idnits 2.17.1 draft-templin-6man-fragrep-04.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 : ---------------------------------------------------------------------------- == The 'Updates: ' line in the draft header should list only the _numbers_ of the RFCs which will be updated by this document (if approved); it should not include the word 'RFC' in the list. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document doesn't use any RFC 2119 keywords, yet seems to have RFC 2119 boilerplate text. -- The document date (21 December 2021) is 828 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to lower-maturity documents in RFCs) == Missing Reference: 'RFCXXXX' is mentioned on line 418, but not defined == Outdated reference: A later version (-63) exists of draft-templin-6man-aero-37 == Outdated reference: A later version (-74) exists of draft-templin-6man-omni-51 Summary: 0 errors (**), 0 flaws (~~), 6 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group F. L. Templin, Ed. 3 Internet-Draft Boeing Research & Technology 4 Updates: RFC8200, RFC8201, RFC4443, RFC1191 (if 21 December 2021 5 approved) 6 Intended status: Standards Track 7 Expires: 24 June 2022 9 IPv6 Fragment Retransmission and Path MTU Discovery Soft Errors 10 draft-templin-6man-fragrep-04 12 Abstract 14 Internet Protocol version 6 (IPv6) provides a fragmentation and 15 reassembly service for end systems allowing for the transmission of 16 packets that exceed the path MTU. However, loss of individual 17 fragments requires retransmission of original packets in their 18 entirety leading to cascading reassembly failures. This document 19 specifies an IPv6 fragment retransmission scheme that matches the 20 loss unit to the retransmission unit. The document further specifies 21 an update to Path MTU Discovery that distinguishes hard link size 22 restrictions from reassembly congestion events. 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 24 June 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 (https://trustee.ietf.org/ 48 license-info) in effect on the date of publication of this document. 49 Please review these documents carefully, as they describe your rights 50 and restrictions with respect to this document. Code Components 51 extracted from this document must include Revised BSD License text as 52 described in Section 4.e of the Trust Legal Provisions and are 53 provided without warranty as described in the Revised BSD License. 55 Table of Contents 57 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 58 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 59 3. Common Use Cases . . . . . . . . . . . . . . . . . . . . . . 4 60 4. IPv6 Fragmentation . . . . . . . . . . . . . . . . . . . . . 4 61 5. IPv6 Fragment Retransmission . . . . . . . . . . . . . . . . 5 62 6. Packet Too Big (PTB) Soft Errors . . . . . . . . . . . . . . 8 63 7. Implementation Status . . . . . . . . . . . . . . . . . . . . 9 64 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9 65 9. Security Considerations . . . . . . . . . . . . . . . . . . . 10 66 10. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 10 67 11. References . . . . . . . . . . . . . . . . . . . . . . . . . 10 68 11.1. Normative References . . . . . . . . . . . . . . . . . . 10 69 11.2. Informative References . . . . . . . . . . . . . . . . . 11 70 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 11 72 1. Introduction 74 Internet Protocol version 6 (IPv6) [RFC8200] provides a fragmentation 75 and reassembly service similar to that found in IPv4 [RFC0791], with 76 the exception that only the source host (i.e., and not routers on the 77 path) may perform fragmentation. When an IPv6 packet is fragmented, 78 the loss unit (i.e., a single IPv6 fragment) becomes smaller than the 79 retransmission unit (i.e., the entire packet) which even under 80 moderate loss conditions could result in cascading reassembly 81 failures that degrade forward progress [RFC8900]. 83 The presumed drawbacks of fragmentation are tempered by the fact that 84 performance increases can often be realized when the source sends 85 packets larger than the path MTU. This is due to the fact that 86 larger packets result in fewer application system calls, plus 87 transmission of a single large packet results in a burst of multiple 88 IPv6 fragments separated by minimal inter-packet delays. These 89 bursts yield high network utilization for the burst duration, while 90 modern reassembly implementations have proven capable of 91 accommodating the bursts. If the loss unit can somehow be made to 92 match the retransmission unit, the performance benefits of IPv6 93 fragmentation can be realized. 95 This document therefore proposes an IPv6 fragment retransmission 96 service where the source marks fragments as retransmission-eligible 97 while the destination may request retransmission of lost fragments. 98 The service provides opportunistic best-effort retransmissions over 99 an imaginary "link" extending from the source to the destination 100 consistent with the Automatic Repeat Request (ARQ) function of common 101 data links [RFC3366]. The service does not attempt to replace true 102 end-to-end reliability, but instead allows the destination to recover 103 missing individual fragments of partial reassemblies before true end- 104 to-end timers would cause retransmission of the entire packet. 106 The original packet source may be either co-located with or many IP 107 network hops before the IPv6 fragmentation source. In the same 108 fashion, the IPv6 reassembly destination may be either co-located 109 with or many IP network hops before the final destination. When 110 conditions suggest that an original source should begin sending 111 smaller packets, the fragmentation source and/or reassembly 112 destination can return a new type of ICMPv6/ICMPv4 Packet Too Big 113 (PTB) message termed a PTB "soft error". 115 PTB "soft errors" are distinguished from classic "hard errors" by a 116 non-zero PTB Code (ICMPv6) or unused (ICMPv4) field value. The 117 fragmentation source can return rate-limited soft errors to recommend 118 smaller packet sizes to the original source while fragmentation of 119 large packets is producing excessive numbers of fragments. 120 Similarly, the reassembly destination can return rate-limited soft 121 errors (i.e., via the fragmentation source to the original source) 122 while reassembly of large packets is causing excessive reassembly 123 congestion. Original sources that receive these soft errors should 124 reduce their packet sizes until the errors subside, but can begin to 125 increase packet sizes again without delay until further soft or hard 126 errors arrive. 128 The following sections discuss common use cases and operational 129 considerations for applying IPv6 fragment retransmission and path MTU 130 discovery soft errors. They further specify new codings for the IPv6 131 fragment header Reserved field, a new ICMPv6 message type and updates 132 to ICMPv6/ICMPv4 PTB messages. This document therefore updates 133 existing standards where necessary. 135 2. Terminology 137 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 138 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 139 "OPTIONAL" in this document are to be interpreted as described in BCP 140 14 [RFC2119][RFC8174] when, and only when, they appear in all 141 capitals, as shown here. 143 3. Common Use Cases 145 A common use case of interest is to improve the state of affairs for 146 IPv6 encapsulation (i.e., "tunneling") [RFC2473] when the original 147 source may be many IP hops away from the tunnel ingress, and the 148 tunnel packet may be fragmented following encapsulation. The tunnel 149 is seen as a "link" on the path from the original source to the final 150 destination, and the goal is to increase the reliability of that link 151 in order to minimize wasteful end-to-end retransmissions. 153 When the original source and IPv6 fragmentation source are co-located 154 on the same platform (physical or virtual) the window of opportunity 155 for successful retransmission of individual fragments may be narrow 156 unless the link persistence timeframe is carefully coordinated with 157 upper layer retransmission timers. (In an uncoordinated case, upper 158 layers may retransmit the entire packet before or at roughly the same 159 time the IPv6 fragmentation source retransmits individual fragments, 160 leading to increased congestion and wasted retransmissions.) 161 However, the same retransmission facility can be applied to both the 162 tunneled and end system source models. 164 Upper layer protocols of the original source can further assign a 165 "Parcel ID" to groups of packets eligible for delivery to final 166 destination applications as a larger aggregate instead of smaller 167 individual packets. The upper layer protocols supply the Parcel ID 168 to lower layers which insert the value as discussed in Section 4, 169 while the destination lower layer protocols deliver the Parcel ID to 170 upper layers. Further details on parcel grouping are out of scope 171 for this document. 173 4. IPv6 Fragmentation 175 IPv6 fragmentation is specified in Section 4.5 of [RFC8200] and is 176 based on the IPv6 Fragment extension header formatted as shown below: 178 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 179 | Next Header | Reserved | Fragment Offset |Res|M| 180 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 181 | Identification | 182 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 184 In this format: 186 * Next Header is a 1-octet IP protocol version of the next header 187 following the Fragment Header. 189 * Reserved is a 1-octet reserved field set to 0 on transmission and 190 ignored on reception. 192 * Fragment Offset is a 13-bit field that provides the offset (in 193 8-octet units) of the data portion that follows from the beginning 194 of the packet. 196 * Res is a 2-bit field set to 0 on transmission and ignored on 197 reception. 199 * M is the "More Fragments" bit telling whether additional fragments 200 follow. 202 * Identification is a 32 bit numerical identification value for the 203 entire IPv6 packet. The value is copied into each fragment of the 204 same IPv6 packet. 206 The fragmentation and reassembly specification in [RFC8200] can be 207 considered as the standard method which adheres to the details of 208 that RFC. This document presents an enhanced method that allows for 209 retransmissions of individual fragments. 211 5. IPv6 Fragment Retransmission 213 Fragmentation implementations that follow this specification reuse 214 the (formerly) Reserved field of the IPv6 Fragment Header. For first 215 fragments (i.e., those with zero Fragment Offset) the Reserved field 216 is replaced with a 7-bit Parcel ID followed by a 1-bit A(RQ) flag as 217 shown below: 219 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 220 | Next Header | Parcel ID |A| Fragment Offset |Res|M| 221 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 222 | Identification | 223 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 225 For non-first fragments (i.e., those with non-zero Fragment Offset), 226 the Reserved field is replaced with a 7-bit "Ordinal" field followed 227 by a 1-bit A(RQ) flag as shown below: 229 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 230 | Next Header | Ordinal |A| Fragment Offset |Res|M| 231 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 232 | Identification | 233 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 235 When a source that follows this specification fragments an IPv6 236 packet it sets the first fragment Parcel ID to a value between 0 and 237 127, sets the A flag to 1 and implicitly considers the first fragment 238 as Ordinal fragment 0. The Parcel ID value 0 indicates that this 239 packet is not part of an upper layer "group", while values between 1 240 and 127 indicate membership in an upper layer protocol packet group 241 coordinated outside the scope of this specification. 243 The source then sets the Ordinal value for each successive non-first 244 fragment to a monotonically-increasing value beginning with 1, i.e., 245 it sets Ordinal to '1' for the first non-first fragment, '2' for the 246 second non-first fragment, '3' for the third non-first fragment, etc. 247 up to either Ordinal '127' or the final fragment (whichever comes 248 first) while also setting the A flag to 1. (If there are additional 249 non-first fragments beyond Ordinal '127', the source instead sets 250 their Ordinals to '0' to indicate that the fragment is not eligible 251 for retransmission.) 253 When a destination that follows this specification receives IPv6 254 fragments with the A flag set to 1, it infers that the source 255 participates in the protocol and maintains a checklist of all Ordinal 256 fragments received for a specific Identification number. (Note that 257 receipt of any IPv6 fragments with the A flag set provides an 258 implicit assertion that all lost Ordinal IPv6 fragments are also 259 eligible for retransmission.) 261 If the destination notices one or more Ordinals missing after most 262 other Ordinals for the same Identification have arrived, it can 263 prepare an ICMPv6 Fragmentation Report (FRAGREP) message [RFC4443] to 264 send back to the source. The message is formatted as follows: 266 0 1 2 3 267 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 1 268 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 269 | Type | Code | Checksum | 270 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 271 | Identification (0) | 272 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 273 ~ Ordinal Bitmap (0) (0-127) ~ 274 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 275 | Identification (1) | 276 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 277 ~ Ordinal Bitmap (1) (0-127) ~ 278 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 279 | ... | 280 | ... | 282 In this format, the destination prepares the FRAGREP message as a 283 list of 20-octet (Identification(i), Bitmap(i)) pairs. The first 4 284 octets in each pair encode the Identification value for the IPv6 285 packet that is subject of the report, while the remaining 16 octets 286 encode a 128-bit Bitmap of Ordinal fragments received for this 287 Identification. For example, if the destination receives the first 288 fragment (i.e., Ordinal number 0) plus non-first fragment Ordinals 1, 289 3, 4, 6, and 8 it sets Bitmap bits 0, 1, 3, 4, 6 and 8 to '1' and 290 sets all other bits to '0'. The destination may include as many 291 (Identification, Bitmap) pairs as necessary without causing the 292 entire message to exceed the minimum IPv6 MTU of 1280 bytes. (If 293 additional pairs are necessary, the destination may prepare and send 294 multiple messages.) 296 The destination next transmits the FRAGREP message to the IPv6 297 fragment source. When the source receives the message, it examines 298 each entry to determine the per-Identification Ordinal fragments that 299 require retransmission. For example, if the source receives a Bitmap 300 for Identification 0x12345678 with bits 0, 1, 3, 4, 6 and 8 set to 301 '1', it would retransmit Ordinal fragments (0x12345678, 2), 302 (0x12345678, 5) and (0x12345678, 7). 304 This implies that the source should retain a cache of recently 305 transmitted fragments for a time that determines "link persistence" 306 [RFC3366]. The link persistence should be at least as long as the 307 round-trip time from the fragmentation source to the reassembly 308 destination, plus an additional small delay to allow for processing 309 overhead and/or delay variance. Then, if the source receives a 310 FRAGREP message requesting retransmission of one or more Ordinals, it 311 can retransmit if it still holds the Ordinals in its cache. 312 Otherwise, the Ordinal will incur a cache miss and the original 313 source will eventually retransmit the original packet in its 314 entirety. After processing all entries in the FRAGREP, the source 315 discards the message. 317 The maximum-sized IPv6 packet that a source can submit for 318 fragmentation is 64KB, and the minimum IPv6 path MTU is 1280B. 319 Assuming the minimum IPv6 path MTU as the nominal size for non-final 320 fragments, the number of Ordinals for each IPv6 packet should 321 therefore easily fit within the available Bitmap bits when the 322 fragments are transmitted over IPv6-only network paths. However, 323 when the path may traverse one or more IPv4 networks (e.g., via 324 tunneling) the path MTU may be significantly smaller. In that case, 325 the number of IPv6 fragments needed may exceed the maximum number of 326 Ordinal retransmission candidates. 328 When the number of IPv6 fragments exceeds 128, the source assigns an 329 Ordinal value in the first 127 non-first fragments, but sets Ordinal 330 to 0 in any remaining non-first fragments then transmits all 331 fragments. When the destination receives the fragments, it may 332 return a FRAGREP to request retransmission of the first fragment and/ 333 or any missing Ordinal non-first fragments, but may not request 334 retransmission of non-first fragments with zero Ordinals for which 335 the best-effort delivery default behavior applies. However, all 336 fragments are presented equally to the reassembly cache regardless of 337 the (formerly) Reserved field settings, where the Reserved values are 338 ignored and successful reassembly is likely. 340 Finally, transmission of IPv6 fragments over IPv6-only paths can be 341 safely conducted without a fragmentation-layer integrity check since 342 IPv6 includes reassembly safeguards and a 32-bit Identification 343 value. Conversely, transmission of IPv6 fragments over IPv4-only or 344 mixed IPv6/IPv4 paths requires a fragmentation-layer integrity check 345 inserted by the source before fragmentation and verified by the 346 destination following reassembly since IPv4 provides only a 16-bit 347 Identification and no reassembly safeguards. (In cases where the 348 full path cannot be determined a priori, an integrity check should 349 always be included as specified in AERO [I-D.templin-6man-aero] and 350 OMNI [I-D.templin-6man-omni].) 352 6. Packet Too Big (PTB) Soft Errors 354 When an IPv6 fragmentation source forwards packets that produce what 355 it considers as excessive numbers fragments (e.g., 32, 48, 64, more), 356 the fragmentation source can also return PTB "soft errors" to the 357 original source (subject to rate limiting). Either the fragmentation 358 source or reassembly destination may also return PTB soft errors if 359 the frequency of retransmissions or reassembly failures exceeds 360 acceptable thresholds. 362 PTB soft errors are distinguished from ordinary "hard errors" through 363 non-zero values in the ICMPv6 "Code" [RFC8201][RFC4443] or ICMPv4 364 "unused" [RFC1191] fields. The following values are currently 365 defined: 367 * 0 - "PTB hard error" - Original sources that receive these 368 messages obey the classic Path MTU Discovery (PMTUD) 369 specifications found in [RFC8201][RFC1191]. 371 * 1 - "PTB soft error (packet lost)" - Original sources that receive 372 these messages should reduce their packet sizes while 373 retransmitting the lost packet data, but need not wait the 374 prescribed 10 minutes before attempting to again increase packet 375 sizes. 377 * 2 - "PTB soft error (packet forwarded)" - Original sources that 378 receive these messages should reduce their packet sizes without 379 invoking retransmission, and also need not wait the prescribed 10 380 minutes before attempting to again increase packet sizes. 382 * 3-255 - reserved for future use. 384 PTB soft errors include as much of the invoking packet as possible 385 without the message exceeding the minimum MTU (i.e., 1280 bytes for 386 IPv6 or 576 bytes for IPv4). Original sources that recognize PTB 387 soft errors should follow common logic to dynamically tune their 388 packet sizes to obtain the best performance. In particular, an 389 original source can gradually increase its packet sizes while PTB 390 soft errors are suppressed then again reduce packet sizes when 391 excessive soft errors arrive. 393 Original sources that do not recognize PTB soft errors (i.e., that do 394 not examine the Code/unused field value) follow the same standards as 395 for hard errors as described above and may therefore miss performance 396 improvement opportunities. 398 7. Implementation Status 400 TBD. 402 8. IANA Considerations 404 A new ICMPv6 Message Type code for "Fragmentation Report (FRAGREP)" 405 is requested. 407 The IANA is instructed to create new registries for "ICMPv6 Packet 408 Too Big Code field" and "ICMPv4 Fragmentation Needed unused field" 409 values. Both registries should have the following initial values: 411 Value Sub-Type name Reference 412 ----- ------------- ---------- 413 0 PTB hard error [RFCXXXX] 414 1 PTB soft error (loss) [RFCXXXX] 415 2 PTB soft error (no loss) [RFCXXXX] 416 3-252 Unassigned 417 253-254 Reserved for Experimentation [RFCXXXX] 418 255 Reserved by IANA [RFCXXXX] 420 Figure 1: Packet Too Big Code/unused Values 422 9. Security Considerations 424 Communications networking security is necessary to preserve 425 confidentiality, integrity and availability. 427 10. Acknowledgements 429 This work was inspired by ongoing AERO/OMNI/DTN investigations. 431 . 433 11. References 435 11.1. Normative References 437 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 438 DOI 10.17487/RFC0791, September 1981, 439 . 441 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 442 DOI 10.17487/RFC1191, November 1990, 443 . 445 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 446 Requirement Levels", BCP 14, RFC 2119, 447 DOI 10.17487/RFC2119, March 1997, 448 . 450 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 451 Control Message Protocol (ICMPv6) for the Internet 452 Protocol Version 6 (IPv6) Specification", STD 89, 453 RFC 4443, DOI 10.17487/RFC4443, March 2006, 454 . 456 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 457 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 458 May 2017, . 460 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 461 (IPv6) Specification", STD 86, RFC 8200, 462 DOI 10.17487/RFC8200, July 2017, 463 . 465 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 466 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 467 DOI 10.17487/RFC8201, July 2017, 468 . 470 11.2. Informative References 472 [I-D.templin-6man-aero] 473 Templin, F. L., "Automatic Extended Route Optimization 474 (AERO)", Work in Progress, Internet-Draft, draft-templin- 475 6man-aero-37, 15 November 2021, 476 . 479 [I-D.templin-6man-omni] 480 Templin, F. L. and T. Whyman, "Transmission of IP Packets 481 over Overlay Multilink Network (OMNI) Interfaces", Work in 482 Progress, Internet-Draft, draft-templin-6man-omni-51, 15 483 November 2021, . 486 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 487 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 488 December 1998, . 490 [RFC3366] Fairhurst, G. and L. Wood, "Advice to link designers on 491 link Automatic Repeat reQuest (ARQ)", BCP 62, RFC 3366, 492 DOI 10.17487/RFC3366, August 2002, 493 . 495 [RFC8900] Bonica, R., Baker, F., Huston, G., Hinden, R., Troan, O., 496 and F. Gont, "IP Fragmentation Considered Fragile", 497 BCP 230, RFC 8900, DOI 10.17487/RFC8900, September 2020, 498 . 500 Author's Address 502 Fred L. Templin (editor) 503 Boeing Research & Technology 504 P.O. Box 3707 505 Seattle, WA 98124 506 United States of America 508 Email: fltemplin@acm.org