idnits 2.17.1 draft-templin-6man-omni-07.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 : ---------------------------------------------------------------------------- ** There are 3 instances of too long lines in the document, the longest one being 1 character in excess of 72. == There are 1 instance of lines with non-RFC6890-compliant IPv4 addresses in the document. If these are example addresses, they should be changed. == There are 1 instance of lines with private range IPv4 addresses in the document. If these are generic example addresses, they should be changed to use any of the ranges defined in RFC 6890 (or successor): 192.0.2.x, 198.51.100.x or 203.0.113.x. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (May 16, 2021) is 1075 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: 'ULA' is mentioned on line 2949, but not defined -- Looks like a reference, but probably isn't: '1' on line 4570 -- Looks like a reference, but probably isn't: '2' on line 4564 -- Looks like a reference, but probably isn't: '3' on line 2371 == Missing Reference: 'RFCXXXX' is mentioned on line 3909, but not defined == Missing Reference: 'N' is mentioned on line 4573, but not defined -- Looks like a reference, but probably isn't: '40' on line 4570 -- Looks like a reference, but probably isn't: '41' on line 4572 == Missing Reference: 'N-2' is mentioned on line 4572, but not defined == Missing Reference: 'N-1' is mentioned on line 4573, but not defined == Unused Reference: 'RFC2474' is defined on line 4098, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-tsvwg-udp-options' is defined on line 4209, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 4218, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-lla-type' is defined on line 4223, but no explicit reference was found in the text == Unused Reference: 'RFC2225' is defined on line 4270, but no explicit reference was found in the text == Unused Reference: 'RFC2529' is defined on line 4286, but no explicit reference was found in the text == Unused Reference: 'RFC3879' is defined on line 4328, but no explicit reference was found in the text == Unused Reference: 'RFC4271' is defined on line 4337, but no explicit reference was found in the text == Unused Reference: 'RFC4389' is defined on line 4351, but no explicit reference was found in the text == Unused Reference: 'RFC5175' is defined on line 4380, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 4423, but no explicit reference was found in the text == Unused Reference: 'RFC8754' is defined on line 4516, but no explicit reference was found in the text ** Obsolete normative reference: RFC 793 (Obsoleted by RFC 9293) == Outdated reference: A later version (-37) exists of draft-ietf-drip-rid-07 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-10 == Outdated reference: A later version (-30) exists of draft-ietf-ipwave-vehicular-networking-20 == Outdated reference: A later version (-32) exists of draft-ietf-tsvwg-udp-options-12 == Outdated reference: A later version (-63) exists of draft-templin-6man-aero-01 -- Obsolete informational reference (is this intentional?): RFC 1146 (Obsoleted by RFC 6247) -- Obsolete informational reference (is this intentional?): RFC 3330 (Obsoleted by RFC 5735) -- Obsolete informational reference (is this intentional?): RFC 7042 (Obsoleted by RFC 9542) Summary: 2 errors (**), 0 flaws (~~), 25 warnings (==), 10 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: Informational A. Whyman 5 Expires: November 17, 2021 MWA Ltd c/o Inmarsat Global Ltd 6 May 16, 2021 8 Transmission of IP Packets over Overlay Multilink Network (OMNI) 9 Interfaces 10 draft-templin-6man-omni-07 12 Abstract 14 Mobile nodes (e.g., aircraft of various configurations, terrestrial 15 vehicles, seagoing vessels, enterprise wireless 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 presented that 19 enables mobile nodes to coordinate with a network-based mobility 20 service and/or with other mobile node peers. This document specifies 21 the transmission of IP packets over Overlay Multilink Network (OMNI) 22 Interfaces. 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 November 17, 2021. 41 Copyright Notice 43 Copyright (c) 2021 IETF Trust and the persons identified as the 44 document authors. All rights reserved. 46 This document is subject to BCP 78 and the IETF Trust's Legal 47 Provisions Relating to IETF Documents 48 (https://trustee.ietf.org/license-info) in effect on the date of 49 publication of this document. Please review these documents 50 carefully, as they describe your rights and restrictions with respect 51 to this document. Code Components extracted from this document must 52 include Simplified BSD License text as described in Section 4.e of 53 the Trust Legal Provisions and are provided without warranty as 54 described in the Simplified BSD License. 56 Table of Contents 58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 59 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 6 60 3. Requirements . . . . . . . . . . . . . . . . . . . . . . . . 11 61 4. Overlay Multilink Network (OMNI) Interface Model . . . . . . 11 62 5. OMNI Interface Maximum Transmission Unit (MTU) . . . . . . . 17 63 6. The OMNI Adaptation Layer (OAL) . . . . . . . . . . . . . . . 18 64 6.1. OAL Source Encapsulation and Fragmentation . . . . . . . 18 65 6.2. OAL *NET Encapsulation and Re-Encapsulation . . . . . . . 23 66 6.3. OAL Destination Decapsulation and Reassembly . . . . . . 25 67 6.4. OAL Header Compression . . . . . . . . . . . . . . . . . 25 68 6.5. OAL Fragment Identification Window Maintenance . . . . . 28 69 6.6. OAL Fragment Retransmission . . . . . . . . . . . . . . . 30 70 6.7. OAL MTU Feedback Messaging . . . . . . . . . . . . . . . 31 71 6.8. OAL Requirements . . . . . . . . . . . . . . . . . . . . 33 72 6.9. OAL Fragmentation Security Implications . . . . . . . . . 34 73 6.10. OAL Super-Packets . . . . . . . . . . . . . . . . . . . . 35 74 7. Frame Format . . . . . . . . . . . . . . . . . . . . . . . . 37 75 8. Link-Local Addresses (LLAs) . . . . . . . . . . . . . . . . . 37 76 9. Unique-Local Addresses (ULAs) . . . . . . . . . . . . . . . . 39 77 10. Global Unicast Addresses (GUAs) . . . . . . . . . . . . . . . 40 78 11. Node Identification . . . . . . . . . . . . . . . . . . . . . 41 79 12. Address Mapping - Unicast . . . . . . . . . . . . . . . . . . 41 80 12.1. Sub-Options . . . . . . . . . . . . . . . . . . . . . . 44 81 12.1.1. Pad1 . . . . . . . . . . . . . . . . . . . . . . . . 46 82 12.1.2. PadN . . . . . . . . . . . . . . . . . . . . . . . . 46 83 12.1.3. Interface Attributes (Types 1 through 3) . . . . . . 46 84 12.1.4. Interface Attributes (Type 4) . . . . . . . . . . . 47 85 12.1.5. MS-Register . . . . . . . . . . . . . . . . . . . . 50 86 12.1.6. MS-Release . . . . . . . . . . . . . . . . . . . . . 50 87 12.1.7. Geo Coordinates . . . . . . . . . . . . . . . . . . 51 88 12.1.8. Dynamic Host Configuration Protocol for IPv6 89 (DHCPv6) Message . . . . . . . . . . . . . . . . . . 52 90 12.1.9. Host Identity Protocol (HIP) Message . . . . . . . . 53 91 12.1.10. PIM-SM Message . . . . . . . . . . . . . . . . . . . 54 92 12.1.11. Reassembly Limit . . . . . . . . . . . . . . . . . . 55 93 12.1.12. Fragmentation Report . . . . . . . . . . . . . . . . 56 94 12.1.13. Node Identification . . . . . . . . . . . . . . . . 57 95 12.1.14. Sub-Type Extension . . . . . . . . . . . . . . . . . 59 96 13. Address Mapping - Multicast . . . . . . . . . . . . . . . . . 62 97 14. Multilink Conceptual Sending Algorithm . . . . . . . . . . . 62 98 14.1. Multiple OMNI Interfaces . . . . . . . . . . . . . . . . 63 99 14.2. MN<->AR Traffic Loop Prevention . . . . . . . . . . . . 63 100 15. Router Discovery and Prefix Registration . . . . . . . . . . 64 101 15.1. Router Discovery in IP Multihop and IPv4-Only Networks . 68 102 15.2. MS-Register and MS-Release List Processing . . . . . . . 70 103 15.3. DHCPv6-based Prefix Registration . . . . . . . . . . . . 72 104 16. Secure Redirection . . . . . . . . . . . . . . . . . . . . . 73 105 17. AR and MSE Resilience . . . . . . . . . . . . . . . . . . . . 74 106 18. Detecting and Responding to MSE Failures . . . . . . . . . . 74 107 19. Transition Considerations . . . . . . . . . . . . . . . . . . 75 108 20. OMNI Interfaces on Open Internetworks . . . . . . . . . . . . 75 109 21. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . . . 77 110 22. (H)HITs and Temporary ULAs . . . . . . . . . . . . . . . . . 78 111 23. Address Selection . . . . . . . . . . . . . . . . . . . . . . 79 112 24. Error Messages . . . . . . . . . . . . . . . . . . . . . . . 79 113 25. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 79 114 25.1. "IEEE 802 Numbers" Registry . . . . . . . . . . . . . . 80 115 25.2. "IPv6 Neighbor Discovery Option Formats" Registry . . . 80 116 25.3. "Ethernet Numbers" Registry . . . . . . . . . . . . . . 80 117 25.4. "ICMPv6 Code Fields: Type 2 - Packet Too Big" Registry . 80 118 25.5. "OMNI Option Sub-Type Values" (New Registry) . . . . . . 81 119 25.6. "OMNI Geo Coordinates Type Values" (New Registry) . . . 81 120 25.7. "OMNI Node Identification ID-Type Values" (New Registry) 81 121 25.8. "OMNI Option Sub-Type Extension Values" (New Registry) . 82 122 25.9. "OMNI RFC4380 UDP/IP Header Option" (New Registry) . . . 82 123 25.10. "OMNI RFC6081 UDP/IP Trailer Option" (New Registry) . . 83 124 25.11. Additional Considerations . . . . . . . . . . . . . . . 83 125 26. Security Considerations . . . . . . . . . . . . . . . . . . . 84 126 27. Implementation Status . . . . . . . . . . . . . . . . . . . . 85 127 28. Document Updates . . . . . . . . . . . . . . . . . . . . . . 85 128 29. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 85 129 30. References . . . . . . . . . . . . . . . . . . . . . . . . . 87 130 30.1. Normative References . . . . . . . . . . . . . . . . . . 87 131 30.2. Informative References . . . . . . . . . . . . . . . . . 89 132 Appendix A. OAL Checksum Algorithm . . . . . . . . . . . . . . . 97 133 Appendix B. VDL Mode 2 Considerations . . . . . . . . . . . . . 97 134 Appendix C. MN / AR Isolation Through L2 Address Mapping . . . . 98 135 Appendix D. Change Log . . . . . . . . . . . . . . . . . . . . . 99 136 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 100 138 1. Introduction 140 Mobile Nodes (MNs) (e.g., aircraft of various configurations, 141 terrestrial vehicles, seagoing vessels, enterprise wireless devices, 142 pedestrians with cellphones, etc.) often have multiple interface 143 connections to wireless and/or wired-line data links used for 144 communicating with networked correspondents. These data links may 145 have diverse performance, cost and availability properties that can 146 change dynamically according to mobility patterns, flight phases, 147 proximity to infrastructure, etc. MNs coordinate their data links in 148 a discipline known as "multilink", in which a single virtual 149 interface is configured over the node's underlying interface 150 connections to the data links. 152 The MN configures a virtual interface (termed the "Overlay Multilink 153 Network Interface (OMNI)") as a thin layer over the underlying 154 interfaces. The OMNI interface is therefore the only interface 155 abstraction exposed to the IP layer and behaves according to the Non- 156 Broadcast, Multiple Access (NBMA) interface principle, while 157 underlying interfaces appear as link layer communication channels in 158 the architecture. The OMNI interface internally employs the "OMNI 159 Adaptation Layer (OAL)" to ensure that original IP packets are 160 delivered without loss due to size restrictions. The OMNI interface 161 connects to a virtual overlay service known as the "OMNI link". The 162 OMNI link spans one or more Internetworks that may include private- 163 use infrastructures and/or the global public Internet itself. 165 Each MN receives a Mobile Network Prefix (MNP) for numbering 166 downstream-attached End User Networks (EUNs) independently of the 167 access network data links selected for data transport. The MN 168 performs router discovery over the OMNI interface (i.e., similar to 169 IPv6 customer edge routers [RFC7084]) and acts as a mobile router on 170 behalf of its EUNs. The router discovery process is iterated over 171 each of the OMNI interface's underlying interfaces in order to 172 register per-link parameters (see Section 15). 174 The OMNI interface provides a multilink nexus for exchanging inbound 175 and outbound traffic via the correct underlying interface(s). The IP 176 layer sees the OMNI interface as a point of connection to the OMNI 177 link. Each OMNI link has one or more associated Mobility Service 178 Prefixes (MSPs), which are typically IP Global Unicast Address (GUA) 179 prefixes from which MNPs are derived. If there are multiple OMNI 180 links, the IPv6 layer will see multiple OMNI interfaces. 182 MNs may connect to multiple distinct OMNI links within the same OMNI 183 domain by configuring multiple OMNI interfaces, e.g., omni0, omni1, 184 omni2, etc. Each OMNI interface is configured over a set of 185 underlying interfaces and provides a nexus for Safety-Based Multilink 186 (SBM) operation. Each OMNI interface within the same OMNI domain 187 configures a common ULA prefix [ULA]::/48, and configures a unique 188 16-bit Subnet ID '*' to construct the sub-prefix [ULA*]::/64 (see: 189 Section 9). The IP layer applies SBM routing to select an OMNI 190 interface, which then applies Performance-Based Multilink (PBM) to 191 select the correct underlying interface. Applications can apply 192 Segment Routing [RFC8402] to select independent SBM topologies for 193 fault tolerance. 195 The OMNI interface interacts with a network-based Mobility Service 196 (MS) through IPv6 Neighbor Discovery (ND) control message exchanges 197 [RFC4861]. The MS provides Mobility Service Endpoints (MSEs) that 198 track MN movements and represent their MNPs in a global routing or 199 mapping system. 201 Many OMNI use cases have been proposed. In particular, the 202 International Civil Aviation Organization (ICAO) Working Group-I 203 Mobility Subgroup is developing a future Aeronautical 204 Telecommunications Network with Internet Protocol Services (ATN/IPS) 205 and has issued a liaison statement requesting IETF adoption [ATN] in 206 support of ICAO Document 9896 [ATN-IPS]. The IETF IP Wireless Access 207 in Vehicular Environments (ipwave) working group has further included 208 problem statement and use case analysis for OMNI in a document now in 209 AD evaluation for RFC publication 210 [I-D.ietf-ipwave-vehicular-networking]. Still other communities of 211 interest include AEEC, RTCA Special Committee 228 (SC-228) and NASA 212 programs that examine commercial aviation, Urban Air Mobility (UAM) 213 and Unmanned Air Systems (UAS). Pedestrians with handheld devices 214 represent another large class of potential OMNI users. 216 In addition to many other aspects, OMNI supports the "6M's" of modern 217 Internetworking including: 219 1. Multilink - a mobile node's ability to coordinate multiple 220 diverse underlying data links as a single logical unit (i.e., the 221 OMNI interface) to achieve the required communications 222 performance and reliability objectives. 224 2. Multinet - the ability to span the OMNI link across multiple 225 diverse network administrative segments while maintaining 226 seamless end-to-end communications between mobile nodes and 227 correspondents such as air traffic controllers, fleet 228 administrators, etc. 230 3. Mobility - a mobile node's ability to change network points of 231 attachment (e.g., moving between wireless base stations) which 232 may result in an underlying interface address change, but without 233 disruptions to ongoing communication sessions with peers over the 234 OMNI link. 236 4. Multicast - the ability to send a single network transmission 237 that reaches multiple nodes belonging to the same interest group, 238 but without disturbing other nodes not subscribed to the interest 239 group. 241 5. Multihop - a mobile node vehicle-to-vehicle relaying capability 242 useful when multiple forwarding hops between vehicles may be 243 necessary to "reach back" to an infrastructure access point 244 connection to the OMNI link. 246 6. MTU assurance - the ability to deliver packets of various robust 247 sizes between peers without loss due to a link size restriction, 248 and to dynamically adjust packets sizes to achieve the optimal 249 performance for each independent traffic flow. 251 This document specifies the transmission of IP packets and MN/MS 252 control messages over OMNI interfaces. The OMNI interface supports 253 either IP protocol version (i.e., IPv4 [RFC0791] or IPv6 [RFC8200]) 254 as the network layer in the data plane, while using IPv6 ND messaging 255 as the control plane independently of the data plane IP protocol(s). 256 The OAL operates as a sublayer between L3 and L2 based on IPv6 257 encapsulation [RFC2473] as discussed in the following sections. 259 2. Terminology 261 The terminology in the normative references applies; especially, the 262 terms "link" and "interface" are the same as defined in the IPv6 263 [RFC8200] and IPv6 Neighbor Discovery (ND) [RFC4861] specifications. 264 Additionally, this document assumes the following IPv6 ND message 265 types: Router Solicitation (RS), Router Advertisement (RA), Neighbor 266 Solicitation (NS), Neighbor Advertisement (NA) and Redirect. 268 The Protocol Constants defined in Section 10 of [RFC4861] are used in 269 their same format and meaning in this document. The terms "All- 270 Routers multicast", "All-Nodes multicast" and "Subnet-Router anycast" 271 are the same as defined in [RFC4291] (with Link-Local scope assumed). 273 The term "IP" is used to refer collectively to either Internet 274 Protocol version (i.e., IPv4 [RFC0791] or IPv6 [RFC8200]) when a 275 specification at the layer in question applies equally to either 276 version. 278 The following terms are defined within the scope of this document: 280 Mobile Node (MN) 281 an end system with a mobile router having multiple distinct 282 upstream data link connections that are grouped together in one or 283 more logical units. The MN's data link connection parameters can 284 change over time due to, e.g., node mobility, link quality, etc. 285 The MN further connects a downstream-attached End User Network 286 (EUN). The term MN used here is distinct from uses in other 287 documents, and does not imply a particular mobility protocol. 289 End User Network (EUN) 290 a simple or complex downstream-attached mobile network that 291 travels with the MN as a single logical unit. The IP addresses 292 assigned to EUN devices remain stable even if the MN's upstream 293 data link connections change. 295 Mobility Service (MS) 296 a mobile routing service that tracks MN movements and ensures that 297 MNs remain continuously reachable even across mobility events. 298 Specific MS details are out of scope for this document. 300 Mobility Service Endpoint (MSE) 301 an entity in the MS (either singular or aggregate) that 302 coordinates the mobility events of one or more MN. 304 Mobility Service Prefix (MSP) 305 an aggregated IP Global Unicast Address (GUA) prefix (e.g., 306 2001:db8::/32, 192.0.2.0/24, etc.) assigned to the OMNI link and 307 from which more-specific Mobile Network Prefixes (MNPs) are 308 delegated. OMNI link administrators typically obtain MSPs from an 309 Internet address registry, however private-use prefixes can 310 alternatively be used subject to certain limitations (see: 311 Section 10). OMNI links that connect to the global Internet 312 advertise their MSPs to their interdomain routing peers. 314 Mobile Network Prefix (MNP) 315 a longer IP prefix delegated from an MSP (e.g., 316 2001:db8:1000:2000::/56, 192.0.2.8/30, etc.) and assigned to a MN. 317 MNs sub-delegate the MNP to devices located in EUNs. Note that 318 OMNI link Relay nodes may also service non-MNP routes (i.e., GUA 319 prefixes not covered by an MSP) but that these correspond to fixed 320 correspondent nodes and not MNs. Other than this distinction, MNP 321 and non-MNP routes are treated exactly the same by the OMNI 322 routing system. 324 Access Network (ANET) 325 a data link service network (e.g., an aviation radio access 326 network, satellite service provider network, cellular operator 327 network, WiFi network, etc.) that connects MNs. Physical and/or 328 data link level security is assumed, and sometimes referred to as 329 "protected spectrum". Private enterprise networks and ground 330 domain aviation service networks may provide multiple secured IP 331 hops between the MN's point of connection and the nearest Access 332 Router. 334 Access Router (AR) 335 a router in the ANET for connecting MNs to correspondents in 336 outside Internetworks. The AR may be located on the same physical 337 link as the MN, or may be located multiple IP hops away. In the 338 latter case, the MN uses encapsulation to communicate with the AR 339 as though it were on the same physical link. 341 ANET interface 342 a MN's attachment to a link in an ANET. 344 Internetwork (INET) 345 a connected network region with a coherent IP addressing plan that 346 provides transit forwarding services between ANETs and nodes that 347 connect directly to the open INET via unprotected media. No 348 physical and/or data link level security is assumed, therefore 349 security must be applied by upper layers. The global public 350 Internet itself is an example. 352 INET interface 353 a node's attachment to a link in an INET. 355 *NET 356 a "wildcard" term used when a given specification applies equally 357 to both ANET and INET cases. 359 OMNI link 360 a Non-Broadcast, Multiple Access (NBMA) virtual overlay configured 361 over one or more INETs and their connected ANETs. An OMNI link 362 can comprise multiple INET segments joined by bridges the same as 363 for any link; the addressing plans in each segment may be mutually 364 exclusive and managed by different administrative entities. 366 OMNI interface 367 a node's attachment to an OMNI link, and configured over one or 368 more underlying *NET interfaces. If there are multiple OMNI links 369 in an OMNI domain, a separate OMNI interface is configured for 370 each link. 372 OMNI Adaptation Layer (OAL) 373 an OMNI interface sublayer service whereby original IP packets 374 admitted into the interface are wrapped in an IPv6 header and 375 subject to fragmentation and reassembly. The OAL is also 376 responsible for generating MTU-related control messages as 377 necessary, and for providing addressing context for spanning 378 multiple segments of a bridged OMNI link. 380 original IP packet 381 a whole IP packet or fragment admitted into the OMNI interface by 382 the network layer prior to OAL encapsulation and fragmentation, or 383 an IP packet delivered to the network layer by the OMNI interface 384 following OAL decapsulation and reassembly. 386 OAL packet 387 an original IP packet encapsulated in OAL headers and trailers 388 before OAL fragmentation, or following OAL reassembly. 390 OAL fragment 391 a portion of an OAL packet following fragmentation but prior to 392 *NET encapsulation, or following *NET encapsulation but prior to 393 OAL reassembly. 395 (OAL) atomic fragment 396 an OAL packet that does not require fragmentation is always 397 encapsulated as an "atomic fragment" with a Fragment Header with 398 Fragment Offset and More Fragments both set to 0, but with a valid 399 Identification value. 401 (OAL) carrier packet 402 an encapsulated OAL fragment following *NET encapsulation or prior 403 to *NET decapsulation. OAL sources and destinations exchange 404 carrier packets over underlying interfaces, and may be separated 405 by one or more OAL intermediate nodes. OAL intermediate nodes may 406 perform re-encapsulation on carrier packets by removing the *NET 407 headers of the first hop network and replacing them with new *NET 408 headers for the next hop network. 410 OAL source 411 an OMNI interface acts as an OAL source when it encapsulates 412 original IP packets to form OAL packets, then performs OAL 413 fragmentation and *NET encapsulation to create carrier packets. 415 OAL destination 416 an OMNI interface acts as an OAL destination when it decapsulates 417 carrier packets, then performs OAL reassembly and decapsulation to 418 derive the original IP packet. 420 OAL intermediate node 421 an OMNI interface acts as an OAL intermediate node when it removes 422 the *NET headers of carrier packets received on a first segment, 423 then re-encapsulates the carrier packets in new *NET headers and 424 forwards them into the next segment. 426 OMNI Option 427 an IPv6 Neighbor Discovery option providing multilink parameters 428 for the OMNI interface as specified in Section 12. 430 Mobile Network Prefix Link Local Address (MNP-LLA) 431 an IPv6 Link Local Address that embeds the most significant 64 432 bits of an MNP in the lower 64 bits of fe80::/64, as specified in 433 Section 8. 435 Mobile Network Prefix Unique Local Address (MNP-ULA) 436 an IPv6 Unique-Local Address derived from an MNP-LLA. 438 Administrative Link Local Address (ADM-LLA) 439 an IPv6 Link Local Address that embeds a 32-bit administratively- 440 assigned identification value in the lower 32 bits of fe80::/96, 441 as specified in Section 8. 443 Administrative Unique Local Address (ADM-ULA) 444 an IPv6 Unique-Local Address derived from an ADM-LLA. 446 Multilink 447 an OMNI interface's manner of managing diverse underlying 448 interface connections to data links as a single logical unit. The 449 OMNI interface provides a single unified interface to upper 450 layers, while underlying interface selections are performed on a 451 per-packet basis considering factors such as DSCP, flow label, 452 application policy, signal quality, cost, etc. Multilinking 453 decisions are coordinated in both the outbound (i.e. MN to 454 correspondent) and inbound (i.e., correspondent to MN) directions. 456 Multinet 457 an OAL intermediate node's manner of bridging multiple diverse IP 458 Internetworks and/or private enterprise networks at the OAL layer 459 below IP. Through intermediate node concatenation of bridged 460 network segments in this way, multiple diverse Internetworks (such 461 as the global public IPv4 and IPv6 Internets) can serve as transit 462 segments in a bridged path for forwarding IP packets end-to-end. 463 This bridging capability provide benefits such as supporting IPv4/ 464 IPv6 transition and coexistence, joining multiple diverse operator 465 networks into a cooperative single service network, etc. 467 Multihop 468 an iterative relaying of IP packets between MNs over an OMNI 469 underlying interface technology (such as omnidirectional wireless) 470 without support of fixed infrastructure. Multihop services entail 471 node-to-node relaying within a Mobile/Vehicular Ad-hoc Network 472 (MANET/VANET) for MN-to-MN communications and/or for "range 473 extension" where MNs within range of communications infrastructure 474 elements provide forwarding services for other MNs. 476 L2 477 The second layer in the OSI network model. Also known as "layer- 478 2", "link-layer", "sub-IP layer", "data link layer", etc. 480 L3 481 The third layer in the OSI network model. Also known as "layer- 482 3", "network-layer", "IP layer", etc. 484 underlying interface 485 a *NET interface over which an OMNI interface is configured. The 486 OMNI interface is seen as a L3 interface by the IP layer, and each 487 underlying interface is seen as a L2 interface by the OMNI 488 interface. The underlying interface either connects directly to 489 the physical communications media or coordinates with another node 490 where the physical media is hosted. 492 Mobility Service Identification (MSID) 493 Each MSE and AR is assigned a unique 32-bit Identification (MSID) 494 (see: Section 8). IDs are assigned according to MS-specific 495 guidelines (e.g., see: [I-D.templin-6man-aero]). 497 Safety-Based Multilink (SBM) 498 A means for ensuring fault tolerance through redundancy by 499 connecting multiple affiliated OMNI interfaces to independent 500 routing topologies (i.e., multiple independent OMNI links). 502 Performance Based Multilink (PBM) 503 A means for selecting underlying interface(s) for packet 504 transmission and reception within a single OMNI interface. 506 OMNI Domain 507 The set of all SBM/PBM OMNI links that collectively provides 508 services for a common set of MSPs. Each OMNI domain consists of a 509 set of affiliated OMNI links that all configure the same ::/48 ULA 510 prefix with a unique 16-bit Subnet ID as discussed in Section 9. 512 3. Requirements 514 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 515 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 516 "OPTIONAL" in this document are to be interpreted as described in BCP 517 14 [RFC2119][RFC8174] when, and only when, they appear in all 518 capitals, as shown here. 520 An implementation is not required to internally use the architectural 521 constructs described here so long as its external behavior is 522 consistent with that described in this document. 524 4. Overlay Multilink Network (OMNI) Interface Model 526 An OMNI interface is a virtual interface configured over one or more 527 underlying interfaces, which may be physical (e.g., an aeronautical 528 radio link, etc.) or virtual (e.g., an Internet or higher-layer 529 "tunnel"). The OMNI interface architectural layering model is the 530 same as in [RFC5558][RFC7847], and augmented as shown in Figure 1. 531 The IP layer therefore sees the OMNI interface as a single L3 532 interface nexus for multiple underlying interfaces that appear as L2 533 communication channels in the architecture. 535 +----------------------------+ 536 | Upper Layer Protocol | 537 Session-to-IP +---->| | 538 Address Binding | +----------------------------+ 539 +---->| IP (L3) | 540 IP Address +---->| | 541 Binding | +----------------------------+ 542 +---->| OMNI Interface | 543 Logical-to- +---->| (OMNI Adaptation Layer) | 544 Physical | +----------------------------+ 545 Interface +---->| L2 | L2 | | L2 | 546 Binding |(IF#1)|(IF#2)| ..... |(IF#n)| 547 +------+------+ +------+ 548 | L1 | L1 | | L1 | 549 | | | | | 550 +------+------+ +------+ 552 Figure 1: OMNI Interface Architectural Layering Model 554 Each underlying interface provides an L2/L1 abstraction according to 555 one of the following models: 557 o INET interfaces connect to an INET either natively or through one 558 or several IPv4 Network Address Translators (NATs). Native INET 559 interfaces have global IP addresses that are reachable from any 560 INET correspondent. NATed INET interfaces typically have private 561 IP addresses and connect to a private network behind one or more 562 NATs that provide INET access. 564 o ANET interfaces connect to a protected ANET that is separated from 565 the open INET by an AR acting as a proxy. The ANET interface may 566 be either on the same L2 link segment as the AR, or separated from 567 the AR by multiple IP hops. 569 o VPNed interfaces use security encapsulation over a *NET to a 570 Virtual Private Network (VPN) gateway. Other than the link-layer 571 encapsulation format, VPNed interfaces behave the same as for 572 Direct interfaces. 574 o Direct (aka "point-to-point") interfaces connect directly to a 575 peer without crossing any *NET paths. An example is a line-of- 576 sight link between a remote pilot and an unmanned aircraft. 578 The OMNI interface forwards original IP packets from the network 579 layer (L3) using the OMNI Adaptation Layer (OAL) (see: Section 5) as 580 an encapsulation and fragmentation sublayer service. This "OAL 581 source" then further encapsulates the resulting OAL packets/fragments 582 in *NET headers to create OAL carrier packets for transmission over 583 underlying interfaces (L2/L1). The target OMNI interface receives 584 the carrier packets from underlying interfaces (L1/L2) and discards 585 the *NET headers. If the resulting OAL packets/fragments are 586 addressed to itself, the OMNI interface acts as an "OAL destination" 587 and performs reassembly if necessary, discards the OAL encapsulation, 588 and delivers the original IP packet to the network layer (L3). If 589 the OAL fragments are addressed to another node, the OMNI interface 590 instead acts as an "OAL intermediate node" by re-encapsulating in new 591 *NET headers and forwarding the new carrier packets over an 592 underlying interface without reassembling or discarding the OAL 593 encapsulation. The OAL source and OAL destination are seen as 594 "neighbors" on the OMNI link, while OAL intermediate nodes are seen 595 as "bridges" capable of multinet concatenation. 597 The OMNI interface can send/receive original IP packets to/from 598 underlying interfaces while including/omitting various encapsulations 599 including OAL, UDP, IP and L2. The network layer can also access the 600 underlying interfaces directly while bypassing the OMNI interface 601 entirely when necessary. This architectural flexibility may be 602 beneficial for underlying interfaces (e.g., some aviation data links) 603 for which encapsulation overhead may be a primary consideration. 604 OMNI interfaces that send original IP packets directly over 605 underlying interfaces without invoking the OAL can only reach peers 606 located on the same OMNI link segment. However, an ANET proxy that 607 receives the original IP packet can forward it further by performing 608 OAL encapsulation with source set to its own address and destination 609 set to the OAL destination corresponding to the final destination 610 (i.e., even if the OAL destination is on a different OMNI link 611 segment). 613 Original IP packets sent directly over underlying interfaces are 614 subject to the same path MTU related issues as for any 615 Internetworking path, and do not include per-packet identifications 616 that can be used for data origin verification and/or link-layer 617 retransmissions. Original IP packets presented directly to an 618 underlying interface that exceed the underlying network path MTU are 619 dropped with an ordinary ICMPv6 Packet Too Big (PTB) message 620 returned. These PTB messages are subject to loss [RFC2923] the same 621 as for any non-OMNI IP interface. 623 The OMNI interface encapsulation/decapsulation layering possibilities 624 are shown in Figure 2 below. In the figure, imaginary vertical lines 625 drawn between the Network Layer and Underlying interfaces denote the 626 encapsulation/decapsulation layering combinations possible. Common 627 combinations include NULL (i.e., direct access to underlying 628 interfaces with or without using the OMNI interface), OMNI/IP, 629 OMNI/UDP/IP, OMNI/UDP/IP/L2, OMNI/OAL/UDP/IP, OMNI/OAL/UDP/L2, etc. 631 +------------------------------------------------------------+ 632 | Network Layer | 633 +--+---------------------------------------------------------+ 634 | OMNI Interface | 635 +--------------------------+------------------------------+ 636 | OAL Encaps/Decaps | 637 +------------------------------+ 638 | OAL Frag/Reass | 639 +------------+---------------+--------------+ 640 | UDP Encaps/Decaps/Compress | 641 +----+---+------------+--------+--+ +--------+ 642 | IP E/D | | IP E/D | | IP E/D | 643 +---+------+-+----+ +--+---+----+ +----+---+--+ 644 |L2 E/D| |L2 E/D| |L2 E/D| |L2 E/D| 645 +-------+------+---+------+----+------+---------------+------+ 646 | Underlying Interfaces | 647 +------------------------------------------------------------+ 649 Figure 2: OMNI Interface Layering 651 The OMNI/OAL model gives rise to a number of opportunities: 653 o MNs receive a MNP from the MS, and coordinate with the MS through 654 IPv6 ND message exchanges. The MN uses the MNP to construct a 655 unique Link-Local Address (MNP-LLA) through the algorithmic 656 derivation specified in Section 8 and assigns the LLA to the OMNI 657 interface. Since MNP-LLAs are uniquely derived from an MNP, no 658 Duplicate Address Detection (DAD) or Multicast Listener Discovery 659 (MLD) messaging is necessary. 661 o since Temporary ULAs are statistically unique, they can be used 662 without DAD, e.g. for MN-to-MN communications until an MNP-LLA is 663 obtained. 665 o underlying interfaces on the same L2 link segment as an AR do not 666 require any L3 addresses (i.e., not even link-local) in 667 environments where communications are coordinated entirely over 668 the OMNI interface. 670 o as underlying interface properties change (e.g., link quality, 671 cost, availability, etc.), any active interface can be used to 672 update the profiles of multiple additional interfaces in a single 673 message. This allows for timely adaptation and service continuity 674 under dynamically changing conditions. 676 o coordinating underlying interfaces in this way allows them to be 677 represented in a unified MS profile with provisions for mobility 678 and multilink operations. 680 o exposing a single virtual interface abstraction to the IPv6 layer 681 allows for multilink operation (including QoS based link 682 selection, packet replication, load balancing, etc.) at L2 while 683 still permitting L3 traffic shaping based on, e.g., DSCP, flow 684 label, etc. 686 o the OMNI interface allows inter-INET traversal when nodes located 687 in different INETs need to communicate with one another. This 688 mode of operation would not be possible via direct communications 689 over the underlying interfaces themselves. 691 o the OAL supports lossless and adaptive path MTU mitigations not 692 available for communications directly over the underlying 693 interfaces themselves. The OAL supports "packing" of multiple IP 694 payload packets within a single OAL packet. 696 o the OAL applies per-packet identification values that allow for 697 link-layer reliability and data origin authentication. 699 o L3 sees the OMNI interface as a point of connection to the OMNI 700 link; if there are multiple OMNI links (i.e., multiple MS's), L3 701 will see multiple OMNI interfaces. 703 o Multiple independent OMNI interfaces can be used for increased 704 fault tolerance through Safety-Based Multilink (SBM), with 705 Performance-Based Multilink (PBM) applied within each interface. 707 Other opportunities are discussed in [RFC7847]. Note that even when 708 the OMNI virtual interface is present, applications can still access 709 underlying interfaces either through the network protocol stack using 710 an Internet socket or directly using a raw socket. This allows for 711 intra-network (or point-to-point) communications without invoking the 712 OMNI interface and/or OAL. For example, when an IPv6 OMNI interface 713 is configured over an underlying IPv4 interface, applications can 714 still invoke IPv4 intra-network communications as long as the 715 communicating endpoints are not subject to mobility dynamics. 716 However, the opportunities discussed above are not realized when the 717 architectural layering is bypassed in this way. 719 Figure 3 depicts the architectural model for a MN with an attached 720 EUN connecting to the MS via multiple independent *NETs. When an 721 underlying interface becomes active, the MN's OMNI interface sends 722 IPv6 ND messages without encapsulation if the first-hop Access Router 723 (AR) is on the same underlying link; otherwise, the interface uses 724 IP-in-IP encapsulation. The IPv6 ND messages traverse the ground 725 domain *NETs until they reach an AR (AR#1, AR#2, ..., AR#n), which 726 then coordinates with an INET Mobility Service Endpoint (MSE#1, 727 MSE#2, ..., MSE#m) and returns an IPv6 ND message response to the MN. 728 The Hop Limit in IPv6 ND messages is not decremented due to 729 encapsulation; hence, the OMNI interface appears to be attached to an 730 ordinary link. 732 +--------------+ (:::)-. 733 | MN |<-->.-(::EUN:::) 734 +--------------+ `-(::::)-' 735 |OMNI interface| 736 +----+----+----+ 737 +--------|IF#1|IF#2|IF#n|------ + 738 / +----+----+----+ \ 739 / | \ 740 / | \ 741 v v v 742 (:::)-. (:::)-. (:::)-. 743 .-(::*NET:::) .-(::*NET:::) .-(::*NET:::) 744 `-(::::)-' `-(::::)-' `-(::::)-' 745 +----+ +----+ +----+ 746 ... |AR#1| .......... |AR#2| ......... |AR#n| ... 747 . +-|--+ +-|--+ +-|--+ . 748 . | | | 749 . v v v . 750 . <----- INET Encapsulation -----> . 751 . . 752 . +-----+ (:::)-. . 753 . |MSE#2| .-(::::::::) +-----+ . 754 . +-----+ .-(::: INET :::)-. |MSE#m| . 755 . (::::: Routing ::::) +-----+ . 756 . `-(::: System :::)-' . 757 . +-----+ `-(:::::::-' . 758 . |MSE#1| +-----+ +-----+ . 759 . +-----+ |MSE#3| |MSE#4| . 760 . +-----+ +-----+ . 761 . . 762 . . 763 . <----- Worldwide Connected Internetwork ----> . 764 ........................................................... 766 Figure 3: MN/MS Coordination via Multiple *NETs 768 After the initial IPv6 ND message exchange, the MN (and/or any nodes 769 on its attached EUNs) can send and receive original IP packets over 770 the OMNI interface. OMNI interface multilink services will forward 771 the packets via ARs in the correct underlying *NETs. The AR 772 encapsulates the packets according to the capabilities provided by 773 the MS and forwards them to the next hop within the worldwide 774 connected Internetwork via optimal routes. 776 5. OMNI Interface Maximum Transmission Unit (MTU) 778 The OMNI interface observes the link nature of tunnels, including the 779 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 780 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 781 The OMNI interface is configured over one or more underlying 782 interfaces as discussed in Section 4, where the interfaces (and their 783 associated *NET paths) may have diverse MTUs. OMNI interface 784 considerations for accommodating original IP packets of various sizes 785 are discussed in the following sections. 787 IPv6 underlying interfaces are REQUIRED to configure a minimum MTU of 788 1280 bytes and a minimum MRU of 1500 bytes [RFC8200]. Therefore, the 789 minimum IPv6 path MTU is 1280 bytes since routers on the path are not 790 permitted to perform network fragmentation even though the 791 destination is required to reassemble more. The network therefore 792 MUST forward original IP packets of at least 1280 bytes without 793 generating an IPv6 Path MTU Discovery (PMTUD) Packet Too Big (PTB) 794 message [RFC8201]. (While the source can apply "source 795 fragmentation" for locally-generated IPv6 packets up to 1500 bytes 796 and larger still if it knows the destination configures a larger MRU, 797 this does not affect the minimum IPv6 path MTU.) 799 IPv4 underlying interfaces are REQUIRED to configure a minimum MTU of 800 68 bytes [RFC0791] and a minimum MRU of 576 bytes [RFC0791][RFC1122]. 801 Therefore, when the Don't Fragment (DF) bit in the IPv4 header is set 802 to 0 the minimum IPv4 path MTU is 576 bytes since routers on the path 803 support network fragmentation and the destination is required to 804 reassemble at least that much. The OMNI interface therefore MUST set 805 DF to 0 in the IPv4 encapsulation headers of carrier packets that are 806 no larger than 576 bytes, and SHOULD set DF to 1 in larger carrier 807 packets. (Note: even if the encapsulation source has a way to 808 determine that the encapsulation destination configures an MRU larger 809 than 576 bytes, it should not assume a larger minimum IPv4 path MTU 810 without careful consideration of the issues discussed in 811 Section 6.9.) 813 The OMNI interface configures an MTU and MRU of 9180 bytes [RFC2492]; 814 the size is therefore not a reflection of the underlying interface or 815 *NET path MTUs, but rather determines the largest original IP packet 816 the OAL (and/or underlying interface) can forward or reassemble. For 817 each OAL destination (i.e., for each OMNI link neighbor), the OAL 818 source may discover "hard" or "soft" Reassembly Limit values smaller 819 than the MRU based on receipt of IPv6 ND messages with OMNI 820 Reassembly Limit sub-options (see: Section 12.1.11). The OMNI 821 interface employs the OAL as an encapsulation sublayer service to 822 transform original IP packets into OAL packets/fragments, and the OAL 823 in turn uses *NET encapsulation to forward carrier packets over the 824 underlying interfaces (see: Section 6). 826 6. The OMNI Adaptation Layer (OAL) 828 When an OMNI interface forwards an original IP packet from the 829 network layer for transmission over one or more underlying 830 interfaces, the OMNI Adaptation Layer (OAL) acting as the OAL source 831 drops the packet and returns a PTB message if the packet exceeds the 832 MRU and/or the hard Reassembly Limit for the intended OAL 833 destination. Otherwise, the OAL source applies encapsulation to form 834 OAL packets and fragmentation to produce resulting OAL fragments 835 suitable for *NET encapsulation and transmission as carrier packets 836 over underlying interfaces as described in Section 6.1. 838 These carrier packets travel over one or more underlying networks 839 bridged by OAL intermediate nodes, which re-encapsulate by removing 840 the *NET headers of the first underlying network and appending *NET 841 headers appropriate for the next underlying network in succession. 842 (This process supports the multinet concatenation capability needed 843 for joining multiple diverse networks.) After re-encapsulation by 844 zero or more OAL intermediate nodes, the carrier packets arrive at 845 the OAL destination. 847 When the OAL destination receives the carrier packets, it discards 848 the *NET headers and reassembles the resulting OAL fragments into an 849 OAL packet as described in Section 6.3. The OAL destination then 850 decapsulates the OAL packet to obtain the original IP packet, which 851 it then delivers to the network layer. 853 The OAL presents an OMNI sublayer abstraction similar to ATM 854 Adaptation Layer 5 (AAL5). Unlike AAL5 which performs segmentation 855 and reassembly with fixed-length 53 octet cells over ATM networks, 856 however, the OAL uses IPv6 encapsulation, fragmentation and 857 reassembly with larger variable-length cells over heterogeneous 858 underlying networks. Detailed operations of the OAL are specified in 859 the following sections. 861 6.1. OAL Source Encapsulation and Fragmentation 863 When the network layer forwards an original IP packet into the OMNI 864 interface, the OAL source inserts an IPv6 encapsulation header but 865 does not decrement the Hop Limit/TTL of the original IP packet since 866 encapsulation occurs at a layer below IP forwarding [RFC2473]. The 867 OAL source copies the "Type of Service/Traffic Class" [RFC2983] and 868 "Congestion Experienced" [RFC3168] values in the original packet's IP 869 header into the corresponding fields in the OAL header, then sets the 870 OAL header "Flow Label" as specified in [RFC6438]. The OAL source 871 finally sets the OAL header IPv6 Hop Limit to a conservative value 872 sufficient to enable loop-free forwarding over multiple concatenated 873 OMNI link segments and sets the Payload Length to the length of the 874 original IP packet. 876 The OAL next selects source and destination addresses for the IPv6 877 header of the resulting OAL packet. MN OMNI interfaces set the OAL 878 IPv6 header source address to a Unique Local Address (ULA) based on 879 the Mobile Network Prefix (MNP-ULA), while AR and MSE OMNI interfaces 880 set the source address to an Administrative ULA (ADM-ULA) (see: 881 Section 9). When a MN OMNI interface does not (yet) have an MNP-ULA, 882 it can use a Temporary ULA and/or Host Identity Tag (HIT) instead 883 (see: Section 22). 885 When the OAL source forwards an original IP packet toward a final 886 destination via an ANET underlying interface, it sets the OAL IPv6 887 header source address to its own ULA and sets the destination to 888 either the Administrative ULA (ADM-ULA) of the ANET peer or the 889 Mobile Network Prefix ULA (MNP-ULA) corresponding to the final 890 destination (see below). The OAL source then fragments the OAL 891 packet if necessary, encapsulates the OAL fragments in any ANET 892 headers and sends the resulting carrier packets to the ANET peer 893 which either reassembles before forwarding if the OAL destination is 894 its own ULA or forwards the fragments toward the true OAL destination 895 without first reassembling otherwise. 897 When the OAL source forwards an original IP packet toward a final 898 destination via an INET underlying interface, it sets the OAL IPv6 899 header source address to its own ULA and sets the destination to the 900 ULA of an OAL destination node on the final *NET segment. The OAL 901 source then fragments the OAL packet if necessary, encapsulates the 902 OAL fragments in any *NET headers and sends the resulting carrier 903 packets toward the OAL destination on the final segment OMNI node 904 which reassembles before forwarding the original IP packets toward 905 the final destination. 907 Following OAL IPv6 encapsulation and address selection, the OAL 908 source next appends a 2 octet trailing Checksum (initialized to 0) at 909 the end of the original IP packet while incrementing the OAL header 910 IPv6 Payload Length field to reflect the addition of the trailer. 911 The format of the resulting OAL packet following encapsulation is 912 shown in Figure 4: 914 +----------+-----+-----+-----+-----+-----+-----+----+ 915 | OAL Hdr | Original IP packet |Csum| 916 +----------+-----+-----+-----+-----+-----+-----+----+ 918 Figure 4: OAL Packet Before Fragmentation 920 The OAL source next selects a 32-bit Identification value for the 921 packet, beginning with an unpredictable value for the initial OAL 922 packet per [RFC7739] and monotonically incrementing for each 923 successive OAL packet until a new initial value is chosen. 925 The OAL source then calculates the checksum per the 8-bit Fletcher 926 algorithm specified in Appendix A over the entire OAL packet 927 beginning with a pseudo-header of the IPv6 header similar to that 928 found in Section 8.1 of [RFC8200] and extending to the end of the 929 (0-initialized) checksum trailer. The OAL IPv6 pseudo-header is 930 formed as shown in Figure 5: 932 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 933 | | 934 + + 935 | | 936 + OAL Source Address + 937 | | 938 + + 939 | | 940 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 941 | | 942 + + 943 | | 944 + OAL Destination Address + 945 | | 946 + + 947 | | 948 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 949 | OAL Payload Length | zero | Next Header | 950 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 951 | Identification | 952 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 954 Figure 5: OAL IPv6 Pseudo-Header 956 After calculating the checksum, the OAL source writes the results 957 over the (0-initialized) trailing checksum octets. The OAL source 958 then inserts a single OMNI Routing Header (ORH) if necessary (see: 959 [I-D.templin-6man-aero]) while incrementing Payload Length to reflect 960 the addition of the ORH (note that the late addition of the ORH is 961 not covered by the checksum). 963 The OAL source next fragments the OAL packet if necessary while 964 assuming the IPv4 minimum path MTU (i.e., 576 bytes) as the worst 965 case for OAL fragmentation regardless of the underlying interface IP 966 protocol version since IPv6/IPv4 protocol translation and/or IPv6-in- 967 IPv4 encapsulation may occur in any *NET path. By always assuming 968 the IPv4 minimum even for IPv6 underlying interfaces, the OAL source 969 may produce smaller fragments with additional encapsulation overhead 970 but will always interoperate and never run the risk of loss due to an 971 MTU restriction or due to presenting an underlying interface with a 972 carrier packet that exceeds its MRU. Additionally, the OAL path 973 could traverse multiple *NET "segments" with intermediate OAL 974 forwarding nodes performing re-encapsulation where the *NET 975 encapsulation of the previous segment is replaced by the *NET 976 encapsulation of the next segment which may be based on a different 977 IP protocol version and/or encapsulation sizes. 979 The OAL source therefore assumes a default minimum path MTU of 576 980 bytes at each *NET segment for the purpose of generating OAL 981 fragments for *NET encapsulation and transmission as carrier packets. 982 In the worst case, each successive *NET segment may re-encapsulate 983 with either a 20 byte IPv4 or 40 byte IPv6 header, an 8 byte UDP 984 header and in some cases an IP security encapsulation (40 bytes 985 maximum assumed). Any *NET segment may also insert a maximum-length 986 (40 byte) ORH as an extension to the existing 40 byte OAL IPv6 header 987 plus 8 byte Fragment Header if an ORH was not already present. 988 Assuming therefore an absolute worst case of (40 + 40 + 8) = 88 bytes 989 for *NET encapsulation plus (40 + 40 + 8) = 88 bytes for OAL 990 encapsulation leaves (576 - 88 - 88) = 400 bytes to accommodate a 991 portion of the original IP packet/fragment. The OAL source therefore 992 sets a minimum Maximum Payload Size (MPS) of 400 bytes as the basis 993 for the minimum-sized OAL fragment that can be assured of traversing 994 all segments without loss due to an MTU/MRU restriction. The Maximum 995 Fragment Size (MFS) for OAL fragmentation is therefore determined by 996 the MPS plus the size of the OAL encapsulation headers. (Note that 997 the OAL source includes the 2 octet trailer as part of the payload 998 during fragmentation, and the OAL destination regards it as ordinary 999 payload until reassembly and checksum verification are complete.) 1001 The OAL source SHOULD maintain "path MPS" values for individual OAL 1002 destinations initialized to the minimum MPS and increased to larger 1003 values (up to the OMNI interface MTU) if better information is known 1004 or discovered. For example, when *NET peers share a common 1005 underlying link or a fixed path with a known larger MTU, the OAL 1006 source can base path MPS on this larger size (i.e., instead of 576 1007 bytes) as long as the *NET peer reassembles before re-encapsulating 1008 and forwarding (while re-fragmenting if necessary). Also, if the OAL 1009 source has a way of knowing the maximum *NET encapsulation size for 1010 all segments along the path it may be able to increase path MPS to 1011 reserve additional room for payload data. The OAL source must 1012 include the uncompressed OAL header size in its path MPS calculation, 1013 since a full header could be included at any time. 1015 The OAL source can also actively probe individual OAL destinations to 1016 discover larger path MPS values using packetization layer probes per 1017 [RFC4821][RFC8899], but care must be taken to avoid setting static 1018 values for dynamically changing paths leading to black holes. The 1019 probe involves sending an OAL packet larger than the current path MPS 1020 and receiving a small acknowledgement message in response (with the 1021 possible receipt of link-layer error message in case the probe was 1022 lost). For this purpose, the OAL source can send an NS message with 1023 one or more OMNI options with large PadN sub-options (see: 1024 Section 12) in order to receive a small NA response from the OAL 1025 destination. While observing the minimum MPS will always result in 1026 robust and secure behavior, the OAL source should optimize path MPS 1027 values when more efficient utilization may result in better 1028 performance (e.g. for wireless aviation data links). 1030 When the OAL source performs fragmentation, it SHOULD produce the 1031 minimum number of non-overlapping fragments under current MPS 1032 constraints, where each non-final fragment MUST be at least as large 1033 as the minimum MPS, while the final fragment MAY be smaller. The OAL 1034 source also converts all original IP packets no larger than the 1035 current MPS into "atomic fragments" by including a Fragment Header 1036 with Fragment Offset and More Fragments both set to 0. The OAL 1037 source finally encapsulates the fragments in *NET headers to form 1038 carrier packets and forwards them over an underlying interface, while 1039 retaining the fragments and their ordinal positions (i.e., as Frag 1040 #0, Frag #1, Frag #2, etc.) for a timeout period in case link-layer 1041 retransmission is requested. The formats of OAL fragments and 1042 carrier packets are shown in Figure 6. 1044 +----------+--+-------------+ 1045 | OAL Hdr |FH| Frag #0 | 1046 +----------+--+-------------+ 1047 +----------+--+-------------+ 1048 | OAL Hdr |FH| Frag #1 | 1049 +----------+--+-------------+ 1050 +----------+--+-------------+ 1051 | OAL Hdr |FH| Frag #2 | 1052 +----------+--+-------------+ 1053 .... 1054 +----------+--+-------------+----+ 1055 | OAL Hdr |FH| Frag #(N-1) |Csum| 1056 +----------+--+-------------+----+ 1057 a) OAL fragments after fragmentation 1058 (FH = Fragment Header; Csum appears only in final fragment) 1060 +--------+--+-----+-----+-----+-----+-----+----+ 1061 |OAL Hdr |FH| Original IP packet |Csum| 1062 +--------+--+-----+-----+-----+-----+-----+----+ 1063 b) An OAL atomic fragment with FH but no fragmentation. 1065 +--------+----------+--+-------------+ 1066 |*NET Hdr| OAL Hdr |FH| Frag #i | 1067 +--------+----------+--+-------------+ 1068 c) OAL carrier packet after *NET encapsulation 1070 Figure 6: OAL Fragments and Carrier Packets 1072 6.2. OAL *NET Encapsulation and Re-Encapsulation 1074 During *NET encapsulation, OAL sources first encapsulate each OAL 1075 fragment in a UDP header as the first *NET encapsulation sublayer if 1076 NAT traversal, packet filtering middlebox traversal and/or OAL header 1077 compression are necessary. The OAL then optionally appends 1078 additional encapsulation sublayer headers, then presents the *NET 1079 packet to an underlying interface. This layering can be seen in 1080 Figure 2. 1082 When a UDP header is included, the OAL source next sets the UDP 1083 source port to a constant value that it will use in each successive 1084 carrier packet it sends to the next OAL hop. For packets sent to an 1085 MSE, the OAL source sets the UDP destination port to 8060, i.e., the 1086 IANA-registered port number for AERO. For packets sent to a MN peer, 1087 the source sets the UDP destination port to the cached port value for 1088 this peer. The OAL source then sets the UDP length to the total 1089 length of the OAL fragment in correspondence with the OAL header 1090 Payload Length (i.e., the UDP length and IPv6 Payload Length must 1091 agree). The OAL source finally sets the UDP checksum to 0 1092 [RFC6935][RFC6936] since the only fields not already covered by the 1093 OAL checksum or underlying *NET CRCs are the Fragment Header fields, 1094 and any corruption in those fields will be garbage collected by the 1095 reassembly algorithm (however, see Section 20 for additional 1096 considerations). The UDP encapsulation header is often used in 1097 association with IP encapsulation, but may also be used between 1098 neighbors on a shared physical link with a true L2 header format such 1099 as for transmission over IEEE 802 Ethernet links. This document 1100 therefore requests a new Ether Type code assignment TBD1 in the IANA 1101 'ieee-802-numbers' registry for direct User Datagram Protocol (UDP) 1102 encapsulation over IEEE 802 Ethernet links (see: Section 25). 1104 For *NET encapsulations over IP, the OAL source next copies the "Type 1105 of Service/Traffic Class" [RFC2983] and "Congestion Experienced" 1106 [RFC3168] values in the OAL IPv6 header into the corresponding fields 1107 in the *NET IP header, then (for IPv6) sets the *NET IPv6 header 1108 "Flow Label" as specified in [RFC6438]. The OAL source then sets the 1109 *NET IP TTL/Hop Limit the same as for any *NET host, i.e., it does 1110 not copy the Hop Limit value from the OAL header. For carrier 1111 packets undergoing re-encapsulation at an OAL intermediate node, the 1112 node decrements the OAL IPv6 header Hop Limit and discards the 1113 carrier packet if the value reaches 0. The node then copies the 1114 "Type of Service/Traffic Class" and "Congestion Experienced" values 1115 from the previous hop *NET encapsulation header into the OAL IPv6 1116 header before setting the next hop *NET IP encapsulation header 1117 values the same as specified for the OAL source above. 1119 Following *NET encapsulation/re-encapsulation, the OAL source sends 1120 the resulting carrier packets over one or more underlying interfaces. 1121 The underlying interfaces often connect directly to physical media on 1122 the local platform (e.g., a laptop computer with WiFi, etc.), but in 1123 some configurations the physical media may be hosted on a separate 1124 Local Area Network (LAN) node. In that case, the OMNI interface can 1125 establish a Layer-2 VLAN or a point-to-point tunnel (at a layer below 1126 the underlying interface) to the node hosting the physical media. 1127 The OMNI interface may also apply encapsulation at the underlying 1128 interface layer (e.g., as for a tunnel virtual interface) such that 1129 carrier packets would appear "double-encapsulated" on the LAN; the 1130 node hosting the physical media in turn removes the LAN encapsulation 1131 prior to transmission or inserts it following reception. Finally, 1132 the underlying interface must monitor the node hosting the physical 1133 media (e.g., through periodic keepalives) so that it can convey 1134 up/down/status information to the OMNI interface. 1136 6.3. OAL Destination Decapsulation and Reassembly 1138 When an OMNI interface receives a carrier packet from an underlying 1139 interface, the OAL destination discards the *NET encapsulation 1140 headers and examines the OAL header of the enclosed OAL fragment. If 1141 the OAL fragment is addressed to a different node, the OAL 1142 destination re-encapsulates and forwards as discussed below. If the 1143 OAL fragment is addressed to itself, the OAL destination creates or 1144 updates a checklist for this (Source, Destination, Identification)- 1145 tuple to track the fragments already received (i.e., by examining the 1146 Payload Length, Fragment Offset, More Fragments and Identification 1147 values supplied by the OAL source). The OAL destination verifies 1148 that all non-final OAL fragments are no smaller than the minimum MPS 1149 and that no fragments overlap or leave "holes" smaller than the 1150 minimum MPS, while dropping any non-conforming fragments. The OAL 1151 destination records each conforming OAL fragment's ordinal position 1152 based on the OAL header Payload Length and Fragment Offset values 1153 (i.e., as Frag #0, Frag #1, Frag #2, etc.) and admits each fragment 1154 into the reassembly cache. 1156 When reassembly is complete, the OAL destination removes the ORH if 1157 present while decrementing Payload Length to reflect the removal of 1158 the ORH. The OAL destination next verifies the resulting OAL 1159 packet's checksum and discards the packet if the checksum is 1160 incorrect. If the OAL packet was accepted, the OAL destination then 1161 removes the OAL header/trailer, then delivers the original IP packet 1162 to the network layer. Note that link layers include a CRC-32 1163 integrity check which provides effective hop-by-hop error detection 1164 in the underlying network for payload sizes up to the OMNI interface 1165 MTU [CRC], but that some hops may traverse intermediate layers such 1166 as tunnels over IPv4 that do not include integrity checks. The 1167 trailing Fletcher checksum therefore allows the OAL destination to 1168 detect OAL packet splicing errors due to reassembly misassociations 1169 and/or to verify integrity for OAL packets whose fragments may have 1170 traversed unprotected underlying network hops [CKSUM]. The Fletcher 1171 checksum algorithm also provides diversity with respect to both lower 1172 layer CRCs and upper layer Internet checksums as part of a 1173 complimentary multi-layer integrity assurance architecture. 1175 6.4. OAL Header Compression 1177 When the OAL source and destination are on the same *NET segment, no 1178 ORH is needed and carrier packet header compression is possible. 1179 When the OAL source and destination exchange initial IPv6 ND messages 1180 as discussed in the following Sections, each caches the observed *NET 1181 UDP source port and source IP (or L2) address associated with the OAL 1182 IPv6 source address found in the full-length OAL IPv6 header. After 1183 the initial IPv6 ND message exchange, the OAL source can begin 1184 applying OAL Header Compression to significantly reduce the 1185 encapsulation overhead required in each carrier packet. 1187 When the OAL source determines that header compression state has been 1188 established (i.e., following the IPv6 ND message exchange), it can 1189 begin sending OAL fragments with significant portions of the IPv6 1190 header and Fragment Header omitted thereby reducing the amount of 1191 encapsulation overhead. For OAL first-fragments (including atomic 1192 fragments), the OMNI Compressed Header - Type 0 (OCH-0) is used and 1193 formatted as shown in Figure 7: 1195 0 1 2 3 1196 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 1197 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ * 1198 | Source port | Destination port | U 1199 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ D 1200 | Length | Checksum | P 1201 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ * 1202 |Version| Traffic Class | Flow Label | 1203 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1204 | Next Header | Reserved |M| Identification (0 -1) | 1205 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1206 | Identification (2-3) | 1207 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+/ 1209 Figure 7: OMNI Compressed Header - Type 0 (OCH-0) 1211 In this format, the UDP header appears in its entirety in the first 8 1212 octets, then followed by the first 4 octets of the IPv6 header with 1213 the remainder omitted. (The IPv6 Version field is set to the value 0 1214 to distinguish this header from a true IP protocol version number and 1215 from OCH-1 - see below.) The compressed IPv6 header is then followed 1216 by a compressed IPv6 Fragment Header with the Fragment Offset field 1217 and two Reserved bits omitted (since these fields always encode the 1218 value 0 in first-fragments), and with the More Fragments (M) bit 1219 relocated to the least significant bit of the first Reserved field. 1220 The OCH-0 header is then followed by the OAL fragment body, and the 1221 UDP length field is reduced by 38 octets (i.e., the difference in 1222 length between full-length IPv6 and Fragment Headers and the length 1223 of the compressed headers). 1225 For OAL non-first fragments (i.e., those with non-zero Fragment 1226 Offsets), the OMNI Compressed Header - Type 1 (OCH-1) is used and 1227 formatted as shown in Figure 8: 1229 0 1 2 3 1230 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 1231 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ * 1232 | Source port | Destination port | U 1233 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ D 1234 | Length | Checksum | P 1235 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ * 1236 |V|R|M| Fragment Offset | Identification (0-1) | 1237 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1238 | Identification (1-3) | 1239 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1241 Figure 8: OMNI Compressed Header - Type 1 (OCH-1) 1243 In this format, the UDP header appears in its entirety in the first 8 1244 octets, but all IPv6 header fields except for the most significant 1245 Version (V) bit are omitted. (The V bit is set to the value 1 to 1246 distinguish this header from a true IP protocol version number and 1247 from OCH-0.) The V bit is followed by a single Reserved (R) bit and 1248 the More Fragments (M) bit in a compressed IPv6 Fragment Header with 1249 the Next Header and first Reserved fields omitted. The OCH-1 header 1250 is then followed by the OAL fragment body, and the UDP length field 1251 is reduced by 42 octets (i.e., the difference in length between full- 1252 length IPv6 and Fragment Headers and the length of the compressed 1253 headers). 1255 When the OAL destination receives a carrier packet with an OCH, it 1256 first determines the OAL IPv6 source and destination addresses by 1257 examining the UDP source port and L2 source address, then determines 1258 the length by examining the UDP length. The OAL destination then 1259 examines the (V)ersion field immediately following the UDP header. 1260 If the (4-bit) Version field encodes the value 0, the OAL destination 1261 processes the remainder of the header as an OCH-0, then reconstitutes 1262 the full-sized IPv6 and Fragment Headers and adds this OAL fragment 1263 to the reassembly buffer if necessary. If the (1-bit) V bit encodes 1264 the value 1, the OAL destination instead processes the remainder of 1265 the header as an OCH-1, then reconstitutes the full-sized IPv6 and 1266 Fragment Headers and adds this OAL fragment to the reassembly buffer. 1267 Note that, since the OCH-1 does not include Traffic Class, Flow Label 1268 or Next Header information, the OAL destination writes the value 0 1269 into these fields when it reconstitutes the full headers. These 1270 values will be correctly populated during reassembly after an OAL 1271 first fragment with an OCH-0 or uncompressed OAL header arrives. 1273 Note: OAL header compression must not interfere with checksum 1274 calculation and verification, which are still applied according to 1275 the OAL pseudo-header per Section 6.1 even though compression is 1276 applied. 1278 6.5. OAL Fragment Identification Window Maintenance 1280 For each active neighbor, OAL nodes maintain a "send" and "accept" 1281 window that determines the range of OAL fragment Identification 1282 values to send to or accept from this neighbor. Both neighbors 1283 maintain an accept window of the sequence of OAL packets they will 1284 currently accept using the same mechanisms specified in the 1285 Transmission Control Protocol (TCP) [RFC0793]. The TCP is therefore 1286 used to manage a stream of OAL packets instead of a stream of bytes 1287 within the OAL context. Both a current and previous accept window is 1288 maintained to support dynamic window start value changes. New window 1289 start values are established through the exchange of authentic IPv6 1290 ND messages. 1292 As noted above, the OAL source establishes a send window of 32-bit 1293 Identifications beginning with an unpredictable value for the initial 1294 message [RFC7739] and monotonically incrementing for each successive 1295 OAL packet until a new initial value is chosen. The OAL source 1296 asserts the new send window starting value by including it as the TCP 1297 Sequence Number in an authentic IPv6 ND NS/RS message with the SYN 1298 flag set. When the OAL destination receives the IPv6 ND message, it 1299 saves the previous accept window starting value and sets the current 1300 accept window starting value to the new Sequence Number value 1301 received for this OAL source. The OAL destination then expects 1302 future OAL packets received from this OAL source to include 1303 Identification values that are within N of either the current or 1304 previous accept window (modulo 2^32) until the neighbor reachable 1305 time expires or the OAL source sends a new IPv6 ND message. 1307 For example, if the OAL destination receives an authentic NS/RS 1308 message with the SYN flag set and with Sequence Number 0x12345678 and 1309 the current accept window begins at 0xfe1284cd, it resets the current 1310 accept window for this OAL source to begin with 0x12345678 and sets 1311 the previous accept window to 0xfe1284cd. The OAL destination then 1312 examines the Identification values in subsequent carrier packets 1313 received from this OAL source. If the Identification values of 1314 subsequent carrier packets fall within the current accept window 1315 (0x12345678 + N) or previous accept window (0xfe1284cd + N) the OAL 1316 destination accepts the packet; otherwise, it silently drops the 1317 packet. 1319 While monitoring the current accept window, the OAL destination must 1320 accept new authentic NS/RS Identification values even if outside the 1321 current window. The TCP Sequence Number resets the OAL destination 1322 current and previous accept windows, allowing a period of overlap in 1323 case OAL packets with Identification values from the previous window 1324 are still in flight. When an acknowledgement is required according 1325 to the TCP protocol, the node that sends the acknowledgement issues 1326 an unsolicited NA message with the ACK flag set and with the 1327 Acknowledgement Number being acknowledged. The node may also request 1328 selective retransmissions by including a Fragmentation Report sub- 1329 option in the OMNI option (see: Section 12.1.12). 1331 This implies that an IPv6 ND message used to initiate a connection 1332 should fit within a single OAL fragment (i.e., within current MPS 1333 constraints), since a fragmented IPv6 ND message with an out-of- 1334 window Identification value could be part of a DoS attack and should 1335 not be reassembled. While larger IPv6 ND messages (up to the OMNI 1336 interface MTU) can certainly be subject to OAL fragmentation, their 1337 Identification should be within the accept windows maintained by the 1338 OAL destination. 1340 The Identification send/accept window duration is therefore 1341 determined by the time between successive IPv6 ND messages, which is 1342 bounded by the neighbor reachability time specified in [RFC4861]. 1343 The window size implies that that the OAL source must send a new NS/ 1344 RS message before more than N OAL packets have been sent within the 1345 current send window, i.e., even if prior to reachability time 1346 expiration. 1348 The above specifications represent a connectionless (i.e., 1349 asymmetric) case when two independent NS/NA or RS/RA exchanges are 1350 needed to establish the send/accept windows of both parties, but a 1351 connection-oriented (i.e., symmetric) case is also supported for NA/ 1352 RA messages. The node that sends the NA/RA message copies the 1353 Identification value received in the NS/RS OMNI Sequence Number into 1354 NA/RA OMNI Acknowledgement Number, writes its own initial 1355 Identification value into the NA/RA OMNI Sequence Number and sets 1356 both the SYN and ACK flags the same as specified for TCP [RFC0793]. 1358 When a node receives an NA/RA message with an OMNI option that 1359 includes SYN/ACK flags, it establishes the send window for the 1360 neighbor based on the Identification value included in the NA/RA OMNI 1361 Acknowledgement Number and establishes the accept window for the 1362 neighbor based on the NA/RA OMNI Sequence Number. The node then 1363 returns a solicited NA message with the ACK flags and without setting 1364 the SYN flag while copying the Sequence Number from the original NA/ 1365 RA into the Acknowledgement Number. When the neighbor receives the 1366 solicited NA, it marks the connection as complete. If the neighbor 1367 does not receive the solicited NA, it may either retransmit the 1368 original NA/RA or wait for a fresh NS/RS to initiate a new exchange. 1369 This symmetric exchange of window start Identification values 1370 parallels the TCP three-way handshake. 1372 Unsolicited NA messages may be sent as acknowledgements at any time 1373 with the ACK flag set and with a valid Identification value in the 1374 Acknowledgment field. Unsolicited NA messages can also include a SYN 1375 flag when a reliable delivery indication is required instead of the 1376 default best-effort behavior; the receiver of the Unsolicited NA 1377 message returns a Solicited NA message with the ACK flag set and the 1378 SYN flag not set. 1380 In order to avoid endless NA/NA looping, a node that receives an NA/ 1381 RA message with the SYN flag set MUST NOT set the SYN flag in its 1382 solicited NA response. 1384 6.6. OAL Fragment Retransmission 1386 When the OAL source sends carrier packets with OAL fragments to an 1387 OAL destination, it should cache them for the window duration in case 1388 retransmission may be necessary. The OAL destination in turn 1389 maintains a checklist for the (Source, Destination, Identification)- 1390 tuple of each new OAL fragment received and notes the ordinal 1391 positions of fragments already received (i.e., as Frag #0, Frag #1, 1392 Frag #2, etc.). 1394 If the OAL destination notices some OAL fragments missing after most 1395 other fragments within the same Identification window have already 1396 arrived, it may send an IPv6 ND unsolicited Neighbor Advertisement 1397 (uNA) message to the OAL source that originated the fragments to 1398 report loss. The OAL destination creates a uNA message with an OMNI 1399 option containing an authentication sub-option to provide 1400 authentication (if the OAL source is on an open Internetwork) 1401 followed by a Fragmentation Report sub-option that includes a list of 1402 (Identification, Bitmap)-tuples for OAL fragments received and 1403 missing from this OAL source (see: Section 12). The OAL destination 1404 signs the message if an authentication sub-option is included, 1405 performs OAL encapsulation (with the its own address as the OAL 1406 source and the source address of the message that prompted the uNA as 1407 the OAL destination) and sends the message to the OAL source. 1409 When the OAL source receives the uNA message, it authenticates the 1410 message using the authentication sub-option (if present) then 1411 examines the Fragmentation Report. For each (Source, Destination, 1412 Identification)-tuple, the OAL source determines whether it still 1413 holds the original OAL fragments in its cache and retransmits any for 1414 which the Bitmap indicates a loss event. For example, if the Bitmap 1415 indicates that the ordinal OAL fragments Frag #3, Frag #7, Frag #10 1416 and Frag #13 from the same OAL packet are missing the OAL source 1417 retransmits these fragments only and no others. 1419 Note that the goal of this service is to provide a light-weight link- 1420 layer Automatic Repeat Request (ARQ) capability in the spirit of 1421 Section 8.1 of [RFC3819]. Rather than provide true end-to-end 1422 reliability, however, the service provides timely link-layer 1423 retransmissions that may improve packet delivery ratios and avoid 1424 some delays inherent in true end-to-end services. 1426 6.7. OAL MTU Feedback Messaging 1428 When the OMNI interface forwards original IP packets from the network 1429 layer, it invokes the OAL and returns internally-generated ICMPv4 1430 Fragmentation Needed [RFC1191] or ICMPv6 Path MTU Discovery (PMTUD) 1431 Packet Too Big (PTB) [RFC8201] messages as necessary. This document 1432 refers to both of these ICMPv4/ICMPv6 message types simply as "PTBs", 1433 and introduces a distinction between PTB "hard" and "soft" errors as 1434 discussed below. 1436 Ordinary PTB messages with ICMPv4 header "unused" field or ICMPv6 1437 header Code field value 0 are hard errors that always indicate that a 1438 packet has been dropped due to a real MTU restriction. In 1439 particular, the OAL source drops the packet and returns a PTB hard 1440 error if the packet exceeds the OAL destination MRU. However, the 1441 OMNI interface can also forward large original IP packets via OAL 1442 encapsulation and fragmentation while at the same time returning PTB 1443 soft error messages (subject to rate limiting) if it deems the 1444 original IP packet too large according to factors such as link 1445 performance characteristics, reassembly congestion, etc. This 1446 ensures that the path MTU is adaptive and reflects the current path 1447 used for a given data flow. The OMNI interface can therefore 1448 continuously forward packets without loss while returning PTB soft 1449 error messages recommending a smaller size if necessary. Original 1450 sources that receive the soft errors in turn reduce the size of the 1451 packets they send (i.e., the same as for hard errors), but can soon 1452 resume sending larger packets if the soft errors subside. 1454 An OAL source sends PTB soft error messages by setting the ICMPv4 1455 header "unused" field or ICMPv6 header Code field to the value 1 if a 1456 original IP packet was deemed lost (e.g., due to reassembly timeout) 1457 or to the value 2 otherwise. The OAL source sets the PTB destination 1458 address to the original IP packet source, and sets the source address 1459 to one of its OMNI interface unicast/anycast addresses that is 1460 routable from the perspective of the original source. The OAL source 1461 then sets the MTU field to a value smaller than the original packet 1462 size but no smaller than 576 for ICMPv4 or 1280 for ICMPv6, writes 1463 the leading portion of the original IP packet into the "packet in 1464 error" field, and returns the PTB soft error to the original source. 1465 When the original source receives the PTB soft error, it temporarily 1466 reduces the size of the packets it sends the same as for hard errors 1467 but may seek to increase future packet sizes dynamically while no 1468 further soft errors are arriving. (If the original source does not 1469 recognize the soft error code, it regards the PTB the same as a hard 1470 error but should heed the retransmission advice given in [RFC8201] 1471 suggesting retransmission based on normal packetization layer 1472 retransmission timers.) 1474 An OAL destination may experience reassembly cache congestion, and 1475 can return uNA messages to the OAL source that originated the 1476 fragments (subject to rate limiting) to advertise reduced hard/soft 1477 Reassembly Limits and/or to report individual reassembly failures. 1478 The OAL destination creates a uNA message with an OMNI option 1479 containing an authentication message sub-option (if the OAL source is 1480 on an open Internetwork) followed optionally by at most one hard and 1481 one soft Reassembly Limit sub-options with reduced hard/soft values, 1482 and with one of them optionally including the leading portion an OAL 1483 first fragment containing the header of an original IP packet whose 1484 source must be notified (see: Section 12). The OAL destination 1485 encapsulates as much of the OAL first fragment (beginning with the 1486 OAL header) as will fit in the "OAL First Fragment" field of sub- 1487 option without causing the entire uNA message to exceed the minimum 1488 MPS, signs the message if an authentication sub-option is included, 1489 performs OAL encapsulation (with the its own address as the OAL 1490 source and the source address of the message that prompted the uNA as 1491 the OAL destination) and sends the message to the OAL source. 1493 When the OAL source receives the uNA message, it records the new 1494 hard/soft Reassembly Limit values for this OAL destination if the 1495 OMNI option includes Reassembly Limit sub-options. If a hard or soft 1496 Reassembly Limit sub-option includes an OAL First Fragment, the OAL 1497 source next sends a corresponding network layer PTB hard or soft 1498 error to the original source to recommend a smaller size. For hard 1499 errors, the OAL source sets the PTB Code field to 0. For soft 1500 errors, the OAL source sets the PTB Code field to 1 if the L flag in 1501 the Reassembly Limit sub-option is 1; otherwise, the OAL source sets 1502 the Code field to 2. The OAL source crafts the PTB by extracting the 1503 leading portion of the original IP packet from the OAL First Fragment 1504 field (i.e., not including the OAL header) and writes it in the 1505 "packet in error" field of a PTB with destination set to the original 1506 IP packet source and source set to one of its OMNI interface unicast/ 1507 anycast addresses that is routable from the perspective of the 1508 original source. For future transmissions, if the original IP packet 1509 is larger than the hard Reassembly Limit for this OAL destination the 1510 OAL source drops the packet and returns a PTB hard error with MTU set 1511 to the hard Reassembly Limit. If the packet is no larger than the 1512 current hard Reassembly Limit but larger than the current soft limit, 1513 the OAL source can also return PTB soft errors (subject to rate 1514 limiting) with Code set to 2 and MTU set to the current soft limit 1515 while still forwarding the packet to the OMNI destination. 1517 Original sources that receive PTB soft errors can dynamically tune 1518 the size of the original IP packets they to send to produce the best 1519 possible throughput and latency, with the understanding that these 1520 parameters may change over time due to factors such as congestion, 1521 mobility, network path changes, etc. The receipt or absence of soft 1522 errors should be seen as hints of when increasing or decreasing 1523 packet sizes may be beneficial. The OMNI interface supports 1524 continuous transmission and reception of packets of various sizes in 1525 the face of dynamically changing network conditions. Moreover, since 1526 PTB soft errors do not indicate a hard limit, original sources that 1527 receive soft errors can begin sending larger packets without waiting 1528 for the recommended 10 minutes specified for PTB hard errors 1529 [RFC1191][RFC8201]. The OMNI interface therefore provides an 1530 adaptive service that accommodates MTU diversity especially well- 1531 suited for dynamic multilink environments. 1533 6.8. OAL Requirements 1535 In light of the above, OAL sources, destinations and intermediate 1536 nodes observe the following normative requirements: 1538 o OAL sources MUST NOT send OAL fragments including original IP 1539 packets larger than the OMNI interface MTU or the OAL destination 1540 hard Reassembly Limit, i.e., whether or not fragmentation is 1541 needed. 1543 o OAL sources MUST NOT perform OAL fragmentation for original IP 1544 packets smaller than the minimum MPS minus the trailer size, and 1545 MUST produce non-final fragments that contain payloads no smaller 1546 than the minimum MPS when performing fragmentation. 1548 o OAL sources MUST NOT send OAL fragments that include any extension 1549 headers other than a single ORH and a single Fragment Header. 1551 o OAL intermediate nodes SHOULD and OAL destinations MUST 1552 unconditionally drop any OAL fragments with offset and length that 1553 would cause the reassembled packet to exceed the OMNI interface 1554 MRU and/or OAL destination hard Reassembly Limit. 1556 o OAL intermediate nodes SHOULD and OAL destinations MUST 1557 unconditionally drop any non-final OAL fragments containing a 1558 payload smaller than the minimum MPS. 1560 o OAL intermediate nodes SHOULD and OAL destinations MUST 1561 unconditionally drop OAL fragments that include any extension 1562 headers other than a single ORH and a single Fragment Header. 1564 o OAL destination nodes MUST drop any new OAL fragments with Offset 1565 and Payload length that would overlap with other fragments and/or 1566 leave holes smaller than the minimum MPS between fragments that 1567 have already been received. 1569 Note: Under the minimum MPS, ordinary 1500 byte original IP packets 1570 would require at most 4 OAL fragments, with each non-final fragment 1571 containing 400 payload bytes and the final fragment containing 302 1572 payload bytes (i.e., the final 300 bytes of the original IP packet 1573 plus the 2 octet trailer). Likewise, maximum-length 9180 byte 1574 original IP packets would require at most 23 fragments. For all 1575 packet sizes, the likelihood of successful reassembly may improve 1576 when the OMNI interface sends all fragments of the same fragmented 1577 OAL packet consecutively over the same underlying interface. 1578 Finally, an assured minimum/path MPS allows continuous operation over 1579 all paths including those that traverse bridged L2 media with 1580 dissimilar MTUs. 1582 Note: Certain legacy network hardware of the past millennium was 1583 unable to accept packet "bursts" resulting from an IP fragmentation 1584 event - even to the point that the hardware would reset itself when 1585 presented with a burst. This does not seem to be a common problem in 1586 the modern era, where fragmentation and reassembly can be readily 1587 demonstrated at line rate (e.g., using tools such as 'iperf3') even 1588 over fast links on average hardware platforms. Even so, the OAL 1589 source could impose an inter-fragment delay while the OAL destination 1590 is reporting reassembly congestion (see: Section 6.7) and decrease 1591 the delay when reassembly congestion subsides. 1593 6.9. OAL Fragmentation Security Implications 1595 As discussed in Section 3.7 of [RFC8900], there are four basic 1596 threats concerning IPv6 fragmentation; each of which is addressed by 1597 effective mitigations as follows: 1599 1. Overlapping fragment attacks - reassembly of overlapping 1600 fragments is forbidden by [RFC8200]; therefore, this threat does 1601 not apply to the OAL. 1603 2. Resource exhaustion attacks - this threat is mitigated by 1604 providing a sufficiently large OAL reassembly cache and 1605 instituting "fast discard" of incomplete reassemblies that may be 1606 part of a buffer exhaustion attack. The reassembly cache should 1607 be sufficiently large so that a sustained attack does not cause 1608 excessive loss of good reassemblies but not so large that (timer- 1609 based) data structure management becomes computationally 1610 expensive. The cache should also be indexed based on the arrival 1611 underlying interface such that congestion experienced over a 1612 first underlying interface does not cause discard of incomplete 1613 reassemblies for uncongested underlying interfaces. 1615 3. Attacks based on predictable fragment identification values - 1616 this threat is mitigated by selecting an unpredictable 1617 Identification value per [RFC7739]. Additionally, inclusion of 1618 the OAL checksum would make it very difficult for an attacker who 1619 could somehow predict a fragment identification value to inject 1620 malicious fragments resulting in undetected reassemblies of bad 1621 data. 1623 4. Evasion of Network Intrusion Detection Systems (NIDS) - since the 1624 OAL source employs a robust MPS, network-based firewalls can 1625 inspect and drop OAL fragments containing malicious data thereby 1626 disabling reassembly by the OAL destination. However, since OAL 1627 fragments may take different paths through the network (some of 1628 which may not employ a firewall) each OAL destination must also 1629 employ a firewall. 1631 Additionally, IPv4 fragmentation includes a 16-bit Identification (IP 1632 ID) field with only 65535 unique values such that at high data rates 1633 the field could wrap and apply to new carrier packets while the 1634 fragments of old packets using the same ID are still alive in the 1635 network [RFC4963]. However, since the largest carrier packet that 1636 will be sent via an IPv4 path with DF = 0 is 576 bytes any IPv4 1637 fragmentation would occur only on links with an IPv4 MTU smaller than 1638 this size, and [RFC3819] recommendations suggest that such links will 1639 have low data rates. Since IPv6 provides a 32-bit Identification 1640 value, IP ID wraparound at high data rates is not a concern for IPv6 1641 fragmentation. 1643 Finally, [RFC6980] documents fragmentation security concerns for 1644 large IPv6 ND messages. These concerns are addressed when the OMNI 1645 interface employs the OAL instead of directly fragmenting the IPv6 ND 1646 message itself. For this reason, OMNI interfaces MUST NOT send IPv6 1647 ND messages larger than the OMNI interface MTU, and MUST employ OAL 1648 encapsulation and fragmentation for IPv6 ND messages larger than the 1649 current MPS for this OAL destination. 1651 6.10. OAL Super-Packets 1653 By default, the OAL source includes a 40-byte IPv6 encapsulation 1654 header for each original IP packet during OAL encapsulation. The OAL 1655 source also calculates and appends a 2 octet trailing checksum then 1656 performs fragmentation such that a copy of the 40-byte IPv6 header 1657 plus an 8-byte IPv6 Fragment Header is included in each OAL fragment 1658 (when an ORH is added, the OAL encapsulation headers become larger 1659 still). However, these encapsulations may represent excessive 1660 overhead in some environments. OAL header compression can 1661 dramatically reduce the amount of encapsulation overhead, however a 1662 complimentary technique known as "packing" (see: 1663 [I-D.ietf-intarea-tunnels]) is also supported so that multiple 1664 original IP packets and/or control messages can be included within a 1665 single OAL "super-packet". 1667 When the OAL source has multiple original IP packets to send to the 1668 same OAL destination with total length no larger than the OAL 1669 destination MRU, it can concatenate them into a super-packet 1670 encapsulated in a single OAL header and trailing checksum. Within 1671 the OAL super-packet, the IP header of the first original IP packet 1672 (iHa) followed by its data (iDa) is concatenated immediately 1673 following the OAL header, then the IP header of the next original 1674 packet (iHb) followed by its data (iDb) is concatenated immediately 1675 following the first original packet, etc. with the trailing checksum 1676 included last. The OAL super-packet format is transposed from 1677 [I-D.ietf-intarea-tunnels] and shown in Figure 9: 1679 <------- Original IP packets -------> 1680 +-----+-----+ 1681 | iHa | iDa | 1682 +-----+-----+ 1683 | 1684 | +-----+-----+ 1685 | | iHb | iDb | 1686 | +-----+-----+ 1687 | | 1688 | | +-----+-----+ 1689 | | | iHc | iDc | 1690 | | +-----+-----+ 1691 | | | 1692 v v v 1693 +----------+-----+-----+-----+-----+-----+-----+----+ 1694 | OAL Hdr | iHa | iDa | iHb | iDb | iHc | iDc |Csum| 1695 +----------+-----+-----+-----+-----+-----+-----+----+ 1696 <--- OAL "Super-Packet" with single OAL Hdr/Csum ---> 1698 Figure 9: OAL Super-Packet Format 1700 When the OAL source prepares a super-packet, it applies OAL 1701 fragmentation and *NET encapsulation then sends the carrier packets 1702 to the OAL destination. When the OAL destination receives the super- 1703 packet it reassembles if necessary, verifies and removes the trailing 1704 checksum, then regards the remaining OAL header Payload Length as the 1705 sum of the lengths of all payload packets. The OAL destination then 1706 selectively extracts each original IP packet (e.g., by setting 1707 pointers into the super-packet buffer and maintaining a reference 1708 count, by copying each packet into a separate buffer, etc.) and 1709 forwards each packet to the network layer. During extraction, the 1710 OAL determines the IP protocol version of each successive original IP 1711 packet 'j' by examining the four most-significant bits of iH(j), and 1712 determines the length of the packet by examining the rest of iH(j) 1713 according to the IP protocol version. 1715 Note that OMNI interfaces must take care to avoid processing super- 1716 packet payload elements that would subvert security. Specifically, 1717 if a super-packet contains a mix of data and control payload packets 1718 (which could include critical security codes), the node MUST NOT 1719 process the data packets before processing the control packets 1721 7. Frame Format 1723 The OMNI interface forwards original IP packets from the network 1724 layer by first invoking the OAL to create OAL packets/fragments if 1725 necessary, then including any *NET encapsulations and finally 1726 engaging the native frame format of the underlying interface. For 1727 example, for Ethernet-compatible interfaces the frame format is 1728 specified in [RFC2464], for aeronautical radio interfaces the frame 1729 format is specified in standards such as ICAO Doc 9776 (VDL Mode 2 1730 Technical Manual), for various forms of tunnels the frame format is 1731 found in the appropriate tunneling specification, etc. 1733 See Figure 2 for a map of the various *NET layering combinations 1734 possible. For any layering combination, the final layer (e.g., UDP, 1735 IP, Ethernet, etc.) must have an assigned number and frame format 1736 representation that is compatible with the selected underlying 1737 interface. 1739 8. Link-Local Addresses (LLAs) 1741 OMNI nodes are assigned OMNI interface IPv6 Link-Local Addresses 1742 (LLAs) through pre-service administrative actions. "MNP-LLAs" embed 1743 the MNP assigned to the mobile node, while "ADM-LLAs" include an 1744 administratively-unique ID that is guaranteed to be unique on the 1745 link. LLAs are configured as follows: 1747 o IPv6 MNP-LLAs encode the most-significant 64 bits of a MNP within 1748 the least-significant 64 bits of the IPv6 link-local prefix 1749 fe80::/64, i.e., in the LLA "interface identifier" portion. The 1750 prefix length for the LLA is determined by adding 64 to the MNP 1751 prefix length. For example, for the MNP 2001:db8:1000:2000::/56 1752 the corresponding MNP-LLA is fe80::2001:db8:1000:2000/120. Non- 1753 MNP routes are also represented the same as for MNP-LLAs, but 1754 include a GUA prefix that is not properly covered by the MSP. 1756 o IPv4-compatible MNP-LLAs are constructed as fe80::ffff:[IPv4], 1757 i.e., the interface identifier consists of 16 '0' bits, followed 1758 by 16 '1' bits, followed by a 32bit IPv4 address/prefix. The 1759 prefix length for the LLA is determined by adding 96 to the MNP 1760 prefix length. For example, the IPv4-Compatible MN OMNI LLA for 1761 192.0.2.0/24 is fe80::ffff:192.0.2.0/120 (also written as 1762 fe80::ffff:c000:0200/120). 1764 o ADM-LLAs are assigned to ARs and MSEs and MUST be managed for 1765 uniqueness. The lower 32 bits of the LLA includes a unique 1766 integer "MSID" value between 0x00000001 and 0xfeffffff, e.g., as 1767 in fe80::1, fe80::2, fe80::3, etc., fe80::feffffff. The ADM-LLA 1768 prefix length is determined by adding 96 to the MSID prefix 1769 length. For example, if the prefix length for MSID 0x10012001 is 1770 16 then the ADM-LLA prefix length is set to 112 and the LLA is 1771 written as fe80::1001:2001/112. The "zero" address for each ADM- 1772 LLA prefix is the Subnet-Router anycast address for that prefix 1773 [RFC4291]; for example, the Subnet-Router anycast address for 1774 fe80::1001:2001/112 is simply fe80::1001:2000. The MSID range 1775 0xff000000 through 0xffffffff is reserved for future use. 1777 Since the prefix 0000::/8 is "Reserved by the IETF" [RFC4291], no 1778 MNPs can be allocated from that block ensuring that there is no 1779 possibility for overlap between the different MNP- and ADM-LLA 1780 constructs discussed above. 1782 Since MNP-LLAs are based on the distribution of administratively 1783 assured unique MNPs, and since ADM-LLAs are guaranteed unique through 1784 administrative assignment, OMNI interfaces set the autoconfiguration 1785 variable DupAddrDetectTransmits to 0 [RFC4862]. 1787 Note: If future protocol extensions relax the 64-bit boundary in IPv6 1788 addressing, the additional prefix bits of an MNP could be encoded in 1789 bits 16 through 63 of the MNP-LLA. (The most-significant 64 bits 1790 would therefore still be in bits 64-127, and the remaining bits would 1791 appear in bits 16 through 48.) However, the analysis provided in 1792 [RFC7421] suggests that the 64-bit boundary will remain in the IPv6 1793 architecture for the foreseeable future. 1795 Note: Even though this document honors the 64-bit boundary in IPv6 1796 addressing, it specifies prefix lengths longer than /64 for routing 1797 purposes. This effectively extends IPv6 routing determination into 1798 the interface identifier portion of the IPv6 address, but it does not 1799 redefine the 64-bit boundary. Modern routing protocol 1800 implementations honor IPv6 prefixes of all lengths, up to and 1801 including /128. 1803 9. Unique-Local Addresses (ULAs) 1805 OMNI domains use IPv6 Unique-Local Addresses (ULAs) as the source and 1806 destination addresses in OAL packet IPv6 encapsulation headers. ULAs 1807 are only routable within the scope of a an OMNI domain, and are 1808 derived from the IPv6 Unique Local Address prefix fc00::/7 followed 1809 by the L bit set to 1 (i.e., as fd00::/8) followed by a 40-bit 1810 pseudo-random Global ID to produce the prefix [ULA]::/48, which is 1811 then followed by a 16-bit Subnet ID then finally followed by a 64 bit 1812 Interface ID as specified in Section 3 of [RFC4193]. All nodes in 1813 the same OMNI domain configure the same 40-bit Global ID as the OMNI 1814 domain identifier. The statistic uniqueness of the 40-bit pseudo- 1815 random Global ID allows different OMNI domains to be joined together 1816 in the future without requiring renumbering. 1818 Each OMNI link instance is identified by a value between 0x0000 and 1819 0xfeff in bits 48-63 of [ULA]::/48; the values 0xff00 through 0xfffe 1820 are reserved for future use, and the value 0xffff denotes the 1821 presence of a Temporary ULA (see below). For example, OMNI ULAs 1822 associated with instance 0 are configured from the prefix 1823 [ULA]:0000::/64, instance 1 from [ULA]:0001::/64, instance 2 from 1824 [ULA]:0002::/64, etc. ULAs and their associated prefix lengths are 1825 configured in correspondence with LLAs through stateless prefix 1826 translation where "MNP-ULAs" are assigned in correspondence to MNP- 1827 LLAs and "ADM-ULAs" are assigned in correspondence to ADM-LLAs. For 1828 example, for OMNI link instance [ULA]:1010::/64: 1830 o the MNP-ULA corresponding to the MNP-LLA fe80::2001:db8:1:2 with a 1831 56-bit MNP length is derived by copying the lower 64 bits of the 1832 LLA into the lower 64 bits of the ULA as 1833 [ULA]:1010:2001:db8:1:2/120 (where, the ULA prefix length becomes 1834 64 plus the IPv6 MNP length). 1836 o the MNP-ULA corresponding to fe80::ffff:192.0.2.0 with a 28-bit 1837 MNP length is derived by simply writing the LLA interface ID into 1838 the lower 64 bits as [ULA]:1010:0:ffff:192.0.2.0/124 (where, the 1839 ULA prefix length is 64 plus 32 plus the IPv4 MNP length). 1841 o the ADM-ULA corresponding to fe80::1000/112 is simply 1842 [ULA]:1010::1000/112. 1844 o the ADM-ULA corresponding to fe80::/128 is simply 1845 [ULA]:1010::/128. 1847 o etc. 1849 Each OMNI interface assigns the Anycast ADM-ULA specific to the OMNI 1850 link instance. For example, the OMNI interface connected to instance 1851 3 assigns the Anycast address [ULA]:0003::/128. Routers that 1852 configure OMNI interfaces advertise the OMNI service prefix (e.g., 1853 [ULA]:0003::/64) into the local routing system so that applications 1854 can direct traffic according to SBM requirements. 1856 The ULA presents an IPv6 address format that is routable within the 1857 OMNI routing system and can be used to convey link-scoped IPv6 ND 1858 messages across multiple hops using IPv6 encapsulation [RFC2473]. 1859 The OMNI link extends across one or more underling Internetworks to 1860 include all ARs and MSEs. All MNs are also considered to be 1861 connected to the OMNI link, however OAL encapsulation is omitted 1862 whenever possible to conserve bandwidth (see: Section 14). 1864 Each OMNI link can be subdivided into "segments" that often 1865 correspond to different administrative domains or physical 1866 partitions. OMNI nodes can use IPv6 Segment Routing [RFC8402] when 1867 necessary to support efficient forwarding to destinations located in 1868 other OMNI link segments. A full discussion of Segment Routing over 1869 the OMNI link appears in [I-D.templin-6man-aero]. 1871 Temporary ULAs are constructed per [RFC8981] based on the prefix 1872 [ULA]:ffff::/64 and used by MNs when they have no other addresses. 1873 Temporary ULAs can be used for MN-to-MN communications outside the 1874 context of any supporting OMNI link infrastructure, and can also be 1875 used as an initial address while the MN is in the process of 1876 procuring an MNP. Temporary ULAs are not routable within the OMNI 1877 routing system, and are therefore useful only for OMNI link "edge" 1878 communications. Temporary ULAs employ optimistic DAD principles 1879 [RFC4429] since they are probabilistically unique. 1881 Note: IPv6 ULAs taken from the prefix fc00::/7 followed by the L bit 1882 set to 0 (i.e., as fc00::/8) are never used for OMNI OAL addressing, 1883 however the range could be used for MSP and MNP addressing under 1884 certain limiting conditions (see: Section 10). 1886 10. Global Unicast Addresses (GUAs) 1888 OMNI domains use IP Global Unicast Address (GUA) prefixes [RFC4291] 1889 as Mobility Service Prefixes (MSPs) from which Mobile Network 1890 Prefixes (MNP) are delegated to Mobile Nodes (MNs). Fixed 1891 correspondent node networks reachable from the OMNI domain are 1892 represented by non-MNP GUA prefixes that are not derived from the 1893 MSP, but are treated in all other ways the same as for MNPs. 1895 For IPv6, GUA prefixes are assigned by IANA [IPV6-GUA] and/or an 1896 associated regional assigned numbers authority such that the OMNI 1897 domain can be interconnected to the global IPv6 Internet without 1898 causing inconsistencies in the routing system. An OMNI domain could 1899 instead use ULAs with the 'L' bit set to 0 (i.e., from the prefix 1900 fc00::/8)[RFC4193], however this would require IPv6 NAT if the domain 1901 were ever connected to the global IPv6 Internet. 1903 For IPv4, GUA prefixes are assigned by IANA [IPV4-GUA] and/or an 1904 associated regional assigned numbers authority such that the OMNI 1905 domain can be interconnected to the global IPv4 Internet without 1906 causing routing inconsistencies. An OMNI domain could instead use 1907 private IPv4 prefixes (e.g., 10.0.0.0/8, etc.) [RFC3330], however 1908 this would require IPv4 NAT if the domain were ever connected to the 1909 global IPv4 Internet. 1911 11. Node Identification 1913 OMNI MNs and MSEs that connect over open Internetworks include a 1914 unique node identification value for themselves in the OMNI options 1915 of their IPv6 ND messages (see: Section 12.1.13). One useful 1916 identification value alternative is the Host Identity Tag (HIT) as 1917 specified in [RFC7401], while Hierarchical HITs (HHITs) 1918 [I-D.ietf-drip-rid] may provide a better alternative in certain 1919 domains such as the Unmanned (Air) Traffic Management (UTM) service 1920 for Unmanned Air Systems (UAS). Another alternative is the 1921 Universally Unique IDentifier (UUID) [RFC4122] which can be self- 1922 generated by a node without supporting infrastructure with very low 1923 probability of collision. 1925 When a MN is truly outside the context of any infrastructure, it may 1926 have no MNP information at all. In that case, the MN can use an IPv6 1927 temporary ULA or (H)HIT as an IPv6 source/destination address for 1928 sustained communications in Vehicle-to-Vehicle (V2V) and (multihop) 1929 Vehicle-to-Infrastructure (V2I) scenarios. The MN can also propagate 1930 the ULA/(H)HIT into the multihop routing tables of (collective) 1931 Mobile/Vehicular Ad-hoc Networks (MANETs/VANETs) using only the 1932 vehicles themselves as communications relays. 1934 When a MN connects to ARs over (non-multihop) protected-spectrum 1935 ANETs, an alternate form of node identification (e.g., MAC address, 1936 serial number, airframe identification value, VIN, etc.) may be 1937 sufficient. The MN can then include OMNI "Node Identification" sub- 1938 options (see: Section 12.1.13) in IPv6 ND messages should the need to 1939 transmit identification information over the network arise. 1941 12. Address Mapping - Unicast 1943 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1944 state and use the link-local address format specified in Section 8. 1945 OMNI interface IPv6 Neighbor Discovery (ND) [RFC4861] messages sent 1946 over physical underlying interfaces without encapsulation observe the 1947 native underlying interface Source/Target Link-Layer Address Option 1948 (S/TLLAO) format (e.g., for Ethernet the S/TLLAO is specified in 1949 [RFC2464]). OMNI interface IPv6 ND messages sent over underlying 1950 interfaces via encapsulation do not include S/TLLAOs which were 1951 intended for encoding physical L2 media address formats and not 1952 encapsulation IP addresses. Furthermore, S/TLLAOs are not intended 1953 for encoding additional interface attributes needed for multilink 1954 coordination. Hence, this document does not define an S/TLLAO format 1955 but instead defines a new option type termed the "OMNI option" 1956 designed for these purposes. 1958 MNs such as aircraft typically have many wireless data link types 1959 (e.g. satellite-based, cellular, terrestrial, air-to-air directional, 1960 etc.) with diverse performance, cost and availability properties. 1961 The OMNI interface would therefore appear to have multiple L2 1962 connections, and may include information for multiple underlying 1963 interfaces in a single IPv6 ND message exchange. OMNI interfaces use 1964 an IPv6 ND option called the OMNI option formatted as shown in 1965 Figure 10: 1967 0 1 2 3 1968 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 1969 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1970 | Type | Length | Preflen | S/T-omIndex | 1971 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1972 | Sequence Number | 1973 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1974 | Acknowledgment Number | 1975 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1976 | |U|A|P|R|S|F| | 1977 | Reserved |R|C|S|S|Y|I| Window | 1978 | |G|K|H|T|N|N| | 1979 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1980 | Urgent Pointer | | 1981 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 1982 | ~ 1983 ~ Sub-Options ~ 1984 | | 1985 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1987 Figure 10: OMNI Option Format 1989 In this format: 1991 o Type is set to TBD2. 1993 o Length is set to the number of 8 octet blocks in the option. The 1994 value 0 is invalid, while the values 1 through 255 (i.e., 8 1995 through 2040 octets, respectively) indicate the total length of 1996 the OMNI option. 1998 o Preflen is an 8 bit field that determines the length of prefix 1999 associated with an LLA. Values 0 through 128 specify a valid 2000 prefix length (all other values are invalid). For IPv6 ND 2001 messages sent from a MN to the MS, Preflen applies to the IPv6 2002 source LLA and provides the length that the MN is requesting or 2003 asserting to the MS. For IPv6 ND messages sent from the MS to the 2004 MN, Preflen applies to the IPv6 destination LLA and indicates the 2005 length that the MS is granting to the MN. For IPv6 ND messages 2006 sent between MS endpoints, Preflen provides the length associated 2007 with the source/target MN that is subject of the ND message. 2009 o S/T-omIndex is an 8 bit field corresponds to the omIndex value for 2010 source or target underlying interface pertaining to this IPv6 ND 2011 message. OMNI interfaces MUST number each distinct underlying 2012 interface with an omIndex value between '1' and '255' that 2013 represents a MN-specific 8-bit mapping for the actual ifIndex 2014 value assigned by network management [RFC2863] (the omIndex value 2015 '0' is reserved for use by the MS). For RS messages, S/T-omIndex 2016 corresponds to the source underlying interface the message 2017 originated from. For NS, RA and NA messages, S/T-omIndex 2018 corresponds to the target underlying interface that the message is 2019 destined to. 2021 o The remaining header fields before the Sub-options are coded 2022 exactly as for the Transmission Control Protocol (TCP) header 2023 specified in Section 3.1 of [RFC0793]. However, the Sequence and 2024 Acknowledgement Numbers do not correspond to a sequence of bytes; 2025 they correspond to a sequence of OAL packets identified by their 2026 corresponding Identification values, where each OAL packet 2027 consists of one or more OAL fragments transmitted as carrier 2028 packets. The connectionless and connection-oriented protocols 2029 used for the transmission of OAL packets are specified in 2030 Section 6.5. 2032 o Sub-Options is a Variable-length field, of length such that the 2033 complete OMNI Option is an integer multiple of 8 octets long. 2034 Contains one or more Sub-Options, as described in Section 12.1. 2036 The OMNI option may appear in any IPv6 ND message type; it is 2037 processed by interfaces that recognize the option and ignored by all 2038 other interfaces. If multiple OMNI option instances appear in the 2039 same IPv6 ND message, the interface processes the Preflen and S/ 2040 T-omIndex fields in the first instance and ignores those fields in 2041 all other instances. The interface processes the Sub-Options of all 2042 OMNI option instances in the same IPv6 ND message in the consecutive 2043 order in which they appear. 2045 The OMNI option(s) in each IPv6 ND message may include full or 2046 partial information for the neighbor. The union of the information 2047 in the most recently received OMNI options is therefore retained, and 2048 the information is aged/removed in conjunction with the corresponding 2049 neighbor cache entry. 2051 12.1. Sub-Options 2053 Each OMNI option includes zero or more Sub-Options. Each consecutive 2054 Sub-Option is concatenated immediately after its predecessor. All 2055 Sub-Options except Pad1 (see below) are in type-length-value (TLV) 2056 encoded in the following format: 2058 0 1 2 2059 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 2060 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 2061 | Sub-Type| Sub-length | Sub-Option Data ... 2062 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 2064 Figure 11: Sub-Option Format 2066 o Sub-Type is a 5-bit field that encodes the Sub-Option type. Sub- 2067 Options defined in this document are: 2069 Sub-Option Name Sub-Type 2070 Pad1 0 2071 PadN 1 2072 Interface Attributes (Type 1) 2 2073 Interface Attributes (Type 2) 3 2074 Interface Attributes (Type 4) 4 2075 MS-Register 5 2076 MS-Release 6 2077 Geo Coordinates 7 2078 DHCPv6 Message 8 2079 HIP Message 9 2080 PIM-SM Message 10 2081 Reassembly Limit 11 2082 Fragmentation Report 12 2083 Node Identification 13 2084 Sub-Type Extension 30 2086 Figure 12 2088 Sub-Types 14-29 are available for future assignment for major 2089 protocol functions. Sub-Type 31 is reserved by IANA. 2091 o Sub-Length is an 11-bit field that encodes the length of the Sub- 2092 Option Data ranging from 0 to 2034 octets. 2094 o Sub-Option Data is a block of data with format determined by Sub- 2095 Type and length determined by Sub-Length. 2097 During transmission, the OMNI interface codes Sub-Type and Sub-Length 2098 together in network byte order in 2 consecutive octets, where Sub- 2099 Option Data may be up to 2034 octets in length. This allows ample 2100 space for coding large objects (e.g., ASCII strings, domain names, 2101 protocol messages, security codes, etc.), while a single OMNI option 2102 is limited to 2040 octets the same as for any IPv6 ND option. If the 2103 Sub-Options to be coded would cause an OMNI option to exceed 2040 2104 octets, the OMNI interface codes any remaining Sub-Options in 2105 additional OMNI option instances in the intended order of processing 2106 in the same IPv6 ND message. Implementations must therefore observe 2107 size limitations, and must refrain from sending IPv6 ND messages 2108 larger than the OMNI interface MTU. If the available OMNI 2109 information would cause a single IPv6 ND message to exceed the OMNI 2110 interface MTU, the OMNI interface codes as much as possible in a 2111 first IPv6 ND message and codes the remainder in additional IPv6 ND 2112 messages. 2114 During reception, the OMNI interface processes each OMNI option Sub- 2115 Option while skipping over and ignoring any unrecognized Sub-Options. 2116 The OMNI interface processes the Sub-Options of all OMNI option 2117 instances in the consecutive order in which they appear in the IPv6 2118 ND message, beginning with the first instance and continuing through 2119 any additional instances to the end of the message. If a Sub-Option 2120 length would cause processing to exceed the OMNI option total length, 2121 the OMNI interface accepts any Sub-Options already processed and 2122 ignores the final Sub-Option. The interface then processes any 2123 remaining OMNI options in the same fashion to the end of the IPv6 ND 2124 message. 2126 Note: large objects that exceed the Sub-Option Data limit of 2034 2127 octets are not supported under the current specification; if this 2128 proves to be limiting in practice, future specifications may define 2129 support for fragmenting large objects across multiple OMNI options 2130 within the same IPv6 ND message. 2132 The following Sub-Option types and formats are defined in this 2133 document: 2135 12.1.1. Pad1 2137 0 2138 0 1 2 3 4 5 6 7 2139 +-+-+-+-+-+-+-+-+ 2140 | S-Type=0|x|x|x| 2141 +-+-+-+-+-+-+-+-+ 2143 Figure 13: Pad1 2145 o Sub-Type is set to 0. If multiple instances appear in OMNI 2146 options of the same message all are processed. 2148 o Sub-Type is followed by 3 'x' bits, set to any value on 2149 transmission (typically all-zeros) and ignored on receipt. Pad1 2150 therefore consists of 1 octet with the most significant 5 bits set 2151 to 0, and with no Sub-Length or Sub-Option Data fields following. 2153 12.1.2. PadN 2155 0 1 2 2156 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 2157 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 2158 | S-Type=1| Sub-length=N | N padding octets ... 2159 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 2161 Figure 14: PadN 2163 o Sub-Type is set to 1. If multiple instances appear in OMNI 2164 options of the same message all are processed. 2166 o Sub-Length is set to N (from 0 to 2034) that encodes the number of 2167 padding octets that follow. 2169 o Sub-Option Data consists of N octets, set to any value on 2170 transmission (typically all-zeros) and ignored on receipt. 2172 12.1.3. Interface Attributes (Types 1 through 3) 2174 Interface Attributes (Type 1) and (Type 2) were defined in 2175 [I-D.templin-6man-omni-interface] and have been moved to historic 2176 status. Their sub-option types (2 and 3) are reserved for future 2177 use. 2179 Interface Attributes (Type 3) was never defined; the number was 2180 skipped to bring (Type 4) into agreement with the corresponding sub- 2181 option Type value. 2183 12.1.4. Interface Attributes (Type 4) 2185 The Interface Attributes (Type 4) sub-option provides L2 forwarding 2186 information for the multilink conceptual sending algorithm discussed 2187 in Section 14. The L2 information is used for selecting among 2188 potentially multiple candidate underlying interfaces that can be used 2189 to forward carrier packets to the neighbor based on factors such as 2190 traffic selectors and link quality. Interface Attributes (Type 4) 2191 further includes link-layer address information to be used for either 2192 OAL encapsulation or direct UDP/IP encapsulation (when OAL 2193 encapsulation can be avoided). 2195 Interface Attributes (Type 4) are the sole Interface Attributes 2196 format in this specification that all OMNI nodes must honor. 2197 Wherever the term "Interface Attributes" occurs throughout this 2198 specification without a "Type" designation, the format given below is 2199 indicated: 2201 0 1 2 3 2202 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 2203 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2204 | S-Type=4| Sub-length=N | omIndex | omType | 2205 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2206 | Provider ID | Link |RES|A|T| SRT | FMT | LHS (0 - 7) | 2207 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2208 | LHS (bits 8 - 31) | ~ 2209 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 2210 ~ ~ 2211 ~ Link Layer Address (L2ADDR) ~ 2212 ~ ~ 2213 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2214 ~ ~ 2215 ~ RFC 6088 Format Traffic Selector ~ 2216 ~ ~ 2217 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2219 Figure 15: Interface Attributes (Type 4) 2221 o Sub-Type is set to 4. If multiple instances with different 2222 omIndex values appear in OMNI options of the same message all are 2223 processed; if multiple instances with the same omIndex value 2224 appear, the first is processed and all others are ignored. 2226 o Sub-Length is set to N (from 4 to 2034) that encodes the number of 2227 Sub-Option Data octets that follow. The 'omIndex', 'omType', 2228 'Provider ID', 'Link', 'R' and 'API' fields are always present; 2229 hence, the remainder of the Sub-Option Data is limited to 2030 2230 octets. 2232 o Sub-Option Data contains an "Interface Attributes (Type 4)" option 2233 encoded as follows: 2235 * omIndex is set to an 8-bit integer value corresponding to a 2236 specific underlying interface the same as specified above for 2237 the OMNI option S/T-omIndex field. The OMNI options of a same 2238 message may include multiple Interface Attributes Sub-Options, 2239 with each distinct omIndex value pertaining to a different 2240 underlying interface. The OMNI option will often include an 2241 Interface Attributes Sub-Option with the same omIndex value 2242 that appears in the S/T-omIndex. In that case, the actual 2243 encapsulation address of the received IPv6 ND message should be 2244 compared with the L2ADDR encoded in the Sub-Option (see below); 2245 if the addresses are different (or, if L2ADDR is absent) the 2246 presence of a NAT is assumed. 2248 * omType is set to an 8-bit integer value corresponding to the 2249 underlying interface identified by omIndex. The value 2250 represents an OMNI interface-specific 8-bit mapping for the 2251 actual IANA ifType value registered in the 'IANAifType-MIB' 2252 registry [http://www.iana.org]. 2254 * Provider ID is set to an OMNI interface-specific 8-bit ID value 2255 for the network service provider associated with this omIndex. 2257 * Link encodes a 4-bit link metric. The value '0' means the link 2258 is DOWN, and the remaining values mean the link is UP with 2259 metric ranging from '1' ("lowest") to '15' ("highest"). 2261 * RES is 2-bit field reserved for future use, set to 0 on 2262 transmit and ignored on receipt. 2264 * A - an "Address" bit. When set to 1, the SRT, FMT, LHS and 2265 L2ADDR fields are included immediately following the flags; 2266 else, they are omitted. 2268 * T - a "Traffic Selector" bit. When set to 1 an [RFC6088] 2269 format Traffic Selctor is included either following the flags 2270 (if A is 0) or following the address information (if A is 1). 2272 * When an "Address" is included, the following fields appear in 2273 consecutive order (else, they are omitted): 2275 + SRT - a 5-bit Segment Routing Topology prefix length value 2276 that (when added to 96) determines the prefix length to 2277 apply to the ULA formed from concatenating [ULA*]::/96 with 2278 the 32 bit LHS MSID value that follows. For example, the 2279 value 16 corresponds to the prefix length 112. 2281 + FMT - a 3-bit "Framework/Mode/Type" code corresponding to 2282 the included Link Layer Address as follows: 2284 - When the most significant bit (i.e., "Framework") is set 2285 to 1, L2ADDR is the INET encapsulation address for the 2286 Source/Target Client itself; otherwise L2ADDR is the 2287 address of the Proxy/Server named in the LHS. 2289 - When the next most significant bit (i.e., "Mode") is set 2290 to 1, the Framework node is (likely) located behind an 2291 INET Network Address Translator (NAT); otherwise, it is 2292 on the open INET. 2294 - When the least significant bit (i.e., "Type") is set to 2295 0, L2ADDR includes a UDP Port Number followed by an IPv4 2296 address; otherwise, it includes a UDP Port Number 2297 followed by an IPv6 address. 2299 + LHS - the 32 bit MSID of the Last Hop Proxy/Server on the 2300 path to the target. When SRT and LHS are both set to 0, the 2301 LHS is considered unspecified in this IPv6 ND message. When 2302 SRT is set to 0 and LHS is non-zero, the prefix length is 2303 set to 128. SRT and LHS together provide guidance to the 2304 OMNI interface forwarding algorithm. Specifically, if SRT/ 2305 LHS is located in the local OMNI link segment then the OMNI 2306 interface can encapsulate according to FMT/L2ADDR (following 2307 any necessary NAT traversal messaging); else, it must 2308 forward according to the OMNI link spanning tree. See 2309 [I-D.templin-6man-aero] for further discussion. 2311 + Link Layer Address (L2ADDR) - Formatted according to FMT, 2312 and identifies the link-layer address (i.e., the 2313 encapsulation address) of the source/target. The UDP Port 2314 Number appears in the first 2 octets and the IP address 2315 appears in the next 4 octets for IPv4 or 16 octets for IPv6. 2316 The Port Number and IP address are recorded in network byte 2317 order, and in ones-compliment "obfuscated" form per 2318 [RFC4380]. The OMNI interface forwarding algorithm uses 2319 FMT/L2ADDR to determine the encapsulation address for 2320 forwarding when SRT/LHS is located in the local OMNI link 2321 segment. Note that if the target is behind a NAT, L2ADDR 2322 will contain the mapped INET address stored in the NAT; 2323 otherwise, L2ADDR will contain the native INET information 2324 of the target itself. 2326 * When a "Traffic Selector" is included, the remainder of the 2327 sub-option (i.e., following the flags if A=0, or following the 2328 address information if A=1) includes a traffic selector 2329 formatted per [RFC6088] beginning with the "TS Format" field. 2331 12.1.5. MS-Register 2333 0 1 2 3 2334 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 2335 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2336 | S-Type=5| Sub-length=4n | MSID[1] (bits 0 - 15) | 2337 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2338 | MSID [1] (bits 16 - 32) | MSID[2] (bits 0 - 15) | 2339 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2340 | MSID [2] (bits 16 - 32) | MSID[3] (bits 0 - 15) | 2341 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2342 ... ... ... ... ... ... 2343 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2344 | MSID [n] (bits 16 - 32) | 2345 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2347 Figure 16: MS-Register Sub-option 2349 o Sub-Type is set to 5. If multiple instances appear in OMNI 2350 options of the same message all are processed. Only the first 2351 MAX_MSID values processed (whether in a single instance or 2352 multiple) are retained and all other MSIDs are ignored. 2354 o Sub-Length is set to 4n, with 508 as the maximum value for n. The 2355 length of the Sub-Option Data section is therefore limited to 2032 2356 octets. 2358 o A list of n 4 octet MSIDs is included in the following 4n octets. 2359 The Anycast MSID value '0' in an RS message MS-Register sub-option 2360 requests the recipient to return the MSID of a nearby MSE in a 2361 corresponding RA response. 2363 12.1.6. MS-Release 2364 0 1 2 3 2365 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 2366 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2367 | S-Type=6| Sub-length=4n | MSID[1] (bits 0 - 15) | 2368 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2369 | MSID [1] (bits 16 - 32) | MSID[2] (bits 0 - 15) | 2370 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2371 | MSID [2] (bits 16 - 32) | MSID[3] (bits 0 - 15) | 2372 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2373 ... ... ... ... ... ... 2374 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2375 | MSID [n] (bits 16 - 32) | 2376 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2378 Figure 17: MS-Release Sub-option 2380 o Sub-Type is set to 6. If multiple instances appear in OMNI 2381 options of the same message all are processed. Only the first 2382 MAX_MSID values processed (whether in a single instance or 2383 multiple) are retained and all other MSIDs are ignored. 2385 o Sub-Length is set to 4n, with 508 as the maximum value for n. The 2386 length of the Sub-Option Data section is therefore limited to 2032 2387 octets. 2389 o A list of n 4 octet MSIDs is included in the following 4n octets. 2390 The Anycast MSID value '0' is ignored in MS-Release sub-options, 2391 i.e., only non-zero values are processed. 2393 12.1.7. Geo Coordinates 2395 0 1 2 3 2396 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 2397 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2398 | S-Type=7| Sub-length=N | Geo Type |Geo Coordinates 2399 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... 2401 Figure 18: Geo Coordinates Sub-option 2403 o Sub-Type is set to 7. If multiple instances appear in OMNI 2404 options of the same message the first is processed and all others 2405 are ignored. 2407 o Sub-Length is set to N (from 0 to 2034) that encodes the number of 2408 Sub-Option Data octets that follow. 2410 o Geo Type is a 1 octet field that encodes a type designator that 2411 determines the format and contents of the Geo Coordinates field 2412 that follows. The following types are currently defined: 2414 * 0 - NULL, i.e., the Geo Coordinates field is zero-length. 2416 o A set of Geo Coordinates of length 0 - 2033 octets. New formats 2417 to be specified in future documents and may include attributes 2418 such as latitude/longitude, altitude, heading, speed, etc. 2420 12.1.8. Dynamic Host Configuration Protocol for IPv6 (DHCPv6) Message 2422 The Dynamic Host Configuration Protocol for IPv6 (DHCPv6) sub-option 2423 may be included in the OMNI options of RS messages sent by MNs and RA 2424 messages returned by MSEs. ARs that act as proxys to forward RS/RA 2425 messages between MNs and MSEs also forward DHCPv6 sub-options 2426 unchanged and do not process DHCPv6 sub-options themselves. Note 2427 that DHCPv6 message sub-option integrity is protected by the Checksum 2428 included in the IPv6 ND message header. 2430 0 1 2 3 2431 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 2432 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2433 | S-Type=8| Sub-length=N | msg-type | id (octet 0) | 2434 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2435 | transaction-id (octets 1-2) | | 2436 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 2437 | | 2438 . DHCPv6 options . 2439 . (variable number and length) . 2440 | | 2441 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2443 Figure 19: DHCPv6 Message Sub-option 2445 o Sub-Type is set to 8. If multiple instances appear in OMNI 2446 options of the same message the first is processed and all others 2447 are ignored. 2449 o Sub-Length is set to N (from 4 to 2034) that encodes the number of 2450 Sub-Option Data octets that follow. The 'msg-type' and 2451 'transaction-id' fields are always present; hence, the length of 2452 the DHCPv6 options is restricted to 2030 octets. 2454 o 'msg-type' and 'transaction-id' are coded according to Section 8 2455 of [RFC8415]. 2457 o A set of DHCPv6 options coded according to Section 21 of [RFC8415] 2458 follows. 2460 12.1.9. Host Identity Protocol (HIP) Message 2462 The Host Identity Protocol (HIP) Message sub-option may be included 2463 in the OMNI options of RS messages sent by MNs and RA messages 2464 returned by ARs. ARs that act as proxys authenticate and remove HIP 2465 messages in RS messages they forward from a MN to an MSE. ARs that 2466 act as proxys insert and sign HIP messages in the RA messages they 2467 forward from an MSE to a MN. 2469 The HIP message sub-option may also be included in any IPv6 ND 2470 message that may traverse an open Internetwork, i.e., where link- 2471 layer authentication is not already assured by lower layers. 2473 0 1 2 3 2474 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 2475 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2476 | S-Type=9| Sub-length=N |0| Packet Type |Version| RES.|1| 2477 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2478 | Checksum | Controls | 2479 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2480 | Sender's Host Identity Tag (HIT) | 2481 | | 2482 | | 2483 | | 2484 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2485 | Receiver's Host Identity Tag (HIT) | 2486 | | 2487 | | 2488 | | 2489 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2490 | | 2491 / HIP Parameters / 2492 / / 2493 | | 2494 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2496 Figure 20: HIP Message Sub-option 2498 o Sub-Type is set to 9. If multiple instances appear in OMNI 2499 options of the same message the first is processed and all others 2500 are ignored. 2502 o Sub-Length is set to N, i.e., the length of the option in octets 2503 beginning immediately following the Sub-Length field and extending 2504 to the end of the HIP parameters. The length of the entire HIP 2505 message is therefore restricted to 2034 octets. 2507 o The HIP message is coded exactly as specified in Section 5 of 2508 [RFC7401], except that the OMNI "Sub-Type" and "Sub-Length" fields 2509 replace the first 2 octets of the HIP message header (i.e., the 2510 Next Header and Header Length fields). Note that, since the IPv6 2511 ND message header already includes a Checksum, the HIP message 2512 Checksum field is set to 0 on transmission and ignored on 2513 reception. (The Checksum field is still included to retain the 2514 [RFC7401] message format.) 2516 12.1.10. PIM-SM Message 2518 The Protocol Independent Multicast - Sparse Mode (PIM-SM) Message 2519 sub-option may be included in the OMNI options of IPv6 ND messages 2520 sent by MNs and MSEs. PIM-SM messages are formatted as specified in 2521 Section 4.9 of [RFC7761], with the exception that the Checksum field 2522 is omitted since the IPv6 ND message is already protected by a 2523 checksum (and possibly also an authentication signature). The PIM-SM 2524 message sub-option format is shown in Figure 21: 2526 0 1 2 3 2527 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 2528 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2529 |S-Type=10| Sub-length=N |PIM Ver| Type | Reserved | 2530 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2531 | | 2532 / PIM-SM Message / 2533 / / 2534 | | 2535 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2537 Figure 21: PIM-SM Message Option Format 2539 o Sub-Type is set to 10. If multiple instances appear in OMNI 2540 options of the same message all are processed. 2542 o Sub-Length is set to N, i.e., the length of the option in octets 2543 beginning immediately following the Sub-Length field and extending 2544 to the end of the PIM-SM message. The length of the entire PIM-SM 2545 message is therefore restricted to 2034 octets. 2547 o The PIM-SM message is coded exactly as specified in Section 4.9 of 2548 [RFC7761], except that the Checksum field is omitted. The "PIM 2549 Ver" field MUST encode the value 2, and the "Type" field encodes 2550 the PIM message type. (See Section 4.9 of [RFC7761] for a list of 2551 PIM-SM message types and formats.) 2553 12.1.11. Reassembly Limit 2555 The Reassembly Limit sub-option may be included in the OMNI options 2556 of IPv6 ND messages. The message consists of a 14-bit Reassembly 2557 Limit value, followed by two flag bits (H, L) optionally followed by 2558 an (N-2)-octet leading portion of an OAL First Fragment that 2559 triggered the message. 2561 0 1 2 3 2562 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 2563 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2564 |S-Type=11| Sub-length=N | Reassembly Limit |H|L| 2565 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2566 | OAL First Fragment (As much of invoking packet | 2567 + as possible without the IPv6 ND message + 2568 | exceeding the minimum IPv6 MTU) | 2569 + + 2571 Figure 22: Reassembly Limit 2573 o Sub-Type is set to 11. If multiple instances appear in OMNI 2574 options of the same message the first occurring "hard" and "soft" 2575 Reassembly Limit values are accepted, and any additional 2576 Reassembly Limit values are ignored. 2578 o Sub-Length is set to 2 if no OAL First Fragment is included, or to 2579 a value N greater than 2 if an OAL First Fragment is included. 2581 o A 14-bit Reassembly Limit follows, and includes a value between 2582 1500 and 9180. If any other value is included, the sub-option is 2583 ignored. The value indicates the hard or soft limit for original 2584 IP packets that the source of the message is currently willing to 2585 reassemble; the source may increase or decrease the hard or soft 2586 limit at any time through the transmission of new IPv6 ND 2587 messages. Until the first IPv6 ND message with a Reassembly Limit 2588 sub-option arrives, OMNI nodes assume initial default hard/soft 2589 limits of 9180 bytes (I.e., the OMNI interface MRU). After IPv6 2590 ND messages with Reassembly Limit sub-options arrive, the OMNI 2591 node retains the most recent hard/soft limit values until new IPv6 2592 ND messages with different values arrive. 2594 o The 'H' flag is set to 1 if the Reassembly Limit is a "Hard" 2595 limit, and set to 0 if the Reassembly Limit is a "Soft" limit. 2597 o The 'L' flag is set to 1 if an OAL First Fragment corresponding to 2598 a reassembly loss event was included; otherwise set to 0. 2600 o If N is greater than 2, the remainder of the Reassembly Limit sub- 2601 option encodes the leading portion of an OAL First Fragment that 2602 prompted this IPv6 ND message. The first fragment is included 2603 beginning with the OAL IPv6 header, and continuing with as much of 2604 the fragment payload as possible without causing the IPv6 ND 2605 message to exceed the minimum IPv6 MTU. (Note that only the OAL 2606 First Fragment is consulted regardless of its size, and without 2607 waiting for additional fragments.) 2609 12.1.12. Fragmentation Report 2611 The Fragmentation Report may be included in the OMNI options of uNA 2612 messages sent from an OAL destination to an OAL source. The message 2613 consists of (N / 8)-many (Identification, Bitmap)-tuples which 2614 include the Identification values of OAL fragments received plus a 2615 Bitmap marking the ordinal positions of individual fragments received 2616 and fragments missing. 2618 0 1 2 3 2619 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 2620 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2621 |S-Type=12| Sub-Length = N | Identification #1 (bits 0 -15)| 2622 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2623 | Identification #1 (bits 15-31)| Bitmap #1 (bits 0 - 15) | 2624 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2625 | Bitmap #1 (bits 16-31) | Identification #2 (bits 0 -15)| 2626 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2627 | Identification #2 (bits 15-31)| Bitmap #2 (bits 0 - 15) | 2628 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2629 | Bitmap #2 (bits 16-31) | Identification #3 (bits 0 -15)| 2630 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2631 | Identification #3 (bits 15-31)| Bitmap #3 (bits 0 - 15) | 2632 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2633 | Bitmap #3 (bits 16-31) | ... | 2634 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... + 2635 | ... | 2636 + ... + 2638 Figure 23: Fragmentation Report 2640 o Sub-Type is set to 12. If multiple instances appear in OMNI 2641 options of the same message all are processed. 2643 o Sub-Length is set to N, i.e., the length of the option in octets 2644 beginning immediately following the Sub-Length field and extending 2645 to the end of the ICMPv6 error message body. N must be an 2646 integral multiple of 8 octets; otherwise, the sub-option is 2647 ignored. The length of the entire sub-option should not cause the 2648 entire IPv6 ND message to exceed the minimum MPS. 2650 o Identification (i) includes the IPv6 Identification value found in 2651 the Fragment Header of a received OAL fragment. (Only those 2652 Identification values included represent fragments for which loss 2653 was unambiguously observed; any Identification values not included 2654 correspond to fragments that were either received in their 2655 entirety or are still in transit.) 2657 o Bitmap (i) includes an ordinal checklist of fragments, with each 2658 bit set to 1 for a fragment received or 0 for a fragment missing. 2659 For example, for a 20-fragment fragmented OAL packet with ordinal 2660 fragments #3, #10, #13 and #17 missing and all other fragments 2661 received, the bitmap would encode: 2663 0 1 2 2664 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 2665 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 2666 |1|1|1|0|1|1|1|1|1|1|0|1|1|0|1|1|1|0|1|1|0|0|0|... 2667 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 2669 Figure 24 2671 (Note that loss of an OAL atomic fragment is indicated by a 2672 Bitmap(i) with all bits set to 0.) 2674 12.1.13. Node Identification 2676 0 1 2 3 2677 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 2678 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2679 |S-Type=13| Sub-length=N | ID-Type | ~ 2680 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 2681 ~ Node Identification Value (N-1 octets) ~ 2682 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2684 Figure 25: Node Identification 2686 o Sub-Type is set to 13. If multiple instances appear in OMNI 2687 options of the same IPv6 ND message the first instance of a 2688 specific ID-Type is processed and all other instances of the same 2689 ID-Type are ignored. (Note therefore that it is possible for a 2690 single IPv6 ND message to convey multiple Node Identifications - 2691 each having a different ID-Type.) 2693 o Sub-Length is set to N (from 1 to 2034) that encodes the number of 2694 Sub-Option Data octets that follow. The ID-Type field is always 2695 present; hence, the maximum Node Identification Value length is 2696 2033 octets. 2698 o ID-Type is a 1 octet field that encodes the type of the Node 2699 Identification Value. The following ID-Type values are currently 2700 defined: 2702 * 0 - Universally Unique IDentifier (UUID) [RFC4122]. Indicates 2703 that Node Identification Value contains a 16 octet UUID. 2705 * 1 - Host Identity Tag (HIT) [RFC7401]. Indicates that Node 2706 Identification Value contains a 16 octet HIT. 2708 * 2 - Hierarchical HIT (HHIT) [I-D.ietf-drip-rid]. Indicates 2709 that Node Identification Value contains a 16 octet HHIT. 2711 * 3 - Network Access Identifier (NAI) [RFC7542]. Indicates that 2712 Node Identification Value contains an N-1 octet NAI. 2714 * 4 - Fully-Qualified Domain Name (FQDN) [RFC1035]. Indicates 2715 that Node Identification Value contains an N-1 octet FQDN. 2717 * 5 - 252 - Unassigned. 2719 * 253-254 - Reserved for experimentation, as recommended in 2720 [RFC3692]. 2722 * 255 - reserved by IANA. 2724 o Node Identification Value is an (N - 1) octet field encoded 2725 according to the appropriate the "ID-Type" reference above. 2727 When a Node Identification Value is needed for DHCPv6 messaging 2728 purposes, it is encoded as a DHCP Unique IDentifier (DUID) using the 2729 "DUID-EN for OMNI" format with enterprise number 45282 (see: 2730 Section 25) as shown in Figure 26: 2732 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 2733 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2734 | DUID-Type (2) | EN (high bits == 0) | 2735 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2736 | EN (low bits = 45282) | ID-Type | | 2737 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 2738 . Node Identification Value . 2739 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2741 Figure 26: DUID-EN for OMNI Format 2743 In this format, the ID-Type and Node Identification Value fields are 2744 coded exactly as in Figure 25 following the 6 octet DUID-EN header, 2745 and the entire "DUID-EN for OMNI" is included in a DHCPv6 message per 2746 [RFC8415]. 2748 12.1.14. Sub-Type Extension 2750 Since the Sub-Type field is only 5 bits in length, future 2751 specifications of major protocol functions may exhaust the remaining 2752 Sub-Type values available for assignment. This document therefore 2753 defines Sub-Type 30 as an "extension", meaning that the actual sub- 2754 option type is determined by examining a 1 octet "Extension-Type" 2755 field immediately following the Sub-Length field. The Sub-Type 2756 Extension is formatted as shown in Figure 27: 2758 0 1 2 3 2759 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 2760 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2761 |S-Type=30| Sub-length=N | Extension-Type| ~ 2762 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 2763 ~ ~ 2764 ~ Extension-Type Body ~ 2765 ~ ~ 2766 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2768 Figure 27: Sub-Type Extension 2770 o Sub-Type is set to 30. If multiple instances appear in OMNI 2771 options of the same message all are processed, where each 2772 individual extension defines its own policy for processing 2773 multiple of that type. 2775 o Sub-Length is set to N (from 1 to 2034) that encodes the number of 2776 Sub-Option Data octets that follow. The Extension-Type field is 2777 always present; hence, the maximum Extension-Type Body length is 2778 2033 octets. 2780 o Extension-Type contains a 1 octet Sub-Type Extension value between 2781 0 and 255. 2783 o Extension-Type Body contains an N-1 octet block with format 2784 defined by the given extension specification. 2786 Extension-Type values 2 through 252 are available for assignment by 2787 future specifications, which must also define the format of the 2788 Extension-Type Body and its processing rules. Extension-Type values 2789 253 and 254 are reserved for experimentation, as recommended in 2791 [RFC3692], and value 255 is reserved by IANA. Extension-Type values 2792 0 and 1 are defined in the following subsections: 2794 12.1.14.1. RFC4380 UDP/IP Header Option 2796 0 1 2 3 2797 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 2798 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2799 |S-Type=30| Sub-length=N | Ext-Type=0 | Header Type | 2800 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2801 ~ Header Option Value ~ 2802 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2804 Figure 28: RFC4380 UDP/IP Header Option (Extension-Type 0) 2806 o Sub-Type is set to 30. 2808 o Sub-Length is set to N (from 2 to 2034) that encodes the number of 2809 Sub-Option Data octets that follow. The Extension-Type and Header 2810 Type fields are always present; hence, the maximum-length Header 2811 Option Value is 2032 octets. 2813 o Extension-Type is set to 0. Each instance encodes exactly one 2814 header option per Section 5.1.1 of [RFC4380], with the leading '0' 2815 octet omitted and the following octet coded as Header Type. If 2816 multiple instances of the same Header Type appear in OMNI options 2817 of the same message the first instance is processed and all others 2818 are ignored. 2820 o Header Type and Header Option Value are coded exactly as specified 2821 in Section 5.1.1 of [RFC4380]; the following types are currently 2822 defined: 2824 * 0 - Origin Indication (IPv4) - value coded per Section 5.1.1 of 2825 [RFC4380]. 2827 * 1 - Authentication Encapsulation - value coded per 2828 Section 5.1.1 of [RFC4380]. 2830 * 2 - Origin Indication (IPv6) - value coded per Section 5.1.1 of 2831 [RFC4380], except that the address is a 16-octet IPv6 address 2832 instead of a 4-octet IPv4 address. 2834 o Header Type values 3 through 252 are available for assignment by 2835 future specifications, which must also define the format of the 2836 Header Option Value and its processing rules. Header Type values 2837 253 and 254 are reserved for experimentation, as recommended in 2838 [RFC3692], and value 255 is Reserved by IANA. 2840 12.1.14.2. RFC6081 UDP/IP Trailer Option 2842 0 1 2 3 2843 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 2844 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2845 |S-Type=30| Sub-length=N | Ext-Type=1 | Trailer Type | 2846 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2847 ~ Trailer Option Value ~ 2848 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2850 Figure 29: RFC6081 UDP/IP Trailer Option (Extension-Type 1) 2852 o Sub-Type is set to 30. 2854 o Sub-Length is set to N (from 2 to 2034) that encodes the number of 2855 Sub-Option Data octets that follow. The Extension-Type and 2856 Trailer Type fields are always present; hence, the maximum-length 2857 Trailer Option Value is 2032 octets. 2859 o Extension-Type is set to 1. Each instance encodes exactly one 2860 trailer option per Section 4 of [RFC6081]. If multiple instances 2861 of the same trailer type appear in OMNI options of the same 2862 message the first instance is processed and all others ignored. 2864 o Trailer Type and Trailer Option Value are coded exactly as 2865 specified in Section 4 of [RFC6081]; the following Trailer Types 2866 are currently defined: 2868 * 0 - Unassigned 2870 * 1 - Nonce Trailer - value coded per Section 4.2 of [RFC6081]. 2872 * 2 - Unassigned 2874 * 3 - Alternate Address Trailer (IPv4) - value coded per 2875 Section 4.3 of [RFC6081]. 2877 * 4 - Neighbor Discovery Option Trailer - value coded per 2878 Section 4.4 of [RFC6081]. 2880 * 5 - Random Port Trailer - value coded per Section 4.5 of 2881 [RFC6081]. 2883 * 6 - Alternate Address Trailer (IPv6) - value coded per 2884 Section 4.3 of [RFC6081], except that each address is a 2885 16-octet IPv6 address instead of a 4-octet IPv4 address. 2887 o Trailer Type values 7 through 252 are available for assignment by 2888 future specifications, which must also define the format of the 2889 Trailer Option Value and its processing rules. Trailer Type 2890 values 253 and 254 are reserved for experimentation, as 2891 recommended in [RFC3692], and value 255 is Reserved by IANA. 2893 13. Address Mapping - Multicast 2895 The multicast address mapping of the native underlying interface 2896 applies. The mobile router on board the MN also serves as an IGMP/ 2897 MLD Proxy for its EUNs and/or hosted applications per [RFC4605] while 2898 using the L2 address of the AR as the L2 address for all multicast 2899 packets. 2901 The MN uses Multicast Listener Discovery (MLDv2) [RFC3810] to 2902 coordinate with the AR, and *NET L2 elements use MLD snooping 2903 [RFC4541]. 2905 14. Multilink Conceptual Sending Algorithm 2907 The MN's IPv6 layer selects the outbound OMNI interface according to 2908 SBM considerations when forwarding original IP packets from local or 2909 EUN applications to external correspondents. Each OMNI interface 2910 maintains a neighbor cache the same as for any IPv6 interface, but 2911 with additional state for multilink coordination. Each OMNI 2912 interface maintains default routes via ARs discovered as discussed in 2913 Section 15, and may configure more-specific routes discovered through 2914 means outside the scope of this specification. 2916 After an original IP packet enters the OMNI interface, one or more 2917 outbound underlying interfaces are selected based on PBM traffic 2918 attributes, and one or more neighbor underlying interfaces are 2919 selected based on the receipt of Interface Attributes sub-options in 2920 IPv6 ND messages (see: Section 12.1.4). Underlying interface 2921 selection for the nodes own local interfaces are based on traffic 2922 selectors, cost, performance, message size, etc. OMNI interface 2923 multilink selections could also be configured to perform replication 2924 across multiple underlying interfaces for increased reliability at 2925 the expense of packet duplication. The set of all Interface 2926 Attributes received in IPv6 ND messages determines the multilink 2927 forwarding profile for selecting the neighbor's underlying 2928 interfaces. 2930 When the OMNI interface sends an original IP packet over a selected 2931 outbound underlying interface, the OAL employs encapsulation and 2932 fragmentation as discussed in Section 5, then performs *NET 2933 encapsulation as determined by the L2 address information received in 2934 Interface Attributes. The OAL also performs encapsulation when the 2935 nearest AR is located multiple hops away as discussed in 2936 Section 15.1. (Note that the OAL MAY employ packing when multiple 2937 original IP packets and/or control messages are available for 2938 forwarding to the same OAL destination.) 2940 OMNI interface multilink service designers MUST observe the BCP 2941 guidance in Section 15 [RFC3819] in terms of implications for 2942 reordering when original IP packets from the same flow may be spread 2943 across multiple underlying interfaces having diverse properties. 2945 14.1. Multiple OMNI Interfaces 2947 MNs may connect to multiple independent OMNI links concurrently in 2948 support of SBM. Each OMNI interface is distinguished by its Anycast 2949 ULA (e.g., [ULA]:0002::, [ULA]:1000::, [ULA]:7345::, etc.). The MN 2950 configures a separate OMNI interface for each link so that multiple 2951 interfaces (e.g., omni0, omni1, omni2, etc.) are exposed to the IPv6 2952 layer. A different Anycast ULA is assigned to each interface, and 2953 the MN injects the service prefixes for the OMNI link instances into 2954 the EUN routing system. 2956 Applications in EUNs can use Segment Routing to select the desired 2957 OMNI interface based on SBM considerations. The Anycast ULA is 2958 written into an original IP packet's IPv6 destination address, and 2959 the actual destination (along with any additional intermediate hops) 2960 is written into the Segment Routing Header. Standard IP routing 2961 directs the packet to the MN's mobile router entity, and the Anycast 2962 ULA identifies the OMNI interface to be used for transmission to the 2963 next hop. When the MN receives the packet, it replaces the IPv6 2964 destination address with the next hop found in the routing header and 2965 transmits the message over the OMNI interface identified by the 2966 Anycast ULA. 2968 Multiple distinct OMNI links can therefore be used to support fault 2969 tolerance, load balancing, reliability, etc. The architectural model 2970 is similar to Layer 2 Virtual Local Area Networks (VLANs). 2972 14.2. MN<->AR Traffic Loop Prevention 2974 After an AR has registered an MNP for a MN (see: Section 15), the AR 2975 will forward packets destined to an address within the MNP to the MN. 2976 The MN will under normal circumstances then forward the packet to the 2977 correct destination within its internal networks. 2979 If at some later time the MN loses state (e.g., after a reboot), it 2980 may begin returning packets destined to an MNP address to the AR as 2981 its default router. The AR therefore must drop any packets 2982 originating from the MN and destined to an address within the MN's 2983 registered MNP. To do so, the AR institutes the following check: 2985 o if the IP destination address belongs to a neighbor on the same 2986 OMNI interface, and if the link-layer source address is the same 2987 as one of the neighbor's link-layer addresses, drop the packet. 2989 15. Router Discovery and Prefix Registration 2991 MNs interface with the MS by sending RS messages with OMNI options 2992 under the assumption that one or more AR on the *NET will process the 2993 message and respond. The MN then configures default routes for the 2994 OMNI interface via the discovered ARs as the next hop. The manner in 2995 which the *NET ensures AR coordination is link-specific and outside 2996 the scope of this document (however, considerations for *NETs that do 2997 not provide ARs that recognize the OMNI option are discussed in 2998 Section 20). 3000 For each underlying interface, the MN sends an RS message with an 3001 OMNI option to coordinate with MSEs identified by MSID values. 3002 Example MSID discovery methods are given in [RFC5214] and include 3003 data link login parameters, name service lookups, static 3004 configuration, a static "hosts" file, etc. When the AR receives an 3005 RS', it selects a nearby MSE (which may be itself) and returns an RA 3006 with the selected MSID in an MS-Register sub-option. The AR selects 3007 only a single nearby MSE while also soliciting the MSEs corresponding 3008 to any non-zero MSIDs. 3010 MNs configure OMNI interfaces that observe the properties discussed 3011 in the previous section. The OMNI interface and its underlying 3012 interfaces are said to be in either the "UP" or "DOWN" state 3013 according to administrative actions in conjunction with the interface 3014 connectivity status. An OMNI interface transitions to UP or DOWN 3015 through administrative action and/or through state transitions of the 3016 underlying interfaces. When a first underlying interface transitions 3017 to UP, the OMNI interface also transitions to UP. When all 3018 underlying interfaces transition to DOWN, the OMNI interface also 3019 transitions to DOWN. 3021 When an OMNI interface transitions to UP, the MN sends RS messages to 3022 register its MNP and an initial set of underlying interfaces that are 3023 also UP. The MN sends additional RS messages to refresh lifetimes 3024 and to register/deregister underlying interfaces as they transition 3025 to UP or DOWN. The MN's OMNI interface sends initial RS messages 3026 over an UP underlying interface with its MNP-LLA as the source (or 3027 with the unspecified address (::) as the source if it does not yet 3028 have an MNP-LLA) and with destination set to link-scoped All-Routers 3029 multicast (ff02::2) [RFC4291]. The OMNI interface includes an OMNI 3030 option per Section 12 with a Preflen assertion, Interface Attributes 3031 appropriate for underlying interfaces, MS-Register/Release sub- 3032 options containing MSID values, Reassembly Limits, an authentication 3033 sub-option and with any other necessary OMNI sub-options (e.g., a 3034 Node Identification sub-option as an identity for the MN). The OMNI 3035 interface then sets the S/T-omIndex field to the index of the 3036 underlying interface over which the RS message is sent. 3038 The OMNI interface then sends the RS over the underlying interface 3039 using OAL encapsulation and fragmentation if necessary. If OAL 3040 encapsulation is used for RS messages sent over an INET interface, 3041 the entire RS message must appear within a single carrier packet so 3042 that it can be authenticated without requiring reassembly. The OMNI 3043 interface selects an unpredictable initial Identification value per 3044 Section 6.5, sets the OAL source address to the ULA corresponding to 3045 the RS source (Or a Temporary ULA if the RS source is the unspecified 3046 address (::)) and sets the OAL destination to site-scoped All-Routers 3047 multicast (ff05::2) then sends the message. 3049 ARs process IPv6 ND messages with OMNI options and act as an MSE 3050 themselves and/or as a proxy for other MSEs. ARs receive RS messages 3051 and create a neighbor cache entry for the MN, then prepare to act as 3052 an MSE themselves and/or coordinate with any MSEs named in the 3053 Register/Release lists in a manner outside the scope of this 3054 document. When an MSE processes the OMNI information, it first 3055 validates the prefix registration information then injects/withdraws 3056 the MNP in the routing/mapping system and caches/discards the new 3057 Preflen, MNP and Interface Attributes. The MSE then informs the AR 3058 of registration success/failure, and the AR returns an RA message to 3059 the MN with an OMNI option per Section 12. 3061 The AR's OMNI interface returns the RA message via the same 3062 underlying interface of the MN over which the RS was received, and 3063 with destination address set to the MNP-LLA (i.e., unicast), with 3064 source address set to its own LLA, and with an OMNI option with S/ 3065 T-omIndex set to the value included in the RS. The OMNI option also 3066 includes a Preflen confirmation, Interface Attributes, MS-Register/ 3067 Release and any other necessary OMNI sub-options (e.g., a Node 3068 Identification sub-option as an identity for the AR). The RA also 3069 includes any information for the link, including RA Cur Hop Limit, M 3070 and O flags, Router Lifetime, Reachable Time and Retrans Timer 3071 values, and includes any necessary options such as: 3073 o PIOs with (A; L=0) that include MSPs for the link [RFC8028]. 3075 o RIOs [RFC4191] with more-specific routes. 3077 o an MTU option that specifies the maximum acceptable packet size 3078 for this underlying interface. 3080 If the RS message arrived as an OAL atomic fragment, the AR prepares 3081 the RA using OAL encapsulation/fragmentation with the same 3082 Identification value that appeared in the RS message, with source set 3083 to the ULA corresponding to the RA source and with destination set to 3084 the ULA corresponding to the RA destination. The AR then sends the 3085 initial RA message to the MN and MAY later send additional periodic 3086 and/or event-driven unsolicited RA messages per [RFC4861]. In that 3087 case, the S/T-omIndex field in the OMNI option of the unsolicited RA 3088 message identifies the target underlying interface of the destination 3089 MN. 3091 The AR can combine the information from multiple MSEs into one or 3092 more "aggregate" RAs sent to the MN in order conserve *NET bandwidth. 3093 Each aggregate RA includes an OMNI option with MS-Register/Release 3094 sub-options with the MSEs represented by the aggregate. If an 3095 aggregate is sent, the RA message contents must consistently 3096 represent the combined information advertised by all represented 3097 MSEs. Note that since the AR uses its own ADM-LLA as the RA source 3098 address, the MN determines the addresses of the represented MSEs by 3099 examining the MS-Register/Release OMNI sub-options. 3101 When the MN receives the RA message, it creates an OMNI interface 3102 neighbor cache entry for each MSID that has confirmed MNP 3103 registration via the L2 address of this AR. If the MN connects to 3104 multiple *NETs, it records the additional L2 AR addresses in each 3105 MSID neighbor cache entry (i.e., as multilink neighbors). The MN 3106 then configures a default route via the MSE that returned the RA 3107 message, and assigns the Subnet Router Anycast address corresponding 3108 to the MNP (e.g., 2001:db8:1:2::) to the OMNI interface. The MN then 3109 manages its underlying interfaces according to their states as 3110 follows: 3112 o When an underlying interface transitions to UP, the MN sends an RS 3113 over the underlying interface with an OMNI option. The OMNI 3114 option contains at least one Interface Attribute sub-option with 3115 values specific to this underlying interface, and may contain 3116 additional Interface Attributes specific to other underlying 3117 interfaces. The option also includes any MS-Register/Release sub- 3118 options. 3120 o When an underlying interface transitions to DOWN, the MN sends an 3121 RS or unsolicited NA message over any UP underlying interface with 3122 an OMNI option containing an Interface Attribute sub-option for 3123 the DOWN underlying interface with Link set to '0'. The MN sends 3124 isolated unsolicited NAs when reliability is not thought to be a 3125 concern (e.g., if redundant transmissions are sent on multiple 3126 underlying interfaces), or may instead set the SYN flag in the 3127 OMNI header to trigger a reliable solicited NA reply. 3129 o When the Router Lifetime for a specific AR nears expiration, the 3130 MN sends an RS over the underlying interface to receive a fresh 3131 RA. If no RA is received, the MN can send RS messages to an 3132 alternate MSID in case the current MSID has failed. If no RS 3133 messages are received even after trying to contact alternate 3134 MSIDs, the MN marks the underlying interface as DOWN. 3136 o When a MN wishes to release from one or more current MSIDs, it 3137 sends an RS or unsolicited NA message over any UP underlying 3138 interfaces with an OMNI option with a Release MSID. Each MSID 3139 then withdraws the MNP from the routing/mapping system and informs 3140 the AR that the release was successful. 3142 o When all of a MNs underlying interfaces have transitioned to DOWN 3143 (or if the prefix registration lifetime expires), any associated 3144 MSEs withdraw the MNP the same as if they had received a message 3145 with a release indication. 3147 The MN is responsible for retrying each RS exchange up to 3148 MAX_RTR_SOLICITATIONS times separated by RTR_SOLICITATION_INTERVAL 3149 seconds until an RA is received. If no RA is received over an UP 3150 underlying interface (i.e., even after attempting to contact 3151 alternate MSEs), the MN declares this underlying interface as DOWN. 3153 The IPv6 layer sees the OMNI interface as an ordinary IPv6 interface. 3154 Therefore, when the IPv6 layer sends an RS message the OMNI interface 3155 returns an internally-generated RA message as though the message 3156 originated from an IPv6 router. The internally-generated RA message 3157 contains configuration information that is consistent with the 3158 information received from the RAs generated by the MS. Whether the 3159 OMNI interface IPv6 ND messaging process is initiated from the 3160 receipt of an RS message from the IPv6 layer is an implementation 3161 matter. Some implementations may elect to defer the IPv6 ND 3162 messaging process until an RS is received from the IPv6 layer, while 3163 others may elect to initiate the process proactively. Still other 3164 deployments may elect to administratively disable the ordinary RS/RA 3165 messaging used by the IPv6 layer over the OMNI interface, since they 3166 are not required to drive the internal RS/RA processing. (Note that 3167 this same logic applies to IPv4 implementations that employ ICMP- 3168 based Router Discovery per [RFC1256].) 3170 Note: The Router Lifetime value in RA messages indicates the time 3171 before which the MN must send another RS message over this underlying 3172 interface (e.g., 600 seconds), however that timescale may be 3173 significantly longer than the lifetime the MS has committed to retain 3174 the prefix registration (e.g., REACHABLETIME seconds). ARs are 3175 therefore responsible for keeping MS state alive on a shorter 3176 timescale than the MN is required to do on its own behalf. 3178 Note: On multicast-capable underlying interfaces, MNs should send 3179 periodic unsolicited multicast NA messages and ARs should send 3180 periodic unsolicited multicast RA messages as "beacons" that can be 3181 heard by other nodes on the link. If a node fails to receive a 3182 beacon after a timeout value specific to the link, it can initiate a 3183 unicast exchange to test reachability. 3185 Note: if an AR acting as a proxy forwards a MN's RS message to 3186 another node acting as an MSE using UDP/IP encapsulation, it must use 3187 a distinct UDP source port number for each MN. This allows the MSE 3188 to distinguish different MNs behind the same AR at the link-layer, 3189 whereas the link-layer addresses would otherwise be 3190 indistinguishable. 3192 Note: when an AR acting as an MSE returns an RA to an INET Client, it 3193 includes an OMNI option with an Interface Attributes sub-option with 3194 omIndex set to 0 and with SRT, FMT, LHS and L2ADDR information for 3195 its INET interface. This provides the Client with partition prefix 3196 context regarding the local OMNI link segment. 3198 Note: The above RS/RA exchanges observe the Identification window 3199 management procedures specified in Section 6.5. In the asymmetric 3200 case, a simple RS/RA exchange establishes only the MN's send window 3201 and AR/MSE's receive window such that an additional NS/NA exchange in 3202 the reverse direction would be required to establish the 3203 corresponding receive/send windows. In the symmetric case, the MN 3204 returns a solicited NA in response to the RA in order to establish 3205 send/receive windows on both sides in a three-message exchange. 3207 15.1. Router Discovery in IP Multihop and IPv4-Only Networks 3209 On some *NETs, a MN may be located multiple IP hops away from the 3210 nearest AR. Forwarding through IP multihop *NETs is conducted 3211 through the application of a routing protocol (e.g., a MANET/VANET 3212 routing protocol over omni-directional wireless interfaces, an inter- 3213 domain routing protocol in an enterprise network, etc.). These *NETs 3214 could be either IPv6-enabled or IPv4-only, while IPv4-only *NETs 3215 could be either multicast-capable or unicast-only (note that for 3216 IPv4-only *NETs the following procedures apply for both single-hop 3217 and multihop cases). 3219 A MN located potentially multiple *NET hops away from the nearest AR 3220 prepares an RS message with source address set to its MNP-LLA (or to 3221 the unspecified address (::) if it does not yet have an MNP-LLA), and 3222 with destination set to link-scoped All-Routers multicast the same as 3223 discussed above. The OMNI interface then employs OAL encapsulation 3224 and fragmentation, and sets the OAL source address to the ULA 3225 corresponding to the RS source (or to a Temporary ULA if the RS 3226 source was the unspecified address (::)) and sets the OAL destination 3227 to site-scoped All-Routers multicast (ff05::2). For IPv6-enabled 3228 *NETs, the MN then encapsulates the message in UDP/IPv6 headers with 3229 source address set to the underlying interface address (or to the ULA 3230 that would be used for OAL encapsulation if the underlying interface 3231 does not yet have an address) and sets the destination to either a 3232 unicast or anycast address of an AR. For IPv4-only *NETs, the MN 3233 instead encapsulates the RS message in UDP/IPv4 headers with source 3234 address set to the IPv4 address of the underlying interface and with 3235 destination address set to either the unicast IPv4 address of an AR 3236 [RFC5214] or an IPv4 anycast address reserved for OMNI. The MN then 3237 sends the encapsulated RS message via the *NET interface, where it 3238 will be forwarded by zero or more intermediate *NET hops. 3240 When an intermediate *NET hop that participates in the routing 3241 protocol receives the encapsulated RS, it forwards the message 3242 according to its routing tables (note that an intermediate node could 3243 be a fixed infrastructure element or another MN). This process 3244 repeats iteratively until the RS message is received by a penultimate 3245 *NET hop within single-hop communications range of an AR, which 3246 forwards the message to the AR. 3248 When the AR receives the message, it decapsulates the RS (while 3249 performing OAL reassembly, if necessary) and coordinates with the MS 3250 the same as for an ordinary link-local RS, since the network layer 3251 Hop Limit will not have been decremented by the multihop forwarding 3252 process. The AR then prepares an RA message with source address set 3253 to its own ADM-LLA and destination address set to the LLA of the 3254 original MN. The AR then performs OAL encapsulation and 3255 fragmentation, with OAL source set to its own ADM-ULA and destination 3256 set to the ULA corresponding to the RA source. The AR then 3257 encapsulates the message in UDP/IPv4 or UDP/IPv6 headers with source 3258 address set to its own address and with destination set to the 3259 encapsulation source of the RS. 3261 The AR then forwards the message to an *NET node within 3262 communications range, which forwards the message according to its 3263 routing tables to an intermediate node. The multihop forwarding 3264 process within the *NET continues repetitively until the message is 3265 delivered to the original MN, which decapsulates the message and 3266 performs autoconfiguration the same as if it had received the RA 3267 directly from the AR as an on-link neighbor. 3269 Note: An alternate approach to multihop forwarding via IPv6 3270 encapsulation would be for the MN and AR to statelessly translate the 3271 IPv6 LLAs into ULAs and forward the RS/RA messages without 3272 encapsulation. This would violate the [RFC4861] requirement that 3273 certain IPv6 ND messages must use link-local addresses and must not 3274 be accepted if received with Hop Limit less than 255. This document 3275 therefore mandates encapsulation since the overhead is nominal 3276 considering the infrequent nature and small size of IPv6 ND messages. 3277 Future documents may consider encapsulation avoidance through 3278 translation while updating [RFC4861]. 3280 Note: An alternate approach to multihop forwarding via IPv4 3281 encapsulation would be to employ IPv6/IPv4 protocol translation. 3282 However, for IPv6 ND messages the LLAs would be truncated due to 3283 translation and the OMNI Router and Prefix Discovery services would 3284 not be able to function. The use of IPv4 encapsulation is therefore 3285 indicated. 3287 Note: An IPv4 anycast address for OMNI in IPv4 networks could be part 3288 of a new IPv4 /24 prefix allocation, but this may be difficult to 3289 obtain given IPv4 address exhaustion. An alternative would be to re- 3290 purpose the prefix 192.88.99.0 which has been set aside from its 3291 former use by [RFC7526]. 3293 15.2. MS-Register and MS-Release List Processing 3295 OMNI links maintain a constant value "MAX_MSID" selected to provide 3296 MNs with an acceptable level of MSE redundancy while minimizing 3297 control message amplification. It is RECOMMENDED that MAX_MSID be 3298 set to the default value 5; if a different value is chosen, it should 3299 be set uniformly by all nodes on the OMNI link. 3301 When a MN sends an RS message with an OMNI option via an underlying 3302 interface to an AR, the MN must convey its knowledge of its 3303 currently-associated MSEs. Initially, the MN will have no associated 3304 MSEs and should therefore send its initial RS messages to the link- 3305 scoped All-Routers multicast address. The AR will then return an RA 3306 message with source address set to the ADM-LLA of the selected MSE 3307 (which may be the address of the AR itself). 3309 As the MN activates additional underlying interfaces, it can 3310 optionally include an MS-Register sub-option with MSIDs for MSEs 3311 discovered from previous RS/RA exchanges. The MN will thus 3312 eventually begin to learn and manage its currently active set of 3313 MSEs, and can register with new MSEs or release from former MSEs with 3314 each successive RS/RA exchange. As the MN's MSE constituency grows, 3315 it alone is responsible for including or omitting MSIDs in the MS- 3316 Register/Release lists it sends in RS messages. The inclusion or 3317 omission of MSIDs determines the MN's interface to the MS and defines 3318 the manner in which MSEs will respond. The only limiting factor is 3319 that the MN should include no more than MAX_MSID values in each list 3320 per each IPv6 ND message, and should avoid duplication of entries in 3321 each list unless it wants to increase likelihood of control message 3322 delivery. 3324 When an AR receives an RS message sent by a MN with an OMNI option, 3325 the option will contain zero or more MS-Register and MS-Release sub- 3326 options containing MSIDs. After processing the OMNI option, the AR 3327 will have a list of zero or more MS-Register MSIDs and a list of zero 3328 or more of MS-Release MSIDs. The AR then processes the lists as 3329 follows: 3331 o For each list, retain the first MAX_MSID values in the list and 3332 discard any additional MSIDs (i.e., even if there are duplicates 3333 within a list). 3335 o Next, for each MSID in the MS-Register list, remove all matching 3336 MSIDs from the MS-Release list. 3338 o Next, proceed as follows: 3340 * If the AR's own MSID appears in the MS-Register list, send an 3341 RA message directly back to the MN and send a proxy copy of the 3342 RS message to each additional MSID in the MS-Register list with 3343 the MS-Register/Release lists omitted. Then, send an 3344 unsolicited NA (uNA) message to each MSID in the MS-Release 3345 list with the MS-Register/Release lists omitted and with an 3346 OMNI option with S/T-omIndex set to 0. 3348 * Otherwise, send a proxy copy of the RS message to each 3349 additional MSID in the MS-Register list with the MS-Register 3350 list omitted. For the first MSID, include the original MS- 3351 Release list; for all other MSIDs, omit the MS-Release list. 3353 Each proxy copy of the RS message will include an OMNI option and OAL 3354 encapsulation header with the ADM-ULA of the AR as the source and the 3355 ADM-ULA of the Register MSE as the destination. When the Register 3356 MSE receives the proxy RS message, if the message includes an MS- 3357 Release list the MSE sends a uNA message to each additional MSID in 3358 the Release list with an OMNI option with S/T-omIndex set to 0. The 3359 Register MSE then sends an RA message back to the (Proxy) AR wrapped 3360 in an OAL encapsulation header with source and destination addresses 3361 reversed, and with RA destination set to the MNP-LLA of the MN. When 3362 the AR receives this RA message, it sends a proxy copy of the RA to 3363 the MN. 3365 Each uNA message (whether sent by the first-hop AR or by a Register 3366 MSE) will include an OMNI option and an OAL encapsulation header with 3367 the ADM-ULA of the Register MSE as the source and the ADM-ULA of the 3368 Release MSE as the destination. The uNA informs the Release MSE that 3369 its previous relationship with the MN has been released and that the 3370 source of the uNA message is now registered. The Release MSE must 3371 then note that the subject MN of the uNA message is now "departed", 3372 and forward any subsequent packets destined to the MN to the Register 3373 MSE. 3375 Note that it is not an error for the MS-Register/Release lists to 3376 include duplicate entries. If duplicates occur within a list, the AR 3377 will generate multiple proxy RS and/or uNA messages - one for each 3378 copy of the duplicate entries. 3380 15.3. DHCPv6-based Prefix Registration 3382 When a MN is not pre-provisioned with an MNP-LLA (or, when the MN 3383 requires additional MNP delegations), it requests the MSE to select 3384 MNPs on its behalf and set up the correct routing state within the 3385 MS. The DHCPv6 service [RFC8415] supports this requirement. 3387 When an MN needs to have the MSE select MNPs, it sends an RS message 3388 with source set to the unspecified address (::) if it has no 3389 MNP_LLAs. If the MN requires only a single MNP delegation, it can 3390 then include a Node Identification sub-option in the OMNI option and 3391 set Preflen to the length of the desired MNP. If the MN requires 3392 multiple MNP delegations and/or more complex DHCPv6 services, it 3393 instead includes a DHCPv6 Message sub-option containing a Client 3394 Identifier, one or more IA_PD options and a Rapid Commit option then 3395 sets the 'msg-type' field to "Solicit", and includes a 3 octet 3396 'transaction-id'. The MN then sets the RS destination to All-Routers 3397 multicast and sends the message using OAL encapsulation and 3398 fragmentation if necessary as discussed above. 3400 When the MSE receives the RS message, it performs OAL reassembly if 3401 necessary. Next, if the RS source is the unspecified address (::) 3402 and/or the OMNI option includes a DHCPv6 message sub-option, the MSE 3403 acts as a "Proxy DHCPv6 Client" in a message exchange with the 3404 locally-resident DHCPv6 server. If the RS did not contain a DHCPv6 3405 message sub-option, the MSE generates a DHCPv6 Solicit message on 3406 behalf of the MN using an IA_PD option with the prefix length set to 3407 the OMNI header Preflen value and with a Client Identifier formed 3408 from the OMNI option Node Identification sub-option; otherwise, the 3409 MSE uses the DHCPv6 Solicit message contained in the OMNI option. 3410 The MSE then sends the DHCPv6 message to the DHCPv6 Server, which 3411 delegates MNPs and returns a DHCPv6 Reply message with PD parameters. 3412 (If the MSE wishes to defer creation of MN state until the DHCPv6 3413 Reply is received, it can instead act as a Lightweight DHCPv6 Relay 3414 Agent per [RFC6221] by encapsulating the DHCPv6 message in a Relay- 3415 forward/reply exchange with Relay Message and Interface ID options. 3416 In the process, the MSE packs any state information needed to return 3417 an RA to the MN in the Relay-forward Interface ID option so that the 3418 information will be echoed back in the Relay-reply.) 3420 When the MSE receives the DHCPv6 Reply, it adds routes to the routing 3421 system and creates MNP-LLAs based on the delegated MNPs. The MSE 3422 then sends an RA back to the MN with the DHCPv6 Reply message 3423 included in an OMNI DHCPv6 message sub-option if and only if the RS 3424 message had included an explicit DHCPv6 Solicit. If the RS message 3425 source was the unspecified address (::), the MSE includes one of the 3426 (newly-created) MNP-LLAs as the RA destination address and sets the 3427 OMNI option Preflen accordingly; otherwise, the MSE includes the RS 3428 source address as the RA destination address. The MSE then sets the 3429 RA source address to its own ADM-LLA then performs OAL encapsulation 3430 and fragmentation and sends the RA to the MN. When the MN receives 3431 the RA, it reassembles and discards the OAL encapsulation, then 3432 creates a default route, assigns Subnet Router Anycast addresses and 3433 uses the RA destination address as its primary MNP-LLA. The MN will 3434 then use this primary MNP-LLA as the source address of any IPv6 ND 3435 messages it sends as long as it retains ownership of the MNP. 3437 Note: After a MN performs a DHCPv6-based prefix registration exchange 3438 with a first MSE, it would need to repeat the exchange with each 3439 additional MSE it registers with. In that case, the MN supplies the 3440 MNP delegation information received from the first MSE when it 3441 engages the additional MSEs. 3443 16. Secure Redirection 3445 If the *NET link model is multiple access, the AR is responsible for 3446 assuring that address duplication cannot corrupt the neighbor caches 3447 of other nodes on the link. When the MN sends an RS message on a 3448 multiple access *NET link, the AR verifies that the MN is authorized 3449 to use the address and returns an RA with a non-zero Router Lifetime 3450 only if the MN is authorized. 3452 After verifying MN authorization and returning an RA, the AR MAY 3453 return IPv6 ND Redirect messages to direct MNs located on the same 3454 *NET link to exchange packets directly without transiting the AR. In 3455 that case, the MNs can exchange packets according to their unicast L2 3456 addresses discovered from the Redirect message instead of using the 3457 dogleg path through the AR. In some *NET links, however, such direct 3458 communications may be undesirable and continued use of the dogleg 3459 path through the AR may provide better performance. In that case, 3460 the AR can refrain from sending Redirects, and/or MNs can ignore 3461 them. 3463 17. AR and MSE Resilience 3465 *NETs SHOULD deploy ARs in Virtual Router Redundancy Protocol (VRRP) 3466 [RFC5798] configurations so that service continuity is maintained 3467 even if one or more ARs fail. Using VRRP, the MN is unaware which of 3468 the (redundant) ARs is currently providing service, and any service 3469 discontinuity will be limited to the failover time supported by VRRP. 3470 Widely deployed public domain implementations of VRRP are available. 3472 MSEs SHOULD use high availability clustering services so that 3473 multiple redundant systems can provide coordinated response to 3474 failures. As with VRRP, widely deployed public domain 3475 implementations of high availability clustering services are 3476 available. Note that special-purpose and expensive dedicated 3477 hardware is not necessary, and public domain implementations can be 3478 used even between lightweight virtual machines in cloud deployments. 3480 18. Detecting and Responding to MSE Failures 3482 In environments where fast recovery from MSE failure is required, ARs 3483 SHOULD use proactive Neighbor Unreachability Detection (NUD) in a 3484 manner that parallels Bidirectional Forwarding Detection (BFD) 3485 [RFC5880] to track MSE reachability. ARs can then quickly detect and 3486 react to failures so that cached information is re-established 3487 through alternate paths. Proactive NUD control messaging is carried 3488 only over well-connected ground domain networks (i.e., and not low- 3489 end *NET links such as aeronautical radios) and can therefore be 3490 tuned for rapid response. 3492 ARs perform proactive NUD for MSEs for which there are currently 3493 active MNs on the *NET. If an MSE fails, ARs can quickly inform MNs 3494 of the outage by sending multicast RA messages on the *NET interface. 3495 The AR sends RA messages to MNs via the *NET interface with an OMNI 3496 option with a Release ID for the failed MSE, and with destination 3497 address set to All-Nodes multicast (ff02::1) [RFC4291]. 3499 The AR SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated 3500 by small delays [RFC4861]. Any MNs on the *NET interface that have 3501 been using the (now defunct) MSE will receive the RA messages and 3502 associate with a new MSE. 3504 19. Transition Considerations 3506 When a MN connects to an *NET link for the first time, it sends an RS 3507 message with an OMNI option. If the first hop AR recognizes the 3508 option, it returns an RA with its ADM-LLA as the source, the MNP-LLA 3509 as the destination and with an OMNI option included. The MN then 3510 engages the AR according to the OMNI link model specified above. If 3511 the first hop AR is a legacy IPv6 router, however, it instead returns 3512 an RA message with no OMNI option and with a non-OMNI unicast source 3513 LLA as specified in [RFC4861]. In that case, the MN engages the *NET 3514 according to the legacy IPv6 link model and without the OMNI 3515 extensions specified in this document. 3517 If the *NET link model is multiple access, there must be assurance 3518 that address duplication cannot corrupt the neighbor caches of other 3519 nodes on the link. When the MN sends an RS message on a multiple 3520 access *NET link with an LLA source address and an OMNI option, ARs 3521 that recognize the option ensure that the MN is authorized to use the 3522 address and return an RA with a non-zero Router Lifetime only if the 3523 MN is authorized. ARs that do not recognize the option instead 3524 return an RA that makes no statement about the MN's authorization to 3525 use the source address. In that case, the MN should perform 3526 Duplicate Address Detection to ensure that it does not interfere with 3527 other nodes on the link. 3529 An alternative approach for multiple access *NET links to ensure 3530 isolation for MN / AR communications is through L2 address mappings 3531 as discussed in Appendix C. This arrangement imparts a (virtual) 3532 point-to-point link model over the (physical) multiple access link. 3534 20. OMNI Interfaces on Open Internetworks 3536 OMNI interfaces configured over IPv6-enabled underlying interfaces on 3537 an open Internetwork without an OMNI-aware first-hop AR receive RA 3538 messages that do not include an OMNI option, while OMNI interfaces 3539 configured over IPv4-only underlying interfaces do not receive any 3540 (IPv6) RA messages at all (although they may receive IPv4 RA messages 3541 [RFC1256]). OMNI interfaces that receive RA messages without an OMNI 3542 option configure addresses, on-link prefixes, etc. on the underlying 3543 interface that received the RA according to standard IPv6 ND and 3544 address resolution conventions [RFC4861] [RFC4862]. OMNI interfaces 3545 configured over IPv4-only underlying interfaces configure IPv4 3546 address information on the underlying interfaces using mechanisms 3547 such as DHCPv4 [RFC2131]. 3549 OMNI interfaces configured over underlying interfaces that connect to 3550 an open Internetwork can apply security services such as VPNs to 3551 connect to an MSE, or can establish a direct link to an MSE through 3552 some other means (see Section 4). In environments where an explicit 3553 VPN or direct link may be impractical, OMNI interfaces can instead 3554 use UDP/IP encapsulation while including authentication signatures in 3555 IPv6 ND messages. 3557 OMNI interfaces use UDP service port number 8060 (see: Section 25.11 3558 and Section 3.6 of [I-D.templin-6man-aero]) according to the simple 3559 UDP/IP encapsulation format specified in [RFC4380] for both IPv4 and 3560 IPv6 underlying interfaces. OMNI interfaces do not include the UDP/ 3561 IP header/trailer extensions specified in [RFC4380][RFC6081], but may 3562 include them as OMNI sub-options instead when necessary. Since the 3563 OAL includes an integrity check over the OAL packet, OAL sources 3564 selectively disable UDP checksums for OAL packets that do not require 3565 UDP/IP address integrity, but enable UDP checksums for others 3566 including non-OAL packets, IPv6 ND messages used to establish link- 3567 layer addresses, etc. If the OAL source discovers that packets with 3568 UDP checksums disabled are being dropped in the path it should enable 3569 UDP checksums in future packets. Further considerations for UDP 3570 encapsulation checksums are found in [RFC6935][RFC6936]. 3572 For MN-to-MSE (e.g., "Vehicle-to-Infrastructure (V2I)") and MSE-to- 3573 MSE neighbor exchanges, the source must include an OMNI option with 3574 an authentication sub-option in all IPv6 ND messages. The source can 3575 apply HIP security services per [RFC7401] using the IPv6 ND message 3576 OMNI option as a "shipping container" to convey an authentication 3577 signature in a (unidirectional) HIP "Notify" message. For MN-to-MN 3578 (e.g., "Vehicle-to-Vehicle (V2V)") neighbor exchanges, two MNs can 3579 attain mutual authentication by exchanging HIP "Initiator/Responder" 3580 messages coded in OMNI options of multiple IPv6 NS/NA messages 3581 according to the HIP protocol. (Alternatively, a simple Hashed 3582 Message Authentication Code (HMAC) can be included in the manner 3583 specified in [RFC4380].) 3585 When HIP authentication is used, the IPv6 ND message source should 3586 include an OMNI option with a HIP message containing a valid 3587 authentication signature. When the source prepares the HIP message, 3588 it includes its own (H)HIT as the Sender's HIT and the neighbor's 3589 (H)HIT if known as the Receiver's HIT (otherwise 0). Before 3590 calculating the HIP signature, the source sets both the ICMPv6 3591 Checksum field and HIP signature fields to 0. The source then 3592 calculates the HIP authentication signature over the full length of 3593 the IPv6 ND message beginning with the ICMPv6 message header and 3594 extending over all included IPv6 ND message options including the 3595 OMNI option itself. The source next writes the authentication 3596 signature into the HIP signature field, then calculates the ICMPv6 3597 message checksum and writes the value into the ICMPv6 Checksum field. 3599 After establishing a VPN or preparing for UDP/IP encapsulation, OMNI 3600 interfaces send control plane messages to interface with the MSE, 3601 including RS/RA messages used according to Section 15 and NS/NA 3602 messages used for route optimization and mobility (see: 3603 [I-D.templin-6man-aero]). The control plane messages must be 3604 authenticated while data plane messages are delivered the same as for 3605 ordinary best-effort traffic with basic source address-based data 3606 origin verification. Data plane communications via OMNI interfaces 3607 that connect over open Internetworks without an explicit VPN should 3608 therefore employ transport- or higher-layer security to ensure 3609 integrity and/or confidentiality. 3611 OMNI interfaces configured over open Internetworks are often located 3612 behind NATs. The OMNI interface accommodates NAT traversal using 3613 UDP/IP encapsulation and the mechanisms discussed in 3614 [I-D.templin-6man-aero]. To support NAT determination, MSEs include 3615 an Origin Indication sub-option in RA messages sent in response to RS 3616 messages received from a Client via UDP/IP encapsulation. 3618 Note: Following the initial IPv6 ND message exchange, OMNI interfaces 3619 configured over open Internetworks maintain neighbor relationships by 3620 transmitting periodic IPv6 ND messages with OMNI options that include 3621 HIP "Update" and/or "Notify" messages. When HMAC authentication is 3622 used instead of HIP, the MN and MSE exchange all IPv6 ND messages 3623 with HMAC signatures included based on a shared-secret. 3625 Note: The [RFC4380] HMAC and/or HIP message [RFC7401] authentication 3626 sub-options appear in the OMNI option, which may occur anywhere 3627 within the IPv6 ND message body. When a node that inserts an 3628 authentication sub-option generates the authentication signature, it 3629 calculates the signature over the entire length of the IPv6 ND 3630 message but with the sub-option authentication field itself set to 0. 3631 The node then writes the resulting signature into the authentication 3632 field then continues to prepare the message for transmission. For 3633 this reason, if an IPv6 ND message includes multiple authentication 3634 sub-options, the first sub-option is consulted and any additional 3635 sub-options are ignored. 3637 21. Time-Varying MNPs 3639 In some use cases, it is desirable, beneficial and efficient for the 3640 MN to receive a constant MNP that travels with the MN wherever it 3641 moves. For example, this would allow air traffic controllers to 3642 easily track aircraft, etc. In other cases, however (e.g., 3643 intelligent transportation systems), the MN may be willing to 3644 sacrifice a modicum of efficiency in order to have time-varying MNPs 3645 that can be changed every so often to defeat adversarial tracking. 3647 The prefix delegation services discussed in Section 15.3 allows OMNI 3648 MNs that desire time-varying MNPs to obtain short-lived prefixes to 3649 send RS messages with source set to the unspecified address (::) and/ 3650 or with an OMNI option with DHCPv6 Option sub-options. The MN would 3651 then be obligated to renumber its internal networks whenever its MNP 3652 (and therefore also its OMNI address) changes. This should not 3653 present a challenge for MNs with automated network renumbering 3654 services, however presents limits for the durations of ongoing 3655 sessions that would prefer to use a constant address. 3657 22. (H)HITs and Temporary ULAs 3659 MNs that generate (H)HITs but do not have pre-assigned MNPs can 3660 request MNP delegations by issuing IPv6 ND messages that use the 3661 (H)HIT instead of a Temporary ULA. In particular, when a MN creates 3662 an RS message it can set the source to the unspecified address (::) 3663 and destination to All-Routers multicast. The IPv6 ND message 3664 includes an OMNI option with a HIP message sub-option, and need not 3665 include a Node Identification sub-option since the MN's HIT appears 3666 in the HIP message. The MN then encapsulates the message in an IPv6 3667 header with the (H)HIT as the source address and with destination set 3668 to either a unicast or anycast ADM-ULA. The MN then sends the 3669 message to the MSE as specified in Section 15.1. 3671 When the MSE receives the message, it notes that the RS source was 3672 the unspecified address (::), then examines the RS encapsulation 3673 source address to determine that the source is a (H)HIT and not a 3674 Temporary ULA. The MSE next invokes the DHCPv6 protocol to request 3675 an MNP prefix delegation while using the HIT as the Client 3676 Identifier, then prepares an RA message with source address set to 3677 its own ADM-LLA and destination set to the MNP-LLA corresponding to 3678 the delegated MNP. The MSE next includes an OMNI option with a HIP 3679 message sub-option and any DHCPv6 prefix delegation parameters. The 3680 MSE then finally encapsulates the RA in an IPv6 header with source 3681 address set to its own ADM-ULA and destination set to the (H)HIT from 3682 the RS encapsulation source address, then returns the encapsulated RA 3683 to the MN. 3685 MNs can also use (H)HITs and/or Temporary ULAs for direct MN-to-MN 3686 communications outside the context of any OMNI link supporting 3687 infrastructure. When two MNs encounter one another they can use 3688 their (H)HITs and/or Temporary ULAs as original IPv6 packet source 3689 and destination addresses to support direct communications. MNs can 3690 also inject their (H)HITs and/or Temporary ULAs into a MANET/VANET 3691 routing protocol to enable multihop communications. MNs can further 3692 exchange IPv6 ND messages (such as NS/NA) using their (H)HITs and/or 3693 Temporary ULAs as source and destination addresses. Note that the 3694 HIP security protocols for establishing secure neighbor relationships 3695 are based on (H)HITs. IPv6 ND messages that use Temporary ULAs 3696 instead use the HMAC authentication service specified in [RFC4380]. 3698 Lastly, when MNs are within the coverage range of OMNI link 3699 infrastructure a case could be made for injecting (H)HITs and/or 3700 Temporary ULAs into the global MS routing system. For example, when 3701 the MN sends an RS to a MSE it could include a request to inject the 3702 (H)HIT / Temporary ULA into the routing system instead of requesting 3703 an MNP prefix delegation. This would potentially enable OMNI link- 3704 wide communications using only (H)HITs or Temporary ULAs, and not 3705 MNPs. This document notes the opportunity, but makes no 3706 recommendation. 3708 23. Address Selection 3710 OMNI MNs use LLAs only for link-scoped communications on the OMNI 3711 link. Typically, MNs use LLAs as source/destination IPv6 addresses 3712 of IPv6 ND messages, but may also use them for addressing ordinary 3713 original IP packets exchanged with an OMNI link neighbor. 3715 OMNI MNs use MNP-ULAs as source/destination IPv6 addresses in the 3716 encapsulation headers of OAL packets. OMNI MNs use Temporary ULAs 3717 for OAL addressing when an MNP-ULA is not available, or as source/ 3718 destination IPv6 addresses for communications within a MANET/VANET 3719 local area. OMNI MNs use HITs instead of Temporary ULAs when 3720 operation outside the context of a specific ULA domain and/or source 3721 address attestation is necessary. 3723 OMNI MNs use MNP-based GUAs as original IP packet source and 3724 destination addresses for communications with Internet destinations 3725 when they are within range of OMNI link supporting infrastructure 3726 that can inject the MNP into the routing system. 3728 24. Error Messages 3730 An OAL destination or intermediate node may need to return ICMPv6 3731 error messages (e.g., Destination Unreachable, Packet Too Big, Time 3732 Exceeded, etc.) [RFC4443] to an OAL source. Since ICMPv6 error 3733 messages do not themselves include authentication codes, the OAL 3734 includes the ICMPv6 error message as an OMNI sub-option in an IPv6 ND 3735 uNA message. The OAL also includes a HIP message sub-option if the 3736 uNA needs to travel over an open Internetwork. 3738 25. IANA Considerations 3740 The following IANA actions are requested in accordance with [RFC8126] 3741 and [RFC8726]: 3743 25.1. "IEEE 802 Numbers" Registry 3745 The IANA is instructed to allocate an official Ethertype number TBD1 3746 from the 'ieee-802-numbers' registry for User Datagram Protocol (UDP) 3747 encapsulation on Ethernet networks. Guidance is found in [RFC7042] 3748 (registration procedure is Expert Review). 3750 25.2. "IPv6 Neighbor Discovery Option Formats" Registry 3752 The IANA is instructed to allocate an official Type number TBD2 from 3753 the "IPv6 Neighbor Discovery Option Formats" registry for the OMNI 3754 option (registration procedure is RFC required). Implementations set 3755 Type to 253 as an interim value [RFC4727]. 3757 25.3. "Ethernet Numbers" Registry 3759 The IANA is instructed to allocate one Ethernet unicast address TBD3 3760 (suggested value '00-52-14') in the 'ethernet-numbers' registry under 3761 "IANA Unicast 48-bit MAC Addresses" (registration procedure is Expert 3762 Review). The registration should appear as follows: 3764 Addresses Usage Reference 3765 --------- ----- --------- 3766 00-52-14 Overlay Multilink Network (OMNI) Interface [RFCXXXX] 3768 Figure 30: IANA Unicast 48-bit MAC Addresses 3770 25.4. "ICMPv6 Code Fields: Type 2 - Packet Too Big" Registry 3772 The IANA is instructed to assign two new Code values in the "ICMPv6 3773 Code Fields: Type 2 - Packet Too Big" registry (registration 3774 procedure is Standards Action or IESG Approval). The registry should 3775 appear as follows: 3777 Code Name Reference 3778 --- ---- --------- 3779 0 PTB Hard Error [RFC4443] 3780 1 PTB Soft Error (loss) [RFCXXXX] 3781 2 PTB Soft Error (no loss) [RFCXXXX] 3783 Figure 31: ICMPv6 Code Fields: Type 2 - Packet Too Big Values 3785 (Note: this registry also to be used to define values for setting the 3786 "unused" field of ICMPv4 "Destination Unreachable - Fragmentation 3787 Needed" messages.) 3789 25.5. "OMNI Option Sub-Type Values" (New Registry) 3791 The OMNI option defines a 5-bit Sub-Type field, for which IANA is 3792 instructed to create and maintain a new registry entitled "OMNI 3793 Option Sub-Type Values". Initial values are given below 3794 (registration procedure is RFC required): 3796 Value Sub-Type name Reference 3797 ----- ------------- ---------- 3798 0 Pad1 [RFCXXXX] 3799 1 PadN [RFCXXXX] 3800 2 Interface Attributes (Type 1) [RFCXXXX] 3801 3 Interface Attributes (Type 2) [RFCXXXX] 3802 4 Interface Attributes (Type 4) [RFCXXXX] 3803 5 MS-Register [RFCXXXX] 3804 6 MS-Release [RFCXXXX] 3805 7 Geo Coordinates [RFCXXXX] 3806 8 DHCPv6 Message [RFCXXXX] 3807 9 HIP Message [RFCXXXX] 3808 11 PIM-SM Message [RFCXXXX] 3809 11 Reassembly Limit [RFCXXXX] 3810 12 Fragmentation Report [RFCXXXX] 3811 13 Node Identification [RFCXXXX] 3812 14-29 Unassigned 3813 30 Sub-Type Extension [RFCXXXX] 3814 31 Reserved by IANA [RFCXXXX] 3816 Figure 32: OMNI Option Sub-Type Values 3818 25.6. "OMNI Geo Coordinates Type Values" (New Registry) 3820 The OMNI Geo Coordinates Sub-Option (see: Section 12.1.7) contains an 3821 8-bit Type field, for which IANA is instructed to create and maintain 3822 a new registry entitled "OMNI Geo Coordinates Type Values". Initial 3823 values are given below (registration procedure is RFC required): 3825 Value Sub-Type name Reference 3826 ----- ------------- ---------- 3827 0 NULL [RFCXXXX] 3828 255 Reserved by IANA [RFCXXXX] 3830 Figure 33: OMNI Geo Coordinates Type 3832 25.7. "OMNI Node Identification ID-Type Values" (New Registry) 3834 The OMNI Node Identification Sub-Option (see: Section 12.1.13) 3835 contains an 8-bit ID-Type field, for which IANA is instructed to 3836 create and maintain a new registry entitled "OMNI Node Identification 3837 ID-Type Values". Initial values are given below (registration 3838 procedure is RFC required): 3840 Value Sub-Type name Reference 3841 ----- ------------- ---------- 3842 0 UUID [RFCXXXX] 3843 1 HIT [RFCXXXX] 3844 2 HHIT [RFCXXXX] 3845 3 Network Access Identifier [RFCXXXX] 3846 4 FQDN [RFCXXXX] 3847 5-252 Unassigned [RFCXXXX] 3848 253-254 Reserved for Experimentation [RFCXXXX] 3849 255 Reserved by IANA [RFCXXXX] 3851 Figure 34: OMNI Node Identification ID-Type Values 3853 25.8. "OMNI Option Sub-Type Extension Values" (New Registry) 3855 The OMNI option defines an 8-bit Extension-Type field for Sub-Type 30 3856 (Sub-Type Extension), for which IANA is instructed to create and 3857 maintain a new registry entitled "OMNI Option Sub-Type Extension 3858 Values". Initial values are given below (registration procedure is 3859 RFC required): 3861 Value Sub-Type name Reference 3862 ----- ------------- ---------- 3863 0 RFC4380 UDP/IP Header Option [RFCXXXX] 3864 1 RFC6081 UDP/IP Trailer Option [RFCXXXX] 3865 2-252 Unassigned 3866 253-254 Reserved for Experimentation [RFCXXXX] 3867 255 Reserved by IANA [RFCXXXX] 3869 Figure 35: OMNI Option Sub-Type Extension Values 3871 25.9. "OMNI RFC4380 UDP/IP Header Option" (New Registry) 3873 The OMNI Sub-Type Extension "RFC4380 UDP/IP Header Option" defines an 3874 8-bit Header Type field, for which IANA is instructed to create and 3875 maintain a new registry entitled "OMNI RFC4380 UDP/IP Header Option". 3876 Initial registry values are given below (registration procedure is 3877 RFC required): 3879 Value Sub-Type name Reference 3880 ----- ------------- ---------- 3881 0 Origin Indication (IPv4) [RFC4380] 3882 1 Authentication Encapsulation [RFC4380] 3883 2 Origin Indication (IPv6) [RFCXXXX] 3884 3-252 Unassigned 3885 253-254 Reserved for Experimentation [RFCXXXX] 3886 255 Reserved by IANA [RFCXXXX] 3888 Figure 36: OMNI RFC4380 UDP/IP Header Option 3890 25.10. "OMNI RFC6081 UDP/IP Trailer Option" (New Registry) 3892 The OMNI Sub-Type Extension for "RFC6081 UDP/IP Trailer Option" 3893 defines an 8-bit Trailer Type field, for which IANA is instructed to 3894 create and maintain a new registry entitled "OMNI RFC6081 UDP/IP 3895 Trailer Option". Initial registry values are given below 3896 (registration procedure is RFC required): 3898 Value Sub-Type name Reference 3899 ----- ------------- ---------- 3900 0 Unassigned 3901 1 Nonce [RFC6081] 3902 2 Unassigned 3903 3 Alternate Address (IPv4) [RFC6081] 3904 4 Neighbor Discovery Option [RFC6081] 3905 5 Random Port [RFC6081] 3906 6 Alternate Address (IPv6) [RFCXXXX] 3907 7-252 Unassigned 3908 253-254 Reserved for Experimentation [RFCXXXX] 3909 255 Reserved by IANA [RFCXXXX] 3911 Figure 37: OMNI RFC6081 Trailer Option 3913 25.11. Additional Considerations 3915 The IANA has assigned the UDP port number "8060" for an earlier 3916 experimental version of AERO [RFC6706]. This document together with 3917 [I-D.templin-6man-aero] reclaims the UDP port number "8060" for 3918 'aero' as the service port for UDP/IP encapsulation. (Note that, 3919 although [RFC6706] was not widely implemented or deployed, any 3920 messages coded to that specification can be easily distinguished and 3921 ignored since they use an invalid ICMPv6 message type number '0'.) 3922 The IANA is therefore instructed to update the reference for UDP port 3923 number "8060" from "RFC6706" to "RFCXXXX" (i.e., this document). 3925 The IANA has assigned a 4 octet Private Enterprise Number (PEN) code 3926 "45282" in the "enterprise-numbers" registry. This document is the 3927 normative reference for using this code in DHCP Unique IDentifiers 3928 based on Enterprise Numbers ("DUID-EN for OMNI Interfaces") (see: 3929 Section 11). The IANA is therefore instructed to change the 3930 enterprise designation for PEN code "45282" from "LinkUp Networks" to 3931 "Overlay Multilink Network Interface (OMNI)". 3933 The IANA has assigned the ifType code "301 - omni - Overlay Multilink 3934 Network Interface (OMNI)" in accordance with Section 6 of [RFC8892]. 3935 The registration appears under the IANA "Structure of Management 3936 Information (SMI) Numbers (MIB Module Registrations) - Interface 3937 Types (ifType)" registry. 3939 No further IANA actions are required. 3941 26. Security Considerations 3943 Security considerations for IPv4 [RFC0791], IPv6 [RFC8200] and IPv6 3944 Neighbor Discovery [RFC4861] apply. OMNI interface IPv6 ND messages 3945 SHOULD include Nonce and Timestamp options [RFC3971] when transaction 3946 confirmation and/or time synchronization is needed. (Note however 3947 that when OAL encapsulation is used the (echoed) OAL Identification 3948 value can provide sufficient transaction confirmation.) 3950 MN OMNI interfaces configured over secured ANET interfaces inherit 3951 the physical and/or link-layer security properties (i.e., "protected 3952 spectrum") of the connected ANETs. MN OMNI interfaces configured 3953 over open INET interfaces can use symmetric securing services such as 3954 VPNs or can by some other means establish a direct link. When a VPN 3955 or direct link may be impractical, however, the security services 3956 specified in [RFC7401] and/or [RFC4380] can be employed. While the 3957 OMNI link protects control plane messaging, applications must still 3958 employ end-to-end transport- or higher-layer security services to 3959 protect the data plane. 3961 Strong network layer security for control plane messages and 3962 forwarding path integrity for data plane messages between MSEs MUST 3963 be supported. In one example, the AERO service 3964 [I-D.templin-6man-aero] constructs a spanning tree between MSEs and 3965 secures the links in the spanning tree with network layer security 3966 mechanisms such as IPsec [RFC4301] or WireGuard. Control plane 3967 messages are then constrained to travel only over the secured 3968 spanning tree paths and are therefore protected from attack or 3969 eavesdropping. Since data plane messages can travel over route 3970 optimized paths that do not strictly follow the spanning tree, 3971 however, end-to-end transport- or higher-layer security services are 3972 still required. Additionally, the OAL Identification value provides 3973 a first level of data origin authentication that mitigates off-path 3974 spoofing. 3976 Identity-based key verification infrastructure services such as iPSK 3977 may be necessary for verifying the identities claimed by MNs. This 3978 requirement should be harmonized with the manner in which (H)HITs are 3979 attested in a given operational environment. 3981 Security considerations for specific access network interface types 3982 are covered under the corresponding IP-over-(foo) specification 3983 (e.g., [RFC2464], [RFC2492], etc.). 3985 Security considerations for IPv6 fragmentation and reassembly are 3986 discussed in Section 6.9. Most importantly, each OAL destination 3987 MUST employ a firewall. 3989 27. Implementation Status 3991 AERO/OMNI Release-3.2 was tagged on March 30, 2021, and is undergoing 3992 internal testing. Additional internal releases expected within the 3993 coming months, with first public release expected end of 1H2021. 3995 28. Document Updates 3997 This document does not itself update other RFCs, but suggests that 3998 the following could be updated through future IETF initiatives: 4000 o [RFC1191] 4002 o [RFC4443] 4004 o [RFC8201] 4006 o [RFC7526] 4008 Updates can be through, e.g., standards action, the errata process, 4009 etc. as appropriate. 4011 29. Acknowledgements 4013 The first version of this document was prepared per the consensus 4014 decision at the 7th Conference of the International Civil Aviation 4015 Organization (ICAO) Working Group-I Mobility Subgroup on March 22, 4016 2019. Consensus to take the document forward to the IETF was reached 4017 at the 9th Conference of the Mobility Subgroup on November 22, 2019. 4018 Attendees and contributors included: Guray Acar, Danny Bharj, 4019 Francois D'Humieres, Pavel Drasil, Nikos Fistas, Giovanni Garofolo, 4020 Bernhard Haindl, Vaughn Maiolla, Tom McParland, Victor Moreno, Madhu 4021 Niraula, Brent Phillips, Liviu Popescu, Jacky Pouzet, Aloke Roy, Greg 4022 Saccone, Robert Segers, Michal Skorepa, Michel Solery, Stephane 4023 Tamalet, Fred Templin, Jean-Marc Vacher, Bela Varkonyi, Tony Whyman, 4024 Fryderyk Wrobel and Dongsong Zeng. 4026 The following individuals are acknowledged for their useful comments: 4027 Stuart Card, Michael Matyas, Robert Moskowitz, Madhu Niraula, Greg 4028 Saccone, Stephane Tamalet, Eric Vyncke. Pavel Drasil, Zdenek Jaron 4029 and Michal Skorepa are especially recognized for their many helpful 4030 ideas and suggestions. Madhuri Madhava Badgandi, Sean Dickson, Don 4031 Dillenburg, Joe Dudkowski, Vijayasarathy Rajagopalan, Ron Sackman and 4032 Katherine Tran are acknowledged for their hard work on the 4033 implementation and technical insights that led to improvements for 4034 the spec. 4036 Discussions on the IETF 6man and atn mailing lists during the fall of 4037 2020 suggested additional points to consider. The authors gratefully 4038 acknowledge the list members who contributed valuable insights 4039 through those discussions. Eric Vyncke and Erik Kline were the 4040 intarea ADs, while Bob Hinden and Ole Troan were the 6man WG chairs 4041 at the time the document was developed; they are all gratefully 4042 acknowledged for their many helpful insights. Many of the ideas in 4043 this document have further built on IETF experiences beginning as 4044 early as Y2K, with insights from colleagues including Brian 4045 Carpenter, Ralph Droms, Christian Huitema, Thomas Narten, Dave 4046 Thaler, Joe Touch, and many others who deserve recognition. 4048 Early observations on IP fragmentation performance implications were 4049 noted in the 1986 Digital Equipment Corporation (DEC) "qe reset" 4050 investigation, where fragment bursts from NFS UDP traffic triggered 4051 hardware resets resulting in communication failures. Jeff Chase, 4052 Fred Glover and Chet Juzsczak of the Ultrix Engineering Group led the 4053 investigation, and determined that setting a smaller NFS mount block 4054 size reduced the amount of fragmentation and suppressed the resets. 4055 Early observations on L2 media MTU issues were noted in the 1988 DEC 4056 FDDI investigation, where Raj Jain, KK Ramakrishnan and Kathy Wilde 4057 represented architectural considerations for FDDI networking in 4058 general including FDDI/Ethernet bridging. Jeff Mogul (who led the 4059 IETF Path MTU Discovery working group) and other DEC colleagues who 4060 supported these early investigations are also acknowledged. 4062 Throughout the 1990's and into the 2000's, many colleagues supported 4063 and encouraged continuation of the work. Beginning with the DEC 4064 Project Sequoia effort at the University of California, Berkeley, 4065 then moving to the DEC research lab offices in Palo Alto CA, then to 4066 the NASA Ames Research Center, then to SRI in Menlo Park, CA, then to 4067 Nokia in Mountain View, CA and finally to the Boeing Company in 2005 4068 the work saw continuous advancement through the encouragement of 4069 many. Those who offered their support and encouragement are 4070 gratefully acknowledged. 4072 This work is aligned with the NASA Safe Autonomous Systems Operation 4073 (SASO) program under NASA contract number NNA16BD84C. 4075 This work is aligned with the FAA as per the SE2025 contract number 4076 DTFAWA-15-D-00030. 4078 This work is aligned with the Boeing Information Technology (BIT) 4079 Mobility Vision Lab (MVL) program. 4081 30. References 4083 30.1. Normative References 4085 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 4086 DOI 10.17487/RFC0791, September 1981, 4087 . 4089 [RFC0793] Postel, J., "Transmission Control Protocol", STD 7, 4090 RFC 793, DOI 10.17487/RFC0793, September 1981, 4091 . 4093 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 4094 Requirement Levels", BCP 14, RFC 2119, 4095 DOI 10.17487/RFC2119, March 1997, 4096 . 4098 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 4099 "Definition of the Differentiated Services Field (DS 4100 Field) in the IPv4 and IPv6 Headers", RFC 2474, 4101 DOI 10.17487/RFC2474, December 1998, 4102 . 4104 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 4105 "SEcure Neighbor Discovery (SEND)", RFC 3971, 4106 DOI 10.17487/RFC3971, March 2005, 4107 . 4109 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 4110 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 4111 November 2005, . 4113 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 4114 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 4115 . 4117 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 4118 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 4119 2006, . 4121 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 4122 Control Message Protocol (ICMPv6) for the Internet 4123 Protocol Version 6 (IPv6) Specification", STD 89, 4124 RFC 4443, DOI 10.17487/RFC4443, March 2006, 4125 . 4127 [RFC4727] Fenner, B., "Experimental Values In IPv4, IPv6, ICMPv4, 4128 ICMPv6, UDP, and TCP Headers", RFC 4727, 4129 DOI 10.17487/RFC4727, November 2006, 4130 . 4132 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 4133 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 4134 DOI 10.17487/RFC4861, September 2007, 4135 . 4137 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 4138 Address Autoconfiguration", RFC 4862, 4139 DOI 10.17487/RFC4862, September 2007, 4140 . 4142 [RFC6088] Tsirtsis, G., Giarreta, G., Soliman, H., and N. Montavont, 4143 "Traffic Selectors for Flow Bindings", RFC 6088, 4144 DOI 10.17487/RFC6088, January 2011, 4145 . 4147 [RFC8028] Baker, F. and B. Carpenter, "First-Hop Router Selection by 4148 Hosts in a Multi-Prefix Network", RFC 8028, 4149 DOI 10.17487/RFC8028, November 2016, 4150 . 4152 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 4153 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 4154 May 2017, . 4156 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 4157 (IPv6) Specification", STD 86, RFC 8200, 4158 DOI 10.17487/RFC8200, July 2017, 4159 . 4161 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 4162 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 4163 DOI 10.17487/RFC8201, July 2017, 4164 . 4166 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 4167 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 4168 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 4169 RFC 8415, DOI 10.17487/RFC8415, November 2018, 4170 . 4172 30.2. Informative References 4174 [ATN] Maiolla, V., "The OMNI Interface - An IPv6 Air/Ground 4175 Interface for Civil Aviation, IETF Liaison Statement 4176 #1676, https://datatracker.ietf.org/liaison/1676/", March 4177 2020. 4179 [ATN-IPS] WG-I, ICAO., "ICAO Document 9896 (Manual on the 4180 Aeronautical Telecommunication Network (ATN) using 4181 Internet Protocol Suite (IPS) Standards and Protocol), 4182 Draft Edition 3 (work-in-progress)", December 2020. 4184 [CKSUM] Stone, J., Greenwald, M., Partridge, C., and J. Hughes, 4185 "Performance of Checksums and CRC's Over Real Data, IEEE/ 4186 ACM Transactions on Networking, Vol. 6, No. 5", October 4187 1998. 4189 [CRC] Jain, R., "Error Characteristics of Fiber Distributed Data 4190 Interface (FDDI), IEEE Transactions on Communications", 4191 August 1990. 4193 [I-D.ietf-drip-rid] 4194 Moskowitz, R., Card, S. W., Wiethuechter, A., and A. 4195 Gurtov, "UAS Remote ID", draft-ietf-drip-rid-07 (work in 4196 progress), January 2021. 4198 [I-D.ietf-intarea-tunnels] 4199 Touch, J. and M. Townsley, "IP Tunnels in the Internet 4200 Architecture", draft-ietf-intarea-tunnels-10 (work in 4201 progress), September 2019. 4203 [I-D.ietf-ipwave-vehicular-networking] 4204 (editor), J. (. J., "IPv6 Wireless Access in Vehicular 4205 Environments (IPWAVE): Problem Statement and Use Cases", 4206 draft-ietf-ipwave-vehicular-networking-20 (work in 4207 progress), March 2021. 4209 [I-D.ietf-tsvwg-udp-options] 4210 Touch, J., "Transport Options for UDP", draft-ietf-tsvwg- 4211 udp-options-12 (work in progress), May 2021. 4213 [I-D.templin-6man-aero] 4214 Templin, F. L., "Automatic Extended Route Optimization 4215 (AERO)", draft-templin-6man-aero-01 (work in progress), 4216 April 2021. 4218 [I-D.templin-6man-dhcpv6-ndopt] 4219 Templin, F. L., "A Unified Stateful/Stateless 4220 Configuration Service for IPv6", draft-templin-6man- 4221 dhcpv6-ndopt-11 (work in progress), January 2021. 4223 [I-D.templin-6man-lla-type] 4224 Templin, F. L., "The IPv6 Link-Local Address Type Field", 4225 draft-templin-6man-lla-type-02 (work in progress), 4226 November 2020. 4228 [I-D.templin-6man-omni-interface] 4229 Templin, F. L. and T. Whyman, "Transmission of IP Packets 4230 over Overlay Multilink Network (OMNI) Interfaces", draft- 4231 templin-6man-omni-interface-99 (work in progress), March 4232 2021. 4234 [IPV4-GUA] 4235 Postel, J., "IPv4 Address Space Registry, 4236 https://www.iana.org/assignments/ipv4-address-space/ipv4- 4237 address-space.xhtml", December 2020. 4239 [IPV6-GUA] 4240 Postel, J., "IPv6 Global Unicast Address Assignments, 4241 https://www.iana.org/assignments/ipv6-unicast-address- 4242 assignments/ipv6-unicast-address-assignments.xhtml", 4243 December 2020. 4245 [RFC1035] Mockapetris, P., "Domain names - implementation and 4246 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 4247 November 1987, . 4249 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 4250 Communication Layers", STD 3, RFC 1122, 4251 DOI 10.17487/RFC1122, October 1989, 4252 . 4254 [RFC1146] Zweig, J. and C. Partridge, "TCP alternate checksum 4255 options", RFC 1146, DOI 10.17487/RFC1146, March 1990, 4256 . 4258 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 4259 DOI 10.17487/RFC1191, November 1990, 4260 . 4262 [RFC1256] Deering, S., Ed., "ICMP Router Discovery Messages", 4263 RFC 1256, DOI 10.17487/RFC1256, September 1991, 4264 . 4266 [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", 4267 RFC 2131, DOI 10.17487/RFC2131, March 1997, 4268 . 4270 [RFC2225] Laubach, M. and J. Halpern, "Classical IP and ARP over 4271 ATM", RFC 2225, DOI 10.17487/RFC2225, April 1998, 4272 . 4274 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 4275 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 4276 . 4278 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 4279 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 4280 December 1998, . 4282 [RFC2492] Armitage, G., Schulter, P., and M. Jork, "IPv6 over ATM 4283 Networks", RFC 2492, DOI 10.17487/RFC2492, January 1999, 4284 . 4286 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 4287 Domains without Explicit Tunnels", RFC 2529, 4288 DOI 10.17487/RFC2529, March 1999, 4289 . 4291 [RFC2863] McCloghrie, K. and F. Kastenholz, "The Interfaces Group 4292 MIB", RFC 2863, DOI 10.17487/RFC2863, June 2000, 4293 . 4295 [RFC2923] Lahey, K., "TCP Problems with Path MTU Discovery", 4296 RFC 2923, DOI 10.17487/RFC2923, September 2000, 4297 . 4299 [RFC2983] Black, D., "Differentiated Services and Tunnels", 4300 RFC 2983, DOI 10.17487/RFC2983, October 2000, 4301 . 4303 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 4304 of Explicit Congestion Notification (ECN) to IP", 4305 RFC 3168, DOI 10.17487/RFC3168, September 2001, 4306 . 4308 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 4309 DOI 10.17487/RFC3330, September 2002, 4310 . 4312 [RFC3692] Narten, T., "Assigning Experimental and Testing Numbers 4313 Considered Useful", BCP 82, RFC 3692, 4314 DOI 10.17487/RFC3692, January 2004, 4315 . 4317 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 4318 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 4319 DOI 10.17487/RFC3810, June 2004, 4320 . 4322 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 4323 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 4324 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 4325 RFC 3819, DOI 10.17487/RFC3819, July 2004, 4326 . 4328 [RFC3879] Huitema, C. and B. Carpenter, "Deprecating Site Local 4329 Addresses", RFC 3879, DOI 10.17487/RFC3879, September 4330 2004, . 4332 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 4333 Unique IDentifier (UUID) URN Namespace", RFC 4122, 4334 DOI 10.17487/RFC4122, July 2005, 4335 . 4337 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 4338 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 4339 DOI 10.17487/RFC4271, January 2006, 4340 . 4342 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 4343 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 4344 December 2005, . 4346 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 4347 Network Address Translations (NATs)", RFC 4380, 4348 DOI 10.17487/RFC4380, February 2006, 4349 . 4351 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 4352 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 4353 2006, . 4355 [RFC4429] Moore, N., "Optimistic Duplicate Address Detection (DAD) 4356 for IPv6", RFC 4429, DOI 10.17487/RFC4429, April 2006, 4357 . 4359 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 4360 "Considerations for Internet Group Management Protocol 4361 (IGMP) and Multicast Listener Discovery (MLD) Snooping 4362 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 4363 . 4365 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 4366 "Internet Group Management Protocol (IGMP) / Multicast 4367 Listener Discovery (MLD)-Based Multicast Forwarding 4368 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 4369 August 2006, . 4371 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 4372 Discovery", RFC 4821, DOI 10.17487/RFC4821, March 2007, 4373 . 4375 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 4376 Errors at High Data Rates", RFC 4963, 4377 DOI 10.17487/RFC4963, July 2007, 4378 . 4380 [RFC5175] Haberman, B., Ed. and R. Hinden, "IPv6 Router 4381 Advertisement Flags Option", RFC 5175, 4382 DOI 10.17487/RFC5175, March 2008, 4383 . 4385 [RFC5213] Gundavelli, S., Ed., Leung, K., Devarapalli, V., 4386 Chowdhury, K., and B. Patil, "Proxy Mobile IPv6", 4387 RFC 5213, DOI 10.17487/RFC5213, August 2008, 4388 . 4390 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 4391 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 4392 DOI 10.17487/RFC5214, March 2008, 4393 . 4395 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 4396 RFC 5558, DOI 10.17487/RFC5558, February 2010, 4397 . 4399 [RFC5798] Nadas, S., Ed., "Virtual Router Redundancy Protocol (VRRP) 4400 Version 3 for IPv4 and IPv6", RFC 5798, 4401 DOI 10.17487/RFC5798, March 2010, 4402 . 4404 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 4405 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 4406 . 4408 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 4409 DOI 10.17487/RFC6081, January 2011, 4410 . 4412 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 4413 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 4414 DOI 10.17487/RFC6221, May 2011, 4415 . 4417 [RFC6247] Eggert, L., "Moving the Undeployed TCP Extensions RFC 4418 1072, RFC 1106, RFC 1110, RFC 1145, RFC 1146, RFC 1379, 4419 RFC 1644, and RFC 1693 to Historic Status", RFC 6247, 4420 DOI 10.17487/RFC6247, May 2011, 4421 . 4423 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 4424 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 4425 DOI 10.17487/RFC6355, August 2011, 4426 . 4428 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 4429 for Equal Cost Multipath Routing and Link Aggregation in 4430 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 4431 . 4433 [RFC6543] Gundavelli, S., "Reserved IPv6 Interface Identifier for 4434 Proxy Mobile IPv6", RFC 6543, DOI 10.17487/RFC6543, May 4435 2012, . 4437 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 4438 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 4439 . 4441 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 4442 UDP Checksums for Tunneled Packets", RFC 6935, 4443 DOI 10.17487/RFC6935, April 2013, 4444 . 4446 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 4447 for the Use of IPv6 UDP Datagrams with Zero Checksums", 4448 RFC 6936, DOI 10.17487/RFC6936, April 2013, 4449 . 4451 [RFC6980] Gont, F., "Security Implications of IPv6 Fragmentation 4452 with IPv6 Neighbor Discovery", RFC 6980, 4453 DOI 10.17487/RFC6980, August 2013, 4454 . 4456 [RFC7042] Eastlake 3rd, D. and J. Abley, "IANA Considerations and 4457 IETF Protocol and Documentation Usage for IEEE 802 4458 Parameters", BCP 141, RFC 7042, DOI 10.17487/RFC7042, 4459 October 2013, . 4461 [RFC7084] Singh, H., Beebee, W., Donley, C., and B. Stark, "Basic 4462 Requirements for IPv6 Customer Edge Routers", RFC 7084, 4463 DOI 10.17487/RFC7084, November 2013, 4464 . 4466 [RFC7401] Moskowitz, R., Ed., Heer, T., Jokela, P., and T. 4467 Henderson, "Host Identity Protocol Version 2 (HIPv2)", 4468 RFC 7401, DOI 10.17487/RFC7401, April 2015, 4469 . 4471 [RFC7421] Carpenter, B., Ed., Chown, T., Gont, F., Jiang, S., 4472 Petrescu, A., and A. Yourtchenko, "Analysis of the 64-bit 4473 Boundary in IPv6 Addressing", RFC 7421, 4474 DOI 10.17487/RFC7421, January 2015, 4475 . 4477 [RFC7526] Troan, O. and B. Carpenter, Ed., "Deprecating the Anycast 4478 Prefix for 6to4 Relay Routers", BCP 196, RFC 7526, 4479 DOI 10.17487/RFC7526, May 2015, 4480 . 4482 [RFC7542] DeKok, A., "The Network Access Identifier", RFC 7542, 4483 DOI 10.17487/RFC7542, May 2015, 4484 . 4486 [RFC7739] Gont, F., "Security Implications of Predictable Fragment 4487 Identification Values", RFC 7739, DOI 10.17487/RFC7739, 4488 February 2016, . 4490 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 4491 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 4492 Multicast - Sparse Mode (PIM-SM): Protocol Specification 4493 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 4494 2016, . 4496 [RFC7847] Melia, T., Ed. and S. Gundavelli, Ed., "Logical-Interface 4497 Support for IP Hosts with Multi-Access Support", RFC 7847, 4498 DOI 10.17487/RFC7847, May 2016, 4499 . 4501 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 4502 Writing an IANA Considerations Section in RFCs", BCP 26, 4503 RFC 8126, DOI 10.17487/RFC8126, June 2017, 4504 . 4506 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 4507 Decraene, B., Litkowski, S., and R. Shakir, "Segment 4508 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 4509 July 2018, . 4511 [RFC8726] Farrel, A., "How Requests for IANA Action Will Be Handled 4512 on the Independent Stream", RFC 8726, 4513 DOI 10.17487/RFC8726, November 2020, 4514 . 4516 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 4517 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 4518 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 4519 . 4521 [RFC8892] Thaler, D. and D. Romascanu, "Guidelines and Registration 4522 Procedures for Interface Types and Tunnel Types", 4523 RFC 8892, DOI 10.17487/RFC8892, August 2020, 4524 . 4526 [RFC8899] Fairhurst, G., Jones, T., Tuexen, M., Ruengeler, I., and 4527 T. Voelker, "Packetization Layer Path MTU Discovery for 4528 Datagram Transports", RFC 8899, DOI 10.17487/RFC8899, 4529 September 2020, . 4531 [RFC8900] Bonica, R., Baker, F., Huston, G., Hinden, R., Troan, O., 4532 and F. Gont, "IP Fragmentation Considered Fragile", 4533 BCP 230, RFC 8900, DOI 10.17487/RFC8900, September 2020, 4534 . 4536 [RFC8981] Gont, F., Krishnan, S., Narten, T., and R. Draves, 4537 "Temporary Address Extensions for Stateless Address 4538 Autoconfiguration in IPv6", RFC 8981, 4539 DOI 10.17487/RFC8981, February 2021, 4540 . 4542 Appendix A. OAL Checksum Algorithm 4544 The OAL Checksum Algorithm adopts the 8-bit Fletcher Checksum 4545 Algorithm specified in Appendix I of [RFC1146] as also analyzed in 4546 [CKSUM]. [RFC6247] declared [RFC1146] historic for the reason that 4547 the algorithms had never seen widespread use with TCP, however this 4548 document adopts the 8-bit Fletcher algorithm for a different purpose. 4549 Quoting from Appendix I of [RFC1146], the OAL Checksum Algorithm 4550 proceeds as follows: 4552 "The 8-bit Fletcher Checksum Algorithm is calculated over a 4553 sequence of data octets (call them D[1] through D[N]) by 4554 maintaining 2 unsigned 1's-complement 8-bit accumulators A and B 4555 whose contents are initially zero, and performing the following 4556 loop where i ranges from 1 to N: 4558 A := A + D[i] 4560 B := B + A 4562 It can be shown that at the end of the loop A will contain the 4563 8-bit 1's complement sum of all octets in the datagram, and that B 4564 will contain (N)D[1] + (N-1)D[2] + ... + D[N]." 4566 To calculate the OAL checksum, the above algorithm is applied over 4567 the N byte concatenation of the OAL pseudo-header, the encapsulated 4568 IP packet and the two-octet trailing checksum field initialized to 0. 4569 Specifically, the algorithm is first applied over the 40 octets of 4570 the OAL pseudo-header as data octets D[1] through D[40], then 4571 continues over the entire length of the original IP packet as data 4572 octets D[41] through D[N-2] and finally concludes with the two 4573 trailing 0 octets as data octets D[N-1] and D[N]. 4575 Appendix B. VDL Mode 2 Considerations 4577 ICAO Doc 9776 is the "Technical Manual for VHF Data Link Mode 2" 4578 (VDLM2) that specifies an essential radio frequency data link service 4579 for aircraft and ground stations in worldwide civil aviation air 4580 traffic management. The VDLM2 link type is "multicast capable" 4581 [RFC4861], but with considerable differences from common multicast 4582 links such as Ethernet and IEEE 802.11. 4584 First, the VDLM2 link data rate is only 31.5Kbps - multiple orders of 4585 magnitude less than most modern wireless networking gear. Second, 4586 due to the low available link bandwidth only VDLM2 ground stations 4587 (i.e., and not aircraft) are permitted to send broadcasts, and even 4588 so only as compact layer 2 "beacons". Third, aircraft employ the 4589 services of ground stations by performing unicast RS/RA exchanges 4590 upon receipt of beacons instead of listening for multicast RA 4591 messages and/or sending multicast RS messages. 4593 This beacon-oriented unicast RS/RA approach is necessary to conserve 4594 the already-scarce available link bandwidth. Moreover, since the 4595 numbers of beaconing ground stations operating within a given spatial 4596 range must be kept as sparse as possible, it would not be feasible to 4597 have different classes of ground stations within the same region 4598 observing different protocols. It is therefore highly desirable that 4599 all ground stations observe a common language of RS/RA as specified 4600 in this document. 4602 Note that links of this nature may benefit from compression 4603 techniques that reduce the bandwidth necessary for conveying the same 4604 amount of data. The IETF lpwan working group is considering possible 4605 alternatives: [https://datatracker.ietf.org/wg/lpwan/documents]. 4607 Appendix C. MN / AR Isolation Through L2 Address Mapping 4609 Per [RFC4861], IPv6 ND messages may be sent to either a multicast or 4610 unicast link-scoped IPv6 destination address. However, IPv6 ND 4611 messaging should be coordinated between the MN and AR only without 4612 invoking other nodes on the *NET. This implies that MN / AR control 4613 messaging should be isolated and not overheard by other nodes on the 4614 link. 4616 To support MN / AR isolation on some *NET links, ARs can maintain an 4617 OMNI-specific unicast L2 address ("MSADDR"). For Ethernet-compatible 4618 *NETs, this specification reserves one Ethernet unicast address TBD3 4619 (see: Section 25). For non-Ethernet statically-addressed *NETs, 4620 MSADDR is reserved per the assigned numbers authority for the *NET 4621 addressing space. For still other *NETs, MSADDR may be dynamically 4622 discovered through other means, e.g., L2 beacons. 4624 MNs map the L3 addresses of all IPv6 ND messages they send (i.e., 4625 both multicast and unicast) to MSADDR instead of to an ordinary 4626 unicast or multicast L2 address. In this way, all of the MN's IPv6 4627 ND messages will be received by ARs that are configured to accept 4628 packets destined to MSADDR. Note that multiple ARs on the link could 4629 be configured to accept packets destined to MSADDR, e.g., as a basis 4630 for supporting redundancy. 4632 Therefore, ARs must accept and process packets destined to MSADDR, 4633 while all other devices must not process packets destined to MSADDR. 4634 This model has well-established operational experience in Proxy 4635 Mobile IPv6 (PMIP) [RFC5213][RFC6543]. 4637 Appendix D. Change Log 4639 << RFC Editor - remove prior to publication >> 4641 Differences from draft-templin-6man-omni-06 to draft-templin-6man- 4642 omni-07: 4644 o Moved Interface Attributes, Type 1 and Type 2 to historic status. 4646 o Incorporated Traffic Selector into Interface Attributes, Type 4. 4648 Differences from draft-templin-6man-omni-05 to draft-templin-6man- 4649 omni-06: 4651 o Adopted TCP as an OAL packet-based connection-oriented protocol. 4653 o Three-Way handshake for establishing symmetric send/receive 4654 windows 4656 o Window length specified, plus "current" and "previous" windows 4658 o New appendix on checksum algorithm, with citations changed 4660 o Security architecture considerations. 4662 o More details on HIP message signatures. 4664 o Require firewalls at OAL destinations. 4666 o Removed "equal-length" requirement for OAL non-final fragments. 4668 Differences from draft-templin-6man-omni-04 to draft-templin-6man- 4669 omni-05: 4671 o Change to S/T-omIndex definition. 4673 Differences from draft-templin-6man-omni-03 to draft-templin-6man- 4674 omni-04: 4676 o Changed reference citations to "draft-templin-6man-aero". 4678 o Included introductory description of the "6M's". 4680 o Included new OMNI sub-option for PIM-SM. 4682 Differences from draft-templin-6man-omni-02 to draft-templin-6man- 4683 omni-03: 4685 o Added citation of RFC8726. 4687 Differences from draft-templin-6man-omni-01 to draft-templin-6man- 4688 omni-02: 4690 o Updated IANA registration policies for OMNI registries. 4692 Differences from draft-templin-6man-omni-00 to draft-templin-6man- 4693 omni-01: 4695 o Changed intended document status to Informational, and removed 4696 documents from "updates" category. 4698 o Updated implementation status. 4700 o Minor edits to HIP message specifications. 4702 o Clarified OAL and *NET IP header field settings during 4703 encapsulation and re-encapsulation. 4705 Differences from earlier versions to draft-templin-6man-omni-00: 4707 o Established working baseline reference. 4709 Authors' Addresses 4711 Fred L. Templin (editor) 4712 The Boeing Company 4713 P.O. Box 3707 4714 Seattle, WA 98124 4715 USA 4717 Email: fltemplin@acm.org 4719 Tony Whyman 4720 MWA Ltd c/o Inmarsat Global Ltd 4721 99 City Road 4722 London EC1Y 1AX 4723 England 4725 Email: tony.whyman@mccallumwhyman.com