idnits 2.17.1 draft-ietf-tram-stun-pmtud-13.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. 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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (September 10, 2019) is 1690 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) -- Possible downref: Non-RFC (?) normative reference: ref. 'ITU.V42.2002' ** Obsolete normative reference: RFC 5389 (Obsoleted by RFC 8489) -- Obsolete informational reference (is this intentional?): RFC 5766 (Obsoleted by RFC 8656) Summary: 1 error (**), 0 flaws (~~), 1 warning (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 TRAM M. Petit-Huguenin 3 Internet-Draft Impedance Mismatch 4 Intended status: Standards Track G. Salgueiro 5 Expires: March 13, 2020 F. Garrido 6 Cisco 7 September 10, 2019 9 Path MTU Discovery Using Session Traversal Utilities for NAT (STUN) 10 draft-ietf-tram-stun-pmtud-13 12 Abstract 14 This document describes a Session Traversal Utilities for NAT (STUN) 15 Usage for Path MTU Discovery (PMTUD) between a client and a server. 17 Status of This Memo 19 This Internet-Draft is submitted in full conformance with the 20 provisions of BCP 78 and BCP 79. 22 Internet-Drafts are working documents of the Internet Engineering 23 Task Force (IETF). Note that other groups may also distribute 24 working documents as Internet-Drafts. The list of current Internet- 25 Drafts is at https://datatracker.ietf.org/drafts/current/. 27 Internet-Drafts are draft documents valid for a maximum of six months 28 and may be updated, replaced, or obsoleted by other documents at any 29 time. It is inappropriate to use Internet-Drafts as reference 30 material or to cite them other than as "work in progress." 32 This Internet-Draft will expire on March 13, 2020. 34 Copyright Notice 36 Copyright (c) 2019 IETF Trust and the persons identified as the 37 document authors. All rights reserved. 39 This document is subject to BCP 78 and the IETF Trust's Legal 40 Provisions Relating to IETF Documents 41 (https://trustee.ietf.org/license-info) in effect on the date of 42 publication of this document. Please review these documents 43 carefully, as they describe your rights and restrictions with respect 44 to this document. Code Components extracted from this document must 45 include Simplified BSD License text as described in Section 4.e of 46 the Trust Legal Provisions and are provided without warranty as 47 described in the Simplified BSD License. 49 Table of Contents 51 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 52 2. Overview of Operations . . . . . . . . . . . . . . . . . . . 4 53 3. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 54 4. Probing Mechanisms . . . . . . . . . . . . . . . . . . . . . 5 55 4.1. Simple Probing Mechanism . . . . . . . . . . . . . . . . 6 56 4.1.1. Sending a Probe Request . . . . . . . . . . . . . . . 6 57 4.1.2. Receiving a Probe Request . . . . . . . . . . . . . . 6 58 4.1.3. Receiving a Probe Response . . . . . . . . . . . . . 7 59 4.2. Complete Probing Mechanism . . . . . . . . . . . . . . . 7 60 4.2.1. Sending a Probe Indications and Report Request . . . 8 61 4.2.2. Receiving an ICMP Packet . . . . . . . . . . . . . . 8 62 4.2.3. Receiving a Probe Indication and Report Request . . . 8 63 4.2.4. Receiving a Report Response . . . . . . . . . . . . . 9 64 4.2.5. Using Checksums as Packet Identifiers . . . . . . . . 9 65 4.2.6. Using Sequence Numbers as Packet Identifiers . . . . 10 66 5. Probe Support Signaling Mechanisms . . . . . . . . . . . . . 10 67 5.1. Explicit Probe Support Signaling Mechanism . . . . . . . 11 68 5.2. Implicit Probe Support Signaling Mechanism . . . . . . . 11 69 6. STUN Attributes . . . . . . . . . . . . . . . . . . . . . . . 11 70 6.1. IDENTIFIERS . . . . . . . . . . . . . . . . . . . . . . . 12 71 6.2. PMTUD-SUPPORTED . . . . . . . . . . . . . . . . . . . . . 12 72 6.3. PADDING . . . . . . . . . . . . . . . . . . . . . . . . . 12 73 7. Security Considerations . . . . . . . . . . . . . . . . . . . 12 74 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 13 75 8.1. New STUN Methods . . . . . . . . . . . . . . . . . . . . 13 76 8.2. New STUN Attributes . . . . . . . . . . . . . . . . . . . 13 77 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 14 78 9.1. Normative References . . . . . . . . . . . . . . . . . . 14 79 9.2. Informative References . . . . . . . . . . . . . . . . . 14 80 Appendix A. Release Notes . . . . . . . . . . . . . . . . . . . 15 81 A.1. Modifications between draft-ietf-tram-stun-pmtud-13 and 82 draft-ietf-tram-stun-pmtud-12 . . . . . . . . . . . . . . 15 83 A.2. Modifications between draft-ietf-tram-stun-pmtud-12 and 84 draft-ietf-tram-stun-pmtud-11 . . . . . . . . . . . . . . 15 85 A.3. Modifications between draft-ietf-tram-stun-pmtud-11 and 86 draft-ietf-tram-stun-pmtud-10 . . . . . . . . . . . . . . 15 87 A.4. Modifications between draft-ietf-tram-stun-pmtud-10 and 88 draft-ietf-tram-stun-pmtud-09 . . . . . . . . . . . . . . 15 89 A.5. Modifications between draft-ietf-tram-stun-pmtud-09 and 90 draft-ietf-tram-stun-pmtud-08 . . . . . . . . . . . . . . 15 91 A.6. Modifications between draft-ietf-tram-stun-pmtud-08 and 92 draft-ietf-tram-stun-pmtud-07 . . . . . . . . . . . . . . 15 93 A.7. Modifications between draft-ietf-tram-stun-pmtud-07 and 94 draft-ietf-tram-stun-pmtud-06 . . . . . . . . . . . . . . 16 95 A.8. Modifications between draft-ietf-tram-stun-pmtud-06 and 96 draft-ietf-tram-stun-pmtud-05 . . . . . . . . . . . . . . 16 98 A.9. Modifications between draft-ietf-tram-stun-pmtud-05 and 99 draft-ietf-tram-stun-pmtud-04 . . . . . . . . . . . . . . 16 100 A.10. Modifications between draft-ietf-tram-stun-pmtud-04 and 101 draft-ietf-tram-stun-pmtud-03 . . . . . . . . . . . . . . 16 102 A.11. Modifications between draft-ietf-tram-stun-pmtud-03 and 103 draft-ietf-tram-stun-pmtud-02 . . . . . . . . . . . . . . 16 104 A.12. Modifications between draft-ietf-tram-stun-pmtud-02 and 105 draft-ietf-tram-stun-pmtud-01 . . . . . . . . . . . . . . 17 106 A.13. Modifications between draft-ietf-tram-stun-pmtud-01 and 107 draft-ietf-tram-stun-pmtud-00 . . . . . . . . . . . . . . 17 108 A.14. Modifications between draft-ietf-tram-stun-pmtud-00 and 109 draft-petithuguenin-tram-stun-pmtud-01 . . . . . . . . . 17 110 A.15. Modifications between draft-petithuguenin-tram-stun- 111 pmtud-01 and draft-petithuguenin-tram-stun-pmtud-00 . . . 17 112 A.16. Modifications between draft-petithuguenin-tram-stun- 113 pmtud-00 and draft-petithuguenin-behave-stun-pmtud-03 . . 18 114 A.17. Modifications between draft-petithuguenin-behave-stun- 115 pmtud-03 and draft-petithuguenin-behave-stun-pmtud-02 . . 18 116 A.18. Modifications between draft-petithuguenin-behave-stun- 117 pmtud-02 and draft-petithuguenin-behave-stun-pmtud-01 . . 18 118 A.19. Modifications between draft-petithuguenin-behave-stun- 119 pmtud-01 and draft-petithuguenin-behave-stun-pmtud-00 . . 18 120 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . 19 121 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 19 123 1. Introduction 125 The Packetization Layer Path MTU Discovery (PMTUD) specification 126 [RFC4821] describes a method to discover the Path MTU but does not 127 describe a practical protocol to do so with UDP. 129 Many UDP-based protocols do not implement the Path MTU discovery 130 mechanism described in [RFC4821]. These protocols can make use of 131 the probing mechanisms described in this document instead of 132 designing their own adhoc extension. These probing mechanisms are 133 implemented with Session Traversal Utilities for NAT (STUN), but 134 their usage is not limited to STUN-based protocols. 136 The STUN usage defined in this document for Path MTU Discovery 137 (PMTUD) between a client and a server permits proper operations of 138 UDP-based applications in the network. It also simplifies 139 troubleshooting and has multiple other applications across a wide 140 variety of technologies. 142 Complementary techniques can be used to discover additional network 143 characteristics, such as the network path (using the STUN Traceroute 144 mechanism described in [I-D.martinsen-tram-stuntrace]) and bandwidth 145 availability (using the mechanism described in 146 [I-D.martinsen-tram-turnbandwidthprobe]). 148 2. Overview of Operations 150 This section is meant to be informative only. It is not intended as 151 a replacement for [RFC4821]. 153 A UDP endpoint that uses this specification to discover the Path MTU 154 over UDP and knows that the endpoint it is communicating with also 155 supports this specification can choose to use either the Simple 156 Probing mechanism (as described in Section 4.1) or the Complete 157 Probing mechanism (as described in Section 4.2). The selection of 158 which Probing Mechanism to use is dependent on performance and 159 security and complexity trade-offs. 161 If the Simple Probing mechanism is chosen, then the Client initiates 162 Probe transactions, as shown in Figure 1, which increase in size 163 until transactions timeout, indicating that the Path MTU has been 164 exceeded. It then uses that information to update the Path MTU. 166 Client Server 167 | | 168 | Probe Request | 169 |---------------->| 170 | | 171 | Probe Response | 172 |<----------------| 173 | | 175 Figure 1: Simple Probing Example 177 If the Complete Probing mechanism (as described in Section 4.2) is 178 chosen, then the Client sends Probe Indications of various sizes (as 179 specified in [RFC4821]) interleaved with UDP packets sent by the UDP 180 protocol. The Client then sends a Report Request for the ordered 181 list of identifiers for the UDP packets and Probe Indications 182 received by the Server. The Client then compares the list returned 183 in the Report Response with its own list of identifiers for the UDP 184 packets and Probe Indications it sent. The Client then uses that 185 comparison to find which Probe Indications were dropped by the 186 network as a result of their size. It then uses that information to 187 update the Path MTU. 189 Because of the possibility of amplification attack, the Complete 190 Probing mechanism must be authenticated. Particular care must be 191 taken to prevent amplification when an external mechanism is used to 192 trigger the Complete Probing mechanism. 194 Client Server 195 | UDP Packet | 196 |------------------>| 197 | | 198 | UDP Packet | 199 |------------------>| 200 | | 201 | Probe Indication | 202 |------------------>| 203 | | 204 | UDP Packet | 205 |------------------>| 206 | | 207 | Probe Indication | 208 |------------------>| 209 | | 210 | Report Request | 211 |------------------>| 212 | Report Response | 213 |<------------------| 214 | | 216 Figure 2: Complete Probing Example 218 3. Terminology 220 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 221 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 222 "OPTIONAL" in this document are to be interpreted as described in BCP 223 14 [RFC2119][RFC8174] when, and only when, they appear in all 224 capitals, as shown here. 226 4. Probing Mechanisms 228 The Probing mechanism is used to discover the Path MTU in one 229 direction only: from the client to the server. 231 Two Probing mechanisms are described: a Simple Probing mechanism and 232 a more complete mechanism that can converge more quickly and find an 233 appropriate Path MTU in the presence of congestion. Additionally, 234 the Simple Probing mechanism does not require authentication except 235 where used as an implicit signaling mechanism, whereas the complete 236 mechanism does. 238 Implementations supporting this specification MUST implement the 239 server side of both the Simple Probing mechanism (Section 4.1) and 240 the Complete Probing mechanism (Section 4.2). 242 Implementations supporting this specification MUST implement the 243 client side of the Complete Probing mechanism. They MAY implement 244 the client side of the Simple Probing mechanism. 246 4.1. Simple Probing Mechanism 248 The Simple Probing mechanism is implemented by sending a Probe 249 Request with a PADDING attribute over UDP with the DF bit set in the 250 IP header for IPv4 packets and IPv6 packets without the Fragment 251 Header included. A router on the path to the server can reject each 252 request with an ICMP message or drop it. 254 4.1.1. Sending a Probe Request 256 A client forms a Probe Request by using the Probe Method and 257 following the rules in Section 7.1 of [RFC5389]. 259 The Probe transaction MUST be authenticated if the Simple Probing 260 mechanism is used in conjunction with the Implicit Probing Support 261 mechanism described in Section 5.2. If not, the Probe transaction 262 MAY be authenticated. 264 The client adds a PADDING attribute with a length that, when added to 265 the IP and UDP headers and the other STUN components, is equal to the 266 Selected Probe Size, as defined in [RFC4821] Section 7.3. The 267 PADDING bits SHOULD be set to zero. The client MUST add the 268 FINGERPRINT attribute so the STUN messages are disambiguated from the 269 other protocol packets. 271 Then the client sends the Probe Request to the server over UDP with 272 the DF bit set for IPv4 packets and IPv6 packets without the Fragment 273 Header included. For the purpose of this transaction, the Rc 274 parameter specified in Section 7.2.1 of [RFC5389] is set to 3. The 275 initial value for RTO stays at 500 ms. 277 A client MUST NOT send a probe if it does not have knowledge that the 278 server supports this specification. This is done either by external 279 signalling or by a mechanism specific to the UDP protocol to which 280 PMTUD capabilities are added or by one of the mechanisms specified in 281 Section 5. 283 4.1.2. Receiving a Probe Request 285 A server receiving a Probe Request MUST process it as specified in 286 [RFC5389]. 288 The server then creates a Probe Response. The server MUST add the 289 FINGERPRINT attribute so the STUN messages are disambiguated from the 290 other protocol packets. The server then sends the response to the 291 client. 293 4.1.3. Receiving a Probe Response 295 A client receiving a Probe Response MUST process it as specified in 296 [RFC5389] and MUST ignore the PADDING attribute. If a response is 297 received this is interpreted as a Probe Success, as defined in 298 [RFC4821] Section 7.6.1. If an ICMP packet "Fragmentation needed" or 299 "Packet Too Big" is received then this is interpreted as a Probe 300 Failure, as defined in [RFC4821] Section 7.6.2. If the Probe 301 transaction times out, then this is interpreted as a Probe 302 Inconclusive, as defined in [RFC4821] Section 7.6.4. 304 4.2. Complete Probing Mechanism 306 The Complete Probing mechanism is implemented by sending one or more 307 Probe Indications with a PADDING attribute over UDP with the DF bit 308 set in the IP header for IPv4 packets and IPv6 packets without the 309 Fragment Header included followed by a Report Request to the same 310 server. A router on the path to the server can reject this 311 Indication with an ICMP message or drop it. The server keeps a 312 chronologically ordered list of identifiers for all packets received 313 (including retransmitted packets) and sends this list back to the 314 client in the Report Response. The client analyzes this list to find 315 which packets were not received. Because UDP packets do not contain 316 an identifier, the Complete Probing mechanism needs a way to identify 317 each packet received. 319 Some application layer protocols may already have a way of 320 identifying each individual UDP packet, in which case these 321 identifiers SHOULD be used in the IDENTIFIERS attribute of the Report 322 Response. While there are other possible packet identification 323 schemes, this document describes two different ways to identify a 324 specific packet when no application layer protocol-specific 325 identification mechanism is available. 327 In the first packet identification mechanism, the server computes a 328 checksum over each packet received and sends back to the sender the 329 list of checksums ordered chronologically. The client compares this 330 list to its own list of checksums. 332 In the second packet identification mechanism, the client prepends 333 the UDP data with a header that provides a sequence number. The 334 server sends back the chronologically ordered list of sequence 335 numbers received that the client then compares with its own list. 337 4.2.1. Sending a Probe Indications and Report Request 339 A client forms a Probe Indication by using the Probe Method and 340 following the rules in [RFC5389] Section 7.1. The client adds to a 341 Probe Indication a PADDING attribute with a size that, when added to 342 the IP and UDP headers and the other STUN components, is equal to the 343 Selected Probe Size, as defined in [RFC4821] Section 7.3. The 344 PADDING bits SHOULD be set to zero. If the authentication mechanism 345 permits it, then the Indication MUST be authenticated. The client 346 MUST add the FINGERPRINT attribute so the STUN messages are 347 disambiguated from the other protocol packets. 349 Then the client sends a Probe Indication to the server over UDP with 350 the DF bit set for IPv4 packets and IPv6 packets without the Fragment 351 Header included. 353 Then the client forms a Report Request by following the rules in 354 [RFC5389] Section 7.1. The Report transaction MUST be authenticated 355 to prevent amplification attacks. The client MUST add the 356 FINGERPRINT attribute so the STUN messages are disambiguated from the 357 other protocol packets. 359 Then the client waits half the RTO after sending the last Probe 360 Indication and then sends the Report Request to the server over UDP. 362 4.2.2. Receiving an ICMP Packet 364 If an ICMP packet "Fragmentation needed" or "Packet Too Big" is 365 received then this is interpreted as a Probe Failure, as defined in 366 [RFC4821] Section 7.5. 368 4.2.3. Receiving a Probe Indication and Report Request 370 A server supporting this specification will keep the identifiers of 371 all packets received in a chronologically ordered list. The packets 372 that are to be associated to a given flow's identifier are selected 373 according to Section 5.2 of [RFC4821]. The same identifier can 374 appear multiple times in the list because of retransmissions. The 375 maximum size of this list is calculated such that when the list is 376 added to the Report Response, the total size of the packet does not 377 exceed the unknown Path MTU, as defined in [RFC5389] Section 7.1. 378 Older identifiers are removed when new identifiers are added to a 379 list that is already full. 381 A server receiving a Report Request MUST process it as specified in 382 [RFC5389] and MUST ignore the PADDING attribute. 384 The server creates a Report Response and adds an IDENTIFIERS 385 attribute that contains the chronologically ordered list of all 386 identifiers received so far. The server MUST add the FINGERPRINT 387 attribute. The server then sends the response to the client. 389 The exact content of the IDENTIFIERS attribute depends on what type 390 of identifiers have been chosen for the protocol. Each protocol 391 adding PMTUD capabilities as specified by this specification MUST 392 describe the format of the contents of the IDENTIFIERS attribute, 393 unless it is using one of the formats described in this 394 specification. See Section 6.1 for details about the IDENTIFIERS 395 attribute. 397 4.2.4. Receiving a Report Response 399 A client receiving a Report Response processes it as specified in 400 [RFC5389]. If the response IDENTIFIERS attribute contains the 401 identifier of a Probe Indication, then this is interpreted as a Probe 402 Success for this probe, as defined in [RFC4821] Section 7.5. If a 403 Probe Indication identifier cannot be found in the Report Response, 404 this is interpreted as a Probe Failure, as defined in [RFC4821] 405 Section 7.5. If a Probe Indication identifier cannot be found in the 406 Report Response but identifiers for other packets sent before or 407 after the Probe Indication can all be found, this is interpreted as a 408 Probe Failure as defined in [RFC4821] Section 7.5. If the Report 409 Transaction times out, this is interpreted as a Full-Stop Timeout, as 410 defined in [RFC4821] Section 3. 412 4.2.5. Using Checksums as Packet Identifiers 414 When using a checksum as a packet identifier, the client keeps a 415 chronologically ordered list of the packets it transmits, along with 416 an associated checksum value. For STUN Probe Indication or Request 417 packets, the associated checksum value is the FINGERPRINT value from 418 the packet; for other packets a checksum value is computed using a 419 similar algorithm to the FINGERPRINT calculation. (i.e., the CRC-32 420 calculated per the algorithm defined in [ITU.V42.2002], such as 421 subsequently been XOR'ed with 32-bit value 0x5354554e). 423 For each STUN Probe Indication or Request, the server retrieves the 424 STUN FINGERPRINT value. For all other packets, the server calculates 425 the checksum as described above. It puts these FINGERPRINT and 426 checksum values in a chronologically ordered list that is sent back 427 in the Report Response. 429 The contents of the IDENTIFIERS attribute is a list of 4 byte 430 numbers, each using the same encoding that is used for the contents 431 of the FINGERPRINT attribute. 433 It could have been possible to use the checksum generated in the UDP 434 checksum for this, but this value is generally not accessible to 435 applications. Also, sometimes the checksum is not calculated or is 436 off-loaded to network hardware. 438 4.2.6. Using Sequence Numbers as Packet Identifiers 440 When using sequence numbers, a small header similar to the TURN 441 ChannelData header, as defined in Section 11.4 of [RFC5766], is added 442 in front of all packets that are not a STUN Probe Indication or 443 Request. The sequence number is monotonically incremented by one for 444 each packet sent. The most significant bit of the sequence number is 445 always 0. The server collects the sequence number of the packets 446 sent, or the 4 first bytes of the transaction ID if a STUN Probe 447 Indication or Request is sent. In that case, the most significant 448 bit of the 4 first bytes is set to 1. 450 0 1 2 3 451 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 1 452 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 453 | Channel Number | Length | 454 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 455 |0| Sequence number | 456 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 457 | | 458 / Application Data / 459 / / 460 | | 461 | +-------------------------------+ 462 | | 463 +-------------------------------+ 465 The Channel Number is always 0xFFFF. The Length field specifies the 466 length in bytes of the sequence number and application data fields. 467 The header values are encoded using network order. 469 The contents of the IDENTIFIERS attribute is a chronologically 470 ordered list of 4 byte numbers, each containing either a sequence 471 number, if the packet was not a STUN Probe Indication or Request, or 472 the 4 first bytes of the transaction ID, with the most significant 473 bit forced to 1, if the packet is a STUN Probe Indication or Request. 475 5. Probe Support Signaling Mechanisms 477 The PMTUD mechanism described in this document is intended to be used 478 by any UDP-based protocols that do not have built-in PMTUD 479 capabilities, irrespective of whether those UDP-based protocols are 480 STUN-based or not. So the manner in which a specific protocol 481 discovers that it is safe to send PMTUD probes is largely dependent 482 on the details of that specific protocol, with the exception of the 483 Implicit Mechanism described below, which applies to any protocol. 485 5.1. Explicit Probe Support Signaling Mechanism 487 Some of these mechanisms can use a separate signalling mechanism (for 488 instance, an SDP attribute in an Offer/Answer exchange [RFC3264]), or 489 an optional flag that can be set in the protocol that is augmented 490 with PMTUD capabilities. STUN Usages that can benefit from PMTUD 491 capabilities can signal in-band that they support probing by 492 inserting a PMTUD-SUPPORTED attribute in some STUN methods. The 493 decision of which methods support this attribute is left to each 494 specific STUN Usage. 496 UDP-based protocols that want to use any of these mechanisms, 497 including the PMTUD-SUPPORTED attribute, to signal PMTUD capabilities 498 MUST ensure that it cannot be used to launch an amplification attack. 500 An amplification attack can be prevented using techniques such as: 502 o Authentication, where the source of the packet and the destination 503 share a secret. 505 o 3 way handshake with some form of unpredictable cookie. 507 o Make sure that the total size of the traffic potentially generated 508 is lower than the size of the request that generated it. 510 5.2. Implicit Probe Support Signaling Mechanism 512 As a result of the fact that all endpoints implementing this 513 specification are both clients and servers, a Probe Request or 514 Indication received by an endpoint acting as a server implicitly 515 signals that this server can now act as a client and MAY send a Probe 516 Request or Indication to probe the Path MTU in the reverse direction 517 toward the former client, that will now be acting as a server. 519 The Probe Request or Indication that are used to implicitly signal 520 probing support in the reverse direction MUST be authenticated to 521 prevent amplification attacks. 523 6. STUN Attributes 524 6.1. IDENTIFIERS 526 The IDENTIFIERS attribute carries a chronologically ordered list of 527 UDP packet identifiers. 529 While Section 4.2.5 and Section 4.2.6 describe two possible methods 530 for acquiring and formatting the identifiers used for this purpose, 531 ultimately each protocol has to define how these identifiers are 532 acquired and formatted. Therefore, the contents of the IDENTIFIERS 533 attribute is opaque. 535 6.2. PMTUD-SUPPORTED 537 The PMTUD-SUPPORTED attribute indicates that its sender supports this 538 mechanism, as incorporated into the STUN usage or protocol being 539 used. This attribute has no value part and thus the attribute length 540 field is 0. 542 6.3. PADDING 544 The PADDING attribute allows for the entire message to be padded to 545 force the STUN message to be divided into IP fragments. PADDING 546 consists entirely of a free-form string, the value of which does not 547 matter. PADDING can be used in either Binding Requests or Binding 548 Responses. 550 PADDING MUST NOT be longer than the length that brings the total IP 551 datagram size to 64K. It SHOULD be equal in length to the MTU of the 552 outgoing interface, rounded up to an even multiple of four bytes. 553 Because STUN messages with PADDING are intended to test the behavior 554 of UDP fragments, they are an exception to the usual rule that STUN 555 messages be less than the MTU of the path. 557 7. Security Considerations 559 The PMTUD mechanism described in this document, when used without the 560 signalling mechanism described in Section 5.1, does not introduce any 561 specific security considerations beyond those described in [RFC4821]. 563 The attacks described in Section 11 of [RFC4821] apply equally to the 564 mechanism described in this document. 566 The amplification attacks introduced by the signalling mechanism 567 described in Section 5.1 can be prevented by using one of the 568 techniques described in that section. 570 The Simple Probing mechanism may be used without authentication 571 because this usage by itself cannot trigger an amplification attack 572 as the Probe Response is smaller than the Probe Request. An 573 unauthenticated Simple Probing mechanism cannot be used in 574 conjunction with the Implicit Probing Support Signaling mechanism in 575 order to prevent amplification attacks. 577 8. IANA Considerations 579 This specification defines two new STUN methods and two new STUN 580 attributes. 582 8.1. New STUN Methods 584 IANA is requested to add the following methods to the STUN Method 585 Registry: 587 0xXXX : Probe 589 0xXXX : Report 591 See Sections Section 4.1 and Section 4.2 for the semantics of these 592 new methods. 594 8.2. New STUN Attributes 596 IANA is requested to add the following attributes to the STUN Method 597 Registry: 599 Comprehension-required range (0x0000-0x7FFF): 600 0xXXXX: IDENTIFIERS 602 Comprehension-optional range (0x8000-0xFFFF) 603 0xXXXX: PMTUD-SUPPORTED 605 0x0026: PADDING 607 The IDENTIFIERS STUN attribute is defined in Section 6.1, the PMTUD- 608 SUPPORTED STUN attribute is defined in Section 6.2; the PADDING STUN 609 attribute is defined in Section 6.3. 611 NOTE: TO BE DELETED BEFORE PUBLICATION. PLEASE NOTE THAT THE PADDING 612 ATTRIBUTE ENTRY IS REPLACING THE ENTRY MADE BY RFC5780 613 (EXPERIMENTAL). THE SAME VALUE AND NAME ARE USED BUT THE REFERENCE 614 SHOULD BE CHANGED TO THIS STANDARDS TRACK DOCUMENT. 616 9. References 618 9.1. Normative References 620 [ITU.V42.2002] 621 International Telecommunications Union, "Error-correcting 622 Procedures for DCEs Using Asynchronous-to-Synchronous 623 Conversion", ITU-T Recommendation V.42, 2002. 625 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 626 Requirement Levels", BCP 14, RFC 2119, 627 DOI 10.17487/RFC2119, March 1997, 628 . 630 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 631 Discovery", RFC 4821, DOI 10.17487/RFC4821, March 2007, 632 . 634 [RFC5389] Rosenberg, J., Mahy, R., Matthews, P., and D. Wing, 635 "Session Traversal Utilities for NAT (STUN)", RFC 5389, 636 DOI 10.17487/RFC5389, October 2008, 637 . 639 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 640 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 641 May 2017, . 643 9.2. Informative References 645 [I-D.martinsen-tram-stuntrace] 646 Martinsen, P. and D. Wing, "STUN Traceroute", draft- 647 martinsen-tram-stuntrace-01 (work in progress), June 2015. 649 [I-D.martinsen-tram-turnbandwidthprobe] 650 Martinsen, P., Andersen, T., Salgueiro, G., and M. Petit- 651 Huguenin, "Traversal Using Relays around NAT (TURN) 652 Bandwidth Probe", draft-martinsen-tram- 653 turnbandwidthprobe-00 (work in progress), May 2015. 655 [RFC3264] Rosenberg, J. and H. Schulzrinne, "An Offer/Answer Model 656 with Session Description Protocol (SDP)", RFC 3264, 657 DOI 10.17487/RFC3264, June 2002, 658 . 660 [RFC5766] Mahy, R., Matthews, P., and J. Rosenberg, "Traversal Using 661 Relays around NAT (TURN): Relay Extensions to Session 662 Traversal Utilities for NAT (STUN)", RFC 5766, 663 DOI 10.17487/RFC5766, April 2010, 664 . 666 Appendix A. Release Notes 668 This section must be removed before publication as an RFC. 670 A.1. Modifications between draft-ietf-tram-stun-pmtud-13 and draft- 671 ietf-tram-stun-pmtud-12 673 o Modifications to address nits 675 A.2. Modifications between draft-ietf-tram-stun-pmtud-12 and draft- 676 ietf-tram-stun-pmtud-11 678 o Modifications following IESG review. Incorporated RFC5780 PADDING 679 attribute (Adam's Discuss) and added IPv6 language (Suresh's 680 Discuss). 682 A.3. Modifications between draft-ietf-tram-stun-pmtud-11 and draft- 683 ietf-tram-stun-pmtud-10 685 o Modifications following IESG review. 687 A.4. Modifications between draft-ietf-tram-stun-pmtud-10 and draft- 688 ietf-tram-stun-pmtud-09 690 o Modifications following reviews for gen-art (Roni Even) and secdir 691 (Carl Wallace). 693 A.5. Modifications between draft-ietf-tram-stun-pmtud-09 and draft- 694 ietf-tram-stun-pmtud-08 696 o Add 3 ways of preventing amplification attacks. 698 A.6. Modifications between draft-ietf-tram-stun-pmtud-08 and draft- 699 ietf-tram-stun-pmtud-07 701 o Updates following Spencer's review. 703 A.7. Modifications between draft-ietf-tram-stun-pmtud-07 and draft- 704 ietf-tram-stun-pmtud-06 706 o Updates following Shepherd review. 708 A.8. Modifications between draft-ietf-tram-stun-pmtud-06 and draft- 709 ietf-tram-stun-pmtud-05 711 o Nits. 713 o Restore missing changelog for previous version. 715 A.9. Modifications between draft-ietf-tram-stun-pmtud-05 and draft- 716 ietf-tram-stun-pmtud-04 718 o Modifications following Brandon Williams review. 720 A.10. Modifications between draft-ietf-tram-stun-pmtud-04 and draft- 721 ietf-tram-stun-pmtud-03 723 o Modifications following Simon Perreault and Brandon Williams 724 reviews. 726 A.11. Modifications between draft-ietf-tram-stun-pmtud-03 and draft- 727 ietf-tram-stun-pmtud-02 729 o Add new Overview of Operations section with ladder diagrams. 731 o Authentication is mandatory for the Complete Probing mechanism, 732 optional for the Simple Probing mechanism. 734 o All the ICE specific text moves to a separate draft to be 735 discussed in the ICE WG. 737 o The TURN usage is removed because probing between a TURN server 738 and TURN client is not useful. 740 o Any usage of PMTUD-SUPPORTED or other signaling mechanisms 741 (formerly knows as discovery mechanisms) must now be 742 authenticated. 744 o Both probing mechanisms are MTI in the server, the complete 745 probing mechanism is MTI in the client. 747 o Make clear that stopping after 3 retransmission is done by 748 changing the STUN parameter. 750 o Define the format of the attributes. 752 o Make clear that the specification is for any UDP protocol that 753 does not already have PMTUD capabilities, not just STUN based 754 protocols. 756 o Change the default delay to send the Report Request to 250 ms 757 after the last Indication if the RTO is unknown. 759 o Each usage of this specification must the format of the 760 IDENTIFIERS attribute contents. 762 o Better define the implicit signaling mechanism. 764 o Extend the Security Consideration section. 766 o Tons of nits. 768 A.12. Modifications between draft-ietf-tram-stun-pmtud-02 and draft- 769 ietf-tram-stun-pmtud-01 771 o Cleaned up references. 773 A.13. Modifications between draft-ietf-tram-stun-pmtud-01 and draft- 774 ietf-tram-stun-pmtud-00 776 o Added Security Considerations Section. 778 o Added IANA Considerations Section. 780 A.14. Modifications between draft-ietf-tram-stun-pmtud-00 and draft- 781 petithuguenin-tram-stun-pmtud-01 783 o Adopted by WG - Text unchanged. 785 A.15. Modifications between draft-petithuguenin-tram-stun-pmtud-01 and 786 draft-petithuguenin-tram-stun-pmtud-00 788 o Moved some Introduction text to the Probing Mechanism section. 790 o Added cross-reference to the other two STUN troubleshooting 791 mechanism drafts. 793 o Updated references. 795 o Added Gonzalo Salgueiro as co-author. 797 A.16. Modifications between draft-petithuguenin-tram-stun-pmtud-00 and 798 draft-petithuguenin-behave-stun-pmtud-03 800 o General refresh for republication. 802 A.17. Modifications between draft-petithuguenin-behave-stun-pmtud-03 803 and draft-petithuguenin-behave-stun-pmtud-02 805 o Changed author address. 807 o Changed the IPR to trust200902. 809 A.18. Modifications between draft-petithuguenin-behave-stun-pmtud-02 810 and draft-petithuguenin-behave-stun-pmtud-01 812 o Defined checksum and sequential numbers as possible packet 813 identifiers. 815 o Updated the reference to RFC 5389 817 o The FINGERPRINT attribute is now mandatory. 819 o Changed the delay between Probe indication and Report request to 820 be RTO/2 or 50 milliseconds. 822 o Added ICMP packet processing. 824 o Added Full-Stop Timeout detection. 826 o Stated that Binding request with PMTUD-SUPPORTED does not start 827 the PMTUD process if already started. 829 A.19. Modifications between draft-petithuguenin-behave-stun-pmtud-01 830 and draft-petithuguenin-behave-stun-pmtud-00 832 o Removed the use of modified STUN transaction but shorten the 833 retransmission for the simple probing mechanism. 835 o Added a complete probing mechanism. 837 o Removed the PADDING-RECEIVED attribute. 839 o Added release notes. 841 Acknowledgements 843 Thanks to Eilon Yardeni, Geir Sandbakken, Paal-Erik Martinsen, 844 Tirumaleswar Reddy, Ram Mohan R, Simon Perreault, Brandon Williams, 845 Tolga Asveren, Spencer Dawkins, Carl Wallace, and Roni Even for their 846 review comments, suggestions and questions that helped to improve 847 this document. 849 Special thanks to Dan Wing, who supported this document since its 850 first publication back in 2008. 852 Authors' Addresses 854 Marc Petit-Huguenin 855 Impedance Mismatch 857 Email: marc@petit-huguenin.org 859 Gonzalo Salgueiro 860 Cisco Systems, Inc. 861 7200-12 Kit Creek Road 862 Research Triangle Park, NC 27709 863 United States 865 Email: gsalguei@cisco.com 867 Felipe Garrido 868 Cisco Systems, Inc. 869 7200-12 Kit Creek Road 870 Research Triangle Park, NC 27709 871 United States 873 Email: fegarrid@cisco.com