idnits 2.17.1 draft-templin-6man-omni-interface-17.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 (April 22, 2020) is 1465 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) -- Looks like a reference, but probably isn't: '00' on line 797 -- Looks like a reference, but probably isn't: '03' on line 791 -- Looks like a reference, but probably isn't: '04' on line 792 -- Looks like a reference, but probably isn't: '07' on line 792 -- Looks like a reference, but probably isn't: '28' on line 793 -- Looks like a reference, but probably isn't: '31' on line 793 -- Looks like a reference, but probably isn't: '63' on line 797 -- Looks like a reference, but probably isn't: '64' on line 799 -- Looks like a reference, but probably isn't: '65' on line 800 -- Looks like a reference, but probably isn't: '66' on line 800 == Missing Reference: 'RFCXXXX' is mentioned on line 1253, but not defined -- Looks like a reference, but probably isn't: '67' on line 1506 -- Looks like a reference, but probably isn't: '70' on line 1507 -- Looks like a reference, but probably isn't: '76' on line 1507 == Unused Reference: 'RFC2225' is defined on line 1393, but no explicit reference was found in the text == Unused Reference: 'RFC3692' is defined on line 1413, but no explicit reference was found in the text == Unused Reference: 'RFC7421' is defined on line 1471, but no explicit reference was found in the text == Outdated reference: A later version (-11) exists of draft-templin-6man-dhcpv6-ndopt-09 == Outdated reference: A later version (-99) exists of draft-templin-intarea-6706bis-45 Summary: 0 errors (**), 0 flaws (~~), 7 warnings (==), 14 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group F. Templin, Ed. 3 Internet-Draft The Boeing Company 4 Intended status: Standards Track A. Whyman 5 Expires: October 24, 2020 MWA Ltd c/o Inmarsat Global Ltd 6 April 22, 2020 8 Transmission of IPv6 Packets over Overlay Multilink Network (OMNI) 9 Interfaces 10 draft-templin-6man-omni-interface-17 12 Abstract 14 Mobile nodes (e.g., aircraft of various configurations, terrestrial 15 vehicles, seagoing vessels, mobile enterprise devices, etc.) 16 communicate with networked correspondents over multiple access 17 network data links and configure mobile routers to connect end user 18 networks. A multilink interface specification is therefore needed 19 for coordination with the network-based mobility service. This 20 document specifies the transmission of IPv6 packets over Overlay 21 Multilink Network (OMNI) Interfaces. 23 Status of This Memo 25 This Internet-Draft is submitted in full conformance with the 26 provisions of BCP 78 and BCP 79. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF). Note that other groups may also distribute 30 working documents as Internet-Drafts. The list of current Internet- 31 Drafts is at https://datatracker.ietf.org/drafts/current/. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 This Internet-Draft will expire on October 24, 2020. 40 Copyright Notice 42 Copyright (c) 2020 IETF Trust and the persons identified as the 43 document authors. All rights reserved. 45 This document is subject to BCP 78 and the IETF Trust's Legal 46 Provisions Relating to IETF Documents 47 (https://trustee.ietf.org/license-info) in effect on the date of 48 publication of this document. Please review these documents 49 carefully, as they describe your rights and restrictions with respect 50 to this document. Code Components extracted from this document must 51 include Simplified BSD License text as described in Section 4.e of 52 the Trust Legal Provisions and are provided without warranty as 53 described in the Simplified BSD License. 55 Table of Contents 57 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 58 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 4 59 3. Requirements . . . . . . . . . . . . . . . . . . . . . . . . 6 60 4. Overlay Multilink Network (OMNI) Interface Model . . . . . . 6 61 5. Maximum Transmission Unit (MTU) and Fragmentation . . . . . . 10 62 6. Frame Format . . . . . . . . . . . . . . . . . . . . . . . . 11 63 7. Link-Local Addresses . . . . . . . . . . . . . . . . . . . . 12 64 8. The SPAN . . . . . . . . . . . . . . . . . . . . . . . . . . 12 65 9. Address Mapping - Unicast . . . . . . . . . . . . . . . . . . 13 66 9.1. Sub-Options . . . . . . . . . . . . . . . . . . . . . . . 14 67 9.1.1. Pad1 . . . . . . . . . . . . . . . . . . . . . . . . 15 68 9.1.2. PadN . . . . . . . . . . . . . . . . . . . . . . . . 16 69 9.1.3. ifIndex-tuple (Type 1) . . . . . . . . . . . . . . . 16 70 9.1.4. ifIndex-tuple (Type 2) . . . . . . . . . . . . . . . 18 71 9.1.5. MS-Register . . . . . . . . . . . . . . . . . . . . . 19 72 9.1.6. MS-Release . . . . . . . . . . . . . . . . . . . . . 19 73 10. Address Mapping - Multicast . . . . . . . . . . . . . . . . . 20 74 11. Conceptual Sending Algorithm . . . . . . . . . . . . . . . . 20 75 11.1. Multiple OMNI Interfaces . . . . . . . . . . . . . . . . 20 76 12. Router Discovery and Prefix Registration . . . . . . . . . . 21 77 13. Secure Redirection . . . . . . . . . . . . . . . . . . . . . 24 78 14. AR and MSE Resilience . . . . . . . . . . . . . . . . . . . . 24 79 15. Detecting and Responding to MSE Failures . . . . . . . . . . 25 80 16. Transition Considerations . . . . . . . . . . . . . . . . . . 25 81 17. OMNI Interfaces on the Open Internet . . . . . . . . . . . . 26 82 18. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . . . 26 83 19. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 27 84 20. Security Considerations . . . . . . . . . . . . . . . . . . . 27 85 21. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 28 86 22. References . . . . . . . . . . . . . . . . . . . . . . . . . 28 87 22.1. Normative References . . . . . . . . . . . . . . . . . . 28 88 22.2. Informative References . . . . . . . . . . . . . . . . . 30 89 Appendix A. Type 1 ifIndex-tuple Traffic Classifier Preference 90 Encoding . . . . . . . . . . . . . . . . . . . . . . 32 91 Appendix B. VDL Mode 2 Considerations . . . . . . . . . . . . . 34 92 Appendix C. MN / AR Isolation Through L2 Address Mapping . . . . 35 93 Appendix D. Change Log . . . . . . . . . . . . . . . . . . . . . 36 94 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 41 96 1. Introduction 98 Mobile Nodes (MNs) (e.g., aircraft of various configurations, 99 terrestrial vehicles, seagoing vessels, mobile enterprise devices, 100 etc.) often have multiple data links for communicating with networked 101 correspondents. These data links may have diverse performance, cost 102 and availability properties that can change dynamically according to 103 mobility patterns, flight phases, proximity to infrastructure, etc. 104 MNs coordinate their data links in a discipline known as "multilink", 105 in which a single virtual interface is configured over the underlying 106 data links. 108 The MN configures a virtual interface (termed the "Overlay Multilink 109 Network (OMNI) interface") as a thin layer over the underlying Access 110 Network (ANET) interfaces. The OMNI interface is therefore the only 111 interface abstraction exposed to the IPv6 layer and behaves according 112 to the Non-Broadcast, Multiple Access (NBMA) interface principle, 113 while underlying interfaces appear as link layer communication 114 channels in the architecture. The OMNI interface connects to a 115 virtual overlay service known as the "OMNI link". The OMNI link 116 spans a worldwide Internetwork that may include private-use 117 infrastructures and/or the global public Internet itself. 119 Each MN receives a Mobile Network Prefix (MNP) for numbering 120 downstream-attached End User Networks (EUNs) independently of the 121 access network data links selected for data transport. The MN 122 performs router discovery over the OMNI interface (i.e., similar to 123 IPv6 customer edge routers [RFC7084]) and acts as a mobile router on 124 behalf of its EUNs. The router discovery process is iterated over 125 each of the OMNI interface's underlying interfaces in order to 126 register per-link parameters (see Section 12). 128 The OMNI interface provides a multilink nexus for exchanging inbound 129 and outbound traffic via the correct underlying interface(s). The 130 IPv6 layer sees the OMNI interface as a point of connection to the 131 OMNI link. Each OMNI link has one or more associated Mobility 132 Service Prefixes (MSPs) from which OMNI link MNPs are derived. If 133 there are multiple OMNI links, the IPv6 layer will see multiple OMNI 134 interfaces. 136 The OMNI interface interacts with a network-based Mobility Service 137 (MS) through IPv6 Neighbor Discovery (ND) control message exchanges 138 [RFC4861]. The MS provides Mobility Service Endpoints (MSEs) that 139 track MN movements and represent their MNPs in a global routing or 140 mapping system. 142 This document specifies the transmission of IPv6 packets [RFC8200] 143 and MN/MS control messaging over OMNI interfaces. 145 2. Terminology 147 The terminology in the normative references applies; especially, the 148 terms "link" and "interface" are the same as defined in the IPv6 149 [RFC8200] and IPv6 Neighbor Discovery (ND) [RFC4861] specifications. 150 Also, the Protocol Constants defined in Section 10 of [RFC4861] are 151 used in their same format and meaning in this document. The terms 152 "All-Routers multicast", "All-Nodes multicast" and "Subnet-Router 153 anycast" are defined in [RFC4291] (with Link-Local scope assumed). 155 The following terms are defined within the scope of this document: 157 Mobile Node (MN) 158 an end system with multiple distinct upstream data link 159 connections that are managed together as a single logical unit. 160 The MN's data link connection parameters can change over time due 161 to, e.g., node mobility, link quality, etc. The MN further 162 connects a downstream-attached End User Network (EUN). The term 163 MN used here is distinct from uses in other documents, and does 164 not imply a particular mobility protocol. 166 End User Network (EUN) 167 a simple or complex downstream-attached mobile network that 168 travels with the MN as a single logical unit. The IPv6 addresses 169 assigned to EUN devices remain stable even if the MN's upstream 170 data link connections change. 172 Mobility Service (MS) 173 a mobile routing service that tracks MN movements and ensures that 174 MNs remain continuously reachable even across mobility events. 175 Specific MS details are out of scope for this document. 177 Mobility Service Endpoint (MSE) 178 an entity in the MS (either singluar or aggregate) that 179 coordinates the mobility events of one or more MN. 181 Mobility Service Prefix (MSP) 182 an aggregated IPv6 prefix (e.g., 2001:db8::/32) advertised to the 183 rest of the Internetwork by the MS, and from which more-specific 184 Mobile Network Prefixes (MNPs) are derived. 186 Mobile Network Prefix (MNP) 187 a longer IPv6 prefix taken from an MSP (e.g., 188 2001:db8:1000:2000::/56) and assigned to a MN. MNs sub-delegate 189 the MNP to devices located in EUNs. 191 Access Network (ANET) 192 a data link service network (e.g., an aviation radio access 193 network, satellite service provider network, cellular operator 194 network, wifi network, etc.) that connects MNs. Physical and/or 195 data link level security between the MN and ANET are assumed. 197 Access Router (AR) 198 a first-hop router in the ANET for connecting MNs to 199 correspondents in outside Internetworks. 201 ANET interface 202 a MN's attachment to a link in an ANET. 204 Internetwork (INET) 205 a connected network region with a coherent IP addressing plan that 206 provides transit forwarding services for ANET MNs and INET 207 correspondents. Examples include private enterprise networks, 208 ground domain aviation service networks and the global public 209 Internet itself. 211 INET interface 212 a node's attachment to a link in an INET. 214 OMNI link 215 a virtual overlay configured over one or more INETs and their 216 connected ANETs. An OMNI link can comprise multiple INET segments 217 joined by bridges the same as for any link; the addressing plans 218 in each segment may be mutually exclusive and managed by different 219 administrative entities. 221 OMNI interface 222 a node's attachment to an OMNI link, and configured over one or 223 more underlying ANET/INET interfaces. 225 OMNI link local address (LLA) 226 an IPv6 link-local address constructed as specified in Section 7, 227 and assigned to an OMNI interface. 229 OMNI Option 230 an IPv6 Neighbor Discovery option providing multilink parameters 231 for the OMNI interface as specified in Section 9. 233 Multilink 234 an OMNI interface's manner of managing diverse underlying data 235 link interfaces as a single logical unit. The OMNI interface 236 provides a single unified interface to upper layers, while 237 underlying data link selections are performed on a per-packet 238 basis considering factors such as DSCP, flow label, application 239 policy, signal quality, cost, etc. Multilinking decisions are 240 coordinated in both the outbound (i.e. MN to correspondent) and 241 inbound (i.e., correspondent to MN) directions. 243 L2 244 The second layer in the OSI network model. Also known as "layer- 245 2", "link-layer", "sub-IP layer", "data link layer", etc. 247 L3 248 The third layer in the OSI network model. Also known as "layer- 249 3", "network-layer", "IPv6 layer", etc. 251 underlying interface 252 an ANET/INET interface over which an OMNI interface is configured. 253 The OMNI interface is seen as a L3 interface by the IP layer, and 254 each underlying interface is seen as a L2 interface by the OMNI 255 interface. 257 Mobility Service Identification (MSID) 258 Each MSE and AR is assigned a unique 32-bit Identification (MSID) 259 as specified in Section 7. 261 Spanning Partitioned Administrative Networks (SPAN) 262 A means for bridging disjoint INET partitions as segments of a 263 unified OMNI link the same as for a bridged campus LAN. The SPAN 264 is a mid-layer IPv6 encapsulation service that supports a unified 265 OMNI link view for all segments. 267 3. Requirements 269 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 270 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 271 "OPTIONAL" in this document are to be interpreted as described in BCP 272 14 [RFC2119][RFC8174] when, and only when, they appear in all 273 capitals, as shown here. 275 An implementation is not required to internally use the architectural 276 constructs described here so long as its external behavior is 277 consistent with that described in this document. 279 4. Overlay Multilink Network (OMNI) Interface Model 281 An OMNI interface is a MN virtual interface configured over one or 282 more underlying interfaces, which may be physical (e.g., an 283 aeronautical radio link) or virtual (e.g., an Internet or higher- 284 layer "tunnel"). The MN receives a MNP from the MS, and coordinates 285 with the MS through IPv6 ND message exchanges. The MN uses the MNP 286 to construct a unique OMNI LLA through the algorithmic derivation 287 specified in Section 7 and assigns the LLA to the OMNI interface. 289 The OMNI interface architectural layering model is the same as in 290 [RFC7847], and augmented as shown in Figure 1. The IP layer 291 therefore sees the OMNI interface as a single L3 interface with 292 multiple underlying interfaces that appear as L2 communication 293 channels in the architecture. 295 +----------------------------+ 296 | Upper Layer Protocol | 297 Session-to-IP +---->| | 298 Address Binding | +----------------------------+ 299 +---->| IP (L3) | 300 IP Address +---->| | 301 Binding | +----------------------------+ 302 +---->| OMNI Interface | 303 Logical-to- +---->| (OMNI LLA) | 304 Physical | +----------------------------+ 305 Interface +---->| L2 | L2 | | L2 | 306 Binding |(IF#1)|(IF#2)| ..... |(IF#n)| 307 +------+------+ +------+ 308 | L1 | L1 | | L1 | 309 | | | | | 310 +------+------+ +------+ 312 Figure 1: OMNI Interface Architectural Layering Model 314 The OMNI virtual interface model gives rise to a number of 315 opportunities: 317 o since OMNI LLAs are uniquely derived from an MNP, no Duplicate 318 Address Detection (DAD) or Muticast Listener Discovery (MLD) 319 messaging is necessary. 321 o ANET interfaces do not require any L3 addresses (i.e., not even 322 link-local) in environments where communications are coordinated 323 entirely over the OMNI interface. (An alternative would be to 324 also assign the same OMNI LLA to all ANET interfaces.) 326 o as ANET interface properties change (e.g., link quality, cost, 327 availability, etc.), any active ANET interface can be used to 328 update the profiles of multiple additional ANET interfaces in a 329 single message. This allows for timely adaptation and service 330 continuity under dynamically changing conditions. 332 o coordinating ANET interfaces in this way allows them to be 333 represented in a unified MS profile with provisions for mobility 334 and multilink operations. 336 o exposing a single virtual interface abstraction to the IPv6 layer 337 allows for multilink operation (including QoS based link 338 selection, packet replication, load balancing, etc.) at L2 while 339 still permitting L3 traffic shaping based on, e.g., DSCP, flow 340 label, etc. 342 o L3 sees the OMNI interface as a point of connection to the OMNI 343 link; if there are multiple OMNI links (i.e., multiple MS's), L3 344 will see multiple OMNI interfaces. 346 Other opportunities are discussed in [RFC7847]. 348 Figure 2 depicts the architectural model for a MN connecting to the 349 MS via multiple independent ANETs. When an underlying interface 350 becomes active, the MN's OMNI interface sends native (i.e., 351 unencapsulated) IPv6 ND messages via the underlying interface. IPv6 352 ND messages traverse the ground domain ANETs until they reach an 353 Access Router (AR#1, AR#2, .., AR#n). The AR then coordinates with a 354 Mobility Service Endpoint (MSE#1, MSE#2, ..., MSE#m) in the INET and 355 returns an IPv6 ND message response to the MN. IPv6 ND messages 356 traverse the ANET at layer 2; hence, the Hop Limit is not 357 decremented. 359 +--------------+ 360 | MN | 361 +--------------+ 362 |OMNI interface| 363 +----+----+----+ 364 +--------|IF#1|IF#2|IF#n|------ + 365 / +----+----+----+ \ 366 / | \ 367 / <---- Native | IP ----> \ 368 v v v 369 (:::)-. (:::)-. (:::)-. 370 .-(::ANET:::) .-(::ANET:::) .-(::ANET:::) 371 `-(::::)-' `-(::::)-' `-(::::)-' 372 +----+ +----+ +----+ 373 ... |AR#1| .......... |AR#2| ......... |AR#n| ... 374 . +-|--+ +-|--+ +-|--+ . 375 . | | | 376 . v v v . 377 . <----- Encapsulation -----> . 378 . . 379 . +-----+ (:::)-. . 380 . |MSE#2| .-(::::::::) +-----+ . 381 . +-----+ .-(::: INET :::)-. |MSE#m| . 382 . (::::: Routing ::::) +-----+ . 383 . `-(::: System :::)-' . 384 . +-----+ `-(:::::::-' . 385 . |MSE#1| +-----+ +-----+ . 386 . +-----+ |MSE#3| |MSE#4| . 387 . +-----+ +-----+ . 388 . . 389 . . 390 . <----- Worldwide Connected Internetwork ----> . 391 ........................................................... 393 Figure 2: MN/MS Coordination via Multiple ANETs 395 After the initial IPv6 ND message exchange, the MN can send and 396 receive unencapsulated IPv6 data packets over the OMNI interface. 397 OMNI interface multilink services will forward the packets via ARs in 398 the correct underlying ANETs. The AR encapsulates the packets 399 according to the capabilities provided by the MS and forwards them to 400 the next hop within the worldwide connected Internetwork via optimal 401 routes. 403 OMNI links span the underlying Internetwork via a mid-layer overlay 404 known as "The SPAN" - see Section 8. Each OMNI link corresponds to a 405 different SPAN overlay which may be carried over a completely 406 separate Internetwork topology and/or differentiated by a SPAN header 407 codepoint. The same as for VLANs, each MN can connect to multiple 408 OMNI links (i.e., multipe SPANs) by configuring a distinct OMNI 409 interface for each link. 411 5. Maximum Transmission Unit (MTU) and Fragmentation 413 All IPv6 interfaces are REQUIRED to configure a minimum Maximum 414 Transmission Unit (MTU) of 1280 bytes [RFC8200]. The network 415 therefore MUST forward packets of at least 1280 bytes without 416 generating an IPv6 Path MTU Discovery (PMTUD) Packet Too Big (PTB) 417 message [RFC8201]. 419 The OMNI interface configures an MTU of 9180 bytes [RFC2492]; the 420 size is therefore not a reflection of the underlying interface MTUs, 421 but rather determines the largest packet the OMNI interface can 422 forward or reassemble. The OMNI interface therefore accommodates IP 423 packets up to 9180 bytes while generating IPv6 Path MTU Discovery 424 (PMTUD) Packet Too Big (PTB) messages [RFC8201] as necessary (see 425 below). 427 OMNI interfaces employ mid-layer IPv6 encapsulation and 428 fragmentation/reassembly per [RFC2473] (also known as "SPAN 429 encapsulation" - see Section 8) to accommodate the 9180 byte MTU. 430 The OMNI interface returns internally-generated PTB messages for 431 packets admitted into the interface that it deems too large (e.g., 432 according to link performance characteristics, reassembly cost, etc.) 433 while either dropping or forwarding the packet as necessary. The 434 OMNI interface performs PMTUD even if the destination appears to be 435 on the same link since an OMNI link node on the path may return a 436 PTB. This ensures that the path MTU is adaptive and reflects the 437 current path used for a given data flow. 439 OMNI interfaces perform SPAN encapsulation and fragmentation/ 440 reassembly as follows: 442 o When an OMNI interface sends a packet toward a final destination 443 via an ANET peer, it sends without SPAN encapsulation if the 444 packet is no larger than the underlying interface MTU. Otherwise, 445 it encapsulates the packet in a SPAN header with source address 446 set to the node's own SPAN address and destination set to the SPAN 447 address of the ANET peer. The OMNI interface then uses IPv6 448 fragmentation to break the encapsulated packet into a minimum 449 number of non-overlapping fragments, where the largest fragment 450 size is determined by the underlying interface MTU and the 451 smallest fragment is no smaller than 640 bytes. The OMNI 452 interface then sends the fragments to the ANET peer, which 453 reassembles before forwarding toward the final destination. 455 o When an OMNI interface sends a packet toward a final destination 456 via an INET interface, it sends packets no larger than 1280 bytes 457 without SPAN encapsulation if the destination is reached via an 458 INET address within the same SPAN segment. Otherwise, it 459 encapsulates the packet in a SPAN header with source address set 460 to the node's SPAN address and destination set to the SPAN address 461 of the next hop OMNI node toward the final destination and (if 462 necessary) with a Segment Routing Header [RFC8754] with the 463 remaining Segment IDs on the path to the final destination. The 464 OMNI interface then uses IPv6 fragmentation to break the 465 encapsulated packet into a minimum number of non-overlapping 466 fragments, where the largest fragment size is 1280 bytes and the 467 smallest fragment is no smaller than 640 bytes. The OMNI 468 interface then sends the fragments to the SPAN destination, which 469 reassembles before forwarding toward the final destination. 471 In order to avoid a "tiny fragment" attack, OMNI interfaces 472 unconditionally drop all SPAN fragments smaller than 640 bytes. In 473 order to set the correct context for reassembly, the OMNI interface 474 that inserts a SPAN header MUST also be the one that inserts the IPv6 475 Fragment Header Identification value. Although all fragmnets of the 476 same fragmented SPAN packet are typically sent via the same 477 underlying interface, this is not strictly required since all 478 fragments will arrive at the OMNI interface that performs reassembly 479 even if they travel over different paths. 481 Note that the OMNI interface can forward large packets via 482 encapsulation and fragmentation while at the same time returning 483 advisory PTB messages, e.g., subject to rate limiting. The receiving 484 node that performs reassembly can also send advisory PTB messages if 485 reassembly conditions become unfavorable. The AERO interface can 486 therefore continuously forward large packets without loss while 487 returning advisory messages recommending a smaller size (but no 488 smaller than 1280). Advisory PTB messages are differentiated from 489 PTB messages that report loss by setting the Code field in the ICMPv6 490 message header to the value 1. This document therefore updates 491 [RFC4443] and [RFC8201]. 493 6. Frame Format 495 The OMNI interface transmits IPv6 packets according to the native 496 frame format of each underlying interface. For example, for 497 Ethernet-compatible interfaces the frame format is specified in 498 [RFC2464], for aeronautical radio interfaces the frame format is 499 specified in standards such as ICAO Doc 9776 (VDL Mode 2 Technical 500 Manual), for tunnels over IPv6 the frame format is specified in 501 [RFC2473], etc. 503 7. Link-Local Addresses 505 OMNI interfaces assign IPv6 Link-Local Addresses (i.e., "OMNI LLAs") 506 using the following constructs: 508 o IPv6 MN OMNI LLAs encode the most-significant 112 bits of a MNP 509 within the least-significant 112 bits of the the IPv6 link-local 510 prefix fe80::/16. For example, for the MNP 511 2001:db8:1000:2000::/56 the corresponding LLA is 512 fe80:2001:db8:1000:2000::. See: [RFC4291], Section 2.5.6) for a 513 discussion of IPv6 link-local addresses. 515 o IPv4-compatible MN OMNI LLAs are assigned as fe80::ffff:[v4addr], 516 i.e., the most significant 16 bits of the prefix fe80::/16, 517 followed by 64 '0' bits, followed by 16 '1' bits, followed by a 518 32bit IPv4 address. For example, the IPv4-Compatible MN OMNI LLA 519 for 192.0.2.1 is fe80::ffff:192.0.2.1 (also written as 520 fe80::ffff:c000:0201). 522 o MS OMNI LLAs are assigned to ARs and MSEs from the range 523 fe80::/96, and MUST be managed for uniqueness. The lower 32 bits 524 of the LLA includes a unique integer "MSID" value between 525 0x00000001 and 0xfeffffff, e.g., as in fe80::1, fe80::2, fe80::3, 526 etc., fe80::feff:ffff. The MSID 0x00000000 corresponds to the 527 link-local Subnet-Router anycast address (fe80::) [RFC4291]. The 528 MSID range 0xff000000 through 0xffffffff is reserved for future 529 use. 531 o The OMNI LLA range fe80::/32 is used as the Teredo service prefix 532 for OMNI interfaces according to the format in Section 4 of 533 [RFC4380] (see Section 17 for further discussion). 535 Since the prefix 0000::/8 is "Reserved by the IETF" [RFC4291], no 536 MNPs can be allocated from that block ensuring that there is no 537 possibility for overlap between the above OMNI LLA constructs. 539 Since MN OMNI LLAs are based on the distribution of administratively 540 assured unique MNPs, and since MS OMNI LLAs are guaranteed unique 541 through administrative assignment, OMNI interfaces set the 542 autoconfiguration variable DupAddrDetectTransmits to 0 [RFC4862]. 544 8. The SPAN 546 OMNI links employ an overlay network instance called "The SPAN" 547 (Spanning Partitioned Administrative Networks) that supports 548 forwarding of encapsulated link-scoped messages over an IPv6 overlay 549 routing instance that spans the entire link without decrementing the 550 (link-local) Hop Limit. The OMNI link reserves the Unique Local 551 Address (ULA) prefix fd80::/10 [RFC4193] used for mapping OMNI LLAs 552 to routable SPAN addresses. 554 SPAN addresses are configured in one-to-one correspondence with MN/MS 555 OMNI LLAs through stateless translation of the prefix. For example, 556 for the SPAN sub-prefix fd80::/16: 558 o the SPAN address corresponding to fe80:2001:db8:1:2:: is simply 559 fd80:2001:db8:1:2:: 561 o the SPAN address corresponding to fe80::ffff:192.0.2.1 is simply 562 fd80::ffff:192.0.2.1 564 o the SPAN address corresponding to fe80::1000 is simply fd80::1000 566 The SPAN address presents an IPv6 address format that is routable 567 within the OMNI link routing system and can be used to convey link- 568 scoped messages across multiple hops using IPv6 encapsulation 569 [RFC2473]. The SPAN extends over the entire OMNI link to include all 570 ARs and MSEs. All MNs are also consideed to be "on the SPAN", 571 however SPAN encapsulation is omitted over ANET links when possible 572 to conserve bandwidth (see: Section 11). 574 The SPAN allows the OMNI link to be subdivided into "segments" that 575 often correspond to administrative domains or physical partitions. 576 OMNI nodes can use IPv6 Segment Routing [RFC8754][RFC8402] when 577 necessary to support efficient packet forwarding to destinations 578 located in other SPAN segments. A full discussion of Segment Routing 579 over the SPAN appears in [I-D.templin-intarea-6706bis]. 581 9. Address Mapping - Unicast 583 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 584 state and use the link-local address format specified in Section 7. 585 IPv6 Neighbor Discovery (ND) [RFC4861] messages on MN OMNI interfaces 586 observe the native Source/Target Link-Layer Address Option (S/TLLAO) 587 formats of the underlying interfaces (e.g., for Ethernet the S/TLLAO 588 is specified in [RFC2464]). 590 MNs such as aircraft typically have many wireless data link types 591 (e.g. satellite-based, cellular, terrestrial, air-to-air directional, 592 etc.) with diverse performance, cost and availability properties. 593 The OMNI interface would therefore appear to have multiple L2 594 connections, and may include information for multiple underlying 595 interfaces in a single IPv6 ND message exchange. 597 OMNI interfaces use an IPv6 ND option called the "OMNI option" 598 formatted as shown in Figure 3: 600 0 1 2 3 601 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 1 602 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 603 | Type | Length | Prefix Length |R| Reserved | 604 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 605 | | 606 ~ Sub-Options ~ 607 | | 608 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 610 Figure 3: OMNI Option Format 612 In this format: 614 o Type is set to TBD. 616 o Length is set to the number of 8 octet blocks in the option. 618 o Prefix Length is set according to the IPv6 source address type. 619 For MN OMNI LLAs, the value is set to the length of the embedded 620 MNP. For IPv4-compatible MN OMNI LLAs, the value is set to 96 621 plus the length of the embedded IPv4 prefix. For MS OMNI LLAs, 622 the value is set to 128. 624 o R (the "Register/Release" bit) is set to 1/0 to request the 625 message recipient to register/release a MN's MNP. The OMNI option 626 may additionally include MSIDs for the recipient to contact to 627 also register/release the MNP. 629 o Reserved is set to the value '0' on transmission and ignored on 630 reception. 632 o Sub-Options is a Variable-length field, of length such that the 633 complete OMNI Option is an integer multiple of 8 octets long. 634 Contains one or more options, as described in Section 8.1. 636 9.1. Sub-Options 638 The OMNI option includes zero or more Sub-Options, some of which may 639 appear multiple times in the same message. Each consecutive Sub- 640 Option is concatenated immediately after its predecessor. All Sub- 641 Options except Pad1 (see below) are type-length-value (TLV) encoded 642 in the following format: 644 0 1 2 645 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 5 646 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 647 | Sub-Type | Sub-length | Sub-Option Data ... 648 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 650 Figure 4: Sub-Option Format 652 o Sub-Type is a 1-byte field that encodes the Sub-Option type. Sub- 653 Options defined in this document are: 655 Option Name Sub-Type 656 Pad1 0 657 PadN 1 658 ifIndex-tuple (Type 1) 2 659 ifIndex-tuple (Type 2) 3 660 MS-Register 4 661 MS-Release 5 663 Figure 5 665 Sub-Types 253 and 254 are reserved for experimentation, as 666 recommended in[RFC3692]]. 668 o Sub-Length is a 1-byte field that encodes the length of the Sub- 669 Option Data, in bytes 671 o Sub-Option Data is a byte string with format determined by Sub- 672 Type 674 During processing, unrecognized Sub-Options are ignored and the next 675 Sub-Option processed until the end of the OMNI option. 677 The following Sub-Option types and formats are defined in this 678 document: 680 9.1.1. Pad1 682 0 683 0 1 2 3 4 5 6 7 684 +-+-+-+-+-+-+-+-+ 685 | Sub-Type=0 | 686 +-+-+-+-+-+-+-+-+ 688 Figure 6: Pad1 690 o Sub-Type is set to 0. 692 o No Sub-Length or Sub-Option Data follows (i.e., the "Sub-Option" 693 consists of a single zero octet). 695 9.1.2. PadN 697 0 1 2 698 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 5 699 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 700 | Sub-Type=1 |Sub-length=N-2 | N-2 padding bytes ... 701 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 703 Figure 7: PadN 705 o Sub-Type is set to 1. 707 o Sub-Length is set to N-2 being the number of padding bytes that 708 follow. 710 o Sub-Option Data consists of N-2 zero-valued octets. 712 9.1.3. ifIndex-tuple (Type 1) 714 0 1 2 3 715 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 1 716 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 717 | Sub-Type=2 | Sub-length=4+N| ifIndex | ifType | 718 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 719 | Provider ID | Link |S|I|RSV| Bitmap(0)=0xff|P00|P01|P02|P03| 720 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 721 |P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15|P16|P17|P18|P19| 722 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 723 |P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| Bitmap(1)=0xff| 724 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 725 |P32|P33|P34|P35|P36|P37|P38|P39| ... 726 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 728 Figure 8: ifIndex-tuple (Type 1) 730 o Sub-Type is set to 2. 732 o Sub-Length is set to 4+N (the number of Sub-Option Data bytes that 733 follow). 735 o Sub-Option Data contains an "ifIndex-tuple" (Type 1) encoded as 736 follows (note that the first four bytes must be present): 738 * ifIndex is set to an 8-bit integer value corresponding to a 739 specific underlying interface. OMNI options MAY include 740 multiple ifIndex-tuples, and MUST number each with an ifIndex 741 value between '1' and '255' that represents a MN-specific 8-bit 742 mapping for the actual ifIndex value assigned to the underlying 743 interface by network management [RFC2863] (the ifIndex value 744 '0' is reserved for use by the MS). Multiple ifIndex-tuples 745 with the same ifIndex value MAY appear in the same OMNI option. 747 * ifType is set to an 8-bit integer value corresponding to the 748 underlying interface identified by ifIndex. The value 749 represents an OMNI interface-specific 8-bit mapping for the 750 actual IANA ifType value registered in the 'IANAifType-MIB' 751 registry [http://www.iana.org]. 753 * Provider ID is set to an OMNI interface-specific 8-bit ID value 754 for the network service provider associated with this ifIndex. 756 * Link encodes a 4-bit link metric. The value '0' means the link 757 is DOWN, and the remaining values mean the link is UP with 758 metric ranging from '1' ("lowest") to '15' ("highest"). 760 * S is set to '1' if this ifIndex-tuple corresponds to the 761 underlying interface that is the source of the ND message. Set 762 to '0' otherwise. 764 * I is set to '0' ("Simplex") if the index for each singleton 765 Bitmap byte in the Sub-Option Data is inferred from its 766 sequential position (i.e., 0, 1, 2, ...), or set to '1' 767 ("Indexed") if each Bitmap is preceded by an Index byte. 768 Figure 8 shows the simplex case for I set to '0'. For I set to 769 '1', each Bitmap is instead preceded by an Index byte that 770 encodes a value "i" = (0 - 255) as the index for its companion 771 Bitmap as follows: 773 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 774 | Index=i | Bitmap(i) |P[*] values ... 775 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 777 Figure 9 779 * RSV is set to the value 0 on transmission and ignored on 780 reception. 782 * The remainder of the Sub-Option Data contains N = (0 - 251) 783 bytes of traffic classifier preferences consisting of a first 784 (indexed) Bitmap (i.e., "Bitmap(i)") followed by 0-8 1-byte 785 blocks of 2-bit P[*] values, followed by a second Bitmap (i), 786 followed by 0-8 blocks of P[*] values, etc. Reading from bit 0 787 to bit 7, the bits of each Bitmap(i) that are set to '1'' 788 indicate the P[*] blocks from the range P[(i*32)] through 789 P[(i*32) + 31] that follow; if any Bitmap(i) bits are '0', then 790 the corresponding P[*] block is instead omitted. For example, 791 if Bitmap(0) contains 0xff then the block with P[00]-P[03], 792 followed by the block with P[04]-P[07], etc., and ending with 793 the block with P[28]-P[31] are included (as showin in 794 Figure 8). The next Bitmap(i) is then consulted with its bits 795 indicating which P[*] blocks follow, etc. out to the end of the 796 Sub-Option. The first 16 P[*] blocks correspond to the 64 797 Differentiated Service Code Point (DSCP) values P[00] - P[63] 798 [RFC2474]. If additional P[*] blocks follow, their values 799 correspond to "pseudo-DSCP" traffic classifier values P[64], 800 P[65], P[66], etc. See Appendix A for further discussion and 801 examples. 803 * Each 2-bit P[*] field is set to the value '0' ("disabled"), '1' 804 ("low"), '2' ("medium") or '3' ("high") to indicate a QoS 805 preference level for underlying interface selection purposes. 806 Not all P[*] values need to be included in all OMNI option 807 instances of a given ifIndex-tuple. Any P[*] values 808 represented in an earlier OMNI option but ommitted in the 809 current OMNI option remain unchanged. Any P[*] values not yet 810 represented in any OMNI option default to "medium". 812 9.1.4. ifIndex-tuple (Type 2) 814 0 1 2 3 815 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 1 816 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 817 | Sub-Type=3 | Sub-length=4+N| ifIndex | ifType | 818 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 819 | Provider ID | Link |S|Resvd| ~ 820 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 821 ~ ~ 822 ~ RFC 6088 Format Traffic Selector ~ 823 ~ ~ 824 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 826 Figure 10: ifIndex-tuple (Type 2) 828 o Sub-Type is set to 3. 830 o Sub-Length is set to 4+N (the number of Sub-Option Data bytes that 831 follow). 833 o Sub-Option Data contains an "ifIndex-tuple" (Type 2) encoded as 834 follows (note that the first four bytes must be present): 836 * ifIndex, ifType, Provider ID, Link and S are set exactly as for 837 Type 1 ifIndex-tuples as specified in Section 9.1.3. 839 * the remainder of the Sub-Option body encodes a variable-length 840 traffic selector formatted per [RFC6088], beginning with the 841 "TS Format" field. 843 9.1.5. MS-Register 845 0 1 2 3 846 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 1 847 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 848 | Sub-Type=4 | Sub-length=4 | MSID (bits 0 - 15) | 849 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 850 | MSID (bits 16 - 32) | 851 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 853 Figure 11: MS-Register Sub-option 855 o Sub-Type is set to 4. 857 o Sub-Length is set to 4. 859 o MSID contains the 32 bit ID of an MSE or AR, in network byte 860 order. OMNI options contain zero or more MS-Register sub-options. 862 9.1.6. MS-Release 864 0 1 2 3 865 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 1 866 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 867 | Sub-Type=5 | Sub-length=4 | MSID (bits 0 - 15) | 868 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 869 | MSID (bits 16 - 32) | 870 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 872 Figure 12: MS-Release Sub-option 874 o Sub-Type is set to 5. 876 o Sub-Length is set to 4. 878 o MSIID contains the 32 bit ID of an MS or AR, in network byte 879 order. OMNI options contain zero or more MS-Release sub-options. 881 10. Address Mapping - Multicast 883 The multicast address mapping of the native underlying interface 884 applies. The mobile router on board the aircraft also serves as an 885 IGMP/MLD Proxy for its EUNs and/or hosted applications per [RFC4605] 886 while using the L2 address of the router as the L2 address for all 887 multicast packets. 889 11. Conceptual Sending Algorithm 891 The MN's IPv6 layer selects the outbound OMNI interface according to 892 standard IPv6 requirements when forwarding data packets from local or 893 EUN applications to external correspondents. The OMNI interface 894 maintains a neighbor cache the same as for any IPv6 interface, but 895 with additional state for multilink coordination. 897 After a packet enters the OMNI interface, an outbound underlying 898 interface is selected based on multilink parameters such as DSCP, 899 application port number, cost, performance, message size, etc. OMNI 900 interface multilink selections could also be configured to perform 901 replication across multiple underlying interfaces for increased 902 reliability at the expense of packet duplication. 904 When an OMNI interface sends a packet over a selected outbound 905 underlying interface, it omits SPAN encapsulation if the packet does 906 not require fragmentation and the neighbor can determine the SPAN 907 addresses through other means (e.g., the packet's destination, 908 neighbor cache information, etc.). Otherwise, the OMNI interface 909 inserts a SPAN header and performs fragmentation if necessary. 911 OMNI interface multilink service designers MUST observe the BCP 912 guidance in Section 15 [RFC3819] in terms of implications for 913 reordering when packets from the same flow may be spread across 914 multiple underlying interfaces having diverse properties. 916 11.1. Multiple OMNI Interfaces 918 MNs may associate with multiple MS instances concurrently. Each MS 919 instance represents a distinct OMNI link distinguished by its 920 associated MSPs. The MN configures a separate OMNI interface for 921 each link so that multiple interfaces (e.g., omni0, omni1, omni2, 922 etc.) are exposed to the IPv6 layer. 924 Depending on local policy and configuration, an MN may choose between 925 alternative active OMNI interfaces using a packet's DSCP, routing 926 information or static configuration. Interface selection based on 927 per-packet source addresses is also enabled when the MSPs for each 928 OMNI interface are known (e.g., discovered through Prefix Information 929 Options (PIOs) and/or Route Information Options (RIOs)). 931 Each OMNI interface can be configured over the same or different sets 932 of underlying interfaces. Each ANET distinguishes between the 933 different OMNI links based on the MSPs represented in per-packet IPv6 934 addresses. 936 Multiple distinct OMNI links can therefore be used to support fault 937 tolerance, load balancing, reliability, etc. The architectural model 938 parallels Layer 2 Virtual Local Area Networks (VLANs), where the MSPs 939 serve as (virtual) VLAN tags. 941 12. Router Discovery and Prefix Registration 943 MNs interface with the MS by sending RS messages with OMNI options 944 under the assumption that a single AR on the ANET will proocess the 945 message and respond. This places a requirement on each ANET, which 946 may be enforced by physical/logical partitioning, L2 AR beaconing, 947 etc. The manner in which the ANET ensures single AR coordination is 948 link-specific and outside the scope of this document. 950 For each underlying interface, the MN sends an RS message with an 951 OMNI option with prefix registration information, ifIndex-tuples, MS- 952 Register/Release suboptions containing MSIDs, and with destination 953 address set to All-Routers multicast (ff02::2) [RFC4291]. Example 954 MSID discovery methods are given in [RFC5214], including data link 955 login parameters, name service lookups, static configuration, etc. 956 Alternatively, MNs can discover indiviual MSIDs by sending an initial 957 RS with MS-Register MSID set to 0x00000000. 959 MNs configure OMNI interfaces that observe the properties discussed 960 in the previous section. The OMNI interface and its underlying 961 interfaces are said to be in either the "UP" or "DOWN" state 962 according to administrative actions in conjunction with the interface 963 connectivity status. An OMNI interface transitions to UP or DOWN 964 through administrative action and/or through state transitions of the 965 underlying interfaces. When a first underlying interface transitions 966 to UP, the OMNI interface also transitions to UP. When all 967 underlying interfaces transition to DOWN, the OMNI interface also 968 transitions to DOWN. 970 When an OMNI interface transitions to UP, the MN sends RS messages to 971 register its MNP and an initial set of underlying interfaces that are 972 also UP. The MN sends additional RS messages to refresh lifetimes 973 and to register/deregister underlying interfaces as they transition 974 to UP or DOWN. The MN sends initial RS messages over an UP 975 underlying interface with its OMNI LLA as the source and with 976 destination set to All-Routers multicast. The RS messages include an 977 OMNI option per Section 9 with valid prefix registration information, 978 ifIndex-tuples appropriate for underlying interfaces and MS-Register/ 979 Release sub-options. 981 ARs process IPv6 ND messages with OMNI options and act as a proxy for 982 MSEs. ARs receive RS messages and create a neighbor cache entry for 983 the MN, then coordinate with any named MSIDs in a manner outside the 984 scope of this document. The AR returns an RA message with 985 destination address set to the MN OMNI LLA (i.e., unicast), with 986 source address set to its MS OMNI LLA, with the P(roxy) bit set in 987 the RA flags [RFC4389], with an OMNI option with valid prefix 988 registration information, ifIndex-tuples, MS-Register/Release sub- 989 options, and with any information for the link that would normally be 990 delivered in a solicited RA message. ARs return RA messages with 991 configuration information in response to a MN's RS messages. The AR 992 sets the RA Cur Hop Limit, M and O flags, Router Lifetime, Reachable 993 Time and Retrans Timer values, and includes any necessary options 994 such as: 996 o PIOs with (A; L=0) that include MSPs for the link [RFC8028]. 998 o RIOs [RFC4191] with more-specific routes. 1000 o an MTU option that specifies the maximum acceptable packet size 1001 for this ANET interface. 1003 The AR coordinates with each Register/Release MSID then sends an 1004 immediate unicast RA response without delay; therefore, the IPv6 ND 1005 MAX_RA_DELAY_TIME and MIN_DELAY_BETWEEN_RAS constants for multicast 1006 RAs do not apply. The AR MAY send periodic and/or event-driven 1007 unsolicited RA messages according to the standard [RFC4861]. 1009 When the MSE processes the OMNI information, it first validates the 1010 prefix registration information. The MSE then injects/withdraws the 1011 MNP in the routing/mapping system and caches/discards the new Prefix 1012 Length, MNP and ifIndex-tuples. The MSE then informs the AR of 1013 registration success/failure, and the AR adds the MSE to the list of 1014 Register/Release MSIDs to return in an RA message OMNI option per 1015 Section 9. 1017 When the MN receives the RA message, it creates an OMNI interface 1018 neighbor cache entry with the AR's address as an L2 address and 1019 records the MSIDs that have confirmed MNP registration via this AR. 1020 If the MN connects to multiple ANETs, it establishes additional AR L2 1021 addresses (i.e., as a Multilink neighbor). The MN then manages its 1022 underlying interfaces according to their states as follows: 1024 o When an underlying interface transitions to UP, the MN sends an RS 1025 over the underlying interface with an OMNI option with R set to 1. 1026 The OMNI option contains at least one ifIndex-tuple with values 1027 specific to this underlying interface, and may contain additional 1028 ifIndex-tuples specific to this and/or other underlying 1029 interfaces. The option also includes any Register/Release MSIDs. 1031 o When an underlying interface transitions to DOWN, the MN sends an 1032 RS or unsolicited NA message over any UP underlying interface with 1033 an OMNI option containing an ifIndex-tuple for the DOWN underlying 1034 interface with Link set to '0'. The MN sends an RS when an 1035 acknowledgement is required, or an unsolicited NA when reliability 1036 is not thought to be a concern (e.g., if redundant transmissions 1037 are sent on multiple underlying interfaces). 1039 o When the Router Lifetime for a specific AR nears expiration, the 1040 MN sends an RS over the underlying interface to receive a fresh 1041 RA. If no RA is received, the MN marks the underlying interface 1042 as DOWN. 1044 o When a MN wishes to release from one or more current MSIDs, it 1045 sends an RS or unsolicited NA message over any UP underlying 1046 interfaces with an OMNI option with a Release MSID. Each MSID 1047 then withdraws the MNP from the routing/mapping system and informs 1048 the AR that the release was successful. 1050 o When all of a MNs underlying interfaces have transitioned to DOWN 1051 (or if the prefix registration lifetime expires), any associated 1052 MSEs withdraw the MNP the same as if they had received a message 1053 with a release indication. 1055 The MN is responsible for retrying each RS exchange up to 1056 MAX_RTR_SOLICITATIONS times separated by RTR_SOLICITATION_INTERVAL 1057 seconds until an RA is received. If no RA is received over a an UP 1058 underlying interface, the MN declares this underlying interface as 1059 DOWN. 1061 The IPv6 layer sees the OMNI interface as an ordinary IPv6 interface. 1062 Therefore, when the IPv6 layer sends an RS message the OMNI interface 1063 returns an internally-generated RA message as though the message 1064 originated from an IPv6 router. The internally-generated RA message 1065 contains configuration information that is consistent with the 1066 information received from the RAs generated by the MS. Whether the 1067 OMNI interface IPv6 ND messaging process is initiated from the 1068 receipt of an RS message from the IPv6 layer is an implementation 1069 matter. Some implementations may elect to defer the IPv6 ND 1070 messaging process until an RS is received from the IPv6 layer, while 1071 others may elect to initiate the process proactively. 1073 Note: The Router Lifetime value in RA messages indicates the time 1074 before which the MN must send another RS message over this underlying 1075 interface (e.g., 600 seconds), however that timescale may be 1076 significantly longer than the lifetime the MS has committed to retain 1077 the prefix registration (e.g., REACHABLETIME seconds). ARs are 1078 therefore responsible for keeping MS state alive on a shorter 1079 timescale than the MN is required to do on its own behalf. 1081 13. Secure Redirection 1083 If the ANET link model is multiple access, the AR is responsible for 1084 assuring that address duplication cannot corrupt the neighbor caches 1085 of other nodes on the link. When the MN sends an RS message on a 1086 multiple access ANET link, the AR verifys that the MN is authorized 1087 to use the address and returns an RA with a non-zero Router Lifetime 1088 only if the MN is authorized. 1090 After verifying MN authorization and returning an RA, the AR MAY 1091 return IPv6 ND Redirect messages to direct MNs located on the same 1092 ANET link to exchange packets directly without transiting the AR. In 1093 that case, the MNs can exchange packets according to their unicast L2 1094 addresses discovered from the Redirect message instead of using the 1095 dogleg path through the AR. In some ANET links, however, such direct 1096 communications may be undesirable and continued use of the dogleg 1097 path through the AR may provide better performance. In that case, 1098 the AR can refrain from sending Redirects, and/or MNs can ignore 1099 them. 1101 14. AR and MSE Resilience 1103 ANETs SHOULD deploy ARs in Virtual Router Redundancy Protocol (VRRP) 1104 [RFC5798] configurations so that service continuity is maintained 1105 even if one or more ARs fail. Using VRRP, the MN is unaware which of 1106 the (redundant) ARs is currently providing service, and any service 1107 discontinuity will be limited to the failover time supported by VRRP. 1108 Widely deployed public domain implementations of VRRP are available. 1110 MSEs SHOULD use high availability clustering services so that 1111 multiple redundant systems can provide coordinated response to 1112 failures. As with VRRP, widely deployed public domain 1113 implementations of high availability clustering services are 1114 available. Note that special-purpose and expensive dedicated 1115 hardware is not necessary, and public domain implementations can be 1116 used even between lightweight virtual machines in cloud deployments. 1118 15. Detecting and Responding to MSE Failures 1120 In environments where fast recovery from MSE failure is required, ARs 1121 SHOULD use proactive Neighbor Unreachability Detection (NUD) in a 1122 manner that parallels Bidirectional Forwarding Detection (BFD) 1123 [RFC5880] to track MSE reachability. ARs can then quickly detect and 1124 react to failures so that cached information is re-established 1125 through alternate paths. Proactive NUD control messaging is carried 1126 only over well-connected ground domain networks (i.e., and not low- 1127 end ANET links such as aeronautical radios) and can therefore be 1128 tuned for rapid response. 1130 ARs perform proactive NUD for MSEs for which there are currently 1131 active MNs on the ANET. If an MSE fails, ARs can quickly inform MNs 1132 of the outage by sending multicast RA messages on the ANET interface. 1133 The AR sends RA messages to the MN via the ANET interface with an 1134 OMNI option with a Release ID for the failed MSE, and with 1135 destination address set to All-Nodes multicast (ff02::1) [RFC4291]. 1137 The AR SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated 1138 by small delays [RFC4861]. Any MNs on the ANET interface that have 1139 been using the (now defunct) MSE will receive the RA messages and 1140 associate with a new MSE. 1142 16. Transition Considerations 1144 When a MN connects to an ANET link for the first time, it sends an RS 1145 message with an OMNI option. If the first hop AR recognizes the 1146 option, it returns an RA with its MS OMNI LLA as the source, the MN 1147 OMNI LLA as the destination, the P(roxy) bit set in the RA flags and 1148 with an OMNI option included. The MN then engages the AR according 1149 to the OMNI link model specified above. If the first hop AR is a 1150 legacy IPv6 router, however, it instead returns an RA message with no 1151 OMNI option and with a non-OMNI unicast source LLA as specified in 1152 [RFC4861]. In that case, the MN engages the ANET according to the 1153 legacy IPv6 link model and without the OMNI extensions specified in 1154 this document. 1156 If the ANET link model is multiple access, there must be assurance 1157 that address duplication cannot corrupt the neighbor caches of other 1158 nodes on the link. When the MN sends an RS message on a multiple 1159 access ANET link with an OMNI LLA source address and an OMNI option, 1160 ARs that recognize the option ensure that the MN is authorized to use 1161 the address and return an RA with a non-zero Router Lifetime only if 1162 the MN is authorized. ARs that do not recognize the option instead 1163 return an RA that makes no statement about the MN's authorization to 1164 use the source address. In that case, the MN should perform 1165 Duplicate Address Detection to ensure that it does not interfere with 1166 other nodes on the link. 1168 An alternative approach for multiple access ANET links to ensure 1169 isolation for MN / AR communications is through L2 address mappings 1170 as discussed in Appendix C. This arrangement imparts a (virtual) 1171 point-to-point link model over the (physical) multiple access link. 1173 17. OMNI Interfaces on the Open Internet 1175 OMNI interfaces that connect to the open Internet via INET interfaces 1176 can apply symmetric security services such as VPNs to establish 1177 secured tunnels to MSEs. In environments where an explicit VPN may 1178 be too restrictive, OMNI interfaces can instead ensure neighbor cache 1179 integrity using SEcure Neighbor Discovery (SEND) [RFC3971] and 1180 Cryptographically Generated Addresses (CGAs) [RFC3972]. 1182 When SEND/CGA are used, the IPv6 ND control plane messages used to 1183 establish neighbor cache state are authenticated while data plane 1184 messages are delivered the same as for ordinary best-effort Internet 1185 traffic. Instead, data plane communications via OMNI interfaces that 1186 connect over the open Internet without an explicit VPN must emply 1187 transport- or higher-layer security to ensure integrity and/or 1188 confidentiality. 1190 In addition to secured OMNI interface RS/RA exchanges, SEND/CGA 1191 supports safe address resolution and neighbor unreachability 1192 detection as discused in Asymmetric Extended Route Optimization 1193 (AERO) [I-D.templin-intarea-6706bis]. This allows for efficient 1194 multilink operations over the open Internet with assured neighbor 1195 cache integrity. 1197 OMNI interfaces in the open Internet are often located behind Network 1198 Address Translators (NATs). The OMNI interface accommodates NAT 1199 traversal using the OMNI LLA prefix fe80::/32 for Teredo IPv6 1200 addresses formatted as discussed in Section 4 of [RFC4380]. Further 1201 specifications for NAT traversal are discussed in 1202 [I-D.templin-intarea-6706bis][RFC6081][RFC4380]. 1204 18. Time-Varying MNPs 1206 In some use cases, it is desireable, beneficial and efficient for the 1207 MN to receive a contstant MNP that travels with the MN wherever it 1208 moves. For example, this would allow air traffic controllers to 1209 easily track aircraft, etc. In other cases, however (e.g., 1210 intelligent transportation systems), the MN may be willing to 1211 sacrifice a modicum of efficiency in order to have time-varying MNPs 1212 that can be changed every so often to defeat adversarial tracking. 1214 Prefix delegation services such as those discussed in 1215 [I-D.templin-6man-dhcpv6-ndopt] and [I-D.templin-intarea-6706bis] 1216 allow OMNI MNs that desire time-varying MNPs to obtain short-lived 1217 prefixes. In that case, the identity of the MN would not be bound to 1218 the MNP but rather to the prefix delegation ID and used as the seed 1219 for Prefix Delegation. The MN would then be obligated to renumber 1220 its internal networks whenever its MNP (and therefore also its OMNI 1221 address) changes. This should not present a challenge for MNs with 1222 automated network renumbering services, however presents limits for 1223 the durations of ongoing sessions that would prefer to use a constant 1224 address. 1226 19. IANA Considerations 1228 The IANA is instructed to allocate an official Type number TBD from 1229 the registry "IPv6 Neighbor Discovery Option Formats" for the OMNI 1230 option. Implementations set Type to 253 as an interim value 1231 [RFC4727]. 1233 The IANA is instructed to allocate one Ethernet unicast address TBD2 1234 (suggest 00-00-5E-00-52-14 [RFC5214]) in the registry "IANA Ethernet 1235 Address Block - Unicast Use". 1237 The OMNI option also defines an 8-bit Sub-Type field, for which IANA 1238 is instructed to create and maintain a new registry entitled "OMNI 1239 option Sub-Type values". Initial values for the OMNI option Sub-Type 1240 values registry are given below; future assignments are to be made 1241 through Expert Review [RFC8126]. 1243 Value Sub-Type name Reference 1244 ----- ------------- ---------- 1245 0 Pad1 [RFCXXXX] 1246 1 PadN [RFCXXXX] 1247 2 ifIndex-tuple (Type 1) [RFCXXXX] 1248 3 ifIndex-tuple (Type 2) [RFCXXXX] 1249 4 MS-Register [RFCXXXX] 1250 5 MS-Release [RFCXXXX] 1251 6-252 Unassigned 1252 253-254 Experimental [RFCXXXX] 1253 255 Reserved [RFCXXXX] 1255 Figure 13: OMNI Option Sub-Type Values 1257 20. Security Considerations 1259 Security considerations for IPv6 [RFC8200] and IPv6 Neighbor 1260 Discovery [RFC4861] apply. OMNI interface IPv6 ND messages SHOULD 1261 include Nonce and Timestamp options [RFC3971] when synchronized 1262 transaction confirmation is needed. 1264 OMNI interfaces configured over secured underlying ANET interfaces 1265 inherit the physical and/or link-layer security aspects of the 1266 connected ANETs. OMNI interfaces configured over open Internet 1267 interfaces must use symmetric securing services such as VPNs or 1268 asymmetric services such as SEND/CGA [RFC3971][RFC3972]. 1270 Security considerations for specific access network interface types 1271 are covered under the corresponding IP-over-(foo) specification 1272 (e.g., [RFC2464], [RFC2492], etc.). 1274 21. Acknowledgements 1276 The first version of this document was prepared per the consensus 1277 decision at the 7th Conference of the International Civil Aviation 1278 Organization (ICAO) Working Group-I Mobility Subgroup on March 22, 1279 2019. Consensus to take the document forward to the IETF was reached 1280 at the 9th Conference of the Mobility Subgroup on November 22, 2019. 1281 Attendees and contributors included: Guray Acar, Danny Bharj, 1282 Francois D'Humieres, Pavel Drasil, Nikos Fistas, Giovanni Garofolo, 1283 Bernhard Haindl, Vaughn Maiolla, Tom McParland, Victor Moreno, Madhu 1284 Niraula, Brent Phillips, Liviu Popescu, Jacky Pouzet, Aloke Roy, Greg 1285 Saccone, Robert Segers, Michal Skorepa, Michel Solery, Stephane 1286 Tamalet, Fred Templin, Jean-Marc Vacher, Bela Varkonyi, Tony Whyman, 1287 Fryderyk Wrobel and Dongsong Zeng. 1289 The following individuals are acknowledged for their useful comments: 1290 Michael Matyas, Madhu Niraula, Greg Saccone, Stephane Tamalet, Eric 1291 Vyncke. Pavel Drasil, Zdenek Jaron and Michal Skorepa are recognized 1292 for their many helpful ideas and suggestions. 1294 This work is aligned with the NASA Safe Autonomous Systems Operation 1295 (SASO) program under NASA contract number NNA16BD84C. 1297 This work is aligned with the FAA as per the SE2025 contract number 1298 DTFAWA-15-D-00030. 1300 22. References 1302 22.1. Normative References 1304 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1305 Requirement Levels", BCP 14, RFC 2119, 1306 DOI 10.17487/RFC2119, March 1997, 1307 . 1309 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 1310 "Definition of the Differentiated Services Field (DS 1311 Field) in the IPv4 and IPv6 Headers", RFC 2474, 1312 DOI 10.17487/RFC2474, December 1998, 1313 . 1315 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 1316 "SEcure Neighbor Discovery (SEND)", RFC 3971, 1317 DOI 10.17487/RFC3971, March 2005, 1318 . 1320 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 1321 RFC 3972, DOI 10.17487/RFC3972, March 2005, 1322 . 1324 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 1325 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 1326 November 2005, . 1328 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 1329 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 1330 . 1332 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 1333 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 1334 2006, . 1336 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 1337 Control Message Protocol (ICMPv6) for the Internet 1338 Protocol Version 6 (IPv6) Specification", STD 89, 1339 RFC 4443, DOI 10.17487/RFC4443, March 2006, 1340 . 1342 [RFC4727] Fenner, B., "Experimental Values In IPv4, IPv6, ICMPv4, 1343 ICMPv6, UDP, and TCP Headers", RFC 4727, 1344 DOI 10.17487/RFC4727, November 2006, 1345 . 1347 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 1348 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 1349 DOI 10.17487/RFC4861, September 2007, 1350 . 1352 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 1353 Address Autoconfiguration", RFC 4862, 1354 DOI 10.17487/RFC4862, September 2007, 1355 . 1357 [RFC6088] Tsirtsis, G., Giarreta, G., Soliman, H., and N. Montavont, 1358 "Traffic Selectors for Flow Bindings", RFC 6088, 1359 DOI 10.17487/RFC6088, January 2011, 1360 . 1362 [RFC8028] Baker, F. and B. Carpenter, "First-Hop Router Selection by 1363 Hosts in a Multi-Prefix Network", RFC 8028, 1364 DOI 10.17487/RFC8028, November 2016, 1365 . 1367 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 1368 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 1369 May 2017, . 1371 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 1372 (IPv6) Specification", STD 86, RFC 8200, 1373 DOI 10.17487/RFC8200, July 2017, 1374 . 1376 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 1377 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 1378 DOI 10.17487/RFC8201, July 2017, 1379 . 1381 22.2. Informative References 1383 [I-D.templin-6man-dhcpv6-ndopt] 1384 Templin, F., "A Unified Stateful/Stateless Configuration 1385 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-09 1386 (work in progress), January 2020. 1388 [I-D.templin-intarea-6706bis] 1389 Templin, F., "Asymmetric Extended Route Optimization 1390 (AERO)", draft-templin-intarea-6706bis-45 (work in 1391 progress), April 2020. 1393 [RFC2225] Laubach, M. and J. Halpern, "Classical IP and ARP over 1394 ATM", RFC 2225, DOI 10.17487/RFC2225, April 1998, 1395 . 1397 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 1398 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 1399 . 1401 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 1402 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 1403 December 1998, . 1405 [RFC2492] Armitage, G., Schulter, P., and M. Jork, "IPv6 over ATM 1406 Networks", RFC 2492, DOI 10.17487/RFC2492, January 1999, 1407 . 1409 [RFC2863] McCloghrie, K. and F. Kastenholz, "The Interfaces Group 1410 MIB", RFC 2863, DOI 10.17487/RFC2863, June 2000, 1411 . 1413 [RFC3692] Narten, T., "Assigning Experimental and Testing Numbers 1414 Considered Useful", BCP 82, RFC 3692, 1415 DOI 10.17487/RFC3692, January 2004, 1416 . 1418 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 1419 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 1420 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 1421 RFC 3819, DOI 10.17487/RFC3819, July 2004, 1422 . 1424 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 1425 Network Address Translations (NATs)", RFC 4380, 1426 DOI 10.17487/RFC4380, February 2006, 1427 . 1429 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 1430 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 1431 2006, . 1433 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 1434 "Internet Group Management Protocol (IGMP) / Multicast 1435 Listener Discovery (MLD)-Based Multicast Forwarding 1436 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 1437 August 2006, . 1439 [RFC5213] Gundavelli, S., Ed., Leung, K., Devarapalli, V., 1440 Chowdhury, K., and B. Patil, "Proxy Mobile IPv6", 1441 RFC 5213, DOI 10.17487/RFC5213, August 2008, 1442 . 1444 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 1445 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 1446 DOI 10.17487/RFC5214, March 2008, 1447 . 1449 [RFC5798] Nadas, S., Ed., "Virtual Router Redundancy Protocol (VRRP) 1450 Version 3 for IPv4 and IPv6", RFC 5798, 1451 DOI 10.17487/RFC5798, March 2010, 1452 . 1454 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 1455 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 1456 . 1458 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 1459 DOI 10.17487/RFC6081, January 2011, 1460 . 1462 [RFC6543] Gundavelli, S., "Reserved IPv6 Interface Identifier for 1463 Proxy Mobile IPv6", RFC 6543, DOI 10.17487/RFC6543, May 1464 2012, . 1466 [RFC7084] Singh, H., Beebee, W., Donley, C., and B. Stark, "Basic 1467 Requirements for IPv6 Customer Edge Routers", RFC 7084, 1468 DOI 10.17487/RFC7084, November 2013, 1469 . 1471 [RFC7421] Carpenter, B., Ed., Chown, T., Gont, F., Jiang, S., 1472 Petrescu, A., and A. Yourtchenko, "Analysis of the 64-bit 1473 Boundary in IPv6 Addressing", RFC 7421, 1474 DOI 10.17487/RFC7421, January 2015, 1475 . 1477 [RFC7847] Melia, T., Ed. and S. Gundavelli, Ed., "Logical-Interface 1478 Support for IP Hosts with Multi-Access Support", RFC 7847, 1479 DOI 10.17487/RFC7847, May 2016, 1480 . 1482 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 1483 Writing an IANA Considerations Section in RFCs", BCP 26, 1484 RFC 8126, DOI 10.17487/RFC8126, June 2017, 1485 . 1487 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 1488 Decraene, B., Litkowski, S., and R. Shakir, "Segment 1489 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 1490 July 2018, . 1492 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 1493 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 1494 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 1495 . 1497 Appendix A. Type 1 ifIndex-tuple Traffic Classifier Preference Encoding 1499 Adaptation of the OMNI option Type 1 ifIndex-tuple's traffic 1500 classifier Bitmap to specific Internetworks such as the Aeronautical 1501 Telecommunications Network with Internet Protocol Services (ATN/IPS) 1502 may include link selection preferences based on other traffic 1503 classifiers (e.g., transport port numbers, etc.) in addition to the 1504 existing DSCP-based preferences. Nodes on specific Internetworks 1505 maintain a map of traffic classifiers to additional P[*] preference 1506 fields beyond the first 64. For example, TCP port 22 maps to P[67], 1507 TCP port 443 maps to P[70], UDP port 8060 maps to P[76], etc. 1509 Implementations use Simplex or Indexed encoding formats for P[*] 1510 encoding in order to encode a given set of traffic classifiers in the 1511 most efficient way. Some use cases may be more efficiently coded 1512 using Simplex form, while others may be more efficient using Indexed. 1513 Once a format is selected for preparation of a single ifIndex-tuple 1514 the same format must be used for the entire Sub-Option. Different 1515 Sub-Options may use different formats. 1517 The following figures show coding examples for various Simplex and 1518 Indexed formats: 1520 0 1 2 3 1521 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 1522 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1523 | Sub-Type=2 | Sub-length=4+N| ifIndex | ifType | 1524 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1525 | Provider ID | Link |S|0|RSV| Bitmap(0)=0xff|P00|P01|P02|P03| 1526 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1527 |P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15|P16|P17|P18|P19| 1528 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1529 |P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| Bitmap(1)=0xff| 1530 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1531 |P32|P33|P34|P35|P36|P37|P38|P39|P40|P41|P42|P43|P44|P45|P46|P47| 1532 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1533 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 1534 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1535 | Bitmap(2)=0xff|P64|P65|P67|P68| ... 1536 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1538 Figure 14: Example 1: Dense Simplex Encoding 1540 0 1 2 3 1541 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 1542 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1543 | Sub-Type=2 | Sub-length=4+N| ifIndex | ifType | 1544 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1545 | Provider ID | Link |S|0|RSV| Bitmap(0)=0x00| Bitmap(1)=0x0f| 1546 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1547 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 1548 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1549 | Bitmap(2)=0x00| Bitmap(3)=0x00| Bitmap(4)=0x00| Bitmap(5)=0x00| 1550 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1551 | Bitmap(6)=0xf0|192|193|194|195|196|197|198|199|200|201|202|203| 1552 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1553 |204|205|206|207| Bitmap(7)=0x00| Bitmap(8)=0x0f|272|273|274|275| 1554 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1555 |276|277|278|279|280|281|282|283|284|285|286|287| Bitmap(9)=0x00| 1556 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1557 |Bitmap(10)=0x00| ... 1558 +-+-+-+-+-+-+-+-+-+-+- 1560 Figure 15: Example 2: Sparse Simplex Encoding 1562 0 1 2 3 1563 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 1564 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1565 | Sub-Type=2 | Sub-length=4+N| ifIndex | ifType | 1566 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1567 | Provider ID | Link |S|1|RSV| Index = 0x00 | Bitmap = 0x80 | 1568 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1569 |P00|P01|P02|P03| Index = 0x01 | Bitmap = 0x01 |P60|P61|P62|P63| 1570 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1571 | Index = 0x10 | Bitmap = 0x80 |512|513|514|515| Index = 0x18 | 1572 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1573 | Bitmap = 0x01 |796|797|798|799| ... 1574 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1576 Figure 16: Example 3: Indexed Encoding 1578 Appendix B. VDL Mode 2 Considerations 1580 ICAO Doc 9776 is the "Technical Manual for VHF Data Link Mode 2" 1581 (VDLM2) that specifies an essential radio frequency data link service 1582 for aircraft and ground stations in worldwide civil aviation air 1583 traffic management. The VDLM2 link type is "multicast capable" 1584 [RFC4861], but with considerable differences from common multicast 1585 links such as Ethernet and IEEE 802.11. 1587 First, the VDLM2 link data rate is only 31.5Kbps - multiple orders of 1588 magnitude less than most modern wireless networking gear. Second, 1589 due to the low available link bandwidth only VDLM2 ground stations 1590 (i.e., and not aircraft) are permitted to send broadcasts, and even 1591 so only as compact layer 2 "beacons". Third, aircraft employ the 1592 services of ground stations by performing unicast RS/RA exchanges 1593 upon receipt of beacons instead of listening for multicast RA 1594 messages and/or sending multicast RS messages. 1596 This beacon-oriented unicast RS/RA approach is necessary to conserve 1597 the already-scarce available link bandwidth. Moreover, since the 1598 numbers of beaconing ground stations operating within a given spatial 1599 range must be kept as sparse as possible, it would not be feasible to 1600 have different classes of ground stations within the same region 1601 observing different protocols. It is therefore highly desirable that 1602 all ground stations observe a common language of RS/RA as specified 1603 in this document. 1605 Note that links of this nature may benefit from compression 1606 techniques that reduce the bandwidth necessary for conveying the same 1607 amount of data. The IETF lpwan working group is considering possible 1608 alternatives: [https://datatracker.ietf.org/wg/lpwan/documents]. 1610 Appendix C. MN / AR Isolation Through L2 Address Mapping 1612 Per [RFC4861], IPv6 ND messages may be sent to either a multicast or 1613 unicast link-scoped IPv6 destination address. However, IPv6 ND 1614 messaging should be coordinated between the MN and AR only without 1615 invoking other nodes on the ANET. This implies that MN / AR 1616 coordinations should be isolated and not overheard by other nodes on 1617 the link. 1619 To support MN / AR isolation on some ANET links, ARs can maintain an 1620 OMNI-specific unicast L2 address ("MSADDR"). For Ethernet-compatible 1621 ANETs, this specification reserves one Ethernet unicast address TBD2 1622 (see: Section 19). For non-Ethernet statically-addressed ANETs, 1623 MSADDR is reserved per the assigned numbers authority for the ANET 1624 addressing space. For still other ANETs, MSADDR may be dynamically 1625 discovered through other means, e.g., L2 beacons. 1627 MNs map the L3 addresses of all IPv6 ND messages they send (i.e., 1628 both multicast and unicast) to MSADDR instead of to an ordinary 1629 unicast or multicast L2 address. In this way, all of the MN's IPv6 1630 ND messages will be received by ARs that are configured to accept 1631 packets destined to MSADDR. Note that multiple ARs on the link could 1632 be configured to accept packets destined to MSADDR, e.g., as a basis 1633 for supporting redundancy. 1635 Therefore, ARs must accept and process packets destined to MSADDR, 1636 while all other devices must not process packets destined to MSADDR. 1637 This model has well-established operational experience in Proxy 1638 Mobile IPv6 (PMIP) [RFC5213][RFC6543]. 1640 Appendix D. Change Log 1642 << RFC Editor - remove prior to publication >> 1644 Differences from draft-templin-6man-omni-interface-14 to draft- 1645 templin-6man-omni-interface-15: 1647 o Prefix length discussions removed. 1649 Differences from draft-templin-6man-omni-interface-12 to draft- 1650 templin-6man-omni-interface-13: 1652 o Teredo 1654 Differences from draft-templin-6man-omni-interface-11 to draft- 1655 templin-6man-omni-interface-12: 1657 o Major simplifications and clarifications on MTU and fragmentation. 1659 o Document now udates RFC4443 and RFC8201. 1661 Differences from draft-templin-6man-omni-interface-10 to draft- 1662 templin-6man-omni-interface-11: 1664 o Removed /64 assumption, resulting in new OMNI address format. 1666 Differences from draft-templin-6man-omni-interface-07 to draft- 1667 templin-6man-omni-interface-08: 1669 o OMNI MNs in the open Internet 1671 Differences from draft-templin-6man-omni-interface-06 to draft- 1672 templin-6man-omni-interface-07: 1674 o Brought back L2 MSADDR mapping text for MN / AR isolation based on 1675 L2 addressing. 1677 o Explanded "Transition Considerations". 1679 Differences from draft-templin-6man-omni-interface-05 to draft- 1680 templin-6man-omni-interface-06: 1682 o Brought back OMNI option "R" flag, and dicussed its use. 1684 Differences from draft-templin-6man-omni-interface-04 to draft- 1685 templin-6man-omni-interface-05: 1687 o Transition considerations, and overhaul of RS/RA addressing with 1688 the inclusion of MSE addresses within the OMNI option instead of 1689 as RS/RA addresses (developed under FAA SE2025 contract number 1690 DTFAWA-15-D-00030). 1692 Differences from draft-templin-6man-omni-interface-02 to draft- 1693 templin-6man-omni-interface-03: 1695 o Added "advisory PTB messages" under FAA SE2025 contract number 1696 DTFAWA-15-D-00030. 1698 Differences from draft-templin-6man-omni-interface-01 to draft- 1699 templin-6man-omni-interface-02: 1701 o Removed "Primary" flag and supporting text. 1703 o Clarified that "Router Lifetime" applies to each ANET interface 1704 independently, and that the union of all ANET interface Router 1705 Lifetimes determines MSE lifetime. 1707 Differences from draft-templin-6man-omni-interface-00 to draft- 1708 templin-6man-omni-interface-01: 1710 o "All-MSEs" OMNI LLA defined. Also reserverd fe80::ff00:0000/104 1711 for future use (most likely as "pseudo-multicast"). 1713 o Non-normative discussion of alternate OMNI LLA construction form 1714 made possible if the 64-bit assumption were relaxed. 1716 Differences from draft-templin-atn-aero-interface-21 to draft- 1717 templin-6man-omni-interface-00: 1719 o Minor clarification on Type-2 ifIndex-tuple encoding. 1721 o Draft filename change (replaces draft-templin-atn-aero-interface). 1723 Differences from draft-templin-atn-aero-interface-20 to draft- 1724 templin-atn-aero-interface-21: 1726 o OMNI option format 1728 o MTU 1730 Differences from draft-templin-atn-aero-interface-19 to draft- 1731 templin-atn-aero-interface-20: 1733 o MTU 1735 Differences from draft-templin-atn-aero-interface-18 to draft- 1736 templin-atn-aero-interface-19: 1738 o MTU 1740 Differences from draft-templin-atn-aero-interface-17 to draft- 1741 templin-atn-aero-interface-18: 1743 o MTU and RA configuration information updated. 1745 Differences from draft-templin-atn-aero-interface-16 to draft- 1746 templin-atn-aero-interface-17: 1748 o New "Primary" flag in OMNI option. 1750 Differences from draft-templin-atn-aero-interface-15 to draft- 1751 templin-atn-aero-interface-16: 1753 o New note on MSE OMNI LLA uniqueness assurance. 1755 o General cleanup. 1757 Differences from draft-templin-atn-aero-interface-14 to draft- 1758 templin-atn-aero-interface-15: 1760 o General cleanup. 1762 Differences from draft-templin-atn-aero-interface-13 to draft- 1763 templin-atn-aero-interface-14: 1765 o General cleanup. 1767 Differences from draft-templin-atn-aero-interface-12 to draft- 1768 templin-atn-aero-interface-13: 1770 o Minor re-work on "Notify-MSE" (changed to Notification ID). 1772 Differences from draft-templin-atn-aero-interface-11 to draft- 1773 templin-atn-aero-interface-12: 1775 o Removed "Request/Response" OMNI option formats. Now, there is 1776 only one OMNI option format that applies to all ND messages. 1778 o Added new OMNI option field and supporting text for "Notify-MSE". 1780 Differences from draft-templin-atn-aero-interface-10 to draft- 1781 templin-atn-aero-interface-11: 1783 o Changed name from "aero" to "OMNI" 1785 o Resolved AD review comments from Eric Vyncke (posted to atn list) 1787 Differences from draft-templin-atn-aero-interface-09 to draft- 1788 templin-atn-aero-interface-10: 1790 o Renamed ARO option to AERO option 1792 o Re-worked Section 13 text to discuss proactive NUD. 1794 Differences from draft-templin-atn-aero-interface-08 to draft- 1795 templin-atn-aero-interface-09: 1797 o Version and reference update 1799 Differences from draft-templin-atn-aero-interface-07 to draft- 1800 templin-atn-aero-interface-08: 1802 o Removed "Classic" and "MS-enabled" link model discussion 1804 o Added new figure for MN/AR/MSE model. 1806 o New Section on "Detecting and responding to MSE failure". 1808 Differences from draft-templin-atn-aero-interface-06 to draft- 1809 templin-atn-aero-interface-07: 1811 o Removed "nonce" field from AR option format. Applications that 1812 require a nonce can include a standard nonce option if they want 1813 to. 1815 o Various editorial cleanups. 1817 Differences from draft-templin-atn-aero-interface-05 to draft- 1818 templin-atn-aero-interface-06: 1820 o New Appendix C on "VDL Mode 2 Considerations" 1822 o New Appendix D on "RS/RA Messaging as a Single Standard API" 1824 o Various significant updates in Section 5, 10 and 12. 1826 Differences from draft-templin-atn-aero-interface-04 to draft- 1827 templin-atn-aero-interface-05: 1829 o Introduced RFC6543 precedent for focusing IPv6 ND messaging to a 1830 reserved unicast link-layer address 1832 o Introduced new IPv6 ND option for Aero Registration 1834 o Specification of MN-to-MSE message exchanges via the ANET access 1835 router as a proxy 1837 o IANA Considerations updated to include registration requests and 1838 set interim RFC4727 option type value. 1840 Differences from draft-templin-atn-aero-interface-03 to draft- 1841 templin-atn-aero-interface-04: 1843 o Removed MNP from aero option format - we already have RIOs and 1844 PIOs, and so do not need another option type to include a Prefix. 1846 o Clarified that the RA message response must include an aero option 1847 to indicate to the MN that the ANET provides a MS. 1849 o MTU interactions with link adaptation clarified. 1851 Differences from draft-templin-atn-aero-interface-02 to draft- 1852 templin-atn-aero-interface-03: 1854 o Sections re-arranged to match RFC4861 structure. 1856 o Multiple aero interfaces 1858 o Conceptual sending algorithm 1860 Differences from draft-templin-atn-aero-interface-01 to draft- 1861 templin-atn-aero-interface-02: 1863 o Removed discussion of encapsulation (out of scope) 1865 o Simplified MTU section 1867 o Changed to use a new IPv6 ND option (the "aero option") instead of 1868 S/TLLAO 1870 o Explained the nature of the interaction between the mobility 1871 management service and the air interface 1873 Differences from draft-templin-atn-aero-interface-00 to draft- 1874 templin-atn-aero-interface-01: 1876 o Updates based on list review comments on IETF 'atn' list from 1877 4/29/2019 through 5/7/2019 (issue tracker established) 1879 o added list of opportunities afforded by the single virtual link 1880 model 1882 o added discussion of encapsulation considerations to Section 6 1884 o noted that DupAddrDetectTransmits is set to 0 1886 o removed discussion of IPv6 ND options for prefix assertions. The 1887 aero address already includes the MNP, and there are many good 1888 reasons for it to continue to do so. Therefore, also including 1889 the MNP in an IPv6 ND option would be redundant. 1891 o Significant re-work of "Router Discovery" section. 1893 o New Appendix B on Prefix Length considerations 1895 First draft version (draft-templin-atn-aero-interface-00): 1897 o Draft based on consensus decision of ICAO Working Group I Mobility 1898 Subgroup March 22, 2019. 1900 Authors' Addresses 1902 Fred L. Templin (editor) 1903 The Boeing Company 1904 P.O. Box 3707 1905 Seattle, WA 98124 1906 USA 1908 Email: fltemplin@acm.org 1910 Tony Whyman 1911 MWA Ltd c/o Inmarsat Global Ltd 1912 99 City Road 1913 London EC1Y 1AX 1914 England 1916 Email: tony.whyman@mccallumwhyman.com