idnits 2.17.1 draft-templin-6man-omni-interface-86.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. == The 'Updates: ' line in the draft header should list only the _numbers_ of the RFCs which will be updated by this document (if approved); it should not include the word 'RFC' in the list. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (March 9, 2021) is 1144 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: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'ULA' is mentioned on line 2265, but not defined -- Looks like a reference, but probably isn't: '00' on line 1791 -- Looks like a reference, but probably isn't: '63' on line 1791 -- Looks like a reference, but probably isn't: '03' on line 1774 -- Looks like a reference, but probably isn't: '04' on line 1775 -- Looks like a reference, but probably isn't: '07' on line 1775 -- Looks like a reference, but probably isn't: '28' on line 1776 -- Looks like a reference, but probably isn't: '31' on line 1776 -- Looks like a reference, but probably isn't: '64' on line 1793 -- Looks like a reference, but probably isn't: '65' on line 1793 -- Looks like a reference, but probably isn't: '66' on line 1793 -- Looks like a reference, but probably isn't: '1' on line 1861 -- Looks like a reference, but probably isn't: '2' on line 1863 -- Looks like a reference, but probably isn't: '3' on line 1863 == Missing Reference: 'RFCXXXX' is mentioned on line 3154, but not defined -- Looks like a reference, but probably isn't: '67' on line 3708 -- Looks like a reference, but probably isn't: '70' on line 3709 -- Looks like a reference, but probably isn't: '76' on line 3709 == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 3417, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-lla-type' is defined on line 3422, but no explicit reference was found in the text == Unused Reference: 'RFC2225' is defined on line 3468, but no explicit reference was found in the text == Unused Reference: 'RFC2529' is defined on line 3488, but no explicit reference was found in the text == Unused Reference: 'RFC3879' is defined on line 3517, but no explicit reference was found in the text == Unused Reference: 'RFC4271' is defined on line 3526, but no explicit reference was found in the text == Unused Reference: 'RFC4389' is defined on line 3540, but no explicit reference was found in the text == Unused Reference: 'RFC5175' is defined on line 3569, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 3606, but no explicit reference was found in the text == Unused Reference: 'RFC8754' is defined on line 3673, but no explicit reference was found in the text == Outdated reference: A later version (-37) exists of draft-ietf-drip-rid-06 == 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-19 == Outdated reference: A later version (-99) exists of draft-templin-intarea-6706bis-87 -- Obsolete informational reference (is this intentional?): RFC 3330 (Obsoleted by RFC 5735) Summary: 1 error (**), 0 flaws (~~), 20 warnings (==), 19 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 Updates: rfc1191, rfc4443, rfc8201 (if A. Whyman 5 approved) MWA Ltd c/o Inmarsat Global Ltd 6 Intended status: Standards Track March 9, 2021 7 Expires: September 10, 2021 9 Transmission of IP Packets over Overlay Multilink Network (OMNI) 10 Interfaces 11 draft-templin-6man-omni-interface-86 13 Abstract 15 Mobile nodes (e.g., aircraft of various configurations, terrestrial 16 vehicles, seagoing vessels, enterprise wireless devices, etc.) 17 communicate with networked correspondents over multiple access 18 network data links and configure mobile routers to connect end user 19 networks. A multilink interface specification is therefore needed 20 for coordination with the network-based mobility service. This 21 document specifies the transmission of IP packets over Overlay 22 Multilink Network (OMNI) 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 September 10, 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 . . . . . . . . . . . . . . . . . . . . . . . . . 5 60 3. Requirements . . . . . . . . . . . . . . . . . . . . . . . . 9 61 4. Overlay Multilink Network (OMNI) Interface Model . . . . . . 9 62 5. OMNI Interface Maximum Transmission Unit (MTU) . . . . . . . 14 63 5.1. OMNI Interface MTU/MRU . . . . . . . . . . . . . . . . . 14 64 5.2. The OMNI Adaptation Layer (OAL) . . . . . . . . . . . . . 15 65 5.3. OAL Addressing Considerations . . . . . . . . . . . . . . 19 66 5.4. OMNI Interface MTU Feedback Messaging . . . . . . . . . . 20 67 5.5. OAL Fragmentation Security Implications . . . . . . . . . 21 68 5.6. OAL Super-Packets . . . . . . . . . . . . . . . . . . . . 23 69 6. Frame Format . . . . . . . . . . . . . . . . . . . . . . . . 24 70 7. Link-Local Addresses (LLAs) . . . . . . . . . . . . . . . . . 25 71 8. Unique-Local Addresses (ULAs) . . . . . . . . . . . . . . . . 26 72 9. Global Unicast Addresses (GUAs) . . . . . . . . . . . . . . . 28 73 10. Node Identification . . . . . . . . . . . . . . . . . . . . . 28 74 11. Address Mapping - Unicast . . . . . . . . . . . . . . . . . . 29 75 11.1. Sub-Options . . . . . . . . . . . . . . . . . . . . . . 31 76 11.1.1. Pad1 . . . . . . . . . . . . . . . . . . . . . . . . 32 77 11.1.2. PadN . . . . . . . . . . . . . . . . . . . . . . . . 33 78 11.1.3. Interface Attributes (Type 1) . . . . . . . . . . . 33 79 11.1.4. Interface Attributes (Type 2) . . . . . . . . . . . 35 80 11.1.5. Traffic Selector . . . . . . . . . . . . . . . . . . 39 81 11.1.6. MS-Register . . . . . . . . . . . . . . . . . . . . 39 82 11.1.7. MS-Release . . . . . . . . . . . . . . . . . . . . . 40 83 11.1.8. Geo Coordinates . . . . . . . . . . . . . . . . . . 41 84 11.1.9. Dynamic Host Configuration Protocol for IPv6 85 (DHCPv6) Message . . . . . . . . . . . . . . . . . . 41 86 11.1.10. Host Identity Protocol (HIP) Message . . . . . . . . 42 87 11.1.11. Node Identification . . . . . . . . . . . . . . . . 43 88 11.1.12. Sub-Type Extension . . . . . . . . . . . . . . . . . 45 89 12. Address Mapping - Multicast . . . . . . . . . . . . . . . . . 48 90 13. Multilink Conceptual Sending Algorithm . . . . . . . . . . . 48 91 13.1. Multiple OMNI Interfaces . . . . . . . . . . . . . . . . 49 92 13.2. MN<->AR Traffic Loop Prevention . . . . . . . . . . . . 50 93 14. Router Discovery and Prefix Registration . . . . . . . . . . 50 94 14.1. Router Discovery in IP Multihop and IPv4-Only Networks . 54 95 14.2. MS-Register and MS-Release List Processing . . . . . . . 56 96 14.3. DHCPv6-based Prefix Registration . . . . . . . . . . . . 58 98 15. Secure Redirection . . . . . . . . . . . . . . . . . . . . . 59 99 16. AR and MSE Resilience . . . . . . . . . . . . . . . . . . . . 60 100 17. Detecting and Responding to MSE Failures . . . . . . . . . . 60 101 18. Transition Considerations . . . . . . . . . . . . . . . . . . 61 102 19. OMNI Interfaces on Open Internetworks . . . . . . . . . . . . 61 103 20. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . . . 63 104 21. (H)HITs and Temporary ULAs . . . . . . . . . . . . . . . . . 63 105 22. Address Selection . . . . . . . . . . . . . . . . . . . . . . 65 106 23. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 65 107 23.1. "IPv6 Neighbor Discovery Option Formats" Registry . . . 65 108 23.2. "Ethernet Numbers" Registry . . . . . . . . . . . . . . 65 109 23.3. "ICMPv6 Code Fields: Type 2 - Packet Too Big" Registry . 65 110 23.4. "OMNI Option Sub-Type Values" (New Registry) . . . . . . 66 111 23.5. "OMNI Node Identification ID-Type Values" (New Registry) 66 112 23.6. "OMNI Option Sub-Type Extension Values" (New Registry) . 67 113 23.7. "OMNI RFC4380 UDP/IP Header Option" (New Registry) . . . 67 114 23.8. "OMNI RFC6081 UDP/IP Trailer Option" (New Registry) . . 68 115 23.9. Additional Considerations . . . . . . . . . . . . . . . 68 116 24. Security Considerations . . . . . . . . . . . . . . . . . . . 69 117 25. Implementation Status . . . . . . . . . . . . . . . . . . . . 70 118 26. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 70 119 27. References . . . . . . . . . . . . . . . . . . . . . . . . . 71 120 27.1. Normative References . . . . . . . . . . . . . . . . . . 71 121 27.2. Informative References . . . . . . . . . . . . . . . . . 73 122 Appendix A. Interface Attribute Preferences Bitmap Encoding . . 80 123 Appendix B. VDL Mode 2 Considerations . . . . . . . . . . . . . 82 124 Appendix C. MN / AR Isolation Through L2 Address Mapping . . . . 83 125 Appendix D. Change Log . . . . . . . . . . . . . . . . . . . . . 83 126 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 86 128 1. Introduction 130 Mobile Nodes (MNs) (e.g., aircraft of various configurations, 131 terrestrial vehicles, seagoing vessels, enterprise wireless devices, 132 pedestrians with cellphones, etc.) often have multiple interface 133 connections to wireless and/or wired-link data links used for 134 communicating with networked correspondents. These data links may 135 have diverse performance, cost and availability properties that can 136 change dynamically according to mobility patterns, flight phases, 137 proximity to infrastructure, etc. MNs coordinate their data links in 138 a discipline known as "multilink", in which a single virtual 139 interface is configured over the node's underlying interface 140 connections to the data links. 142 The MN configures a virtual interface (termed the "Overlay Multilink 143 Network Interface (OMNI)") as a thin layer over the underlying 144 interfaces. The OMNI interface is therefore the only interface 145 abstraction exposed to the IP layer and behaves according to the Non- 146 Broadcast, Multiple Access (NBMA) interface principle, while 147 underlying interfaces appear as link layer communication channels in 148 the architecture. The OMNI interface internally employs the "OMNI 149 Adaptation Layer (OAL)" to ensure that packets are delivered without 150 loss due to size restrictions. The OMNI interface connects to a 151 virtual overlay service known as the "OMNI link". The OMNI link 152 spans one or more Internetworks that may include private-use 153 infrastructures and/or the global public Internet itself. 155 Each MN receives a Mobile Network Prefix (MNP) for numbering 156 downstream-attached End User Networks (EUNs) independently of the 157 access network data links selected for data transport. The MN 158 performs router discovery over the OMNI interface (i.e., similar to 159 IPv6 customer edge routers [RFC7084]) and acts as a mobile router on 160 behalf of its EUNs. The router discovery process is iterated over 161 each of the OMNI interface's underlying interfaces in order to 162 register per-link parameters (see Section 14). 164 The OMNI interface provides a multilink nexus for exchanging inbound 165 and outbound traffic via the correct underlying interface(s). The IP 166 layer sees the OMNI interface as a point of connection to the OMNI 167 link. Each OMNI link has one or more associated Mobility Service 168 Prefixes (MSPs), which are typically IP Global Unicast Address (GUA) 169 prefixes from which OMNI link MNPs are derived. If there are 170 multiple OMNI links, the IPv6 layer will see multiple OMNI 171 interfaces. 173 MNs may connect to multiple distinct OMNI links within the same OMNI 174 domain by configuring multiple OMNI interfaces, e.g., omni0, omni1, 175 omni2, etc. Each OMNI interface is configured over a set of 176 underlying interfaces and provides a nexus for Safety-Based Multilink 177 (SBM) operation. Each OMNI interface within the same OMNI domain 178 configures a common ULA prefix [ULA]::/48, and configures a unique 179 16-bit Subnet ID '*' to construct the sub-prefix [ULA*]::/64 (see: 180 Section 8). The IP layer applies SBM routing to select an OMNI 181 interface, which then applies Performance-Based Multilink (PBM) to 182 select the correct underlying interface. Applications can apply 183 Segment Routing [RFC8402] to select independent SBM topologies for 184 fault tolerance. 186 The OMNI interface interacts with a network-based Mobility Service 187 (MS) through IPv6 Neighbor Discovery (ND) control message exchanges 188 [RFC4861]. The MS provides Mobility Service Endpoints (MSEs) that 189 track MN movements and represent their MNPs in a global routing or 190 mapping system. 192 Many OMNI use cases have been proposed. In particular, the 193 International Civil Aviation Organization (ICAO) Working Group-I 194 Mobility Subgroup is developing a future Aeronautical 195 Telecommunications Network with Internet Protocol Services (ATN/IPS) 196 and has issued a liaison statement requesting IETF adoption [ATN] in 197 support of ICAO Document 9896 [ATN-IPS]. The IETF IP Wireless Access 198 in Vehicular Environments (ipwave) working group has further included 199 problem statement and use case analysis for OMNI in a document now in 200 AD evaluation for RFC publication 201 [I-D.ietf-ipwave-vehicular-networking]. Still other communities of 202 interest include AEEC, RTCA Special Committee 228 (SC-228) and NASA 203 programs that examine commercial aviation, Urban Air Mobility (UAM) 204 and Unmanned Air Systems (UAS). Pedestrians with handheld devices 205 represent another large class of potential OMNI users. 207 This document specifies the transmission of IP packets and MN/MS 208 control messages over OMNI interfaces. The OMNI interface supports 209 either IP protocol version (i.e., IPv4 [RFC0791] or IPv6 [RFC8200]) 210 as the network layer in the data plane, while using IPv6 ND messaging 211 as the control plane independently of the data plane IP protocol(s). 212 The OAL operates as a mid-layer between L3 and L2 based on IP-in-IPv6 213 encapsulation per [RFC2473] as discussed in the following sections. 214 OMNI interfaces enable multilink, mobility, multihop and multicast 215 services, with provisions for both Vehicle-to-Infrastructure (V2I) 216 communications and Vehicle-to-Vehicle (V2V) communications outside 217 the context of infrastructure. 219 2. Terminology 221 The terminology in the normative references applies; especially, the 222 terms "link" and "interface" are the same as defined in the IPv6 223 [RFC8200] and IPv6 Neighbor Discovery (ND) [RFC4861] specifications. 224 Additionally, this document assumes the following IPv6 ND message 225 types: Router Solicitation (RS), Router Advertisement (RA), Neighbor 226 Solicitation (NS), Neighbor Advertisement (NA) and Redirect. 228 The Protocol Constants defined in Section 10 of [RFC4861] are used in 229 their same format and meaning in this document. The terms "All- 230 Routers multicast", "All-Nodes multicast" and "Subnet-Router anycast" 231 are the same as defined in [RFC4291] (with Link-Local scope assumed). 233 The term "IP" is used to refer collectively to either Internet 234 Protocol version (i.e., IPv4 [RFC0791] or IPv6 [RFC8200]) when a 235 specification at the layer in question applies equally to either 236 version. 238 The following terms are defined within the scope of this document: 240 Mobile Node (MN) 241 an end system with a mobile router having multiple distinct 242 upstream data link connections that are grouped together in one or 243 more logical units. The MN's data link connection parameters can 244 change over time due to, e.g., node mobility, link quality, etc. 245 The MN further connects a downstream-attached End User Network 246 (EUN). The term MN used here is distinct from uses in other 247 documents, and does not imply a particular mobility protocol. 249 End User Network (EUN) 250 a simple or complex downstream-attached mobile network that 251 travels with the MN as a single logical unit. The IP addresses 252 assigned to EUN devices remain stable even if the MN's upstream 253 data link connections change. 255 Mobility Service (MS) 256 a mobile routing service that tracks MN movements and ensures that 257 MNs remain continuously reachable even across mobility events. 258 Specific MS details are out of scope for this document. 260 Mobility Service Endpoint (MSE) 261 an entity in the MS (either singular or aggregate) that 262 coordinates the mobility events of one or more MN. 264 Mobility Service Prefix (MSP) 265 an aggregated IP Global Unicast Address (GUA) prefix (e.g., 266 2001:db8::/32, 192.0.2.0/24, etc.) assigned to the OMNI link and 267 from which more-specific Mobile Network Prefixes (MNPs) are 268 delegated. OMNI link administrators typically obtain MSPs from an 269 Internet address registry, however private-use prefixes can 270 alternatively be used subject to certain limitations (see: 271 Section 9). OMNI links that connect to the global Internet 272 advertise their MSPs to their interdomain routing peers. 274 Mobile Network Prefix (MNP) 275 a longer IP prefix delegated from an MSP (e.g., 276 2001:db8:1000:2000::/56, 192.0.2.8/30, etc.) and assigned to a MN. 277 MNs sub-delegate the MNP to devices located in EUNs. 279 Access Network (ANET) 280 a data link service network (e.g., an aviation radio access 281 network, satellite service provider network, cellular operator 282 network, WiFi network, etc.) that connects MNs. Physical and/or 283 data link level security is assumed, and sometimes referred to as 284 "protected spectrum". Private enterprise networks and ground 285 domain aviation service networks may provide multiple secured IP 286 hops between the MN's point of connection and the nearest Access 287 Router. 289 Access Router (AR) 290 a router in the ANET for connecting MNs to correspondents in 291 outside Internetworks. The AR may be located on the same physical 292 link as the MN, or may be located multiple IP hops away. In the 293 latter case, the MN uses encapsulation to communicate with the AR 294 as though it were on the same physical link. 296 ANET interface 297 a MN's attachment to a link in an ANET. 299 Internetwork (INET) 300 a connected network region with a coherent IP addressing plan that 301 provides transit forwarding services between ANETs and nodes that 302 connect directly to the open INET via unprotected media. No 303 physical and/or data link level security is assumed, therefore 304 security must be applied by upper layers. The global public 305 Internet itself is an example. 307 INET interface 308 a node's attachment to a link in an INET. 310 *NET 311 a "wildcard" term used when a given specification applies equally 312 to both ANET and INET cases. 314 OMNI link 315 a Non-Broadcast, Multiple Access (NBMA) virtual overlay configured 316 over one or more INETs and their connected ANETs. An OMNI link 317 can comprise multiple INET segments joined by bridges the same as 318 for any link; the addressing plans in each segment may be mutually 319 exclusive and managed by different administrative entities. 321 OMNI interface 322 a node's attachment to an OMNI link, and configured over one or 323 more underlying *NET interfaces. If there are multiple OMNI links 324 in an OMNI domain, a separate OMNI interface is configured for 325 each link. 327 OMNI Adaptation Layer (OAL) 328 an OMNI interface process whereby packets admitted into the 329 interface are wrapped in a mid-layer IPv6 header and fragmented/ 330 reassembled if necessary to support the OMNI link Maximum 331 Transmission Unit (MTU). The OAL is also responsible for 332 generating MTU-related control messages as necessary, and for 333 providing addressing context for spanning multiple segments of a 334 bridged OMNI link. 336 OMNI Option 337 an IPv6 Neighbor Discovery option providing multilink parameters 338 for the OMNI interface as specified in Section 11. 340 Mobile Network Prefix Link Local Address (MNP-LLA) 341 an IPv6 Link Local Address that embeds the most significant 64 342 bits of an MNP in the lower 64 bits of fe80::/64, as specified in 343 Section 7. 345 Mobile Network Prefix Unique Local Address (MNP-ULA) 346 an IPv6 Unique-Local Address derived from an MNP-LLA. 348 Administrative Link Local Address (ADM-LLA) 349 an IPv6 Link Local Address that embeds a 32-bit administratively- 350 assigned identification value in the lower 32 bits of fe80::/96, 351 as specified in Section 7. 353 Administrative Unique Local Address (ADM-ULA) 354 an IPv6 Unique-Local Address derived from an ADM-LLA. 356 Multilink 357 an OMNI interface's manner of managing diverse underlying 358 interface connections to data links as a single logical unit. The 359 OMNI interface provides a single unified interface to upper 360 layers, while underlying interface selections are performed on a 361 per-packet basis considering factors such as DSCP, flow label, 362 application policy, signal quality, cost, etc. Multilinking 363 decisions are coordinated in both the outbound (i.e. MN to 364 correspondent) and inbound (i.e., correspondent to MN) directions. 366 Multihop 367 an iterative relaying of IP packets between MNs over an OMNI 368 underlying interface technology (such as omnidirectional wireless) 369 without support of fixed infrastructure. Multihop services entail 370 node-to-node relaying within a Mobile/Vehicular Ad-hoc Network 371 (MANET/VANET) for MN-to-MN communications and/or for "range 372 extension" where MNs within range of communications infrastructure 373 elements provide forwarding services for other MNs. 375 L2 376 The second layer in the OSI network model. Also known as "layer- 377 2", "link-layer", "sub-IP layer", "data link layer", etc. 379 L3 380 The third layer in the OSI network model. Also known as "layer- 381 3", "network-layer", "IP layer", etc. 383 underlying interface 384 a *NET interface over which an OMNI interface is configured. The 385 OMNI interface is seen as a L3 interface by the IP layer, and each 386 underlying interface is seen as a L2 interface by the OMNI 387 interface. The underlying interface either connects directly to 388 the physical communications media or coordinates with another node 389 where the physical media is hosted. 391 Mobility Service Identification (MSID) 392 Each MSE and AR is assigned a unique 32-bit Identification (MSID) 393 (see: Section 7). IDs are assigned according to MS-specific 394 guidelines (e.g., see: [I-D.templin-intarea-6706bis]). 396 Safety-Based Multilink (SBM) 397 A means for ensuring fault tolerance through redundancy by 398 connecting multiple affiliated OMNI interfaces to independent 399 routing topologies (i.e., multiple independent OMNI links). 401 Performance Based Multilink (PBM) 402 A means for selecting underlying interface(s) for packet 403 transmission and reception within a single OMNI interface. 405 OMNI Domain 406 The set of all SBM/PBM OMNI links that collectively provides 407 services for a common set of MSPs. Each OMNI domain consists of a 408 set of affiliated OMNI links that all configure the same ::/48 ULA 409 prefix with a unique 16-bit Subnet ID as discussed in Section 8. 411 3. Requirements 413 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 414 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 415 "OPTIONAL" in this document are to be interpreted as described in BCP 416 14 [RFC2119][RFC8174] when, and only when, they appear in all 417 capitals, as shown here. 419 An implementation is not required to internally use the architectural 420 constructs described here so long as its external behavior is 421 consistent with that described in this document. 423 4. Overlay Multilink Network (OMNI) Interface Model 425 An OMNI interface is a MN virtual interface configured over one or 426 more underlying interfaces, which may be physical (e.g., an 427 aeronautical radio link) or virtual (e.g., an Internet or higher- 428 layer "tunnel"). The MN receives a MNP from the MS, and coordinates 429 with the MS through IPv6 ND message exchanges. The MN uses the MNP 430 to construct a unique Link-Local Address (MNP-LLA) through the 431 algorithmic derivation specified in Section 7 and assigns the LLA to 432 the OMNI interface. 434 The OMNI interface architectural layering model is the same as in 435 [RFC5558][RFC7847], and augmented as shown in Figure 1. The IP layer 436 therefore sees the OMNI interface as a single L3 interface with 437 multiple underlying interfaces that appear as L2 communication 438 channels in the architecture. 440 +----------------------------+ 441 | Upper Layer Protocol | 442 Session-to-IP +---->| | 443 Address Binding | +----------------------------+ 444 +---->| IP (L3) | 445 IP Address +---->| | 446 Binding | +----------------------------+ 447 +---->| OMNI Interface | 448 Logical-to- +---->| (LLA) | 449 Physical | +----------------------------+ 450 Interface +---->| L2 | L2 | | L2 | 451 Binding |(IF#1)|(IF#2)| ..... |(IF#n)| 452 +------+------+ +------+ 453 | L1 | L1 | | L1 | 454 | | | | | 455 +------+------+ +------+ 457 Figure 1: OMNI Interface Architectural Layering Model 459 Each underlying interface provides an L2/L1 abstraction according to 460 one of the following models: 462 o INET interfaces connect to an INET either natively or through one 463 or several IPv4 Network Address Translators (NATs). Native INET 464 interfaces have global IP addresses that are reachable from any 465 INET correspondent. NATed INET interfaces typically have private 466 IP addresses and connect to a private network behind one or more 467 NATs that provide INET access. 469 o ANET interfaces connect to a protected ANET that is separated from 470 the open INET by an AR acting as a proxy. The ANET interface may 471 be either on the same L2 link segment as the AR, or separated from 472 the AR by multiple IP hops. 474 o VPNed interfaces use security encapsulation over a *NET to a 475 Virtual Private Network (VPN) gateway. Other than the link-layer 476 encapsulation format, VPNed interfaces behave the same as for 477 Direct interfaces. 479 o Direct (aka "point-to-point") interfaces connect directly to a 480 peer without crossing any *NET paths. An example is a line-of- 481 sight link between a remote pilot and an unmanned aircraft. 483 The OMNI virtual interface model gives rise to a number of 484 opportunities: 486 o since MNP-LLAs are uniquely derived from an MNP, no Duplicate 487 Address Detection (DAD) or Multicast Listener Discovery (MLD) 488 messaging is necessary. 490 o since Temporary ULAs are statistically unique, they can be used 491 without DAD, e.g. for MN-to-MN communications until an MNP-LLA is 492 obtained. 494 o *NET interfaces on the same L2 link segment as an AR do not 495 require any L3 addresses (i.e., not even link-local) in 496 environments where communications are coordinated entirely over 497 the OMNI interface. (An alternative would be to also assign the 498 same LLA to all *NET interfaces.) 500 o as underlying interface properties change (e.g., link quality, 501 cost, availability, etc.), any active interface can be used to 502 update the profiles of multiple additional interfaces in a single 503 message. This allows for timely adaptation and service continuity 504 under dynamically changing conditions. 506 o coordinating underlying interfaces in this way allows them to be 507 represented in a unified MS profile with provisions for mobility 508 and multilink operations. 510 o exposing a single virtual interface abstraction to the IPv6 layer 511 allows for multilink operation (including QoS based link 512 selection, packet replication, load balancing, etc.) at L2 while 513 still permitting L3 traffic shaping based on, e.g., DSCP, flow 514 label, etc. 516 o the OMNI interface allows inter-INET traversal when nodes located 517 in different INETs need to communicate with one another. This 518 mode of operation would not be possible via direct communications 519 over the underlying interfaces themselves. 521 o the OMNI Adaptation Layer (OAL) within the OMNI interface supports 522 lossless and adaptive path MTU mitigations not available for 523 communications directly over the underlying interfaces themselves. 524 The OAL supports "packing" of multiple IP payload packets within a 525 single OAL packet. 527 o L3 sees the OMNI interface as a point of connection to the OMNI 528 link; if there are multiple OMNI links (i.e., multiple MS's), L3 529 will see multiple OMNI interfaces. 531 o Multiple independent OMNI interfaces can be used for increased 532 fault tolerance through Safety-Based Multilink (SBM), with 533 Performance-Based Multilink (PBM) applied within each interface. 535 Other opportunities are discussed in [RFC7847]. Note that even when 536 the OMNI virtual interface is present, applications can still access 537 underlying interfaces either through the network protocol stack using 538 an Internet socket or directly using a raw socket. This allows for 539 intra-network (or point-to-point) communications without invoking the 540 OMNI interface and/or OAL. For example, when an IPv6 OMNI interface 541 is configured over an underlying IPv4 interface, applications can 542 still invoke IPv4 intra-network communications as long as the 543 communicating endpoints are not subject to mobility dynamics. 544 However, the opportunities discussed above are not available when the 545 architectural layering is bypassed in this way. 547 Figure 2 depicts the architectural model for a MN with an attached 548 EUN connecting to the MS via multiple independent *NETs. When an 549 underlying interface becomes active, the MN's OMNI interface sends 550 IPv6 ND messages without encapsulation if the first-hop Access Router 551 (AR) is on the same underlying link; otherwise, the interface uses 552 IP-in-IP encapsulation. The IPv6 ND messages traverse the ground 553 domain *NETs until they reach an AR (AR#1, AR#2, ..., AR#n), which 554 then coordinates with an INET Mobility Service Endpoint (MSE#1, 555 MSE#2, ..., MSE#m) and returns an IPv6 ND message response to the MN. 556 The Hop Limit in IPv6 ND messages is not decremented due to 557 encapsulation; hence, the OMNI interface appears to be attached to an 558 ordinary link. 560 +--------------+ (:::)-. 561 | MN |<-->.-(::EUN:::) 562 +--------------+ `-(::::)-' 563 |OMNI interface| 564 +----+----+----+ 565 +--------|IF#1|IF#2|IF#n|------ + 566 / +----+----+----+ \ 567 / | \ 568 / | \ 569 v v v 570 (:::)-. (:::)-. (:::)-. 571 .-(::*NET:::) .-(::*NET:::) .-(::*NET:::) 572 `-(::::)-' `-(::::)-' `-(::::)-' 573 +----+ +----+ +----+ 574 ... |AR#1| .......... |AR#2| ......... |AR#n| ... 575 . +-|--+ +-|--+ +-|--+ . 576 . | | | 577 . v v v . 578 . <----- INET Encapsulation -----> . 579 . . 580 . +-----+ (:::)-. . 581 . |MSE#2| .-(::::::::) +-----+ . 582 . +-----+ .-(::: INET :::)-. |MSE#m| . 583 . (::::: Routing ::::) +-----+ . 584 . `-(::: System :::)-' . 585 . +-----+ `-(:::::::-' . 586 . |MSE#1| +-----+ +-----+ . 587 . +-----+ |MSE#3| |MSE#4| . 588 . +-----+ +-----+ . 589 . . 590 . . 591 . <----- Worldwide Connected Internetwork ----> . 592 ........................................................... 594 Figure 2: MN/MS Coordination via Multiple *NETs 596 After the initial IPv6 ND message exchange, the MN (and/or any nodes 597 on its attached EUNs) can send and receive IP data packets over the 598 OMNI interface. OMNI interface multilink services will forward the 599 packets via ARs in the correct underlying *NETs. The AR encapsulates 600 the packets according to the capabilities provided by the MS and 601 forwards them to the next hop within the worldwide connected 602 Internetwork via optimal routes. 604 OMNI links span one or more underlying Internetwork via the OMNI 605 Adaptation Layer (OAL) which is based on a mid-layer overlay 606 encapsulation using [RFC2473]. Each OMNI link corresponds to a 607 different overlay (differentiated by an address codepoint) which may 608 be carried over a completely separate underlying topology. Each MN 609 can facilitate SBM by connecting to multiple OMNI links using a 610 distinct OMNI interface for each link. 612 Note: OMNI interface underlying interfaces often connect directly to 613 physical media on the local platform (e.g., a laptop computer with 614 WiFi, etc.), but in some configurations the physical media may be 615 hosted on a separate Local Area Network (LAN) node. In that case, 616 the OMNI interface can establish a Layer-2 VLAN or a point-to-point 617 tunnel (at a layer below the underlying interface) to the node 618 hosting the physical media. The OMNI interface may also apply 619 encapsulation at a layer above the underlying interface such that 620 packets would appear "double-encapsulated" on the LAN; the node 621 hosting the physical media in turn removes the LAN encapsulation 622 prior to transmission or inserts it following reception. Finally, 623 the underlying interface must monitor the node hosting the physical 624 media (e.g., through periodic keepalives) so that it can convey 625 up/down/status information to the OMNI interface. 627 5. OMNI Interface Maximum Transmission Unit (MTU) 629 The OMNI interface observes the link nature of tunnels, including the 630 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 631 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 632 The OMNI interface is configured over one or more underlying 633 interfaces as discussed in Section 4, where the interfaces (and their 634 associated *NET paths) may have diverse MTUs. The OMNI interface 635 accommodates MTU diversity by encapsulating original IP packets using 636 the OMNI Adaptation Layer (OAL) mid-layer encapsulation and 637 fragmentation/reassembly service. The OAL then further encapsulates 638 the resulting OAL packets/fragments in *NET headers for transmission 639 over underlying interfaces. OMNI interface considerations for 640 accommodating original IP packets of various sizes are discussed in 641 the following sections. 643 5.1. OMNI Interface MTU/MRU 645 IPv6 underlying interfaces are REQUIRED to configure a minimum MTU of 646 1280 bytes and a minimum MRU of 1500 bytes [RFC8200]. Therefore, the 647 minimum IPv6 path MTU is 1280 bytes since routers on the path are not 648 permitted to perform network fragmentation even though the 649 destination is required to reassemble more. The network therefore 650 MUST forward packets of at least 1280 bytes without generating an 651 IPv6 Path MTU Discovery (PMTUD) Packet Too Big (PTB) message 652 [RFC8201]. (Note: the source can apply "source fragmentation" for 653 locally-generated IPv6 packets up to 1500 bytes and larger still if 654 it if has a way to determine that the destination configures a larger 655 MRU, but this does not affect the minimum IPv6 path MTU.) 656 IPv4 underlying interfaces are REQUIRED to configure a minimum MTU of 657 68 bytes [RFC0791] and a minimum MRU of 576 bytes [RFC0791][RFC1122]. 658 Therefore, when the Don't Fragment (DF) bit in the IPv4 header is set 659 to 0 the minimum IPv4 path MTU is 576 bytes since routers on the path 660 support network fragmentation and the destination is required to 661 reassemble at least that much. The OMNI interface therefore MUST set 662 DF to 0 in the IPv4 encapsulation headers of OAL/*NET packets that 663 are no larger than 576 bytes, and MUST set DF to 1 in larger packets. 664 (Note: even if the encapsulation source has a way to determine that 665 the encapsulation destination configures an MRU larger than 576 666 bytes, it should not assume a larger minimum IPv4 path MTU without 667 careful consideration of the issues discussed in Section 5.5.) 669 The OMNI interface configures both an MTU and MRU of 9180 bytes 670 [RFC2492]; the size is therefore not a reflection of the underlying 671 interface or *NET path MTUs, but rather determines the largest 672 original "inner-layer" IP packet the OMNI interface can forward or 673 reassemble. The OMNI interface employs the OAL as a "mid-layer" 674 encapsulation, fragmentation and reassembly service for original IP 675 packets, and the OAL in turn uses "outer-layer" *NET encapsulation to 676 forward OAL/*NET packets/fragments over the underlying interfaces 677 (see: Section 5.2). 679 5.2. The OMNI Adaptation Layer (OAL) 681 When the network layer admits an original inner IP packet/fragment 682 into the OMNI interface, the OAL inserts a mid-layer IPv6 683 encapsulation per [RFC2473] if fragmentation, integrity verification 684 and/or explicit OAL addressing is required. In other cases (e.g., 685 for packet transmissions between ANET peers that share a common 686 underlying link), the OAL MAY omit the encapsulation. When OAL 687 encapsulation is used, the OAL sets the header fields per standard 688 IPv6 procedures but does not decrement the inner IP header Hop Limit/ 689 TTL since encapsulation occurs at a layer below IP forwarding. The 690 OAL next inserts a 2 octet trailer (initialized to 0) at the end of 691 the packet while incrementing the OAL header Payload Length to 692 reflect the addition of the trailer. The OAL then calculates the 2's 693 complement (mod 256) Fletcher's checksum [CKSUM][RFC2328][RFC0905] 694 over the entire OAL packet beginning with an IPv6 pseudo-header based 695 on the OAL header Payload Length (see Section 8.1 of [RFC8200]), then 696 writes the results over the trailer octets. The OAL then inserts a 697 single OMNI Routing Header (ORH) if necessary (see: 698 [I-D.templin-intarea-6706bis]) while incrementing Payload Length to 699 reflect the addition of the ORH, then fragments the OAL packet if 700 necessary. The OAL finally forwards the packet/fragments over an 701 underlying interface while adding any necessary *NET encapsulations. 702 The formats of OAL packets and fragments are shown in Figure 3. 704 +----------+-----+-----+-----+-----+-----+-----+----+ 705 | OAL Hdr | Original IP packet |Csum| 706 +----------+-----+-----+-----+-----+-----+-----+----+ 707 a) OAL packet after encapsulation but before fragmentation 709 +--------+----------+--+---------+ 710 |*NET Hdr| OAL Hdr |FH| Frag #1 | 711 +--------+----------+--+---------+ 712 +--------+----------+--+---------+ 713 |*NET Hdr| OAL Hdr |FH| Frag #2 | 714 +--------+----------+--+---------+ 715 +--------+----------+--+---------+ 716 |*NET Hdr| OAL Hdr |FH| Frag #3 | 717 +--------+----------+--+---------+ 718 .... 719 +--------+----------+--+---------+----+ 720 |*NET hdr| OAL Hdr |FH| Frag #n |Csum| 721 +--------+----------+--+---------+----+ 722 b) Fragments after OAL fragmentation and *NET encapsulation 723 (FH = Fragment Header; Csum appears only in final fragment) 725 Figure 3: OAL Packets and Fragments 727 When an OMNI interface receives a *NET encapsulated packet from an 728 underlying interface, the OAL discards the *NET encapsulation headers 729 and examines the OAL header if present. If the packet is addressed 730 to a different node, the OAL re-encapsulates and forwards as 731 discussed below; otherwise, the OAL performs reassembly if necessary 732 then removes the ORH if present while decrementing Payload Length to 733 reflect the removal of the ORH. If the OAL header is present, the 734 OAL next verifies the checksum and discards the packet if the 735 checksum is incorrect. If the packet was accepted, the OAL then 736 removes the OAL header/trailer, then delivers the original packet to 737 the IP layer. Note that link layers include a CRC-32 integrity check 738 which provides effective hop-by-hop error detection in the underlying 739 network for packet/fragment sizes up to the OMNI interface MTU [CRC], 740 but that some hops may traverse intermediate layers such as tunnels 741 over IPv4 that do not include integrity checks. The OAL source 742 therefore includes a trailing Fletcher checksum to allow the OAL 743 destination to detect packet splicing errors for fragmented OAL 744 packets and/or to verify integrity for packets that may have 745 traversed unprotected underlying network hops [CKSUM]. The Fletcher 746 checksum also provides diversity with respect to both lower layer 747 CRCs and upper layer Internet checksums as part of a complimentary 748 multi-layer integrity assurance architecture. 750 The OMNI interface assumes the IPv4 minimum path MTU (i.e., 576 751 bytes) as the worst case for OAL fragmentation regardless of the 752 underlying interface IP protocol version since IPv6/IPv4 protocol 753 translation and/or IPv6-in-IPv4 encapsulation may occur in any *NET 754 path. By always assuming the IPv4 minimum even for IPv6 underlying 755 interfaces, the OAL may produce smaller fragments with additional 756 encapsulation overhead but will always interoperate and never run the 757 risk of loss due to an MTU restriction or due to presenting a 758 destination interface with a packet that exceeds its MRU. 759 Additionally, the OAL path could traverse multiple *NET "segments" 760 with intermediate OAL forwarding nodes performing re-encapsulation 761 where the *NET encapsulation of the previous segment is replaced by 762 the *NET encapsulation of the next segment which may be based on a 763 different IP protocol version and/or encapsulation sizes. 765 The OAL therefore assumes a default minimum path MTU of 576 bytes at 766 each *NET segment for the purpose of generating OAL fragments. In 767 the worst case, each successive *NET segment may re-encapsulate with 768 either a 20 byte IPv4 or 40 byte IPv6 header, an 8 byte UDP header 769 and in some cases an IP security encapsulation (40 bytes maximum 770 assumed). Any *NET segment may also insert a maximum-length (40 771 byte) ORH as an extension to the existing 40 byte OAL IPv6 header 772 plus 8 byte Fragment Header if an ORH was not already present. 773 Assuming therefore an absolute worst case of (40 + 40 + 8) = 88 bytes 774 for *NET encapsulation plus (40 + 40 + 8) = 88 bytes for OAL 775 encapsulation leaves (576 - 88 - 88) = 400 bytes to accommodate a 776 portion of the original IP packet/fragment. OMNI interfaces 777 therefore set a minimum Maximum Payload Size (MPS) of 400 bytes as 778 the basis for the minimum-sized OAL fragment that can be assured of 779 traversing all segments without loss due to an MTU/MRU restriction. 780 The OAL fragmentation Maximum Fragment Size (MFS) is therefore 781 determined by the MPS plus the size of the OAL encapsulation headers. 782 (Note that the OAL source must include the 2 octet trailer as part of 783 the payload during fragmentation, and the OAL destination must regard 784 it as ordinary payload until reassembly and checksum verification are 785 complete.) 787 In light of the above, OAL source, intermediate and destination nodes 788 observe the following normative requirements: 790 o OAL sources MUST NOT send OAL packets/fragments including original 791 IP packets larger than the OMNI interface MTU, i.e., whether or 792 not fragmentation is needed. 794 o OAL sources MUST NOT perform OAL fragmentation for original IP 795 packets smaller than the minimum MPS minus the trailer size, and 796 MUST produce non-final fragments that contain a payload no smaller 797 than the minimum MPS when performing fragmentation. 799 o OAL sources MUST NOT send OAL packets/fragments that include any 800 extension headers other than a single ORH and a single Fragment 801 Header. 803 o OAL intermediate nodes SHOULD and OAL destinations MUST 804 unconditionally drop OAL packets/fragments including original IP 805 packets larger than the OMNI interface MRU, i.e., whether or not 806 reassembly is needed. 808 o OAL intermediate nodes SHOULD and OAL destinations MUST 809 unconditionally drop any non-final OAL fragments containing a 810 payload smaller than the minimum MPS. 812 o OAL intermediate nodes SHOULD and OAL destinations MUST 813 unconditionally drop OAL packets/fragments that include any 814 extension headers other than a single ORH and a single Fragment 815 Header. 817 The OAL source MAY maintain "path MPS" values for selected OAL 818 destinations initialized to the minimum MPS and increased to larger 819 values (up to the OMNI interface MTU) if better information is known 820 or discovered. For example, when *NET peers share a common 821 underlying link or a fixed path with a known larger MTU, the OAL can 822 base path MPS on this larger size (i.e., instead of 576 bytes) as 823 long as the *NET peer reassembles before re-encapsulating and 824 forwarding (while re-fragmenting if necessary). Also, if the OAL 825 source has a way of knowing the maximum *NET encapsulation size for 826 all segments along the path it may be able to increase path MPS to 827 reserve additional room for payload data. 829 The OAL source can also actively probe OAL destinations to discover 830 larger path MPS values (e.g., per [RFC4821][RFC8899]), but care must 831 be taken to avoid setting static values for dynamically changing 832 paths leading to black holes. The probe involves sending an OAL 833 packet larger than the current path MPS and receiving a small 834 acknowledgement message in response. For this purpose, the OAL 835 source can send a large NS message with OMNI options with PadN sub- 836 options (see: Section 11) in order to receive a small NA response 837 from the OAL destination. While observing the minimum MPS will 838 always result in robust and secure behavior, the OAL should optimize 839 path MPS values when more efficient utilization may result in better 840 performance (e.g. for wireless aviation data links). 842 Under the minimum MPS, a common 1500 byte inner IP packet would 843 require 4 OAL fragments, with each non-final fragment containing 400 844 payload bytes and the final fragment containing 302 payload bytes 845 (i.e., the final 300 bytes of the original IP packet plus the 2 octet 846 trailer). For all packet sizes, the likelihood of successful 847 reassembly may improve when the OMNI interface sends all fragments of 848 the same fragmented OAL packet consecutively over the same underlying 849 interface. Finally, an assured minimum/path MPS allows continuous 850 operation over all paths including those that traverse bridged L2 851 media with dissimilar MTUs. 853 Note: Some early network hardware of the past millennium was unable 854 to accept packet "bursts" resulting from an IP fragmentation event - 855 even to the point that the hardware would reset itself when presented 856 with a burst. This does not seem to be a common problem in the 857 modern era, where fragmentation and reassembly can be readily 858 demonstrated at line rate (e.g., using tools such as 'iperf3') even 859 over fast links on average hardware platforms. Even so, if fragile 860 hardware were suspected the OAL could impose an inter-fragment delay 861 and either increase the delay if loss was detected or decrease the 862 delay to improve performance when there is little/no loss. However, 863 no loss feedback loop exists at the OAL and even if there were any 864 inter-fragment delay would need to be on the order of a small number 865 of nanoseconds which is far smaller than the minimum intervals 866 supported in software sleeps and/or timers. Therefore, all hardware 867 must be able to robustly accept line-rate bursts of IP fragments on 868 each directly-connected link in order to avoid sustained loss. 870 5.3. OAL Addressing Considerations 872 The OMNI interface performs OAL encapsulation and selects source and 873 destination addresses for the OAL IPv6 header according to the node's 874 *NET orientation. The OMNI interface sets the OAL IPv6 header 875 addresses to Unique Local Addresses (ULAs) based on either 876 Administrative (ADM-ULA) or Mobile Network Prefix (MNP-ULA) values 877 (see: Section 8). When an ADM-ULA or MNP-ULA is not available, the 878 OMNI interface can use Temporary ULAs and/or Host Identity Tags 879 (HITs) instead (see: Section 21). The following sections discuss the 880 addressing considerations for OMNI Interfaces configured over *NET 881 interfaces. 883 When an OMNI interface sends an original IP packet toward a final 884 destination via an ANET interface, it sends without OAL encapsulation 885 if the packet (including any outer-layer ANET encapsulations) is no 886 larger than the underlying interface MTU for on-link ANET peers or 887 the minimum ANET path MTU for peers separated by multiple IP hops. 888 Otherwise, the OAL inserts an IPv6 header per [RFC2473] with source 889 address set to the node's own ULA and destination set to either the 890 Administrative ULA (ADM-ULA) of the ANET peer or the Mobile Network 891 Prefix ULA (MNP-ULA) corresponding to the final destination (see 892 below). The OAL then fragments if necessary, encapsulates the OAL 893 packet/fragments in any ANET headers and sends them to the ANET peer 894 which either reassembles before forwarding if the OAL destination is 895 its own ULA or forwards the fragments toward the final destination 896 without first reassembling otherwise. 898 When an OMNI interface sends an original IP packet toward a final 899 destination via an INET interface, it sends packets (including any 900 outer-layer INET encapsulations) no larger than the minimum INET path 901 MTU without OAL encapsulation if the destination is reached via an 902 INET address within the same OMNI link segment. Otherwise, the OAL 903 inserts an IPv6 header per [RFC2473] with source address set to the 904 node's ULA and destination set to the ULA of an OMNI node on the 905 final *NET segment. The OAL then fragments if necessary, 906 encapsulates the OAL packet/fragments in any INET headers and sends 907 them toward the final segment OMNI node, which reassembles before 908 forwarding toward the final destination. 910 5.4. OMNI Interface MTU Feedback Messaging 912 When the OMNI interface admits original IP packets, it invokes the 913 OAL and returns internally-generated ICMPv4 Fragmentation Needed 914 [RFC1191] or ICMPv6 Path MTU Discovery (PMTUD) Packet Too Big (PTB) 915 [RFC8201] messages as necessary. This document refers to both of 916 these ICMPv4/ICMPv6 message types simply as "PTBs", and introduces a 917 distinction between PTB "hard" and "soft" errors as discussed below. 919 Ordinary PTB messages with ICMPv4 header "unused" field or ICMPv6 920 header Code field value 0 are hard errors that always indicate that a 921 packet has been dropped due to a real MTU restriction. However, the 922 OMNI interface can also forward large original IP packets via OAL 923 encapsulation and fragmentation while at the same time returning PTB 924 soft error messages (subject to rate limiting) if it deems the 925 original IP packet too large according to factors such as link 926 performance characteristics, reassembly congestion, etc. This 927 ensures that the path MTU is adaptive and reflects the current path 928 used for a given data flow. The OMNI interface can therefore 929 continuously forward packets without loss while returning PTB soft 930 error messages recommending a smaller size if necessary. Original 931 sources that receive the soft errors in turn reduce the size of the 932 packets they send, i.e., the same as for hard errors. 934 The OMNI interface sets the ICMPv4 header "unused" field or ICMPv6 935 header Code field to the value 1 in PTB soft error messages. The 936 OMNI interface sets the PTB destination address to the source address 937 of the original packet, and sets the source address to the MNP Subnet 938 Router Anycast address of the MN. The OMNI interface then sets the 939 MTU field to a value no smaller than 576 for ICMPv4 or 1280 for 940 ICMPv6, and returns the PTB soft error to the original source. 942 When the original source receives the PTB, it reduces its path MTU 943 estimate the same as for hard errors but does not regard the message 944 as a loss indication. (If the original source does not recognize the 945 soft error code, it regards the PTB the same as a hard error but 946 should heed the retransmission advice given in [RFC8201] suggesting 947 retransmission based on normal packetization layer retransmission 948 timers.) This document therefore updates [RFC1191][RFC4443] and 949 [RFC8201]. Furthermore, implementations of [RFC4821] must be aware 950 that PTB hard or soft errors may arrive at any time even if after a 951 successful MTU probe (this is the same consideration as for an 952 ordinary path fluctuation following a successful probe). 954 An OMNI interface that reassembles OAL fragments may experience 955 congestion-oriented loss in its reassembly cache and can optionally 956 send PTB soft errors to the original source and/or ICMP "Time 957 Exceeded" messages to the source of the OAL fragments. In 958 environments where the messages may contribute to unacceptable 959 additional congestion, however, the OMNI interface can refrain from 960 sending PTB soft errors and simply regard the loss as an ordinary 961 unreported congestion event for which the original source will 962 eventually compensate. 964 Applications that receive PT soft errors can dynamically tune the 965 size of the packets they to send to produce the best possible 966 throughput and latency, with the understanding that these parameters 967 may change over time due to factors such as congestion, mobility, 968 network path changes, etc. The receipt or absence of soft errors 969 should be seen as hints of when increasing or decreasing packet sizes 970 may be beneficial. 972 In summary, the OMNI interface supports continuous transmission and 973 reception of packets of various sizes in the face of dynamically 974 changing network conditions. Moreover, since PTB soft errors do not 975 indicate loss, original sources that receive soft errors can quickly 976 scan for path MTU increases without waiting for the minimum 10 977 minutes specified for loss-oriented PTB hard errors 978 [RFC1191][RFC8201]. The OMNI interface therefore provides a lossless 979 and adaptive service that accommodates MTU diversity especially well- 980 suited for dynamic multilink environments. 982 5.5. OAL Fragmentation Security Implications 984 As discussed in Section 3.7 of [RFC8900], there are four basic 985 threats concerning IPv6 fragmentation; each of which is addressed by 986 effective mitigations as follows: 988 1. Overlapping fragment attacks - reassembly of overlapping 989 fragments is forbidden by [RFC8200]; therefore, this threat does 990 not apply to the OAL. 992 2. Resource exhaustion attacks - this threat is mitigated by 993 providing a sufficiently large OAL reassembly cache and 994 instituting "fast discard" of incomplete reassemblies that may be 995 part of a buffer exhaustion attack. The reassembly cache should 996 be sufficiently large so that a sustained attack does not cause 997 excessive loss of good reassemblies but not so large that (timer- 998 based) data structure management becomes computationally 999 expensive. The cache should also be indexed based on the arrival 1000 underlying interface such that congestion experienced over a 1001 first underlying interface does not cause discard of incomplete 1002 reassemblies for uncongested underlying interfaces. 1004 3. Attacks based on predictable fragment identification values - 1005 this threat is mitigated by selecting a suitably random ID value 1006 per [RFC7739]. Additionally, inclusion of the trailing Fletcher 1007 checksum would make it very difficult for an attacker who could 1008 somehow predict a fragment identification value to inject 1009 malicious fragments resulting in undetected reassemblies of bad 1010 data. 1012 4. Evasion of Network Intrusion Detection Systems (NIDS) - this 1013 threat is mitigated by setting a minimum MPS for OAL 1014 fragmentation, which defeats all "tiny fragment"-based attacks. 1016 Additionally, IPv4 fragmentation includes a 16-bit Identification (IP 1017 ID) field with only 65535 unique values such that at high data rates 1018 the field could wrap and apply to new packets while the fragments of 1019 old packets using the same ID are still alive in the network 1020 [RFC4963]. However, since the largest OAL fragment that will be sent 1021 via an IPv4 *NET path with DF = 0 is 576 bytes any IPv4 fragmentation 1022 would occur only on links with an IPv4 MTU smaller than this size, 1023 and [RFC3819] recommendations suggest that such links will have low 1024 data rates. Since IPv6 provides a 32-bit Identification value, IP ID 1025 wraparound at high data rates is not a concern for IPv6 1026 fragmentation. 1028 Finally, [RFC6980] documents fragmentation security concerns for 1029 large IPv6 ND messages. These concerns are addressed when the OMNI 1030 interface employs the OAL instead of directly fragmenting the IPv6 ND 1031 message itself. For this reason, OMNI interfaces MUST NOT admit IPv6 1032 ND messages larger than the OMNI interface MTU, and MUST employ the 1033 OAL for IPv6 ND messages admitted into the OMNI interface the same as 1034 discussed above. 1036 5.6. OAL Super-Packets 1038 By default, the OAL source includes a 40-byte IPv6 encapsulation 1039 header for each original IP packet during OAL encapsulation. The OAL 1040 source also calculates and appends a 2 octet trailing Fletcher 1041 checksum then performs fragmentation such that a copy of the 40-byte 1042 IPv6 header plus an 8-byte IPv6 Fragment Header is included in each 1043 OAL fragment (when an ORH is added, the OAL encapsulation headers 1044 become larger still). However, these encapsulations may represent 1045 excessive overhead in some environments. A technique known as 1046 "packing" discussed in [I-D.ietf-intarea-tunnels] is therefore 1047 supported so that multiple original IP packets can be included within 1048 a single OAL "super-packet". 1050 When the OAL source has multiple original IP packets with total 1051 length no larger than the OMNI interface MTU to send to the same 1052 destination, it can concatenate them into a super-packet encapsulated 1053 in a single OAL header and trailing checksum. Within the OAL super- 1054 packet, the IP header of the first original packet (iHa) followed by 1055 its data (iDa) is concatenated immediately following the OAL header, 1056 then the IP header of the next original packet (iHb) followed by its 1057 data (iDb) is concatenated immediately following the first original 1058 packet, etc. with the trailing checksum included last. The OAL 1059 super-packet format is transposed from [I-D.ietf-intarea-tunnels] and 1060 shown in Figure 4: 1062 <------- Original IP packets -------> 1063 +-----+-----+ 1064 | iHa | iDa | 1065 +-----+-----+ 1066 | 1067 | +-----+-----+ 1068 | | iHb | iDb | 1069 | +-----+-----+ 1070 | | 1071 | | +-----+-----+ 1072 | | | iHc | iDc | 1073 | | +-----+-----+ 1074 | | | 1075 v v v 1076 +----------+-----+-----+-----+-----+-----+-----+----+ 1077 | OAL Hdr | iHa | iDa | iHb | iDb | iHc | iDc |Csum| 1078 +----------+-----+-----+-----+-----+-----+-----+----+ 1079 <--- OAL "Super-Packet" with single OAL Hdr/Csum ---> 1081 Figure 4: OAL Super-Packet Format 1083 When the OAL source prepares a super-packet, it applies OAL 1084 fragmentation if necessary then sends the packet or fragments to the 1085 OAL destination. When the OAL destination receives the super-packet 1086 it reassembles if necessary, verifies and removes the trailing 1087 checksum, then regards the remaining OAL header Payload Length as the 1088 sum of the lengths of all payload packets. The OAL destination then 1089 selectively extracts each original IP packet (e.g., by setting 1090 pointers into the super-packet buffer and maintaining a reference 1091 count, by copying each packet into a separate buffer, etc.) and 1092 forwards each packet or processes it locally as appropriate. During 1093 extraction, the OAL determines the IP protocol version of each 1094 successive original IP packet 'j' by examining the four most- 1095 significant bits of iH(j), and determines the length of the packet by 1096 examining the rest of iH(j) according to the IP protocol version. 1098 Note that OMNI interfaces must take care to avoid processing super- 1099 packet payload elements that would subvert security. Specifically, 1100 if a super-packet contains a mix of data and control payload packets 1101 (which could include critical security codes), the node MUST NOT 1102 process the data packets before processing the control packets 1104 6. Frame Format 1106 The OMNI interface transmits original IP packets by first invoking 1107 the OAL, next including any *NET encapsulations and finally engaging 1108 the native frame format of the underlying interface. For example, 1109 for Ethernet-compatible interfaces the frame format is specified in 1110 [RFC2464], for aeronautical radio interfaces the frame format is 1111 specified in standards such as ICAO Doc 9776 (VDL Mode 2 Technical 1112 Manual), for various forms of tunnels the frame format is found in 1113 the appropriate tunneling specification, etc. 1115 The OMNI interface SHOULD minimize the amount of *NET encapsulation 1116 for increased efficiency. For example, while an OMNI node may need 1117 to use UDP/IP as a *NET encapsulation over underlying interfaces 1118 connected to an open Internetwork, it may be able to omit the UDP 1119 header over underlying interfaces connected to *NETs that do not 1120 include NATs or packet filtering gateways. Similarly, when an OMNI 1121 MN shares a common underlying link with an AR, the MN may be able to 1122 avoid including any *NET encapsulations and instead directly engage 1123 the underlying interface native frame format. Further considerations 1124 for *NET encapsulation are discussed throughout the document and in 1125 [I-D.templin-intarea-6706bis]. 1127 7. Link-Local Addresses (LLAs) 1129 OMNI nodes are assigned OMNI interface IPv6 Link-Local Addresses 1130 (LLAs) through pre-service administrative actions. "MNP-LLAs" embed 1131 the MNP assigned to the mobile node, while "ADM-LLAs" include an 1132 administratively-unique ID that is guaranteed to be unique on the 1133 link. LLAs are configured as follows: 1135 o IPv6 MNP-LLAs encode the most-significant 64 bits of a MNP within 1136 the least-significant 64 bits of the IPv6 link-local prefix 1137 fe80::/64, i.e., in the LLA "interface identifier" portion. The 1138 prefix length for the LLA is determined by adding 64 to the MNP 1139 prefix length. For example, for the MNP 2001:db8:1000:2000::/56 1140 the corresponding MNP-LLA is fe80::2001:db8:1000:2000/120. 1142 o IPv4-compatible MNP-LLAs are constructed as fe80::ffff:[IPv4], 1143 i.e., the interface identifier consists of 16 '0' bits, followed 1144 by 16 '1' bits, followed by a 32bit IPv4 address/prefix. The 1145 prefix length for the LLA is determined by adding 96 to the MNP 1146 prefix length. For example, the IPv4-Compatible MN OMNI LLA for 1147 192.0.2.0/24 is fe80::ffff:192.0.2.0/120 (also written as 1148 fe80::ffff:c000:0200/120). 1150 o ADM-LLAs are assigned to ARs and MSEs and MUST be managed for 1151 uniqueness. The lower 32 bits of the LLA includes a unique 1152 integer "MSID" value between 0x00000001 and 0xfeffffff, e.g., as 1153 in fe80::1, fe80::2, fe80::3, etc., fe80::feffffff. The ADM-LLA 1154 prefix length is determined by adding 96 to the MSID prefix 1155 length. For example, if the prefix length for MSID 0x10012001 is 1156 16 then the ADM-LLA prefix length is set to 112 and the LLA is 1157 written as fe80::1001:2001/112. The "zero" address for each ADM- 1158 LLA prefix is the Subnet-Router anycast address for that prefix 1159 [RFC4291]; for example, the Subnet-Router anycast address for 1160 fe80::1001:2001/112 is simply fe80::1001:2000. The MSID range 1161 0xff000000 through 0xffffffff is reserved for future use. 1163 Since the prefix 0000::/8 is "Reserved by the IETF" [RFC4291], no 1164 MNPs can be allocated from that block ensuring that there is no 1165 possibility for overlap between the different MNP- and ADM-LLA 1166 constructs discussed above. 1168 Since MNP-LLAs are based on the distribution of administratively 1169 assured unique MNPs, and since ADM-LLAs are guaranteed unique through 1170 administrative assignment, OMNI interfaces set the autoconfiguration 1171 variable DupAddrDetectTransmits to 0 [RFC4862]. 1173 Note: If future protocol extensions relax the 64-bit boundary in IPv6 1174 addressing, the additional prefix bits of an MNP could be encoded in 1175 bits 16 through 63 of the MNP-LLA. (The most-significant 64 bits 1176 would therefore still be in bits 64-127, and the remaining bits would 1177 appear in bits 16 through 48.) However, the analysis provided in 1178 [RFC7421] suggests that the 64-bit boundary will remain in the IPv6 1179 architecture for the foreseeable future. 1181 Note: Even though this document honors the 64-bit boundary in IPv6 1182 addressing, it specifies prefix lengths longer than /64 for routing 1183 purposes. This effectively extends IPv6 routing determination into 1184 the interface identifier portion of the IPv6 address, but it does not 1185 redefine the 64-bit boundary. Modern routing protocol 1186 implementations honor IPv6 prefixes of all lengths, up to and 1187 including /128. 1189 8. Unique-Local Addresses (ULAs) 1191 OMNI domains use IPv6 Unique-Local Addresses (ULAs) as the source and 1192 destination addresses in OAL IPv6 encapsulation headers. ULAs are 1193 only routable within the scope of a an OMNI domain, and are derived 1194 from the IPv6 Unique Local Address prefix fc00::/7 followed by the L 1195 bit set to 1 (i.e., as fd00::/8) followed by a 40-bit pseudo-random 1196 Global ID to produce the prefix [ULA]::/48, which is then followed by 1197 a 16-bit Subnet ID then finally followed by a 64 bit Interface ID as 1198 specified in Section 3 of [RFC4193]. All nodes in the same OMNI 1199 domain configure the same 40-bit Global ID as the OMNI domain 1200 identifier. The statistic uniqueness of the 40-bit pseudo-random 1201 Global ID allows different OMNI domains to be joined together in the 1202 future without requiring renumbering. 1204 Each OMNI link instance is identified by a value between 0x0000 and 1205 0xfeff in bits 48-63 of [ULA]::/48; the values 0xff00 through 0xfffe 1206 are reserved for future use, and the value 0xffff denotes the 1207 presence of a Temporary ULA (see below). For example, OMNI ULAs 1208 associated with instance 0 are configured from the prefix 1209 [ULA]:0000::/64, instance 1 from [ULA]:0001::/64, instance 2 from 1210 [ULA]:0002::/64, etc. ULAs and their associated prefix lengths are 1211 configured in correspondence with LLAs through stateless prefix 1212 translation where "MNP-ULAs" are assigned in correspondence to MNP- 1213 LLAs and "ADM-ULAs" are assigned in correspondence to ADM-LLAs. For 1214 example, for OMNI link instance [ULA]:1010::/64: 1216 o the MNP-ULA corresponding to the MNP-LLA fe80::2001:db8:1:2 with a 1217 56-bit MNP length is derived by copying the lower 64 bits of the 1218 LLA into the lower 64 bits of the ULA as 1219 [ULA]:1010:2001:db8:1:2/120 (where, the ULA prefix length becomes 1220 64 plus the IPv6 MNP length). 1222 o the MNP-ULA corresponding to fe80::ffff:192.0.2.0 with a 28-bit 1223 MNP length is derived by simply writing the LLA interface ID into 1224 the lower 64 bits as [ULA]:1010:0:ffff:192.0.2.0/124 (where, the 1225 ULA prefix length is 64 plus 32 plus the IPv4 MNP length). 1227 o the ADM-ULA corresponding to fe80::1000/112 is simply 1228 [ULA]:1010::1000/112. 1230 o the ADM-ULA corresponding to fe80::/128 is simply 1231 [ULA]:1010::/128. 1233 o etc. 1235 Each OMNI interface assigns the Anycast ADM-ULA specific to the OMNI 1236 link instance. For example, the OMNI interface connected to instance 1237 3 assigns the Anycast address [ULA]:0003::/128. Routers that 1238 configure OMNI interfaces advertise the OMNI service prefix (e.g., 1239 [ULA]:0003::/64) into the local routing system so that applications 1240 can direct traffic according to SBM requirements. 1242 The ULA presents an IPv6 address format that is routable within the 1243 OMNI routing system and can be used to convey link-scoped IPv6 ND 1244 messages across multiple hops using IPv6 encapsulation [RFC2473]. 1245 The OMNI link extends across one or more underling Internetworks to 1246 include all ARs and MSEs. All MNs are also considered to be 1247 connected to the OMNI link, however OAL encapsulation is omitted 1248 whenever possible to conserve bandwidth (see: Section 13). 1250 Each OMNI link can be subdivided into "segments" that often 1251 correspond to different administrative domains or physical 1252 partitions. OMNI nodes can use IPv6 Segment Routing [RFC8402] when 1253 necessary to support efficient packet forwarding to destinations 1254 located in other OMNI link segments. A full discussion of Segment 1255 Routing over the OMNI link appears in [I-D.templin-intarea-6706bis]. 1257 Temporary ULAs are constructed per [RFC8981] based on the prefix 1258 [ULA]:ffff::/64 and used by MNs when they have no other addresses. 1259 Temporary ULAs can be used for MN-to-MN communications outside the 1260 context of any supporting OMNI link infrastructure, and can also be 1261 used as an initial address while the MN is in the process of 1262 procuring an MNP. Temporary ULAs are not routable within the OMNI 1263 routing system, and are therefore useful only for OMNI link "edge" 1264 communications. Temporary ULAs employ optimistic DAD principles 1265 [RFC4429] since they are probabilistically unique. 1267 Note: IPv6 ULAs taken from the prefix fc00::/7 followed by the L bit 1268 set to 0 (i.e., as fc00::/8) are never used for OMNI OAL addressing, 1269 however the range could be used for MSP and MNP addressing under 1270 certain limiting conditions (see: Section 9). 1272 9. Global Unicast Addresses (GUAs) 1274 OMNI domains use IP Global Unicast Address (GUA) prefixes [RFC4291] 1275 as Mobility Service Prefixes (MSPs) from which Mobile Network 1276 Prefixes (MNP) are delegated to Mobile Nodes (MNs). 1278 For IPv6, GUA prefixes are assigned by IANA [IPV6-GUA] and/or an 1279 associated regional assigned numbers authority such that the OMNI 1280 domain can be interconnected to the global IPv6 Internet without 1281 causing inconsistencies in the routing system. An OMNI domain could 1282 instead use ULAs with the 'L' bit set to 0 (i.e., from the prefix 1283 fc00::/8)[RFC4193], however this would require IPv6 NAT if the domain 1284 were ever connected to the global IPv6 Internet. 1286 For IPv4, GUA prefixes are assigned by IANA [IPV4-GUA] and/or an 1287 associated regional assigned numbers authority such that the OMNI 1288 domain can be interconnected to the global IPv4 Internet without 1289 causing routing inconsistencies. An OMNI domain could instead use 1290 private IPv4 prefixes (e.g., 10.0.0.0/8, etc.) [RFC3330], however 1291 this would require IPv4 NAT if the domain were ever connected to the 1292 global IPv4 Internet. 1294 10. Node Identification 1296 OMNI MNs and MSEs that connect over open Internetworks generate a 1297 Host Identity Tag (HIT) as specified in [RFC7401] and use the value 1298 as a robust general-purpose node identification value. Hierarchical 1299 HITs (HHITs) [I-D.ietf-drip-rid] may provide a useful alternative in 1300 certain domains such as the Unmanned (Air) Traffic Management (UTM) 1301 service for Unmanned Air Systems (UAS). MNs and MSEs can then use 1302 their (H)HITs in IPv6 ND control message exchanges. 1304 When a MN is truly outside the context of any infrastructure, it may 1305 have no MNP information at all. In that case, the MN can use its 1306 (H)HIT as an IPv6 source/destination address for sustained 1307 communications in Vehicle-to-Vehicle (V2V) and (multihop) Vehicle-to- 1308 Infrastructure (V2I) scenarios. The MN can also propagate the (H)HIT 1309 into the multihop routing tables of (collective) Mobile/Vehicular Ad- 1310 hoc Networks (MANETs/VANETs) using only the vehicles themselves as 1311 communications relays. 1313 When a MN connects to ARs over (non-multihop) protected-spectrum 1314 ANETs, an alternate form of node identification (e.g., MAC address, 1315 serial number, airframe identification value, VIN, etc.) may be 1316 sufficient. In that case, the MN should still generate a (H)HIT and 1317 maintain it in conjunction with any other node identifiers. The MN 1318 can then include OMNI "Node Identification" sub-options (see: 1319 Section 11.1.11) in IPv6 ND messages should the need to transmit 1320 identification information over the network arise. 1322 11. Address Mapping - Unicast 1324 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1325 state and use the link-local address format specified in Section 7. 1326 OMNI interface IPv6 Neighbor Discovery (ND) [RFC4861] messages sent 1327 over physical underlying interfaces without encapsulation observe the 1328 native underlying interface Source/Target Link-Layer Address Option 1329 (S/TLLAO) format (e.g., for Ethernet the S/TLLAO is specified in 1330 [RFC2464]). OMNI interface IPv6 ND messages sent over underlying 1331 interfaces via encapsulation do not include S/TLLAOs which were 1332 intended for encoding physical L2 media address formats and not 1333 encapsulation IP addresses. Furthermore, S/TLLAOs are not intended 1334 for encoding additional interface attributes needed for multilink 1335 coordination. Hence, this document does not define an S/TLLAO format 1336 but instead defines a new option type termed the "OMNI option" 1337 designed for these purposes. 1339 MNs such as aircraft typically have many wireless data link types 1340 (e.g. satellite-based, cellular, terrestrial, air-to-air directional, 1341 etc.) with diverse performance, cost and availability properties. 1342 The OMNI interface would therefore appear to have multiple L2 1343 connections, and may include information for multiple underlying 1344 interfaces in a single IPv6 ND message exchange. OMNI interfaces use 1345 an IPv6 ND option called the OMNI option formatted as shown in 1346 Figure 5: 1348 0 1 2 3 1349 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 1350 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1351 | Type | Length | Preflen | S/T-omIndex | 1352 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1353 | | 1354 ~ Sub-Options ~ 1355 | | 1356 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1358 Figure 5: OMNI Option Format 1360 In this format: 1362 o Type is set to TBD1. 1364 o Length is set to the number of 8 octet blocks in the option. The 1365 value 0 is invalid, while the values 1 through 255 (i.e., 8 1366 through 2040 octets, respectively) indicate the total length of 1367 the OMNI option. 1369 o Preflen is an 8 bit field that determines the length of prefix 1370 associated with an LLA. Values 0 through 128 specify a valid 1371 prefix length (all other values are invalid). For IPv6 ND 1372 messages sent from a MN to the MS, Preflen applies to the IPv6 1373 source LLA and provides the length that the MN is requesting or 1374 asserting to the MS. For IPv6 ND messages sent from the MS to the 1375 MN, Preflen applies to the IPv6 destination LLA and indicates the 1376 length that the MS is granting to the MN. For IPv6 ND messages 1377 sent between MS endpoints, Preflen provides the length associated 1378 with the source/target MN that is subject of the ND message. 1380 o S/T-omIndex is an 8 bit field corresponds to the omIndex value for 1381 source or target underlying interface used to convey this IPv6 ND 1382 message. OMNI interfaces MUST number each distinct underlying 1383 interface with an omIndex value between '1' and '255' that 1384 represents a MN-specific 8-bit mapping for the actual ifIndex 1385 value assigned by network management [RFC2863] (the omIndex value 1386 '0' is reserved for use by the MS). For RS and NS messages, S/ 1387 T-omIndex corresponds to the source underlying interface the 1388 message originated from. For RA and NA messages, S/T-omIndex 1389 corresponds to the target underlying interface that the message is 1390 destined to. (For NS messages used for Neighbor Unreachability 1391 Detection (NUD), S/T-omIndex instead identifies the neighbor's 1392 underlying interface to be used as the target interface to return 1393 the NA.) 1395 o Sub-Options is a Variable-length field, of length such that the 1396 complete OMNI Option is an integer multiple of 8 octets long. 1397 Contains one or more Sub-Options, as described in Section 11.1. 1399 The OMNI option may appear in any IPv6 ND message type; it is 1400 processed by interfaces that recognize the option and ignored by all 1401 other interfaces. If multiple OMNI option instances appear in the 1402 same IPv6 ND message, the interface processes the Preflen and S/ 1403 T-omIndex fields in the first instance and ignores those fields in 1404 all other instances. The interface processes the Sub-Options of all 1405 OMNI option instances in the same IPv6 ND message in the consecutive 1406 order in which they appear. 1408 The OMNI option(s) in each IPv6 ND message may include full or 1409 partial information for the neighbor. The union of the information 1410 in the most recently received OMNI options is therefore retained, and 1411 the information is aged/removed in conjunction with the corresponding 1412 neighbor cache entry. 1414 11.1. Sub-Options 1416 Each OMNI option includes zero or more Sub-Options. Each consecutive 1417 Sub-Option is concatenated immediately after its predecessor. All 1418 Sub-Options except Pad1 (see below) are in type-length-value (TLV) 1419 encoded in the following format: 1421 0 1 2 1422 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 1423 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1424 | Sub-Type| Sub-length | Sub-Option Data ... 1425 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1427 Figure 6: Sub-Option Format 1429 o Sub-Type is a 5-bit field that encodes the Sub-Option type. Sub- 1430 Options defined in this document are: 1432 Sub-Option Name Sub-Type 1433 Pad1 0 1434 PadN 1 1435 Interface Attributes (Type 1) 2 1436 Interface Attributes (Type 2) 3 1437 Traffic Selector 4 1438 MS-Register 5 1439 MS-Release 6 1440 Geo Coordinates 7 1441 DHCPv6 Message 8 1442 HIP Message 9 1443 Node Identification 10 1444 Sub-Type Extension 30 1446 Figure 7 1448 Sub-Types 11-29 are available for future assignment for major 1449 protocol functions. Sub-Type 31 is reserved by IANA. 1451 o Sub-Length is an 11-bit field that encodes the length of the Sub- 1452 Option Data ranging from 0 to 2034 octets. 1454 o Sub-Option Data is a block of data with format determined by Sub- 1455 Type and length determined by Sub-Length. 1457 During transmission, the OMNI interface codes Sub-Type and Sub-Length 1458 together in network byte order in 2 consecutive octets, where Sub- 1459 Option Data may be up to 2034 octets in length. This allows ample 1460 space for coding large objects (e.g., ASCII strings, domain names, 1461 protocol messages, security codes, etc.), while a single OMNI option 1462 is limited to 2040 octets the same as for any IPv6 ND option. If the 1463 Sub-Options to be coded would cause an OMNI option to exceed 2040 1464 octets, the OMNI interface codes any remaining Sub-Options in 1465 additional OMNI option instances in the intended order of processing 1466 in the same IPv6 ND message. Implementations must therefore observe 1467 size limitations, and must refrain from sending IPv6 ND messages 1468 larger than the OMNI interface MTU. If the available OMNI 1469 information would cause a single IPv6 ND message to exceed the OMNI 1470 interface MTU, the OMNI interface codes as much as possible in a 1471 first IPv6 ND message and codes the remainder in additional IPv6 ND 1472 messages. 1474 During reception, the OMNI interface processes each OMNI option Sub- 1475 Option while skipping over and ignoring any unrecognized Sub-Options. 1476 The OMNI interface processes the Sub-Options of all OMNI option 1477 instances in the consecutive order in which they appear in the IPv6 1478 ND message, beginning with the first instance and continuing through 1479 any additional instances to the end of the message. If a Sub-Option 1480 length would cause processing to exceed the OMNI option total length, 1481 the OMNI interface accepts any Sub-Options already processed and 1482 ignores the final Sub-Option. The interface then processes any 1483 remaining OMNI options in the same fashion to the end of the IPv6 ND 1484 message. 1486 Note: large objects that exceed the Sub-Option Data limit of 2034 1487 octets are not supported under the current specification; if this 1488 proves to be limiting in practice, future specifications may define 1489 support for fragmenting large objects across multiple OMNI options 1490 within the same IPv6 ND message. 1492 The following Sub-Option types and formats are defined in this 1493 document: 1495 11.1.1. Pad1 1497 0 1498 0 1 2 3 4 5 6 7 1499 +-+-+-+-+-+-+-+-+ 1500 | S-Type=0|x|x|x| 1501 +-+-+-+-+-+-+-+-+ 1503 Figure 8: Pad1 1505 o Sub-Type is set to 0. If multiple instances appear in OMNI 1506 options of the same message all are processed. 1508 o Sub-Type is followed by 3 'x' bits, set to any value on 1509 transmission (typically all-zeros) and ignored on receipt. Pad1 1510 therefore consists of 1 octet with the most significant 5 bits set 1511 to 0, and with no Sub-Length or Sub-Option Data fields following. 1513 11.1.2. PadN 1515 0 1 2 1516 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 1517 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1518 | S-Type=1| Sub-length=N | N padding octets ... 1519 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1521 Figure 9: PadN 1523 o Sub-Type is set to 1. If multiple instances appear in OMNI 1524 options of the same message all are processed. 1526 o Sub-Length is set to N (from 0 to 2034) that encodes the number of 1527 padding octets that follow. 1529 o Sub-Option Data consists of N octets, set to any value on 1530 transmission (typically all-zeros) and ignored on receipt. 1532 11.1.3. Interface Attributes (Type 1) 1534 The Interface Attributes (Type 1) sub-option provides a basic set of 1535 attributes for underlying interfaces. Interface Attributes (Type 1) 1536 is deprecated throughout the rest of this specification, and 1537 Interface Attributes (Type 2) (see: Section 11.1.4) are indicated 1538 wherever the term "Interface Attributes" appears without an 1539 associated Type designation. 1541 Nodes SHOULD NOT include Interface Attributes (Type 1) sub-options in 1542 IPv6 ND messages they send, and MUST ignore any in IPv6 ND messages 1543 they receive. If an Interface Attributes (Type 1) is included, it 1544 must have the following format: 1546 0 1 2 3 1547 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 1548 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1549 | Sub-Type=2| Sub-length=N | omIndex | omType | 1550 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1551 | Provider ID | Link | Resvd |P00|P01|P02|P03|P04|P05|P06|P07| 1552 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1553 |P08|P09|P10|P11|P12|P13|P14|P15|P16|P17|P18|P19|P20|P21|P22|P23| 1554 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1555 |P24|P25|P26|P27|P28|P29|P30|P31|P32|P33|P34|P35|P36|P37|P38|P39| 1556 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1557 |P40|P41|P42|P43|P44|P45|P46|P47|P48|P49|P50|P51|P52|P53|P54|P55| 1558 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1559 |P56|P57|P58|P59|P60|P61|P62|P63| 1560 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1562 Figure 10: Interface Attributes (Type 1) 1564 o Sub-Type is set to 2. If multiple instances with different 1565 omIndex values appear in OMNI option of the same message all are 1566 processed; if multiple instances with the same omIndex value 1567 appear, the first is processed and all others are ignored 1569 o Sub-Length is set to N (from 4 to 2034) that encodes the number of 1570 Sub-Option Data octets that follow. 1572 o omIndex is a 1-octet field containing a value from 0 to 255 1573 identifying the underlying interface for which the attributes 1574 apply. 1576 o omType is a 1-octet field containing a value from 0 to 255 1577 corresponding to the underlying interface identified by omIndex. 1579 o Provider ID is a 1-octet field containing a value from 0 to 255 1580 corresponding to the underlying interface identified by omIndex. 1582 o Link encodes a 4-bit link metric. The value '0' means the link is 1583 DOWN, and the remaining values mean the link is UP with metric 1584 ranging from '1' ("lowest") to '15' ("highest"). 1586 o Resvd is reserved for future use. Set to 0 on transmission and 1587 ignored on reception. 1589 o A 16-octet ""Preferences" field immediately follows 'Resvd', with 1590 values P[00] through P[63] corresponding to the 64 Differentiated 1591 Service Code Point (DSCP) values [RFC2474]. Each 2-bit P[*] field 1592 is set to the value '0' ("disabled"), '1' ("low"), '2' ("medium") 1593 or '3' ("high") to indicate a QoS preference for underlying 1594 interface selection purposes. 1596 11.1.4. Interface Attributes (Type 2) 1598 The Interface Attributes (Type 2) sub-option provides L2 forwarding 1599 information for the multilink conceptual sending algorithm discussed 1600 in Section 13. The L2 information is used for selecting among 1601 potentially multiple candidate underlying interfaces that can be used 1602 to forward packets to the neighbor based on factors such as DSCP 1603 preferences and link quality. Interface Attributes (Type 2) further 1604 includes link-layer address information to be used for either OAL 1605 encapsulation or direct UDP/IP encapsulation (when OAL encapsulation 1606 can be avoided). 1608 Interface Attributes (Type 2) are the sole Interface Attributes 1609 format in this specification that all OMNI nodes must honor. 1610 Wherever the term "Interface Attributes" occurs throughout this 1611 specification without a "Type" designation, the format given below is 1612 indicated: 1614 0 1 2 3 1615 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 1616 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1617 | S-Type=3| Sub-length=N | omIndex | omType | 1618 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1619 | Provider ID | Link |R| API | SRT | FMT | LHS (0 - 7) | 1620 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1621 | LHS (bits 8 - 31) | ~ 1622 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 1623 ~ ~ 1624 ~ Link Layer Address (L2ADDR) ~ 1625 ~ ~ 1626 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1627 | Bitmap(0)=0xff|P00|P01|P02|P03|P04|P05|P06|P07|P08|P09|P10|P11| 1628 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1629 |P12|P13|P14|P15|P16|P17|P18|P19|P20|P21|P22|P23|P24|P25|P26|P27| 1630 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1631 |P28|P29|P30|P31| Bitmap(1)=0xff|P32|P33|P34|P35|P36| ... 1632 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1634 Figure 11: Interface Attributes (Type 2) 1636 o Sub-Type is set to 3. If multiple instances with different 1637 omIndex values appear in OMNI options of the same message all are 1638 processed; if multiple instances with the same omIndex value 1639 appear, the first is processed and all others are ignored. 1641 o Sub-Length is set to N (from 4 to 2034) that encodes the number of 1642 Sub-Option Data octets that follow. The 'omIndex', 'omType', 1643 'Provider ID', 'Link', 'R' and 'API' fields are always present; 1644 hence, the remainder of the Sub-Option Data is limited to 2030 1645 octets. 1647 o Sub-Option Data contains an "Interface Attributes (Type 2)" option 1648 encoded as follows: 1650 * omIndex is set to an 8-bit integer value corresponding to a 1651 specific underlying interface the same as specified above for 1652 the OMNI option S/T-omIndex field. The OMNI options of a same 1653 message may include multiple Interface Attributes Sub-Options, 1654 with each distinct omIndex value pertaining to a different 1655 underlying interface. The OMNI option will often include an 1656 Interface Attributes Sub-Option with the same omIndex value 1657 that appears in the S/T-omIndex. In that case, the actual 1658 encapsulation address of the received IPv6 ND message should be 1659 compared with the L2ADDR encoded in the Sub-Option (see below); 1660 if the addresses are different (or, if L2ADDR is absent) the 1661 presence of a NAT is assumed. 1663 * omType is set to an 8-bit integer value corresponding to the 1664 underlying interface identified by omIndex. The value 1665 represents an OMNI interface-specific 8-bit mapping for the 1666 actual IANA ifType value registered in the 'IANAifType-MIB' 1667 registry [http://www.iana.org]. 1669 * Provider ID is set to an OMNI interface-specific 8-bit ID value 1670 for the network service provider associated with this omIndex. 1672 * Link encodes a 4-bit link metric. The value '0' means the link 1673 is DOWN, and the remaining values mean the link is UP with 1674 metric ranging from '1' ("lowest") to '15' ("highest"). 1676 * R is reserved for future use. 1678 * API - a 3-bit "Address/Preferences/Indexed" code that 1679 determines the contents of the remainder of the sub-option as 1680 follows: 1682 + When the most significant bit (i.e., "Address") is set to 1, 1683 the SRT, FMT, LHS and L2ADDR fields are included immediately 1684 following the API code; else, they are omitted. 1686 + When the next most significant bit (i.e., "Preferences") is 1687 set to 1, a preferences block is included next; else, it is 1688 omitted. (Note that if "Address" is set the preferences 1689 block immediately follows L2ADDR; else, it immediately 1690 follows the API code.) 1692 + When a preferences block is present and the least 1693 significant bit (i.e., "Indexed") is set to 0, the block is 1694 encoded in "Simplex" form as shown in Figure 10; else it is 1695 encoded in "Indexed" form as discussed below. 1697 * When API indicates that an "Address" is included, the following 1698 fields appear in consecutive order (else, they are omitted): 1700 + SRT - a 5-bit Segment Routing Topology prefix length value 1701 that (when added to 96) determines the prefix length to 1702 apply to the ULA formed from concatenating [ULA*]::/96 with 1703 the 32 bit LHS MSID value that follows. For example, the 1704 value 16 corresponds to the prefix length 112. 1706 + FMT - a 3-bit "Framework/Mode/Type" code corresponding to 1707 the included Link Layer Address as follows: 1709 - When the most significant bit (i.e., "Framework") is set 1710 to 1, L2ADDR is the INET encapsulation address for the 1711 Source/Target Client itself; otherwise L2ADDR is the 1712 address of the Server/Proxy named in the LHS. 1714 - When the next most significant bit (i.e., "Mode") is set 1715 to 1, the Framework node is (likely) located behind an 1716 INET Network Address Translator (NAT); otherwise, it is 1717 on the open INET. 1719 - When the least significant bit (i.e., "Type") is set to 1720 0, L2ADDR includes a UDP Port Number followed by an IPv4 1721 address; otherwise, it includes a UDP Port Number 1722 followed by an IPv6 address. 1724 + LHS - the 32 bit MSID of the Last Hop Server/Proxy on the 1725 path to the target. When SRT and LHS are both set to 0, the 1726 LHS is considered unspecified in this IPv6 ND message. When 1727 SRT is set to 0 and LHS is non-zero, the prefix length is 1728 set to 128. SRT and LHS together provide guidance to the 1729 OMNI interface forwarding algorithm. Specifically, if SRT/ 1730 LHS is located in the local OMNI link segment then the OMNI 1731 interface can encapsulate according to FMT/L2ADDR (following 1732 any necessary NAT traversal messaging); else, it must 1733 forward according to the OMNI link spanning tree. See 1734 [I-D.templin-intarea-6706bis] for further discussion. 1736 + Link Layer Address (L2ADDR) - Formatted according to FMT, 1737 and identifies the link-layer address (i.e., the 1738 encapsulation address) of the source/target. The UDP Port 1739 Number appears in the first 2 octets and the IP address 1740 appears in the next 4 octets for IPv4 or 16 octets for IPv6. 1741 The Port Number and IP address are recorded in network byte 1742 order, and in ones-compliment "obfuscated" form per 1743 [RFC4380]. The OMNI interface forwarding algorithm uses 1744 FMT/L2ADDR to determine the encapsulation address for 1745 forwarding when SRT/LHS is located in the local OMNI link 1746 segment. Note that if the target is behind a NAT, L2ADDR 1747 will contain the mapped INET address stored in the NAT; 1748 otherwise, L2ADDR will contain the native INET information 1749 of the target itself. 1751 * When API indicates that "Preferences" are included, a 1752 preferences block appears as the remainder of the Sub-Option as 1753 a series of Bitmaps and P[*] values. In "Simplex" form, the 1754 index for each singleton Bitmap octet is inferred from its 1755 sequential position (i.e., 0, 1, 2, ...) as shown in Figure 11. 1756 In "Indexed" form, each Bitmap is preceded by an Index octet 1757 that encodes a value "i" = (0 - 255) as the index for its 1758 companion Bitmap as follows: 1760 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1761 | Index=i | Bitmap(i) |P[*] values ... 1762 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 1764 Figure 12 1766 * The preferences consist of a first (simplex/indexed) Bitmap 1767 (i.e., "Bitmap(i)") followed by 0-8 single-octet blocks of 1768 2-bit P[*] values, followed by a second Bitmap (i), followed by 1769 0-8 blocks of P[*] values, etc. Reading from bit 0 to bit 7, 1770 the bits of each Bitmap(i) that are set to '1'' indicate the 1771 P[*] blocks from the range P[(i*32)] through P[(i*32) + 31] 1772 that follow; if any Bitmap(i) bits are '0', then the 1773 corresponding P[*] block is instead omitted. For example, if 1774 Bitmap(0) contains 0xff then the block with P[00]-P[03], 1775 followed by the block with P[04]-P[07], etc., and ending with 1776 the block with P[28]-P[31] are included (as shown in 1777 Figure 10). The next Bitmap(i) is then consulted with its bits 1778 indicating which P[*] blocks follow, etc. out to the end of the 1779 Sub-Option. 1781 * Each 2-bit P[*] field is set to the value '0' ("disabled"), '1' 1782 ("low"), '2' ("medium") or '3' ("high") to indicate a QoS 1783 preference for underlying interface selection purposes. Not 1784 all P[*] values need to be included in the OMNI option of each 1785 IPv6 ND message received. Any P[*] values represented in an 1786 earlier OMNI option but omitted in the current OMNI option 1787 remain unchanged. Any P[*] values not yet represented in any 1788 OMNI option default to "medium". 1790 * The first 16 P[*] blocks correspond to the 64 Differentiated 1791 Service Code Point (DSCP) values P[00] - P[63] [RFC2474]. Any 1792 additional P[*] blocks that follow correspond to "pseudo-DSCP" 1793 traffic classifier values P[64], P[65], P[66], etc. See 1794 Appendix A for further discussion and examples. 1796 11.1.5. Traffic Selector 1798 0 1 2 3 1799 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 1800 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1801 | S-Type=4| Sub-length=N | omIndex | ~ 1802 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 1803 ~ ~ 1804 ~ RFC 6088 Format Traffic Selector ~ 1805 ~ ~ 1806 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1808 Figure 13: Traffic Selector 1810 o Sub-Type is set to 4. If multiple instances appear in OMNI 1811 options of the same message all are processed, i.e., even if the 1812 same omIndex value appears multiple times. 1814 o Sub-Length is set to N (from 1 to 2034) that encodes the number of 1815 Sub-Option Data octets that follow. 1817 o Sub-Option Data contains a 1 octet omIndex encoded exactly as 1818 specified in Section 11.1.3, followed by an N-1 octet traffic 1819 selector formatted per [RFC6088] beginning with the "TS Format" 1820 field. The largest traffic selector for a given omIndex is 1821 therefore 2033 octets. 1823 11.1.6. MS-Register 1824 0 1 2 3 1825 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 1826 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1827 | S-Type=5| Sub-length=4n | MSID[1] (bits 0 - 15) | 1828 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1829 | MSID [1] (bits 16 - 32) | MSID[2] (bits 0 - 15) | 1830 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1831 | MSID [2] (bits 16 - 32) | MSID[3] (bits 0 - 15) | 1832 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1833 ... ... ... ... ... ... 1834 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1835 | MSID [n] (bits 16 - 32) | 1836 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1838 Figure 14: MS-Register Sub-option 1840 o Sub-Type is set to 5. If multiple instances appear in OMNI 1841 options of the same message all are processed. Only the first 1842 MAX_MSID values processed (whether in a single instance or 1843 multiple) are retained and all other MSIDs are ignored. 1845 o Sub-Length is set to 4n, with 508 as the maximum value for n. The 1846 length of the Sub-Option Data section is therefore limited to 2032 1847 octets. 1849 o A list of n 4 octet MSIDs is included in the following 4n octets. 1850 The Anycast MSID value '0' in an RS message MS-Register sub-option 1851 requests the recipient to return the MSID of a nearby MSE in a 1852 corresponding RA response. 1854 11.1.7. MS-Release 1856 0 1 2 3 1857 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 1858 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1859 | S-Type=6| Sub-length=4n | MSID[1] (bits 0 - 15) | 1860 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1861 | MSID [1] (bits 16 - 32) | MSID[2] (bits 0 - 15) | 1862 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1863 | MSID [2] (bits 16 - 32) | MSID[3] (bits 0 - 15) | 1864 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1865 ... ... ... ... ... ... 1866 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1867 | MSID [n] (bits 16 - 32) | 1868 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1870 Figure 15: MS-Release Sub-option 1872 o Sub-Type is set to 6. If multiple instances appear in OMNI 1873 options of the same message all are processed. Only the first 1874 MAX_MSID values processed (whether in a single instance or 1875 multiple) are retained and all other MSIDs are ignored. 1877 o Sub-Length is set to 4n, with 508 as the maximum value for n. The 1878 length of the Sub-Option Data section is therefore limited to 2032 1879 octets. 1881 o A list of n 4 octet MSIDs is included in the following 4n octets. 1882 The Anycast MSID value '0' is ignored in MS-Release sub-options, 1883 i.e., only non-zero values are processed. 1885 11.1.8. Geo Coordinates 1887 0 1 2 3 1888 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 1889 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1890 | S-Type=7| Sub-length=N | Geo Coordinates 1891 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... 1893 Figure 16: Geo Coordinates Sub-option 1895 o Sub-Type is set to 7. If multiple instances appear in OMNI 1896 options of the same message the first is processed and all others 1897 are ignored. 1899 o Sub-Length is set to N (from 0 to 2034) that encodes the number of 1900 Sub-Option Data octets that follow. 1902 o A set of Geo Coordinates of maximum length 2034 octets. Format(s) 1903 to be specified in future documents; should include Latitude/ 1904 Longitude, plus any additional attributes such as altitude, 1905 heading, speed, etc. 1907 11.1.9. Dynamic Host Configuration Protocol for IPv6 (DHCPv6) Message 1909 The Dynamic Host Configuration Protocol for IPv6 (DHCPv6) sub-option 1910 may be included in the OMNI options of RS messages sent by MNs and RA 1911 messages returned by MSEs. ARs that act as proxys to forward RS/RA 1912 messages between MNs and MSEs also forward DHCPv6 sub-options 1913 unchanged and do not process DHCPv6 sub-options themselves. 1915 0 1 2 3 1916 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 1917 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1918 | S-Type=8| Sub-length=N | msg-type | id (octet 0) | 1919 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1920 | transaction-id (octets 1-2) | | 1921 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1922 | | 1923 . DHCPv6 options . 1924 . (variable number and length) . 1925 | | 1926 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1928 Figure 17: DHCPv6 Message Sub-option 1930 o Sub-Type is set to 8. If multiple instances appear in OMNI 1931 options of the same message the first is processed and all others 1932 are ignored. 1934 o Sub-Length is set to N (from 4 to 2034) that encodes the number of 1935 Sub-Option Data octets that follow. The 'msg-type' and 1936 'transaction-id' fields are always present; hence, the length of 1937 the DHCPv6 options is restricted to 2030 octets. 1939 o 'msg-type' and 'transaction-id' are coded according to Section 8 1940 of [RFC8415]. 1942 o A set of DHCPv6 options coded according to Section 21 of [RFC8415] 1943 follows. 1945 11.1.10. Host Identity Protocol (HIP) Message 1947 The Host Identity Protocol (HIP) Message sub-option may be included 1948 in the OMNI options of RS messages sent by MNs and RA messages 1949 returned by ARs. ARs that act as proxys authenticate and remove HIP 1950 messages in RS messages they forward from a MN to an MSE. ARs that 1951 act as proxys insert and sign HIP messages in the RA messages they 1952 forward from an MSE to a MN. 1954 0 1 2 3 1955 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 1956 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1957 | S-Type=9| Sub-length=N |0| Packet Type |Version| RES.|1| 1958 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1959 | Checksum | Controls | 1960 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1961 | Sender's Host Identity Tag (HIT) | 1962 | | 1963 | | 1964 | | 1965 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1966 | Receiver's Host Identity Tag (HIT) | 1967 | | 1968 | | 1969 | | 1970 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1971 | | 1972 / HIP Parameters / 1973 / / 1974 | | 1975 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1977 Figure 18: HIP Message Sub-option 1979 o Sub-Type is set to 9. If multiple instances appear in OMNI 1980 options of the same message the first is processed and all others 1981 are ignored. 1983 o Sub-Length is set to N, i.e., the length of the option in octets 1984 beginning immediately following the Sub-Length field and extending 1985 to the end of the HIP parameters. The length of the entire HIP 1986 message is therefore restricted to 2034 octets. 1988 o The HIP message is coded exactly as specified in Section 5 of 1989 [RFC7401], except that the OMNI "Sub-Type" and "Sub-Length" fields 1990 replace the first 2 octets of the HIP message header (i.e., the 1991 Next Header and Header Length fields). 1993 11.1.11. Node Identification 1994 0 1 2 3 1995 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 1996 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1997 |S-Type=10| Sub-length=N | ID-Type | ~ 1998 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 1999 ~ Node Identification Value (N-1 octets) ~ 2000 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2002 Figure 19: Node Identification 2004 o Sub-Type is set to 10. If multiple instances appear in OMNI 2005 options of the same IPv6 ND message the first instance of a 2006 specific ID-Type is processed and all other instances of the same 2007 ID-Type are ignored. (Note therefore that it is possible for a 2008 single IPv6 ND message to convey multiple Node Identifications - 2009 each having a different ID-Type.) 2011 o Sub-Length is set to N (from 1 to 2034) that encodes the number of 2012 Sub-Option Data octets that follow. The ID-Type field is always 2013 present; hence, the maximum Node Identification Value length is 2014 2033 octets. 2016 o ID-Type is a 1 octet field that encodes the type of the Node 2017 Identification Value. The following ID-Type values are currently 2018 defined: 2020 * 0 - Universally Unique IDentifier (UUID) [RFC4122]. Indicates 2021 that Node Identification Value contains a 16 octet UUID. 2023 * 1 - Host Identity Tag (HIT) [RFC7401]. Indicates that Node 2024 Identification Value contains a 16 octet HIT. 2026 * 2 - Hierarchical HIT (HHIT) [I-D.ietf-drip-rid]. Indicates 2027 that Node Identification Value contains a 16 octet HHIT. 2029 * 3 - Network Access Identifier (NAI) [RFC7542]. Indicates that 2030 Node Identification Value contains an N-1 octet NAI. 2032 * 4 - Fully-Qualified Domain Name (FQDN) [RFC1035]. Indicates 2033 that Node Identification Value contains an N-1 octet FQDN. 2035 * 5 - 252 - Unassigned. 2037 * 253-254 - Reserved for experimentation, as recommended in 2038 [RFC3692]. 2040 * 255 - reserved by IANA. 2042 o Node Identification Value is an (N - 1) octet field encoded 2043 according to the appropriate the "ID-Type" reference above. 2045 When a Node Identification Value is needed for DHCPv6 messaging 2046 purposes, it is encoded as a DHCP Unique IDentifier (DUID) using the 2047 "DUID-EN for OMNI" format with enterprise number 45282 (see: 2048 Section 23) as shown in Figure 20: 2050 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 2051 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2052 | DUID-Type (2) | EN (high bits == 0) | 2053 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2054 | EN (low bits = 45282) | ID-Type | | 2055 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 2056 . Node Identification Value . 2057 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2059 Figure 20: DUID-EN for OMNI Format 2061 In this format, the ID-Type and Node Identification Value fields are 2062 coded exactly as in Figure 19 following the 6 octet DUID-EN header, 2063 and the entire "DUID-EN for OMNI" is included in a DHCPv6 message per 2064 [RFC8415]. 2066 11.1.12. Sub-Type Extension 2068 Since the Sub-Type field is only 5 bits in length, future 2069 specifications of major protocol functions may exhaust the remaining 2070 Sub-Type values available for assignment. This document therefore 2071 defines Sub-Type 30 as an "extension", meaning that the actual sub- 2072 option type is determined by examining a 1 octet "Extension-Type" 2073 field immediately following the Sub-Length field. The Sub-Type 2074 Extension is formatted as shown in Figure 21: 2076 0 1 2 3 2077 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 2078 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2079 |S-Type=30| Sub-length=N | Extension-Type| ~ 2080 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 2081 ~ ~ 2082 ~ Extension-Type Body ~ 2083 ~ ~ 2084 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2086 Figure 21: Sub-Type Extension 2088 o Sub-Type is set to 30. If multiple instances appear in OMNI 2089 options of the same message all are processed, where each 2090 individual extension defines its own policy for processing 2091 multiple of that type. 2093 o Sub-Length is set to N (from 1 to 2034) that encodes the number of 2094 Sub-Option Data octets that follow. The Extension-Type field is 2095 always present; hence, the maximum Extension-Type Body length is 2096 2033 octets. 2098 o Extension-Type contains a 1 octet Sub-Type Extension value between 2099 0 and 255. 2101 o Extension-Type Body contains an N-1 octet block with format 2102 defined by the given extension specification. 2104 Extension-Type values 2 through 252 are available for assignment by 2105 future specifications, which must also define the format of the 2106 Extension-Type Body and its processing rules. Extension-Type values 2107 253 and 254 are reserved for experimentation, as recommended in 2108 [RFC3692], and value 255 is reserved by IANA. Extension-Type values 2109 0 and 1 are defined in the following subsections: 2111 11.1.12.1. RFC4380 UDP/IP Header Option 2113 0 1 2 3 2114 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 2115 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2116 |S-Type=30| Sub-length=N | Ext-Type=0 | Header Type | 2117 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2118 ~ Header Option Value ~ 2119 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2121 Figure 22: RFC4380 UDP/IP Header Option (Extension-Type 0) 2123 o Sub-Type is set to 30. 2125 o Sub-Length is set to N (from 2 to 2034) that encodes the number of 2126 Sub-Option Data octets that follow. The Extension-Type and Header 2127 Type fields are always present; hence, the maximum-length Header 2128 Option Value is 2032 octets. 2130 o Extension-Type is set to 0. Each instance encodes exactly one 2131 header option per Section 5.1.1 of [RFC4380], with the leading '0' 2132 octet omitted and the following octet coded as Header Type. If 2133 multiple instances of the same Header Type appear in OMNI options 2134 of the same message the first instance is processed and all others 2135 are ignored. 2137 o Header Type and Header Option Value are coded exactly as specified 2138 in Section 5.1.1 of [RFC4380]; the following types are currently 2139 defined: 2141 * 0 - Origin Indication (IPv4) - value coded per Section 5.1.1 of 2142 [RFC4380]. 2144 * 1 - Authentication Encapsulation - value coded per 2145 Section 5.1.1 of [RFC4380]. 2147 * 2 - Origin Indication (IPv6) - value coded per Section 5.1.1 of 2148 [RFC4380], except that the address is a 16-octet IPv6 address 2149 instead of a 4-octet IPv4 address. 2151 o Header Type values 3 through 252 are available for assignment by 2152 future specifications, which must also define the format of the 2153 Header Option Value and its processing rules. Header Type values 2154 253 and 254 are reserved for experimentation, as recommended in 2155 [RFC3692], and value 255 is Reserved by IANA. 2157 11.1.12.2. RFC6081 UDP/IP Trailer Option 2159 0 1 2 3 2160 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 2161 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2162 |S-Type=30| Sub-length=N | Ext-Type=1 | Trailer Type | 2163 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2164 ~ Trailer Option Value ~ 2165 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2167 Figure 23: RFC6081 UDP/IP Trailer Option (Extension-Type 1) 2169 o Sub-Type is set to 30. 2171 o Sub-Length is set to N (from 2 to 2034) that encodes the number of 2172 Sub-Option Data octets that follow. The Extension-Type and 2173 Trailer Type fields are always present; hence, the maximum-length 2174 Trailer Option Value is 2032 octets. 2176 o Extension-Type is set to 1. Each instance encodes exactly one 2177 trailer option per Section 4 of [RFC6081]. If multiple instances 2178 of the same trailer type appear in OMNI options of the same 2179 message the first instance is processed and all others ignored. 2181 o Trailer Type and Trailer Option Value are coded exactly as 2182 specified in Section 4 of [RFC6081]; the following Trailer Types 2183 are currently defined: 2185 * 0 - Unassigned 2187 * 1 - Nonce Trailer - value coded per Section 4.2 of [RFC6081]. 2189 * 2 - Unassigned 2191 * 3 - Alternate Address Trailer (IPv4) - value coded per 2192 Section 4.3 of [RFC6081]. 2194 * 4 - Neighbor Discovery Option Trailer - value coded per 2195 Section 4.4 of [RFC6081]. 2197 * 5 - Random Port Trailer - value coded per Section 4.5 of 2198 [RFC6081]. 2200 * 6 - Alternate Address Trailer (IPv6) - value coded per 2201 Section 4.3 of [RFC6081], except that each address is a 2202 16-octet IPv6 address instead of a 4-octet IPv4 address. 2204 o Trailer Type values 7 through 252 are available for assignment by 2205 future specifications, which must also define the format of the 2206 Trailer Option Value and its processing rules. Trailer Type 2207 values 253 and 254 are reserved for experimentation, as 2208 recommended in [RFC3692], and value 255 is Reserved by IANA. 2210 12. Address Mapping - Multicast 2212 The multicast address mapping of the native underlying interface 2213 applies. The mobile router on board the MN also serves as an IGMP/ 2214 MLD Proxy for its EUNs and/or hosted applications per [RFC4605] while 2215 using the L2 address of the AR as the L2 address for all multicast 2216 packets. 2218 The MN uses Multicast Listener Discovery (MLDv2) [RFC3810] to 2219 coordinate with the AR, and *NET L2 elements use MLD snooping 2220 [RFC4541]. 2222 13. Multilink Conceptual Sending Algorithm 2224 The MN's IPv6 layer selects the outbound OMNI interface according to 2225 SBM considerations when forwarding data packets from local or EUN 2226 applications to external correspondents. Each OMNI interface 2227 maintains a neighbor cache the same as for any IPv6 interface, but 2228 with additional state for multilink coordination. Each OMNI 2229 interface maintains default routes via ARs discovered as discussed in 2230 Section 14, and may configure more-specific routes discovered through 2231 means outside the scope of this specification. 2233 After a packet enters the OMNI interface, one or more outbound 2234 underlying interfaces are selected based on PBM traffic attributes, 2235 and one or more neighbor underlying interfaces are selected based on 2236 the receipt of Interface Attributes sub-options in IPv6 ND messages 2237 (see: Figure 10). Underlying interface selection for the nodes own 2238 local interfaces are based on attributes such as DSCP, application 2239 port number, cost, performance, message size, etc. OMNI interface 2240 multilink selections could also be configured to perform replication 2241 across multiple underlying interfaces for increased reliability at 2242 the expense of packet duplication. The set of all Interface 2243 Attributes received in IPv6 ND messages determines the multilink 2244 forwarding profile for selecting the neighbor's underlying 2245 interfaces. 2247 When the OMNI interface sends a packet over a selected outbound 2248 underlying interface, the OAL includes or omits a mid-layer 2249 encapsulation header as necessary as discussed in Section 5 and as 2250 determined by the L2 address information received in Interface 2251 Attributes. The OAL also performs encapsulation when the nearest AR 2252 is located multiple hops away as discussed in Section 14.1. (Note 2253 that the OAL MAY employ packing when multiple packets are available 2254 for forwarding to the same destination.) 2256 OMNI interface multilink service designers MUST observe the BCP 2257 guidance in Section 15 [RFC3819] in terms of implications for 2258 reordering when packets from the same flow may be spread across 2259 multiple underlying interfaces having diverse properties. 2261 13.1. Multiple OMNI Interfaces 2263 MNs may connect to multiple independent OMNI links concurrently in 2264 support of SBM. Each OMNI interface is distinguished by its Anycast 2265 ULA (e.g., [ULA]:0002::, [ULA]:1000::, [ULA]:7345::, etc.). The MN 2266 configures a separate OMNI interface for each link so that multiple 2267 interfaces (e.g., omni0, omni1, omni2, etc.) are exposed to the IPv6 2268 layer. A different Anycast ULA is assigned to each interface, and 2269 the MN injects the service prefixes for the OMNI link instances into 2270 the EUN routing system. 2272 Applications in EUNs can use Segment Routing to select the desired 2273 OMNI interface based on SBM considerations. The Anycast ULA is 2274 written into the IPv6 destination address, and the actual destination 2275 (along with any additional intermediate hops) is written into the 2276 Segment Routing Header. Standard IP routing directs the packets to 2277 the MN's mobile router entity, and the Anycast ULA identifies the 2278 OMNI interface to be used for transmission to the next hop. When the 2279 MN receives the message, it replaces the IPv6 destination address 2280 with the next hop found in the routing header and transmits the 2281 message over the OMNI interface identified by the Anycast ULA. 2283 Multiple distinct OMNI links can therefore be used to support fault 2284 tolerance, load balancing, reliability, etc. The architectural model 2285 is similar to Layer 2 Virtual Local Area Networks (VLANs). 2287 13.2. MN<->AR Traffic Loop Prevention 2289 After an AR has registered an MNP for a MN (see: Section 14), the AR 2290 will forward packets destined to an address within the MNP to the MN. 2291 The MN will under normal circumstances then forward the packet to the 2292 correct destination within its internal networks. 2294 If at some later time the MN loses state (e.g., after a reboot), it 2295 may begin returning packets destined to an MNP address to the AR as 2296 its default router. The AR therefore must drop any packets 2297 originating from the MN and destined to an address within the MN's 2298 registered MNP. To do so, the AR institutes the following check: 2300 o if the IP destination address belongs to a neighbor on the same 2301 OMNI interface, and if the link-layer source address is the same 2302 as one of the neighbor's link-layer addresses, drop the packet. 2304 14. Router Discovery and Prefix Registration 2306 MNs interface with the MS by sending RS messages with OMNI options 2307 under the assumption that one or more AR on the *NET will process the 2308 message and respond. The MN then configures default routes for the 2309 OMNI interface via the discovered ARs as the next hop. The manner in 2310 which the *NET ensures AR coordination is link-specific and outside 2311 the scope of this document (however, considerations for *NETs that do 2312 not provide ARs that recognize the OMNI option are discussed in 2313 Section 19). 2315 For each underlying interface, the MN sends an RS message with an 2316 OMNI option to coordinate with MSEs identified by MSID values. 2317 Example MSID discovery methods are given in [RFC5214] and include 2318 data link login parameters, name service lookups, static 2319 configuration, a static "hosts" file, etc. The MN can also send an 2320 RS with an MS-Register sub-option that includes the Anycast MSID 2321 value '0', i.e., instead of or in addition to any non-zero MSIDs. 2322 When the AR receives an RS with a MSID '0', it selects a nearby MSE 2323 (which may be itself) and returns an RA with the selected MSID in an 2324 MS-Register sub-option. The AR selects only a single wildcard MSE 2325 (i.e., even if the RS MS-Register sub-option included multiple '0' 2326 MSIDs) while also soliciting the MSEs corresponding to any non-zero 2327 MSIDs. 2329 MNs configure OMNI interfaces that observe the properties discussed 2330 in the previous section. The OMNI interface and its underlying 2331 interfaces are said to be in either the "UP" or "DOWN" state 2332 according to administrative actions in conjunction with the interface 2333 connectivity status. An OMNI interface transitions to UP or DOWN 2334 through administrative action and/or through state transitions of the 2335 underlying interfaces. When a first underlying interface transitions 2336 to UP, the OMNI interface also transitions to UP. When all 2337 underlying interfaces transition to DOWN, the OMNI interface also 2338 transitions to DOWN. 2340 When an OMNI interface transitions to UP, the MN sends RS messages to 2341 register its MNP and an initial set of underlying interfaces that are 2342 also UP. The MN sends additional RS messages to refresh lifetimes 2343 and to register/deregister underlying interfaces as they transition 2344 to UP or DOWN. The MN's OMNI interface sends initial RS messages 2345 over an UP underlying interface with its MNP-LLA as the source and 2346 with destination set to link-scoped All-Routers multicast (ff02::2) 2347 [RFC4291]. The OMNI interface includes an OMNI option per Section 11 2348 with a Preflen assertion, Interface Attributes appropriate for 2349 underlying interfaces, MS-Register/Release sub-options containing 2350 MSID values, and with any other necessary OMNI sub-options (e.g., a 2351 Node Identification sub-option as an identity for the MN). The OMNI 2352 interface then sets the S/T-omIndex field to the index of the 2353 underlying interface over which the RS message is sent. The OMNI 2354 interface then sends the RS over the underlying interface, using OAL 2355 encapsulation and fragmentation if necessary. If OAL encapsulation 2356 is used, the OMNI interface sets the OAL source address to the ULA 2357 corresponding to the RS source and sets the OAL destination to site- 2358 scoped All-Routers multicast (ff05::2). 2360 ARs process IPv6 ND messages with OMNI options and act as an MSE 2361 themselves and/or as a proxy for other MSEs. ARs receive RS messages 2362 (while performing OAL reassembly if necessary) and create a neighbor 2363 cache entry for the MN, then coordinate with any MSEs named in the 2364 Register/Release lists in a manner outside the scope of this 2365 document. When an MSE processes the OMNI information, it first 2366 validates the prefix registration information then injects/withdraws 2367 the MNP in the routing/mapping system and caches/discards the new 2368 Preflen, MNP and Interface Attributes. The MSE then informs the AR 2369 of registration success/failure, and the AR returns an RA message to 2370 the MN with an OMNI option per Section 11. 2372 The AR's OMNI interface returns the RA message via the same 2373 underlying interface of the MN over which the RS was received, and 2374 with destination address set to the MNP-LLA (i.e., unicast), with 2375 source address set to its own LLA, and with an OMNI option with S/ 2376 T-omIndex set to the value included in the RS. The OMNI option also 2377 includes a Preflen confirmation, Interface Attributes, MS-Register/ 2378 Release and any other necessary OMNI sub-options (e.g., a Node 2379 Identification sub-option as an identity for the AR). The RA also 2380 includes any information for the link, including RA Cur Hop Limit, M 2381 and O flags, Router Lifetime, Reachable Time and Retrans Timer 2382 values, and includes any necessary options such as: 2384 o PIOs with (A; L=0) that include MSPs for the link [RFC8028]. 2386 o RIOs [RFC4191] with more-specific routes. 2388 o an MTU option that specifies the maximum acceptable packet size 2389 for this underlying interface. 2391 The OMNI interface then sends the RA, using OAL encapsulation and 2392 fragmentation if necessary. If OAL encapsulation is used, the OMNI 2393 interface sets the OAL source address to the ULA corresponding to the 2394 RA source and sets the OAL destination to the ULA corresponding to 2395 the RA destination. The AR MAY also send periodic and/or event- 2396 driven unsolicited RA messages per [RFC4861]. In that case, the S/ 2397 T-omIndex field in the OMNI option of the unsolicited RA message 2398 identifies the target underlying interface of the destination MN. 2400 The AR can combine the information from multiple MSEs into one or 2401 more "aggregate" RAs sent to the MN in order conserve *NET bandwidth. 2402 Each aggregate RA includes an OMNI option with MS-Register/Release 2403 sub-options with the MSEs represented by the aggregate. If an 2404 aggregate is sent, the RA message contents must consistently 2405 represent the combined information advertised by all represented 2406 MSEs. Note that since the AR uses its own ADM-LLA as the RA source 2407 address, the MN determines the addresses of the represented MSEs by 2408 examining the MS-Register/Release OMNI sub-options. 2410 When the MN receives the RA message, it creates an OMNI interface 2411 neighbor cache entry for each MSID that has confirmed MNP 2412 registration via the L2 address of this AR. If the MN connects to 2413 multiple *NETs, it records the additional L2 AR addresses in each 2414 MSID neighbor cache entry (i.e., as multilink neighbors). The MN 2415 then configures a default route via the MSE that returned the RA 2416 message, and assigns the Subnet Router Anycast address corresponding 2417 to the MNP (e.g., 2001:db8:1:2::) to the OMNI interface. The MN then 2418 manages its underlying interfaces according to their states as 2419 follows: 2421 o When an underlying interface transitions to UP, the MN sends an RS 2422 over the underlying interface with an OMNI option. The OMNI 2423 option contains at least one Interface Attribute sub-option with 2424 values specific to this underlying interface, and may contain 2425 additional Interface Attributes specific to other underlying 2426 interfaces. The option also includes any MS-Register/Release sub- 2427 options. 2429 o When an underlying interface transitions to DOWN, the MN sends an 2430 RS or unsolicited NA message over any UP underlying interface with 2431 an OMNI option containing an Interface Attribute sub-option for 2432 the DOWN underlying interface with Link set to '0'. The MN sends 2433 an RS when an acknowledgement is required, or an unsolicited NA 2434 when reliability is not thought to be a concern (e.g., if 2435 redundant transmissions are sent on multiple underlying 2436 interfaces). 2438 o When the Router Lifetime for a specific AR nears expiration, the 2439 MN sends an RS over the underlying interface to receive a fresh 2440 RA. If no RA is received, the MN can send RS messages to an 2441 alternate MSID in case the current MSID has failed. If no RS 2442 messages are received even after trying to contact alternate 2443 MSIDs, the MN marks the underlying interface as DOWN. 2445 o When a MN wishes to release from one or more current MSIDs, it 2446 sends an RS or unsolicited NA message over any UP underlying 2447 interfaces with an OMNI option with a Release MSID. Each MSID 2448 then withdraws the MNP from the routing/mapping system and informs 2449 the AR that the release was successful. 2451 o When all of a MNs underlying interfaces have transitioned to DOWN 2452 (or if the prefix registration lifetime expires), any associated 2453 MSEs withdraw the MNP the same as if they had received a message 2454 with a release indication. 2456 The MN is responsible for retrying each RS exchange up to 2457 MAX_RTR_SOLICITATIONS times separated by RTR_SOLICITATION_INTERVAL 2458 seconds until an RA is received. If no RA is received over an UP 2459 underlying interface (i.e., even after attempting to contact 2460 alternate MSEs), the MN declares this underlying interface as DOWN. 2462 The IPv6 layer sees the OMNI interface as an ordinary IPv6 interface. 2463 Therefore, when the IPv6 layer sends an RS message the OMNI interface 2464 returns an internally-generated RA message as though the message 2465 originated from an IPv6 router. The internally-generated RA message 2466 contains configuration information that is consistent with the 2467 information received from the RAs generated by the MS. Whether the 2468 OMNI interface IPv6 ND messaging process is initiated from the 2469 receipt of an RS message from the IPv6 layer is an implementation 2470 matter. Some implementations may elect to defer the IPv6 ND 2471 messaging process until an RS is received from the IPv6 layer, while 2472 others may elect to initiate the process proactively. Still other 2473 deployments may elect to administratively disable the ordinary RS/RA 2474 messaging used by the IPv6 layer over the OMNI interface, since they 2475 are not required to drive the internal RS/RA processing. (Note that 2476 this same logic applies to IPv4 implementations that employ ICMP- 2477 based Router Discovery per [RFC1256].) 2479 Note: The Router Lifetime value in RA messages indicates the time 2480 before which the MN must send another RS message over this underlying 2481 interface (e.g., 600 seconds), however that timescale may be 2482 significantly longer than the lifetime the MS has committed to retain 2483 the prefix registration (e.g., REACHABLETIME seconds). ARs are 2484 therefore responsible for keeping MS state alive on a shorter 2485 timescale than the MN is required to do on its own behalf. 2487 Note: On multicast-capable underlying interfaces, MNs should send 2488 periodic unsolicited multicast NA messages and ARs should send 2489 periodic unsolicited multicast RA messages as "beacons" that can be 2490 heard by other nodes on the link. If a node fails to receive a 2491 beacon after a timeout value specific to the link, it can initiate a 2492 unicast exchange to test reachability. 2494 Note: if an AR acting as a proxy forwards a MN's RS message to 2495 another node acting as an MSE using UDP/IP encapsulation, it must use 2496 a distinct UDP source port number for each MN. This allows the MSE 2497 to distinguish different MNs behind the same AR at the link-layer, 2498 whereas the link-layer addresses would otherwise be 2499 indistinguishable. 2501 Note: when an AR acting as an MSE returns an RA to an INET Client, it 2502 includes an OMNI option with an Interface Attributes sub-option with 2503 omIndex set to 0 and with SRT, FMT, LHS and L2ADDR information for 2504 its INET interface. This provides the Client with partition prefix 2505 context regarding the local OMNI link segment. 2507 14.1. Router Discovery in IP Multihop and IPv4-Only Networks 2509 On some *NETs, a MN may be located multiple IP hops away from the 2510 nearest AR. Forwarding through IP multihop *NETs is conducted 2511 through the application of a routing protocol (e.g., a MANET/VANET 2512 routing protocol over omni-directional wireless interfaces, an inter- 2513 domain routing protocol in an enterprise network, etc.). These *NETs 2514 could be either IPv6-enabled or IPv4-only, while IPv4-only *NETs 2515 could be either multicast-capable or unicast-only (note that for 2516 IPv4-only *NETs the following procedures apply for both single-hop 2517 and multihop cases). 2519 A MN located potentially multiple *NET hops away from the nearest AR 2520 prepares an RS message with source address set to its MNP-LLA (or to 2521 the unspecified address (::) if it does not yet have an MNP-LLA), and 2522 with destination set to link-scoped All-Routers multicast the same as 2523 discussed above. If OAL encapsulation and fragmentation are 2524 necessary, the OMNI interface sets the OAL source address to the ULA 2525 corresponding to the RS source (or to a Temporary ULA if the RS 2526 source was the unspecified address (::)) and sets the OAL destination 2527 to site-scoped All-Routers multicast (ff05::2). For IPv6-enabled 2528 *NETs, the MN then encapsulates the message in UDP/IPv6 headers with 2529 source address set to the underlying interface address (or to the ULA 2530 that would be used for OAL encapsulation if the underlying interface 2531 does not yet have an address) and sets the destination to either a 2532 unicast or anycast address of an AR. For IPv4-only *NETs, the MN 2533 instead encapsulates the RS message in an IPv4 header with source 2534 address set to the IPv4 address of the underlying interface and with 2535 destination address set to either the unicast IPv4 address of an AR 2536 [RFC5214] or an IPv4 anycast address reserved for OMNI. The MN then 2537 sends the encapsulated RS message via the *NET interface, where it 2538 will be forwarded by zero or more intermediate *NET hops. 2540 When an intermediate *NET hop that participates in the routing 2541 protocol receives the encapsulated RS, it forwards the message 2542 according to its routing tables (note that an intermediate node could 2543 be a fixed infrastructure element or another MN). This process 2544 repeats iteratively until the RS message is received by a penultimate 2545 *NET hop within single-hop communications range of an AR, which 2546 forwards the message to the AR. 2548 When the AR receives the message, it decapsulates the RS (while 2549 performing OAL reassembly, if necessary) and coordinates with the MS 2550 the same as for an ordinary link-local RS, since the inner Hop Limit 2551 will not have been decremented by the multihop forwarding process. 2552 The AR then prepares an RA message with source address set to its own 2553 ADM-LLA and destination address set to the LLA of the original MN. 2554 The AR then performs OAL encapsulation and fragmentation if 2555 necessary, with OAL source set to its own ADM-ULA and destination set 2556 to the ULA corresponding to the RA source. The AR then encapsulates 2557 the message in an IPv4/IPv6 header with source address set to its own 2558 IPv4/ULA address and with destination set to the encapsulation source 2559 of the RS. 2561 The AR then forwards the message to an *NET node within 2562 communications range, which forwards the message according to its 2563 routing tables to an intermediate node. The multihop forwarding 2564 process within the *NET continues repetitively until the message is 2565 delivered to the original MN, which decapsulates the message and 2566 performs autoconfiguration the same as if it had received the RA 2567 directly from the AR as an on-link neighbor. 2569 Note: An alternate approach to multihop forwarding via IPv6 2570 encapsulation would be for the MN and AR to statelessly translate the 2571 IPv6 LLAs into ULAs and forward the RS/RA messages without 2572 encapsulation. This would violate the [RFC4861] requirement that 2573 certain IPv6 ND messages must use link-local addresses and must not 2574 be accepted if received with Hop Limit less than 255. This document 2575 therefore mandates encapsulation since the overhead is nominal 2576 considering the infrequent nature and small size of IPv6 ND messages. 2577 Future documents may consider encapsulation avoidance through 2578 translation while updating [RFC4861]. 2580 Note: An alternate approach to multihop forwarding via IPv4 2581 encapsulation would be to employ IPv6/IPv4 protocol translation. 2582 However, for IPv6 ND messages the LLAs would be truncated due to 2583 translation and the OMNI Router and Prefix Discovery services would 2584 not be able to function. The use of IPv4 encapsulation is therefore 2585 indicated. 2587 Note: An IPv4 anycast address for OMNI in IPv4 networks could be part 2588 of a new IPv4 /24 prefix allocation, but this may be difficult to 2589 obtain given IPv4 address exhaustion. An alternative would be to re- 2590 purpose the prefix 192.88.99.0 which has been set aside from its 2591 former use by [RFC7526]. 2593 14.2. MS-Register and MS-Release List Processing 2595 OMNI links maintain a constant value "MAX_MSID" selected to provide 2596 MNs with an acceptable level of MSE redundancy while minimizing 2597 control message amplification. It is RECOMMENDED that MAX_MSID be 2598 set to the default value 5; if a different value is chosen, it should 2599 be set uniformly by all nodes on the OMNI link. 2601 When a MN sends an RS message with an OMNI option via an underlying 2602 interface to an AR, the MN must convey its knowledge of its 2603 currently-associated MSEs. Initially, the MN will have no associated 2604 MSEs and should therefore include an MS-Register sub-option with the 2605 single "anycast" MSID value 0 which requests the AR to select and 2606 assign an MSE. The AR will then return an RA message with source 2607 address set to the ADM-LLA of the selected MSE. 2609 As the MN activates additional underlying interfaces, it can 2610 optionally include an MS-Register sub-option with MSID value 0, or 2611 with non-zero MSIDs for MSEs discovered from previous RS/RA 2612 exchanges. The MN will thus eventually begin to learn and manage its 2613 currently active set of MSEs, and can register with new MSEs or 2614 release from former MSEs with each successive RS/RA exchange. As the 2615 MN's MSE constituency grows, it alone is responsible for including or 2616 omitting MSIDs in the MS-Register/Release lists it sends in RS 2617 messages. The inclusion or omission of MSIDs determines the MN's 2618 interface to the MS and defines the manner in which MSEs will 2619 respond. The only limiting factor is that the MN should include no 2620 more than MAX_MSID values in each list per each IPv6 ND message, and 2621 should avoid duplication of entries in each list unless it wants to 2622 increase likelihood of control message delivery. 2624 When an AR receives an RS message sent by a MN with an OMNI option, 2625 the option will contain zero or more MS-Register and MS-Release sub- 2626 options containing MSIDs. After processing the OMNI option, the AR 2627 will have a list of zero or more MS-Register MSIDs and a list of zero 2628 or more of MS-Release MSIDs. The AR then processes the lists as 2629 follows: 2631 o For each list, retain the first MAX_MSID values in the list and 2632 discard any additional MSIDs (i.e., even if there are duplicates 2633 within a list). 2635 o Next, for each MSID in the MS-Register list, remove all matching 2636 MSIDs from the MS-Release list. 2638 o Next, proceed according to whether the AR's own MSID or the value 2639 0 appears in the MS-Register list as follows: 2641 * If yes, send an RA message directly back to the MN and send a 2642 proxy copy of the RS message to each additional MSID in the MS- 2643 Register list with the MS-Register/Release lists omitted. 2644 Then, send an unsolicited NA (uNA) message to each MSID in the 2645 MS-Release list with the MS-Register/Release lists omitted and 2646 with an OMNI option with S/T-omIndex set to 0. 2648 * If no, send a proxy copy of the RS message to each additional 2649 MSID in the MS-Register list with the MS-Register list omitted. 2650 For the first MSID, include the original MS-Release list; for 2651 all other MSIDs, omit the MS-Release list. 2653 Each proxy copy of the RS message will include an OMNI option and OAL 2654 encapsulation header with the ADM-ULA of the AR as the source and the 2655 ADM-ULA of the Register MSE as the destination. When the Register 2656 MSE receives the proxy RS message, if the message includes an MS- 2657 Release list the MSE sends a uNA message to each additional MSID in 2658 the Release list with an OMNI option with S/T-omIndex set to 0. The 2659 Register MSE then sends an RA message back to the (Proxy) AR wrapped 2660 in an OAL encapsulation header with source and destination addresses 2661 reversed, and with RA destination set to the MNP-LLA of the MN. When 2662 the AR receives this RA message, it sends a proxy copy of the RA to 2663 the MN. 2665 Each uNA message (whether sent by the first-hop AR or by a Register 2666 MSE) will include an OMNI option and an OAL encapsulation header with 2667 the ADM-ULA of the Register MSE as the source and the ADM-ULA of the 2668 Release MSE as the destination. The uNA informs the Release MSE that 2669 its previous relationship with the MN has been released and that the 2670 source of the uNA message is now registered. The Release MSE must 2671 then note that the subject MN of the uNA message is now "departed", 2672 and forward any subsequent packets destined to the MN to the Register 2673 MSE. 2675 Note that it is not an error for the MS-Register/Release lists to 2676 include duplicate entries. If duplicates occur within a list, the AR 2677 will generate multiple proxy RS and/or uNA messages - one for each 2678 copy of the duplicate entries. 2680 14.3. DHCPv6-based Prefix Registration 2682 When a MN is not pre-provisioned with an MNP-LLA (or, when the MN 2683 requires additional MNP delegations), it requests the MSE to select 2684 MNPs on its behalf and set up the correct routing state within the 2685 MS. The DHCPv6 service [RFC8415] supports this requirement. 2687 When an MN needs to have the MSE select MNPs, it sends an RS message 2688 with source set to the unspecified address (::) if it has no 2689 MNP_LLAs. If the MN requires only a single MNP delegation, it can 2690 then include a Node Identification sub-option in the OMNI option and 2691 set Preflen to the length of the desired MNP. If the MN requires 2692 multiple MNP delegations and/or more complex DHCPv6 services, it 2693 instead includes a DHCPv6 Message sub-option containing a Client 2694 Identifier, one or more IA_PD options and a Rapid Commit option then 2695 sets the 'msg-type' field to "Solicit", and includes a 3 octet 2696 'transaction-id'. The MN then sets the RS destination to All-Routers 2697 multicast and sends the message using OAL encapsulation and 2698 fragmentation if necessary as discussed above. 2700 When the MSE receives the RS message, it performs OAL reassembly if 2701 necessary. Next, if the RS source is the unspecified address (::) 2702 and/or the OMNI option includes a DHCPv6 message sub-option, the MSE 2703 acts as a "Proxy DHCPv6 Client" in a message exchange with the 2704 locally-resident DHCPv6 server. If the RS did not contain a DHCPv6 2705 message sub-option, the MSE generates a DHCPv6 Solicit message on 2706 behalf of the MN using an IA_PD option with the prefix length set to 2707 the OMNI header Preflen value and with a Client Identifier formed 2708 from the OMNI option Node Identification sub-option; otherwise, the 2709 MSE uses the DHCPv6 Solicit message contained in the OMNI option. 2710 The MSE then sends the DHCPv6 message to the DHCPv6 Server, which 2711 delegates MNPs and returns a DHCPv6 Reply message with PD parameters. 2712 (If the MSE wishes to defer creation of MN state until the DHCPv6 2713 Reply is received, it can instead act as a Lightweight DHCPv6 Relay 2714 Agent per [RFC6221] by encapsulating the DHCPv6 message in a Relay- 2715 forward/reply exchange with Relay Message and Interface ID options. 2716 In the process, the MSE packs any state information needed to return 2717 an RA to the MN in the Relay-forward Interface ID option so that the 2718 information will be echoed back in the Relay-reply.) 2720 When the MSE receives the DHCPv6 Reply, it adds routes to the routing 2721 system and creates MNP-LLAs based on the delegated MNPs. The MSE 2722 then sends an RA back to the MN with the DHCPv6 Reply message 2723 included in an OMNI DHCPv6 message sub-option if and only if the RS 2724 message had included an explicit DHCPv6 Solicit. If the RS message 2725 source was the unspecified address (::), the MSE includes one of the 2726 (newly-created) MNP-LLAs as the RA destination address and sets the 2727 OMNI option Preflen accordingly; otherwise, the MSE includes the RS 2728 source address as the RA destination address. The MSE then sets the 2729 RA source address to its own ADM-LLA then performs OAL encapsulation 2730 and fragmentation if necessary and sends the RA to the MN. When the 2731 MN receives the RA, it reassembles and discards the OAL encapsulation 2732 if necessary, then creates a default route, assigns Subnet Router 2733 Anycast addresses and uses the RA destination address as its primary 2734 MNP-LLA. The MN will then use this primary MNP-LLA as the source 2735 address of any IPv6 ND messages it sends as long as it retains 2736 ownership of the MNP. 2738 Note: After a MN performs a DHCPv6-based prefix registration exchange 2739 with a first MSE, it would need to repeat the exchange with each 2740 additional MSE it registers with. In that case, the MN supplies the 2741 MNP delegation information received from the first MSE when it 2742 engages the additional MSEs. 2744 15. Secure Redirection 2746 If the *NET link model is multiple access, the AR is responsible for 2747 assuring that address duplication cannot corrupt the neighbor caches 2748 of other nodes on the link. When the MN sends an RS message on a 2749 multiple access *NET link, the AR verifies that the MN is authorized 2750 to use the address and returns an RA with a non-zero Router Lifetime 2751 only if the MN is authorized. 2753 After verifying MN authorization and returning an RA, the AR MAY 2754 return IPv6 ND Redirect messages to direct MNs located on the same 2755 *NET link to exchange packets directly without transiting the AR. In 2756 that case, the MNs can exchange packets according to their unicast L2 2757 addresses discovered from the Redirect message instead of using the 2758 dogleg path through the AR. In some *NET links, however, such direct 2759 communications may be undesirable and continued use of the dogleg 2760 path through the AR may provide better performance. In that case, 2761 the AR can refrain from sending Redirects, and/or MNs can ignore 2762 them. 2764 16. AR and MSE Resilience 2766 *NETs SHOULD deploy ARs in Virtual Router Redundancy Protocol (VRRP) 2767 [RFC5798] configurations so that service continuity is maintained 2768 even if one or more ARs fail. Using VRRP, the MN is unaware which of 2769 the (redundant) ARs is currently providing service, and any service 2770 discontinuity will be limited to the failover time supported by VRRP. 2771 Widely deployed public domain implementations of VRRP are available. 2773 MSEs SHOULD use high availability clustering services so that 2774 multiple redundant systems can provide coordinated response to 2775 failures. As with VRRP, widely deployed public domain 2776 implementations of high availability clustering services are 2777 available. Note that special-purpose and expensive dedicated 2778 hardware is not necessary, and public domain implementations can be 2779 used even between lightweight virtual machines in cloud deployments. 2781 17. Detecting and Responding to MSE Failures 2783 In environments where fast recovery from MSE failure is required, ARs 2784 SHOULD use proactive Neighbor Unreachability Detection (NUD) in a 2785 manner that parallels Bidirectional Forwarding Detection (BFD) 2786 [RFC5880] to track MSE reachability. ARs can then quickly detect and 2787 react to failures so that cached information is re-established 2788 through alternate paths. Proactive NUD control messaging is carried 2789 only over well-connected ground domain networks (i.e., and not low- 2790 end *NET links such as aeronautical radios) and can therefore be 2791 tuned for rapid response. 2793 ARs perform proactive NUD for MSEs for which there are currently 2794 active MNs on the *NET. If an MSE fails, ARs can quickly inform MNs 2795 of the outage by sending multicast RA messages on the *NET interface. 2796 The AR sends RA messages to MNs via the *NET interface with an OMNI 2797 option with a Release ID for the failed MSE, and with destination 2798 address set to All-Nodes multicast (ff02::1) [RFC4291]. 2800 The AR SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA messages separated 2801 by small delays [RFC4861]. Any MNs on the *NET interface that have 2802 been using the (now defunct) MSE will receive the RA messages and 2803 associate with a new MSE. 2805 18. Transition Considerations 2807 When a MN connects to an *NET link for the first time, it sends an RS 2808 message with an OMNI option. If the first hop AR recognizes the 2809 option, it returns an RA with its ADM-LLA as the source, the MNP-LLA 2810 as the destination and with an OMNI option included. The MN then 2811 engages the AR according to the OMNI link model specified above. If 2812 the first hop AR is a legacy IPv6 router, however, it instead returns 2813 an RA message with no OMNI option and with a non-OMNI unicast source 2814 LLA as specified in [RFC4861]. In that case, the MN engages the *NET 2815 according to the legacy IPv6 link model and without the OMNI 2816 extensions specified in this document. 2818 If the *NET link model is multiple access, there must be assurance 2819 that address duplication cannot corrupt the neighbor caches of other 2820 nodes on the link. When the MN sends an RS message on a multiple 2821 access *NET link with an LLA source address and an OMNI option, ARs 2822 that recognize the option ensure that the MN is authorized to use the 2823 address and return an RA with a non-zero Router Lifetime only if the 2824 MN is authorized. ARs that do not recognize the option instead 2825 return an RA that makes no statement about the MN's authorization to 2826 use the source address. In that case, the MN should perform 2827 Duplicate Address Detection to ensure that it does not interfere with 2828 other nodes on the link. 2830 An alternative approach for multiple access *NET links to ensure 2831 isolation for MN / AR communications is through L2 address mappings 2832 as discussed in Appendix C. This arrangement imparts a (virtual) 2833 point-to-point link model over the (physical) multiple access link. 2835 19. OMNI Interfaces on Open Internetworks 2837 OMNI interfaces configured over IPv6-enabled underlying interfaces on 2838 an open Internetwork without an OMNI-aware first-hop AR receive RA 2839 messages that do not include an OMNI option, while OMNI interfaces 2840 configured over IPv4-only underlying interfaces do not receive any 2841 (IPv6) RA messages at all. OMNI interfaces that receive RA messages 2842 without an OMNI option configure addresses, on-link prefixes, etc. on 2843 the underlying interface that received the RA according to standard 2844 IPv6 ND and address resolution conventions [RFC4861] [RFC4862]. OMNI 2845 interfaces configured over IPv4-only underlying interfaces configure 2846 IPv4 address information on the underlying interfaces using 2847 mechanisms such as DHCPv4 [RFC2131]. 2849 OMNI interfaces configured over underlying interfaces that connect to 2850 an open Internetwork can apply security services such as VPNs to 2851 connect to an MSE, or can establish a direct link to an MSE through 2852 some other means (see Section 4). In environments where an explicit 2853 VPN or direct link may be impractical, OMNI interfaces can instead 2854 use UDP/IP encapsulation per [RFC6081][RFC4380] and HIP-based message 2855 authentication per [RFC7401]. 2857 OMNI interfaces use UDP service port number 8060 (see: Section 23.9 2858 and Section 3.6 of [I-D.templin-intarea-6706bis]) according to the 2859 simple UDP/IP encapsulation format specified in [RFC4380] for both 2860 IPv4 and IPv6 underlying interfaces. OMNI interfaces do not include 2861 the UDP/IP header/trailer extensions specified in [RFC4380][RFC6081], 2862 but may include them as OMNI sub-options instead when necessary. 2863 Since the OAL includes an integrity check over the OAL packet, OAL 2864 sources selectively disable UDP checksums for OAL packets that do not 2865 require UDP/IP address integrity, but enable UDP checksums for others 2866 including non-OAL packets, IPv6 ND messages used to establish link- 2867 layer addresses, etc. If the OAL source discovers that packets with 2868 UDP checksums disabled are being dropped in the path it should enable 2869 UDP checksums in future packets. Further considerations for UDP 2870 encapsulation checksums are found in [RFC6935][RFC6936]. 2872 For "Vehicle-to-Infrastructure (V2I)" coordination, the MN codes a 2873 HIP "Initiator" message in an OMNI option of an IPv6 RS message and 2874 the AR responds with a HIP "Responder" message coded in an OMNI 2875 option of an IPv6 RA message. HIP security services are applied per 2876 [RFC7401], using the RS/RA messages as simple "shipping containers" 2877 to convey the HIP parameters. In that case, a "two-message HIP 2878 exchange" through a single RS/RA exchange may be sufficient for 2879 mutual authentication. For "Vehicle-to-Vehicle (V2V)" coordination, 2880 two MNs can coordinate directly with one another with HIP "Initiator/ 2881 Responder" messages coded in OMNI options of IPv6 NS/NA messages. In 2882 that case, a four-message HIP exchange (i.e., two back-to-back NS/NA 2883 exchanges) may be necessary for the two MNs to attain mutual 2884 authentication. 2886 After establishing a VPN or preparing for UDP/IP encapsulation, OMNI 2887 interfaces send control plane messages to interface with the MS, 2888 including RS/RA messages used according to Section 14 and NS/NA 2889 messages used for route optimization and mobility (see: 2890 [I-D.templin-intarea-6706bis]). The control plane messages must be 2891 authenticated while data plane messages are delivered the same as for 2892 ordinary best-effort traffic with basic source address-based data 2893 origin verification. Data plane communications via OMNI interfaces 2894 that connect over open Internetworks without an explicit VPN should 2895 therefore employ transport- or higher-layer security to ensure 2896 integrity and/or confidentiality. 2898 OMNI interfaces configured over open Internetworks are often located 2899 behind NATs. The OMNI interface accommodates NAT traversal using 2900 UDP/IP encapsulation and the mechanisms discussed in 2902 [I-D.templin-intarea-6706bis]. To support NAT determination, ARs 2903 include an Origin Indication sub-option in RA messages sent in 2904 response to RS messages received from a Client via UDP/IP 2905 encapsulation. 2907 Note: Following the initial HIP Initiator/Responder exchange, OMNI 2908 interfaces configured over open Internetworks maintain HIP 2909 associations through the transmission of IPv6 ND messages that 2910 include OMNI options with HIP "Update" and "Notify" messages. OMNI 2911 interfaces use the HIP "Update" message when an acknowledgement is 2912 required, and use the "Notify" message in unacknowledged isolated 2913 IPv6 ND messages (e.g., unsolicited NAs). 2915 Note: ARs that act as proxys on an open Internetwork authenticate and 2916 remove HIP message OMNI sub-options from RSes they forward from a MN 2917 to an MSE, and insert and sign HIP message and Origin Indication sub- 2918 options in RAs they forward from an MSE to an MN. Conversely, ARs 2919 that act as proxys forward without processing any DHCPv6 information 2920 in RS/RA message exchanges between MNs and MSEs. The AR is therefore 2921 responsible for MN authentication while the MSE is responsible for 2922 registering/delegating MNPs. 2924 20. Time-Varying MNPs 2926 In some use cases, it is desirable, beneficial and efficient for the 2927 MN to receive a constant MNP that travels with the MN wherever it 2928 moves. For example, this would allow air traffic controllers to 2929 easily track aircraft, etc. In other cases, however (e.g., 2930 intelligent transportation systems), the MN may be willing to 2931 sacrifice a modicum of efficiency in order to have time-varying MNPs 2932 that can be changed every so often to defeat adversarial tracking. 2934 The prefix delegation services discussed in Section 14.3 allows OMNI 2935 MNs that desire time-varying MNPs to obtain short-lived prefixes to 2936 send RS messages with source set to the unspecified address (::) and/ 2937 or with an OMNI option with DHCPv6 Option sub-options. The MN would 2938 then be obligated to renumber its internal networks whenever its MNP 2939 (and therefore also its OMNI address) changes. This should not 2940 present a challenge for MNs with automated network renumbering 2941 services, however presents limits for the durations of ongoing 2942 sessions that would prefer to use a constant address. 2944 21. (H)HITs and Temporary ULAs 2946 MNs that generate (H)HITs but do not have pre-assigned MNPs can 2947 request MNP delegations by issuing IPv6 ND messages that use the 2948 (H)HIT instead of a Temporary ULA. In particular, when a MN creates 2949 an RS message it can set the source to the unspecified address (::) 2950 and destination to All-Routers multicast. The IPv6 ND message 2951 includes an OMNI option with a HIP "Initiator" message sub-option, 2952 and need not include a Node Identification sub-option since the MN's 2953 HIT appears in the HIP message. The MN then encapsulates the message 2954 in an IPv6 header with the (H)HIT as the source address and with 2955 destination set to either a unicast or anycast ADM-ULA. The MN then 2956 sends the message to the AR as specified in Section 14.1. 2958 When the AR receives the message, it notes that the RS source was the 2959 unspecified address (::), then examines the RS encapsulation source 2960 address to determine that the source is a (H)HIT and not a Temporary 2961 ULA. The AR next invokes the DHCPv6 protocol to request an MNP 2962 prefix delegation while using the HIT as the Client Identifier, then 2963 prepares an RA message with source address set to its own ADM-LLA and 2964 destination set to the MNP-LLA corresponding to the delegated MNP. 2965 The AR next includes an OMNI option with a HIP "Responder" message 2966 and any DHCPv6 prefix delegation parameters. The AR then finally 2967 encapsulates the RA in an IPv6 header with source address set to its 2968 own ADM-ULA and destination set to the (H)HIT from the RS 2969 encapsulation source address, then returns the encapsulated RA to the 2970 MN. 2972 MNs can also use (H)HITs and/or Temporary ULAs for direct MN-to-MN 2973 communications outside the context of any OMNI link supporting 2974 infrastructure. When two MNs encounter one another they can use 2975 their (H)HITs and/or Temporary ULAs as IPv6 packet source and 2976 destination addresses to support direct communications. MNs can also 2977 inject their (H)HITs and/or Temporary ULAs into a MANET/VANET routing 2978 protocol to enable multihop communications. MNs can further exchange 2979 IPv6 ND messages (such as NS/NA) using their (H)HITs and/or Temporary 2980 ULAs as source and destination addresses. Note that the HIP security 2981 protocols for establishing secure neighbor relationships are based on 2982 (H)HITs; therefore, Temporary ULAs would presumably utilize some 2983 alternate form of message authentication such as the [RFC4380] 2984 authentication service. 2986 Lastly, when MNs are within the coverage range of OMNI link 2987 infrastructure a case could be made for injecting (H)HITs and/or 2988 Temporary ULAs into the global MS routing system. For example, when 2989 the MN sends an RS to a MSE it could include a request to inject the 2990 (H)HIT / Temporary ULA into the routing system instead of requesting 2991 an MNP prefix delegation. This would potentially enable OMNI link- 2992 wide communications using only (H)HITs or Temporary ULAs, and not 2993 MNPs. This document notes the opportunity, but makes no 2994 recommendation. 2996 22. Address Selection 2998 OMNI MNs use LLAs only for link-scoped communications on the OMNI 2999 link. Typically, MNs use LLAs as source/destination IPv6 addresses 3000 of IPv6 ND messages, but may also use them for addressing ordinary 3001 data packets exchanged with an OMNI link neighbor. 3003 OMNI MNs use MNP-ULAs as source/destination IPv6 addresses in the OAL 3004 headers of OAL-encapsulated packets. OMNI MNs use Temporary ULAs for 3005 OAL addressing when an MNP-ULA is not available, or as source/ 3006 destination IPv6 addresses for communications within a MANET/VANET 3007 local area. OMNI MNs use HITs instead of Temporary ULAs when 3008 operation outside the context of a specific ULA domain and/or source 3009 address attestation is necessary. 3011 OMNI MNs use MNP-based GUAs for communications with Internet 3012 destinations when they are within range of OMNI link supporting 3013 infrastructure that can inject the MNP into the routing system. 3015 23. IANA Considerations 3017 The following IANA actions are requested: 3019 23.1. "IPv6 Neighbor Discovery Option Formats" Registry 3021 The IANA is instructed to allocate an official Type number TBD1 from 3022 the registry "IPv6 Neighbor Discovery Option Formats" for the OMNI 3023 option. Implementations set Type to 253 as an interim value 3024 [RFC4727]. 3026 23.2. "Ethernet Numbers" Registry 3028 The IANA is instructed to allocate one Ethernet unicast address TBD2 3029 (suggested value '00-52-14') in the 'ethernet-numbers' registry under 3030 "IANA Unicast 48-bit MAC Addresses" as follows: 3032 Addresses Usage Reference 3033 --------- ----- --------- 3034 00-52-14 Overlay Multilink Network (OMNI) Interface [RFCXXXX] 3036 Figure 24: IANA Unicast 48-bit MAC Addresses 3038 23.3. "ICMPv6 Code Fields: Type 2 - Packet Too Big" Registry 3040 The IANA is instructed to assign a new Code value "1" in the "ICMPv6 3041 Code Fields: Type 2 - Packet Too Big" registry. The registry should 3042 appear as follows: 3044 Code Name Reference 3045 --- ---- --------- 3046 0 Diagnostic Packet Too Big [RFC4443] 3047 1 Advisory Packet Too Big [RFCXXXX] 3049 Figure 25: ICMPv6 Code Fields: Type 2 - Packet Too Big Values 3051 23.4. "OMNI Option Sub-Type Values" (New Registry) 3053 The OMNI option defines a 5-bit Sub-Type field, for which IANA is 3054 instructed to create and maintain a new registry entitled "OMNI 3055 Option Sub-Type Values". Initial values are given below (future 3056 assignments are to be made through Standards Action [RFC8126]): 3058 Value Sub-Type name Reference 3059 ----- ------------- ---------- 3060 0 Pad1 [RFCXXXX] 3061 1 PadN [RFCXXXX] 3062 2 Interface Attributes (Type 1) [RFCXXXX] 3063 3 Interface Attributes (Type 2) [RFCXXXX] 3064 4 Traffic Selector [RFCXXXX] 3065 5 MS-Register [RFCXXXX] 3066 6 MS-Release [RFCXXXX] 3067 7 Geo Coordinates [RFCXXXX] 3068 8 DHCPv6 Message [RFCXXXX] 3069 9 HIP Message [RFCXXXX] 3070 10 Node Identification [RFCXXXX] 3071 11-29 Unassigned 3072 30 Sub-Type Extension [RFCXXXX] 3073 31 Reserved by IANA [RFCXXXX] 3075 Figure 26: OMNI Option Sub-Type Values 3077 23.5. "OMNI Node Identification ID-Type Values" (New Registry) 3079 The OMNI Node Identification Sub-Option (see: Section 11.1.11) 3080 contains an 8-bit ID-Type field, for which IANA is instructed to 3081 create and maintain a new registry entitled "OMNI Node Identification 3082 ID-Type Values". Initial values are given below (future assignments 3083 are to be made through Expert Review [RFC8126]): 3085 Value Sub-Type name Reference 3086 ----- ------------- ---------- 3087 0 UUID [RFCXXXX] 3088 1 HIT [RFCXXXX] 3089 2 HHIT [RFCXXXX] 3090 3 Network Access Identifier [RFCXXXX] 3091 4 FQDN [RFCXXXX] 3092 5-252 Unassigned [RFCXXXX] 3093 253-254 Reserved for Experimentation [RFCXXXX] 3094 255 Reserved by IANA [RFCXXXX] 3096 Figure 27: OMNI Node Identification ID-Type Values 3098 23.6. "OMNI Option Sub-Type Extension Values" (New Registry) 3100 The OMNI option defines an 8-bit Extension-Type field for Sub-Type 30 3101 (Sub-Type Extension), for which IANA is instructed to create and 3102 maintain a new registry entitled "OMNI Option Sub-Type Extension 3103 Values". Initial values are given below (future assignments are to 3104 be made through Expert Review [RFC8126]): 3106 Value Sub-Type name Reference 3107 ----- ------------- ---------- 3108 0 RFC4380 UDP/IP Header Option [RFCXXXX] 3109 1 RFC6081 UDP/IP Trailer Option [RFCXXXX] 3110 2-252 Unassigned 3111 253-254 Reserved for Experimentation [RFCXXXX] 3112 255 Reserved by IANA [RFCXXXX] 3114 Figure 28: OMNI Option Sub-Type Extension Values 3116 23.7. "OMNI RFC4380 UDP/IP Header Option" (New Registry) 3118 The OMNI Sub-Type Extension "RFC4380 UDP/IP Header Option" defines an 3119 8-bit Header Type field, for which IANA is instructed to create and 3120 maintain a new registry entitled "OMNI RFC4380 UDP/IP Header Option". 3121 Initial registry values are given below (future assignments are to be 3122 made through Expert Review [RFC8126]): 3124 Value Sub-Type name Reference 3125 ----- ------------- ---------- 3126 0 Origin Indication (IPv4) [RFC4380] 3127 1 Authentication Encapsulation [RFC4380] 3128 2 Origin Indication (IPv6) [RFCXXXX] 3129 3-252 Unassigned 3130 253-254 Reserved for Experimentation [RFCXXXX] 3131 255 Reserved by IANA [RFCXXXX] 3133 Figure 29: OMNI RFC4380 UDP/IP Header Option 3135 23.8. "OMNI RFC6081 UDP/IP Trailer Option" (New Registry) 3137 The OMNI Sub-Type Extension for "RFC6081 UDP/IP Trailer Option" 3138 defines an 8-bit Trailer Type field, for which IANA is instructed to 3139 create and maintain a new registry entitled "OMNI RFC6081 UDP/IP 3140 Trailer Option". Initial registry values are given below (future 3141 assignments are to be made through Expert Review [RFC8126]): 3143 Value Sub-Type name Reference 3144 ----- ------------- ---------- 3145 0 Unassigned 3146 1 Nonce [RFC6081] 3147 2 Unassigned 3148 3 Alternate Address (IPv4) [RFC6081] 3149 4 Neighbor Discovery Option [RFC6081] 3150 5 Random Port [RFC6081] 3151 6 Alternate Address (IPv6) [RFCXXXX] 3152 7-252 Unassigned 3153 253-254 Reserved for Experimentation [RFCXXXX] 3154 255 Reserved by IANA [RFCXXXX] 3156 Figure 30: OMNI RFC6081 Trailer Option 3158 23.9. Additional Considerations 3160 The IANA has assigned the UDP port number "8060" for an earlier 3161 experimental version of AERO [RFC6706]. This document together with 3162 [I-D.templin-intarea-6706bis] reclaims the UDP port number "8060" for 3163 'aero' as the service port for UDP/IP encapsulation. (Note that, 3164 although [RFC6706] was not widely implemented or deployed, any 3165 messages coded to that specification can be easily distinguished and 3166 ignored since they use the invalid ICMPv6 message type number '0'.) 3167 The IANA is therefore instructed to update the reference for UDP port 3168 number "8060" from "RFC6706" to "RFCXXXX" (i.e., this document). 3170 The IANA has assigned a 4 octet Private Enterprise Number (PEN) code 3171 "45282" in the "enterprise-numbers" registry. This document is the 3172 normative reference for using this code in DHCP Unique IDentifiers 3173 based on Enterprise Numbers ("DUID-EN for OMNI Interfaces") (see: 3174 Section 10). The IANA is therefore instructed to change the 3175 enterprise designation for PEN code "45282" from "LinkUp Networks" to 3176 "Overlay Multilink Network Interface (OMNI)". 3178 The IANA has assigned the ifType code "301 - omni - Overlay Multilink 3179 Network Interface (OMNI)" in accordance with Section 6 of [RFC8892]. 3180 The registration appears under the IANA "Structure of Management 3181 Information (SMI) Numbers (MIB Module Registrations) - Interface 3182 Types (ifType)" registry. 3184 No further IANA actions are required. 3186 24. Security Considerations 3188 Security considerations for IPv4 [RFC0791], IPv6 [RFC8200] and IPv6 3189 Neighbor Discovery [RFC4861] apply. OMNI interface IPv6 ND messages 3190 SHOULD include Nonce and Timestamp options [RFC3971] when transaction 3191 confirmation and/or time synchronization is needed. 3193 MN OMNI interfaces configured over secured ANET interfaces inherit 3194 the physical and/or link-layer security properties (i.e., "protected 3195 spectrum") of the connected ANETs. MN OMNI interfaces configured 3196 over open INET interfaces can use symmetric securing services such as 3197 VPNs or can by some other means establish a direct link. When a VPN 3198 or direct link may be impractical, however, the security services 3199 specified in [RFC7401] and/or [RFC4380] can be employed. While the 3200 OMNI link protects control plane messaging, applications must still 3201 employ end-to-end transport- or higher-layer security services to 3202 protect the data plane. 3204 Strong network layer security for control plane messages and 3205 forwarding path integrity for data plane messages between MSEs MUST 3206 be supported. In one example, the AERO service 3207 [I-D.templin-intarea-6706bis] constructs a spanning tree between MSEs 3208 and secures the links in the spanning tree with network layer 3209 security mechanisms such as IPsec [RFC4301] or Wireguard. Control 3210 plane messages are then constrained to travel only over the secured 3211 spanning tree paths and are therefore protected from attack or 3212 eavesdropping. Since data plane messages can travel over route 3213 optimized paths that do not strictly follow the spanning tree, 3214 however, end-to-end transport- or higher-layer security services are 3215 still required. 3217 Identity-based key verification infrastructure services such as iPSK 3218 may be necessary for verifying the identities claimed by MNs. This 3219 requirement should be harmonized with the manner in which (H)HITs are 3220 attested in a given operational environment. 3222 Security considerations for specific access network interface types 3223 are covered under the corresponding IP-over-(foo) specification 3224 (e.g., [RFC2464], [RFC2492], etc.). 3226 Security considerations for IPv6 fragmentation and reassembly are 3227 discussed in Section 5.5. 3229 25. Implementation Status 3231 AERO/OMNI Release-3.0.2 was tagged on October 15, 2020, and is 3232 undergoing internal testing. Additional internal releases expected 3233 within the coming months, with first public release expected end of 3234 1H2021. 3236 26. Acknowledgements 3238 The first version of this document was prepared per the consensus 3239 decision at the 7th Conference of the International Civil Aviation 3240 Organization (ICAO) Working Group-I Mobility Subgroup on March 22, 3241 2019. Consensus to take the document forward to the IETF was reached 3242 at the 9th Conference of the Mobility Subgroup on November 22, 2019. 3243 Attendees and contributors included: Guray Acar, Danny Bharj, 3244 Francois D'Humieres, Pavel Drasil, Nikos Fistas, Giovanni Garofolo, 3245 Bernhard Haindl, Vaughn Maiolla, Tom McParland, Victor Moreno, Madhu 3246 Niraula, Brent Phillips, Liviu Popescu, Jacky Pouzet, Aloke Roy, Greg 3247 Saccone, Robert Segers, Michal Skorepa, Michel Solery, Stephane 3248 Tamalet, Fred Templin, Jean-Marc Vacher, Bela Varkonyi, Tony Whyman, 3249 Fryderyk Wrobel and Dongsong Zeng. 3251 The following individuals are acknowledged for their useful comments: 3252 Stuart Card, Michael Matyas, Robert Moskowitz, Madhu Niraula, Greg 3253 Saccone, Stephane Tamalet, Eric Vyncke. Pavel Drasil, Zdenek Jaron 3254 and Michal Skorepa are especially recognized for their many helpful 3255 ideas and suggestions. Madhuri Madhava Badgandi, Sean Dickson, Don 3256 Dillenburg, Joe Dudkowski, Vijayasarathy Rajagopalan, Ron Sackman and 3257 Katherine Tran are acknowledged for their hard work on the 3258 implementation and technical insights that led to improvements for 3259 the spec. 3261 Discussions on the IETF 6man and atn mailing lists during the fall of 3262 2020 suggested additional points to consider. The authors gratefully 3263 acknowledge the list members who contributed valuable insights 3264 through those discussions. Eric Vyncke and Erik Kline were the 3265 intarea ADs, while Bob Hinden and Ole Troan were the 6man WG chairs 3266 at the time the document was developed; they are all gratefully 3267 acknowledged for their many helpful insights. 3269 Early observations on IP fragmentation performance implications were 3270 noted in the 1986 Digital Equipment Corporation (DEC) "qe reset" 3271 investigation, where fragment bursts from NFS traffic caused hardware 3272 resets resulting in sustained retransmission storms. Jeff Chase and 3273 Chet Juzsczak of the Ultrix Engineering Group supported the 3274 investigation. Early observations on L2 media MTU mismatch issues 3275 were noted in the 1988 DEC FDDI investigation, where Raj Jain and KK 3276 Ramakrishnan represented architectural considerations for FDDI/ 3277 Ethernet bridging. 3279 This work is aligned with the NASA Safe Autonomous Systems Operation 3280 (SASO) program under NASA contract number NNA16BD84C. 3282 This work is aligned with the FAA as per the SE2025 contract number 3283 DTFAWA-15-D-00030. 3285 This work is aligned with the Boeing Information Technology (BIT) 3286 Mobility Vision Lab (MVL) program. 3288 27. References 3290 27.1. Normative References 3292 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3293 DOI 10.17487/RFC0791, September 1981, 3294 . 3296 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3297 Requirement Levels", BCP 14, RFC 2119, 3298 DOI 10.17487/RFC2119, March 1997, 3299 . 3301 [RFC2474] Nichols, K., Blake, S., Baker, F., and D. Black, 3302 "Definition of the Differentiated Services Field (DS 3303 Field) in the IPv4 and IPv6 Headers", RFC 2474, 3304 DOI 10.17487/RFC2474, December 1998, 3305 . 3307 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3308 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3309 DOI 10.17487/RFC3971, March 2005, 3310 . 3312 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3313 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3314 November 2005, . 3316 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3317 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3318 . 3320 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3321 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3322 2006, . 3324 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3325 Control Message Protocol (ICMPv6) for the Internet 3326 Protocol Version 6 (IPv6) Specification", STD 89, 3327 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3328 . 3330 [RFC4727] Fenner, B., "Experimental Values In IPv4, IPv6, ICMPv4, 3331 ICMPv6, UDP, and TCP Headers", RFC 4727, 3332 DOI 10.17487/RFC4727, November 2006, 3333 . 3335 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3336 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3337 DOI 10.17487/RFC4861, September 2007, 3338 . 3340 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3341 Address Autoconfiguration", RFC 4862, 3342 DOI 10.17487/RFC4862, September 2007, 3343 . 3345 [RFC6088] Tsirtsis, G., Giarreta, G., Soliman, H., and N. Montavont, 3346 "Traffic Selectors for Flow Bindings", RFC 6088, 3347 DOI 10.17487/RFC6088, January 2011, 3348 . 3350 [RFC7401] Moskowitz, R., Ed., Heer, T., Jokela, P., and T. 3351 Henderson, "Host Identity Protocol Version 2 (HIPv2)", 3352 RFC 7401, DOI 10.17487/RFC7401, April 2015, 3353 . 3355 [RFC8028] Baker, F. and B. Carpenter, "First-Hop Router Selection by 3356 Hosts in a Multi-Prefix Network", RFC 8028, 3357 DOI 10.17487/RFC8028, November 2016, 3358 . 3360 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3361 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3362 May 2017, . 3364 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3365 (IPv6) Specification", STD 86, RFC 8200, 3366 DOI 10.17487/RFC8200, July 2017, 3367 . 3369 [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., 3370 "Path MTU Discovery for IP version 6", STD 87, RFC 8201, 3371 DOI 10.17487/RFC8201, July 2017, 3372 . 3374 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3375 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3376 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3377 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3378 . 3380 27.2. Informative References 3382 [ATN] Maiolla, V., "The OMNI Interface - An IPv6 Air/Ground 3383 Interface for Civil Aviation, IETF Liaison Statement 3384 #1676, https://datatracker.ietf.org/liaison/1676/", March 3385 2020. 3387 [ATN-IPS] WG-I, ICAO., "ICAO Document 9896 (Manual on the 3388 Aeronautical Telecommunication Network (ATN) using 3389 Internet Protocol Suite (IPS) Standards and Protocol), 3390 Draft Edition 3 (work-in-progress)", December 2020. 3392 [CKSUM] Stone, J., Greenwald, M., Partridge, C., and J. Hughes, 3393 "Performance of Checksums and CRC's Over Real Data, IEEE/ 3394 ACM Transactions on Networking, Vol. 6, No. 5", October 3395 1998. 3397 [CRC] Jain, R., "Error Characteristics of Fiber Distributed Data 3398 Interface (FDDI), IEEE Transactions on Communications", 3399 August 1990. 3401 [I-D.ietf-drip-rid] 3402 Moskowitz, R., Card, S., Wiethuechter, A., and A. Gurtov, 3403 "UAS Remote ID", draft-ietf-drip-rid-06 (work in 3404 progress), December 2020. 3406 [I-D.ietf-intarea-tunnels] 3407 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3408 Architecture", draft-ietf-intarea-tunnels-10 (work in 3409 progress), September 2019. 3411 [I-D.ietf-ipwave-vehicular-networking] 3412 Jeong, J., "IPv6 Wireless Access in Vehicular Environments 3413 (IPWAVE): Problem Statement and Use Cases", draft-ietf- 3414 ipwave-vehicular-networking-19 (work in progress), July 3415 2020. 3417 [I-D.templin-6man-dhcpv6-ndopt] 3418 Templin, F., "A Unified Stateful/Stateless Configuration 3419 Service for IPv6", draft-templin-6man-dhcpv6-ndopt-11 3420 (work in progress), January 2021. 3422 [I-D.templin-6man-lla-type] 3423 Templin, F., "The IPv6 Link-Local Address Type Field", 3424 draft-templin-6man-lla-type-02 (work in progress), 3425 November 2020. 3427 [I-D.templin-intarea-6706bis] 3428 Templin, F., "Asymmetric Extended Route Optimization 3429 (AERO)", draft-templin-intarea-6706bis-87 (work in 3430 progress), January 2021. 3432 [IPV4-GUA] 3433 Postel, J., "IPv4 Address Space Registry, 3434 https://www.iana.org/assignments/ipv4-address-space/ipv4- 3435 address-space.xhtml", December 2020. 3437 [IPV6-GUA] 3438 Postel, J., "IPv6 Global Unicast Address Assignments, 3439 https://www.iana.org/assignments/ipv6-unicast-address- 3440 assignments/ipv6-unicast-address-assignments.xhtml", 3441 December 2020. 3443 [RFC0905] "ISO Transport Protocol specification ISO DP 8073", 3444 RFC 905, DOI 10.17487/RFC0905, April 1984, 3445 . 3447 [RFC1035] Mockapetris, P., "Domain names - implementation and 3448 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3449 November 1987, . 3451 [RFC1122] Braden, R., Ed., "Requirements for Internet Hosts - 3452 Communication Layers", STD 3, RFC 1122, 3453 DOI 10.17487/RFC1122, October 1989, 3454 . 3456 [RFC1191] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191, 3457 DOI 10.17487/RFC1191, November 1990, 3458 . 3460 [RFC1256] Deering, S., Ed., "ICMP Router Discovery Messages", 3461 RFC 1256, DOI 10.17487/RFC1256, September 1991, 3462 . 3464 [RFC2131] Droms, R., "Dynamic Host Configuration Protocol", 3465 RFC 2131, DOI 10.17487/RFC2131, March 1997, 3466 . 3468 [RFC2225] Laubach, M. and J. Halpern, "Classical IP and ARP over 3469 ATM", RFC 2225, DOI 10.17487/RFC2225, April 1998, 3470 . 3472 [RFC2328] Moy, J., "OSPF Version 2", STD 54, RFC 2328, 3473 DOI 10.17487/RFC2328, April 1998, 3474 . 3476 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 3477 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 3478 . 3480 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3481 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3482 December 1998, . 3484 [RFC2492] Armitage, G., Schulter, P., and M. Jork, "IPv6 over ATM 3485 Networks", RFC 2492, DOI 10.17487/RFC2492, January 1999, 3486 . 3488 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3489 Domains without Explicit Tunnels", RFC 2529, 3490 DOI 10.17487/RFC2529, March 1999, 3491 . 3493 [RFC2863] McCloghrie, K. and F. Kastenholz, "The Interfaces Group 3494 MIB", RFC 2863, DOI 10.17487/RFC2863, June 2000, 3495 . 3497 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 3498 DOI 10.17487/RFC3330, September 2002, 3499 . 3501 [RFC3692] Narten, T., "Assigning Experimental and Testing Numbers 3502 Considered Useful", BCP 82, RFC 3692, 3503 DOI 10.17487/RFC3692, January 2004, 3504 . 3506 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3507 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3508 DOI 10.17487/RFC3810, June 2004, 3509 . 3511 [RFC3819] Karn, P., Ed., Bormann, C., Fairhurst, G., Grossman, D., 3512 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and L. 3513 Wood, "Advice for Internet Subnetwork Designers", BCP 89, 3514 RFC 3819, DOI 10.17487/RFC3819, July 2004, 3515 . 3517 [RFC3879] Huitema, C. and B. Carpenter, "Deprecating Site Local 3518 Addresses", RFC 3879, DOI 10.17487/RFC3879, September 3519 2004, . 3521 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 3522 Unique IDentifier (UUID) URN Namespace", RFC 4122, 3523 DOI 10.17487/RFC4122, July 2005, 3524 . 3526 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3527 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3528 DOI 10.17487/RFC4271, January 2006, 3529 . 3531 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3532 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3533 December 2005, . 3535 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3536 Network Address Translations (NATs)", RFC 4380, 3537 DOI 10.17487/RFC4380, February 2006, 3538 . 3540 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3541 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3542 2006, . 3544 [RFC4429] Moore, N., "Optimistic Duplicate Address Detection (DAD) 3545 for IPv6", RFC 4429, DOI 10.17487/RFC4429, April 2006, 3546 . 3548 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3549 "Considerations for Internet Group Management Protocol 3550 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3551 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3552 . 3554 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3555 "Internet Group Management Protocol (IGMP) / Multicast 3556 Listener Discovery (MLD)-Based Multicast Forwarding 3557 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3558 August 2006, . 3560 [RFC4821] Mathis, M. and J. Heffner, "Packetization Layer Path MTU 3561 Discovery", RFC 4821, DOI 10.17487/RFC4821, March 2007, 3562 . 3564 [RFC4963] Heffner, J., Mathis, M., and B. Chandler, "IPv4 Reassembly 3565 Errors at High Data Rates", RFC 4963, 3566 DOI 10.17487/RFC4963, July 2007, 3567 . 3569 [RFC5175] Haberman, B., Ed. and R. Hinden, "IPv6 Router 3570 Advertisement Flags Option", RFC 5175, 3571 DOI 10.17487/RFC5175, March 2008, 3572 . 3574 [RFC5213] Gundavelli, S., Ed., Leung, K., Devarapalli, V., 3575 Chowdhury, K., and B. Patil, "Proxy Mobile IPv6", 3576 RFC 5213, DOI 10.17487/RFC5213, August 2008, 3577 . 3579 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3580 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3581 DOI 10.17487/RFC5214, March 2008, 3582 . 3584 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3585 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3586 . 3588 [RFC5798] Nadas, S., Ed., "Virtual Router Redundancy Protocol (VRRP) 3589 Version 3 for IPv4 and IPv6", RFC 5798, 3590 DOI 10.17487/RFC5798, March 2010, 3591 . 3593 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3594 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3595 . 3597 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3598 DOI 10.17487/RFC6081, January 2011, 3599 . 3601 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3602 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3603 DOI 10.17487/RFC6221, May 2011, 3604 . 3606 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 3607 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 3608 DOI 10.17487/RFC6355, August 2011, 3609 . 3611 [RFC6543] Gundavelli, S., "Reserved IPv6 Interface Identifier for 3612 Proxy Mobile IPv6", RFC 6543, DOI 10.17487/RFC6543, May 3613 2012, . 3615 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3616 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3617 . 3619 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3620 UDP Checksums for Tunneled Packets", RFC 6935, 3621 DOI 10.17487/RFC6935, April 2013, 3622 . 3624 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3625 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3626 RFC 6936, DOI 10.17487/RFC6936, April 2013, 3627 . 3629 [RFC6980] Gont, F., "Security Implications of IPv6 Fragmentation 3630 with IPv6 Neighbor Discovery", RFC 6980, 3631 DOI 10.17487/RFC6980, August 2013, 3632 . 3634 [RFC7084] Singh, H., Beebee, W., Donley, C., and B. Stark, "Basic 3635 Requirements for IPv6 Customer Edge Routers", RFC 7084, 3636 DOI 10.17487/RFC7084, November 2013, 3637 . 3639 [RFC7421] Carpenter, B., Ed., Chown, T., Gont, F., Jiang, S., 3640 Petrescu, A., and A. Yourtchenko, "Analysis of the 64-bit 3641 Boundary in IPv6 Addressing", RFC 7421, 3642 DOI 10.17487/RFC7421, January 2015, 3643 . 3645 [RFC7526] Troan, O. and B. Carpenter, Ed., "Deprecating the Anycast 3646 Prefix for 6to4 Relay Routers", BCP 196, RFC 7526, 3647 DOI 10.17487/RFC7526, May 2015, 3648 . 3650 [RFC7542] DeKok, A., "The Network Access Identifier", RFC 7542, 3651 DOI 10.17487/RFC7542, May 2015, 3652 . 3654 [RFC7739] Gont, F., "Security Implications of Predictable Fragment 3655 Identification Values", RFC 7739, DOI 10.17487/RFC7739, 3656 February 2016, . 3658 [RFC7847] Melia, T., Ed. and S. Gundavelli, Ed., "Logical-Interface 3659 Support for IP Hosts with Multi-Access Support", RFC 7847, 3660 DOI 10.17487/RFC7847, May 2016, 3661 . 3663 [RFC8126] Cotton, M., Leiba, B., and T. Narten, "Guidelines for 3664 Writing an IANA Considerations Section in RFCs", BCP 26, 3665 RFC 8126, DOI 10.17487/RFC8126, June 2017, 3666 . 3668 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3669 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3670 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3671 July 2018, . 3673 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3674 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 3675 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 3676 . 3678 [RFC8892] Thaler, D. and D. Romascanu, "Guidelines and Registration 3679 Procedures for Interface Types and Tunnel Types", 3680 RFC 8892, DOI 10.17487/RFC8892, August 2020, 3681 . 3683 [RFC8899] Fairhurst, G., Jones, T., Tuexen, M., Ruengeler, I., and 3684 T. Voelker, "Packetization Layer Path MTU Discovery for 3685 Datagram Transports", RFC 8899, DOI 10.17487/RFC8899, 3686 September 2020, . 3688 [RFC8900] Bonica, R., Baker, F., Huston, G., Hinden, R., Troan, O., 3689 and F. Gont, "IP Fragmentation Considered Fragile", 3690 BCP 230, RFC 8900, DOI 10.17487/RFC8900, September 2020, 3691 . 3693 [RFC8981] Gont, F., Krishnan, S., Narten, T., and R. Draves, 3694 "Temporary Address Extensions for Stateless Address 3695 Autoconfiguration in IPv6", RFC 8981, 3696 DOI 10.17487/RFC8981, February 2021, 3697 . 3699 Appendix A. Interface Attribute Preferences Bitmap Encoding 3701 Adaptation of the OMNI option Interface Attributes Preferences Bitmap 3702 encoding to specific Internetworks such as the Aeronautical 3703 Telecommunications Network with Internet Protocol Services (ATN/IPS) 3704 may include link selection preferences based on other traffic 3705 classifiers (e.g., transport port numbers, etc.) in addition to the 3706 existing DSCP-based preferences. Nodes on specific Internetworks 3707 maintain a map of traffic classifiers to additional P[*] preference 3708 fields beyond the first 64. For example, TCP port 22 maps to P[67], 3709 TCP port 443 maps to P[70], UDP port 8060 maps to P[76], etc. 3711 Implementations use Simplex or Indexed encoding formats for P[*] 3712 encoding in order to encode a given set of traffic classifiers in the 3713 most efficient way. Some use cases may be more efficiently coded 3714 using Simplex form, while others may be more efficient using Indexed. 3715 Once a format is selected for preparation of a single Interface 3716 Attribute the same format must be used for the entire Interface 3717 Attribute sub-option. Different sub-options may use different 3718 formats. 3720 The following figures show coding examples for various Simplex and 3721 Indexed formats: 3723 0 1 2 3 3724 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 3725 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3726 | Sub-Type=3| Sub-length=N | omIndex | omType | 3727 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3728 | Provider ID | Link |R| API | Bitmap(0)=0xff|P00|P01|P02|P03| 3729 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3730 |P04|P05|P06|P07|P08|P09|P10|P11|P12|P13|P14|P15|P16|P17|P18|P19| 3731 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3732 |P20|P21|P22|P23|P24|P25|P26|P27|P28|P29|P30|P31| Bitmap(1)=0xff| 3733 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3734 |P32|P33|P34|P35|P36|P37|P38|P39|P40|P41|P42|P43|P44|P45|P46|P47| 3735 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3736 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 3737 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3738 | Bitmap(2)=0xff|P64|P65|P67|P68| ... 3739 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 3741 Figure 31: Example 1: Dense Simplex Encoding 3743 0 1 2 3 3744 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 3745 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3746 | Sub-Type=3| Sub-length=N | omIndex | omType | 3747 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3748 | Provider ID | Link |R| API | Bitmap(0)=0x00| Bitmap(1)=0x0f| 3749 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3750 |P48|P49|P50|P51|P52|P53|P54|P55|P56|P57|P58|P59|P60|P61|P62|P63| 3751 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3752 | Bitmap(2)=0x00| Bitmap(3)=0x00| Bitmap(4)=0x00| Bitmap(5)=0x00| 3753 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3754 | Bitmap(6)=0xf0|192|193|194|195|196|197|198|199|200|201|202|203| 3755 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3756 |204|205|206|207| Bitmap(7)=0x00| Bitmap(8)=0x0f|272|273|274|275| 3757 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3758 |276|277|278|279|280|281|282|283|284|285|286|287| Bitmap(9)=0x00| 3759 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3760 |Bitmap(10)=0x00| ... 3761 +-+-+-+-+-+-+-+-+-+-+- 3763 Figure 32: Example 2: Sparse Simplex Encoding 3765 0 1 2 3 3766 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 3767 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3768 | Sub-Type=3| Sub-length=N | omIndex | omType | 3769 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3770 | Provider ID | Link |R| API | Index = 0x00 | Bitmap = 0x80 | 3771 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3772 |P00|P01|P02|P03| Index = 0x01 | Bitmap = 0x01 |P60|P61|P62|P63| 3773 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3774 | Index = 0x10 | Bitmap = 0x80 |512|513|514|515| Index = 0x18 | 3775 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 3776 | Bitmap = 0x01 |796|797|798|799| ... 3777 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- 3779 Figure 33: Example 3: Indexed Encoding 3781 Appendix B. VDL Mode 2 Considerations 3783 ICAO Doc 9776 is the "Technical Manual for VHF Data Link Mode 2" 3784 (VDLM2) that specifies an essential radio frequency data link service 3785 for aircraft and ground stations in worldwide civil aviation air 3786 traffic management. The VDLM2 link type is "multicast capable" 3787 [RFC4861], but with considerable differences from common multicast 3788 links such as Ethernet and IEEE 802.11. 3790 First, the VDLM2 link data rate is only 31.5Kbps - multiple orders of 3791 magnitude less than most modern wireless networking gear. Second, 3792 due to the low available link bandwidth only VDLM2 ground stations 3793 (i.e., and not aircraft) are permitted to send broadcasts, and even 3794 so only as compact layer 2 "beacons". Third, aircraft employ the 3795 services of ground stations by performing unicast RS/RA exchanges 3796 upon receipt of beacons instead of listening for multicast RA 3797 messages and/or sending multicast RS messages. 3799 This beacon-oriented unicast RS/RA approach is necessary to conserve 3800 the already-scarce available link bandwidth. Moreover, since the 3801 numbers of beaconing ground stations operating within a given spatial 3802 range must be kept as sparse as possible, it would not be feasible to 3803 have different classes of ground stations within the same region 3804 observing different protocols. It is therefore highly desirable that 3805 all ground stations observe a common language of RS/RA as specified 3806 in this document. 3808 Note that links of this nature may benefit from compression 3809 techniques that reduce the bandwidth necessary for conveying the same 3810 amount of data. The IETF lpwan working group is considering possible 3811 alternatives: [https://datatracker.ietf.org/wg/lpwan/documents]. 3813 Appendix C. MN / AR Isolation Through L2 Address Mapping 3815 Per [RFC4861], IPv6 ND messages may be sent to either a multicast or 3816 unicast link-scoped IPv6 destination address. However, IPv6 ND 3817 messaging should be coordinated between the MN and AR only without 3818 invoking other nodes on the *NET. This implies that MN / AR control 3819 messaging should be isolated and not overheard by other nodes on the 3820 link. 3822 To support MN / AR isolation on some *NET links, ARs can maintain an 3823 OMNI-specific unicast L2 address ("MSADDR"). For Ethernet-compatible 3824 *NETs, this specification reserves one Ethernet unicast address TBD2 3825 (see: Section 23). For non-Ethernet statically-addressed *NETs, 3826 MSADDR is reserved per the assigned numbers authority for the *NET 3827 addressing space. For still other *NETs, MSADDR may be dynamically 3828 discovered through other means, e.g., L2 beacons. 3830 MNs map the L3 addresses of all IPv6 ND messages they send (i.e., 3831 both multicast and unicast) to MSADDR instead of to an ordinary 3832 unicast or multicast L2 address. In this way, all of the MN's IPv6 3833 ND messages will be received by ARs that are configured to accept 3834 packets destined to MSADDR. Note that multiple ARs on the link could 3835 be configured to accept packets destined to MSADDR, e.g., as a basis 3836 for supporting redundancy. 3838 Therefore, ARs must accept and process packets destined to MSADDR, 3839 while all other devices must not process packets destined to MSADDR. 3840 This model has well-established operational experience in Proxy 3841 Mobile IPv6 (PMIP) [RFC5213][RFC6543]. 3843 Appendix D. Change Log 3845 << RFC Editor - remove prior to publication >> 3847 Differences from draft-templin-6man-omni-interface-35 to draft- 3848 templin-6man-omni-interface-36: 3850 o Major clarifications on aspects such as "hard/soft" PTB error 3851 messages 3853 o Made generic so that either IP protocol version (IPv4 or IPv6) can 3854 be used in the data plane. 3856 Differences from draft-templin-6man-omni-interface-31 to draft- 3857 templin-6man-omni-interface-32: 3859 o MTU 3860 o Support for multi-hop ANETS such as ISATAP. 3862 Differences from draft-templin-6man-omni-interface-29 to draft- 3863 templin-6man-omni-interface-30: 3865 o Moved link-layer addressing information into the OMNI option on a 3866 per-ifIndex basis 3868 o Renamed "ifIndex-tuple" to "Interface Attributes" 3870 Differences from draft-templin-6man-omni-interface-27 to draft- 3871 templin-6man-omni-interface-28: 3873 o Updates based on implementation experience. 3875 Differences from draft-templin-6man-omni-interface-25 to draft- 3876 templin-6man-omni-interface-26: 3878 o Further clarification on "aggregate" RA messages. 3880 o Expanded Security Considerations to discuss expectations for 3881 security in the Mobility Service. 3883 Differences from draft-templin-6man-omni-interface-20 to draft- 3884 templin-6man-omni-interface-21: 3886 o Safety-Based Multilink (SBM) and Performance-Based Multilink 3887 (PBM). 3889 Differences from draft-templin-6man-omni-interface-18 to draft- 3890 templin-6man-omni-interface-19: 3892 o SEND/CGA. 3894 Differences from draft-templin-6man-omni-interface-17 to draft- 3895 templin-6man-omni-interface-18: 3897 o Teredo 3899 Differences from draft-templin-6man-omni-interface-14 to draft- 3900 templin-6man-omni-interface-15: 3902 o Prefix length discussions removed. 3904 Differences from draft-templin-6man-omni-interface-12 to draft- 3905 templin-6man-omni-interface-13: 3907 o Teredo 3908 Differences from draft-templin-6man-omni-interface-11 to draft- 3909 templin-6man-omni-interface-12: 3911 o Major simplifications and clarifications on MTU and fragmentation. 3913 o Document now updates RFC4443 and RFC8201. 3915 Differences from draft-templin-6man-omni-interface-10 to draft- 3916 templin-6man-omni-interface-11: 3918 o Removed /64 assumption, resulting in new OMNI address format. 3920 Differences from draft-templin-6man-omni-interface-07 to draft- 3921 templin-6man-omni-interface-08: 3923 o OMNI MNs in the open Internet 3925 Differences from draft-templin-6man-omni-interface-06 to draft- 3926 templin-6man-omni-interface-07: 3928 o Brought back L2 MSADDR mapping text for MN / AR isolation based on 3929 L2 addressing. 3931 o Expanded "Transition Considerations". 3933 Differences from draft-templin-6man-omni-interface-05 to draft- 3934 templin-6man-omni-interface-06: 3936 o Brought back OMNI option "R" flag, and discussed its use. 3938 Differences from draft-templin-6man-omni-interface-04 to draft- 3939 templin-6man-omni-interface-05: 3941 o Transition considerations, and overhaul of RS/RA addressing with 3942 the inclusion of MSE addresses within the OMNI option instead of 3943 as RS/RA addresses (developed under FAA SE2025 contract number 3944 DTFAWA-15-D-00030). 3946 Differences from draft-templin-6man-omni-interface-02 to draft- 3947 templin-6man-omni-interface-03: 3949 o Added "advisory PTB messages" under FAA SE2025 contract number 3950 DTFAWA-15-D-00030. 3952 Differences from draft-templin-6man-omni-interface-01 to draft- 3953 templin-6man-omni-interface-02: 3955 o Removed "Primary" flag and supporting text. 3957 o Clarified that "Router Lifetime" applies to each ANET interface 3958 independently, and that the union of all ANET interface Router 3959 Lifetimes determines MSE lifetime. 3961 Differences from draft-templin-6man-omni-interface-00 to draft- 3962 templin-6man-omni-interface-01: 3964 o "All-MSEs" OMNI LLA defined. Also reserved fe80::ff00:0000/104 3965 for future use (most likely as "pseudo-multicast"). 3967 o Non-normative discussion of alternate OMNI LLA construction form 3968 made possible if the 64-bit assumption were relaxed. 3970 First draft version (draft-templin-atn-aero-interface-00): 3972 o Draft based on consensus decision of ICAO Working Group I Mobility 3973 Subgroup March 22, 2019. 3975 Authors' Addresses 3977 Fred L. Templin (editor) 3978 The Boeing Company 3979 P.O. Box 3707 3980 Seattle, WA 98124 3981 USA 3983 Email: fltemplin@acm.org 3985 Tony Whyman 3986 MWA Ltd c/o Inmarsat Global Ltd 3987 99 City Road 3988 London EC1Y 1AX 3989 England 3991 Email: tony.whyman@mccallumwhyman.com