idnits 2.17.1 draft-ietf-ipsecme-iptfs-09.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 == Line 1295 has weird spacing: '...4 any any...' == Line 1311 has weird spacing: '...4 any any...' -- The document date (July 5, 2021) is 1027 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) == Missing Reference: '--800--' is mentioned on line 1113, but not defined -- Looks like a reference, but probably isn't: '60' on line 1113 == Missing Reference: '-240-' is mentioned on line 1113, but not defined == Missing Reference: '--4000----------------------' is mentioned on line 1113, but not defined -- Obsolete informational reference (is this intentional?): RFC 8229 (Obsoleted by RFC 9329) Summary: 0 errors (**), 0 flaws (~~), 6 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group C. Hopps 3 Internet-Draft LabN Consulting, L.L.C. 4 Intended status: Standards Track July 5, 2021 5 Expires: January 6, 2022 7 IP-TFS: Aggregation and Fragmentation Mode for ESP and its Use for IP 8 Traffic Flow Security 9 draft-ietf-ipsecme-iptfs-09 11 Abstract 13 This document describes a mechanism for aggregation and fragmentation 14 of IP packets when they are being encapsulated in ESP payload. This 15 new payload type can be used for various purposes such as decreasing 16 encapsulation overhead for small IP packets; however, the focus in 17 this document is to enhance IPsec traffic flow security (IP-TFS) by 18 adding Traffic Flow Confidentiality (TFC) to encrypted IP 19 encapsulated traffic. TFC is provided by obscuring the size and 20 frequency of IP traffic using a fixed-sized, constant-send-rate IPsec 21 tunnel. The solution allows for congestion control as well as non- 22 constant send-rate usage. 24 Status of This Memo 26 This Internet-Draft is submitted in full conformance with the 27 provisions of BCP 78 and BCP 79. 29 Internet-Drafts are working documents of the Internet Engineering 30 Task Force (IETF). Note that other groups may also distribute 31 working documents as Internet-Drafts. The list of current Internet- 32 Drafts is at https://datatracker.ietf.org/drafts/current/. 34 Internet-Drafts are draft documents valid for a maximum of six months 35 and may be updated, replaced, or obsoleted by other documents at any 36 time. It is inappropriate to use Internet-Drafts as reference 37 material or to cite them other than as "work in progress." 39 This Internet-Draft will expire on January 6, 2022. 41 Copyright Notice 43 Copyright (c) 2021 IETF Trust and the persons identified as the 44 document authors. All rights reserved. 46 This document is subject to BCP 78 and the IETF Trust's Legal 47 Provisions Relating to IETF Documents 48 (https://trustee.ietf.org/license-info) in effect on the date of 49 publication of this document. Please review these documents 50 carefully, as they describe your rights and restrictions with respect 51 to this document. Code Components extracted from this document must 52 include Simplified BSD License text as described in Section 4.e of 53 the Trust Legal Provisions and are provided without warranty as 54 described in the Simplified BSD License. 56 Table of Contents 58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 59 1.1. Terminology & Concepts . . . . . . . . . . . . . . . . . 4 60 2. The AGGFRAG Tunnel . . . . . . . . . . . . . . . . . . . . . 4 61 2.1. Tunnel Content . . . . . . . . . . . . . . . . . . . . . 4 62 2.2. Payload Content . . . . . . . . . . . . . . . . . . . . . 5 63 2.2.1. Data Blocks . . . . . . . . . . . . . . . . . . . . . 6 64 2.2.2. End Padding . . . . . . . . . . . . . . . . . . . . . 6 65 2.2.3. Fragmentation, Sequence Numbers and All-Pad Payloads 6 66 2.2.4. Empty Payload . . . . . . . . . . . . . . . . . . . . 8 67 2.2.5. IP Header Value Mapping . . . . . . . . . . . . . . . 8 68 2.2.6. IP Time-To-Live (TTL) and Tunnel errors . . . . . . . 9 69 2.2.7. Effective MTU of the Tunnel . . . . . . . . . . . . . 9 70 2.3. Exclusive SA Use . . . . . . . . . . . . . . . . . . . . 9 71 2.4. Modes of Operation . . . . . . . . . . . . . . . . . . . 10 72 2.4.1. Non-Congestion Controlled Mode . . . . . . . . . . . 10 73 2.4.2. Congestion Controlled Mode . . . . . . . . . . . . . 10 74 2.5. Summary of Receiver Processing . . . . . . . . . . . . . 12 75 3. Congestion Information . . . . . . . . . . . . . . . . . . . 12 76 3.1. ECN Support . . . . . . . . . . . . . . . . . . . . . . . 13 77 4. Configuration of AGGFRAG Tunnels for IP-TFS . . . . . . . . . 14 78 4.1. Bandwidth . . . . . . . . . . . . . . . . . . . . . . . . 14 79 4.2. Fixed Packet Size . . . . . . . . . . . . . . . . . . . . 14 80 4.3. Congestion Control . . . . . . . . . . . . . . . . . . . 14 81 5. IKEv2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 82 5.1. USE_AGGFRAG Notification Message . . . . . . . . . . . . 14 83 6. Packet and Data Formats . . . . . . . . . . . . . . . . . . . 15 84 6.1. AGGFRAG_PAYLOAD Payload . . . . . . . . . . . . . . . . . 15 85 6.1.1. Non-Congestion Control AGGFRAG_PAYLOAD Payload Format 16 86 6.1.2. Congestion Control AGGFRAG_PAYLOAD Payload Format . . 16 87 6.1.3. Data Blocks . . . . . . . . . . . . . . . . . . . . . 18 88 6.1.4. IKEv2 USE_AGGFRAG Notification Message . . . . . . . 20 89 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 21 90 7.1. AGGFRAG_PAYLOAD Sub-Type Registry . . . . . . . . . . . . 21 91 7.2. USE_AGGFRAG Notify Message Status Type . . . . . . . . . 21 92 8. Security Considerations . . . . . . . . . . . . . . . . . . . 21 93 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 22 94 9.1. Normative References . . . . . . . . . . . . . . . . . . 22 95 9.2. Informative References . . . . . . . . . . . . . . . . . 22 96 Appendix A. Example Of An Encapsulated IP Packet Flow . . . . . 24 97 Appendix B. A Send and Loss Event Rate Calculation . . . . . . . 25 98 Appendix C. Comparisons of IP-TFS . . . . . . . . . . . . . . . 25 99 C.1. Comparing Overhead . . . . . . . . . . . . . . . . . . . 25 100 C.1.1. IP-TFS Overhead . . . . . . . . . . . . . . . . . . . 26 101 C.1.2. ESP with Padding Overhead . . . . . . . . . . . . . . 26 102 C.2. Overhead Comparison . . . . . . . . . . . . . . . . . . . 27 103 C.3. Comparing Available Bandwidth . . . . . . . . . . . . . . 28 104 C.3.1. Ethernet . . . . . . . . . . . . . . . . . . . . . . 28 105 Appendix D. Acknowledgements . . . . . . . . . . . . . . . . . . 30 106 Appendix E. Contributors . . . . . . . . . . . . . . . . . . . . 30 107 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 30 109 1. Introduction 111 Traffic Analysis ([RFC4301], [AppCrypt]) is the act of extracting 112 information about data being sent through a network. While directly 113 obscuring the data with encryption [RFC4303], the traffic pattern 114 itself exposes information due to variations in its shape and timing 115 ([RFC8546], [AppCrypt]). Hiding the size and frequency of traffic is 116 referred to as Traffic Flow Confidentiality (TFC) per [RFC4303]. 118 [RFC4303] provides for TFC by allowing padding to be added to 119 encrypted IP packets and allowing for transmission of all-pad packets 120 (indicated using protocol 59). This method has the major limitation 121 that it can significantly under-utilize the available bandwidth. 123 This document defines an aggregation and fragmentation (AGGFRAG) mode 124 for ESP, and its use for IP Traffic Flow Security (IP-TFS). This 125 solution provides for full TFC without the aforementioned bandwidth 126 limitation. This is accomplished by using a constant-send-rate IPsec 127 [RFC4303] tunnel with fixed-sized encapsulating packets; however, 128 these fixed-sized packets can contain partial, whole or multiple IP 129 packets to maximize the bandwidth of the tunnel. A non-constant 130 send-rate is allowed, but the confidentiality properties of its use 131 are outside the scope of this document. 133 For a comparison of the overhead of IP-TFS with the RFC4303 134 prescribed TFC solution see Appendix C. 136 Additionally, IP-TFS provides for operating fairly within congested 137 networks [RFC2914]. This is important for when the IP-TFS user is 138 not in full control of the domain through which the IP-TFS tunnel 139 path flows. 141 The mechanisms, such as the AGGFRAG mode, defined in this document 142 are generic with the intent of allowing for non-TFS uses, but such 143 uses are outside the scope of this document. 145 1.1. Terminology & Concepts 147 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 148 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 149 "OPTIONAL" in this document are to be interpreted as described in BCP 150 14 [RFC2119] [RFC8174] when, and only when, they appear in all 151 capitals, as shown here. 153 This document assumes familiarity with IP security concepts including 154 TFC as described in [RFC4301]. 156 2. The AGGFRAG Tunnel 158 As mentioned in Section 1, AGGFRAG mode utilizes an IPsec [RFC4303] 159 tunnel as its transport. For the purpose of IP-TFS, fixed-sized 160 encapsulating packets are sent at a constant rate on the AGGFRAG 161 tunnel. 163 The primary input to the tunnel algorithm is the requested bandwidth 164 to be used by the tunnel. Two values are then required to provide 165 for this bandwidth use, the fixed size of the encapsulating packets, 166 and rate at which to send them. 168 The fixed packet size MAY either be specified manually or be 169 determined through other methods such as the Packetization Layer MTU 170 Discovery (PLMTUD) ([RFC4821], [RFC8899]) or Path MTU discovery 171 (PMTUD) ([RFC1191], [RFC8201]). PMTUD is known to have issues so 172 PLMTUD is considered the more robust option. For PLMTUD, congestion 173 control payloads can be used as in-band probes (see Section 6.1.2 and 174 [RFC8899]). 176 Given the encapsulating packet size and the requested bandwidth to be 177 used, the corresponding packet send rate can be calculated. The 178 packet send rate is the requested bandwidth to be used divided by the 179 size of the encapsulating packet. 181 The egress (receiving) side of the AGGFRAG tunnel MUST allow for and 182 expect the ingress (sending) side of the AGGFRAG tunnel to vary the 183 size and rate of sent encapsulating packets, unless constrained by 184 other policy. 186 2.1. Tunnel Content 188 As previously mentioned, one issue with the TFC padding solution in 189 [RFC4303] is the large amount of wasted bandwidth as only one IP 190 packet can be sent per encapsulating packet. In order to maximize 191 bandwidth, IP-TFS breaks this one-to-one association by introducing 192 an AGGFRAG mode for ESP. 194 AGGFRAG mode aggregates as well as fragments the inner IP traffic 195 flow into encapsulating IPsec tunnel packets. For IP-TFS, the IPsec 196 encapsulating tunnel packets are a fixed size. Padding is only added 197 to the the tunnel packets if there is no data available to be sent at 198 the time of tunnel packet transmission, or if fragmentation has been 199 disabled by the receiver. 201 This is accomplished using a new Encapsulating Security Payload (ESP, 202 [RFC4303]) Next Header field value AGGFRAG_PAYLOAD (Section 6.1). 204 Other non-IP-TFS uses of this AGGFRAG mode have been suggested, such 205 as increased performance through packet aggregation, as well as 206 handling MTU issues using fragmentation. These uses are not defined 207 here, but are also not restricted by this document. 209 2.2. Payload Content 211 The AGGFRAG_PAYLOAD payload content defined in this document is 212 comprised of a 4 or 24 octet header followed by either a partial 213 datablock, a full datablock, or multiple partial or full datablocks. 214 The following diagram illustrates this payload within the ESP packet. 215 See Section 6.1 for the exact formats of the AGGFRAG_PAYLOAD payload. 217 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 . Outer Encapsulating Header ... . 219 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 . ESP Header... . 221 +---------------------------------------------------------------+ 222 | [AGGFRAG subtype/flags] : BlockOffset | 223 +---------------------------------------------------------------+ 224 : [Optional Congestion Info] : 225 +---------------------------------------------------------------+ 226 | DataBlocks ... ~ 227 ~ ~ 228 ~ | 229 +---------------------------------------------------------------| 230 . ESP Trailer... . 231 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Figure 1: Layout of an AGGFRAG mode IPsec Packet 235 The "BlockOffset" value is either zero or some offset into or past 236 the end of the "DataBlocks" data. 238 If the "BlockOffset" value is zero it means that the "DataBlocks" 239 data begins with a new data block. 241 Conversely, if the "BlockOffset" value is non-zero it points to the 242 start of the new data block, and the initial "DataBlocks" data 243 belongs to the data block that is still being re-assembled. 245 If the "BlockOffset" points past the end of the "DataBlocks" data 246 then the next data block occurs in a subsequent encapsulating packet. 248 Having the "BlockOffset" always point at the next available data 249 block allows for recovering the next inner packet in the presence of 250 outer encapsulating packet loss. 252 An example AGGFRAG mode packet flow can be found in Appendix A. 254 2.2.1. Data Blocks 256 +---------------------------------------------------------------+ 257 | Type | rest of IPv4, IPv6 or pad. 258 +-------- 260 Figure 2: Layout of a DataBlock 262 A data block is defined by a 4-bit type code followed by the data 263 block data. The type values have been carefully chosen to coincide 264 with the IPv4/IPv6 version field values so that no per-data block 265 type overhead is required to encapsulate an IP packet. Likewise, the 266 length of the data block is extracted from the encapsulated IPv4's 267 "Total Length" or IPv6's "Payload Length" fields. 269 2.2.2. End Padding 271 Since a data block's type is identified in its first 4-bits, the only 272 time padding is required is when there is no data to encapsulate. 273 For this end padding a "Pad Data Block" is used. 275 2.2.3. Fragmentation, Sequence Numbers and All-Pad Payloads 277 In order for a receiver to reassemble fragmented inner-packets, the 278 sender MUST send the inner-packet fragments back-to-back in the 279 logical outer packet stream (i.e., using consecutive ESP sequence 280 numbers). However, the sender is allowed to insert "all-pad" 281 payloads (i.e., payloads with a "BlockOffset" of zero and a single 282 pad "DataBlock") in between the packets carrying the inner-packet 283 fragment payloads. This interleaving of all-pad payloads allows the 284 sender to always send a tunnel packet, regardless of the 285 encapsulation computational requirements. 287 When a receiver is reassembling an inner-packet, and it receives an 288 "all-pad" payload, it increments the expected sequence number that 289 the next inner-packet fragment is expected to arrive in. 291 Given the above, the receiver will need to handle out-of-order 292 arrival of outer ESP packets prior to reassembly processing. ESP 293 already provides for optionally detecting replay attacks. Detecting 294 replay attacks normally utilizes a window method. A similar sequence 295 number based sliding window can be used to correct re-ordering of the 296 outer packet stream. Receiving a larger (newer) sequence number 297 packet advances the window, and received older ESP packets whose 298 sequence numbers the window has passed by are dropped. A good choice 299 for the size of this window depends on the amount of re-ordering the 300 user may normally experience. 302 As the amount of reordering that may be present is hard to predict, 303 the window size SHOULD be configurable by the user. Implementations 304 MAY also dynamically adjust the reordering window based on actual 305 reordering seen in arriving packets. Finally, note that as IP-TFS is 306 sending a continuous stream of packets there is no requirement for 307 timers (although there's no prohibition either) as newly arrived 308 packets will cause the window to advance and older packets will then 309 be processed as they leave the window. Implementations that are 310 concerned about memory use when packets are delayed (e.g., when an SA 311 deletion is delayed), or non-IP-TFS uses of AGGFRAG mode, can of 312 course use timers to drop packets as well. 314 While ESP guarantees an increasing sequence number with subsequently 315 sent packets, it does not actually require the sequence numbers to be 316 generated with no gaps (e.g., sending only even numbered sequence 317 numbers would be allowed as long as they are always increasing). 318 Gaps in the sequence numbers will not work for this document so the 319 sequence number stream MUST increase monotonically by 1 for each 320 subsequent packet. 322 When using the AGGFRAG_PAYLOAD in conjunction with replay detection, 323 the window size for both MAY be reduced to the smaller of the two 324 window sizes. This is because packets outside of the smaller window 325 but inside the larger would still be dropped by the mechanism with 326 the smaller window size. However, there is also no requirement to 327 make these values the same. Indeed, in some cases, such as slow 328 tunnels where a very small or zero reorder window size is 329 appropriate, the user may want a large replay detection window to log 330 replayed packets. Additionally, large replay windows can be 331 implemented with very little overhead compared to large reorder 332 windows. 334 Finally, as sequence numbers are reset when switching SAs (e.g., when 335 re-keying a child SA), senders MUST NOT send initial fragments of an 336 inner packet using one SA and subsequent fragments in a different SA. 338 2.2.3.1. Optional Extra Padding 340 When the tunnel bandwidth is not being fully utilized, a sender MAY 341 pad-out the current encapsulating packet in order to deliver an inner 342 packet un-fragmented in the following outer packet. The benefit 343 would be to avoid inner-packet fragmentation in the presence of a 344 bursty offered load (non-bursty traffic will naturally not fragment). 345 Senders MAY also choose to allow for a minimum fragment size to be 346 configured (e.g., as a percentage of the AGGFRAG_PAYLOAD payload 347 size) to avoid fragmentation at the cost of tunnel bandwidth. The 348 cost with these methods is complexity and added delay of inner 349 traffic. The main advantage to avoiding fragmentation is to minimize 350 inner packet loss in the presence of outer packet loss. When this is 351 worthwhile (e.g., how much loss and what type of loss is required, 352 given different inner traffic shapes and utilization, for this to 353 make sense), and what values to use for the allowable/added delay may 354 be worth researching, but is outside the scope of this document. 356 While use of padding to avoid fragmentation does not impact 357 interoperability, used inappropriately it can reduce the effective 358 throughput of a tunnel. Senders implementing either of the above 359 approaches will need to take care to not reduce the effective 360 capacity, and overall utility, of the tunnel through the overuse of 361 padding. 363 2.2.4. Empty Payload 365 To support reporting of congestion control information (described 366 later) using a non-AGGFRAG_PAYLOAD enabled SA, it is allowed to send 367 an AGGFRAG_PAYLOAD payload with no data blocks (i.e., the ESP payload 368 length is equal to the AGGFRAG_PAYLOAD header length). This special 369 payload is called an empty payload. 371 Currently this situation is only applicable in non-IKEv2 use cases. 373 2.2.5. IP Header Value Mapping 375 [RFC4301] provides some direction on when and how to map various 376 values from an inner IP header to the outer encapsulating header, 377 namely the Don't-Fragment (DF) bit ([RFC0791] and [RFC8200]), the 378 Differentiated Services (DS) field [RFC2474] and the Explicit 379 Congestion Notification (ECN) field [RFC3168]. Unlike [RFC4301], 380 AGGFRAG mode may and often will be encapsulating more than one IP 381 packet per ESP packet. To deal with this, these mappings are 382 restricted further. 384 2.2.5.1. DF bit 386 AGGFRAG mode never maps the inner DF bit as it is unrelated to the 387 AGGFRAG tunnel functionality; AGGFRAG mode never needs to IP fragment 388 the inner packets and the inner packets will not affect the 389 fragmentation of the outer encapsulation packets. 391 2.2.5.2. ECN value 393 The ECN value need not be mapped as any congestion related to the 394 constant-send-rate IP-TFS tunnel is unrelated (by design) to the 395 inner traffic flow. The sender MAY still set the ECN value of inner 396 packets based on the normal ECN specification [RFC3168]. 398 2.2.5.3. DS field 400 By default the DS field SHOULD NOT be copied, although a sender MAY 401 choose to allow for configuration to override this behavior. A 402 sender SHOULD also allow the DS value to be set by configuration. 404 2.2.6. IP Time-To-Live (TTL) and Tunnel errors 406 [RFC4301] specifies how to modify the inner packet TTL [RFC0791]. 408 Any errors (e.g., ICMP errors arriving back at the tunnel ingress due 409 to tunnel traffic) are handled the same as with non-AGGFRAG IPsec 410 tunnels. 412 2.2.7. Effective MTU of the Tunnel 414 Unlike [RFC4301], there is normally no effective MTU (EMTU) on an 415 AGGFRAG tunnel as all IP packet sizes are properly transmitted 416 without requiring IP fragmentation prior to tunnel ingress. That 417 said, a sender MAY allow for explicitly configuring an MTU for the 418 tunnel. 420 If fragmentation has been disabled on the AGGFRAG tunnel, then the 421 tunnel's EMTU and behaviors are the same as normal IPsec tunnels 422 [RFC4301]. 424 2.3. Exclusive SA Use 426 This document does not specify mixed use of an AGGFRAG_PAYLOAD 427 enabled SA. A sender MUST only send AGGFRAG_PAYLOAD payloads over an 428 SA configured for AGGFRAG mode. 430 2.4. Modes of Operation 432 Just as with normal IPsec/ESP tunnels, AGGFRAG tunnels are 433 unidirectional. Bidirectional IP-TFS functionality is achieved by 434 setting up 2 AGGFRAG tunnels, one in either direction. 436 An AGGFRAG tunnel used for IP-TFS can operate in 2 modes, a non- 437 congestion controlled mode and congestion controlled mode. 439 2.4.1. Non-Congestion Controlled Mode 441 In the non-congestion controlled mode, IP-TFS sends fixed-sized 442 packets over an AGGFRAG tunnel at a constant rate. The packet send 443 rate is constant and is not automatically adjusted regardless of any 444 network congestion (e.g., packet loss). 446 For similar reasons as given in [RFC7510] the non-congestion 447 controlled mode should only be used where the user has full 448 administrative control over the path the tunnel will take. This is 449 required so the user can guarantee the bandwidth and also be sure as 450 to not be negatively affecting network congestion [RFC2914]. In this 451 case packet loss should be reported to the administrator (e.g., via 452 syslog, YANG notification, SNMP traps, etc) so that any failures due 453 to a lack of bandwidth can be corrected. 455 Non-congestion control mode is also appropriate if ESP over TCP is in 456 use [RFC8229]. 458 2.4.2. Congestion Controlled Mode 460 With the congestion controlled mode, IP-TFS adapts to network 461 congestion by lowering the packet send rate to accommodate the 462 congestion, as well as raising the rate when congestion subsides. 463 Since overhead is per packet, by allowing for maximal fixed-size 464 packets and varying the send rate transport overhead is minimized. 466 The output of the congestion control algorithm will adjust the rate 467 at which the ingress sends packets. While this document does not 468 require a specific congestion control algorithm, best current 469 practice RECOMMENDS that the algorithm conform to [RFC5348]. 470 Congestion control principles are documented in [RFC2914] as well. 471 [RFC4342] provides an example of the [RFC5348] algorithm which 472 matches the requirements of IP-TFS (i.e., designed for fixed-size 473 packet and send rate varied based on congestion. 475 The required inputs for the TCP friendly rate control algorithm 476 described in [RFC5348] are the receiver's loss event rate and the 477 sender's estimated round-trip time (RTT). These values are provided 478 by IP-TFS using the congestion information header fields described in 479 Section 3. In particular, these values are sufficient to implement 480 the algorithm described in [RFC5348]. 482 At a minimum, the congestion information MUST be sent, from the 483 receiver and from the sender, at least once per RTT. Prior to 484 establishing an RTT the information SHOULD be sent constantly from 485 the sender and the receiver so that an RTT estimate can be 486 established. Not receiving this information over multiple 487 consecutive RTT intervals should be considered a congestion event 488 that causes the sender to adjust its sending rate lower. For 489 example, [RFC4342] calls this the "no feedback timeout" and it is 490 equal to 4 RTT intervals. When a "no feedback timeout" has occurred 491 [RFC4342] halves the sending rate. 493 An implementation MAY choose to always include the congestion 494 information in its AGGFRAG payload header if sending on an IP-TFS 495 enabled SA. Since IP-TFS normally will operate with a large packet 496 size, the congestion information should represent a small portion of 497 the available tunnel bandwidth. An implementation choosing to always 498 send the data MAY also choose to only update the "LossEventRate" and 499 "RTT" header field values it sends every "RTT" though. 501 When choosing a congestion control algorithm (or a selection of 502 algorithms) note that IP-TFS is not providing for reliable delivery 503 of IP traffic, and so per packet ACKs are not required and are not 504 provided. 506 It is worth noting that the variable send-rate of a congestion 507 controlled AGGFRAG tunnel, is not private; however, this send-rate is 508 being driven by network congestion, and as long as the encapsulated 509 (inner) traffic flow shape and timing are not directly affecting the 510 (outer) network congestion, the variations in the tunnel rate will 511 not weaken the provided inner traffic flow confidentiality. 513 2.4.2.1. Circuit Breakers 515 In additional to congestion control, implementations MAY choose to 516 define and implement circuit breakers [RFC8084] as a recovery method 517 of last resort. Enabling circuit breakers is also a reason a user 518 may wish to enable congestion information reports even when using the 519 non-congestion controlled mode of operation. The definition of 520 circuit breakers are outside the scope of this document. 522 2.5. Summary of Receiver Processing 524 An AGGFRAG enabled SA receiver has a few tasks to perform. 526 The receiver first reorders, possibly out-of-order ESP packets 527 received on an SA into in-sequence-order AGGFRAG_PAYLOAD payloads 528 (Section 2.2.3). If congestion control is enabled, the receiver 529 considers a packet lost when it's sequence number is abandoned (e.g., 530 pushed out of the re-ordering window, or timed-out) by the reordering 531 algorithm. 533 Additionally, if congestion control is enabled, the receiver sends 534 congestion control data (Section 6.1.2) back to the sender as 535 described in Section 2.4.2 and Section 3. 537 Finally, the receiver processes the now in-order AGGFRAG_PAYLOAD 538 payload stream to extract the inner-packets (Section 2.2.3, 539 Section 6.1). 541 3. Congestion Information 543 In order to support the congestion control mode, the sender needs to 544 know the loss event rate and to approximate the RTT [RFC5348]. In 545 order to obtain these values, the receiver sends congestion control 546 information on it's SA back to the sender. Thus, to support 547 congestion control the receiver must have a paired SA back to the 548 sender (this is always the case when the tunnel was created using 549 IKEv2). If the SA back to the sender is a non-AGGFRAG_PAYLOAD 550 enabled SA then an AGGFRAG_PAYLOAD empty payload (i.e., header only) 551 is used to convey the information. 553 In order to calculate a loss event rate compatible with [RFC5348], 554 the receiver needs to have a round-trip time estimate. Thus the 555 sender communicates this estimate in the "RTT" header field. On 556 startup this value will be zero as no RTT estimate is yet known. 558 In order for the sender to estimate its "RTT" value, the sender 559 places a timestamp value in the "TVal" header field. On first 560 receipt of this "TVal", the receiver records the new "TVal" value 561 along with the time it arrived locally, subsequent receipt of the 562 same "TVal" MUST NOT update the recorded time. 564 When the receiver sends its CC header it places this latest recorded 565 "TVal" in the "TEcho" header field, along with 2 delay values, "Echo 566 Delay" and "Transmit Delay". The "Echo Delay" value is the time 567 delta from the recorded arrival time of "TVal" and the current clock 568 in microseconds. The second value, "Transmit Delay", is the 569 receiver's current transmission delay on the tunnel (i.e., the 570 average time between sending packets on its half of the AGGFRAG 571 tunnel). 573 When the sender receives back its "TVal" in the "TEcho" header field 574 it calculates 2 RTT estimates. The first is the actual delay found 575 by subtracting the "TEcho" value from its current clock and then 576 subtracting "Echo Delay" as well. The second RTT estimate is found 577 by adding the received "Transmit Delay" header value to the senders 578 own transmission delay (i.e., the average time between sending 579 packets on its half of the AGGFRAG tunnel). The larger of these 2 580 RTT estimates SHOULD be used as the "RTT" value. 582 The two RTT estimates are required to handle different combinations 583 of faster or slower tunnel packet paths with faster or slower fixed 584 tunnel rates. Choosing the larger of the two values guarantees that 585 the "RTT" is never considered faster than the aggregate transmission 586 delay based on the IP-TFS send rate (the second estimate), as well as 587 never being considered faster than the actual RTT along the tunnel 588 packet path (the first estimate). 590 The receiver also calculates, and communicates in the "LossEventRate" 591 header field, the loss event rate for use by the sender. This is 592 slightly different from [RFC4342] which periodically sends all the 593 loss interval data back to the sender so that it can do the 594 calculation. See Appendix B for a suggested way to calculate the 595 loss event rate value. Initially this value will be zero (indicating 596 no loss) until enough data has been collected by the receiver to 597 update it. 599 3.1. ECN Support 601 In additional to normal packet loss information AGGFRAG mode supports 602 use of the ECN bits in the encapsulating IP header [RFC3168] for 603 identifying congestion. If ECN use is enabled and a packet arrives 604 at the egress (receiving) side with the Congestion Experienced (CE) 605 value set, then the receiver considers that packet as being dropped, 606 although it does not drop it. The receiver MUST set the E bit in any 607 AGGFRAG_PAYLOAD payload header containing a "LossEventRate" value 608 derived from a CE value being considered. 610 As noted in [RFC3168] the ECN bits are not protected by IPsec and 611 thus may constitute a covert channel. For this reason, ECN use 612 SHOULD NOT be enabled by default. 614 4. Configuration of AGGFRAG Tunnels for IP-TFS 616 IP-TFS is meant to be deployable with a minimal amount of 617 configuration. All IP-TFS specific configuration should be specified 618 at the unidirectional tunnel ingress (sending) side. It is intended 619 that non-IKEv2 operation is supported, at least, with local static 620 configuration. 622 4.1. Bandwidth 624 Bandwidth is a local configuration option. For non-congestion 625 controlled mode, the bandwidth SHOULD be configured. For congestion 626 controlled mode, the bandwidth can be configured or the congestion 627 control algorithm discovers and uses the maximum bandwidth available. 628 No standardized configuration method is required. 630 4.2. Fixed Packet Size 632 The fixed packet size to be used for the tunnel encapsulation packets 633 MAY be configured manually or can be automatically determined using 634 other methods such as PLMTUD ([RFC4821], [RFC8899]) or PMTUD 635 ([RFC1191], [RFC8201]). As PMTUD is known to have issues, PLMTUD is 636 considered the more robust option. No standardized configuration 637 method is required. 639 4.3. Congestion Control 641 Congestion control is a local configuration option. No standardized 642 configuration method is required. 644 5. IKEv2 646 5.1. USE_AGGFRAG Notification Message 648 As mentioned previously AGGFRAG tunnels utilize ESP payloads of type 649 AGGFRAG_PAYLOAD. 651 When using IKEv2, a new "USE_AGGFRAG" Notification Message enables 652 the AGGFRAG_PAYLOAD payload on a child SA pair. The method used is 653 similar to how USE_TRANSPORT_MODE is negotiated, as described in 654 [RFC7296]. 656 To request use of the AGGFRAG_PAYLOAD payload on the Child SA pair, 657 the initiator includes the USE_AGGFRAG notification in an SA payload 658 requesting a new Child SA (either during the initial IKE_AUTH or 659 during CREATE_CHILD_SA exchanges). If the request is accepted then 660 the response MUST also include a notification of type USE_AGGFRAG. 661 If the responder declines the request the child SA will be 662 established without AGGFRAG_PAYLOAD payload use enabled. If this is 663 unacceptable to the initiator, the initiator MUST delete the child 664 SA. 666 As the use of the AGGFRAG_PAYLOAD payload is currently only defined 667 for non-transport mode tunnels, the USE_AGGFRAG notification MUST NOT 668 be combined with USE_TRANSPORT notification. 670 The USE_AGGFRAG notification contains a 1 octet payload of flags that 671 specify requirements from the sender of the notification. If any 672 requirement flags are not understood or cannot be supported by the 673 receiver then the receiver SHOULD NOT enable use of AGGFRAG_PAYLOAD 674 (either by not responding with the USE_AGGFRAG notification, or in 675 the case of the initiator, by deleting the child SA if the now 676 established non-AGGFRAG_PAYLOAD using SA is unacceptable). 678 The notification type and payload flag values are defined in 679 Section 6.1.4. 681 6. Packet and Data Formats 683 The packet and data formats defined below are generic with the intent 684 of allowing for non-IP-TFS uses, but such uses are outside the scope 685 of this document. 687 6.1. AGGFRAG_PAYLOAD Payload 689 ESP Next Header value: 0x5 691 An AGGFRAG payload is identified by the ESP Next Header value 692 AGGFRAG_PAYLOAD which has the value 0x5. The value 5 was chosen to 693 not conflict with other used values. The first octet of this payload 694 indicates the format of the remaining payload data. 696 0 1 2 3 4 5 6 7 697 +-+-+-+-+-+-+-+-+-+-+- 698 | Sub-type | ... 699 +-+-+-+-+-+-+-+-+-+-+- 701 Sub-type: 702 An 8-bit value indicating the payload format. 704 This document defines 2 payload sub-types. These payload formats are 705 defined in the following sections. 707 6.1.1. Non-Congestion Control AGGFRAG_PAYLOAD Payload Format 709 The non-congestion control AGGFRAG_PAYLOAD payload is comprised of a 710 4 octet header followed by a variable amount of "DataBlocks" data as 711 shown below. 713 1 2 3 714 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 1 715 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 716 | Sub-Type (0) | Reserved | BlockOffset | 717 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 718 | DataBlocks ... 719 +-+-+-+-+-+-+-+-+-+-+- 721 Sub-type: 722 An octet indicating the payload format. For this non-congestion 723 control format, the value is 0. 725 Reserved: 726 An octet set to 0 on generation, and ignored on receipt. 728 BlockOffset: 729 A 16-bit unsigned integer counting the number of octets of 730 "DataBlocks" data before the start of a new data block. If the 731 start of a new data block occurs in a subsequent payload the 732 "BlockOffset" will point past the end of the "DataBlocks" data. 733 In this case all the "DataBlocks" data belongs to the current data 734 block being assembled. When the "BlockOffset" extends into 735 subsequent payloads it continues to only count "DataBlocks" data 736 (i.e., it does not count subsequent packets non-"DataBlocks" data 737 such as header octets). 739 DataBlocks: 740 Variable number of octets that begins with the start of a data 741 block, or the continuation of a previous data block, followed by 742 zero or more additional data blocks. 744 6.1.2. Congestion Control AGGFRAG_PAYLOAD Payload Format 746 The congestion control AGGFRAG_PAYLOAD payload is comprised of a 24 747 octet header followed by a variable amount of "DataBlocks" data as 748 shown below. 750 1 2 3 751 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 1 752 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 753 | Sub-type (1) | Reserved |P|E| BlockOffset | 754 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 755 | LossEventRate | 756 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 757 | RTT | Echo Delay ... 758 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 759 ... Echo Delay | Transmit Delay | 760 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 761 | TVal | 762 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 763 | TEcho | 764 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 765 | DataBlocks ... 766 +-+-+-+-+-+-+-+-+-+-+- 768 Sub-type: 769 An octet indicating the payload format. For this congestion 770 control format, the value is 1. 772 Reserved: 773 A 6-bit field set to 0 on generation, and ignored on receipt. 775 P: 776 A 1-bit value if set indicates that PLMTUD probing is in progress. 777 This information can be used to avoid treating missing packets as 778 loss events by the CC algorithm when running the PLMTUD probe 779 algorithm. 781 E: 782 A 1-bit value if set indicates that Congestion Experienced (CE) 783 ECN bits were received and used in deriving the reported 784 "LossEventRate". 786 BlockOffset: 787 The same value as the non-congestion controlled payload format 788 value. 790 LossEventRate: 791 A 32-bit value specifying the inverse of the current loss event 792 rate as calculated by the receiver. A value of zero indicates no 793 loss. Otherwise the loss event rate is "1/LossEventRate". 795 RTT: 796 A 22-bit value specifying the sender's current round-trip time 797 estimate in microseconds. The value MAY be zero prior to the 798 sender having calculated a round-trip time estimate. The value 799 SHOULD be set to zero on non-AGGFRAG_PAYLOAD enabled SAs. If the 800 value is equal to or larger than "0x3FFFFF" it MUST be set to 801 "0x3FFFFF". 803 Echo Delay: 804 A 21-bit value specifying the delay in microseconds incurred 805 between the receiver first receiving the "TVal" value which it is 806 sending back in "TEcho". If the value is equal to or larger than 807 "0x1FFFFF" it MUST be set to "0x1FFFFF". 809 Transmit Delay: 810 A 21-bit value specifying the transmission delay in microseconds. 811 This is the fixed (or average) delay on the receiver between it 812 sending packets on the IPTFS tunnel. If the value is equal to or 813 larger than "0x1FFFFF" it MUST be set to "0x1FFFFF". 815 TVal: 816 An opaque 32-bit value that will be echoed back by the receiver in 817 later packets in the "TEcho" field, along with an "Echo Delay" 818 value of how long that echo took. 820 TEcho: 821 The opaque 32-bit value from a received packet's "TVal" field. 822 The received "TVal" is placed in "TEcho" along with an "Echo 823 Delay" value indicating how long it has been since receiving the 824 "TVal" value. 826 DataBlocks: 827 Variable number of octets that begins with the start of a data 828 block, or the continuation of a previous data block, followed by 829 zero or more additional data blocks. For the special case of 830 sending congestion control information on an non-IP-TFS enabled SA 831 this value MUST be empty (i.e., be zero octets long). 833 6.1.3. Data Blocks 835 1 2 3 836 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 1 837 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 838 | Type | IPv4, IPv6 or pad... 839 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 841 Type: 842 A 4-bit field where 0x0 identifies a pad data block, 0x4 indicates 843 an IPv4 data block, and 0x6 indicates an IPv6 data block. 845 6.1.3.1. IPv4 Data Block 847 1 2 3 848 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 1 849 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 850 | 0x4 | IHL | TypeOfService | TotalLength | 851 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 852 | Rest of the inner packet ... 853 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 855 These values are the actual values within the encapsulated IPv4 856 header. In other words, the start of this data block is the start of 857 the encapsulated IP packet. 859 Type: 860 A 4-bit value of 0x4 indicating IPv4 (i.e., first nibble of the 861 IPv4 packet). 863 TotalLength: 864 The 16-bit unsigned integer "Total Length" field of the IPv4 inner 865 packet. 867 6.1.3.2. IPv6 Data Block 869 1 2 3 870 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 1 871 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 872 | 0x6 | TrafficClass | FlowLabel | 873 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 874 | PayloadLength | Rest of the inner packet ... 875 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 877 These values are the actual values within the encapsulated IPv6 878 header. In other words, the start of this data block is the start of 879 the encapsulated IP packet. 881 Type: 882 A 4-bit value of 0x6 indicating IPv6 (i.e., first nibble of the 883 IPv6 packet). 885 PayloadLength: 886 The 16-bit unsigned integer "Payload Length" field of the inner 887 IPv6 inner packet. 889 6.1.3.3. Pad Data Block 891 1 2 3 892 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 1 893 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 894 | 0x0 | Padding ... 895 +-+-+-+-+-+-+-+-+-+-+- 897 Type: 898 A 4-bit value of 0x0 indicating a padding data block. 900 Padding: 901 Extends to end of the encapsulating packet. 903 6.1.4. IKEv2 USE_AGGFRAG Notification Message 905 As discussed in Section 5.1, a notification message USE_AGGFRAG is 906 used to negotiate use of the ESP AGGFRAG_PAYLOAD Next Header value. 908 The USE_AGGFRAG Notification Message State Type is (TBD2). 910 The notification payload contains 1 octet of requirement flags. 911 There are currently 2 requirement flags defined. This may be revised 912 by later specifications. 914 +-+-+-+-+-+-+-+-+ 915 |0|0|0|0|0|0|C|D| 916 +-+-+-+-+-+-+-+-+ 918 0: 919 6 bits - reserved, MUST be zero on send, unless defined by later 920 specifications. 922 C: 923 Congestion Control bit. If set, then the sender is requiring that 924 congestion control information MUST be returned to it periodically 925 as defined in Section 3. 927 D: 928 Don't Fragment bit. If set, indicates the sender of the notify 929 message does not support receiving packet fragments (i.e., inner 930 packets MUST be sent using a single "Data Block"). This value 931 only applies to what the sender is capable of receiving; the 932 sender MAY still send packet fragments unless similarly restricted 933 by the receiver in it's USE_AGGFRAG notification. 935 7. IANA Considerations 937 7.1. AGGFRAG_PAYLOAD Sub-Type Registry 939 This document requests IANA create a registry called "AGGFRAG_PAYLOAD 940 Sub-Type Registry" under a new category named "ESP AGGFRAG_PAYLOAD 941 Parameters". The registration policy for this registry is "Expert 942 Review" ([RFC8126] and [RFC7120]). 944 Name: 945 AGGFRAG_PAYLOAD Sub-Type Registry 947 Description: 948 AGGFRAG_PAYLOAD Payload Formats. 950 Reference: 951 This document 953 This initial content for this registry is as follows: 955 Sub-Type Name Reference 956 -------------------------------------------------------- 957 0 Non-Congestion Control Format This document 958 1 Congestion Control Format This document 959 3-255 Reserved 961 7.2. USE_AGGFRAG Notify Message Status Type 963 This document requests a status type USE_AGGFRAG be allocated from 964 the "IKEv2 Notify Message Types - Status Types" registry. 966 Value: 967 TBD2 969 Name: 970 USE_AGGFRAG 972 Reference: 973 This document 975 8. Security Considerations 977 This document describes an aggregation and fragmentation mechanism 978 and it use to add TFC to IP traffic. The use described is expected 979 to increase the security of the traffic being transported. Other 980 than the additional security afforded by using this mechanism, IP-TFS 981 utilizes the security protocols [RFC4303] and [RFC7296] and so their 982 security considerations apply to IP-TFS as well. 984 As noted in (Section 3.1) the ECN bits are not protected by IPsec and 985 thus may constitute a covert channel. For this reason, ECN use 986 SHOULD NOT be enabled by default. 988 As noted previously in Section 2.4.2, for TFC to be fully maintained 989 the encapsulated traffic flow should not be affecting network 990 congestion in a predictable way, and if it would be then non- 991 congestion controlled mode use should be considered instead. 993 9. References 995 9.1. Normative References 997 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 998 Requirement Levels", BCP 14, RFC 2119, 999 DOI 10.17487/RFC2119, March 1997, 1000 . 1002 [RFC4303] Kent, S., "IP Encapsulating Security Payload (ESP)", 1003 RFC 4303, DOI 10.17487/RFC4303, December 2005, 1004 . 1006 [RFC7296] Kaufman, C., Hoffman, P., Nir, Y., Eronen, P., and T. 1007 Kivinen, "Internet Key Exchange Protocol Version 2 1008 (IKEv2)", STD 79, RFC 7296, DOI 10.17487/RFC7296, October 1009 2014, . 1011 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 1012 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 1013 May 2017, . 1015 9.2. Informative References 1017 [AppCrypt] 1018 Schneier, B., "Applied Cryptography: Protocols, 1019 Algorithms, and Source Code in C", 11 2017. 1021 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 1022 DOI 10.17487/RFC0791, September 1981, 1023 . 1025 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 1026 DOI 10.17487/RFC1191, November 1990, 1027 . 1029 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 1030 "Definition of the Differentiated Services Field (DS 1031 Field) in the IPv4 and IPv6 Headers", RFC 2474, 1032 DOI 10.17487/RFC2474, December 1998, 1033 . 1035 [RFC2914] Floyd, S., "Congestion Control Principles", BCP 41, 1036 RFC 2914, DOI 10.17487/RFC2914, September 2000, 1037 . 1039 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 1040 of Explicit Congestion Notification (ECN) to IP", 1041 RFC 3168, DOI 10.17487/RFC3168, September 2001, 1042 . 1044 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 1045 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 1046 December 2005, . 1048 [RFC4342] Floyd, S., Kohler, E., and J. Padhye, "Profile for 1049 Datagram Congestion Control Protocol (DCCP) Congestion 1050 Control ID 3: TCP-Friendly Rate Control (TFRC)", RFC 4342, 1051 DOI 10.17487/RFC4342, March 2006, 1052 . 1054 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 1055 Discovery", RFC 4821, DOI 10.17487/RFC4821, March 2007, 1056 . 1058 [RFC5348] Floyd, S., Handley, M., Padhye, J., and J. Widmer, "TCP 1059 Friendly Rate Control (TFRC): Protocol Specification", 1060 RFC 5348, DOI 10.17487/RFC5348, September 2008, 1061 . 1063 [RFC7120] Cotton, M., "Early IANA Allocation of Standards Track Code 1064 Points", BCP 100, RFC 7120, DOI 10.17487/RFC7120, January 1065 2014, . 1067 [RFC7510] Xu, X., Sheth, N., Yong, L., Callon, R., and D. Black, 1068 "Encapsulating MPLS in UDP", RFC 7510, 1069 DOI 10.17487/RFC7510, April 2015, 1070 . 1072 [RFC8084] Fairhurst, G., "Network Transport Circuit Breakers", 1073 BCP 208, RFC 8084, DOI 10.17487/RFC8084, March 2017, 1074 . 1076 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 1077 Writing an IANA Considerations Section in RFCs", BCP 26, 1078 RFC 8126, DOI 10.17487/RFC8126, June 2017, 1079 . 1081 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 1082 (IPv6) Specification", STD 86, RFC 8200, 1083 DOI 10.17487/RFC8200, July 2017, 1084 . 1086 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 1087 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 1088 DOI 10.17487/RFC8201, July 2017, 1089 . 1091 [RFC8229] Pauly, T., Touati, S., and R. Mantha, "TCP Encapsulation 1092 of IKE and IPsec Packets", RFC 8229, DOI 10.17487/RFC8229, 1093 August 2017, . 1095 [RFC8546] Trammell, B. and M. Kuehlewind, "The Wire Image of a 1096 Network Protocol", RFC 8546, DOI 10.17487/RFC8546, April 1097 2019, . 1099 [RFC8899] Fairhurst, G., Jones, T., Tuexen, M., Ruengeler, I., and 1100 T. Voelker, "Packetization Layer Path MTU Discovery for 1101 Datagram Transports", RFC 8899, DOI 10.17487/RFC8899, 1102 September 2020, . 1104 Appendix A. Example Of An Encapsulated IP Packet Flow 1106 Below an example inner IP packet flow within the encapsulating tunnel 1107 packet stream is shown. Notice how encapsulated IP packets can start 1108 and end anywhere, and more than one or less than 1 may occur in a 1109 single encapsulating packet. 1111 Offset: 0 Offset: 100 Offset: 2900 Offset: 1400 1112 [ ESP1 (1500) ][ ESP2 (1500) ][ ESP3 (1500) ][ ESP4 (1500) ] 1113 [--800--][--800--][60][-240-][--4000----------------------][pad] 1115 Figure 3: Inner and Outer Packet Flow 1117 The encapsulated IP packet flow (lengths include IP header and 1118 payload) is as follows: an 800 octet packet, an 800 octet packet, a 1119 60 octet packet, a 240 octet packet, a 4000 octet packet. 1121 The "BlockOffset" values in the 4 AGGFRAG payload headers for this 1122 packet flow would thus be: 0, 100, 2900, 1400 respectively. The 1123 first encapsulating packet ESP1 has a zero "BlockOffset" which points 1124 at the IP data block immediately following the AGGFRAG header. The 1125 following packet ESP2s "BlockOffset" points inward 100 octets to the 1126 start of the 60 octet data block. The third encapsulating packet 1127 ESP3 contains the middle portion of the 4000 octet data block so the 1128 offset points past its end and into the forth encapsulating packet. 1129 The fourth packet ESP4s offset is 1400 pointing at the padding which 1130 follows the completion of the continued 4000 octet packet. 1132 Appendix B. A Send and Loss Event Rate Calculation 1134 The current best practice indicates that congestion control SHOULD be 1135 done in a TCP friendly way. A TCP friendly congestion control 1136 algorithm is described in [RFC5348]. For this IP-TFS use case (as 1137 with [RFC4342]) the (fixed) packet size is used as the segment size 1138 for the algorithm. The main formula in the algorithm for the send 1139 rate is then as follows: 1141 1 1142 X = ----------------------------------------------- 1143 R * (sqrt(2*p/3) + 12*sqrt(3*p/8)*p*(1+32*p^2)) 1145 Where "X" is the send rate in packets per second, "R" is the round 1146 trip time estimate and "p" is the loss event rate (the inverse of 1147 which is provided by the receiver). 1149 In addition the algorithm in [RFC5348] also uses an "X_recv" value 1150 (the receiver's receive rate). For IP-TFS one MAY set this value 1151 according to the sender's current tunnel send-rate ("X"). 1153 The IP-TFS receiver, having the RTT estimate from the sender can use 1154 the same method as described in [RFC5348] and [RFC4342] to collect 1155 the loss intervals and calculate the loss event rate value using the 1156 weighted average as indicated. The receiver communicates the inverse 1157 of this value back to the sender in the AGGFRAG_PAYLOAD payload 1158 header field "LossEventRate". 1160 The IP-TFS sender now has both the "R" and "p" values and can 1161 calculate the correct sending rate. If following [RFC5348] the 1162 sender should also use the slow start mechanism described therein 1163 when the IP-TFS SA is first established. 1165 Appendix C. Comparisons of IP-TFS 1167 C.1. Comparing Overhead 1169 For comparing overhead the overhead of ESP for both normal and 1170 AGGFRAG tunnel packets must be calculated, and so an algorithm for 1171 encryption and authentication must be chosen. For the data below 1172 AES-GCM-256 was selected. This leads to an IP+ESP overhead of 54. 1174 54 = 20 (IP) + 8 (ESPH) + 2 (ESPF) + 8 (IV) + 16 (ICV) 1176 Additionally, for IP-TFS, non-congestion control AGGFRAG_PAYLOAD 1177 headers were chosen which adds 4 octets for a total overhead of 58. 1179 C.1.1. IP-TFS Overhead 1181 For comparison the overhead of AGGFRAG payload is 58 octets per outer 1182 packet. Therefore the octet overhead per inner packet is 58 divided 1183 by the number of outer packets required (fractional allowed). The 1184 overhead as a percentage of inner packet size is a constant based on 1185 the Outer MTU size. 1187 OH = 58 / Outer Payload Size / Inner Packet Size 1188 OH % of Inner Packet Size = 100 * OH / Inner Packet Size 1189 OH % of Inner Packet Size = 5800 / Outer Payload Size 1191 Type IP-TFS IP-TFS IP-TFS 1192 MTU 576 1500 9000 1193 PSize 518 1442 8942 1194 ------------------------------- 1195 40 11.20% 4.02% 0.65% 1196 576 11.20% 4.02% 0.65% 1197 1500 11.20% 4.02% 0.65% 1198 9000 11.20% 4.02% 0.65% 1200 Figure 4: IP-TFS Overhead as Percentage of Inner Packet Size 1202 C.1.2. ESP with Padding Overhead 1204 The overhead per inner packet for constant-send-rate padded ESP 1205 (i.e., traditional IPsec TFC) is 36 octets plus any padding, unless 1206 fragmentation is required. 1208 When fragmentation of the inner packet is required to fit in the 1209 outer IPsec packet, overhead is the number of outer packets required 1210 to carry the fragmented inner packet times both the inner IP overhead 1211 (20) and the outer packet overhead (54) minus the initial inner IP 1212 overhead plus any required tail padding in the last encapsulation 1213 packet. The required tail padding is the number of required packets 1214 times the difference of the Outer Payload Size and the IP Overhead 1215 minus the Inner Payload Size. So: 1217 Inner Paylaod Size = IP Packet Size - IP Overhead 1218 Outer Payload Size = MTU - IPsec Overhead 1220 Inner Payload Size 1221 NF0 = ---------------------------------- 1222 Outer Payload Size - IP Overhead 1224 NF = CEILING(NF0) 1226 OH = NF * (IP Overhead + IPsec Overhead) 1227 - IP Overhead 1228 + NF * (Outer Payload Size - IP Overhead) 1229 - Inner Payload Size 1231 OH = NF * (IPsec Overhead + Outer Payload Size) 1232 - (IP Overhead + Inner Payload Size) 1234 OH = NF * (IPsec Overhead + Outer Payload Size) 1235 - Inner Packet Size 1237 C.2. Overhead Comparison 1239 The following tables collect the overhead values for some common L3 1240 MTU sizes in order to compare them. The first table is the number of 1241 octets of overhead for a given L3 MTU sized packet. The second table 1242 is the percentage of overhead in the same MTU sized packet. 1244 XXX rerun these. 1246 Type ESP+Pad ESP+Pad ESP+Pad IP-TFS IP-TFS IP-TFS 1247 L3 MTU 576 1500 9000 576 1500 9000 1248 PSize 522 1446 8946 518 1442 8942 1249 ----------------------------------------------------------- 1250 40 482 1406 8906 4.5 1.6 0.3 1251 128 394 1318 8818 14.3 5.1 0.8 1252 256 266 1190 8690 28.7 10.3 1.7 1253 518 4 928 8428 58.0 20.8 3.4 1254 576 576 870 8370 64.5 23.2 3.7 1255 1442 286 4 7504 161.5 58.0 9.4 1256 1500 228 1500 7446 168.0 60.3 9.7 1257 8942 1426 1558 4 1001.2 359.7 58.0 1258 9000 1368 1500 9000 1007.7 362.0 58.4 1260 Figure 5: Overhead comparison in octets 1262 Type ESP+Pad ESP+Pad ESP+Pad IP-TFS IP-TFS IP-TFS 1263 MTU 576 1500 9000 576 1500 9000 1264 PSize 522 1446 8946 518 1442 8942 1265 ----------------------------------------------------------- 1266 40 1205.0% 3515.0% 22265.0% 11.20% 4.02% 0.65% 1267 128 307.8% 1029.7% 6889.1% 11.20% 4.02% 0.65% 1268 256 103.9% 464.8% 3394.5% 11.20% 4.02% 0.65% 1269 518 0.8% 179.2% 1627.0% 11.20% 4.02% 0.65% 1270 576 100.0% 151.0% 1453.1% 11.20% 4.02% 0.65% 1271 1442 19.8% 0.3% 520.4% 11.20% 4.02% 0.65% 1272 1500 15.2% 100.0% 496.4% 11.20% 4.02% 0.65% 1273 8942 15.9% 17.4% 0.0% 11.20% 4.02% 0.65% 1274 9000 15.2% 16.7% 100.0% 11.20% 4.02% 0.65% 1276 Figure 6: Overhead as Percentage of Inner Packet Size 1278 C.3. Comparing Available Bandwidth 1280 Another way to compare the two solutions is to look at the amount of 1281 available bandwidth each solution provides. The following sections 1282 consider and compare the percentage of available bandwidth. For the 1283 sake of providing a well understood baseline normal (unencrypted) 1284 Ethernet as well as normal ESP values are included. 1286 C.3.1. Ethernet 1288 In order to calculate the available bandwidth the per packet overhead 1289 is calculated first. The total overhead of Ethernet is 14+4 octets 1290 of header and CRC plus and additional 20 octets of framing (preamble, 1291 start, and inter-packet gap) for a total of 38 octets. Additionally 1292 the minimum payload is 46 octets. 1294 Size E + P E + P E + P IPTFS IPTFS IPTFS Enet ESP 1295 MTU 590 1514 9014 590 1514 9014 any any 1296 OH 92 92 92 96 96 96 38 74 1297 ------------------------------------------------------------ 1298 40 614 1538 9038 47 42 40 84 114 1299 128 614 1538 9038 151 136 129 166 202 1300 256 614 1538 9038 303 273 258 294 330 1301 518 614 1538 9038 614 552 523 574 610 1302 576 1228 1538 9038 682 614 582 614 650 1303 1442 1842 1538 9038 1709 1538 1457 1498 1534 1304 1500 1842 3076 9038 1777 1599 1516 1538 1574 1305 8942 11052 10766 9038 10599 9537 9038 8998 9034 1306 9000 11052 10766 18076 10667 9599 9096 9038 9074 1308 Figure 7: L2 Octets Per Packet 1310 Size E + P E + P E + P IPTFS IPTFS IPTFS Enet ESP 1311 MTU 590 1514 9014 590 1514 9014 any any 1312 OH 92 92 92 96 96 96 38 74 1313 -------------------------------------------------------------- 1314 40 2.0M 0.8M 0.1M 26.4M 29.3M 30.9M 14.9M 11.0M 1315 128 2.0M 0.8M 0.1M 8.2M 9.2M 9.7M 7.5M 6.2M 1316 256 2.0M 0.8M 0.1M 4.1M 4.6M 4.8M 4.3M 3.8M 1317 518 2.0M 0.8M 0.1M 2.0M 2.3M 2.4M 2.2M 2.1M 1318 576 1.0M 0.8M 0.1M 1.8M 2.0M 2.1M 2.0M 1.9M 1319 1442 678K 812K 138K 731K 812K 857K 844K 824K 1320 1500 678K 406K 138K 703K 781K 824K 812K 794K 1321 8942 113K 116K 138K 117K 131K 138K 139K 138K 1322 9000 113K 116K 69K 117K 130K 137K 138K 137K 1324 Figure 8: Packets Per Second on 10G Ethernet 1326 Size E + P E + P E + P IPTFS IPTFS IPTFS Enet ESP 1327 590 1514 9014 590 1514 9014 any any 1328 92 92 92 96 96 96 38 74 1329 ---------------------------------------------------------------------- 1330 40 6.51% 2.60% 0.44% 84.36% 93.76% 98.94% 47.62% 35.09% 1331 128 20.85% 8.32% 1.42% 84.36% 93.76% 98.94% 77.11% 63.37% 1332 256 41.69% 16.64% 2.83% 84.36% 93.76% 98.94% 87.07% 77.58% 1333 518 84.36% 33.68% 5.73% 84.36% 93.76% 98.94% 93.17% 87.50% 1334 576 46.91% 37.45% 6.37% 84.36% 93.76% 98.94% 93.81% 88.62% 1335 1442 78.28% 93.76% 15.95% 84.36% 93.76% 98.94% 97.43% 95.12% 1336 1500 81.43% 48.76% 16.60% 84.36% 93.76% 98.94% 97.53% 95.30% 1337 8942 80.91% 83.06% 98.94% 84.36% 93.76% 98.94% 99.58% 99.18% 1338 9000 81.43% 83.60% 49.79% 84.36% 93.76% 98.94% 99.58% 99.18% 1340 Figure 9: Percentage of Bandwidth on 10G Ethernet 1342 A sometimes unexpected result of using an AGGFRAG tunnel (or any 1343 packet aggregating tunnel) is that, for small to medium sized 1344 packets, the available bandwidth is actually greater than native 1345 Ethernet. This is due to the reduction in Ethernet framing overhead. 1346 This increased bandwidth is paid for with an increase in latency. 1347 This latency is the time to send the unrelated octets in the outer 1348 tunnel frame. The following table illustrates the latency for some 1349 common values on a 10G Ethernet link. The table also includes 1350 latency introduced by padding if using ESP with padding. 1352 ESP+Pad ESP+Pad IP-TFS IP-TFS 1353 1500 9000 1500 9000 1355 ------------------------------------------ 1356 40 1.12 us 7.12 us 1.17 us 7.17 us 1357 128 1.05 us 7.05 us 1.10 us 7.10 us 1358 256 0.95 us 6.95 us 1.00 us 7.00 us 1359 518 0.74 us 6.74 us 0.79 us 6.79 us 1360 576 0.70 us 6.70 us 0.74 us 6.74 us 1361 1442 0.00 us 6.00 us 0.05 us 6.05 us 1362 1500 1.20 us 5.96 us 0.00 us 6.00 us 1364 Figure 10: Added Latency 1366 Notice that the latency values are very similar between the two 1367 solutions; however, whereas IP-TFS provides for constant high 1368 bandwidth, in some cases even exceeding native Ethernet, ESP with 1369 padding often greatly reduces available bandwidth. 1371 Appendix D. Acknowledgements 1373 We would like to thank Don Fedyk for help in reviewing and editing 1374 this work. We would also like to thank Michael Richardson, Sean 1375 Turner and Valery Smyslov for reviews and many suggestions for 1376 improvements, as well as Joseph Touch for the transport area review 1377 and suggested improvements. 1379 Appendix E. Contributors 1381 The following people made significant contributions to this document. 1383 Lou Berger 1384 LabN Consulting, L.L.C. 1386 Email: lberger@labn.net 1388 Author's Address 1390 Christian Hopps 1391 LabN Consulting, L.L.C. 1393 Email: chopps@chopps.org