idnits 2.17.1 draft-templin-6man-aero-21.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (July 2, 2021) is 1028 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: 'ULA' is mentioned on line 1015, but not defined == Missing Reference: 'ULA0' is mentioned on line 1015, but not defined == Missing Reference: 'ULA1' is mentioned on line 1015, but not defined == Missing Reference: 'ULA2' is mentioned on line 1015, but not defined == Unused Reference: 'RFC3971' is defined on line 4098, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 4103, but no explicit reference was found in the text == Unused Reference: 'RFC4193' is defined on line 4111, but no explicit reference was found in the text == Unused Reference: 'RFC7739' is defined on line 4139, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 4160, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-crh-helper-opt' is defined on line 4169, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 4174, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 4197, but no explicit reference was found in the text == Unused Reference: 'OVPN' is defined on line 4226, but no explicit reference was found in the text == Unused Reference: 'RFC2003' is defined on line 4236, but no explicit reference was found in the text == Unused Reference: 'RFC2004' is defined on line 4240, but no explicit reference was found in the text == Unused Reference: 'RFC2983' is defined on line 4257, but no explicit reference was found in the text == Unused Reference: 'RFC3168' is defined on line 4261, but no explicit reference was found in the text == Unused Reference: 'RFC4122' is defined on line 4275, but no explicit reference was found in the text == Unused Reference: 'RFC4389' is defined on line 4297, but no explicit reference was found in the text == Unused Reference: 'RFC4982' is defined on line 4324, but no explicit reference was found in the text == Unused Reference: 'RFC6139' is defined on line 4366, but no explicit reference was found in the text == Unused Reference: 'RFC6273' is defined on line 4386, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 4395, but no explicit reference was found in the text == Unused Reference: 'RFC6438' is defined on line 4400, but no explicit reference was found in the text == Unused Reference: 'RFC6935' is defined on line 4409, but no explicit reference was found in the text == Unused Reference: 'RFC6936' is defined on line 4414, but no explicit reference was found in the text == Outdated reference: A later version (-74) exists of draft-templin-6man-omni-03 == Outdated reference: A later version (-31) exists of draft-bonica-6man-comp-rtg-hdr-24 == Outdated reference: A later version (-04) exists of draft-bonica-6man-crh-helper-opt-03 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-10 == Outdated reference: A later version (-30) exists of draft-ietf-ipwave-vehicular-networking-20 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-10 -- Obsolete informational reference (is this intentional?): RFC 3330 (Obsoleted by RFC 5735) -- Obsolete informational reference (is this intentional?): RFC 6106 (Obsoleted by RFC 8106) -- Obsolete informational reference (is this intentional?): RFC 6347 (Obsoleted by RFC 9147) Summary: 0 errors (**), 0 flaws (~~), 33 warnings (==), 5 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 Boeing Research & Technology 4 Intended status: Informational July 2, 2021 5 Expires: January 3, 2022 7 Automatic Extended Route Optimization (AERO) 8 draft-templin-6man-aero-21 10 Abstract 12 This document specifies an Automatic Extended Route Optimization 13 (AERO) service for IP internetworking over Overlay Multilink Network 14 (OMNI) interfaces. AERO/OMNI use an IPv6 link-local address format 15 that supports operation of the IPv6 Neighbor Discovery (IPv6 ND) 16 protocol. Prefix delegation/registration services are employed for 17 network admission and to manage the IP forwarding and routing 18 systems. Secure multilink operation, mobility management, multicast, 19 traffic selector signaling and route optimization are naturally 20 supported through dynamic neighbor cache updates. AERO is a widely- 21 applicable mobile internetworking service especially well-suited to 22 aviation services, intelligent transportation systems, mobile end 23 user devices and many other applications. 25 Status of This Memo 27 This Internet-Draft is submitted in full conformance with the 28 provisions of BCP 78 and BCP 79. 30 Internet-Drafts are working documents of the Internet Engineering 31 Task Force (IETF). Note that other groups may also distribute 32 working documents as Internet-Drafts. The list of current Internet- 33 Drafts is at https://datatracker.ietf.org/drafts/current/. 35 Internet-Drafts are draft documents valid for a maximum of six months 36 and may be updated, replaced, or obsoleted by other documents at any 37 time. It is inappropriate to use Internet-Drafts as reference 38 material or to cite them other than as "work in progress." 40 This Internet-Draft will expire on January 3, 2022. 42 Copyright Notice 44 Copyright (c) 2021 IETF Trust and the persons identified as the 45 document authors. All rights reserved. 47 This document is subject to BCP 78 and the IETF Trust's Legal 48 Provisions Relating to IETF Documents 49 (https://trustee.ietf.org/license-info) in effect on the date of 50 publication of this document. Please review these documents 51 carefully, as they describe your rights and restrictions with respect 52 to this document. Code Components extracted from this document must 53 include Simplified BSD License text as described in Section 4.e of 54 the Trust Legal Provisions and are provided without warranty as 55 described in the Simplified BSD License. 57 Table of Contents 59 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 60 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 6 61 3. Automatic Extended Route Optimization (AERO) . . . . . . . . 13 62 3.1. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 14 63 3.2. The AERO Service over OMNI Links . . . . . . . . . . . . 15 64 3.2.1. AERO/OMNI Reference Model . . . . . . . . . . . . . . 15 65 3.2.2. Addressing and Node Identification . . . . . . . . . 18 66 3.2.3. AERO Routing System . . . . . . . . . . . . . . . . . 18 67 3.2.4. OMNI Link Forwarding . . . . . . . . . . . . . . . . 20 68 3.2.5. Segment Routing Topologies (SRTs) . . . . . . . . . . 22 69 3.2.6. Segment Routing For OMNI Link Selection . . . . . . . 22 70 3.2.7. OMNI Multilink Forwarding . . . . . . . . . . . . . . 23 71 3.3. OMNI Interface Characteristics . . . . . . . . . . . . . 33 72 3.4. OMNI Interface Initialization . . . . . . . . . . . . . . 35 73 3.4.1. AERO Proxy/Server and Relay Behavior . . . . . . . . 35 74 3.4.2. AERO Client Behavior . . . . . . . . . . . . . . . . 36 75 3.4.3. AERO Bridge Behavior . . . . . . . . . . . . . . . . 36 76 3.5. OMNI Interface Neighbor Cache Maintenance . . . . . . . . 36 77 3.5.1. OMNI ND Messages . . . . . . . . . . . . . . . . . . 38 78 3.5.2. OMNI Neighbor Advertisement Message Flags . . . . . . 40 79 3.5.3. OMNI Neighbor Window Synchronization . . . . . . . . 40 80 3.6. OMNI Interface Encapsulation and Re-encapsulation . . . . 41 81 3.7. OMNI Interface Decapsulation . . . . . . . . . . . . . . 41 82 3.8. OMNI Interface Data Origin Authentication . . . . . . . . 41 83 3.9. OMNI Interface MTU . . . . . . . . . . . . . . . . . . . 42 84 3.10. OMNI Interface Forwarding Algorithm . . . . . . . . . . . 43 85 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 44 86 3.10.2. Proxy/Server and Relay Forwarding Algorithm . . . . 46 87 3.10.3. Bridge Forwarding Algorithm . . . . . . . . . . . . 49 88 3.11. OMNI Interface Error Handling . . . . . . . . . . . . . . 50 89 3.12. AERO Router Discovery, Prefix Delegation and 90 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 52 91 3.12.1. AERO Service Model . . . . . . . . . . . . . . . . . 53 92 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 54 93 3.12.3. AERO Proxy/Server Behavior . . . . . . . . . . . . . 56 94 3.13. AERO Proxy/Server Coordination . . . . . . . . . . . . . 59 95 3.13.1. Detecting and Responding to Proxy/Server Failures . 62 96 3.14. AERO Route Optimization . . . . . . . . . . . . . . . . . 63 97 3.14.1. Route Optimization Initiation . . . . . . . . . . . 63 98 3.14.2. Relaying the NS(AR) *NET Packet(s) . . . . . . . . . 64 99 3.14.3. Processing the NS(AR) and Sending the NA(AR) . . . . 65 100 3.14.4. Relaying the NA(AR) . . . . . . . . . . . . . . . . 66 101 3.14.5. Processing the NA(AR) . . . . . . . . . . . . . . . 66 102 3.14.6. Forwarding Packets to Route Optimized Targets . . . 67 103 3.15. Neighbor Unreachability Detection (NUD) . . . . . . . . . 68 104 3.16. Mobility Management and Quality of Service (QoS) . . . . 69 105 3.16.1. Mobility Update Messaging . . . . . . . . . . . . . 70 106 3.16.2. Announcing Link-Layer Address and/or QoS Preference 107 Changes . . . . . . . . . . . . . . . . . . . . . . 71 108 3.16.3. Bringing New Links Into Service . . . . . . . . . . 72 109 3.16.4. Deactivating Existing Links . . . . . . . . . . . . 72 110 3.16.5. Moving Between Proxy/Servers . . . . . . . . . . . . 72 111 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 73 112 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 74 113 3.17.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 75 114 3.17.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 76 115 3.18. Operation over Multiple OMNI Links . . . . . . . . . . . 76 116 3.19. DNS Considerations . . . . . . . . . . . . . . . . . . . 77 117 3.20. Transition/Coexistence Considerations . . . . . . . . . . 77 118 3.21. Detecting and Reacting to Proxy/Server and Bridge 119 Failures . . . . . . . . . . . . . . . . . . . . . . . . 78 120 3.22. AERO Clients on the Open Internet . . . . . . . . . . . . 78 121 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 81 122 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 81 123 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 81 124 6. Security Considerations . . . . . . . . . . . . . . . . . . . 82 125 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 84 126 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 86 127 8.1. Normative References . . . . . . . . . . . . . . . . . . 86 128 8.2. Informative References . . . . . . . . . . . . . . . . . 88 129 Appendix A. Non-Normative Considerations . . . . . . . . . . . . 94 130 A.1. Implementation Strategies for Route Optimization . . . . 94 131 A.2. Implicit Mobility Management . . . . . . . . . . . . . . 94 132 A.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 95 133 A.4. AERO Critical Infrastructure Considerations . . . . . . . 95 134 A.5. AERO Server Failure Implications . . . . . . . . . . . . 96 135 A.6. AERO Client / Server Architecture . . . . . . . . . . . . 96 136 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 98 137 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 102 139 1. Introduction 141 Automatic Extended Route Optimization (AERO) fulfills the 142 requirements of Distributed Mobility Management (DMM) [RFC7333] and 143 route optimization [RFC5522] for aeronautical networking and other 144 network mobility use cases including intelligent transportation 145 systems and enterprise mobile device users. AERO is a secure 146 internetworking and mobility management service that employs the 147 Overlay Multilink Network Interface (OMNI) [I-D.templin-6man-omni] 148 Non-Broadcast, Multiple Access (NBMA) virtual link model. The OMNI 149 link is a virtual overlay configured over one or more underlying 150 Internetworks, and nodes on the link can exchange original IP packets 151 as single-hop neighbors. The OMNI Adaptation Layer (OAL) supports 152 multilink operation for increased reliability and path optimization 153 while providing fragmentation and reassembly services to support 154 Maximum Transmission Unit (MTU) diversity. In terms of precedence, 155 this specification may provide first-principle insights into a 156 representative mobility service architecture as context for 157 understanding the OMNI specification. 159 The AERO service comprises Clients, Proxy/Servers and Relays that are 160 seen as OMNI link neighbors as well as Bridges that interconnect 161 diverse Internetworks as OMNI link segments through OAL forwarding at 162 a layer below IP. Each node's OMNI interface uses an IPv6 link-local 163 address format that supports operation of the IPv6 Neighbor Discovery 164 (IPv6 ND) protocol [RFC4861]. A node's OMNI interface can be 165 configured over multiple underlying interfaces, and therefore appears 166 as a single interface with multiple link-layer addresses. Each link- 167 layer address is subject to change due to mobility and/or multilink 168 fluctuations, and link-layer address changes are signaled by ND 169 messaging the same as for any IPv6 link. 171 AERO provides a secure cloud-based service where mobile node Clients 172 may use Proxy/Servers acting as default routers and mobility anchor 173 points while fixed nodes may use any Relay on the link for efficient 174 communications. Fixed nodes forward original IP packets destined to 175 other AERO nodes via the nearest Relay, which forwards them through 176 the cloud. Mobile node Clients discover shortest paths to OMNI link 177 neighbors through AERO route optimization. Both unicast and 178 multicast communications are supported, and Clients may efficiently 179 move between locations while maintaining continuous communications 180 with correspondents and without changing their IP Address. 182 AERO Bridges peer with Proxy/Servers in a secured private BGP overlay 183 routing instance to establish a Segment Routing Topology (SRT) 184 spanning tree over the underlying Internetworks of multiple disjoint 185 administrative domains as a single unified OMNI link. Each OMNI link 186 instance is characterized by the set of Mobility Service Prefixes 187 (MSPs) common to all mobile nodes. Relays provide an optimal route 188 from (fixed) correspondent nodes on the underlying Internetwork to 189 (mobile or fixed) nodes on the OMNI link. To the underlying 190 Internetwork, the Relay is the source of a route to the MSP; hence 191 uplink traffic to the mobile node is naturally routed to the nearest 192 Relay. 194 AERO can be used with OMNI links that span private-use Internetworks 195 and/or public Internetworks such as the global Internet. In the 196 latter case, some end systems may be located behind global Internet 197 Network Address Translators (NATs). A means for robust traversal of 198 NATs while avoiding "triangle routing" and critical infrastructure 199 traffic concentration is therefore provided. 201 AERO assumes the use of PIM Sparse Mode in support of multicast 202 communication. In support of Source Specific Multicast (SSM) when a 203 Mobile Node is the source, AERO route optimization ensures that a 204 shortest-path multicast tree is established with provisions for 205 mobility and multilink operation. In all other multicast scenarios 206 there are no AERO dependencies. 208 AERO provides a secure aeronautical internetworking service for both 209 manned and unmanned aircraft, where the aircraft is treated as a 210 mobile node that can connect an Internet of Things (IoT). AERO is 211 also applicable to a wide variety of other use cases. For example, 212 it can be used to coordinate the links of mobile nodes (e.g., 213 cellphones, tablets, laptop computers, etc.) that connect into a home 214 enterprise network via public access networks with VPN or non-VPN 215 services enabled according to the appropriate security model. AERO 216 can also be used to facilitate terrestrial vehicular and urban air 217 mobility (as well as pedestrian communication services) for future 218 intelligent transportation systems 219 [I-D.ietf-ipwave-vehicular-networking][I-D.templin-ipwave-uam-its]. 220 Other applicable use cases are also in scope. 222 Along with OMNI, AERO provides secured optimal routing support for 223 the "6M's" of modern Internetworking, including: 225 1. Multilink - a mobile node's ability to coordinate multiple 226 diverse underlying data links as a single logical unit (i.e., the 227 OMNI interface) to achieve the required communications 228 performance and reliability objectives. 230 2. Multinet - the ability to span the OMNI link over a segment 231 routing topology with multiple diverse network administrative 232 domains while maintaining seamless end-to-end communications 233 between mobile Clients and correspondents such as air traffic 234 controllers, fleet administrators, etc. 236 3. Mobility - a mobile node's ability to change network points of 237 attachment (e.g., moving between wireless base stations) which 238 may result in an underlying interface address change, but without 239 disruptions to ongoing communication sessions with peers over the 240 OMNI link. 242 4. Multicast - the ability to send a single network transmission 243 that reaches multiple nodes belonging to the same interest group, 244 but without disturbing other nodes not subscribed to the interest 245 group. 247 5. Multihop - a mobile node vehicle-to-vehicle relaying capability 248 useful when multiple forwarding hops between vehicles may be 249 necessary to "reach back" to an infrastructure access point 250 connection to the OMNI link. 252 6. MTU assurance - the ability to deliver packets of various robust 253 sizes between peers without loss due to a link size restriction, 254 and to dynamically adjust packets sizes to achieve the optimal 255 performance for each independent traffic flow. 257 The following numbered sections present the AERO specification. The 258 appendices at the end of the document are non-normative. 260 2. Terminology 262 The terminology in the normative references applies; especially, the 263 terminology in the OMNI specification [I-D.templin-6man-omni] is used 264 extensively throughout. The following terms are defined within the 265 scope of this document: 267 IPv6 Neighbor Discovery (IPv6 ND) 268 a control message service for coordinating neighbor relationships 269 between nodes connected to a common link. AERO uses the IPv6 ND 270 messaging service specified in [RFC4861]. 272 IPv6 Prefix Delegation 273 a networking service for delegating IPv6 prefixes to nodes on the 274 link. The nominal service is DHCPv6 [RFC8415], however alternate 275 services (e.g., based on IPv6 ND messaging) are also in scope. A 276 minimal form of prefix delegation known as "prefix registration" 277 can be used if the Client knows its prefix in advance and can 278 represent it in the source address of an IPv6 ND message. 280 Access Network (ANET) 281 a node's first-hop data link service network (e.g., a radio access 282 network, cellular service provider network, corporate enterprise 283 network, etc.) that often provides link-layer security services 284 such as IEEE 802.1X and physical-layer security (e.g., "protected 285 spectrum") to prevent unauthorized access internally and with 286 border network-layer security services such as firewalls and 287 proxys that prevent unauthorized outside access. 289 ANET interface 290 a node's attachment to a link in an ANET. 292 Internetwork (INET) 293 a network topology with a coherent IP routing and addressing plan 294 and that provides a transit backbone service for its connected end 295 systems. INETs also provide an underlay service over which the 296 AERO virtual link is configured. Example INETs include corporate 297 enterprise networks, aviation networks, and the public Internet 298 itself. When there is no administrative boundary between an ANET 299 and the INET, the ANET and INET are one and the same. 301 INET interface 302 a node's attachment to a link in an INET. 304 *NET 305 a "wildcard" term referring to either ANET or INET when it is not 306 necessary to draw a distinction between the two. 308 *NET interface 309 a node's attachment to a link in a *NET. 311 *NET Partition 312 frequently, *NETs such as large corporate enterprise networks are 313 sub-divided internally into separate isolated partitions (a 314 technique also known as "network segmentation"). Each partition 315 is fully connected internally but disconnected from other 316 partitions, and there is no requirement that separate partitions 317 maintain consistent Internet Protocol and/or addressing plans. 318 (Each *NET partition is seen as a separate OMNI link segment as 319 discussed below.) 321 *NET address 322 an IP address assigned to a node's interface connection to a *NET. 324 *NET encapsulation 325 the encapsulation of a packet in an outer header or headers that 326 can be routed within the scope of the local *NET partition. 328 OMNI link 329 the same as defined in [I-D.templin-6man-omni]. The OMNI link 330 employs IPv6 encapsulation [RFC2473] to traverse intermediate 331 nodes in a spanning tree over underlying *NET segments the same as 332 a bridged campus LAN. AERO nodes on the OMNI link appear as 333 single-hop neighbors at the network layer even though they may be 334 separated by many underlying *NET hops; AERO nodes can employ 335 Segment Routing [RFC8402] to navigate between different OMNI 336 links, and/or to cause packets to visit selected waypoints within 337 the same OMNI link. 339 OMNI Interface 340 a node's attachment to an OMNI link. Since OMNI interface 341 addresses are managed for uniqueness, OMNI interfaces do not 342 require Duplicate Address Detection (DAD) and therefore set the 343 administrative variable 'DupAddrDetectTransmits' to zero 344 [RFC4862]. 346 OMNI Adaptation Layer (OAL) 347 an OMNI interface service that subjects original IP packets 348 admitted into the interface to mid-layer IPv6 header encapsulation 349 followed by fragmentation and reassembly. The OAL is also 350 responsible for generating MTU-related control messages as 351 necessary, and for providing addressing context for spanning 352 multiple segments of a bridged OMNI link. 354 original IP packet 355 a whole IP packet or fragment admitted into the OMNI interface by 356 the network layer prior to OAL encapsulation and fragmentation, or 357 an IP packet delivered to the network layer by the OMNI interface 358 following OAL decapsulation and reassembly. 360 OAL packet 361 an original IP packet encapsulated in OAL headers and trailers 362 before OAL fragmentation, or following OAL reassembly. 364 OAL fragment 365 a portion of an OAL packet following fragmentation but prior to 366 *NET encapsulation, or following *NET encapsulation but prior to 367 OAL reassembly. 369 (OAL) atomic fragment 370 an OAL packet that can be forwarded without fragmentation, but 371 still includes a Fragment Header with a valid Identification value 372 and with Fragment Offset and More Fragments both set to 0. 374 (OAL) carrier packet 375 an encapsulated OAL fragment following *NET encapsulation or prior 376 to *NET decapsulation. OAL sources and destinations exchange 377 carrier packets over underlying interfaces, and may be separated 378 by one or more OAL intermediate nodes. OAL intermediate nodes re- 379 encapsulate carrier packets during forwarding by removing the *NET 380 headers of the previous hop underlying network and replacing them 381 with new *NET headers for the next hop underlying network. 383 OAL source 384 an OMNI interface acts as an OAL source when it encapsulates 385 original IP packets to form OAL packets, then performs OAL 386 fragmentation and *NET encapsulation to create carrier packets. 388 OAL destination 389 an OMNI interface acts as an OAL destination when it decapsulates 390 carrier packets, then performs OAL reassembly and decapsulation to 391 derive the original IP packet. 393 OAL intermediate node 394 an OMNI interface acts as an OAL intermediate node when it removes 395 the *NET headers of carrier packets received from a first hop, 396 then re-encapsulates the carrier packets in new *NET headers and 397 forwards them to the next hop. OAL intermediate nodes decrement 398 the Hop Limit of the OAL IPv6 header during re-encapsulation, and 399 discard the packet if the Hop Limit reaches 0. OAL intermediate 400 nodes do not decrement the Hop Limit/TTL of the original IP 401 packet. 403 underlying interface 404 a *NET interface over which an OMNI interface is configured. 406 Mobility Service Prefix (MSP) 407 an aggregated IP Global Unicast Address (GUA) prefix (e.g., 408 2001:db8::/32, 192.0.2.0/24, etc.) assigned to the OMNI link and 409 from which more-specific Mobile Network Prefixes (MNPs) are 410 delegated. OMNI link administrators typically obtain MSPs from an 411 Internet address registry, however private-use prefixes can 412 alternatively be used subject to certain limitations (see: 413 [I-D.templin-6man-omni]). OMNI links that connect to the global 414 Internet advertise their MSPs to their interdomain routing peers. 416 Mobile Network Prefix (MNP) 417 a longer IP prefix delegated from an MSP (e.g., 418 2001:db8:1000:2000::/56, 192.0.2.8/30, etc.) and delegated to an 419 AERO Client or Relay. 421 Mobile Network Prefix Link Local Address (MNP-LLA) 422 an IPv6 Link Local Address that embeds the most significant 64 423 bits of an MNP in the lower 64 bits of fe80::/64, as specified in 424 [I-D.templin-6man-omni]. 426 Mobile Network Prefix Unique Local Address (MNP-ULA) 427 an IPv6 Unique-Local Address derived from an MNP-LLA. 429 Administrative Link Local Address (ADM-LLA) 430 an IPv6 Link Local Address that embeds a 32-bit administratively- 431 assigned identification value in the lower 32 bits of fe80::/96, 432 as specified in [I-D.templin-6man-omni]. 434 Administrative Unique Local Address (ADM-ULA) 435 an IPv6 Unique-Local Address derived from an ADM-LLA. 437 AERO node 438 a node that is connected to an OMNI link and participates in the 439 AERO internetworking and mobility service. 441 AERO Client ("Client") 442 an AERO node that connects over one or more underlying interfaces 443 and requests MNP delegation/registration service from AERO Proxy/ 444 Servers. The Client assigns an MNP-LLA to the OMNI interface for 445 use in IPv6 ND exchanges with other AERO nodes and forwards 446 original IP packets to correspondents according to OMNI interface 447 neighbor cache state. 449 AERO Proxy/Server ("Proxy/Server") 450 a node that provides a proxying service between AERO Clients and 451 external peers on its Client-facing ANET interfaces (i.e., in the 452 same fashion as for an enterprise network proxy) as well as 453 default forwarding and mobility anchor point services for 454 coordination with correspondents on its INET-facing interfaces. 455 (Proxy/Servers in the open INET instead configure only an INET 456 interface and no ANET interfaces.) The Proxy/Server configures an 457 OMNI interface and assigns an ADM-LLA to support the operation of 458 IPv6 ND services, while advertising all of its associated MNPs via 459 BGP peerings with Bridges. 461 AERO Relay ("Relay") 462 a Proxy/Server that provides forwarding services between nodes 463 reached via the OMNI link and correspondents on other links/ 464 networks. AERO Relays configure an OMNI interface and assign an 465 ADM-LLA the same as Proxy/Servers, and also run a dynamic routing 466 protocol to discover any non-MNP IP GUA routes in service on its 467 other links/networks. The Relay advertises the MSP(s) to its 468 other links/networks, and redistributes routes discovered on other 469 links/networks into the OMNI link routing system the same as for 470 Proxy/Servers. 472 AERO Bridge ("Bridge") 473 a BGP hub autonomous system node that also provides OAL forwarding 474 services for nodes on an OMNI link. Bridges forwards carrier 475 packets between OMNI link segments as OAL intermediate nodes while 476 decrementing the OAL IPv6 header Hop Limit but without 477 decrementing the network layer IP TTL/Hop Limit. Bridges peer 478 with Proxy/Servers and other Bridges to form a spanning tree over 479 all OMNI link segments and to discover the set of all MNP and non- 480 MNP prefixes in service. Bridges process carrier packets received 481 over the secured spanning tree that are addressed to themselves, 482 while forwarding all other carrier packets to the next hop also 483 via the secured spanning tree. Bridges forward carrier packets 484 received over the unsecured spanning tree to the next hop either 485 via the unsecured spanning tree or via direct encapsulation if the 486 next hop is on the same OMNI link segment. 488 Hub Proxy/Server 489 a single Proxy/Server selected by the Client that provides a 490 designated router and mobility anchor point service for all of the 491 Client's underlying interfaces. Clients normally select the first 492 FHS Proxy/Server they coordinate with to serve in the Hub role, as 493 all FHS Proxy/Servers are equally capable candidates to serve in 494 that capacity. 496 First-Hop Segment (FHS) Proxy/Server 497 a Proxy/Server for an underlying interface of the source Client 498 that forwards packets sent by the source Client over that 499 interface into the segment routing topology. FHS Proxy/Servers 500 act as intermediate forwarding nodes to facilitate RS/RA exchanges 501 between a Client and its Hub Proxy/Server. 503 Last-Hop Segment (LHS) Proxy/Server 504 a Proxy/Server for an underlying interface of the target Client 505 that forwards packets received from the segment routing topology 506 to the target Client over that interface. 508 Segment Routing Topology (SRT) 509 a multinet OMNI link forwarding region between the FHS Proxy/ 510 Server and LHS Proxy/Server. FHS/LHS Proxy/Servers and SRT 511 Bridges span the OMNI link on behalf of source/target Client 512 pairs. The SRT maintains a spanning tree established through BGP 513 peerings between Bridges and Proxy/Servers. Each SRT segment 514 includes Bridges in a "hub" and Proxy/Servers in "spokes", while 515 adjacent segments are interconnected by Bridge-Bridge peerings. 516 The BGP peerings are configured over both secured and unsecured 517 underlying network paths such that a secured spanning tree is 518 available for critical control messages while other messages can 519 use the unsecured spanning tree. 521 link-layer address 522 an IP address used as an encapsulation header source or 523 destination address from the perspective of the OMNI interface. 524 When an upper layer protocol (e.g., UDP) is used as part of the 525 encapsulation, the port number is also considered as part of the 526 link-layer address. 528 network layer address 529 the source or destination address of an original IP packet 530 presented to the OMNI interface. 532 end user network (EUN) 533 an internal virtual or external edge IP network that an AERO 534 Client or Relay connects to the rest of the network via the OMNI 535 interface. The Client/Relay sees each EUN as a "downstream" 536 network, and sees the OMNI interface as the point of attachment to 537 the "upstream" network. 539 Mobile Node (MN) 540 an AERO Client and all of its downstream-attached networks that 541 move together as a single unit, i.e., an end system that connects 542 an Internet of Things. 544 Mobile Router (MR) 545 a MN's on-board router that forwards original IP packets between 546 any downstream-attached networks and the OMNI link. The MR is the 547 MN entity that hosts the AERO Client. 549 Route Optimization Source (ROS) 550 the AERO node nearest the source that initiates route 551 optimization. The ROS may be a FHS Proxy/Server or Relay for the 552 source, or may be the source Client itself. 554 Route Optimization responder (ROR) 555 the AERO node that responds to route optimization requests on 556 behalf of the target. The ROR may be a Proxy/Server for a target 557 MNP Client or a Relay for a non-MNP target. 559 Potential Router List (PRL) 560 a geographically and/or topologically referenced list of addresses 561 of all Proxy/Servers within the same OMNI link. Each OMNI link 562 has its own PRL. 564 Distributed Mobility Management (DMM) 565 a BGP-based overlay routing service coordinated by Proxy/Servers 566 and Bridges that tracks all Proxy/Server-to-Client associations. 568 Mobility Service (MS) 569 the collective set of all Proxy/Servers, Bridges and Relays that 570 provide the AERO Service to Clients. 572 Mobility Service Endpoint MSE) 573 an individual Proxy/Server, Bridge or Relay in the Mobility 574 Service. 576 Multilink Forwarding Information Base (MFIB) 577 A forwarding table on each AERO/OMNI source, destination and 578 intermediate node that includes Multilink Forwarding Vectors (MFV) 579 with both next hop forwarding instructions and context for 580 reconstructing compressed headers for specific underlying 581 interface pairs used to communicate with peers. 583 Multilink Forwarding Vector (MFV) 584 An MFIB entry that includes soft state for each underlying 585 interface pairwise communication session between peer OMNI nodes. 586 MFVs are identified by both a next-hop and previous-hop MFV Index 587 (MFVI), with the next-hop established based on an IPv6 ND 588 solicitation and the previous hop established based on the 589 solicited IPv6 ND advertisement response. 591 Multilink Forwarding Vector Index (MVFI) 592 A 4 octet value selected by an AERO/OMNI node when it creates an 593 MFV, then advertised to either a next-hop or previous-hop. AERO/ 594 OMNI intermediate nodes assign two distinct local MFVIs for each 595 MFV and advertise one to the next-hop and the other to the 596 previous-hop. AERO/OMNI end systems assign and advertise a single 597 MFVI. AERO/OMNI nodes also discover the remote MFVIs advertised 598 by other nodes that indicate a value the other node is willing to 599 accept. 601 Throughout the document, the simple terms "Client", "Proxy/Server", 602 "Bridge" and "Relay" refer to "AERO Client", "AERO Proxy/Server", 603 "AERO Bridge" and "AERO Relay", respectively. Capitalization is used 604 to distinguish these terms from other common Internetworking uses in 605 which they appear without capitalization. 607 The terminology of IPv6 ND [RFC4861] and DHCPv6 [RFC8415] (including 608 the names of node variables, messages and protocol constants) is used 609 throughout this document. The terms "All-Routers multicast", "All- 610 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 611 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 612 generically refer to either Internet Protocol version, i.e., IPv4 613 [RFC0791] or IPv6 [RFC8200]. 615 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 616 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 617 "OPTIONAL" in this document are to be interpreted as described in BCP 618 14 [RFC2119][RFC8174] when, and only when, they appear in all 619 capitals, as shown here. 621 3. Automatic Extended Route Optimization (AERO) 623 The following sections specify the operation of IP over OMNI links 624 using the AERO service: 626 3.1. AERO Node Types 628 AERO Clients can be deployed as fixed infrastructure nodes close to 629 end systems, or as Mobile Nodes (MNs) that can change their network 630 attachment points dynamically. AERO Clients configure OMNI 631 interfaces over underlying interfaces with addresses that may change 632 due to mobility. AERO Clients register their Mobile Network Prefixes 633 (MNPs) with the AERO service, and distribute the MNPs to nodes on 634 EUNs. AERO Bridges, Proxy/Servers and Relays are critical 635 infrastructure elements in fixed (i.e., non-mobile) INET deployments 636 and hence have permanent and unchanging INET addresses. Together, 637 they constitute the AERO service which provides an OMNI link virtual 638 overlay for connecting AERO Clients. 640 AERO Bridges (together with Proxy/Servers) provide the secured 641 backbone supporting infrastructure for a Segment Routing Topology 642 (SRT) spanning tree for the OMNI link. Bridges forward carrier 643 packets both within the same SRT segment and between disjoint SRT 644 segments based on an IPv6 encapsulation mid-layer known as the OMNI 645 Adaptation Layer (OAL) [I-D.templin-6man-omni]. The OMNI interface 646 and OAL provide a virtual bridging service, since the inner IP TTL/ 647 Hop Limit is not decremented. Each Bridge also peers with Proxy/ 648 Servers and other Bridges in a dynamic routing protocol instance to 649 provide a Distributed Mobility Management (DMM) service for the list 650 of active MNPs (see Section 3.2.3). Bridges present the OMNI link as 651 a set of one or more Mobility Service Prefixes (MSPs) and configure 652 secured tunnels with Proxy/Servers, Relays and other Bridges; they 653 further maintain forwarding table entries for each MNP or non-MNP 654 prefix in service on the OMNI link. 656 AERO Proxy/Servers in distributed SRT segments provide default 657 forwarding and mobility/multilink services for AERO Client mobile 658 nodes. Each Proxy/Server also peers with Bridges in a dynamic 659 routing protocol instance to advertise its list of associated MNPs 660 (see Section 3.2.3). Hub Proxy/Servers provide prefix delegation/ 661 registration services and track the mobility/multilink profiles of 662 each of their associated Clients, where each delegated prefix becomes 663 an MNP taken from an MSP. Proxy/Servers at ANET/INET boundaries 664 provide a forwarding service for ANET Clients to communicate with 665 peers in external INETs while Proxy/Servers in the open INET provide 666 an authentication service for INET Client IPv6 ND messages but 667 limited forwarding services. Source Clients securely coordinate with 668 target Clients by sending control messages via a First-Hop Segment 669 (FHS) Proxy/Server which forwards them over the SRT spanning tree to 670 a Last-Hop Segment (LHS) Proxy/Server which finally forwards them to 671 the target. 673 AERO Relays are Proxy/Servers that provide forwarding services to 674 exchange original IP packets between the OMNI link and other links/ 675 networks. Relays run a dynamic routing protocol to discover any non- 676 MNP prefixes in service on other links/networks. The Relay 677 redistributes OMNI link MSP(s) into other links/networks, and 678 redistributes non-MNP prefixes via OMNI link Bridge BGP peerings. 680 3.2. The AERO Service over OMNI Links 682 3.2.1. AERO/OMNI Reference Model 684 Figure 1 presents the basic OMNI link reference model: 686 +----------------+ 687 | AERO Bridge B1 | 688 | Nbr: S1, S2, P1| 689 |(X1->S1; X2->S2)| 690 | MSP M1 | 691 +-------+--------+ 692 +--------------+ | +--------------+ 693 | AERO P/S S1 | | | AERO P/S S2 | 694 | Nbr: C1, B1 | | | Nbr: C2, B1 | 695 | default->B1 | | | default->B1 | 696 | X1->C1 | | | X2->C2 | 697 +-------+------+ | +------+-------+ 698 | OMNI link | | 699 X===+===+==================+===================+===+===X 700 | | 701 +-----+--------+ +--------+-----+ 702 |AERO Client C1| |AERO Client C2| 703 | Nbr: S1 | | Nbr: S2 | 704 | default->S1 | | default->S2 | 705 | MNP X1 | | MNP X2 | 706 +------+-------+ +-----+--------+ 707 | | 708 .-. .-. 709 ,-( _)-. ,-( _)-. 710 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 711 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 712 `-(______)-' +-------+ +-------+ `-(______)-' 714 Figure 1: AERO/OMNI Reference Model 716 In this model: 718 o the OMNI link is an overlay network service configured over one or 719 more underlying SRT segments which may be managed by different 720 administrative authorities and have incompatible protocols and/or 721 addressing plans. 723 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 724 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 725 via BGP peerings over secured tunnels to Proxy/Servers (S1, S2). 726 Bridges provide the backbone for an SRT spanning tree for the OMNI 727 link. 729 o AERO Proxy/Servers S1 and S2 configure secured tunnels with Bridge 730 B1 and also provide mobility, multilink, multicast and default 731 router services for the MNPs of their associated Clients C1 and 732 C2. (Proxy/Servers that act as Relays can also advertise non-MNP 733 routes for non-mobile correspondent nodes the same as for MNP 734 Clients.) 736 o AERO Clients C1 and C2 associate with Proxy/Servers S1 and S2, 737 respectively. They receive MNP delegations X1 and X2, and also 738 act as default routers for their associated physical or internal 739 virtual EUNs. Simple hosts H1 and H2 attach to the EUNs served by 740 Clients C1 and C2, respectively. 742 An OMNI link configured over a single *NET appears as a single 743 unified link with a consistent underlying network addressing plan; 744 all nodes on the link can exchange carrier packets via simple *NET 745 encapsulation (i.e., following any necessary NAT traversal) since the 746 underlying *NET is connected. In common practice, however, OMNI 747 links are often configured over an SRT spanning tree that bridges 748 multiple distinct *NET segments managed under different 749 administrative authorities (e.g., as for worldwide aviation service 750 providers such as ARINC, SITA, Inmarsat, etc.). Individual *NETs may 751 also be partitioned internally, in which case each internal partition 752 appears as a separate segment. 754 The addressing plan of each SRT segment is consistent internally but 755 will often bear no relation to the addressing plans of other 756 segments. Each segment is also likely to be separated from others by 757 network security devices (e.g., firewalls, proxys, packet filtering 758 gateways, etc.), and disjoint segments often have no common physical 759 link connections. Therefore, nodes can only be assured of exchanging 760 carrier packets directly with correspondents in the same segment, and 761 not with those in other segments. The only means for joining the 762 segments therefore is through inter-domain peerings between AERO 763 Bridges. 765 The OMNI link spans multi-segment SRT topologies using the OMNI 766 Adaptation Layer (OAL) [I-D.templin-6man-omni] to provide the network 767 layer with a virtual abstraction similar to a bridged campus LAN. 769 The OAL is an OMNI interface sublayer that inserts a mid-layer IPv6 770 encapsulation header for inter-segment forwarding (i.e., bridging) 771 without decrementing the network-layer TTL/Hop Limit of the original 772 IP packet. An example OMNI link SRT is shown in Figure 2: 774 . . . . . . . . . . . . . . . . . . . . . . . 775 . . 776 . .-(::::::::) . 777 . .-(::::::::::::)-. +-+ . 778 . (:::: Segment A :::)--|B|---+ . 779 . `-(::::::::::::)-' +-+ | . 780 . `-(::::::)-' | . 781 . | . 782 . .-(::::::::) | . 783 . .-(::::::::::::)-. +-+ | . 784 . (:::: Segment B :::)--|B|---+ . 785 . `-(::::::::::::)-' +-+ | . 786 . `-(::::::)-' | . 787 . | . 788 . .-(::::::::) | . 789 . .-(::::::::::::)-. +-+ | . 790 . (:::: Segment C :::)--|B|---+ . 791 . `-(::::::::::::)-' +-+ | . 792 . `-(::::::)-' | . 793 . | . 794 . ..(etc).. x . 795 . . 796 . . 797 . <- Segment Routing Topology (SRT) -> . 798 . . . . . . . . . . . . . .. . . . . . . . . 800 Figure 2: OMNI Link Segment Routing Topology (SRT) 802 Bridge, Proxy/Server and Relay OMNI interfaces are configured over 803 both secured tunnels and open INET underlying interfaces within their 804 respective SRT segments. Within each segment, Bridges configure 805 "hub-and-spokes" BGP peerings with Proxy/Server/Relays as "spokes". 806 Adjacent SRT segments are joined by Bridge-to-Bridge peerings to 807 collectively form a spanning tree over the entire SRT. The "secured" 808 spanning tree supports authentication and integrity for critical 809 control plane messages. The "unsecured" spanning tree conveys 810 ordinary carrier packets without security codes and that must be 811 treated by destinations according to data origin authentication 812 procedures. AERO nodes can employ route optimization to cause 813 carrier packets to take more direct paths between OMNI link neighbors 814 without having to follow strict spanning tree paths. 816 3.2.2. Addressing and Node Identification 818 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 819 fe80::/64 [RFC4291] to assign LLAs used for network-layer addresses 820 in link-scoped IPv6 ND and data messages. AERO Clients use LLAs 821 constructed from MNPs (i.e., "MNP-LLAs") while other AERO nodes use 822 LLAs constructed from administrative identification values ("ADM- 823 LLAs") as specified in [I-D.templin-6man-omni]. Non-MNP routes are 824 also represented the same as for MNP-LLAs, but may include a prefix 825 that is not properly covered by an MSP. 827 AERO nodes also use the Unique Local Address (ULA) prefix fd00::/8 828 followed by a pseudo-random 40-bit OMNI domain identifier to form the 829 prefix [ULA]::/48, then include a 16-bit OMNI link identifier '*' to 830 form the prefix [ULA*]::/64 [RFC4291]. The AERO node then uses the 831 prefix [ULA*]::/64 to form "MNP-ULAs" or "ADM-ULA"s as specified in 832 [I-D.templin-6man-omni] to support OAL addressing. (The prefix 833 [ULA*]::/64 appearing alone and with no suffix represents "default".) 834 AERO Clients also use Temporary ULAs constructed per 835 [I-D.templin-6man-omni], where the addresses are typically used only 836 in initial control message exchanges until a stable MNP-LLA/ULA is 837 assigned. 839 AERO MSPs, MNPs and non-MNP routes are typically based on Global 840 Unicast Addresses (GUAs), but in some cases may be based on private- 841 use addresses. A GUA block is also reserved for OMNI link anycast 842 purposes. See [I-D.templin-6man-omni] for a full specification of 843 LLAs, ULAs and GUAs used by AERO nodes on OMNI links. 845 Finally, AERO Clients and Proxy/Servers configure node identification 846 values as specified in [I-D.templin-6man-omni]. 848 3.2.3. AERO Routing System 850 The AERO routing system comprises a private Border Gateway Protocol 851 (BGP) [RFC4271] service coordinated between Bridges and Proxy/ 852 Servers. The service supports carrier packet forwarding at a layer 853 below IP and does not interact with the public Internet BGP routing 854 system, but supports redistribution of information for other links 855 and networks discovered by Relays. 857 In a reference deployment, each Proxy/Server is configured as an 858 Autonomous System Border Router (ASBR) for a stub Autonomous System 859 (AS) using a 32-bit AS Number (ASN) [RFC4271] that is unique within 860 the BGP instance, and each Proxy/Server further uses eBGP to peer 861 with one or more Bridges but does not peer with other Proxy/Servers. 862 Each SRT segment in the OMNI link must include one or more Bridges in 863 a "hub" AS, which peer with the Proxy/Servers within that segment as 864 "spoke" ASes. All Bridges within the same segment are members of the 865 same hub AS, and use iBGP to maintain a consistent view of all active 866 routes currently in service. The Bridges of different segments peer 867 with one another using eBGP. 869 Bridges maintain forwarding table entries only for the MNP-ULAs 870 corresponding to MNP and non-MNP routes that are currently active, 871 while carrier packets destined to all other MNP-ULAs are dropped with 872 a Destination Unreachable message returned due to the black-hole 873 route. In this way, Proxy/Servers and Relays have only partial 874 topology knowledge (i.e., they only maintain routing information for 875 their directly associated Clients and non-AERO links) and they 876 forward all other carrier packets to Bridges which have full topology 877 knowledge. 879 Each OMNI link segment assigns a unique ADM-ULA sub-prefix of 880 [ULA*]::/96 known as the "SRT prefix". For example, a first segment 881 could assign [ULA*]::1000/116, a second could assign 882 [ULA*]::2000/116, a third could assign [ULA*]::3000/116, etc. Within 883 each segment, each Proxy/Server configures an ADM-ULA within the 884 segment's SRT prefix, e.g., the Proxy/Servers within [ULA*]::2000/116 885 could assign the ADM-ULAs [ULA*]::2011/116, [ULA*]::2026/116, 886 [ULA*]::2003/116, etc. 888 The administrative authorities for each segment must therefore 889 coordinate to assure mutually-exclusive ADM-ULA prefix assignments, 890 but internal provisioning of ADM-ULAs an independent local 891 consideration for each administrative authority. For each ADM-ULA 892 prefix, the Bridge(s) that connect that segment assign the all-zero's 893 address of the prefix as a Subnet Router Anycast address. For 894 example, the Subnet Router Anycast address for [ULA*]::1023/116 is 895 simply [ULA*]::1000. 897 ADM-ULA prefixes are statically represented in Bridge forwarding 898 tables. Bridges join multiple SRT segments into a unified OMNI link 899 over multiple diverse network administrative domains. They support a 900 virtual bridging service by first establishing forwarding table 901 entries for their ADM-ULA prefixes either via standard BGP routing or 902 static routes. For example, if three Bridges ('A', 'B' and 'C') from 903 different segments serviced [ULA*]::1000/116, [ULA*]::2000/116 and 904 [ULA*]::3000/116 respectively, then the forwarding tables in each 905 Bridge are as follows: 907 A: [ULA*]::1000/116->local, [ULA*]::2000/116->B, [ULA*]::3000/116->C 909 B: [ULA*]::1000/116->A, [ULA*]::2000/116->local, [ULA*]::3000/116->C 911 C: [ULA*]::1000/116->A, [ULA*]::2000/116->B, [ULA*]::3000/116->local 912 These forwarding table entries rarely change, since they correspond 913 to fixed infrastructure elements in their respective segments. 915 MNP (and non-MNP) ULAs are instead dynamically advertised in the AERO 916 routing system by Proxy/Servers and Relays that provide service for 917 their corresponding MNPs. For example, if three Proxy/Servers ('D', 918 'E' and 'F') service the MNPs 2001:db8:1000:2000::/56, 919 2001:db8:3000:4000::/56 and 2001:db8:5000:6000::/56 then the routing 920 system would include: 922 D: [ULA*]:2001:db8:1000:2000/120 924 E: [ULA*]:2001:db8:3000:4000/120 926 F: [ULA*]:2001:db8:5000:6000/120 928 A full discussion of the BGP-based routing system used by AERO is 929 found in [I-D.ietf-rtgwg-atn-bgp]. 931 3.2.4. OMNI Link Forwarding 933 When the network layer forwards an original IP packet into an OMNI 934 interface, the OMNI Adaptation Layer (OAL) encapsulates the packet to 935 produce an OAL packet [I-D.templin-6man-omni]. This OAL source then 936 fragments the OAL packet while including an identical Identification 937 value for each fragment that must be within the window for the LHS 938 Proxy/Server or the target Client itself. The OAL source also 939 includes an identical Compressed Routing Header with 32-bit ID fields 940 (CRH-32) [I-D.bonica-6man-comp-rtg-hdr] with each fragment if 941 necessary as discussed in Section 3.2.7 and Section 3.14. The OAL 942 source finally encapsulates each resulting OAL fragment in *NET 943 headers to form an OAL carrier packet, with source address set to its 944 own *NET address (e.g., 192.0.2.100) and destination set to the *NET 945 address of the next hop OAL intermediate node or destination (e.g., 946 192.0.2.1). 948 The carrier packet encapsulation format in the above example is shown 949 in Figure 3: 951 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 952 | *NET Header | 953 | src = 192.0.2.100 | 954 | dst = 192.0.2.1 | 955 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 956 | OAL IPv6 Header | 957 | src = [ULA*]::2001:db8:1:2 | 958 | dst= [ULA*]::3000:0000 | 959 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 960 | CRH-32 (if necessary) | 961 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 962 | OAL Fragment Header | 963 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 964 | Original IP Header | 965 | (first-fragment only) | 966 | src = 2001:db8:1:2::1 | 967 | dst = 2001:db8:1234:5678::1 | 968 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 969 | | 970 ~ ~ 971 ~ Original Packet Body/Fragment ~ 972 ~ ~ 973 | | 974 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 976 Figure 3: Carrier Packet Format 978 In this format, the original IP header and packet body/fragment are 979 encapsulated in an OAL IPv6 header prepared according to [RFC2473], 980 the CRH-32 is a Routing Header extension of the OAL header, the 981 Fragment Header identifies each fragment, and the *NET header is 982 prepared as discussed in Section 3.6. The OAL source transmits each 983 such carrier packet into the SRT spanning tree, where they are 984 forwarded over possibly multiple OAL intermediate nodes until they 985 arrive at the OAL destination. 987 The OMNI link control plane service distributes both Client MNP-ULA 988 prefix information that may change dynamically due to regional node 989 mobility and per-segment ADM-ULA prefix information that rarely 990 changes. OMNI link Bridges and Proxy/Servers use the information to 991 establish and maintain a forwarding plane spanning tree that connects 992 all nodes on the link. The spanning tree supports a carrier packet 993 virtual bridging service according to link-layer (instead of network- 994 layer) information, but may often include longer paths than 995 necessary. Each OMNI interface therefore also includes a Multilink 996 Forwarding Information Base (MFIB) with Multilink Forwarding Vectors 997 (MFVs) that can often provide "shortcuts" instead of always following 998 strict spanning tree paths. As a result, the spanning tree is always 999 available but OMNI interfaces can often use the MFIB to greatly 1000 improve performance and reduce load on critical infrastructure 1001 elements. 1003 3.2.5. Segment Routing Topologies (SRTs) 1005 The 64-bit sub-prefixes of [ULA]::/48 identify up to 2^16 distinct 1006 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 1007 OMNI link overlay instance using a distinct set of ULAs, and emulates 1008 a bridged campus LAN service for the OMNI link. In some cases (e.g., 1009 when redundant topologies are needed for fault tolerance and 1010 reliability) it may be beneficial to deploy multiple SRTs that act as 1011 independent overlay instances. A communication failure in one 1012 instance therefore will not affect communications in other instances. 1014 Each SRT is identified by a distinct value in bits 48-63 of 1015 [ULA]::/48, i.e., as [ULA0]::/64, [ULA1]::/64, [ULA2]::/64, etc. 1016 Each OMNI interface is identified by a unique interface name (e.g., 1017 omni0, omni1, omni2, etc.) and assigns an OMNI IPv6 anycast address 1018 used for OMNI interface determination in Safety-Based Multilink (SBM) 1019 as discussed in [I-D.templin-6man-omni]. Each OMNI interface further 1020 applies Performance-Based Multilink (PBM) internally. 1022 The Bridges and Proxy/Servers of each independent SRT engage in BGP 1023 peerings to form a spanning tree with the Bridges in non-leaf nodes 1024 and the Proxy/Servers in leaf nodes. The spanning tree is configured 1025 over both secured and unsecured underlying network paths. The 1026 secured spanning tree is used to convey secured control messages 1027 between Hub, FHS and LHS Proxy/Servers, while the unsecured spanning 1028 tree forwards data messages and/or unsecured control messages. 1030 Each SRT segment is identified by a unique ADM-ULA prefix used by all 1031 Proxy/Servers and Bridges in the segment. Each AERO node must 1032 therefore discover an SRT prefix that correspondents can use to 1033 determine the correct segment, and must publish the SRT prefix in 1034 IPv6 ND messages. 1036 3.2.6. Segment Routing For OMNI Link Selection 1038 Original IPv6 source can direct IPv6 packets to an AERO node by 1039 including a standard IPv6 Segment Routing Header (SRH) [RFC8754] with 1040 the OMNI IPv6 anycast address for the selected OMNI link as either 1041 the IPv6 destination or as an intermediate hop within the SRH. This 1042 allows the original source to determine the specific OMNI link SRT an 1043 original IPv6 packet will traverse when there may be multiple 1044 alternatives. 1046 When an AERO node processes the SRH and forwards the original IPv6 1047 packet to the correct OMNI interface, the OMNI interface writes the 1048 next IPv6 address from the SRH into the IPv6 destination address and 1049 decrements Segments Left. If decrementing would cause Segments Left 1050 to become 0, the OMNI interface deletes the SRH before forwarding. 1051 This form of Segment Routing supports Safety-Based Multilink (SBM). 1053 3.2.7. OMNI Multilink Forwarding 1055 OMNI interfaces include a supplemental forwarding table termed the 1056 Multilink Forwarding Information Base (MFIB) that provides shorter 1057 paths for carrier packet forwarding based on OMNI neighbor underlying 1058 interface pairs. The MFIB contains Multilink Forwarding Vectors 1059 (MFVs) indexed by 4-octet values known as MFV Indexes (MFVIs). 1061 OMNI interface "OAL source", "OAL intermediate" and "OAL destination" 1062 nodes create MFVs/MFVIs when they process an IPv6 ND solicitation 1063 message with Job code "00" (Initialize; Build B) or a solicited 1064 advertisement with Job code "01" (Follow B; Build A) (see: 1065 [I-D.templin-6man-omni]). The OAL source of the solicitation (and 1066 OAL destination of the solicited advertisement) are considered to 1067 reside in the "First Hop Segment (FHS)", while the OAL destination of 1068 the solicitation (and OAL source of the solicited advertisement) are 1069 considered to reside in the "Last Hop Segment (LHS)". 1071 When an OAL node processes a solicitation with Job code "00", it 1072 creates an MFV, records the solicitation's source and destination 1073 LLAs and assigns a "B" MFVI. When the "B" MVFI is referenced, the 1074 MVF presents the LLAs in (dst,src) order the opposite of how they 1075 appeared in the original solicitation. 1077 When an OAL node processes a solicited advertisement with Job code 1078 "01", it locates the MFV created by the solicitation and assigns an 1079 "A" MFVI. When the "A" MFVI is referenced, the MFV presents the LLAs 1080 in (src,dst) order the same as they appeared in the original 1081 solicitation. 1083 OAL nodes generate random 32-bit values as candidate A/B MFVIs which 1084 must first be tested for local uniqueness. If a candidate MFVI s 1085 already in use (or if the value is 0), the OAL node repeats the 1086 process until it obtains a unique non-zero value. (Since the number 1087 of MFVs in service at each OAL node is likely to be much smaller than 1088 2**32, the process will generate a unique value after a small number 1089 of tries.) An MFVI generated by a first OAL node SHOULD NOT be 1090 tested for uniqueness on other OAL nodes, since the uniqueness 1091 property is node-local only. 1093 OAL nodes maintain A/B MFVIs as follows: 1095 o "B1" - a locally-unique MFVI maintained independently by each OAL 1096 node on the path from the FHS OAL source to the last LHS 1097 intermediate node before the OAL destination. The OAL node 1098 generates and assigns a "B1" MFVI to a newly-created MFV when it 1099 processes a solicitation message with Job code "00". When the OAL 1100 node receives future carrier packets that include this value, it 1101 can unambiguously locate the correct MFV and determine 1102 directionality without examining addresses. 1104 o "A1" - a locally unique MFVI maintained independently by each OAL 1105 node on the path from the LHS OAL source to the last FHS 1106 intermediate node before the OAL destination. The OAL node 1107 generates and assigns an "A1" MFVI to the MVF that configures the 1108 corresponding "B1" MFVI when it processes a solicited 1109 advertisement message with Job code "01". When the OAL node 1110 receives future carrier packets that include this value, it can 1111 unambiguously locate the correct MFV and determine directionality 1112 without examining addresses. 1114 o "A2" - the A1 MFVI of a remote OAL node discovered by an FHS OAL 1115 source or OAL intermediate node when it processes an advertisement 1116 message with Job code "01" that originated from an LHS OAL source. 1117 A2 values MUST NOT be tested for uniqueness within the OAL node's 1118 local context. 1120 o "B2" - the B1 MFVI of a remote OAL node discovered by an LHS OAL 1121 source or OAL intermediate node when it processes a solicitation 1122 message with Job code "00" that originated from an FHS OAL source. 1123 B2 values MUST NOT be tested for uniqueness within the OAL node's 1124 local context. 1126 When an FHS OAL source has an original IP packet to send to an LHS 1127 OAL destination, (i.e., one for which there is no existing NCE) it 1128 first selects a source and target underlying interface pair. The OAL 1129 source uses its cached information for the target underlying 1130 interface as LHS information then prepares a solicitation message 1131 with an OMNI Multilink Forwarding Parameters sub-option with Job code 1132 "00" and with source set to its own {ADM,MNP}-LLA. If the LHS FMT- 1133 Forward and FMT-Mode bits are both clear, the OAL source sets the 1134 destination to the ADM-LLA of the LHS Proxy/Server; otherwise, it 1135 sets the destination to the MNP-LLA of the target Client. The OAL 1136 source then sets window synchronization information in the OMNI 1137 header and creates a NCE for the selected destination {ADM,MNP}-LLA 1138 in the INCOMPLETE state. The OAL source next creates an MFV based on 1139 the solicitation source and destination LLAs, then generates a "B1" 1140 MFVI and assigns it to the MFV while also including it as the first B 1141 entry in the MFVI List. The OAL source then populates the 1142 solicitation Multilink Forwarding Parameters based on any FHS/LHS 1143 information it knows locally. OAL intermediate nodes on the path to 1144 the OAL destination may populate addtional FHS/LHS information on a 1145 hop-by-hop basis. 1147 If the OAL source is the FHS Proxy/Server, it then performs OAL 1148 encapsulation/fragmentation while setting the source to its own ADM- 1149 ULA and setting the destination to the FHS Subnet Router Anycast ULA 1150 determined by applying the FHS SRT prefix length to its ADM-ULA. The 1151 FHS Proxy/Server next examines the LHS FMT code. If FMT-Forward is 1152 clear and FMT-Mode is set, the FHS Proxy/Server checks for a NCE for 1153 the ADM-LLA of the LHS Proxy/Server. If there is no NCE, the LHS 1154 Proxy/Server creates one in the INCOMPLETE state. If a new NCE was 1155 created (or if the existing NCE requires fresh window 1156 synchronization), the FHS Proxy/Server then writes window 1157 synchronization parameters into the OMNI Multilink Forwarding 1158 Parameters Tunnel Window Synchronization fields. The FHS Proxy/ 1159 Server then selects an appropriate Identification value and *NET 1160 headers and forwards the resulting carrier packets into the secured 1161 spanning tree which will deliver them to a Bridge interface that 1162 assigns the FHS Subnet Router Anycast ULA. 1164 If the OAL source is the FHS Client, it instead includes an 1165 authentication signature if necessary, performs OAL encapsulation/ 1166 fragmentation, sets the source to its own ADM-ULA and sets the 1167 destination to the ADM-ULA of the FHS Proxy/Server. The FHS Client 1168 then selects an appropriate Identification value and *NET headers and 1169 forwards the carrier packets to the FHS Proxy/Server. When the FHS 1170 Proxy/Server receives the carrier packets, it verifies the 1171 Identification, reassembles/decapsulates to obtain the solicitation 1172 then verifies the authentication signature. The FHS Proxy/Server 1173 then creates an MFV (i.e., the same as the FHS Client had done) while 1174 assigning the current B entry in the MFVI List (i.e., the one 1175 included by the FHS Client) as the "B2" MFVI for this MVF. The FHS 1176 Proxy/Server then generates a new unique "B1" MFVI, then both assigns 1177 it to the MFV and writes it as the next B entry in the OMNI Multilink 1178 Forwarding Parameters MFVI List (while also writing any FHS Client 1179 and Proxy/Server addressing information). The FHS Proxy/Server then 1180 checks LHS FMT-Forward/Mode to determine whether to create a NCE for 1181 the LHS Proxy/Server ADM-LLA and include Tunnel Window 1182 Synchronization parameters the same as above. The FHS Proxy/Server 1183 then re-encapsulates/re-fragments while setting the source to its own 1184 ADM-ULA and destination address to the FHS Subnet Router Anycast ULA. 1185 The FHS Proxy/Server finally includes an appropriate Identification 1186 value and *NET headers and forwards the carrier packets into the 1187 secured spanning tree the same as above. 1189 Bridges in the spanning tree forward carrier packets not explicitly 1190 addressed to themselves, while forwarding those that arrived via the 1191 secured spanning tree to the next hop also via the secured spanning 1192 tree and forwarding all others via the unsecured spanning tree. When 1193 an FHS Bridge receives a carrier packet over the secured spanning 1194 tree addressed to its ADM-ULA or the FHS Subnet Router Anycast ULA, 1195 it instead reassembles/decapsulates to obtain the solicitation. The 1196 FHS Bridge next creates an MFV (i.e., the same as the FHS Proxy/ 1197 Server had done) while assigning the current B entry in the MFVI List 1198 as the MFV "B2" index. The FHS Bridge also caches the solicitation 1199 Multilink Forwarding Parameters FHS information in the MFV, and also 1200 caches the first B entry in the MFVI List as "FHS-Client" when FHS 1201 FMT-Forward/Mode are both set to enable future direct forwarding to 1202 this FHS Client. The FHS Bridge then generates a "B1" MFVI for the 1203 MFV and also writes it as the next B entry in the solicitation's MFVI 1204 List. 1206 The FHS Bridge then examines the SRT prefixes corresponding to both 1207 FHS and LHS. If the FHS Bridge has a local interface connection to 1208 both the FHS and LHS (whether they are the same or different 1209 segments), the FHS/LHS Bridge caches the solicitation LHS information 1210 and writes its ADM-ULA suffix and LHS INADDR into the solicitation 1211 OMNI Multilink Forwarding Parameters LHS fields. The FHS/LHS Bridge 1212 then re-encapsulates the solicitation with its own ADM-ULA as the 1213 source and with the ADM-ULA of the LHS Proxy/Server as the 1214 destination. If the FHS and LHS prefixes are different, the FHS 1215 Bridge instead re-encapsulates with its own ADM-ULA as the source and 1216 with the LHS Subnet Router Anycast ULA as the destination. The FHS 1217 Bridge selects an appropriate Identification and *NET headers as 1218 above then forwards the carrier packets into the secured spanning 1219 tree. 1221 When the FHS and LHS Bridges are different, the LHS Bridge will 1222 receive carrier packets over the secured spanning tree from the FHS 1223 Bridge. The LHS Bridge reassembles/decapsulates to obtain the 1224 solicitation then creates an MFV (i.e., the same as the FHS Bridge 1225 had done) while assigning the current B entry in the MFVI List as the 1226 MFV "B2" index. The LHS Bridge also caches the ADM-ULA of the FHS 1227 Bridge as the spanning tree address for "B2", caches the solicitation 1228 Multilink Forwarding Parameters LHS information then generates a "B1" 1229 MFVI for the MFV while also writing it as the next B entry in the 1230 MFVI List. The LHS Bridge also writes its own ADM-ULA suffix and LHS 1231 INADDR into the OMNI Multilink Forwarding Parameters. The LHS Bridge 1232 then re-encapsulates with its own ADM-ULA as the source and the ADM- 1233 ULA of the LHS Proxy/Server as the destination, then selects an 1234 appropriate Identification and *NET headers and forwards the carrier 1235 packets into the secured spanning tree. 1237 When the LHS Proxy/Server receives the carrier packets from the 1238 secured spanning tree, it reassembles/decapsulates to obtain the 1239 solicitation then verifies that the LHS information supplied by the 1240 FHS source is consistent with its own cached information. If the 1241 information is consistent, the LHS Proxy/Server then creates an MFV 1242 and assigns the current B entry in the MFVI List as the "B2" MFVI the 1243 same as for the prior hop. If the solicitation destination is the 1244 MNP-LLA of the target Client, the LHS Proxy/Server also generates a 1245 "B1" MFVI and assigns it both to the MFVI and as the next B entry in 1246 the MFVI List. The LHS Proxy/Server then examines FHS FMT; if FMT- 1247 Forward is clear and FMT-Mode is set, the LHS Proxy/Server creates a 1248 NCE for the ADM-LLA of the FHS Proxy/Server (if necessary) and sets 1249 the state to STALE, then caches any Tunnel Window Synchronization 1250 parameters. 1252 If the solicitation destination is its own ADM-LLA, the LHS Proxy/ 1253 Server next prepares to return a solicited advertisement with Job 1254 code "01". If the solicitation source was the MNP-LLA of the FHS 1255 Client, the LHS Proxy/Server first creates or updates an NCE for the 1256 MNP-LLA with state set to STALE. The LHS Proxy/Server next caches 1257 the solicitation OMNI header window synchronization parameters and 1258 Multilink Forwarding Parameters information (including the MFVI List) 1259 in the NCE corresponding to the LLA source. When the LHS Proxy/ 1260 Server forwards future carrier packets based on the NCE, it can 1261 populate reverse-path forwarding information in a CRH-32 routing 1262 header to enable forwarding based on the cached MFVI List B entries 1263 instead of ULA addresses. 1265 The LHS Proxy/Server then creates an advertisement with Job code "01" 1266 while copying the solicitation OMNI Multilink Forwarding Parameters 1267 FHS/LHS information into the corresponding fields in the 1268 advertisement. The LHS Proxy/Server then generates an "A1" MFVI and 1269 both assigns it to the MFV and includes it as the first A entry in 1270 advertisement's MFVI List (see: [I-D.templin-6man-omni] for details 1271 on MFVI List A/B processing). The LHS Proxy/Server then includes 1272 end-to-end window synchronization parameters in the OMIN header (if 1273 necessary) and also tunnel window synchronization parameters in the 1274 Multilink Forwarding Parameters Tunnel block (if necessary). The LHS 1275 Proxy/Server then encapsulates the advertisement, sets the source to 1276 its own ADM-ULA, sets the destination to the ADM-ULA of the LHS 1277 Bridge, selects an appropriate Identification value and *NET headers 1278 then forwards the carrier packets into the secured spanning tree. 1280 If the solicitation destination was the MNP-LLA of the LHS Client, 1281 the LHS Proxy/Server instead includes an authentication signature in 1282 the solicitation, then re-encapsulates/re-fragments with its own ADM- 1283 ULA as the source and the MNP-ULA of the LHS Client as the 1284 destination. The LHS Proxy/Server then selects an appropriate 1285 Identification value and *NET headers and forwards the carrier 1286 packets to the LHS Client. When the LHS Client receives the carrier 1287 packets, it verifies the Identification and reassembles/decapsulates 1288 to obtain the solicitation. The LHS Client then creates a NCE for 1289 the solicitation LLA source address in the STALE state. If LHS FMT- 1290 Forward is set, FHS FMT-Forward is clear and the solicitation source 1291 was an MNP-LLA, the Client also creates a NCE for the ADM-LLA of the 1292 FHS Proxy/Server in the STALE state and caches any Tunnel Window 1293 Synchronization parameters. The Client then caches the solicitation 1294 OMNI header window synchronization parameters and Multilink 1295 Forwarding Parameters in the NCE corresponding to the solicitation 1296 LLA source, then creates an MFV and assigns both the current MFVI 1297 List B entry as "B2" and a locally generated "A1" MFVI the same as 1298 for previous hops (the LHS Client also includes the "A1" value in the 1299 solicited advertisement - see above and below). The LHS Client also 1300 caches the previous MFVI List B entry as "LHS-Bridge" since it can 1301 include this value when it sends future carrier packets directly to 1302 the Bridge (following appropriate neighbor coordination). 1304 The LHS Client then prepares an advertisement using exactly the same 1305 procedures as for the LHS Proxy/Server above, except that it uses its 1306 MNP-LLA as the source. The LHS Client also includes an 1307 authentication signature, then encapsulates the advertisement with 1308 source set to its own ADM-ULA and destination set to the ADM-ULA of 1309 the LHS Proxy/Server. The LHS Client then includes an appropriate 1310 Identification and *NET headers and forwards the carrier packets to 1311 the LHS Proxy/Server. When the LHS Proxy/Server receives the carrier 1312 packets, it verifies the Identifications, reassembles/decapsulates to 1313 obtain the advertisement, verifies the authentication signature, then 1314 uses the current MVFI List B entry to locate the MFV. The LHS Proxy/ 1315 Server then writes the current MFVI List A entry as the "A2" value 1316 for the MVF, generates an "A1" MFVI and both assigns it to the MFV 1317 and writes it as the next MFVI List A entry. The LHS Proxy/Server 1318 then examines the FHS/LHS FMT codes to determine if it needs to 1319 include Tunnel window synchronization parameters. The LHS Proxy/ 1320 Server then re-encapsulates/re-fragments the advertisement, sets the 1321 OAL source to its own ADM-ULA and destination to the ADM-ULA of the 1322 LHS Bridge, includes an appropriate Identification and *NET headers 1323 and forwards the carrier packets into the secured spanning tree. 1325 When the LHS Bridge receives the carrier packets, it reassembles/ 1326 decapsulates to obtain the advertisement then uses the current MFVI 1327 List B entry to locate the MFV. The LHS Bridge then writes the 1328 current MFVI List A entry as the MFV "A2" index and generates a new 1329 "A1" value which it both assigns the MFV and writes as the next MFVI 1330 List A entry. (The LHS Bridge also caches the first A entry in the 1331 MFVI List as "LHS-Client" when LHS FMT-Forward/Mode are both set to 1332 enable future direct forwarding to this LHS Client.) If the LHS 1333 Bridge is connected directly to both the FHS and LHS segments 1334 (whether the segments are the same or different), the FHS/LHS Bridge 1335 will have already cached the FHS/LHS information based on the 1336 original solicitation. The FHS/LHS Bridge then re-encapsulates the 1337 solicitation with its own ADM-ULA as the source and with the ADM-ULA 1338 of the FHS Proxy/Server as the destination. If the FHS and LHS 1339 prefixes are different, the FHS Bridge instead re-encapsulates/re- 1340 fragments with its own ADM-ULA as the source and with the ADM-ULA of 1341 the FHS Bridge as the destination. The LHS Bridge selects an 1342 appropriate Identification and *NET headers then forwards the carrier 1343 packets into the secured spanning tree. 1345 When the FHS and LHS Bridges are different, the FHS Bridge will 1346 receive the carrier packets from the LHS Bridge over the secured 1347 spanning tree. The FHS Bridge reassembles/decapsulates to obtain the 1348 advertisement, then locates the MFV based on the current MFVI List B 1349 entry. The FHS Bridge then assigns the current MFVI List A entry as 1350 the MFV "A2" index and caches the ADM-ULA of the LHS Bridge as the 1351 spanning tree address for "A2". The FHS Bridge then generates an 1352 "A1" MVFI and both assigns it to the MVF and writes it as the next 1353 MFVI List A entry while also writing its ADM-ULA and INADDR in the 1354 advertisement FHS Bridge fields. The FHS Bridge then re- 1355 encapsulates/re-fragments with its own ADM-ULA as the source, with 1356 the ADM-ULA of the FHS Proxy/Server as the destination, then selects 1357 an appropriate Identification value and *NET headers and forwards the 1358 carrier packets into the secured spanning tree. 1360 When the FHS Proxy/Server receives the carrier packets from the 1361 secured spanning tree, it reassembles/decapsulates to obtain the 1362 advertisement then locates the MFV based on the current MFVI List B 1363 entry. The FHS Proxy/Server then assigns the current MFVI List A 1364 entry as the "A2" MFVI the same as for the prior hop. If the 1365 advertisement destination is its own ADM-LLA, the FHS Proxy/Server 1366 then caches the advertisement Multilink Forwarding Parameters with 1367 the MFV and examines LHS FMT. If FMT-Forward is clear, the FHS 1368 Proxy/Server locates the NCE for the ADM-LLA of the LHS Proxy/Server 1369 and sets the state to REACHABLE then caches any Tunnel Window 1370 Synchronization parameters. If the advertisement source is the MNP- 1371 LLA of the LHS Client, the FHS Proxy/Server then locates the LHS 1372 Client NCE and sets the state to REACHABLE then caches the OMNI 1373 header window synchronization parameters and prepares to return an NA 1374 acknowledgement, if necessary. 1376 If the advertisement destination is the MNP-LLA of the FHS Client, 1377 the FHS Proxy/Server also searches for and updates the NCE for the 1378 ADM-LLA of the LHS Proxy/Server if necessary the same as above. The 1379 FHS Proxy/Server then generates an "A1" MFVI and assigns it both to 1380 the MFVI and as the next MFVI List A entry, then includes an 1381 authentication signature in the advertisement message. The FHS 1382 Proxy/Server then re-encapsulates/re-fragments with its own ADM-ULA 1383 as the source, with the MNP-ULA of the FHS Client as the destination, 1384 then selects an appropriate Identification value and *NET headers and 1385 forwards the carrier packets to the FHS Client. 1387 When the FHS Client receives the carrier packets, it verifies the 1388 Identification, reassembles/decapsulates to obtain the advertisement 1389 then locates the MFV based on the current MFVI List B entry. The FHS 1390 Client then assigns the current MFVI List A entry as the "A2" MFVI 1391 the same as for the prior hop. The FHS Client then caches the 1392 advertisement Multilink Forwarding Parameters (including the MFVI 1393 List) with the MFV and examines LHS FMT. If FMT-Forward is clear, 1394 the FHS Client locates the NCE for the ADM-LLA of the LHS Proxy/ 1395 Server and sets the state to REACHABLE then caches any Tunnel Window 1396 Synchronization parameters. If the advertisement source is the MNP- 1397 LLA of the LHS Client, the FHS Proxy/Server then locates the LHS 1398 Client NCE and sets the state to REACHABLE then caches the OMNI 1399 header window synchronization parameters and prepares to return an NA 1400 acknowledgement, if necessary. The FHS Client also caches the 1401 previous MFVI List A entry as "FHS-Bridge" since it can include this 1402 value when it sends future carrier packets directly to the Bridge 1403 (following appropriate neighbor coordination). 1405 When either the FHS Client or FHS Proxy/Server needs to return an NA 1406 acknowledgement to complete window synchronization, it prepares an 1407 acknowledgement message with an OMNI Multilink Forwarding Parameters 1408 sub-option with Job code set to "10" (Follow A; Record B). The FHS 1409 node then includes Tunnel Window Synchronization parameters if 1410 necessary and sets the MFVI List to the cached list of A entries 1411 received in the LHS advertisement, but need not set any other FHS/LHS 1412 information. If the FHS node is the Client, it next includes an 1413 authentication signature then encapsulates/fragments with its own 1414 MNP-ULA as the source and the ADM-ULA of the FHS Proxy/Server as the 1415 destination, then selects an appropriate Identification value and 1416 *NET headers and forwards the carrier packets to the FHS Proxy/ 1417 Server. The FHS Proxy/Server then verifies the Identification, 1418 reassembles/decapsulates, verifies the authentication signature and 1419 uses the current MFVI List A entry to locate the MFV. The FHS Proxy/ 1420 Server then writes its "B1" MFVI as the next MFVI List B entry and 1421 determines whether it needs to include Tunnel Window Synchronization 1422 parameters the same as it had done when it forwarded the original 1423 solicitation. 1425 The FHS Proxy/Server then re-encapsulates/re-fragments with its own 1426 ADM-ULA as the source and the ADM-ULA of the FHS Bridge as the 1427 destination, then selects an appropriate Identification and *NET 1428 headers and forwards the carrier packets into the secured spanning 1429 tree. When the FHS Bridge receives the carrier packets, it 1430 reassembles/decapsulates then uses the current MFVI List A entry to 1431 locate the MFV. The FHS Bridge then writes its "B1" MFVI as the next 1432 MFVI List B entry. The FHS Bridge then re-encapsulates/re-fragments 1433 with its own ADM-ULA as the source and the ADM-ULA of the LHS Proxy/ 1434 Server as the destination. If the FHS Bridge is also the LHS Bridge, 1435 it sets the ADM-ULA of the LHS Proxy/Server as the destination; 1436 otherwise it sets the ADM-ULA of the LHS Bridge. The FHS Bridge then 1437 selects an appropriate Identification and *NET headers and forwards 1438 the carrier packets into the secured spanning tree. If an LHS Bridge 1439 receives the carrier packets, it processes them exactly the same as 1440 the FHS Bridge had done while setting the carrier packet source to 1441 its own ADM-ULA and destination to the ADM-ULA of the LHS Proxy/ 1442 Server. 1444 When the LHS Proxy/Server receives the carrier packets, it 1445 reassembles/decapsulates to obtain the NA acknowledgement message. 1446 The LHS Proxy/Server then locates the MFV based on the current MFVI 1447 List A entry then determines whether it is a tunnel ingress the same 1448 as for the original solicitation. If it is a tunnel ingress, the LHS 1449 Proxy/Server updates the NCE for the tunnel far-end based on the 1450 Tunnel Window Synchronization parameters in the NA. If the NA 1451 destination is its own ADM-LLA, the LHS Proxy/Server next updates the 1452 NCE for the NA source LLA based on the OMNI header Window 1453 Synchronization parameters and MAY compare the MVFI List to the 1454 version it had cached in the MFV based on the original solicitation. 1456 If the NA destination is the MNP-LLA of the LHS Client, the LHS 1457 Proxy/Server instead writes its "B1" MFV as the next MFVI List B 1458 entry, includes an authentication signature, re-encapsulates/re- 1459 fragments with its own ADM-ULA as the source and the MNP-ULA of the 1460 Client as the destination then selects an appropriate Identification 1461 and *NET headers and forwards the resulting carrier packets to the 1462 LHS Client. When the LHS Client receives the carrier packets, it 1463 verifies the Identification, reassembles/decapsulates to obtain the 1464 NA acknowledgement, verifies the authentication signature then 1465 processes the message exactly the same as for the LHS Proxy/Server 1466 case above. 1468 Following the solicitation/advertisement/acknowledgement exchange, 1469 OAL end systems and tunnel endpoints can begin exchanging ordinary 1470 carrier packets with Identification values within their respective 1471 send/receive windows without requiring security signatures and/or 1472 secured spanning tree traversal. Either peer can refresh window 1473 synchronization parameters and/or send other carrier packets 1474 requiring security at any time using the same secured procedures 1475 described above. OAL end systems and intermediate nodes can also use 1476 their own A1/B1 MFVIs when they receive carrier packets to 1477 unambiguously locate the correct MFV and determine directionality and 1478 can use any discovered A2/B2 MFVIs to forward carrier packets to 1479 other OAL nodes that configure the corresponding A1/B1 MFVIs. When 1480 an OAL node uses an MFVI included in a carrier packet to locate an 1481 MFV, it need not also examine the carrier packet addresses. 1483 OAL sources can also begin including CRH-32s in carrier packets with 1484 a list of A/B MFVIs that OAL intermediate nodes can use for shortest- 1485 path carrier packet forwarding based on MFVIs instead of spanning 1486 tree addresses. OAL sources and intermediate nodes can also begin 1487 forwarding carrier packets with compressed headers (see: 1488 [I-D.templin-6man-omni]) that include only a single A/B MFVI 1489 meaningful to the next hop, since all nodes in the path up to (and 1490 sometimes including) the OAL destination have already established MFV 1491 forwarding information. Note that when an FHS OAL source receives a 1492 solicited advertisement with Job code "01', the message will contain 1493 an MFVI List with A entries populated in the reverse order needed for 1494 populating a CRH-32 routing header. The FHS OAL source must 1495 therefore write the MFVI List A entries last-to-first when it 1496 populates a CRH-32, or must select the correct A entry to include in 1497 a compressed OAL header based on the intended OAL intermediate node 1498 or destination. 1500 When a Bridge receives unsecured carrier packets destined to a local 1501 segment Client that has asserted direct reachability, the Bridge 1502 employs NAT traversal procedures to enable direct carrier packet 1503 forwarding while bypassing the local Proxy/Server based on the 1504 Client's advertised MFVIs and discovered NATed L2ADDR information. 1505 If the Client cannot be reached directly (or if NAT traversal has not 1506 yet converged), the Bridge instead forwards carrier packets directly 1507 to the local Proxy/Server. 1509 When a Proxy/Server receives carrier packets destined to a local 1510 Client or forwards carrier packets received from a local Client, it 1511 first locates the correct MFV. If the carrier packets include a 1512 secured IPv6 ND message, the Proxy/Server uses the Client's MVF 1513 established through RS/RA exchanges to re-encapsulate/re-fragment 1514 while forwarding outbound secured carrier packets via the secured 1515 spanning tree and forwarding inbound secured carrier packets while 1516 including an authentication signature. For ordinary carrier packets, 1517 the Proxy/Server uses the same MFV if directed by MFVI and/or OAL 1518 addressing. Otherwise it locates an MFV established through an NS/NA 1519 exchange between the Client and the remote peer, and forwards the 1520 carrier packets without first reassembling/decapsulating. 1522 When a Proxy/Server or Client configured as a tunnel ingress receives 1523 a carrier packet with a full header with an MNP-ULA source or a 1524 compressed header with an MFVI that matches an MFV, the ingress 1525 encapsulates the carrier packet in a new OAL full or compressed 1526 header with the inner header containing Identification values 1527 appropriate for the end-to-end window and the outer header containing 1528 an Identification value appropriate for the tunnel endpoints. When a 1529 Proxy/Server or Client configured as a tunnel egress receives an 1530 encapsulated carrier packet, it verifies the Identification in the 1531 outer header, then discards the outer header and forwards the inner 1532 carrier packet to the final destination. 1534 When a source Client forwards carrier packets it can employ header 1535 compression according to the MFVIs established through an NS/NA 1536 exchange with a remote or local peer. When the source Client 1537 forwards to a remote peer, it can forward carrier packets to a local 1538 SRT Bridge (following the establishment of L2ADDR information) while 1539 bypassing the Proxy/Server. When a target Client receives carrier 1540 packets that match a local MFV, the Client first verifies the 1541 Identification then decompresses the headers if necessary, 1542 reassembles if necessary to obtain the OAL packet then decapsulates 1543 and delivers the IP packet to upper layers. 1545 When synchronized peer Clients in the same SRT segment with FMT- 1546 Forward and FMT-Mode set discover each other's NATed L2ADDR addresses 1547 through NAT traversal, they can exchange carrier packets directly 1548 with header compression using MFVIs discovered as above. The FHS 1549 Client will have cached the A MFVI for the LHS Client, which will 1550 have cached the B MVFI for the FHS Client. 1552 3.3. OMNI Interface Characteristics 1554 OMNI interfaces are virtual interfaces configured over one or more 1555 underlying interfaces classified as follows: 1557 o INET interfaces connect to an INET either natively or through one 1558 or more NATs. Native INET interfaces have global IP addresses 1559 that are reachable from any INET correspondent. The INET-facing 1560 interfaces of Proxy/Servers are native interfaces, as are Relay 1561 and Bridge interfaces. NATed INET interfaces connect to a private 1562 network behind one or more NATs that provide INET access. Clients 1563 that are behind a NAT are required to send periodic keepalive 1564 messages to keep NAT state alive when there are no carrier packets 1565 flowing. 1567 o ANET interfaces connect to an ANET that is separated from the open 1568 INET by an FHS Proxy/Server. Clients can issue control messages 1569 over the ANET without including an authentication signature since 1570 the ANET is secured at the network layer or below. Proxy/Servers 1571 can actively issue control messages over the INET on behalf of 1572 ANET Clients to reduce ANET congestion. 1574 o VPNed interfaces use security encapsulation over the INET to a 1575 Virtual Private Network (VPN) server that also acts as an FHS 1576 Proxy/Server. Other than the link-layer encapsulation format, 1577 VPNed interfaces behave the same as Direct interfaces. 1579 o Direct (i.e., single-hop point-to-point) interfaces connect a 1580 Client directly to an FHS Proxy/Server without crossing any ANET/ 1581 INET paths. An example is a line-of-sight link between a remote 1582 pilot and an unmanned aircraft. The same Client considerations 1583 apply as for VPNed interfaces. 1585 OMNI interfaces use OAL encapsulation and fragmentation as discussed 1586 in Section 3.2.4. OMNI interfaces use *NET encapsulation (see: 1587 Section 3.6) to exchange carrier packets with OMNI link neighbors 1588 over INET or VPNed interfaces as well as over ANET interfaces for 1589 which the Client and FHS Proxy/Server may be multiple IP hops away. 1590 OMNI interfaces do not use link-layer encapsulation over Direct 1591 underlying interfaces or ANET interfaces when the Client and FHS 1592 Proxy/Server are known to be on the same underlying link. 1594 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1595 state the same as for any interface. OMNI interfaces use IPv6 ND 1596 messages including Router Solicitation (RS), Router Advertisement 1597 (RA), Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1598 neighbor cache management. In environments where spoofing may be a 1599 threat, OMNI neighbors should employ OAL Identification window 1600 synchronization in their IPv6 ND message exchanges. 1602 OMNI interfaces send IPv6 ND messages with an OMNI option formatted 1603 as specified in [I-D.templin-6man-omni]. The OMNI option includes 1604 prefix registration information, Multilink Forwarding Parameters 1605 containing link information parameters for the OMNI interface's 1606 underlying interfaces and any other per-neighbor information. 1608 A Client's OMNI interface may be configured over multiple underlying 1609 interfaces. For example, common mobile handheld devices have both 1610 wireless local area network ("WLAN") and cellular wireless links. 1611 These links are often used "one at a time" with low-cost WLAN 1612 preferred and highly-available cellular wireless as a standby, but a 1613 simultaneous-use capability could provide benefits. In a more 1614 complex example, aircraft frequently have many wireless data link 1615 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1616 directional, etc.) with diverse performance and cost properties. 1618 If a Client's multiple underlying interfaces are used "one at a time" 1619 (i.e., all other interfaces are in standby mode while one interface 1620 is active), then successive IPv6 ND messages all include OMNI option 1621 Multilink Forwarding Parameters sub-options with the same underlying 1622 interface index. In that case, the Client would appear to have a 1623 single underlying interface but with a dynamically changing link- 1624 layer address. 1626 If the Client has multiple active underlying interfaces, then from 1627 the perspective of IPv6 ND it would appear to have multiple link- 1628 layer addresses. In that case, IPv6 ND message OMNI options MAY 1629 include Multilink Forwarding Parameters sub-options with different 1630 underlying interface indexes. 1632 Bridge and Proxy/Server OMNI interfaces are configured over 1633 underlying interfaces that provide both secured tunnels for carrying 1634 IPv6 ND and BGP protocol control plane messages and open INET access 1635 for carrying unsecured messages. The OMNI interface configures both 1636 an ADM-LLA and its corresponding ADM-ULA, and acts as an OAL source 1637 to encapsulate and fragment original IP packets while presenting the 1638 resulting carrier packets over the secured or unsecured underlying 1639 paths. Note that Bridge and Proxy/Server BGP protocol TCP sessions 1640 are run directly over the OMNI interface and use ADM-ULA source and 1641 destination addresses. The OMNI interface employs the OAL to 1642 encapsulate the original IP packets for these sessions as carrier 1643 packets (i.e., even though the OAL header may use the same ADM-ULAs 1644 as the original IP header) and forwards them over the secured 1645 underlying path. 1647 3.4. OMNI Interface Initialization 1649 AERO Proxy/Servers and Clients configure OMNI interfaces as their 1650 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1651 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1652 to ensure that original IP packets with destination addresses covered 1653 by an MNP not explicitly associated with another interface are 1654 directed to an OMNI interface. 1656 OMNI interface initialization procedures for Proxy/Servers, Clients 1657 and Bridges are discussed in the following sections. 1659 3.4.1. AERO Proxy/Server and Relay Behavior 1661 When a Proxy/Server enables an OMNI interface, it assigns an 1662 ADM-{LLA,ULA} appropriate for the given OMNI link SRT segment. The 1663 Proxy/Server also configures secured tunnels with one or more 1664 neighboring Bridges and engages in BGP routing protocol sessions with 1665 one or more Bridges. 1667 The OMNI interface provides a single interface abstraction to the IP 1668 layer, but internally includes an NBMA nexus for sending carrier 1669 packets to OMNI interface neighbors over underlying INET interfaces 1670 and secured tunnels. The Proxy/Server further configures a service 1671 to facilitate IPv6 ND exchanges with AERO Clients and manages per- 1672 Client neighbor cache entries and IP forwarding table entries based 1673 on control message exchanges. 1675 Relays are simply Proxy/Servers that run a dynamic routing protocol 1676 to redistribute routes between the OMNI interface and INET/EUN 1677 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1678 networks on the INET/EUN interfaces (i.e., the same as a Client would 1679 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1680 interfaces. The Relay further provides an attachment point of the 1681 OMNI link to a non-MNP-based global topology. 1683 3.4.2. AERO Client Behavior 1685 When a Client enables an OMNI interface, it assigns either an 1686 MNP-{LLA, ULA} or a Temporary ULA and sends RS messages over its 1687 underlying interfaces to an FHS Proxy/Server, which returns an RA 1688 message with corresponding parameters. The RS/RA messages may pass 1689 through one or more NATs in the case of a Client's INET interface. 1690 (Note: if the Client used a Temporary ULA in its initial RS message, 1691 it will discover an MNP-{LLA, ULA} in the corresponding RA that it 1692 receives from the FHS Proxy/Server and begin using these new 1693 addresses. If the Client is operating outside the context of AERO 1694 infrastructure such as in a Mobile Ad-hoc Network (MANET), however, 1695 it may continue using Temporary ULAs for Client-to-Client 1696 communications until it encounters an infrastructure element that can 1697 provide an MNP.) 1699 3.4.3. AERO Bridge Behavior 1701 AERO Bridges configure an OMNI interface and assign an ADM-ULA and 1702 corresponding Subnet Router Anycast address for each OMNI link SRT 1703 segment they connect to. Bridges configure secured tunnels with 1704 Proxy/Servers in the same SRT segment and other Bridges in the same 1705 (or an adjacent) SRT segment. Bridges then engage in a BGP routing 1706 protocol session with neighbors over the secured spanning tree (see: 1707 Section 3.2.3). 1709 3.5. OMNI Interface Neighbor Cache Maintenance 1711 Each OMNI interface maintains a conceptual neighbor cache that 1712 includes a Neighbor Cache Entry (NCE) for each of its active 1713 neighbors on the OMNI link per [RFC4861]. Each NCE is indexed by the 1714 LLA of the neighbor, while the OAL encapsulation ULA determines the 1715 context for Identification verification. Clients and Proxy/Servers 1716 maintain NCEs through RS/RA exchanges, and also maintain NCEs for any 1717 active correspondent peers through NS/NA exchanges. 1719 Bridges also maintain NCEs for Clients within their local segments 1720 based on NS/NA(WIN) route optimization. When a Bridge creates/ 1721 updates a NCE for a local segment Client based on NS/NA(WIN) route 1722 optimization, it also maintains MVFI and L2ADDR state for messages 1723 destined to this local segment Client. 1725 Hub Proxy/Servers add an additional state DEPARTED to the list of NCE 1726 states found in Section 7.3.2 of [RFC4861]. When a Client terminates 1727 its association, the Proxy/Server OMNI interface sets a "DepartTime" 1728 variable for the NCE to "DEPART_TIME" seconds. DepartTime is 1729 decremented unless a new IPv6 ND message causes the state to return 1730 to REACHABLE. While a NCE is in the DEPARTED state, the Proxy/Server 1731 forwards carrier packets destined to the target Client to the 1732 Client's new location instead. When DepartTime decrements to 0, the 1733 NCE is deleted. It is RECOMMENDED that DEPART_TIME be set to the 1734 default constant value REACHABLE_TIME plus 10 seconds (40 seconds by 1735 default) to allow a window for carrier packets in flight to be 1736 delivered while stale route optimization state may be present. 1738 Hub Proxy/Servers act as RORs on behalf of their associated Clients 1739 according to the Proxy Neighbor Advertisement specification in 1740 Section 7.2.8 of [RFC4861]. When a Hub Proxy/Server ROR receives an 1741 authentic NS(AR) message, it first searches for a NCE for the target 1742 Client and accepts the message only if there is an entry. The Hub 1743 Proxy/Server then returns a solicited NA(AR) message while creating 1744 or updating a "Report List" entry in the target Client's NCE that 1745 caches both the LLA and ULA of ROS with a "ReportTime" variable set 1746 to REPORT_TIME seconds. The ROR resets ReportTime when it receives a 1747 new authentic NS(AR) message, and otherwise decrements ReportTime 1748 while no authentic NS(AR) messages have been received. It is 1749 RECOMMENDED that REPORT_TIME be set to the default constant value 1750 REACHABLE_TIME plus 10 seconds (40 seconds by default) to allow a 1751 window for route optimization to converge before ReportTime 1752 decrements below REACHABLE_TIME. 1754 When the ROS receives a solicited NA(AR) message response to its 1755 NS(AR), it creates or updates a NCE for the target network-layer and 1756 link-layer addresses. The ROS then (re)sets ReachableTime for the 1757 NCE to REACHABLE_TIME seconds and performs reachability tests over 1758 specific underlying interface pairs to determine paths for forwarding 1759 carrier packets directly to the target. The ROS otherwise decrements 1760 ReachableTime while no further solicited NA messages arrive. It is 1761 RECOMMENDED that REACHABLE_TIME be set to the default constant value 1762 30 seconds as specified in [RFC4861]. 1764 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1765 of NS messages sent when a correspondent may have gone unreachable, 1766 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1767 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1768 to limit the number of unsolicited NAs that can be sent based on a 1769 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1770 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1771 same as specified in [RFC4861]. 1773 Different values for the above constants MAY be administratively set; 1774 however, if different values are chosen, all nodes on the link MUST 1775 consistently configure the same values. Most importantly, 1776 DEPART_TIME and REPORT_TIME SHOULD be set to a value that is 1777 sufficiently longer than REACHABLE_TIME to avoid packet loss due to 1778 stale route optimization state. 1780 3.5.1. OMNI ND Messages 1782 OMNI interfaces prepare IPv6 ND messages the same as for standard 1783 IPv6 ND, but also include a new option type termed the OMNI option 1784 [I-D.templin-6man-omni]. OMNI interfaces prepare IPv6 ND messages 1785 the same as for standard IPv6 ND, and include one or more OMNI 1786 options and any other options then completely populate all option 1787 information. If the OMNI interface includes an authentication 1788 signature, it sets the IPv6 ND message Checksum field to 0 and 1789 calculates the authentication signature over the entire length of the 1790 message (beginning with a pseudo-header of the IPv6 header) but does 1791 not then proceed to calculate the IPv6 ND message checksum itself. 1792 If the OMNI interface forwards the message to a next hop over the 1793 secured spanning tree path, it omits both the authentication 1794 signature and checksum since lower layers already ensure 1795 authentication and integrity. In all other cases, the OMNI interface 1796 calculates the standard IPv6 ND message checksum and writes the value 1797 in the Checksum field. OMNI interfaces verify authentication and 1798 integrity of each IPv6 ND message received according to the specific 1799 check(s) included, and process the message further only following 1800 verification. 1802 OMNI options include per-neighbor information that provides multilink 1803 forwarding, link-layer address and traffic selector information for 1804 the neighbor's underlying interfaces. This information is stored in 1805 the neighbor cache and provides the basis for the forwarding 1806 algorithm specified in Section 3.10. The information is cumulative 1807 and reflects the union of the OMNI information from the most recent 1808 IPv6 ND messages received from the neighbor; it is therefore not 1809 required that each IPv6 ND message contain all neighbor information. 1811 The OMNI option is distinct from any Source/Target Link-Layer Address 1812 Options (S/TLLAOs) that may appear in an IPv6 ND message according to 1813 the appropriate IPv6 over specific link layer specification (e.g., 1814 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1815 pertains to encapsulation addresses while the latter pertains to the 1816 native L2 address format of the underlying media 1818 OMNI interface IPv6 ND messages may also include other IPv6 ND 1819 options. In particular, solicitation messages may include Nonce and/ 1820 or Timestamp options if required for verification of advertisement 1821 replies. If an OMNI IPv6 ND solicitation message includes a Nonce 1822 option, the advertisement reply must echo the same Nonce. If an OMNI 1823 IPv6 ND solicitation message includes a Timestamp option, the 1824 advertisement reply should also include a Timestamp option. 1826 AERO Clients send RS messages to the All-Routers multicast address or 1827 an ADM-LLA while using unicast link-layer addresses. AERO Proxy/ 1828 Servers respond by returning unicast RA messages. During the RS/RA 1829 exchange, AERO Clients and Proxy/Servers include state 1830 synchronization parameters to establish Identification windows and 1831 other state. 1833 AERO nodes use NS/NA messages for the following purposes: 1835 o NS/NA(AR) messages are used for address resolution only. The ROS 1836 sends an NS(AR) to the solicited-node multicast address of the 1837 target, and the current ROR with addressing information for the 1838 target returns a unicast NA(AR). The NA(AR) contains current, 1839 consistent and authentic target address resolution information, 1840 but only an implicit third-party assertion of target reachability. 1841 NS/NA(AR) messages must be secured. 1843 o NS/NA(WIN) messages are used for establishing and maintaining 1844 window synchronization state (and/or any other state such as 1845 Interface Attributes). The source sends an NS(WIN) to the unicast 1846 address of the target, and the target returns a unicast NA(WIN). 1847 The NS/NA(WIN) exchange synchronizes the sequence number windows 1848 for Identification values the neighbors will include in subsequent 1849 carrier packets, and asserts reachability for the target without 1850 necessarily testing a specific underlying interface pair. NS/ 1851 NA(WIN) messages must be secured. 1853 o NS/NA(NUD) messages are used for determining target reachability. 1854 The source sends an NS(NUD) to the unicast address of the target 1855 while naming a specific underlying interface pair, and the target 1856 returns a unicast NA(NUD). NS/NA(NUD) messages that use an in- 1857 window sequence number and do not update any other state need not 1858 be secured but should include an IPv6 ND message checksum. NS/ 1859 NA(NUD) messages may also be used in combination with window 1860 synchronization (i.e., NUD+WIN), in which case the messages must 1861 be secured. 1863 o Unsolicited NA (uNA) messages are used to signal addressing and/or 1864 other neighbor state changes (e.g., address changes due to 1865 mobility, signal degradation, traffic selector updates, etc.). uNA 1866 messages that include state update information must be secured. 1868 o NS/NA(DAD) messages are not used in AERO, since Duplicate Address 1869 Detection is not required. 1871 Additionally, nodes may send NA/RA messages with the OMNI option PNG 1872 flag set to receive a solicited NA response from the neighbor. The 1873 solicited NA response MUST set the ACK flag (without also setting the 1874 SYN or PNG flags) and include the Identification used in the PNG 1875 message in the Acknowledgement. 1877 3.5.2. OMNI Neighbor Advertisement Message Flags 1879 As discussed in Section 4.4 of [RFC4861] NA messages include three 1880 flag bits R, S and O. OMNI interface NA messages treat the flags as 1881 follows: 1883 o R: The R ("Router") flag is set to 1 in the NA messages sent by 1884 all AERO/OMNI node types. Simple hosts that would set R to 0 do 1885 not occur on the OMNI link itself, but may occur on the downstream 1886 links of Clients and Relays. 1888 o S: The S ("Solicited") flag is set exactly as specified in 1889 Section 4.4. of [RFC4861], i.e., it is set to 1 for Solicited NAs 1890 and set to 0 for uNAs (both unicast and multicast). 1892 o O: The O ("Override") flag is set to 0 for solicited NAs returned 1893 by a Proxy/Server ROR and set to 1 for all other solicited and 1894 unsolicited NAs. For further study is whether solicited NAs for 1895 anycast targets apply for OMNI links. Since MNP-LLAs must be 1896 uniquely assigned to Clients to support correct IPv6 ND protocol 1897 operation, however, no role is currently seen for assigning the 1898 same MNP-LLA to multiple Clients. 1900 3.5.3. OMNI Neighbor Window Synchronization 1902 In secured environments (e.g., between secured spanning tree 1903 neighbors, between neighbors on the same secured ANET, etc.), OMNI 1904 interface neighbors can exchange OAL packets using randomly- 1905 initialized and monotonically-increasing Identification values 1906 (modulo 2*32) without window synchronization. In environments where 1907 spoofing is considered a threat, OMNI interface neighbors instead 1908 invoke window synchronization in NS/NA(WIN) message exchanges to 1909 maintain send/receive window state in their respective neighbor cache 1910 entries as specified in [I-D.templin-6man-omni]. 1912 3.6. OMNI Interface Encapsulation and Re-encapsulation 1914 The OMNI interface admits original IP packets then acts as an OAL 1915 source to perform OAL encapsulation and fragmentation as specified in 1916 [I-D.templin-6man-omni] while including a CRH-32 if necessary as 1917 specified in Section 3.2.4. The OAL encapsulates original IP packets 1918 to form OAL packets subject to fragmentation, then encapsulates the 1919 resulting OAL fragments in *NET headers as carrier packets. 1921 For carrier packets undergoing re-encapsulation at an OAL 1922 intermediate node, the OMNI interface decrements the OAL IPv6 header 1923 Hop Limit and discards the carrier packet if the Hop Limit reaches 0. 1924 The intermediate node next removes the *NET encapsulation headers 1925 from the first segment and re-encapsulates the packet in new *NET 1926 encapsulation headers for the next segment. 1928 When an FHS Bridge receives a carrier packet with a compressed header 1929 that must be forwarded to an LHS Bridge over the unsecured spanning 1930 tree, it reconstructs the headers based on MFV state, inserts a 1931 CRH-32 immediately following the OAL header and adjusts the OAL 1932 payload length and destination address field. The FHS Bridge 1933 includes a single MFVI in the CRH-32 that will be meaningful to the 1934 LHS Bridge. When the LHS Bridge receives the carrier packet, it 1935 locates the MFV for the next hop based on the CRH-32 MFVI then re- 1936 applies header compression (resulting in the removal of the CRH-32) 1937 and forwards the carrier packet to the next hop. 1939 3.7. OMNI Interface Decapsulation 1941 OMNI interfaces (acting as OAL destinations) decapsulate and 1942 reassemble OAL packets into original IP packets destined either to 1943 the AERO node itself or to a destination reached via an interface 1944 other than the OMNI interface the original IP packet was received on. 1945 When carrier packets containing OAL fragments addressed to itself 1946 arrive, the OMNI interface discards the NET encapsulation headers and 1947 reassembles as discussed in Section 3.9. 1949 3.8. OMNI Interface Data Origin Authentication 1951 AERO nodes employ simple data origin authentication procedures. In 1952 particular: 1954 o AERO Bridges and Proxy/Servers accept carrier packets received 1955 from the secured spanning tree. 1957 o AERO Proxy/Servers and Clients accept carrier packets and original 1958 IP packets that originate from within the same secured ANET. 1960 o AERO Clients and Relays accept original IP packets from downstream 1961 network correspondents based on ingress filtering. 1963 o AERO Clients, Relays, Proxy/Servers and Bridges verify carrier 1964 packet UDP/IP encapsulation addresses according to 1965 [I-D.templin-6man-omni]. 1967 o AERO nodes accept carrier packets addressed to themselves with 1968 Identification values within the current window for the OAL source 1969 neighbor (when window synchronization is used) and drop any 1970 carrier packets with out-of-window Identification values. (AERO 1971 nodes may forward carrier packets not addressed to themselves 1972 without verifying the Identification value.) 1974 AERO nodes silently drop any packets that do not satisfy the above 1975 data origin authentication procedures. Further security 1976 considerations are discussed in Section 6. 1978 3.9. OMNI Interface MTU 1980 The OMNI interface observes the link nature of tunnels, including the 1981 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1982 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 1983 The OMNI interface employs an OMNI Adaptation Layer (OAL) that 1984 accommodates multiple underlying links with diverse MTUs while 1985 observing both a minimum and per-path Maximum Payload Size (MPS). 1986 The functions of the OAL and the OMNI interface MTU/MRU/MPS are 1987 specified in [I-D.templin-6man-omni] with MTU/MRU both set to the 1988 constant value 9180 bytes, with minimum MPS set to 400 bytes, and 1989 with potentially larger per-path MPS values depending on the 1990 underlying path. 1992 When the network layer presents an original IP packet to the OMNI 1993 interface, the OAL source encapsulates and fragments the original IP 1994 packet if necessary. When the network layer presents the OMNI 1995 interface with multiple original IP packets bound to the same OAL 1996 destination, the OAL source can concatenate them together into a 1997 single OAL super-packet as discussed in [I-D.templin-6man-omni]. The 1998 OAL source then fragments the OAL packet if necessary according to 1999 the minimum/path MPS such that the OAL headers appear in each 2000 fragment while the original IP packet header appears only in the 2001 first fragment. The OAL source then encapsulates each OAL fragment 2002 in *NET headers for transmission as carrier packets over an 2003 underlying interface connected to either a physical link (such as 2004 Ethernet, WiFi and the like) or a virtual link such as an Internet or 2005 higher-layer tunnel (see the definition of link in [RFC8200]). 2007 Note: Although a CRH-32 may be inserted or removed by a Bridge in the 2008 path (see: Section 3.10.3), this does not interfere with the 2009 destination's ability to reassemble since the CRH-32 is not included 2010 in the fragmentable part and its removal/transformation does not 2011 invalidate fragment header information. 2013 3.10. OMNI Interface Forwarding Algorithm 2015 Original IP packets enter a node's OMNI interface either from the 2016 network layer (i.e., from a local application or the IP forwarding 2017 system) while carrier packets enter from the link layer (i.e., from 2018 an OMNI interface neighbor). All original IP packets and carrier 2019 packets entering a node's OMNI interface first undergo data origin 2020 authentication as discussed in Section 3.8. Those that satisfy data 2021 origin authentication are processed further, while all others are 2022 dropped silently. 2024 Original IP packets that enter the OMNI interface from the network 2025 layer are forwarded to an OMNI interface neighbor using OAL 2026 encapsulation and fragmentation to produce carrier packets for 2027 transmission over underlying interfaces. (If routing indicates that 2028 the original IP packet should instead be forwarded back to the 2029 network layer, the packet is dropped to avoid looping). Carrier 2030 packets that enter the OMNI interface from the link layer are either 2031 re-encapsulated and re-admitted into the OMNI link, or reassembled 2032 and forwarded to the network layer where they are subject to either 2033 local delivery or IP forwarding. In all cases, the OAL MUST NOT 2034 decrement the original IP packet TTL/Hop-count since its forwarding 2035 actions occur below the network layer. 2037 OMNI interfaces may have multiple underlying interfaces and/or 2038 neighbor cache entries for neighbors with multiple underlying 2039 interfaces (see Section 3.3). The OAL uses Interface Attributes and/ 2040 or Traffic Selectors (e.g., port number, flow specification, etc.) to 2041 select an outbound underlying interface for each OAL packet and also 2042 to select segment routing and/or link-layer destination addresses 2043 based on the neighbor's underlying interfaces. AERO implementations 2044 SHOULD permit network management to dynamically adjust Traffic 2045 Selector values at runtime. 2047 If an OAL packet matches the Traffic Selectors of multiple outgoing 2048 interfaces and/or neighbor interfaces, the OMNI interface replicates 2049 the packet and sends one copy via each of the (outgoing / neighbor) 2050 interface pairs; otherwise, it sends a single copy of the OAL packet 2051 via an interface with the best matching Traffic Selector. (While not 2052 strictly required, the likelihood of successful reassembly may 2053 improve when the OMNI interface sends all fragments of the same 2054 fragmented OAL packet consecutively over the same underlying 2055 interface pair to avoid complicating factors such as delay variance 2056 and reordering.) AERO nodes keep track of which underlying 2057 interfaces are currently "reachable" or "unreachable", and only use 2058 "reachable" interfaces for forwarding purposes. 2060 The following sections discuss the OMNI interface forwarding 2061 algorithms for Clients, Proxy/Servers and Bridges. In the following 2062 discussion, an original IP packet's destination address is said to 2063 "match" if it is the same as a cached address, or if it is covered by 2064 a cached prefix (which may be encoded in an MNP-LLA). 2066 3.10.1. Client Forwarding Algorithm 2068 When an original IP packet enters a Client's OMNI interface from the 2069 network layer the Client searches for a NCE that matches the 2070 destination. If there is a match, the Client selects one or more 2071 "reachable" neighbor interfaces in the entry for forwarding purposes. 2072 If there is no NCE, the Client instead either enqueues the original 2073 IP packet and invokes route optimization while forwarding the 2074 original IP packet toward an FHS Proxy/Server. The Client (acting as 2075 an OAL source) performs OAL encapsulation and sets the OAL 2076 destination address to the MNP-ULA of the target if there is a 2077 matching NCE; otherwise, it sets the OAL destination to the ADM-ULA 2078 of the FHS Proxy/Server. If the Client has multiple original IP 2079 packets to send to the same neighbor, it can concatenate them in a 2080 single super-packet [I-D.templin-6man-omni]. The OAL source then 2081 performs fragmentation to create OAL fragments (see: Section 3.9), 2082 appends any *NET encapsulation, and sends the resulting carrier 2083 packets over underlying interfaces to the neighbor acting as an OAL 2084 destination. 2086 If the neighbor interface selected for forwarding is located on the 2087 same OMNI link segment and not behind a NAT, the Client forwards the 2088 carrier packets directly according to the L2ADDR information for the 2089 neighbor. If the neighbor interface is behind a NAT on the same OMNI 2090 link segment, the Client instead forwards the initial carrier packets 2091 to the LHS Proxy/Server (while inserting an CRH-32 if necessary) and 2092 initiates NAT traversal procedures. If the Client's intended source 2093 underlying interface is also behind a NAT and located on the same 2094 OMNI link segment, it sends a "direct bubble" over the interface per 2095 [RFC6081][RFC4380] to the L2ADDR found in the neighbor cache in order 2096 to establish state in its own NAT by generating traffic toward the 2097 neighbor (note that no response to the bubble is expected). 2099 The Client next sends an NS(NUD) message toward the MNP-ULA of the 2100 neighbor via the LHS Proxy/Server as discussed in Section 3.15. If 2101 the Client receives an NA(NUD) from the neighbor over the underlying 2102 interface, it marks the neighbor interface as "trusted" and sends 2103 future carrier packets directly to the L2ADDR information for the 2104 neighbor instead of indirectly via the LHS Proxy/Server. The Client 2105 must honor the neighbor cache maintenance procedure by sending 2106 additional direct bubbles and/or NS/NA(NUD) messages as discussed in 2107 [RFC6081][RFC4380] in order to keep NAT state alive as long as 2108 carrier packets are still flowing. 2110 When a carrier packet enters a Client's OMNI interface from the link- 2111 layer, if the OAL destination matches one of the Client's ULAs the 2112 Client (acting as an OAL destination) verifies that the 2113 Identification is in-window for this OAL source, then reassembles and 2114 decapsulates as necessary and delivers the original IP packet to the 2115 network layer. If the OAL destination does not match, the Client 2116 drops the original IP packet and MAY return a network-layer ICMP 2117 Destination Unreachable message subject to rate limiting (see: 2118 Section 3.11). 2120 Note: When a Client performs an NS/NA(WIN) exchange with a peer over 2121 the spanning tree, MFIB information will be established in all FHS 2122 and LHS intermediate nodes in the path. The Client can then 2123 opportunistically "skip ahead" in the chain of hops to bypass 2124 intermediate nodes that are not required to forward packets. The 2125 Client will have the map of MFVIs held at each hop, and can include 2126 the MFVI for the first hop it visits in a compressed OMNI header or 2127 in a CRH-32 routing header. 2129 Note: Clients and their FHS Proxy/Server (and other Client) peers can 2130 exchange original IP packets over ANET underlying interfaces without 2131 invoking the OAL, since the ANET is secured at the link and physical 2132 layers. By forwarding original IP packets without invoking the OAL, 2133 however, the ANET peers can engage only in classical path MTU 2134 discovery since the packets are subject to loss and/or corruption due 2135 to the various per-link MTU limitations that may occur within the 2136 ANET. Moreover, the original IP packets do not include either the 2137 OAL integrity check or per-packet Identification values that can be 2138 used for data origin authentication and link-layer retransmissions. 2139 The tradeoff therefore involves an assessment of the per-packet 2140 encapsulation overhead saved by bypassing the OAL vs. inheritance of 2141 classical network "brittleness". (Note however that ANET peers can 2142 send small original IP packets without invoking the OAL, while 2143 invoking the OAL for larger packets. This presents the beneficial 2144 aspects of both small packet efficiency and large packet robustness, 2145 with delay variance and reordering as possible side effects.) 2147 3.10.2. Proxy/Server and Relay Forwarding Algorithm 2149 When the Proxy/Server receives an original IP packet from the network 2150 layer, it drops the packet if routing indicates that it should be 2151 forwarded back to the network layer to avoid looping. Otherwise, the 2152 Proxy/Server regards the original IP packet the same as if it had 2153 arrived as carrier packets with OAL destination set to its own ADM- 2154 ULA. When the Proxy/Server receives carrier packets on underlying 2155 interfaces with OAL destination set to its own ADM-ULA, it performs 2156 OAL reassembly if necessary to obtain the original IP packet. 2158 The Proxy/Server next searches for a NCE that matches the original IP 2159 destination and proceeds as follows: 2161 o if the packet is an NA(WIN) message for a local Client NCE, the 2162 Proxy/Server examines the Multilink Forwarding Parameters 2163 information and rewrites the fields if the NA(WIN) was not already 2164 processed by a (local segment) Bridge as discussed in 2165 Section 3.2.7. 2167 o else, if the original IP packet destination matches a NCE, the 2168 Proxy/Sever uses one or more "reachable" neighbor interfaces in 2169 the entry for packet forwarding using OAL encapsulation and 2170 fragmentation according to the cached link-layer address 2171 information. If the neighbor interface is in a different OMNI 2172 link segment, the Proxy/Server performs OAL encapsulation and 2173 fragmentation, inserts an CRH-32 if necessary and forwards the 2174 resulting carrier packets via the spanning tree to a Bridge; 2175 otherwise, it forwards the carrier packets directly to the 2176 neighbor via INET encapsulation. If the neighbor is behind a NAT, 2177 this FHS Proxy/Server instead forwards initial carrier packets via 2178 a Bridge (or more directly via an LHS Proxy/Server) while sending 2179 an NS(NUD) to the neighbor. When the Proxy/Server receives the 2180 NA(NUD), it can begin forwarding carrier packets directly to the 2181 neighbor the same as discussed in Section 3.10.1 while sending 2182 additional NS(NUD) messages as necessary to maintain NAT state. 2183 Note that no direct bubbles are necessary since the Proxy/Server 2184 is by definition not located behind a NAT. 2186 o else, if the original IP destination matches a non-MNP route in 2187 the IP forwarding table or an ADM-LLA assigned to the Proxy/ 2188 Server's OMNI interface, the Proxy/Server acting as a Relay 2189 presents the original IP packet to the network layer for local 2190 delivery or IP forwarding. 2192 o else, the Proxy/Server initiates address resolution as discussed 2193 in Section 3.14, while submitting the original IP packets for OAL 2194 encapsulation and forwarding the resulting carrier packets into 2195 the secured spanning tree subject to rate limiting. 2197 When the Proxy/Server receives a carrier packet with OAL destination 2198 set to an MNP-ULA that does not match the MSP, it accepts the carrier 2199 packet only if data origin authentication succeeds and if there is a 2200 network layer routing table entry for a GUA route that matches the 2201 MNP-ULA. If there is no route, the Proxy/Server drops the carrier 2202 packet; otherwise, it reassembles and decapsulates to obtain the 2203 original IP packet then acts as a Relay to present it to the network 2204 layer where it will be delivered according to standard IP forwarding. 2206 When a Proxy/Server receives a carrier packet from one of its Client 2207 neighbors with OAL destination set to another node, it forwards the 2208 packets via a matching NCE or via the spanning tree if there is no 2209 matching entry. When the Proxy/Server receives a carrier packet with 2210 OAL destination set to the MNP-ULA of one of its Client neighbors 2211 established through RS/RA exchanges, it accepts the carrier packet 2212 only if data origin authentication succeeds. If the NCE state is 2213 DEPARTED, the Proxy/Server changes the OAL destination address to the 2214 ADM-ULA of the new Proxy/Server, then re-encapsulates the carrier 2215 packet and forwards it to a Bridge which will eventually deliver it 2216 to the new Proxy/Server. 2218 If the neighbor cache state for the MNP-ULA is REACHABLE, the Proxy/ 2219 Server forwards the carrier packets to the Client which then must 2220 reassemble. (Note that the Proxy/Server does not reassemble carrier 2221 packets not explicitly addressed to its own ADM-ULA, since some of 2222 the carrier packets of the same original IP packet could be forwarded 2223 through a different Proxy/Server.) In that case, the Client may 2224 receive fragments that are smaller than its link MTU but that can 2225 still be reassembled. 2227 FHS Clients maintain a single Hub Proxy/Server and one or more FHS 2228 Proxy/Servers. FHS Clients can forward carrier packets to the FHS 2229 Proxy/Server for a specific outbound underlying interface while also 2230 initiating route optimization, and the FHS Proxy/Server will 2231 reassemble, re-encapsulate and re-fragment then send the resulting 2232 carrier packets into the secured spanning tree subject to rate 2233 limiting while route optimization is in progress. The secured 2234 spanning tree carrier packets will arrive at the Hub LHS Proxy/Server 2235 for an LHS Client, which reassembles, re-encapsulates, re-fragments 2236 and forwards to the LHS Client. Once route optimization has 2237 converged, the FHS and LHS Clients can enact "shortcuts" to avoid 2238 slow-path forwarding of carrier packets over the secured spanning 2239 tree. 2241 Proxy/Servers process carrier packets with OAL destinations that do 2242 not match their ADM-ULA in the same manner as for traditional IP 2243 forwarding within the OAL, i.e., nodes use IP forwarding to forward 2244 packets not explicitly addressed to themselves. Proxy/Servers 2245 process carrier packets with their ADM-ULA as the destination by 2246 first examining the packet for a CRH-32 header or a compressed OAL 2247 header. In that case, the Proxy/Server examines the next MFVI in the 2248 carrier packet to locate the MFV entry in the MFIB for next hop 2249 forwarding (i.e., without examining IP addresses). When the Proxy/ 2250 Server forwards the carrier packet, it changes the destination 2251 address according to the MFVI value for the next hop found either in 2252 the CRH-32 header or in the node's own MFIB. 2254 Note: Proxy/Servers may receive carrier packets with CRH-32s that 2255 include additional forwarding information. Proxy/Servers use the 2256 forwarding information to determine the correct NCE and underlying 2257 interface for forwarding to the target Client, then remove the CRH-32 2258 and forward the carrier packet. If necessary, the Proxy/Server 2259 reassembles first before re-encapsulating (and possibly also re- 2260 fragmenting) then forwards to the target Client. For a full 2261 discussion see: Section 3.14.6. 2263 Note: Clients and their FHS Proxy/Server peers can exchange original 2264 IP packets over ANET underlying interfaces without invoking the OAL, 2265 since the ANET is secured at the link and physical layers. By 2266 forwarding original IP packets without invoking the OAL, however, the 2267 Client and Proxy/Server can engage only in classical path MTU 2268 discovery since the packets are subject to loss and/or corruption due 2269 to the various per-link MTU limitations that may occur within the 2270 ANET. Moreover, the original IP packets do not include either the 2271 OAL integrity check or per-packet Identification values that can be 2272 used for data origin authentication and link-layer retransmissions. 2273 The tradeoff therefore involves an assessment of the per-packet 2274 encapsulation overhead saved by bypassing the OAL vs. inheritance of 2275 classical network "brittleness". (Note however that ANET peers can 2276 send small original IP packets without invoking the OAL, while 2277 invoking the OAL for larger packets. This presents the beneficial 2278 aspects of both small packet efficiency and large packet robustness.) 2280 Note: When a Proxy/Server receives a (non-OAL) original IP packet 2281 from an ANET Client, or a carrier packet with OAL destination set to 2282 its own ADM-ULA from any Client, the Proxy/Server reassembles if 2283 necessary then performs ROS functions on behalf of the Client. The 2284 Client may at some later time begin sending carrier packets to the 2285 OAL address of the actual target instead of the Proxy/Server, at 2286 which point it may begin functioning as an ROS on its own behalf and 2287 thereby "override" the Proxy/Server's ROS role. 2289 Note; Proxy/Servers drop any original IP packets (received either 2290 directly from an ANET Client or following reassembly of carrier 2291 packets received from an ANET/INET Client) with a destination that 2292 corresponds to the Client's delegated MNP. Similarly, Proxy/Servers 2293 drop any carrier packet received with both a source and destination 2294 that correspond to the Client's delegated MNP regardless of their 2295 OMNI link point of origin. These checks are necessary to prevent 2296 Clients from either accidentally or intentionally establishing 2297 endless loops that could congest Proxy/Servers and/or ANET/INET 2298 links. 2300 Note: Proxy/Servers forward secure control plane carrier packets via 2301 the SRT secured spanning tree and forward other carrier packets via 2302 the unsecured spanning tree. When a Proxy/Server receives a carrier 2303 packet from the secured spanning tree, it considers the message as 2304 authentic without having to verify upper layer authentication 2305 signatures. When a Proxy/Server receives a carrier packet from the 2306 unsecured spanning tree, it verifies any upper layer authentication 2307 signatures and/or forwards the unsecured message toward the 2308 destination which must apply data origin authentication. 2310 Note: If the Proxy/Server has multiple original IP packets to send to 2311 the same neighbor, it can concatenate them in a single OAL super- 2312 packet [I-D.templin-6man-omni]. 2314 3.10.3. Bridge Forwarding Algorithm 2316 Bridges forward spanning tree carrier packets while decrementing the 2317 OAL header Hop Count but not the original IP header Hop Count/TTL. 2318 Bridges convey carrier packets that encapsulate critical IPv6 ND 2319 control messages or routing protocol control messages via the secured 2320 spanning tree, and may convey other carrier packets via the unsecured 2321 spanning tree or via more direct paths according to MFIB information. 2322 When the Bridge receives a carrier packet, it removes the outer *NET 2323 header and searches for an MFIB entry that matches an MFVI or an IP 2324 forwarding table entry that matches the OAL destination address. 2326 Bridges process carrier packets with OAL destinations that do not 2327 match their ADM-ULA or the SRT Subnet Router Anycast address in the 2328 same manner as for traditional IP forwarding within the OAL, i.e., 2329 nodes use IP forwarding to forward packets not explicitly addressed 2330 to themselves. Bridges process carrier packets with their ADM-ULA or 2331 the SRT Subnet Router Anycast address as the destination by first 2332 examining the packet for a CRH-32 header or a compressed OAL header. 2333 In that case, the Bridge examines the next MFVI in the carrier packet 2334 to locate the MFV entry in the MFIB for next hop forwarding (i.e., 2335 without examining IP addresses). When the Bridge forwards the 2336 carrier packet, it changes the destination address according to the 2337 MFVI value for the next hop found either in the CRH-32 header or in 2338 the node's own MFIB. 2340 Bridges forward carrier packets received from a first segment via the 2341 SRT secured spanning tree to the next segment also via the secured 2342 spanning tree. Bridges forward carrier packets received from a first 2343 segment via the unsecured spanning tree to the next segment also via 2344 the unsecured spanning tree. Bridges use a single IPv6 routing table 2345 that always determines the same next hop for a given OAL destination, 2346 where the secured/unsecured spanning tree is determined through the 2347 selection of the underlying interface to be used for transmission 2348 (i.e., a secured tunnel or an open INET interface). 2350 3.11. OMNI Interface Error Handling 2352 When an AERO node admits an original IP packet into the OMNI 2353 interface, it may receive link-layer or network-layer error 2354 indications. The AERO node may also receive OMNI link error 2355 indications in OAL-encapsulated uNA messages that include 2356 authentication signatures. 2358 A link-layer error indication is an ICMP error message generated by a 2359 router in the INET on the path to the neighbor or by the neighbor 2360 itself. The message includes an IP header with the address of the 2361 node that generated the error as the source address and with the 2362 link-layer address of the AERO node as the destination address. 2364 The IP header is followed by an ICMP header that includes an error 2365 Type, Code and Checksum. Valid type values include "Destination 2366 Unreachable", "Time Exceeded" and "Parameter Problem" 2367 [RFC0792][RFC4443]. (OMNI interfaces ignore link-layer IPv4 2368 "Fragmentation Needed" and IPv6 "Packet Too Big" messages for carrier 2369 packets that are no larger than the minimum/path MPS as discussed in 2370 Section 3.9, however these messages may provide useful hints of probe 2371 failures during path MPS probing.) 2373 The ICMP header is followed by the leading portion of the carrier 2374 packet that generated the error, also known as the "packet-in-error". 2375 For ICMPv6, [RFC4443] specifies that the packet-in-error includes: 2376 "As much of invoking packet as possible without the ICMPv6 packet 2377 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 2378 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 2379 "Internet Header + 64 bits of Original Data Datagram", however 2380 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 2381 ICMP datagram SHOULD contain as much of the original datagram as 2382 possible without the length of the ICMP datagram exceeding 576 2383 bytes". 2385 The link-layer error message format is shown in Figure 4: 2387 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2388 ~ ~ 2389 | IP Header of link layer | 2390 | error message | 2391 ~ ~ 2392 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2393 | ICMP Header | 2394 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 2395 ~ ~ P 2396 | carrier packet *NET and OAL | a 2397 | encapsulation headers | c 2398 ~ ~ k 2399 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 2400 ~ ~ t 2401 | original IP packet headers | 2402 | (first-fragment only) | i 2403 ~ ~ n 2404 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2405 ~ ~ e 2406 | Portion of the body of | r 2407 | the original IP packet | r 2408 | (all fragments) | o 2409 ~ ~ r 2410 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 2412 Figure 4: OMNI Interface Link-Layer Error Message Format 2414 The AERO node rules for processing these link-layer error messages 2415 are as follows: 2417 o When an AERO node receives a link-layer Parameter Problem message, 2418 it processes the message the same as described as for ordinary 2419 ICMP errors in the normative references [RFC0792][RFC4443]. 2421 o When an AERO node receives persistent link-layer Time Exceeded 2422 messages, the IP ID field may be wrapping before earlier fragments 2423 awaiting reassembly have been processed. In that case, the node 2424 should begin including integrity checks and/or institute rate 2425 limits for subsequent packets. 2427 o When an AERO node receives persistent link-layer Destination 2428 Unreachable messages in response to carrier packets that it sends 2429 to one of its neighbor correspondents, the node should process the 2430 message as an indication that a path may be failing, and 2431 optionally initiate NUD over that path. If it receives 2432 Destination Unreachable messages over multiple paths, the node 2433 should allow future carrier packets destined to the correspondent 2434 to flow through a default route and re-initiate route 2435 optimization. 2437 o When an AERO Client receives persistent link-layer Destination 2438 Unreachable messages in response to carrier packets that it sends 2439 to one of its neighbor Proxy/Servers, the Client should mark the 2440 path as unusable and use another path. If it receives Destination 2441 Unreachable messages on many or all paths, the Client should 2442 associate with a new Proxy/Server and release its association with 2443 the old Proxy/Server as specified in Section 3.16.5. 2445 o When an AERO Proxy/Server receives persistent link-layer 2446 Destination Unreachable messages in response to carrier packets 2447 that it sends to one of its neighbor Clients, the Proxy/Server 2448 should mark the underlying path as unusable and use another 2449 underlying path. 2451 o When an AERO Proxy/Server receives link-layer Destination 2452 Unreachable messages in response to a carrier packet that it sends 2453 to one of its permanent neighbors, it treats the messages as an 2454 indication that the path to the neighbor may be failing. However, 2455 the dynamic routing protocol should soon reconverge and correct 2456 the temporary outage. 2458 When an AERO Bridge receives a carrier packet for which the network- 2459 layer destination address is covered by an MSP, the Bridge drops the 2460 packet if there is no more-specific routing information for the 2461 destination and returns an OMNI interface Destination Unreachable 2462 message subject to rate limiting. 2464 When an AERO node receives a carrier packet for which reassembly is 2465 currently congested, it returns an OMNI interface Packet Too Big 2466 (PTB) message as discussed in [I-D.templin-6man-omni] (note that the 2467 PTB messages could indicate either "hard" or "soft" errors). 2469 AERO nodes include ICMPv6 error messages intended for the OAL source 2470 as sub-options in the OMNI option of secured uNA messages. When the 2471 OAL source receives the uNA message, it can extract the ICMPv6 error 2472 message enclosed in the OMNI option and either process it locally or 2473 translate it into a network-layer error to return to the original 2474 source. 2476 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 2478 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 2479 coordinated as discussed in the following Sections. 2481 3.12.1. AERO Service Model 2483 Each AERO Proxy/Server on the OMNI link is configured to facilitate 2484 Client prefix delegation/registration requests. Each Proxy/Server is 2485 provisioned with a database of MNP-to-Client ID mappings for all 2486 Clients enrolled in the AERO service, as well as any information 2487 necessary to authenticate each Client. The Client database is 2488 maintained by a central administrative authority for the OMNI link 2489 and securely distributed to all Proxy/Servers, e.g., via the 2490 Lightweight Directory Access Protocol (LDAP) [RFC4511], via static 2491 configuration, etc. Clients receive the same service regardless of 2492 the Proxy/Servers they select. 2494 Clients associate each of their underlying interfaces with a FHS 2495 Proxy/Server. Each FHS Proxy/Server may locally service one or more 2496 of the Client's underlying interfaces, and the Client selects one 2497 among them to serve as the Hub Proxy/Server. The Hub Proxy/Server is 2498 responsible for short-term packet forwarding, for acting as a 2499 mobility anchor point and for acting as an ROR for NS(AR) messages 2500 directed to the Client. All of the Client's other FHS Proxy/Servers 2501 forward proxyed copies of RS/RA messages between the Hub Proxy/Server 2502 and Client without assuming the Hub role functions themselves. 2504 Each Client associates with a single Hub Proxy/Server at a time, 2505 while all FHS Proxy/Servers are candidates for providing the Hub role 2506 for other Clients. An FHS Proxy/Server assumes the Hub role when it 2507 receives an RS message with its own ADM-LLA or All-Routers multicast 2508 as the destination. An FHS Proxy/Server assumes the proxy role when 2509 it receives an RS message with the ADM-LLA of another Proxy/Server as 2510 the destination. 2512 AERO Clients and Proxy/Servers use IPv6 ND messages to maintain 2513 neighbor cache entries. AERO Proxy/Servers configure their OMNI 2514 interfaces as advertising NBMA interfaces, and therefore send unicast 2515 RA messages with a short Router Lifetime value (e.g., ReachableTime 2516 seconds) in response to a Client's RS message. Thereafter, Clients 2517 send additional RS messages to keep Proxy/Server state alive. 2519 AERO Clients and Hub Proxy/Servers include prefix delegation and/or 2520 registration parameters in RS/RA messages (see 2521 [I-D.templin-6man-omni]). The IPv6 ND messages are exchanged between 2522 the Client and Hub Proxy/Server (via any FHS Proxy/Servers acting as 2523 proxies) according to the prefix management schedule required by the 2524 service. If the Client knows its MNP in advance, it can employ 2525 prefix registration by including its MNP-LLA as the source address of 2526 an RS message and with an OMNI option with valid prefix registration 2527 information for the MNP. If the Hub Proxy/Server accepts the 2528 Client's MNP assertion, it injects the MNP into the routing system 2529 and establishes the necessary neighbor cache state. If the Client 2530 does not have a pre-assigned MNP, it can instead employ prefix 2531 delegation by including the unspecified address (::) as the source 2532 address of an RS message and with an OMNI option with prefix 2533 delegation parameters to request an MNP. 2535 The following sections specify the Client and Proxy/Server behavior. 2537 3.12.2. AERO Client Behavior 2539 AERO Clients discover the addresses of candidate FHS Proxy/Servers by 2540 resolving the Potential Router List (PRL) in a similar manner as 2541 described in [RFC5214]. Discovery methods include static 2542 configuration (e.g., a flat-file map of Proxy/Server addresses and 2543 locations), or through an automated means such as Domain Name System 2544 (DNS) name resolution [RFC1035]. Alternatively, the Client can 2545 discover Proxy/Server addresses through a layer 2 data link login 2546 exchange, or through a unicast RA response to a multicast/anycast RS 2547 as described below. In the absence of other information, the Client 2548 can resolve the DNS Fully-Qualified Domain Name (FQDN) 2549 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 2550 text string and "[domainname]" is a DNS suffix for the OMNI link 2551 (e.g., "example.com"). 2553 To associate with a Hub Proxy/Server over a first underlying 2554 interface, the Client acts as a requesting router to request MNPs by 2555 preparing an RS message with prefix management parameters. If the 2556 Client already knows the Proxy/Server's ADM-LLA, it includes the LLA 2557 as the network-layer destination address; otherwise, the Client 2558 includes the (link-local) All-Routers multicast as the network-layer 2559 destination. The Client can use its MNP-LLA as the network-layer 2560 source address and include an OMNI option with prefix registration 2561 information. If the Client does not yet have an MNP-LLA, it instead 2562 sets the network-layer source address to unspecified (::) and 2563 includes prefix delegation parameters in the OMNI option (see: 2564 [I-D.templin-6man-omni]). 2566 The Client next includes an authentication sub-option if necessary 2567 and Multilink Forwarding Parameters corresponding to the underlying 2568 interface over which it will send the RS message. Next, the Client 2569 submits the RS for OAL encapsulation and fragmentation if necessary 2570 with its own MNP-ULA and the Proxy/Server's ADM-ULA or an OMNI IPv6 2571 anycast address as the OAL addresses while selecting an 2572 Identification value and invoking window synchronization as specified 2573 in [I-D.templin-6man-omni]. 2575 The Client then sends the RS (either directly via Direct interfaces, 2576 via a VPN for VPNed interfaces, via an access router for ANET 2577 interfaces or via INET encapsulation for INET interfaces) then waits 2578 up to RetransTimer milliseconds for an RA message reply (see 2579 Section 3.12.3) (retrying up to MAX_RTR_SOLICITATIONS). If the 2580 Client receives no RAs, or if it receives an RA with Router Lifetime 2581 set to 0, the Client SHOULD abandon attempts through the first 2582 candidate Hub Proxy/Server and try another FHS Proxy/Server. 2583 Otherwise, the Client processes the prefix information found in the 2584 RA message. 2586 When the Client processes an RA, it first performs OAL reassembly and 2587 decapsulation if necessary then creates a NCE with the Hub Proxy/ 2588 Server's ADM-LLA as the network-layer address and the Hub Proxy/ 2589 Server's encapsulation and/or link-layer addresses as the link-layer 2590 address. The Client then caches the Multilink Forwarding Parameters 2591 information. The Client next records the RA Router Lifetime field 2592 value in the NCE as the time for which the Hub Proxy/Server has 2593 committed to maintaining the MNP in the routing system via this 2594 underlying interface, and caches the other RA configuration 2595 information including Cur Hop Limit, M and O flags, Reachable Time 2596 and Retrans Timer. The Client then autoconfigures MNP-LLAs for any 2597 delegated MNPs and assigns them to the OMNI interface. The Client 2598 also caches any MSPs included in Route Information Options (RIOs) 2599 [RFC4191] as MSPs to associate with the OMNI link, and assigns the 2600 MTU value in the MTU option to the underlying interface. 2602 The Client then registers its additional underlying interfaces with 2603 FHS Proxy/Servers for those interfaces discovered by sending RS 2604 messages via each additional interface but with the ADM-LLA of the 2605 Hub Proxy/Server as the destination. The additional FHS Proxy/ 2606 Servers will assume the proxy role and forward proxyed copies of the 2607 RS/RA exchanges between the Client and Hub Proxy/Server. The Client 2608 finally sub-delegates the MNPs to its attached EUNs and/or the 2609 Client's own internal virtual interfaces as described in 2610 [I-D.templin-v6ops-pdhost] to support the Client's downstream 2611 attached "Internet of Things (IoT)". The Client then sends 2612 additional RS messages over each underlying interface before the 2613 Router Lifetime received for that interface expires. 2615 After the Client registers its underlying interfaces, it may wish to 2616 change one or more registrations, e.g., if an interface changes 2617 address or becomes unavailable, if traffic selectors change, etc. To 2618 do so, the Client prepares an RS message to send over any available 2619 underlying interface as above. The RS includes an OMNI option with 2620 prefix registration/delegation information and with Multilink 2621 Forwarding Parameters specific to the selected underlying interface. 2622 When the Client receives the Hub Proxy/Server's RA response, it has 2623 assurance that the Proxy/Server has been updated with the new 2624 information. 2626 If the Client wishes to discontinue use of a Hub Proxy/Server it 2627 issues an RS message over any underlying interface with an OMNI 2628 option with a prefix release indication. When the Hub Proxy/Server 2629 processes the message, it releases the MNP, sets the NCE state for 2630 the Client to DEPARTED and returns an RA reply with Router Lifetime 2631 set to 0. After a short delay (e.g., 2 seconds), the Hub Proxy/ 2632 Server withdraws the MNP from the routing system. 2634 3.12.3. AERO Proxy/Server Behavior 2636 AERO Proxy/Servers act as both IP routers and IPv6 ND proxies, and 2637 support a prefix delegation/registration service for Clients. Proxy/ 2638 Servers arrange to add their ADM-LLAs to the PRL maintained in a 2639 static map of Proxy/Server addresses for the link, the DNS resource 2640 records for the FQDN "linkupnetworks.[domainname]", etc. before 2641 entering service. The PRL should be arranged such that Clients can 2642 discover the addresses of Proxy/Servers that are geographically and/ 2643 or topologically "close" to their underlying network connections. 2645 When an FHS Proxy/Server receives a prospective Client's RS message, 2646 it SHOULD return an immediate RA reply with Router Lifetime set to 0 2647 if it is currently too busy or otherwise unable to service the 2648 Client. Otherwise, the Proxy/Server performs OAL reassembly if 2649 necessary, then decapsulates and authenticates the RS message. If 2650 the RS message destination is All-Routers multicast or the Proxy/ 2651 Server's own ADM-LLA, the Proxy/Server assumes the Hub role. If the 2652 RS message destination is the ADM-LLA of another node, the Proxy/ 2653 Server assumes the proxy role and forwards the RS to the Hub Proxy/ 2654 server via the secured spanning tree. 2656 The Hub Proxy/Server then determines the correct MNPs to provide to 2657 the Client by processing the MNP-LLA prefix parameters and/or the 2658 DHCPv6 OMNI sub-option. When the Hub Proxy/Server returns the MNPs, 2659 it also creates a forwarding table entry for the MNP-ULA 2660 corresponding to each MNP resulting in a BGP update (see: 2661 Section 3.2.3). For IPv6, the Hub Proxy/Server creates an IPv6 2662 forwarding table entry for each MNP. For IPv4, the Hub Proxy/Server 2663 creates an IPv6 forwarding table entry with the IPv4-compatibility 2664 MNP-ULA prefix corresponding to the IPv4 address. 2666 The Hub Proxy/Server next creates a NCE for the Client using the base 2667 MNP-LLA as the network-layer address. Next, the Hub Proxy/Server 2668 updates the NCE by recording the information in the Multilink 2669 Forwarding Parameters sub-option in the RS OMNI option. The Hub 2670 Proxy/Server also records the actual OAL/*NET addresses and RS 2671 message window synchronization parameters (if any) in the NCE. 2673 Next, the Hub Proxy/Server prepares an RA message using its ADM-LLA 2674 as the network-layer source address and the network-layer source 2675 address of the RS message as the network-layer destination address. 2676 The Hub Proxy/Server sets the Router Lifetime to the time for which 2677 it will maintain both this underlying interface individually and the 2678 NCE as a whole. The Hub Proxy/Server also sets Cur Hop Limit, M and 2679 O flags, Reachable Time and Retrans Timer to values appropriate for 2680 the OMNI link. The Hub Proxy/Server includes the MNPs, any other 2681 prefix management parameters and an OMNI option with a Multilink 2682 Forwarding Parameters sub-option with FHS addressing information 2683 filled out. The Hub Proxy/Server then includes one or more RIOs that 2684 encode the MSPs for the OMNI link, plus an MTU option (see 2685 Section 3.9). The Hub Proxy/Server finally forwards the message to 2686 the Client using OAL encapsulation/fragmentation if necessary while 2687 including an acknowledgement if the RS invoked window 2688 synchronization. 2690 After the initial RS/RA exchange, the Hub Proxy/Server maintains a 2691 ReachableTime timer for each of the Client's underlying interfaces 2692 individually (and for the Client's NCE collectively) set to expire 2693 after ReachableTime seconds. If the Client (or an FHS Proxy/Server) 2694 issues additional RS messages, the Hub Proxy/Server sends an RA 2695 response and resets ReachableTime. If the Hub Proxy/Server receives 2696 an IPv6 ND message with a prefix release indication it sets the 2697 Client's NCE to the DEPARTED state and withdraws the MNP from the 2698 routing system after a short delay (e.g., 2 seconds). If 2699 ReachableTime expires before a new RS is received on an individual 2700 underlying interface, the Hub Proxy/Server marks the interface as 2701 DOWN. If ReachableTime expires before any new RS is received on any 2702 individual underlying interface, the Hub Proxy/Server sets the NCE 2703 state to STALE and sets a 10 second timer. If the Hub Proxy/Server 2704 has not received a new RS or uNA message with a prefix release 2705 indication before the 10 second timer expires, it deletes the NCE and 2706 withdraws the MNP from the routing system. 2708 The Hub Proxy/Server processes any IPv6 ND messages pertaining to the 2709 Client and returns an NA/RA reply in response to solicitations. The 2710 Hub Proxy/Server may also issue unsolicited RA messages, e.g., with 2711 reconfigure parameters to cause the Client to renegotiate its prefix 2712 delegation/registrations, with Router Lifetime set to 0 if it can no 2713 longer service this Client, etc. Finally, If the NCE is in the 2714 DEPARTED state, the Hub Proxy/Server deletes the entry after 2715 DepartTime expires. 2717 The Hub Proxy/Server may also receive carrier packets via the secured 2718 spanning tree that contain initial data packets sent while route 2719 optimization is in progress. The Hub Proxy/Server reassembles, then 2720 re-encapsulates/re-fragments and forwards the packets to the target 2721 Client. Although these fragments will have traversed the secured 2722 spanning tree, the security only assures correct reassembly and does 2723 not assure message content security. 2725 Note: Clients SHOULD notify former Hub Proxy/Servers of their 2726 departures, but Hub Proxy/Servers are responsible for expiring 2727 neighbor cache entries and withdrawing routes even if no departure 2728 notification is received (e.g., if the Client leaves the network 2729 unexpectedly). Hub Proxy/Servers SHOULD therefore set Router 2730 Lifetime to ReachableTime seconds in solicited RA messages to 2731 minimize persistent stale cache information in the absence of Client 2732 departure notifications. A short Router Lifetime also ensures that 2733 proactive RS/RA messaging between Clients and FHS Proxy/Servers will 2734 keep any NAT state alive (see above). 2736 Note: All Proxy/Servers on an OMNI link MUST advertise consistent 2737 values in the RA Cur Hop Limit, M and O flags, Reachable Time and 2738 Retrans Timer fields the same as for any link, since unpredictable 2739 behavior could result if different Proxy/Servers on the same link 2740 advertised different values. 2742 3.12.3.1. DHCPv6-Based Prefix Registration 2744 When a Client is not pre-provisioned with an MNP-LLA, it will need 2745 for the Hub Proxy/Server to select one or more MNPs on its behalf and 2746 set up the correct state in the AERO routing service. (A Client with 2747 a pre-provisioned MNP may also request the Hub Proxy/Server to select 2748 additional MNPs.) The DHCPv6 service [RFC8415] is used to support 2749 this requirement. 2751 When a Client needs to have the Hub Proxy/Server select MNPs, it 2752 sends an RS message with source address set to the unspecified 2753 address (::) and with an OMNI option that includes a DHCPv6 message 2754 sub-option with DHCPv6 Prefix Delegation (DHCPv6-PD) parameters. 2755 When the Hub Proxy/Server receives the RS message, it extracts the 2756 DHCPv6-PD message from the OMNI option. 2758 The Hub Proxy/Server then acts as a "Proxy DHCPv6 Client" in a 2759 message exchange with the locally-resident DHCPv6 server, which 2760 delegates MNPs and returns a DHCPv6-PD Reply message. (If the Hub 2761 Proxy/Server wishes to defer creation of MN state until the DHCPv6-PD 2762 Reply is received, it can instead act as a Lightweight DHCPv6 Relay 2763 Agent per [RFC6221] by encapsulating the DHCPv6-PD message in a 2764 Relay-forward/reply exchange with Relay Message and Interface ID 2765 options.) 2767 When the Hub Proxy/Server receives the DHCPv6-PD Reply, it adds a 2768 route to the routing system and creates an MNP-LLA based on the 2769 delegated MNP. The Hub Proxy/Server then sends an RA back to the 2770 Client with the (newly-created) MNP-LLA as the destination address 2771 and with the DHCPv6-PD Reply message coded in the OMNI option. When 2772 the Client receives the RA, it creates a default route, assigns the 2773 Subnet Router Anycast address and sets its MNP-LLA based on the 2774 delegated MNP. 2776 Note: See [I-D.templin-6man-omni] for an MNP delegation alternative 2777 that avoids including a DHCPv6 message sub-option in the RS. Namely, 2778 when the Client requests a single MNP it can set the RS source to the 2779 unspecified address (::) and include a Node Identification sub-option 2780 and Preflen in the OMNI option (but with no DHCPv6 message sub- 2781 option). When the Hub Proxy/Server receives the RS message, it 2782 forwards a self-generated DHCPv6 Solicit message to the DHCPv6 server 2783 on behalf of the Client. When the Hub Proxy/Server receives the 2784 DHCPv6 Reply, it prepares an RA message with an OMNI option with 2785 Preflen information (but with no DHCPv6 message sub-option), then 2786 places the (newly-created) MNP-LLA in the RA destination address and 2787 returns the message to the Client. 2789 3.13. AERO Proxy/Server Coordination 2791 OMNI link Clients register with FHS Proxy/Servers for each underlying 2792 interface. Each of the Client's FHS Proxy/Servers must inform a 2793 single Hub Proxy/Server of all of the Client's additional underlying 2794 interfaces. For Clients on Direct and VPNed underlying interfaces, 2795 the FHS Proxy/Server for each interface is directly connected, for 2796 Clients on ANET underlying interfaces the FHS Proxy/Server is located 2797 on the ANET/INET boundary, and for Clients on INET underlying 2798 interfaces the FHS Proxy/Server is located somewhere in the connected 2799 Internetwork. When FHS Proxy/Server "A" processes a Client 2800 registration, it must either assume the Hub role or forward a proxyed 2801 registration to another FHS Proxy/Server acting as the Hub. Proxy/ 2802 Servers satisfy these requirements as follows: 2804 o when Proxy/Server "A" receives a Client RS message, it first 2805 verifies that the OAL Identification is within the window for the 2806 NCE that matches the MNP-ULA for this Client neighbor and 2807 authenticates the message. (If no NCE was found, Proxy/Server "A 2808 instead creates one in the STALE state and returns an RA message 2809 with an authentication signature if necessary and any window 2810 synchronization parameters.) Proxy/Server "A" then examines the 2811 network-layer destination address. If the destination address is 2812 the ADM-LLA of a different Proxy/Server "B", Proxy/Server "A" 2813 prepares a separate proxyed version of the RS message with an OAL 2814 header with source set to its own ADM-ULA and destination set to 2815 Proxy/Server B's ADM-ULA. Proxy/Server "A" also writes its own 2816 information over the Multilink Forwarding Parameters sub-option 2817 supplied by the Client then sets the S/T-omIndex to the value for 2818 this Client underlying interface, then forwards the message into 2819 the OMNI link secured spanning tree. 2821 o when Proxy/Server "B" receives the RS, it assume the Hub role and 2822 creates or updates a NCE for the Client with FHS Proxy/Server 2823 "A"'s Multilink Forwarding Parameters as the link-layer address 2824 information for this S/T-omIndex and caches any window 2825 synchronization parameters supplied by the Client. Hub Proxy/ 2826 Server "B" then prepares an RA message with source set to its own 2827 LLA and destination set to the Client's MNP-LLA, and with any 2828 window synchronization acknowledgements. Hub Proxy/Server "B" 2829 then encapsulates the RA in an OAL header with source set to its 2830 own ADM-ULA and destination set to the ADM-ULA of FHS Proxy/Server 2831 "A, performs fragmentation if necessary, then sends the resulting 2832 carrier packets into the secured spanning tree. 2834 o when Proxy/Server "A" reassembles the RA, it locates the Client 2835 NCE based on the RA destination LLA. Proxy/Server "A" then re- 2836 encapsulates the RA message with OAL source set to its own ADM-ULA 2837 and OAL destination set to the MNP-ULA of the Client, includes an 2838 authentication signature if necessary, and echoes the Multilink 2839 Forwarding Parameters sub-option. Proxy/Server "A" then fragments 2840 if necessary and returns the fragments to the Client. 2842 o The Client repeats this process over each of its additional 2843 underlying interfaces while treating each FHS Proxy/Server "C", 2844 "D", "E", etc. as a proxy to facilitate RS/RA exchanges between 2845 the Hub and the Client. 2847 After the initial RS/RA exchanges each FHS Proxy/Server forwards any 2848 of the Client's carrier packets with OAL destinations for which there 2849 is no matching NCE to a Bridge using OAL encapsulation with its own 2850 ADM-ULA as the source and with destination determined by the Client. 2851 The Proxy/Server instead forwards any carrier packets destined to a 2852 neighbor cache target directly to the target according to the OAL/ 2853 link-layer information - the process of establishing neighbor cache 2854 entries is specified in Section 3.14. 2856 While the Client is still associated with each FHS Proxy/Server "A", 2857 "A" can send NS, RS and/or unsolicited NA messages to update the 2858 neighbor cache entries of other AERO nodes on behalf of the Client 2859 and/or to convey Multilink Forwarding Parameter updates. This allows 2860 for higher-frequency Proxy-initiated RS/RA messaging over well- 2861 connected INET infrastructure supplemented by lower-frequency Client- 2862 initiated RS/RA messaging over constrained ANET data links. 2864 If the Hub Proxy/Server "A" ceases to send solicited RAs, Proxy/ 2865 Servers "B", "C", "D" send unsolicited RAs over the Client's 2866 underlying interface with destination set to (link-local) All-Nodes 2867 multicast and with Router Lifetime set to zero to inform Clients that 2868 the Hub Proxy/Server has failed. Although Proxy/Servers "B", "C" and 2869 "D" can engage in IPv6 ND exchanges on behalf of the Client, the 2870 Client can also send IPv6 ND messages on its own behalf, e.g., if it 2871 is in a better position to convey state changes. The IPv6 ND 2872 messages sent by the Client include the Client's MNP-LLA as the 2873 source in order to differentiate them from the IPv6 ND messages sent 2874 by Proxy/Server "A". 2876 If the Client becomes unreachable over all underlying interface it 2877 serves, the Hub Proxy/Server sets the NCE state to DEPARTED and 2878 retains the entry for DepartTime seconds. While the state is 2879 DEPARTED, the Hub Proxy/Server forwards any carrier packets destined 2880 to the Client to a Bridge via OAL encapsulation. When DepartTime 2881 expires, the Hub Proxy/Server deletes the NCE and discards any 2882 further carrier packets destined to the former Client. 2884 In some ANETs that employ a Proxy/Server, the Client's MNP can be 2885 injected into the ANET routing system. In that case, the Client can 2886 send original IP packets without invoking the OAL so that the ANET 2887 routing system transports the original IP packets to the Proxy. This 2888 can be very beneficial, e.g., if the Client connects to the ANET via 2889 low-end data links such as some aviation wireless links. 2891 If the ANET first-hop access router is on the same underlying link as 2892 the Client and recognizes the AERO/OMNI protocol, the Client can 2893 avoid OAL encapsulation for both its control and data messages. When 2894 the Client connects to the link, it can send an unencapsulated RS 2895 message with source address set to its own MNP-LLA (or to a Temporary 2896 LLA), and with destination address set to the ADM-LLA of the Client's 2897 selected Proxy/Server or to (link-local) All-Routers multicast. The 2898 Client includes an OMNI option formatted as specified in 2899 [I-D.templin-6man-omni]. The Client then sends the unencapsulated RS 2900 message, which will be intercepted by the AERO-Aware access router. 2902 The ANET access router then performs OAL encapsulation on the RS 2903 message and forwards it to a Proxy/Server at the ANET/INET boundary. 2904 When the access router and Proxy/Server are one and the same node, 2905 the Proxy/Server would share and underlying link with the Client but 2906 its message exchanges with outside correspondents would need to pass 2907 through a security gateway at the ANET/INET border. The method for 2908 deploying access routers and Proxys (i.e. as a single node or 2909 multiple nodes) is an ANET-local administrative consideration. 2911 Note: When a Proxy/Server alters the IPv6 ND message contents before 2912 forwarding (e.g., such as altering the OMNI option contents), the 2913 IPv6 ND message checksum and/or authentication signature are 2914 invalidated. If the Proxy/Server forwards the message over the 2915 secured spanning tree, however, it need not re-calculate the 2916 checksum/signature since they will not be examined by the next hop. 2918 Note: When a Proxy/Server receives a secured Client NS message, it 2919 performs the same proxying procedures as for described for RS 2920 messages above. The proxying procedures for NS/NA message exchanges 2921 is specified in Section 3.14. 2923 3.13.1. Detecting and Responding to Proxy/Server Failures 2925 In environments where fast recovery from Proxy/Server failure is 2926 required, FHS Proxy/Servers SHOULD use proactive Neighbor 2927 Unreachability Detection (NUD) to track Hub Proxy/Server reachability 2928 in a similar fashion as for Bidirectional Forwarding Detection (BFD) 2929 [RFC5880]. Each FHS Proxy/Server can then quickly detect and react 2930 to failures so that cached information is re-established through 2931 alternate paths. The NS/NA(NUD) control messaging is carried only 2932 over well-connected ground domain networks (i.e., and not low-end 2933 aeronautical radio links) and can therefore be tuned for rapid 2934 response. 2936 FHS Proxy/Servers perform continuous NS/NA(NUD) exchanges with the 2937 Hub Proxy/Server in rapid succession, e.g., one exchange per second. 2938 The FHS Proxy/Server sends the NS(NUD) message via the spanning tree 2939 with its own ADM-LLA as the source and the ADM-LLA of the Hub Proxy/ 2940 Server as the destination, and the Hub Proxy/Server responds with an 2941 NA(NUD). When the FHS Proxy/Server is also sending RS messages to a 2942 Hub Proxy/Server on behalf of Clients, the resulting RA responses can 2943 be considered as equivalent hints of forward progress. This means 2944 that the FHS Proxy/Server need not also send a periodic NS(NUD) if it 2945 has already sent an RS within the same period. If the Hub Proxy/ 2946 Server fails (i.e., if the FHS Proxy/Server ceases to receive 2947 advertisements), the FHS Proxy/Server can quickly inform Clients by 2948 sending unsolicited RA messages 2950 The FHS Proxy/Server sends unsolicited RA messages with source 2951 address set to the Hub Proxy/Server's address, destination address 2952 set to (link-local) All-Nodes multicast, and Router Lifetime set to 2953 0. The FHS Proxy/Server SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA 2954 messages separated by small delays [RFC4861]. Any Clients that had 2955 been using the failed Hub Proxy/Server will receive the RA messages 2956 and select one of its other FHS Proxy/Servers to assume the Hub role 2957 (i.e., by sending an RS with destination set to the ADM-LLA of the 2958 new Hub). 2960 3.14. AERO Route Optimization 2962 AERO nodes invoke route optimization when they need to forward 2963 packets to new target destinations. Route optimization is based on 2964 IPv6 ND Address Resolution messaging between a Route Optimization 2965 Source (ROS) and the target Client's current Hub Proxy/Server acting 2966 as a Route Optimization Responder (ROR). Route optimization is 2967 initiated by the first eligible ROS closest to the source as follows: 2969 o For Clients on VPNed and Direct interfaces, the Client's FHS 2970 Proxy/Server is the ROS. 2972 o For Clients on ANET interfaces, either the Client or the FHS 2973 Proxy/Server may be the ROS. 2975 o For Clients on INET interfaces, the Client itself is the ROS. 2977 o For correspondent nodes on INET/EUN interfaces serviced by a 2978 Relay, the Relay is the ROS. 2980 The route optimization procedure is conducted between the ROS and the 2981 LHS Hub Proxy/Server/Relay for the target selected by routing as the 2982 ROR. In this arrangement, the ROS is always the Client or Proxy/ 2983 Server (or Relay) nearest the source over the selected source 2984 underlying interface, while the ROR is always the target's current 2985 Hub Proxy/Server. 2987 The AERO routing system directs a route optimization request sent by 2988 the ROS to the ROR, which returns a route optimization reply which 2989 must include information that is current, consistent and authentic. 2990 The ROS is responsible for periodically refreshing the route 2991 optimization, and the ROR is responsible for quickly informing the 2992 ROS of any changes. 2994 The procedures are specified in the following sections. 2996 3.14.1. Route Optimization Initiation 2998 When an original IP packet from a source node destined to a target 2999 node arrives, the ROS checks for a NCE with an MNP-LLA that matches 3000 the target destination. If there is a NCE in the REACHABLE state, 3001 the ROS invokes the OAL and forwards the resulting carrier packets 3002 according to the cached state then returns from processing. 3003 Otherwise, if there is no NCE the ROS creates one in the INCOMPLETE 3004 state. 3006 The ROS next invokes the OAL and forwards the resulting carrier 3007 packets into the secured spanning tree, then sends an NS message for 3008 Address Resolution (NS(AR)) to receive a solicited NA(AR) message 3009 from the ROR. While route optimization is in progress, the ROS may 3010 forward additional original IP packets into the secured spanning tree 3011 but if so must impose rate limiting to minimize secured spanning tree 3012 traffic as well as ROR reassembly. 3014 The NS(AR) message must be sent securely, and includes: 3016 o the LLA of the ROS as the source address. 3018 o the MNP-LLA corresponding to the original IP packet's destination 3019 as the Target Address, e.g., for 2001:db8:1:2::10:2000 the Target 3020 Address is fe80::2001:db8:1:2. 3022 o the Solicited-Node multicast address [RFC4291] formed from the 3023 lower 24 bits of the original IP packet's destination as the 3024 destination address, e.g., for 2001:db8:1:2::10:2000 the NS(AR) 3025 destination address is ff02:0:0:0:0:1:ff10:2000. 3027 The NS(AR) message also includes an OMNI option with an 3028 authentication sub-option if necessary and with Preflen set to the 3029 prefix length associated with the NS(AR) source. The ROS then 3030 selects an Identification value and submits the NS(AR) message for 3031 OAL encapsulation with OAL source set to its own ULA and OAL 3032 destination set to the ULA corresponding to the target. (The ROS 3033 does not include any window synchronization parameters, since it will 3034 not exchange other packet types with the ROR.) The ROS then sends 3035 the resulting carrier packet into the SRT secured spanning tree 3036 without decrementing the network-layer TTL/Hop Limit field. 3038 When the ROS is an INET Client, it must instead forward the resulting 3039 carrier packet to the ADM-ULA of one of its current Proxy/Servers. 3040 The Proxy/Server then verifies the NS(AR) authentication signature, 3041 then re-encapsulates with the OAL source set to its own ADM-ULA and 3042 OAL destination set to the ULA corresponding to the target and 3043 forwards the resulting carrier packets into the secured spanning tree 3044 on behalf of the Client. 3046 3.14.2. Relaying the NS(AR) *NET Packet(s) 3048 When the Bridge receives the carrier packet containing the RS from 3049 the ROS, it discards the *NET headers and determines the next hop by 3050 consulting its standard IPv6 forwarding table for the OAL header 3051 destination address. The Bridge then decrements the OAL header Hop- 3052 Limit, then re-encapsulates and forwards the carrier packet(s) via 3053 the secured spanning tree the same as for any IPv6 router, where it 3054 may traverse multiple OMNI link segments. The final-hop Bridge will 3055 deliver the carrier packet via the secured spanning tree to the ROR 3056 for the target. 3058 3.14.3. Processing the NS(AR) and Sending the NA(AR) 3060 When the ROR for the target receives the secured carrier packet, it 3061 examines the NS(AR) target to determine whether it has a matching NCE 3062 and/or non-MNP route. If there is no match, the ROR drops the 3063 message. Otherwise, the ROR continues processing as follows: 3065 o if the NS(AR) target matches a Client NCE in the DEPARTED state, 3066 the ROR re-encapsulates while setting the OAL source to the ULA of 3067 the ROS and OAL destination address to the ADM-ULA of the Client's 3068 new Proxy/Server. The ROR then forwards the resulting carrier 3069 packet over the secured spanning tree then returns from 3070 processing. 3072 o If the NS(AR) target matches the MNP-LLA of a Client NCE in the 3073 REACHABLE state, the ROR notes whether the NS (AR) arrived from 3074 the secured spanning tree then provides route optimization 3075 information on behalf of the Client. If the message arrived via 3076 the secured spanning tree the ROR need not perform further 3077 authentication; otherwise, it must verify the message 3078 authentication signature before accepting. 3080 o If the NS(AR) target matches one of its non-MNP routes, the ROR 3081 serves as both a Relay and a route optimization target, since the 3082 Relay forwards IP packets toward the (fixed network) target at the 3083 network layer. 3085 The ROR next checks the target NCE for a Report List entry that 3086 matches the NS(AR) source LLA/ULA of the ROS. If there is a Report 3087 List entry, the ROR refreshes ReportTime for this ROR; otherwise, the 3088 ROR creates a new entry for the ROS and records both the LLA and ULA. 3090 The ROR then prepares a (solicited) NA(AR) message to return to the 3091 ROS with the source address set to its own ADM-LLA, the destination 3092 address set to the NS(AR) LLA source address and the Target Address 3093 set to the target Client's MNP-LLA. The ROR includes an OMNI option 3094 with Preflen set to the prefix length associated with the NA(AR) 3095 source address, with S/T-omIndex set to the value that appeared in 3096 the NS(AR) and with Interface Attributes sub-options for all of the 3097 target's underlying interfaces with current information for each 3098 interface. 3100 For each Interface Attributes sub-option, the ROR sets the L2ADDR 3101 according to its own INET address for VPNed, Direct, ANET and NATed 3102 Client interfaces, or to the Client's INET address for native Client 3103 interfaces. The ROR then includes the lower 32 bits of its ADM-ULA 3104 as the LHS, encodes the ADM-ULA SRT prefix length in the SRT field 3105 and sets FMT as specified in Section 3.3. 3107 The ROR then sets the NA(AR) message R flag to 1 (as a router) and S 3108 flag to 1 (as a response to a solicitation) and sets the O flag to 0 3109 (as a proxy). The ROR finally submits the NA(AR) for OAL 3110 encapsulation with source set to its own ULA and destination set to 3111 the same ULA that appeared in the NS(AR) OAL source, then performs 3112 OAL encapsulation using the same Identification value that appeared 3113 in the NS(AR) and finally forwards the resulting (*NET-encapsulated) 3114 carrier packet via the secured spanning tree without decrementing the 3115 network-layer TTL/Hop Limit field. 3117 3.14.4. Relaying the NA(AR) 3119 When the Bridge receives NA(AR) carrier packet from the ROR, it 3120 discards the *NET header and determines the next hop by consulting 3121 its standard IPv6 forwarding table for the OAL header destination 3122 address. The Bridge then decrements the OAL header Hop-Limit, re- 3123 encapsulates the carrier packet and forwards it via the SRT secured 3124 spanning tree, where it may traverse multiple OMNI link segments. 3125 The final-hop Bridge will deliver the carrier packet via the secured 3126 spanning tree to a Proxy/Server for the ROS. 3128 3.14.5. Processing the NA(AR) 3130 When the ROS receives the NA(AR) message, it first searches for a NCE 3131 that matches the NA(AR) target address. The ROS then processes the 3132 message the same as for standard IPv6 Address Resolution [RFC4861]. 3133 In the process, it caches all OMNI option information in the target 3134 NCE (including all Interface Attributes), and caches the NA(AR) 3135 ADM-{LLA,ULA} source addresses as the addresses of the ROR. If the 3136 ROS receives additional NA(AR) or uNA messages for this target Client 3137 with the same ADM-LLA source address but a different ADM-ULA source 3138 address, it configures the ADM-LLA corresponding to the new ADM-ULA, 3139 then caches the new ADM-{LLA,ULA} and deprecates the former 3140 ADM-{LLA,ULA}. 3142 When the ROS is a Client, the SRT secured spanning tree will first 3143 deliver the solicited NA(AR) message to the local Proxy/Server, which 3144 re-encapsulates and forwards the message to the Client. If the 3145 Client is on a well-managed ANET, physical security and protected 3146 spectrum ensures security for the unmodified NA(AR); if the Client is 3147 on the open INET the Proxy/Server must instead include an 3148 authentication signature (while adjusting the OMNI option size, if 3149 necessary). The Proxy/Server uses its own ADM-ULA as the OAL source 3150 and the MNP-ULA of the Client as the OAL destination. 3152 3.14.6. Forwarding Packets to Route Optimized Targets 3154 After the ROS receives the route optimization NA(AR) and updates the 3155 target NCE, it sends additional NS(AR) messages to the ADM-ULA of the 3156 ROR to refresh the NCE ReachableTime before expiration while it still 3157 has sustained interest in this target. While the NCE remains 3158 REACHABLE, the ROS can forward packets along paths that use best 3159 underlying interface pairs based on local preferences and target 3160 Interface Attributes. The ROS selects target underlying interfaces 3161 according to traffic selectors and/or any other traffic 3162 discriminators, but must first establish window synchronization state 3163 for each target if necessary. 3165 The ROS initiates window synchronization through a secured uncast NS/ 3166 NA(WIN) exchange as specified in Section 3.2.7. The NS/NA(WIN) 3167 exchange is conducted over a first underlying interface pair and 3168 registers only those interfaces. If the ROS and target have 3169 additional underlying interface pairs serviced by the same source/ 3170 destination LLAs, they may register new interfaces by sending 3171 additional NS/NA(WIN) messages but need not include window 3172 synchronization parameters. If the ROS and target have additional 3173 underlying interface pairs services by different source/destination 3174 LLAs, they must include window synchronization parameters when they 3175 send NS/NA(WIN) messages to establish NCE state for the new source/ 3176 destination LLAs. 3178 After window synchronization state has been established, the ROS and 3179 target Client can begin forwarding carrier packets while performing 3180 additional NS/NA(WIN) exchanges as above to update window state, 3181 register new interfaces and/or test reachability. The ROS sends 3182 carrier packets to the FHS Bridge discovered through the NS/NA(WIN) 3183 exchange which verifies the Identification is in window for the 3184 target Client. The FHS Bridge then forwards the carrier packets over 3185 the unsecured spanning tree to the LHS Bridge, which forwards them 3186 via LHS encapsulation to the LHS Proxy/Server or directly to the 3187 target Client itself. The target Client in turn sends packets to the 3188 ROS in the reverse direction while forwarding through the Bridges to 3189 minimize Proxy/Server load whenever possible. 3191 While the ROS continues to actively forward packets to the target 3192 Client, it is responsible for updating window synchronization state 3193 and per-interface reachability before expiration. Window 3194 synchronization state is shared by all underlying interfaces in the 3195 ROS' NCE that use the same destination LLA so that a single NS/ 3196 NA(WIN) exchange applies for all interfaces regardless of the 3197 (single) interface used to conduct the exchange. However, the window 3198 synchronization exchange only confirms target Client reachability 3199 over the specific interface used to conduct the exchange. 3201 Reachability for other underlying interfaces that share the same 3202 window synchronization state must be determined individually using 3203 NS/NA(NUD) messages which need not be secured as long as they use in- 3204 window Identifications and do not update other state information. 3206 3.15. Neighbor Unreachability Detection (NUD) 3208 AERO nodes perform Neighbor Unreachability Detection (NUD) per 3209 [RFC4861] either reactively in response to persistent link-layer 3210 errors (see Section 3.11) or proactively to confirm reachability. 3211 The NUD algorithm is based on periodic control message exchanges and 3212 may further be seeded by IPv6 ND hints of forward progress, but care 3213 must be taken to avoid inferring reachability based on spoofed 3214 information. For example, IPv6 ND message exchanges that include 3215 authentication codes and/or in-window Identifications may be 3216 considered as acceptable hints of forward progress, while spurious 3217 random carrier packets should be ignored. 3219 AERO nodes can perform NS/NA(NUD) exchanges over the OMNI link 3220 secured spanning tree (i.e. the same as described above for NS/ 3221 NA(WIN)) to test reachability without risk of DoS attacks from nodes 3222 pretending to be a neighbor. These NS/NA(NUD) messages use the 3223 unicast LLAs and ULAs of the parties involved in the NUD test. When 3224 only reachability information is required without updating any other 3225 NCE state, AERO nodes can instead perform NS/NA(NUD) exchanges 3226 directly between neighbors without employing the secured spanning 3227 tree as long as they include in-window Identifications and either an 3228 authentication signature or checksum. 3230 When an ROR directs an ROS to a target neighbor with one or more 3231 link-layer addresses, the ROS probes each unsecured target underlying 3232 interface either proactively or on-demand of carrier packets directed 3233 to the path by multilink forwarding to maintain the interface's state 3234 as reachable. Probing is performed through NS(NUD) messages over the 3235 SRT secured or unsecured spanning tree, or through NS(NUD) messages 3236 sent directly to an underlying interface of the target itself. While 3237 testing a target underlying interface, the ROS can optionally 3238 continue to forward carrier packets via alternate interfaces and/or 3239 maintain a small queue of carrier packets until target reachability 3240 is confirmed. 3242 NS(NUD) messages are encapsulated, fragmented and transmitted as 3243 carrier packets the same as for ordinary original IP data packets, 3244 however the encapsulated destinations are the LLA of the ROS and 3245 either the ADM-LLA of the LHS Proxy/Server or the MNP-LLA of the 3246 target itself. The ROS encapsulates the NS(NUD) message the same as 3247 described in Section 3.2.7 and sets the NS(NUD) OMNI header S/ 3248 T-omIndex to identify the underlying interface used for forwarding 3249 (or to 0 if any underlying interface can be used). The ROS then 3250 fragments the OAL packet and forwards the resulting carrier packets 3251 into the unsecured spanning tree or via direct encapsulation for 3252 local segment targets. 3254 When the target receives the NS(NUD) carrier packets, it verifies 3255 that it has a NCE for this ROS and that the Identification is in- 3256 window, then submits the carrier packets for reassembly. The node 3257 then verifies the authentication signature or checksum, then searches 3258 for Interface Attributes in its NCE for the ROS that match the 3259 NS(NUD) S/T-omIndex for the NA(NUD) reply. The node then prepares 3260 the NA(NUD) with the source and destination LLAs reversed, 3261 encapsulates and sets the OAL source and destination, sets the 3262 NA(NUD) S/T-omIndex to the index of the underlying interface the 3263 NS(NUD) arrived on and sets the Target Address to the same value 3264 included in the NS(NUD). The target next sets the R flag to 1, the S 3265 flag to 1 and the O flag to 1, then selects an in-window 3266 Identification for the ROS and performs fragmentation. The node then 3267 forwards the carrier packets into the unsecured spanning tree, 3268 directly to the ROS if it is in the local segment or directly to a 3269 Bridge in the local segment. 3271 When the ROS receives the NA(NUD), it marks the target underlying 3272 interface tested as "reachable". Note that underlying interface 3273 states are maintained independently of the overall NCE REACHABLE 3274 state, and that a single NCE may have multiple target underlying 3275 interfaces in various states "reachable" and otherwise while the NCE 3276 state as a whole remains REACHABLE. 3278 Note also that the exchange of NS/NA(NUD) messages has the useful 3279 side-benefit of opening holes in NATs that may be useful for NAT 3280 traversal. For example, a Client that discovers the address of a 3281 Bridge on the local SRT segment during an NS/NA(WIN) exchange with a 3282 peer that established MFIB state can send an NS(NUD) message directly 3283 to the INET address of the Bridge while including an authentication 3284 signature. The NS(NUD) will open a hole in any NATs on the path from 3285 the Client to the Bridge, and the Bridge can verify the 3286 authentication signature before returning a direct NA(NUD) to the 3287 Client's NATed L2ADDR while also including an authentication 3288 signature. Future carrier packets exchanged between the Client and 3289 peer can then be forwarded directly via the Bridge while bypassing 3290 the Client's FHS Proxy/Server. 3292 3.16. Mobility Management and Quality of Service (QoS) 3294 AERO is a Distributed Mobility Management (DMM) service. Each Proxy/ 3295 Server is responsible for only a subset of the Clients on the OMNI 3296 link, as opposed to a Centralized Mobility Management (CMM) service 3297 where there is a single network mobility collective entity for all 3298 Clients. Clients coordinate with their associated FHS and Hub Proxy/ 3299 Servers via RS/RA exchanges to maintain the DMM profile, and the AERO 3300 routing system tracks all current Client/Proxy/Server peering 3301 relationships. 3303 Hub Proxy/Servers provide ROR, default routing and mobility anchor 3304 point services for their dependent Clients, while FHS Proxy/Servers 3305 provide a proxy conduit between the Client and the Hub. Clients are 3306 responsible for maintaining neighbor relationships with their Proxy/ 3307 Servers through periodic RS/RA exchanges, which also serves to 3308 confirm neighbor reachability. When a Client's underlying interface 3309 attributes change, the Client is responsible for updating the Hub 3310 Proxy/Server with this new information while using the FHS Proxy/ 3311 Server as a first-hop conduit. The FHS Proxy/Server can also act as 3312 a proxy to perform some IPv6 ND exchanges on the Client's behalf 3313 without consuming bandwidth on the Client underlying interface. 3315 Mobility management considerations are specified in the following 3316 sections. 3318 3.16.1. Mobility Update Messaging 3320 RORs accommodate Client mobility and/or multilink change events by 3321 sending secured uNA messages to each ROS in the target Client's 3322 Report List. When an ROR sends a uNA message, it sets the IPv6 3323 source address to the its own ADM-LLA, sets the destination address 3324 to the ROS LLA (i.e., an MNP-LLA if the ROS is a Client and an ADM- 3325 LLA if the ROS is a Proxy/Server) and sets the Target Address to the 3326 Client's MNP-LLA. The ROR also includes an OMNI option with Preflen 3327 set to the prefix length associated with the Client's MNP-LLA, with 3328 Interface Attributes for the target Client's underlying interfaces 3329 and with the OMNI header S/T-omIndex set to 0. The ROR then sets the 3330 uNA R flag to 1, S flag to 0 and O flag to 1, then encapsulates the 3331 message in an OAL header with source set to its own ADM-ULA and 3332 destination set to the ROS ULA (i.e., the ADM-ULA of the ROS Proxy/ 3333 Server) and sends the message into the secured spanning tree. 3335 As discussed in Section 7.2.6 of [RFC4861], the transmission and 3336 reception of uNA messages is unreliable but provides a useful 3337 optimization. In well-connected Internetworks with robust data links 3338 uNA messages will be delivered with high probability, but in any case 3339 the ROR can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs to 3340 each ROS to increase the likelihood that at least one will be 3341 received. Alternatively, the ROR can set the PNG flag in the uNA 3342 OMNI option header to request a solicited NA acknowledgement as 3343 specified in [I-D.templin-6man-omni]. 3345 When the ROS Proxy/Server receives a uNA message prepared as above, 3346 it ignores the message if the OAL destination is not its own ADM-ULA. 3347 If the uNA destination was its own ADM-LLA, the ROS Proxy/Server uses 3348 the included OMNI option information to update its NCE for the target 3349 but does not reset ReachableTime since the receipt of an unsolicited 3350 NA message from the ROR does not provide confirmation that any 3351 forward paths to the target Client are working. If the destination 3352 was the MNP-LLA of the ROS Client, the Proxy/Server instead re- 3353 encapsulates with the OAL source set to its own ADM-ULA, OAL 3354 destination set to the MNP-ULA of the ROS Client with an 3355 authentication signature if necessary, and with an in-window 3356 Identification for this Client. Finally, if the uNA message PNG flag 3357 was set, the ROS returns a solicited NA acknowledgement as specified 3358 in [I-D.templin-6man-omni]. 3360 In addition to sending uNA messages to the current set of ROSs for 3361 the target Client, the ROR also sends uNAs to the former Proxy/Server 3362 associated with the underlying interface for which the link-layer 3363 address has changed. These uNA messages update former Proxy/Servers 3364 that cannot easily detect (e.g., without active probing) when a 3365 formerly-active Client has departed. When the ROR sends the uNA, it 3366 sets the source address to its ADM-LLA, sets the destination address 3367 to the former Proxy/Server's ADM-LLA, and sets the Target Address to 3368 the Client's MNP-LLA. The ROR also includes an OMNI option with 3369 Preflen set to the prefix length associated with the Client's MNP- 3370 LLA, with Interface Attributes for the changed underlying interface, 3371 and with the OMNI header S/T-omIndex set to 0. The ROR then sets the 3372 uNA R flag to 1, S flag to 0 and O flag to 1, then encapsulates the 3373 message in an OAL header with source set to its own ADM-ULA and 3374 destination set to the ADM-ULA of the former Proxy/Server and sends 3375 the message into the secured spanning tree. 3377 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 3379 When a Client needs to change its underlying Interface Attributes 3380 (e.g., due to a mobility event), the Client sends an RS message to 3381 its Hub Proxy/Server (i.e., the ROR) via a first-hop FHS Proxy/ 3382 Server, if necessary. The RS includes an OMNI option with a 3383 Multilink Forwarding Parameters sub-option with the new link quality 3384 and address information. Note that the first FHS Proxy/Server may 3385 change due to the underlying interface change; any stale state in 3386 former FHS Proxy/Servers will simply expire after ReachableTime 3387 expires with no effect on the Hub Proxy/Server. 3389 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 3390 sending carrier packets containing user data in case one or more RAs 3391 are lost. If all RAs are lost, the Client SHOULD re-associate with a 3392 new Proxy/Server. 3394 When the Proxy/Server receives the Client's changes, it sends uNA 3395 messages to all nodes in the Report List the same as described in the 3396 previous section. 3398 3.16.3. Bringing New Links Into Service 3400 When a Client needs to bring new underlying interfaces into service 3401 (e.g., when it activates a new data link), it sends an RS message to 3402 the Hub Proxy/Server via a FHS Proxy/Server for the underlying 3403 interface (if necessary) with an OMNI option that includes Multilink 3404 Forwarding Parameters with appropriate link quality values and with 3405 link-layer address information for the new link. 3407 3.16.4. Deactivating Existing Links 3409 When a Client needs to deactivate an existing underlying interface, 3410 it sends an RS message to an FHS Proxy/Server with an OMNI option 3411 with appropriate Multilink Forwarding Parameter values for the 3412 deactivated link - in particular, the link quality value 0 assures 3413 that neighbors will cease to use the link. 3415 If the Client needs to send RS messages over an underlying interface 3416 other than the one being deactivated, it MUST include Interface 3417 Attributes with appropriate link quality values for any underlying 3418 interfaces being deactivated. 3420 Note that when a Client deactivates an underlying interface, 3421 neighbors that have received the RS/uNA messages need not purge all 3422 references for the underlying interface from their neighbor cache 3423 entries. The Client may reactivate or reuse the underlying interface 3424 and/or its omIndex at a later point in time, when it will send new RS 3425 messages to an FHS Proxy/Server with fresh interface parameters to 3426 update any neighbors. 3428 3.16.5. Moving Between Proxy/Servers 3430 The Client performs the procedures specified in Section 3.12.2 when 3431 it first associates with a new Hub Proxy/Server or renews its 3432 association with an existing Hub Proxy/Server. 3434 When an FHS Proxy/Server receives the Client's RS message destined to 3435 a new Hub Proxy/Server, it forwards the RS and also sends uNA 3436 messages to inform the old Hub Proxy/Server that the Client has 3437 DEPARTED. The FHS Proxy/Server sets the uNA source to the ADM-LLA of 3438 the new Hub Proxy/Server, sets the destination to the ADM-LLA of the 3439 old Hub Proxy/Server, sets the OAL source to its own ADM-ULA and sets 3440 the OAL destination to the ADM-ULA of the old Hub Proxy/Server. The 3441 FHS Proxy/Server then submits the uNA for OAL encapsulation and 3442 fragmentation, then forwards the resulting carrier packets into the 3443 secured spanning tree. 3445 When the old Hub Proxy/Server receives the uNA, it changes the 3446 Client's NCE state to DEPARTED, sets the interface attributes 3447 information for the Client to point to the new Hub Proxy/Server, and 3448 resets DepartTime. After a short delay (e.g., 2 seconds) the old Hub 3449 Proxy/Server withdraws the Client's MNP from the routing system. 3450 After DepartTime expires, the old Hub Proxy/Server deletes the 3451 Client's NCE. 3453 The old Hub Proxy/Server also iteratively sends uNA messages to each 3454 ROS in the Client's Report List with its own ADM-LLA as the source 3455 and the LLA of the ROS as the destination. The old Proxy/Server then 3456 encapsulates the uNA with OAL source address set to the ADM-ULA of 3457 the new Hub Proxy/Server and OAL destination address set to the ADM- 3458 ULA of the ROS Proxy/Server and sends the carrier packets over the 3459 secured spanning tree. When the ROS Proxy/Server receives the uNA, 3460 it forwards the message to the ROS Client if the destination is an 3461 MNP-LLA. The ROS then examines the uNA Target Address to locate the 3462 target Client's NCE and the ADM-LLA source address to identify the 3463 old Hub Proxy/Server. The ROS then caches the ULA source address as 3464 the ADM-{LLA/ULA} for the new Hub Proxy/Server for this target NCE 3465 and marks the entry as STALE. While in the STALE state, the ROS 3466 sends new NS(AR) messages using its own ULA as the OAL source and the 3467 ADM-ULA of the new Hub Proxy/Server as the OAL destination address. 3468 The new Hub Proxy/Server will then process the NS(AR) and return an 3469 NA(AR) response. 3471 Clients SHOULD NOT move rapidly between Hub Proxy/Servers in order to 3472 avoid causing excessive oscillations in the AERO routing system. 3473 Examples of when a Client might wish to change to a different Hub 3474 Proxy/Server include a Hub Proxy/Server that has gone unreachable, 3475 topological movements of significant distance, movement to a new 3476 geographic region, movement to a new OMNI link segment, etc. 3478 3.17. Multicast 3480 Clients provide an IGMP (IPv4) [RFC2236] or MLD (IPv6) [RFC3810] 3481 proxy service for its EUNs and/or hosted applications [RFC4605] and 3482 act as a Protocol Independent Multicast - Sparse-Mode (PIM-SM, or 3483 simply "PIM") Designated Router (DR) [RFC7761] on the OMNI link. 3484 Proxy/Servers act as OMNI link PIM routers for Clients on ANET, VPNed 3485 or Direct interfaces, and Relays also act as OMNI link PIM routers on 3486 behalf of nodes on other links/networks. 3488 Clients on VPNed, Direct or ANET underlying interfaces for which the 3489 ANET has deployed native multicast services forward IGMP/MLD messages 3490 into the ANET. The IGMP/MLD messages may be further forwarded by a 3491 first-hop ANET access router acting as an IGMP/MLD-snooping switch 3492 [RFC4541], then ultimately delivered to an ANET Proxy/Server. The 3493 Proxy/Server then acts as an ROS to send NS(AR) messages to an ROR. 3494 Clients on INET and ANET underlying interfaces without native 3495 multicast services instead send NS(AR) messages as an ROS to cause 3496 their Proxy/Server forward the message to an ROR. When the ROR 3497 receives an NA(AR) response, it initiates PIM protocol messaging 3498 according to the Source-Specific Multicast (SSM) and Any-Source 3499 Multicast (ASM) operational modes as discussed in the following 3500 sections. 3502 3.17.1. Source-Specific Multicast (SSM) 3504 When an ROS "X" (i.e., either a Client or Proxy Server) acting as PIM 3505 router receives a Join/Prune message from a node on its downstream 3506 interfaces containing one or more ((S)ource, (G)roup) pairs, it 3507 updates its Multicast Routing Information Base (MRIB) accordingly. 3508 For each S belonging to a prefix reachable via X's non-OMNI 3509 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 3510 on those interfaces per [RFC7761]. 3512 For each S belonging to a prefix reachable via X's OMNI interface, X 3513 sends an NS(AR) message (see: Section 3.14) using its own LLA as the 3514 source address, the solicited node multicast address corresponding to 3515 S as the destination and the LLA of S as the target address. X then 3516 encapsulates the NS(AR) in an OAL header with source address set to 3517 its own ULA and destination address set to the ULA for S, then 3518 forwards the message into the secured spanning tree which delivers it 3519 to ROR "Y" that services S. The resulting NA(AR) will return an OMNI 3520 option with Interface Attributes for any underlying interfaces that 3521 are currently servicing S. 3523 When X processes the NA(AR) it selects one or more underlying 3524 interfaces for S and performs an NS/NA(WIN) exchange over the secured 3525 spanning tree while including a PIM Join/Prune message for each 3526 multicast group of interest in the OMNI option. If S is located 3527 behind any Proxys "Z"*, each Z* then updates its MRIB accordingly and 3528 maintains the LLA of X as the next hop in the reverse path. Since 3529 Bridges forward messages not addressed to themselves without 3530 examining them, this means that the (reverse) multicast tree path is 3531 simply from each Z* (and/or S) to X with no other multicast-aware 3532 routers in the path. 3534 Following the initial combined Join/Prune and NS/NA(WIN) messaging, X 3535 maintains a NCE for each S the same as if X was sending unicast data 3536 traffic to S. In particular, X performs additional NS/NA(WIN) 3537 exchanges to keep the NCE alive for up to t_periodic seconds 3539 [RFC7761]. If no new Joins are received within t_periodic seconds, X 3540 allows the NCE to expire. Finally, if X receives any additional 3541 Join/Prune messages for (S,G) it forwards the messages over the 3542 secured spanning tree. 3544 Client C that holds an MNP for source S may later depart from a first 3545 Proxy/Server Z1 and/or connect via a new Proxy/Server Z2. In that 3546 case, Y sends a uNA message to X the same as specified for unicast 3547 mobility in Section 3.16. When X receives the uNA message, it 3548 updates its NCE for the LLA for source S and sends new Join messages 3549 in NS/NA(WIN) exchanges addressed to the new target Client underlying 3550 interface connection for S. There is no requirement to send any 3551 Prune messages to old Proxy/Server Z1 since source S will no longer 3552 source any multicast data traffic via Z1. Instead, the multicast 3553 state for (S,G) in Proxy/Server Z1 will soon expire since no new 3554 Joins will arrive. 3556 3.17.2. Any-Source Multicast (ASM) 3558 When an ROS X acting as a PIM router receives Join/Prune messages 3559 from a node on its downstream interfaces containing one or more (*,G) 3560 pairs, it updates its Multicast Routing Information Base (MRIB) 3561 accordingly. X first performs an NS/NA(AR) exchange to receive route 3562 optimization information for Rendezvous Point (RP) R for each G. X 3563 then includes a copy of each Join/Prune message in the OMNI option of 3564 an NS(WIN) message with its own LLA as the source address and the LLA 3565 for R as the destination address, then encapsulates the NS(WIN) 3566 message in an OAL header with its own ULA as the source and the ADM- 3567 ULA of R's Proxy/Server as the destination then sends the message 3568 into the secured spanning tree. 3570 For each source S that sends multicast traffic to group G via R, 3571 Client S* that aggregates S (or its Proxy/Server) encapsulates the 3572 original IP packets in PIM Register messages, includes the PIM 3573 Register messages in the OMNI options of uNA messages, performs OAL 3574 encapsulation and fragmentation then forwards the resulting carrier 3575 packets with Identification values within the receive window for 3576 Client R* that aggregates R. Client R* may then elect to send a PIM 3577 Join to S* in the OMNI option of a uNA over the secured spanning 3578 tree. This will result in an (S,G) tree rooted at S* with R as the 3579 next hop so that R will begin to receive two copies of the original 3580 IP packet; one native copy from the (S, G) tree and a second copy 3581 from the pre-existing (*, G) tree that still uses uNA PIM Register 3582 encapsulation. R can then issue a uNA PIM Register-stop message over 3583 the secured spanning tree to suppress the Register-encapsulated 3584 stream. At some later time, if Client S* moves to a new Proxy/ 3585 Server, it resumes sending original IP packets via uNA PIM Register 3586 encapsulation via the new Proxy/Server. 3588 At the same time, as multicast listeners discover individual S's for 3589 a given G, they can initiate an (S,G) Join for each S under the same 3590 procedures discussed in Section 3.17.1. Once the (S,G) tree is 3591 established, the listeners can send (S, G) Prune messages to R so 3592 that multicast original IP packets for group G sourced by S will only 3593 be delivered via the (S, G) tree and not from the (*, G) tree rooted 3594 at R. All mobility considerations discussed for SSM apply. 3596 3.17.3. Bi-Directional PIM (BIDIR-PIM) 3598 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 3599 approach to ASM that treats the Rendezvous Point (RP) as a Designated 3600 Forwarder (DF). Further considerations for BIDIR-PIM are out of 3601 scope. 3603 3.18. Operation over Multiple OMNI Links 3605 An AERO Client can connect to multiple OMNI links the same as for any 3606 data link service. In that case, the Client maintains a distinct 3607 OMNI interface for each link, e.g., 'omni0' for the first link, 3608 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 3609 would include its own distinct set of Bridges and Proxy/Servers, 3610 thereby providing redundancy in case of failures. 3612 Each OMNI link could utilize the same or different ANET connections. 3613 The links can be distinguished at the link-layer via the SRT prefix 3614 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 3615 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 3616 MSPs on each link. This gives rise to the opportunity for supporting 3617 multiple redundant networked paths (see: Section 3.2.5). 3619 The Client's IP layer can select the outgoing OMNI interface 3620 appropriate for a given traffic profile while (in the reverse 3621 direction) correspondent nodes must have some way of steering their 3622 original IP packets destined to a target via the correct OMNI link. 3624 In a first alternative, if each OMNI link services different MSPs the 3625 Client can receive a distinct MNP from each of the links. IP routing 3626 will therefore assure that the correct OMNI link is used for both 3627 outbound and inbound traffic. This can be accomplished using 3628 existing technologies and approaches, and without requiring any 3629 special supporting code in correspondent nodes or Bridges. 3631 In a second alternative, if each OMNI link services the same MSP(s) 3632 then each link could assign a distinct "OMNI link Anycast" address 3633 that is configured by all Bridges on the link. Correspondent nodes 3634 can then perform Segment Routing to select the correct SRT, which 3635 will then direct the original IP packet over multiple hops to the 3636 target. 3638 3.19. DNS Considerations 3640 AERO Client MNs and INET correspondent nodes consult the Domain Name 3641 System (DNS) the same as for any Internetworking node. When 3642 correspondent nodes and Client MNs use different IP protocol versions 3643 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 3644 A records for IPv4 address mappings to MNs which must then be 3645 populated in Relay NAT64 mapping caches. In that way, an IPv4 3646 correspondent node can send original IPv4 packets to the IPv4 address 3647 mapping of the target MN, and the Relay will translate the IPv4 3648 header and destination address into an IPv6 header and IPv6 3649 destination address of the MN. 3651 When an AERO Client registers with an AERO Proxy/Server, the Proxy/ 3652 Server can return the address(es) of DNS servers in RDNSS options 3653 [RFC6106]. The DNS server provides the IP addresses of other MNs and 3654 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 3656 3.20. Transition/Coexistence Considerations 3658 OAL encapsulation ensures that dissimilar INET partitions can be 3659 joined into a single unified OMNI link, even though the partitions 3660 themselves may have differing protocol versions and/or incompatible 3661 addressing plans. However, a commonality can be achieved by 3662 incrementally distributing globally routable (i.e., native) IP 3663 prefixes to eventually reach all nodes (both mobile and fixed) in all 3664 OMNI link segments. This can be accomplished by incrementally 3665 deploying AERO Bridges on each INET partition, with each Bridge 3666 distributing its MNPs and/or discovering non-MNP IP GUA prefixes on 3667 its INET links. 3669 This gives rise to the opportunity to eventually distribute native IP 3670 addresses to all nodes, and to present a unified OMNI link view even 3671 if the INET partitions remain in their current protocol and 3672 addressing plans. In that way, the OMNI link can serve the dual 3673 purpose of providing a mobility/multilink service and a transition/ 3674 coexistence service. Or, if an INET partition is transitioned to a 3675 native IP protocol version and addressing scheme that is compatible 3676 with the OMNI link MNP-based addressing scheme, the partition and 3677 OMNI link can be joined by Bridges. 3679 Relays that connect INETs/EUNs with dissimilar IP protocol versions 3680 may need to employ a network address and protocol translation 3681 function such as NAT64 [RFC6146]. 3683 3.21. Detecting and Reacting to Proxy/Server and Bridge Failures 3685 In environments where rapid failure recovery is required, Proxy/ 3686 Servers and Bridges SHOULD use Bidirectional Forwarding Detection 3687 (BFD) [RFC5880]. Nodes that use BFD can quickly detect and react to 3688 failures so that cached information is re-established through 3689 alternate nodes. BFD control messaging is carried only over well- 3690 connected ground domain networks (i.e., and not low-end radio links) 3691 and can therefore be tuned for rapid response. 3693 Proxy/Servers and Bridges maintain BFD sessions in parallel with 3694 their BGP peerings. If a Proxy/Server or Bridge fails, BGP peers 3695 will quickly re-establish routes through alternate paths the same as 3696 for common BGP deployments. Similarly, Proxys maintain BFD sessions 3697 with their associated Bridges even though they do not establish BGP 3698 peerings with them. 3700 3.22. AERO Clients on the Open Internet 3702 AERO Clients that connect to the open Internet via INET interfaces 3703 can establish a VPN or direct link to securely connect to a FHS/Hub 3704 Proxy/Server in a "tethered" arrangement with all of the Client's 3705 traffic transiting the Proxy/Server which acts as a router. 3706 Alternatively, the Client can associate with an INET FHS/Hub Proxy/ 3707 Server using UDP/IP encapsulation and control message securing 3708 services as discussed in the following sections. 3710 When a Client's OMNI interface enables an INET underlying interface, 3711 it first examines the INET address. For IPv4, the Client assumes it 3712 is on the open Internet if the INET address is not a special-use IPv4 3713 address per [RFC3330]. Similarly for IPv6, the Client assumes it is 3714 on the open Internet if the INET address is a Global Unicast Address 3715 (GUA) [RFC4291]. Otherwise, the Client should assume it is behind 3716 one or several NATs. 3718 The Client then prepares an RS message with IPv6 source address set 3719 to its MNP-LLA, with IPv6 destination set to (link-local) All-Routers 3720 multicast and with an OMNI option with underlying interface 3721 attributes. If the Client believes that it is on the open Internet, 3722 it SHOULD include its IP address and UDP port number in the Multilink 3723 Forwarding Parameters sub-option corresponding to the underlying 3724 interface. If the underlying address is IPv4, the Client includes 3725 the Port Number and IPv4 address written in obfuscated form [RFC4380] 3726 as discussed in Section 3.3. If the underlying interface address is 3727 IPv6, the Client instead includes the Port Number and IPv6 address in 3728 obfuscated form. The Client finally includes an authentication 3729 signature per [I-D.templin-6man-omni] to provide message 3730 authentication, selects an Identification value and window 3731 synchronization parameters, and submits the RS for OAL encapsulation. 3732 The Client then encapsulates the OAL atomic fragment in UDP/IP 3733 headers to form a carrier packet, sets the UDP/IP source to its INET 3734 address and UDP port, sets the UDP/IP destination to the FHS Proxy/ 3735 Server's INET address and the AERO service port number (8060), then 3736 sends the carrier packet to the Proxy/Server. 3738 When the FHS Proxy/Server receives the RS, it discards the OAL 3739 encapsulation, authenticates the RS message, and examines the 3740 destination address. If the destination is the ADM-LLA of another 3741 Proxy/Server, the FHS Proxy/Server assumes the proxy role and 3742 forwards the message into the secured spanning tree. If the 3743 destination is All-Routers multicast or its own ADM-LLA, the FHS 3744 Proxy/Server instead assumes the Hub role, creates a NCE and 3745 registers the Client's MNP, window synchronization state and INET 3746 interface information according to the OMNI option parameters. If 3747 the Multilink Forwarding Paramters sub-option includes a non-zero 3748 L2ADDR, the Hub Proxy/Server compares the encapsulation IP address 3749 and UDP port number with the (unobfuscated) values. If the values 3750 are the same, the Hub Proxy/Server caches the Client's information as 3751 an "INET" address meaning that the Client is likely to accept direct 3752 messages without requiring NAT traversal exchanges. If the values 3753 are different (or, if the OMNI option did not include an L2ADDR) the 3754 Hub Proxy/Server instead caches the Client's information as a 3755 "mapped" address meaning that NAT traversal exchanges may be 3756 necessary. 3758 The Hub Proxy/Server then prepares an RA message with IPv6 source and 3759 destination set corresponding to the addresses in the RS, and with an 3760 OMNI option with an Origin Indication sub-option per 3761 [I-D.templin-6man-omni] with the mapped and obfuscated Port Number 3762 and IP address observed in the encapsulation headers. The Proxy/ 3763 Server also includes a Multilink Forwarding Parameters sub-option, an 3764 authentication signature sub-option per [I-D.templin-6man-omni] and/ 3765 or a symmetric window synchronization/acknowledgement if necessary. 3766 The Hub Proxy/Server then performs OAL encapsulation then 3767 encapsulates the carrier packet in UDP/IP headers with addresses set 3768 per the L2ADDR information in the NCE for the Client. 3770 When the Client receives the RA, it authenticates the message then 3771 process the window synchronization/acknowledgement and compares the 3772 mapped Port Number and IP address from the Multilink Forwarding 3773 Parameters sub-option with its own address. If the addresses are the 3774 same, the Client assumes the open Internet / Cone NAT principle; if 3775 the addresses are different, the Client instead assumes that further 3776 qualification procedures are necessary to detect the type of NAT and 3777 performs NAT traversal on-demand according to standard procedures 3778 [RFC6081][RFC4380]. The Client also caches the RA rest of the 3779 Multilink Forwarding Parameters information to discover the FHS 3780 Proxy/Server's local spanning tree segment. The Client finally 3781 arranges to return an explicit/implicit acknowledgement, and sends 3782 periodic RS messages to receive fresh RA messages before the Router 3783 Lifetime received on each INET interface expires. 3785 When the Client sends messages to target IP addresses, it also 3786 invokes route optimization per Section 3.14. For route optimized 3787 targets in the same OMNI link segment, if the target's L2ADDR is on 3788 the open INET, the Client forwards carrier packets directly to the 3789 target INET address. If the target is behind a NAT, the Client first 3790 establishes NAT state for the L2ADDR using the "direct bubble" and 3791 NS/NA(NUD) mechanisms discussed in Section 3.10.1. The Client 3792 continues to send carrier packets via the local Bridge discovered 3793 during window synchronization until NAT state is populated, then 3794 begins forwarding carrier packets via the direct path through the NAT 3795 to the target. For targets in different OMNI link segments, the 3796 Client forwards carrier packets to the local Bridge. 3798 The Client can send original IP packets to route-optimized neighbors 3799 in the same OMNI link segment no larger than the minimum/path MPS in 3800 one piece and with OAL encapsulation as atomic fragments. For larger 3801 original IP packets, the Client applies OAL encapsulation then 3802 fragments if necessary according to Section 3.9, with OAL header with 3803 source set to its own MNP-ULA and destination set to the MNP-ULA of 3804 the target, and with an in-window Identification value. The Client 3805 then encapsulates each resulting carrier packet in UDP/IP *NET 3806 headers and sends them to the neighbor. 3808 INET Clients exchange NS/NA(WIN) messages to associate with a new 3809 peer as discussed in Section 3.2.7. The exchange establishes MFIB 3810 state in the Client, peer and all OMNI intermediate nodes in the 3811 path. After MFIB state is established, INET Clients and peers can 3812 exchange carrier packets with compressed headers that include an MFVI 3813 which is updated on a hop-by-hop basis, while employing "shortcuts" 3814 to skip any unnecessary hops. 3816 Note: The NAT traversal procedures specified in this document are 3817 applicable for Cone, Address-Restricted and Port-Restricted NATs 3818 only. While future updates to this document may specify procedures 3819 for other NAT variations (e.g., hairpinning and various forms of 3820 Symmetric NATs), it should be noted that continuous communications 3821 are always possible through Proxy/Server forwarding even for these 3822 other NAT variations. 3824 3.23. Time-Varying MNPs 3826 In some use cases, it is desirable, beneficial and efficient for the 3827 Client to receive a constant MNP that travels with the Client 3828 wherever it moves. For example, this would allow air traffic 3829 controllers to easily track aircraft, etc. In other cases, however 3830 (e.g., intelligent transportation systems), the MN may be willing to 3831 sacrifice a modicum of efficiency in order to have time-varying MNPs 3832 that can be changed every so often to defeat adversarial tracking. 3834 The DHCPv6 service offers a way for Clients that desire time-varying 3835 MNPs to obtain short-lived prefixes (e.g., on the order of a small 3836 number of minutes). In that case, the identity of the Client would 3837 not be bound to the MNP but rather to a Node Identification value 3838 (see: [I-D.templin-6man-omni]) to be used as the Client ID seed for 3839 MNP prefix delegation. The Client would then be obligated to 3840 renumber its internal networks whenever its MNP (and therefore also 3841 its MNP-LLA) changes. This should not present a challenge for 3842 Clients with automated network renumbering services, however presents 3843 limits for the durations of ongoing sessions that would prefer to use 3844 a constant address. 3846 4. Implementation Status 3848 An early AERO implementation based on OpenVPN (https://openvpn.net/) 3849 was announced on the v6ops mailing list on January 10, 2018 and an 3850 initial public release of the AERO proof-of-concept source code was 3851 announced on the intarea mailing list on August 21, 2015. 3853 AERO Release-3.2 was tagged on March 30, 2021, and is undergoing 3854 internal testing. Additional internal releases expected within the 3855 coming months, with first public release expected end of 1H2021. 3857 Many AERO/OMNI functions are implemented and undergoing final 3858 integration. OAL fragmentation/reassembly buffer management code has 3859 been cleared for public release and will be presented at the June 3860 2021 ICAO mobility subgroup meeting. 3862 5. IANA Considerations 3864 The IANA has assigned the UDP port number "8060" for an earlier 3865 experimental first version of AERO [RFC6706]. This document together 3866 with [I-D.templin-6man-omni] reclaims UDP port number "8060" as the 3867 service port for UDP/IP encapsulation. This document makes no 3868 request of IANA, since [I-D.templin-6man-omni] already provides 3869 instructions. (Note: although [RFC6706] was not widely implemented 3870 or deployed, it need not be obsoleted since its messages use the 3871 invalid ICMPv6 message type number '0' which implementations of this 3872 specification can easily distinguish and ignore.) 3874 No further IANA actions are required. 3876 6. Security Considerations 3878 AERO Bridges configure secured tunnels with AERO Proxy/Servers and 3879 Relays within their local OMNI link segments. Applicable secured 3880 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3881 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3882 segments in turn configure secured tunnels for their neighboring AERO 3883 Bridges in a secured spanning tree topology. Therefore, control 3884 messages exchanged between any pair of OMNI link neighbors over the 3885 secured spanning tree are already protected. 3887 To prevent spoofing vectors, Proxy/Servers MUST discard without 3888 responding to any unsecured NS(AR) messages. Also, Proxy/Servers 3889 MUST discard without forwarding any original IP packets received from 3890 one of their own Clients (whether directly or following OAL 3891 reassembly) with a source address that does not match the Client's 3892 MNP and/or a destination address that does match the Client's MNP. 3893 Finally, Proxy/Servers MUST discard without forwarding any carrier 3894 packets with an OAL source and destination that both match the same 3895 MNP. 3897 For INET partitions that require strong security in the data plane, 3898 two options for securing communications include 1) disable route 3899 optimization so that all traffic is conveyed over secured tunnels, or 3900 2) enable on-demand secure tunnel creation between Client neighbors. 3901 Option 1) would result in longer routes than necessary and impose 3902 traffic concentration on critical infrastructure elements. Option 2) 3903 could be coordinated between Clients using NS/NA messages with OMNI 3904 Host Identity Protocol (HIP) "Initiator/Responder" message sub- 3905 options [RFC7401][I-D.templin-6man-omni] to create a secured tunnel 3906 on-demand. 3908 AERO Clients that connect to secured ANETs need not apply security to 3909 their IPv6 ND messages, since the messages will be authenticated and 3910 forwarded by a perimeter Proxy/Server that applies security on its 3911 INET-facing interface as part of the spanning tree (see above). AERO 3912 Clients connected to the open INET can use network and/or transport 3913 layer security services such as VPNs or can by some other means 3914 establish a direct link to a Proxy/Server. When a VPN or direct link 3915 may be impractical, however, INET Clients and Proxy/Servers SHOULD 3916 include and verify authentication signatures for their IPv6 ND 3917 messages as specified in [I-D.templin-6man-omni]. 3919 Application endpoints SHOULD use transport-layer (or higher-layer) 3920 security services such as TLS/SSL, DTLS or SSH [RFC4251] to assure 3921 the same level of protection as for critical secured Internet 3922 services. AERO Clients that require host-based VPN services SHOULD 3923 use network and/or transport layer security services such as IPsec, 3924 TLS/SSL, DTLS, etc. AERO Proxys and Proxy/Servers can also provide a 3925 network-based VPN service on behalf of the Client, e.g., if the 3926 Client is located within a secured enclave and cannot establish a VPN 3927 on its own behalf. 3929 AERO Proxy/Servers and Bridges present targets for traffic 3930 amplification Denial of Service (DoS) attacks. This concern is no 3931 different than for widely-deployed VPN security gateways in the 3932 Internet, where attackers could send spoofed packets to the gateways 3933 at high data rates. This can be mitigated through the AERO/OMNI data 3934 origin authentication procedures, as well as connecting Proxy/Servers 3935 and Bridges over dedicated links with no connections to the Internet 3936 and/or when connections to the Internet are only permitted through 3937 well-managed firewalls. Traffic amplification DoS attacks can also 3938 target an AERO Client's low data rate links. This is a concern not 3939 only for Clients located on the open Internet but also for Clients in 3940 secured enclaves. AERO Proxy/Servers and Proxys can institute rate 3941 limits that protect Clients from receiving packet floods that could 3942 DoS low data rate links. 3944 AERO Relays must implement ingress filtering to avoid a spoofing 3945 attack in which spurious messages with ULA addresses are injected 3946 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3947 that their connectivity is not used by unauthorized nodes on their 3948 EUNs to gain access to a protected network, i.e., AERO Clients that 3949 act as routers MUST NOT provide routing services for unauthorized 3950 nodes. (This concern is no different than for ordinary hosts that 3951 receive an IP address delegation but then "share" the address with 3952 other nodes via some form of Internet connection sharing such as 3953 tethering.) 3955 The PRL MUST be well-managed and secured from unauthorized tampering, 3956 even though the list contains only public information. The PRL can 3957 be conveyed to the Client in a similar fashion as in [RFC5214] (e.g., 3958 through layer 2 data link login messaging, secure upload of a static 3959 file, DNS lookups, etc.). 3961 The AERO service for open INET Clients depends on a public key 3962 distribution service in which Client public keys and identities are 3963 maintained in a shared database accessible to all open INET Proxy/ 3964 Servers. Similarly, each Client must be able to determine the public 3965 key of each Proxy/Server, e.g. by consulting an online database. 3966 When AERO nodes register their public keys indexed by a unique Host 3967 Identity Tag (HIT) [RFC7401] in a distributed database such as the 3968 DNS, and use the HIT as an identity for applying IPv6 ND message 3969 authentication signatures, a means for determining public key 3970 attestation is available. 3972 Security considerations for IPv6 fragmentation and reassembly are 3973 discussed in [I-D.templin-6man-omni]. In environments where spoofing 3974 is considered a threat, OMNI nodes SHOULD employ Identification 3975 window synchronization and OAL destinations SHOULD configure an (end- 3976 system-based) firewall. 3978 SRH authentication facilities are specified in [RFC8754]. Security 3979 considerations for accepting link-layer ICMP messages and reflected 3980 packets are discussed throughout the document. 3982 7. Acknowledgements 3984 Discussions in the IETF, aviation standards communities and private 3985 exchanges helped shape some of the concepts in this work. 3986 Individuals who contributed insights include Mikael Abrahamsson, Mark 3987 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Scott Burleigh, 3988 Brian Carpenter, Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian 3989 Farrel, Nick Green, Sri Gundavelli, Brian Haberman, Bernhard Haindl, 3990 Joel Halpern, Tom Herbert, Bob Hinden, Sascha Hlusiak, Lee Howard, 3991 Christian Huitema, Zdenek Jaron, Andre Kostur, Hubert Kuenig, Ted 3992 Lemon, Andy Malis, Satoru Matsushima, Tomek Mrugalski, Thomas Narten, 3993 Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal Skorepa, 3994 Dave Thaler, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, 3995 Lloyd Wood and James Woodyatt. Members of the IESG also provided 3996 valuable input during their review process that greatly improved the 3997 document. Special thanks go to Stewart Bryant, Joel Halpern and 3998 Brian Haberman for their shepherding guidance during the publication 3999 of the AERO first edition. 4001 This work has further been encouraged and supported by Boeing 4002 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 4003 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 4004 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 4005 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 4006 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 4007 Gene MacLean III, Kyle Mikos, Rob Muszkiewicz, Sean O'Sullivan, Vijay 4008 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 4009 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 4010 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 4011 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 4012 Benson, Madhuri Madhava Badgandi, Vijayasarathy Rajagopalan, Katie 4013 Tran and Eric Yeh are especially acknowledged for their work on the 4014 AERO implementation. Chuck Klabunde is honored and remembered for 4015 his early leadership, and we mourn his untimely loss. 4017 This work was inspired by the support and encouragement of countless 4018 outstanding colleagues, managers and program directors over the span 4019 of many decades. Beginning in the late 1980s,' the Digital Equipment 4020 Corporation (DEC) Ultrix Engineering and DECnet Architects groups 4021 identified early issues with fragmentation and bridging links with 4022 diverse MTUs. In the early 1990s, engagements at DEC Project Sequoia 4023 at UC Berkeley and the DEC Western Research Lab in Palo Alto included 4024 investigations into large-scale networked filesystems, ATM vs 4025 Internet and network security proxies. In the mid-1990s to early 4026 2000s employment at the NASA Ames Research Center (Sterling Software) 4027 and SRI International supported early investigations of IPv6, ONR UAV 4028 Communications and the IETF. An employment at Nokia where important 4029 IETF documents were published gave way to a present-day engagement 4030 with The Boeing Company. The work matured at Boeing through major 4031 programs including Future Combat Systems, Advanced Airplane Program, 4032 DTN for the International Space Station, Mobility Vision Lab, CAST, 4033 Caravan, Airplane Internet of Things, the NASA UAS/CNS program, the 4034 FAA/ICAO ATN/IPS program and many others. An attempt to name all who 4035 gave support and encouragement would double the current document size 4036 and result in many unintentional omissions - but to all a humble 4037 thanks. 4039 Earlier works on NBMA tunneling approaches are found in 4040 [RFC2529][RFC5214][RFC5569]. 4042 Many of the constructs presented in this second edition of AERO are 4043 based on the author's earlier works, including: 4045 o The Internet Routing Overlay Network (IRON) 4046 [RFC6179][I-D.templin-ironbis] 4048 o Virtual Enterprise Traversal (VET) 4049 [RFC5558][I-D.templin-intarea-vet] 4051 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 4052 [RFC5320][I-D.templin-intarea-seal] 4054 o AERO, First Edition [RFC6706] 4056 Note that these works cite numerous earlier efforts that are not also 4057 cited here due to space limitations. The authors of those earlier 4058 works are acknowledged for their insights. 4060 This work is aligned with the NASA Safe Autonomous Systems Operation 4061 (SASO) program under NASA contract number NNA16BD84C. 4063 This work is aligned with the FAA as per the SE2025 contract number 4064 DTFAWA-15-D-00030. 4066 This work is aligned with the Boeing Commercial Airplanes (BCA) 4067 Internet of Things (IoT) and autonomy programs. 4069 This work is aligned with the Boeing Information Technology (BIT) 4070 MobileNet program. 4072 8. References 4074 8.1. Normative References 4076 [I-D.templin-6man-omni] 4077 Templin, F. L. and T. Whyman, "Transmission of IP Packets 4078 over Overlay Multilink Network (OMNI) Interfaces", draft- 4079 templin-6man-omni-03 (work in progress), April 2021. 4081 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 4082 DOI 10.17487/RFC0791, September 1981, 4083 . 4085 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 4086 RFC 792, DOI 10.17487/RFC0792, September 1981, 4087 . 4089 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 4090 Requirement Levels", BCP 14, RFC 2119, 4091 DOI 10.17487/RFC2119, March 1997, 4092 . 4094 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 4095 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 4096 December 1998, . 4098 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 4099 "SEcure Neighbor Discovery (SEND)", RFC 3971, 4100 DOI 10.17487/RFC3971, March 2005, 4101 . 4103 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 4104 RFC 3972, DOI 10.17487/RFC3972, March 2005, 4105 . 4107 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 4108 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 4109 November 2005, . 4111 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 4112 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 4113 . 4115 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 4116 Network Address Translations (NATs)", RFC 4380, 4117 DOI 10.17487/RFC4380, February 2006, 4118 . 4120 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 4121 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 4122 DOI 10.17487/RFC4861, September 2007, 4123 . 4125 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 4126 Address Autoconfiguration", RFC 4862, 4127 DOI 10.17487/RFC4862, September 2007, 4128 . 4130 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 4131 DOI 10.17487/RFC6081, January 2011, 4132 . 4134 [RFC7401] Moskowitz, R., Ed., Heer, T., Jokela, P., and T. 4135 Henderson, "Host Identity Protocol Version 2 (HIPv2)", 4136 RFC 7401, DOI 10.17487/RFC7401, April 2015, 4137 . 4139 [RFC7739] Gont, F., "Security Implications of Predictable Fragment 4140 Identification Values", RFC 7739, DOI 10.17487/RFC7739, 4141 February 2016, . 4143 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 4144 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 4145 May 2017, . 4147 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 4148 (IPv6) Specification", STD 86, RFC 8200, 4149 DOI 10.17487/RFC8200, July 2017, 4150 . 4152 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 4153 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 4154 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 4155 RFC 8415, DOI 10.17487/RFC8415, November 2018, 4156 . 4158 8.2. Informative References 4160 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 4161 2016. 4163 [I-D.bonica-6man-comp-rtg-hdr] 4164 Bonica, R., Kamite, Y., Alston, A., Henriques, D., and L. 4165 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 4166 bonica-6man-comp-rtg-hdr-24 (work in progress), January 4167 2021. 4169 [I-D.bonica-6man-crh-helper-opt] 4170 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 4171 Routing Header (CRH) Helper Option", draft-bonica-6man- 4172 crh-helper-opt-03 (work in progress), April 2021. 4174 [I-D.ietf-intarea-frag-fragile] 4175 Bonica, R., Baker, F., Huston, G., Hinden, R. M., Troan, 4176 O., and F. Gont, "IP Fragmentation Considered Fragile", 4177 draft-ietf-intarea-frag-fragile-17 (work in progress), 4178 September 2019. 4180 [I-D.ietf-intarea-tunnels] 4181 Touch, J. and M. Townsley, "IP Tunnels in the Internet 4182 Architecture", draft-ietf-intarea-tunnels-10 (work in 4183 progress), September 2019. 4185 [I-D.ietf-ipwave-vehicular-networking] 4186 (editor), J. (. J., "IPv6 Wireless Access in Vehicular 4187 Environments (IPWAVE): Problem Statement and Use Cases", 4188 draft-ietf-ipwave-vehicular-networking-20 (work in 4189 progress), March 2021. 4191 [I-D.ietf-rtgwg-atn-bgp] 4192 Templin, F. L., Saccone, G., Dawra, G., Lindem, A., and V. 4193 Moreno, "A Simple BGP-based Mobile Routing System for the 4194 Aeronautical Telecommunications Network", draft-ietf- 4195 rtgwg-atn-bgp-10 (work in progress), January 2021. 4197 [I-D.templin-6man-dhcpv6-ndopt] 4198 Templin, F. L., "A Unified Stateful/Stateless 4199 Configuration Service for IPv6", draft-templin-6man- 4200 dhcpv6-ndopt-11 (work in progress), January 2021. 4202 [I-D.templin-intarea-seal] 4203 Templin, F. L., "The Subnetwork Encapsulation and 4204 Adaptation Layer (SEAL)", draft-templin-intarea-seal-68 4205 (work in progress), January 2014. 4207 [I-D.templin-intarea-vet] 4208 Templin, F. L., "Virtual Enterprise Traversal (VET)", 4209 draft-templin-intarea-vet-40 (work in progress), May 2013. 4211 [I-D.templin-ipwave-uam-its] 4212 Templin, F. L., "Urban Air Mobility Implications for 4213 Intelligent Transportation Systems", draft-templin-ipwave- 4214 uam-its-04 (work in progress), January 2021. 4216 [I-D.templin-ironbis] 4217 Templin, F. L., "The Interior Routing Overlay Network 4218 (IRON)", draft-templin-ironbis-16 (work in progress), 4219 March 2014. 4221 [I-D.templin-v6ops-pdhost] 4222 Templin, F. L., "IPv6 Prefix Delegation and Multi- 4223 Addressing Models", draft-templin-v6ops-pdhost-27 (work in 4224 progress), January 2021. 4226 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 4228 [RFC1035] Mockapetris, P., "Domain names - implementation and 4229 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 4230 November 1987, . 4232 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 4233 RFC 1812, DOI 10.17487/RFC1812, June 1995, 4234 . 4236 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 4237 DOI 10.17487/RFC2003, October 1996, 4238 . 4240 [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, 4241 DOI 10.17487/RFC2004, October 1996, 4242 . 4244 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 4245 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 4246 . 4248 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 4249 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 4250 . 4252 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 4253 Domains without Explicit Tunnels", RFC 2529, 4254 DOI 10.17487/RFC2529, March 1999, 4255 . 4257 [RFC2983] Black, D., "Differentiated Services and Tunnels", 4258 RFC 2983, DOI 10.17487/RFC2983, October 2000, 4259 . 4261 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 4262 of Explicit Congestion Notification (ECN) to IP", 4263 RFC 3168, DOI 10.17487/RFC3168, September 2001, 4264 . 4266 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 4267 DOI 10.17487/RFC3330, September 2002, 4268 . 4270 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 4271 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 4272 DOI 10.17487/RFC3810, June 2004, 4273 . 4275 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 4276 Unique IDentifier (UUID) URN Namespace", RFC 4122, 4277 DOI 10.17487/RFC4122, July 2005, 4278 . 4280 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 4281 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 4282 January 2006, . 4284 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 4285 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 4286 DOI 10.17487/RFC4271, January 2006, 4287 . 4289 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 4290 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 4291 2006, . 4293 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 4294 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 4295 December 2005, . 4297 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 4298 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 4299 2006, . 4301 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 4302 Control Message Protocol (ICMPv6) for the Internet 4303 Protocol Version 6 (IPv6) Specification", STD 89, 4304 RFC 4443, DOI 10.17487/RFC4443, March 2006, 4305 . 4307 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 4308 Protocol (LDAP): The Protocol", RFC 4511, 4309 DOI 10.17487/RFC4511, June 2006, 4310 . 4312 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 4313 "Considerations for Internet Group Management Protocol 4314 (IGMP) and Multicast Listener Discovery (MLD) Snooping 4315 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 4316 . 4318 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 4319 "Internet Group Management Protocol (IGMP) / Multicast 4320 Listener Discovery (MLD)-Based Multicast Forwarding 4321 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 4322 August 2006, . 4324 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 4325 Algorithms in Cryptographically Generated Addresses 4326 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 4327 . 4329 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 4330 "Bidirectional Protocol Independent Multicast (BIDIR- 4331 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 4332 . 4334 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 4335 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 4336 DOI 10.17487/RFC5214, March 2008, 4337 . 4339 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 4340 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 4341 February 2010, . 4343 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 4344 Route Optimization Requirements for Operational Use in 4345 Aeronautics and Space Exploration Mobile Networks", 4346 RFC 5522, DOI 10.17487/RFC5522, October 2009, 4347 . 4349 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 4350 RFC 5558, DOI 10.17487/RFC5558, February 2010, 4351 . 4353 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 4354 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 4355 January 2010, . 4357 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 4358 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 4359 . 4361 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 4362 "IPv6 Router Advertisement Options for DNS Configuration", 4363 RFC 6106, DOI 10.17487/RFC6106, November 2010, 4364 . 4366 [RFC6139] Russert, S., Ed., Fleischman, E., Ed., and F. Templin, 4367 Ed., "Routing and Addressing in Networks with Global 4368 Enterprise Recursion (RANGER) Scenarios", RFC 6139, 4369 DOI 10.17487/RFC6139, February 2011, 4370 . 4372 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 4373 NAT64: Network Address and Protocol Translation from IPv6 4374 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 4375 April 2011, . 4377 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 4378 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 4379 . 4381 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 4382 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 4383 DOI 10.17487/RFC6221, May 2011, 4384 . 4386 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 4387 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 4388 DOI 10.17487/RFC6273, June 2011, 4389 . 4391 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 4392 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 4393 January 2012, . 4395 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 4396 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 4397 DOI 10.17487/RFC6355, August 2011, 4398 . 4400 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 4401 for Equal Cost Multipath Routing and Link Aggregation in 4402 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 4403 . 4405 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 4406 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 4407 . 4409 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 4410 UDP Checksums for Tunneled Packets", RFC 6935, 4411 DOI 10.17487/RFC6935, April 2013, 4412 . 4414 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 4415 for the Use of IPv6 UDP Datagrams with Zero Checksums", 4416 RFC 6936, DOI 10.17487/RFC6936, April 2013, 4417 . 4419 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 4420 Korhonen, "Requirements for Distributed Mobility 4421 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 4422 . 4424 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 4425 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 4426 Multicast - Sparse Mode (PIM-SM): Protocol Specification 4427 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 4428 2016, . 4430 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 4431 Decraene, B., Litkowski, S., and R. Shakir, "Segment 4432 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 4433 July 2018, . 4435 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 4436 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 4437 . 4439 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 4440 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 4441 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 4442 . 4444 [WG] Wireguard, "WireGuard, https://www.wireguard.com", August 4445 2020. 4447 Appendix A. Non-Normative Considerations 4449 AERO can be applied to a multitude of Internetworking scenarios, with 4450 each having its own adaptations. The following considerations are 4451 provided as non-normative guidance: 4453 A.1. Implementation Strategies for Route Optimization 4455 Route optimization as discussed in Section 3.14 results in the route 4456 optimization source (ROS) creating a NCE for the target neighbor. 4457 The NCE state is set to REACHABLE for at most ReachableTime seconds. 4458 In order to refresh the NCE lifetime before the ReachableTime timer 4459 expires, the specification requires implementations to issue a new 4460 NS/NA(AR) exchange to reset ReachableTime while data packets are 4461 still flowing. However, the decision of when to initiate a new NS/ 4462 NA(AR) exchange and to perpetuate the process is left as an 4463 implementation detail. 4465 One possible strategy may be to monitor the NCE watching for data 4466 packets for (ReachableTime - 5) seconds. If any data packets have 4467 been sent to the neighbor within this timeframe, then send an NS(AR) 4468 to receive a new NA(AR). If no data packets have been sent, wait for 4469 5 additional seconds and send an immediate NS(AR) if any data packets 4470 are sent within this "expiration pending" 5 second window. If no 4471 additional data packets are sent within the 5 second window, reset 4472 the NCE state to STALE. 4474 The monitoring of the neighbor data packet traffic therefore becomes 4475 an ongoing process during the NCE lifetime. If the NCE expires, 4476 future data packets will trigger a new NS/NA(AR) exchange while the 4477 packets themselves are delivered over a longer path until route 4478 optimization state is re-established. 4480 A.2. Implicit Mobility Management 4482 OMNI interface neighbors MAY provide a configuration option that 4483 allows them to perform implicit mobility management in which no IPv6 4484 ND messaging is used. In that case, the Client only transmits 4485 packets over a single interface at a time, and the neighbor always 4486 observes packets arriving from the Client from the same link-layer 4487 source address. 4489 If the Client's underlying interface address changes (either due to a 4490 readdressing of the original interface or switching to a new 4491 interface) the neighbor immediately updates the NCE for the Client 4492 and begins accepting and sending packets according to the Client's 4493 new address. This implicit mobility method applies to use cases such 4494 as cellphones with both WiFi and Cellular interfaces where only one 4495 of the interfaces is active at a given time, and the Client 4496 automatically switches over to the backup interface if the primary 4497 interface fails. 4499 A.3. Direct Underlying Interfaces 4501 When a Client's OMNI interface is configured over a Direct interface, 4502 the neighbor at the other end of the Direct link can receive packets 4503 without any encapsulation. In that case, the Client sends packets 4504 over the Direct link according to traffic selectors. If the Direct 4505 interface is selected, then the Client's IP packets are transmitted 4506 directly to the peer without going through an ANET/INET. If other 4507 interfaces are selected, then the Client's IP packets are transmitted 4508 via a different interface, which may result in the inclusion of 4509 Proxy/Servers and Bridges in the communications path. Direct 4510 interfaces must be tested periodically for reachability, e.g., via 4511 NUD. 4513 A.4. AERO Critical Infrastructure Considerations 4515 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 4516 IP routers or virtual machines in the cloud. Bridges must be 4517 provisioned, supported and managed by the INET administrative 4518 authority, and connected to the Bridges of other INETs via inter- 4519 domain peerings. Cost for purchasing, configuring and managing 4520 Bridges is nominal even for very large OMNI links. 4522 AERO INET Proxy/Servers can be standard dedicated server platforms, 4523 but most often will be deployed as virtual machines in the cloud. 4524 The only requirements for INET Proxy/Servers are that they can run 4525 the AERO/OMNI code and have at least one network interface connection 4526 to the INET. INET Proxy/Servers must be provisioned, supported and 4527 managed by the INET administrative authority. Cost for purchasing, 4528 configuring and managing cloud Proxy/Servers is nominal especially 4529 for virtual machines. 4531 AERO ANET Proxy/Servers are most often standard dedicated server 4532 platforms with one underlying interface connected to the ANET and a 4533 second interface connected to an INET. As with INET Proxy/Servers, 4534 the only requirements are that they can run the AERO/OMNI code and 4535 have at least one interface connection to the INET. ANET Proxy/ 4536 Servers must be provisioned, supported and managed by the ANET 4537 administrative authority. Cost for purchasing, configuring and 4538 managing Proxys is nominal, and borne by the ANET administrative 4539 authority. 4541 AERO Relays are simply Proxy/Servers connected to INETs and/or EUNs 4542 that provide forwarding services for non-MNP destinations. The Relay 4543 connects to the OMNI link and engages in eBGP peering with one or 4544 more Bridges as a stub AS. The Relay then injects its MNPs and/or 4545 non-MNP prefixes into the BGP routing system, and provisions the 4546 prefixes to its downstream-attached networks. The Relay can perform 4547 ROS/ROR services the same as for any Proxy/Server, and can route 4548 between the MNP and non-MNP address spaces. 4550 A.5. AERO Server Failure Implications 4552 AERO Proxy/Servers may appear as a single point of failure in the 4553 architecture, but such is not the case since all Proxy/Servers on the 4554 link provide identical services and loss of a Proxy/Server does not 4555 imply immediate and/or comprehensive communication failures. Proxy/ 4556 Server failure is quickly detected and conveyed by Bidirectional 4557 Forward Detection (BFD) and/or proactive NUD allowing Clients to 4558 migrate to new Proxy/Servers. 4560 If a Proxy/Server fails, ongoing packet forwarding to Clients will 4561 continue by virtue of the neighbor cache entries that have already 4562 been established in route optimization sources (ROSs). If a Client 4563 also experiences mobility events at roughly the same time the Proxy/ 4564 Server fails, uNA messages may be lost but neighbor cache entries in 4565 the DEPARTED state will ensure that packet forwarding to the Client's 4566 new locations will continue for up to DepartTime seconds. 4568 If a Client is left without a Proxy/Server for a considerable length 4569 of time (e.g., greater than ReachableTime seconds) then existing 4570 neighbor cache entries will eventually expire and both ongoing and 4571 new communications will fail. The original source will continue to 4572 retransmit until the Client has established a new Proxy/Server 4573 relationship, after which time continuous communications will resume. 4575 Therefore, providing many Proxy/Servers on the link with high 4576 availability profiles provides resilience against loss of individual 4577 Proxy/Servers and assurance that Clients can establish new Proxy/ 4578 Server relationships quickly in event of a Proxy/Server failure. 4580 A.6. AERO Client / Server Architecture 4582 The AERO architectural model is client / server in the control plane, 4583 with route optimization in the data plane. The same as for common 4584 Internet services, the AERO Client discovers the addresses of AERO 4585 Proxy/Servers and connects to one or more of them. The AERO service 4586 is analogous to common Internet services such as google.com, 4587 yahoo.com, cnn.com, etc. However, there is only one AERO service for 4588 the link and all Proxy/Servers provide identical services. 4590 Common Internet services provide differing strategies for advertising 4591 server addresses to clients. The strategy is conveyed through the 4592 DNS resource records returned in response to name resolution queries. 4593 As of January 2020 Internet-based 'nslookup' services were used to 4594 determine the following: 4596 o When a client resolves the domainname "google.com", the DNS always 4597 returns one A record (i.e., an IPv4 address) and one AAAA record 4598 (i.e., an IPv6 address). The client receives the same addresses 4599 each time it resolves the domainname via the same DNS resolver, 4600 but may receive different addresses when it resolves the 4601 domainname via different DNS resolvers. But, in each case, 4602 exactly one A and one AAAA record are returned. 4604 o When a client resolves the domainname "ietf.org", the DNS always 4605 returns one A record and one AAAA record with the same addresses 4606 regardless of which DNS resolver is used. 4608 o When a client resolves the domainname "yahoo.com", the DNS always 4609 returns a list of 4 A records and 4 AAAA records. Each time the 4610 client resolves the domainname via the same DNS resolver, the same 4611 list of addresses are returned but in randomized order (i.e., 4612 consistent with a DNS round-robin strategy). But, interestingly, 4613 the same addresses are returned (albeit in randomized order) when 4614 the domainname is resolved via different DNS resolvers. 4616 o When a client resolves the domainname "amazon.com", the DNS always 4617 returns a list of 3 A records and no AAAA records. As with 4618 "yahoo.com", the same three A records are returned from any 4619 worldwide Internet connection point in randomized order. 4621 The above example strategies show differing approaches to Internet 4622 resilience and service distribution offered by major Internet 4623 services. The Google approach exposes only a single IPv4 and a 4624 single IPv6 address to clients. Clients can then select whichever IP 4625 protocol version offers the best response, but will always use the 4626 same IP address according to the current Internet connection point. 4627 This means that the IP address offered by the network must lead to a 4628 highly-available server and/or service distribution point. In other 4629 words, resilience is predicated on high availability within the 4630 network and with no client-initiated failovers expected (i.e., it is 4631 all-or-nothing from the client's perspective). However, Google does 4632 provide for worldwide distributed service distribution by virtue of 4633 the fact that each Internet connection point responds with a 4634 different IPv6 and IPv4 address. The IETF approach is like google 4635 (all-or-nothing from the client's perspective), but provides only a 4636 single IPv4 or IPv6 address on a worldwide basis. This means that 4637 the addresses must be made highly-available at the network level with 4638 no client failover possibility, and if there is any worldwide service 4639 distribution it would need to be conducted by a network element that 4640 is reached via the IP address acting as a service distribution point. 4642 In contrast to the Google and IETF philosophies, Yahoo and Amazon 4643 both provide clients with a (short) list of IP addresses with Yahoo 4644 providing both IP protocol versions and Amazon as IPv4-only. The 4645 order of the list is randomized with each name service query 4646 response, with the effect of round-robin load balancing for service 4647 distribution. With a short list of addresses, there is still 4648 expectation that the network will implement high availability for 4649 each address but in case any single address fails the client can 4650 switch over to using a different address. The balance then becomes 4651 one of function in the network vs function in the end system. 4653 The same implications observed for common highly-available services 4654 in the Internet apply also to the AERO client/server architecture. 4655 When an AERO Client connects to one or more ANETs, it discovers one 4656 or more AERO Proxy/Server addresses through the mechanisms discussed 4657 in earlier sections. Each Proxy/Server address presumably leads to a 4658 fault-tolerant clustering arrangement such as supported by Linux-HA, 4659 Extended Virtual Synchrony or Paxos. Such an arrangement has 4660 precedence in common Internet service deployments in lightweight 4661 virtual machines without requiring expensive hardware deployment. 4662 Similarly, common Internet service deployments set service IP 4663 addresses on service distribution points that may relay requests to 4664 many different servers. 4666 For AERO, the expectation is that a combination of the Google/IETF 4667 and Yahoo/Amazon philosophies would be employed. The AERO Client 4668 connects to different ANET access points and can receive 1-2 Proxy/ 4669 Server ADM-LLAs at each point. It then selects one AERO Proxy/Server 4670 address, and engages in RS/RA exchanges with the same Proxy/Server 4671 from all ANET connections. The Client remains with this Proxy/Server 4672 unless or until the Proxy/Server fails, in which case it can switch 4673 over to an alternate Proxy/Server. The Client can likewise switch 4674 over to a different Proxy/Server at any time if there is some reason 4675 for it to do so. So, the AERO expectation is for a balance of 4676 function in the network and end system, with fault tolerance and 4677 resilience at both levels. 4679 Appendix B. Change Log 4681 << RFC Editor - remove prior to publication >> 4683 Changes from draft-templin-6man-aero-20 to draft-templin-6man-aero- 4684 21: 4686 o Major updates to Hub-and-Spokes Proxy/Server coordination. 4688 Changes from draft-templin-6man-aero-19 to draft-templin-6man-aero- 4689 20: 4691 o Major updates especially in Section 3.2.7. 4693 Changes from draft-templin-6man-aero-18 to draft-templin-6man-aero- 4694 19: 4696 o Major revision update for review. 4698 Changes from draft-templin-6man-aero-17 to draft-templin-6man-aero- 4699 18: 4701 o Interim version with extensive new text - cleanup planned for next 4702 release. 4704 Changes from draft-templin-6man-aero-16 to draft-templin-6man-aero- 4705 17: 4707 o Final editorial review pass resulting in multiple changes. 4708 Document now submit for final approval (with reference to rfcdiff 4709 from previous version). 4711 Changes from draft-templin-6man-aero-15 to draft-templin-6man-aero- 4712 16: 4714 o Final editorial review pass resulting in multiple changes. 4715 Document now submit for final approval (with reference to rfcdiff 4716 from previous version). 4718 Changes from draft-templin-6man-aero-14 to draft-templin-6man-aero- 4719 15: 4721 o Final editorial review pass resulting in multiple changes. 4722 Document now submit for final approval (with reference to rfcdiff 4723 from previous version). 4725 Changes from draft-templin-6man-aero-13 to draft-templin-6man-aero- 4726 14: 4728 o Final editorial review pass resulting in multiple changes. 4729 Document now submit for final approval (with reference to rfcdiff 4730 from previous version). 4732 Changes from draft-templin-6man-aero-12 to draft-templin-6man-aero- 4733 13: 4735 o Final editorial review pass resulting in multiple changes. 4736 Document now submit for final approval (with reference to rfcdiff 4737 from previous version). 4739 Changes from draft-templin-6man-aero-11 to draft-templin-6man-aero- 4740 12: 4742 o Final editorial review pass resulting in multiple changes. 4743 Document now submit for final approval (with reference to rfcdiff 4744 from previous version). 4746 Changes from draft-templin-6man-aero-10 to draft-templin-6man-aero- 4747 11: 4749 o Final editorial review pass resulting in multiple changes. 4750 Document now submit for final approval (with reference to rfcdiff 4751 from previous version). 4753 Changes from draft-templin-6man-aero-09 to draft-templin-6man-aero- 4754 10: 4756 o Final editorial review pass resulting in multiple changes. 4757 Document now submit for final approval (with reference to rfcdiff 4758 from previous version). 4760 Changes from draft-templin-6man-aero-08 to draft-templin-6man-aero- 4761 09: 4763 o Final editorial review pass resulting in multiple changes. 4764 Document now submit for final approval (with reference to rfcdiff 4765 from previous version). 4767 Changes from draft-templin-6man-aero-07 to draft-templin-6man-aero- 4768 08: 4770 o Final editorial review pass resulting in multiple changes. 4771 Document now submit for final approval (with reference to rfcdiff 4772 from previous version). 4774 Changes from draft-templin-6man-aero-06 to draft-templin-6man-aero- 4775 07: 4777 o Final editorial review pass resulting in multiple changes. 4778 Document now submit for final approval (with reference to rfcdiff 4779 from previous version). 4781 Changes from draft-templin-6man-aero-05 to draft-templin-6man-aero- 4782 06: 4784 o Final editorial review pass resulting in multiple changes. 4785 Document now submit for final approval. 4787 Changes from draft-templin-6man-aero-04 to draft-templin-6man-aero- 4788 05: 4790 o Changed to use traffic selectors instead of the former multilink 4791 selection strategy. 4793 Changes from draft-templin-6man-aero-03 to draft-templin-6man-aero- 4794 04: 4796 o Removed documents from "Obsoletes" list. 4798 o Introduced the concept of "secured" and "unsecured" spanning tree. 4800 o Additional security considerations. 4802 o Additional route optimization considerations. 4804 Changes from draft-templin-6man-aero-02 to draft-templin-6man-aero- 4805 03: 4807 o Support for extended route optimization from ROR to target over 4808 target's underlying interfaces. 4810 Changes from draft-templin-6man-aero-01 to draft-templin-6man-aero- 4811 02: 4813 o Changed reference citations to "draft-templin-6man-omni". 4815 o Several important updates to IPv6 ND cache states and route 4816 optimization message addressing. 4818 o Included introductory description of the "6M's". 4820 o Updated Multicast specification. 4822 Changes from draft-templin-6man-aero-00 to draft-templin-6man-aero- 4823 01: 4825 o Changed category to "Informational". 4827 o Updated implementation status. 4829 Changes from earlier versions to draft-templin-6man-aero-00: 4831 o Established working baseline reference. 4833 Author's Address 4835 Fred L. Templin (editor) 4836 Boeing Research & Technology 4837 P.O. Box 3707 4838 Seattle, WA 98124 4839 USA 4841 Email: fltemplin@acm.org