6-failure-detection] but STUN provides the 305 same support with a more likely chance for widespread 306 deployment. REAP currently only provides IPv6 support. It it 307 obviously possible to turn a protocol in any other one. 308 Custom MIP messages could be created. 310 o If one peer does not support STUN then the optimal results of 311 M-ICE cannot be provided. There is, however, the ability to make 312 use of STUN LITE when a host is on the public address space and 313 known not to be behind a firewall. 314 o Obtaining Relay Addresses from STUN [I-D.ietf-behave-turn], 315 formerly known as TURN, is intentionally not used in this 316 document. For MIP, the Home Agent tunneling functionality is used 317 instead of TURN. 318 o This document makes use of the UDP-encapsulated of MIP packets, as 319 specified in [I-D.ietf-mip6-nemo-v4traversal]. 320 o This document focuses only on the data exchange between the two 321 end points rather than on the communication between a mobile node 322 and the Home Agent or on the ability to allow MIP signaling 323 messages to traverse NATs and firewalls. 324 o Each STUN connectivity check is covered by a message 325 authentication code (MAC) generated based on keying material 326 derived from information carried in MIP messages, see Section 11. 327 Alternatively, keying material could be derived from the return 328 routability test procedure. 330 Note that the ICE description assumes usage within a VoIP environment 331 where individual flows are controlled. However, the protocol 332 interaction described in this document operates at a lower layer 333 where application specific message flows are not visible. When a 334 CANDIDATE PAIR, consisting of two TRANSPORT ADDRESSES, is created 335 then it will typically refer to multiple flows then traffic between 336 two end points experiences UDP encapsulation (due to the need to 337 traverse a NAT or a stateful packet filtering firewall). 339 The descriptions in the ICE specification related to SIP, ANAT, RTP, 340 RTCP, third party call control, preconditions, forking, etc. are not 341 applicable to MIP and are not included in this document. 343 From an editorial point of view it would be possible to copy-and- 344 paste relevant parts of the ICE specification and to remove VoIP 345 specific descriptions but for this version of the document we did 346 not follow this approach. 348 The main accomplishment of this document is the reuse of the well- 349 established ICE specification that builds on STUN. STUN enjoys 350 widespread implementation support and maximum code re-use was one of 351 the design criteria for this document. 353 4. Sending the Initial Offer 355 In order to send the initial offer in an offer/answer exchange, an 356 agent must (1) gather candidates, (2) prioritize them, (3) choose 357 default candidates, and then (4) formulate and send them to the other 358 peer. 360 Section 4 of ICE [I-D.ietf-mmusic-ice] is applicable to this document 361 with the following two exceptions: First, TURN is not used in this 362 document but instead similar functionality is accomplished via a Home 363 Agent. Second, the description regarding encoding of candidates in 364 SDP is not applicable and replaced by a MIP specific encoding 365 described in Section 11. 367 5. Receiving the Initial Offer 369 When an agent receives an initial offer, it will check if the offerer 370 supports sufficient ICE functionality to proceed (i.e., if both 371 offerer and answerer are lite implementations, ICE cannot proceed), 372 determine its own role, gather candidates, prioritize them, choose 373 default candidates, encode and send an answer, and for full 374 implementations, form the check lists and begin connectivity checks. 376 Again, the description regarding encoding of candidates in SDP is not 377 applicable to this document and is replaced by a MIP specific 378 encoding described in Section 11. Note that only the encoding is 379 different but not the semantic. As such, the description in Section 380 5 of [I-D.ietf-mmusic-ice] is applicable to this document. 382 6. Receipt of the Initial Answer 384 Section 6 of ICE [I-D.ietf-mmusic-ice] describes the procedures that 385 an agent follows when it receives the answer from the peer. It 386 verifies that its peer supports ICE, determines its role, and for 387 full implementations, forms the check list and begins performing 388 periodic checks. 390 7. Performing Connectivity Checks 392 Section 7 of ICE [I-D.ietf-mmusic-ice] describes how connectivity 393 checks are performed using STUN [I-D.ietf-behave-rfc3489bis] and the 394 content of that section is fully applicable to this document. 396 8. Concluding ICE Processing 398 The description in Section 8 of ICE [I-D.ietf-mmusic-ice] illustrates 399 processing rules that apply only to full implementations. Concluding 400 ICE involves nominating pairs by the controlling agent and updating 401 of state machinery 403 9. Subsequent Offer/Answer Exchanges 405 Either agent may generate a subsequent offer at any time. The rules 406 in Section 9 of ICE [I-D.ietf-mmusic-ice] will cause the controlling 407 agent to send an updated offer at the conclusion of ICE processing 408 when ICE has selected different candidate pairs from the default 409 pairs. Section 9 of ICE [I-D.ietf-mmusic-ice] defines rules for 410 construction of subsequent offers and answers. 412 Note that the term "media stream" in Section 9 of ICE 413 [I-D.ietf-mmusic-ice] translates to an individual UDP-encapsulated 414 data flow exchanged between the two MIP end points. 416 10. Keepalives 418 Section 10 of ICE [I-D.ietf-mmusic-ice] describes a keepalive 419 mechanism. The RTP description, such as RTP No-Op and RTP comfort 420 noise, is not applicable to this document. Other useful keepalive 421 techniques are described in [I-D.marjou-behave-app-rtp-keepalive] and 422 may be useful for MIP; a recommendation will be made in a subsequent 423 version of this document. 425 11. Attribute Encoding 427 To accomplish the same functionality this specification needs to 428 reuse the semantic, but not necessarily the encoding, of seven 429 attributes defined in the ICE specification [I-D.ietf-mmusic-ice], 430 namely "candidate", "remote-candidates", "ice-lite", "ice-mismatch", 431 "ice-ufrag", "ice-pwd" and "ice-options". 433 Section 15.1 to Section 15.5 of ICE [I-D.ietf-mmusic-ice] describe 434 the semantic of the attributes. 436 MIP-ICE Mobility Options Format: 438 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 439 | Type | Header Len |# of candidates| 440 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 441 | Checksum |L|M| Reserved | 442 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 443 | | 444 + + 445 | Ice-pwd | 446 + + 447 | | 448 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 449 | ice-ufrag | ice-options (var) ... 450 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 451 | 452 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 453 . . 454 . Candidate 1 . 455 . . 456 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 457 . . 458 . Candidate 2 . 459 . . 460 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 461 . . 462 . Candidate n . 463 . . 464 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 466 Where L: ice-lite 467 M: ice-mismatch 468 # of candidates: the number of candidates carried by this option 470 Ice-options: 472 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 473 | Length | Data ... 475 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 477 Candidate: 479 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 480 | ver | Length | type | comp-id | 481 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 482 | transport | Reserved | 483 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 484 | Priority | 485 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 486 . . 487 . Connection-address . 488 | | 489 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 490 | port | rel-port | 491 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 492 . . 493 . Rel-address . 494 | | 495 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 497 The fields have the following meaning: 499 ver: IP address version contained in Connection-address and 500 Rel-address fields 502 type: cand-type as defined in ICE 504 comp-id: component-id as defined in ICE 506 transport: transport address 508 priority: sender priority assigned to the connection-address, as 509 defined in ICE 511 connection-address: IP address, 32 bit if ver=4 and 128 bit if 512 ver=6 514 port: port number 516 rel-port: port number, as defined in ICE 518 rel-address: IP address, as defined in ICE 520 Figure 2: Attribute Encoding 522 12. Demultiplexing MIP and STUN messages 524 When MIP and STUN messages are run over the same port it is necessary 525 to demultiplex them. For this usage it is necessary to have a 526 FINGERPRINT attribute in place, as defined in 527 [I-D.ietf-behave-rfc3489bis]. 529 A STUN packet always has the fixed value 0x2112A442 in its Magic 530 Cookie field (bits 32-64 from the beginning of the UDP payload). 532 0 1 2 3 533 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 1 534 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 535 |0 0| STUN Message Type | Message Length | 536 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 537 |0 0 1 0 0 0 0 1 0 0 0 1 0 0 1 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 1 0| 538 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 539 . . . 541 Figure 3: STUN Header 543 In this same offset from the UDP header, the MIP header has the 544 Checksum field and the start of the Message Data field. The 545 concatenation of the Checksum field and the first 16 bits of the 546 Message Data field may coincide with the STUN Magic Cookie. 548 0 1 2 3 549 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 1 550 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 551 | Payload Proto | Header Len | MH Type | Reserved | 552 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 553 | Checksum | | 554 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 555 | | 556 . . 557 . Message Data . 558 . . 559 | | 560 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 561 0 1 2 3 562 . . . 564 Figure 4: MIP Header 566 When the value in that place equals the value of the STUN Magic 567 Cookie, the presence of the STUN FINGERPRINT attribute tells 568 unambigously whether this is a STUN message or not. 570 A future version will also discuss the demultiplexing when UDP 571 encapsulation is not used. 573 13. Example 575 [Editor's Note: An example will be provided in the next version of 576 the document.] 578 14. Security Considerations 580 There are several types of attacks possible in an M-ICE system. This 581 section considers these attacks and their countermeasures. 583 14.1. Attacks on Connectivity Checks 585 An attacker might attempt to disrupt the STUN connectivity checks. 586 Ultimately, all of these attacks fool an agent into thinking 587 something incorrect about the results of the connectivity checks. 588 The possible false conclusions an attacker can try and cause are: 590 False Invalid: 592 An attacker can fool a pair of agents into thinking a candidate 593 pair is invalid, when it isn't. This can be used to cause an 594 agent to prefer a different candidate (such as one injected by the 595 attacker), or to disrupt a call by forcing all candidates to fail. 597 False Valid: 599 An attacker can fool a pair of agents into thinking a candidate 600 pair is valid, when it isn't. This can cause an agent to proceed 601 with a session, but then not be able to receive any data traffic. 603 False Peer-Reflexive Candidate: 605 An attacker can cause an agent to discover a new peer reflexive 606 candidate, when it shouldn't have. This can be used to redirect 607 data traffic to a DoS target or to the attacker, for eavesdropping 608 or other purposes. 610 False Valid on False Candidate: 612 An attacker has already convinced an agent that there is a 613 candidate with an address that doesn't actually route to that 614 agent (for example, by injecting a false peer reflexive candidate 615 or false server reflexive candidate). It must then launch an 616 attack that forces the agents to believe that this candidate is 617 valid. 619 Of the various techniques for creating faked STUN messages described 620 in [I-D.ietf-behave-rfc3489bis], many are not applicable for the 621 connectivity checks. Compromises of STUN servers are not much of a 622 concern, since the STUN servers are embedded in endpoints and 623 distributed throughout the network. Thus, compromising the peer's 624 embedded STUN server is equivalent to compromising the end point, and 625 if that happens, far more problematic attacks are possible than those 626 against ICE. 628 Injection of fake responses and relaying modified requests all can be 629 handled in ICE with the countermeasures discussed below. 631 To force the false invalid result, the attacker has to wait for the 632 connectivity check from one of the agents to be sent. When it is, 633 the attacker needs to inject a fake response with an unrecoverable 634 error response, such as a 600. However, since the candidate is, in 635 fact, valid, the original request may reach the peer agent, and 636 result in a success response. The attacker needs to force this 637 packet or its response to be dropped, through a DoS attack, layer 2 638 network disruption, or other technique. If it doesn't do this, the 639 success response will also reach the originator, alerting it to a 640 possible attack. Fortunately, this attack is mitigated completely 641 through the STUN message integrity mechanism. The attacker needs to 642 inject a fake response, and in order for this response to be 643 processed, the attacker needs the password. If the candidates are 644 exchange in MIP messages and therefore secured, the attacker will not 645 have the password. 647 Forcing the fake valid result works in a similar way. The agent 648 needs to wait for the Binding Request from each agent, and inject a 649 fake success response. The attacker won't need to worry about 650 disrupting the actual response since, if the candidate is not valid, 651 it presumably wouldn't be received anyway. However, like the fake 652 invalid attack, this attack is mitigated completely through the STUN 653 message integrity and offer/answer security techniques. 655 Forcing the false peer reflexive candidate result can be done either 656 with fake requests or responses, or with replays. We consider the 657 fake requests and responses case first. It requires the attacker to 658 send a Binding Request to one agent with a source IP address and port 659 for the false candidate. In addition, the attacker must wait for a 660 Binding Request from the other agent, and generate a fake response 661 with a XOR-MAPPED-ADDRESS attribute containing the false candidate. 662 Like the other attacks described here, this attack is mitigated by 663 the STUN message integrity mechanisms and secure offer/answer 664 exchanges. 666 Forcing the false peer reflexive candidate result with packet replays 667 is different. The attacker waits until one of the agents sends a 668 check. It intercepts this request, and replays it towards the other 669 agent with a faked source IP address. It must also prevent the 670 original request from reaching the remote agent, either by launching 671 a DoS attack to cause the packet to be dropped, or forcing it to be 672 dropped using layer 2 mechanisms. The replayed packet is received at 673 the other agent, and accepted, since the integrity check passes (the 674 integrity check cannot and does not cover the source IP address and 675 port). It is then responded to. This response will contain a XOR- 676 MAPPED-ADDRESS with the false candidate, and will be sent to that 677 false candidate. The attacker must then receive it and relay it 678 towards the originator. 680 The other agent will then initiate a connectivity check towards that 681 false candidate. This validation needs to succeed. This requires 682 the attacker to force a false valid on a false candidate. Injecting 683 of fake requests or responses to achieve this goal is prevented using 684 the integrity mechanisms of STUN and the offer/answer exchange. 685 Thus, this attack can only be launched through replays. To do that, 686 the attacker must intercept the check towards this false candidate, 687 and replay it towards the other agent. Then, it must intercept the 688 response and replay that back as well. 690 This attack is very hard to launch unless the attacker is identified 691 by the fake candidate. This is because it requires the attacker to 692 intercept and replay packets sent by two different hosts. If both 693 agents are on different networks (for example, across the public 694 Internet), this attack can be hard to coordinate, since it needs to 695 occur against two different endpoints on different parts of the 696 network at the same time. 698 If the attacker them self is identified by the fake candidate the 699 attack is easier to coordinate. However, since MIP utilizes IPsec 700 ESP to protect the data traffic end-to-end, the attacker will not be 701 able to inspect any application data, they will only be able to 702 discard them. However, this attack requires the agent to disrupt 703 packets in order to block the connectivity check from reaching the 704 target. In that case, if the goal is to disrupt the end-to-end 705 communication, its much easier to just disrupt it with the same 706 mechanism, rather than attack ICE. 708 14.2. Attacks on Address Gathering 710 ICE endpoints make use of STUN for gathering candidates from a STUN 711 server in the network. This is corresponds to the Binding Discovery 712 usage of STUN described in [I-D.ietf-behave-rfc3489bis]. As a 713 consequence, the attacks against STUN itself that are described in 714 that specification can still be used against the binding discovery 715 usage when utilized with ICE. 717 However, the additional mechanisms provided by ICE actually 718 counteract such attacks, making binding discovery with STUN more 719 secure when combined with ICE. 721 Consider an attacker which is able to provide an agent with a faked 722 mapped address in a STUN Binding Request that is used for address 723 gathering. This is the primary attack primitive described in 724 [I-D.ietf-behave-rfc3489bis]. This address will be used as a server 725 reflexive candidate in the ICE exchange. For this candidate to 726 actually be used for media, the attacker must also attack the 727 connectivity checks, and in particular, force a false valid on a 728 false candidate. This attack is very hard to launch if the false 729 address identifies a fourth party (neither the offerer, answerer, or 730 attacker), since it requires attacking the checks generated by each 731 agent in the session. 733 If the attacker elects not to attack the connectivity checks, the 734 worst it can do is prevent the server reflexive candidate from being 735 used. However, if the peer agent has at least one candidate that is 736 reachable by the agent under attack, the STUN connectivity checks 737 themselves will provide a peer reflexive candidate that can be used 738 for the exchange of media. Peer reflexive candidates are generally 739 preferred over server reflexive candidates. As such, an attack 740 solely on the STUN address gathering will normally have no impact on 741 a session at all. 743 14.3. Attacks on the Offer/Answer Exchanges 745 An attacker that can modify or disrupt the offer/answer exchanges 746 themselves can readily launch a variety of attacks with M-ICE. They 747 could direct data traffic to a target of a DoS attack, they could 748 insert themselves into the data exchange, and so on. The security 749 considerations of MIP apply. 751 14.4. Insider Attacks 753 In addition to attacks where the attacker is a third party trying to 754 insert fake offers, answers or STUN messages, there are several 755 attacks possible with ICE when the attacker is an authenticated and 756 valid participant in the M-ICE exchange. 758 14.4.1. MIP Amplification Attack 760 In this attack, the attacker initiates communication to other agents, 761 and maliciously includes the IP address and port of a DoS target as 762 the destination for data traffic signaled in the MIP exchange. 764 This could causes substantial amplification; a single offer/answer 765 exchange can create a continuing flood of data packets, possibly at 766 high rates (consider video sources). This attack is not specific to 767 ICE, but ICE can help provide remediation. 769 Specifically, if ICE is used, the agent receiving the malicious SDP 770 will first perform connectivity checks to the target of media before 771 sending media there. If this target is a third party host, the 772 checks will not succeed, and media is never sent. 774 Unfortunately, ICE doesn't help if its not used, in which case an 775 attacker could simply send the offer without the ICE parameters. 776 However, in environments where the set of clients are known, and 777 limited to ones that support ICE, the server can reject any offers or 778 answers that don't indicate ICE support. 780 14.4.2. STUN Amplification Attack 782 The STUN amplification attack is similar to the MIP amplification 783 attack. However, instead of data packets being directed to the 784 target, STUN connectivity checks are directed to the target. The 785 attacker sends an offer with a large number of candidates, say 50. 786 The answerer receives the offer, and starts its checks, which are 787 directed at the target, and consequently, never generate a response. 788 The answerer will start a new connectivity check every 20ms, and each 789 check is a STUN transaction consisting of 7 transmissions of a 790 message 65 bytes in length (plus 28 bytes for the IP/UDP header) that 791 runs for 7.9 seconds, for a total of 58 bytes/second per transaction 792 on average. In the worst case, there can be 395 transactions in 793 progress at once (7.9 seconds divided by 20ms), for a total of 182 794 kbps, just for STUN requests. 796 It is impossible to eliminate the amplification, but the volume can 797 be reduced through a variety of heuristics. Agents SHOULD limit the 798 total number of connectivity checks they perform to 100. 799 Additionally, agents MAY limit the number of candidates they'll 800 accept in an offer or answer. 802 15. IAB Considerations 804 The IAB has studied the problem of "Unilateral Self Address Fixing", 805 which is the general process by which a agent attempts to determine 806 its address in another realm on the other side of a NAT through a 807 collaborative protocol reflection mechanism [RFC3424]. M-ICE is an 808 example of a protocol that performs this type of function. 809 Interestingly, the process for M-ICE is not unilateral, but 810 bilateral, and the difference has a significant impact on the issues 811 raised by IAB. M-ICE can be considered a B-SAF (Bilateral Self- 812 Address Fixing) protocol, rather than an UNSAF protocol. Regardless, 813 the IAB has mandated that any protocols developed for this purpose 814 document a specific set of considerations. This section meets those 815 requirements. 817 15.1. Problem Definition 819 From RFC 3424 [RFC3424] any UNSAF proposal must provide: 821 Precise definition of a specific, limited-scope problem that is to be 822 solved with the UNSAF proposal. A short term fix should not be 823 generalized to solve other problems; this is why "short term fixes 824 usually aren't". 826 The specific problems being solved by M-ICE are: 828 Provide a means for two peers to determine the set of transport 829 addresses which can be used for communication. 831 Provide a means for resolving many of the limitations of other UNSAF 832 mechanisms by wrapping them in an additional layer of processing (the 833 M-ICE methodology). 835 Provide a means for a agent to determine an address that is reachable 836 by another peer with which it wishes to communicate. 838 15.2. Exit Strategy 840 From RFC 3424, any UNSAF proposal must provide: 842 Description of an exit strategy/transition plan. The better short 843 term fixes are the ones that will naturally see less and less use as 844 the appropriate technology is deployed. 846 M-ICE itself doesn't easily get phased out. However, it is useful 847 even in a globally connected Internet, to serve as a means for 848 detecting whether communication paths are disrupted. M-ICE also 849 helps prevent certain security attacks which have nothing to do with 850 NAT. However, what M-ICE does is help phase out other UNSAF 851 mechanisms. M-ICE effectively selects amongst those mechanisms, 852 prioritizing ones that are better, and deprioritizing ones that are 853 worse. Local IPv6 addresses can be preferred. As NATs begin to 854 dissipate as IPv6 is introduced, server reflexive and relayed 855 candidates (both forms of UNSAF mechanisms) simply never get used, 856 because higher priority connectivity exists to the native host 857 candidates. Therefore, the servers get used less and less, and can 858 eventually be remove when their usage goes to zero. 860 Indeed, M-ICE can assist in the transition from IPv4 to IPv6. It can 861 be used to determine whether to use IPv6 or IPv4 when two dual-stack 862 hosts communicate. It can also allow a network with both 6to4 and 863 native v6 connectivity to determine which address to use when 864 communicating with a peer. 866 15.3. Brittleness Introduced by M-ICE 868 From RFC3424, any UNSAF proposal must provide: 870 Discussion of specific issues that may render systems more "brittle". 871 For example, approaches that involve using data at multiple network 872 layers create more dependencies, increase debugging challenges, and 873 make it harder to transition. 875 M-ICE uses ICE that is utilizes [I-D.ietf-behave-rfc3489bis] instead 876 of traditional STUN, RFC 3489 [RFC3489]). RFC 3489 has several 877 points of brittleness. One of them is the discovery process which 878 requires a agent to try and classify the type of NAT it is behind. 879 This process is error-prone. With M-ICE, that discovery process is 880 simply not used. Rather than unilaterally assessing the validity of 881 the address, its validity is dynamically determined by measuring 882 connectivity to a peer. The process of determining connectivity is 883 very robust. 885 Another point of brittleness in traditional STUN is that it assumes 886 that the STUN server is on the public Internet. Interestingly, with 887 M-ICE, that is not necessary. There can be a multitude of STUN 888 servers in a variety of address realms. ICE will discover the one 889 that has provided a usable address. 891 The most troubling point of brittleness in traditional STUN is that 892 it does not work in all network topologies. In cases where there is 893 a shared NAT between each agent and the STUN server, traditional STUN 894 may not work. With ICE, that restriction is removed. 896 Traditional STUN also introduces some security considerations. 897 Fortunately, those security considerations are also mitigated by ICE. 899 Consequently, ICE serves to repair the brittleness introduced in 900 other UNSAF mechanisms, and does not introduce any additional 901 brittleness into the system. 903 With M-ICE Home Agents are used and they are assumed to be located on 904 the public Internet to allow MIP to work. 906 15.4. Requirements for a Long Term Solution 908 From RFC 3424, any UNSAF proposal must provide: 910 Identify requirements for longer term, sound technical solutions -- 911 contribute to the process of finding the right longer term solution. 913 M-ICE provides a long term solution by utilizing ICE concepts that 914 have received a lot of peer review in the VoIP community and to apply 915 them to MIP. The only other possible long term solutions are (a) to 916 get rid of middleboxes, such as NATs and firewalls or to (b) interact 917 with them. Regarding (b) extensions for STUN to allow the protocol 918 to be deployed on NATs and firewalls is currently being investigated 919 in [I-D.wing-behave-nat-control-stun-usage]. 921 15.5. Issues with Existing NAPT Boxes 923 From RFC 3424, any UNSAF proposal must provide: 925 Discussion of the impact of the noted practical issues with existing, 926 deployed NA[P]Ts and experience reports. 928 A number of NAT boxes are now being deployed into the market which 929 try and provide "generic" ALG functionality. These generic ALGs hunt 930 for IP addresses, either in text or binary form within a packet, and 931 rewrite them if they match a binding. This interferes with 932 traditional STUN. However, the update to STUN 933 [I-D.ietf-behave-rfc3489bis] uses an encoding which hides these 934 binary addresses from generic ALGs. 936 Existing NAPT boxes have non-deterministic and typically short 937 expiration times for UDP-based bindings. This requires 938 implementations to send periodic keepalives to maintain those 939 bindings. ICE uses a default of 15s, which is a very conservative 940 estimate. Eventually, over time, as NAT boxes become compliant to 941 behave [RFC4787], this minimum keepalive will become deterministic 942 and well-known, and the ICE timers can be adjusted. Having a way to 943 discover and control the minimum keepalive interval would be far 944 better still. 946 16. Contributors 948 We would like to thank Thomas Schreck for his contributions to 949 various aspects in this document. 951 17. Acknowledgments 953 The authors would like to thank Jonathan Rosenberg for his work on 954 the ICE specification. This document copy-and-pastes text from the 955 ICE specification. Hence, all the credits go to Jonathan. 957 Finally, Dan Wing and Philip Matthews helped us with the work on HIP- 958 ICE. 960 18. References 962 18.1. Normative References 964 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 965 Requirement Levels", March 1997. 967 [I-D.ietf-mmusic-ice] 968 Rosenberg, J., "Interactive Connectivity Establishment 969 (ICE): A Protocol for Network Address Translator (NAT) 970 Traversal for Offer/Answer Protocols", 971 draft-ietf-mmusic-ice-16 (work in progress), June 2007. 973 [RFC3775] Johnson, D., Perkins, C., and J. Arkko, "Mobility Support 974 in IPv6", RFC 3775, June 2004. 976 18.2. Informative References 978 [I-D.ietf-behave-turn] 979 Rosenberg, J., "Obtaining Relay Addresses from Simple 980 Traversal Underneath NAT (STUN)", 981 draft-ietf-behave-turn-03 (work in progress), March 2007. 983 [I-D.ietf-shim6-failure-detection] 984 Arkko, J. and I. Beijnum, "Failure Detection and Locator 985 Pair Exploration Protocol for IPv6 Multihoming", 986 draft-ietf-shim6-failure-detection-08 (work in progress), 987 June 2007. 989 [RFC3489] Rosenberg, J., Weinberger, J., Huitema, C., and R. Mahy, 990 "STUN - Simple Traversal of User Datagram Protocol (UDP) 991 Through Network Address Translators (NATs)", RFC 3489, 992 March 2003. 994 [I-D.ietf-behave-rfc3489bis] 995 Rosenberg, J., Huitema, C., Mahy, R., Matthews, P., and D. 996 Wing, "Session Traversal Utilities for (NAT) (STUN)", 997 draft-ietf-behave-rfc3489bis-07 (work in progress), 998 July 2007. 1000 [RFC3424] Daigle, L. and IAB, "IAB Considerations for UNilateral 1001 Self-Address Fixing (UNSAF) Across Network Address 1002 Translation", RFC 3424, November 2002. 1004 [I-D.wing-behave-nat-control-stun-usage] 1005 Wing, D. and J. Rosenberg, "Discovering, Querying, and 1006 Controlling Firewalls and NATs using STUN", 1007 draft-wing-behave-nat-control-stun-usage-02 (work in 1008 progress), June 2007. 1010 [RFC4474] Peterson, J. and C. Jennings, "Enhancements for 1011 Authenticated Identity Management in the Session 1012 Initiation Protocol (SIP)", RFC 4474, August 2006. 1014 [I-D.ietf-monami6-multiplecoa] 1015 Wakikawa, R., "Multiple Care-of Addresses Registration", 1016 draft-ietf-monami6-multiplecoa-02 (work in progress), 1017 March 2007. 1019 [I-D.ietf-mip6-nemo-v4traversal] 1020 Soliman, H., "Mobile IPv6 support for dual stack Hosts and 1021 Routers (DSMIPv6)", draft-ietf-mip6-nemo-v4traversal-04 1022 (work in progress), March 2007. 1024 [I-D.ietf-mip4-dsmipv4] 1025 Tsirtsis, G., "Dual Stack Mobile IPv4", 1026 draft-ietf-mip4-dsmipv4-02 (work in progress), May 2007. 1028 [I-D.marjou-behave-app-rtp-keepalive] 1029 Marjou, X., "Application Mechanism for maintaining alive 1030 the Network Address Translator (NAT) mappings associated 1031 to RTP flows.", draft-marjou-behave-app-rtp-keepalive-01 1032 (work in progress), February 2007. 1034 [RFC4787] Audet, F. and C. Jennings, "Network Address Translation 1035 (NAT) Behavioral Requirements for Unicast UDP", BCP 127, 1036 RFC 4787, January 2007. 1038 [I-D.bajko-mip6-rrtfw] 1039 Bajko, G., "Firewall friendly RTT for MIPv6", 1040 draft-bajko-mip6-rrtfw-01 (work in progress), 1041 October 2006. 1043 Authors' Addresses 1045 Hannes Tschofenig 1046 Nokia Siemens Networks 1047 Otto-Hahn-Ring 6 1048 Munich, Bavaria 81739 1049 Germany 1051 Email: Hannes.Tschofenig@nsn.com 1052 URI: http://www.tschofenig.com 1054 Gabor Bajko 1055 Nokia 1057 Full Copyright Statement 1059 Copyright (C) The IETF Trust (2007). 1061 This document is subject to the rights, licenses and restrictions 1062 contained in BCP 78, and except as set forth therein, the authors 1063 retain all their rights. 1065 This document and the information contained herein are provided on an 1066 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 1067 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 1068 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 1069 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 1070 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 1071 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 1073 Intellectual Property 1075 The IETF takes no position regarding the validity or scope of any 1076 Intellectual Property Rights or other rights that might be claimed to 1077 pertain to the implementation or use of the technology described in 1078 this document or the extent to which any license under such rights 1079 might or might not be available; nor does it represent that it has 1080 made any independent effort to identify any such rights. Information 1081 on the procedures with respect to rights in RFC documents can be 1082 found in BCP 78 and BCP 79. 1084 Copies of IPR disclosures made to the IETF Secretariat and any 1085 assurances of licenses to be made available, or the result of an 1086 attempt made to obtain a general license or permission for the use of 1087 such proprietary rights by implementers or users of this 1088 specification can be obtained from the IETF on-line IPR repository at 1089 http://www.ietf.org/ipr. 1091 The IETF invites any interested party to bring to its attention any 1092 copyrights, patents or patent applications, or other proprietary 1093 rights that may cover technology that may be required to implement 1094 this standard. Please address the information to the IETF at 1095 ietf-ipr@ietf.org. 1097 Acknowledgment 1099 Funding for the RFC Editor function is provided by the IETF 1100 Administrative Support Activity (IASA).