idnits 2.17.1 draft-templin-6man-aero-36.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 (October 25, 2021) is 913 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Missing Reference: 'ULA' is mentioned on line 947, but not defined == Missing Reference: 'ULA0' is mentioned on line 947, but not defined == Missing Reference: 'ULA1' is mentioned on line 947, but not defined == Missing Reference: 'ULA2' is mentioned on line 947, but not defined == Unused Reference: 'RFC3971' is defined on line 4068, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 4073, but no explicit reference was found in the text == Unused Reference: 'RFC4191' is defined on line 4077, but no explicit reference was found in the text == Unused Reference: 'RFC4193' is defined on line 4081, but no explicit reference was found in the text == Unused Reference: 'RFC7739' is defined on line 4109, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 4130, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-crh-helper-opt' is defined on line 4138, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 4143, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 4166, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-v6ops-pdhost' is defined on line 4190, but no explicit reference was found in the text == Unused Reference: 'OVPN' is defined on line 4195, but no explicit reference was found in the text == Unused Reference: 'RFC2003' is defined on line 4205, but no explicit reference was found in the text == Unused Reference: 'RFC2004' is defined on line 4209, but no explicit reference was found in the text == Unused Reference: 'RFC2983' is defined on line 4226, but no explicit reference was found in the text == Unused Reference: 'RFC3168' is defined on line 4230, but no explicit reference was found in the text == Unused Reference: 'RFC3330' is defined on line 4235, but no explicit reference was found in the text == Unused Reference: 'RFC4122' is defined on line 4244, but no explicit reference was found in the text == Unused Reference: 'RFC4982' is defined on line 4293, but no explicit reference was found in the text == Unused Reference: 'RFC6139' is defined on line 4335, but no explicit reference was found in the text == Unused Reference: 'RFC6273' is defined on line 4355, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 4364, but no explicit reference was found in the text == Unused Reference: 'RFC6438' is defined on line 4369, but no explicit reference was found in the text == Unused Reference: 'RFC6935' is defined on line 4378, but no explicit reference was found in the text == Unused Reference: 'RFC6936' is defined on line 4383, but no explicit reference was found in the text == Outdated reference: A later version (-74) exists of draft-templin-6man-omni-48 ** Downref: Normative reference to an Informational draft: draft-templin-6man-omni (ref. 'I-D.templin-6man-omni') ** Downref: Normative reference to an Informational RFC: RFC 7739 == Outdated reference: A later version (-31) exists of draft-bonica-6man-comp-rtg-hdr-26 == 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-24 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-11 -- 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: 2 errors (**), 0 flaws (~~), 34 warnings (==), 4 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: Standards Track October 25, 2021 5 Expires: April 28, 2022 7 Automatic Extended Route Optimization (AERO) 8 draft-templin-6man-aero-36 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 path selection and route optimization are naturally supported 20 through dynamic neighbor cache updates. AERO is a widely-applicable 21 mobile internetworking service especially well-suited to aviation 22 services, intelligent transportation systems, mobile end user devices 23 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 April 28, 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. Segment Routing Topologies (SRTs) . . . . . . . . . . 20 68 3.2.5. Segment Routing For OMNI Link Selection . . . . . . . 21 69 3.3. OMNI Interface Characteristics . . . . . . . . . . . . . 21 70 3.4. OMNI Interface Initialization . . . . . . . . . . . . . . 23 71 3.4.1. AERO Proxy/Server and Relay Behavior . . . . . . . . 24 72 3.4.2. AERO Client Behavior . . . . . . . . . . . . . . . . 24 73 3.4.3. AERO Bridge Behavior . . . . . . . . . . . . . . . . 24 74 3.5. OMNI Interface Neighbor Cache Maintenance . . . . . . . . 25 75 3.5.1. OMNI ND Messages . . . . . . . . . . . . . . . . . . 27 76 3.5.2. OMNI Neighbor Advertisement Message Flags . . . . . . 28 77 3.5.3. OMNI Neighbor Window Synchronization . . . . . . . . 29 78 3.6. OMNI Interface Encapsulation and Fragmentation . . . . . 29 79 3.7. OMNI Interface Decapsulation . . . . . . . . . . . . . . 31 80 3.8. OMNI Interface Data Origin Authentication . . . . . . . . 32 81 3.9. OMNI Interface MTU . . . . . . . . . . . . . . . . . . . 32 82 3.10. OMNI Interface Forwarding Algorithm . . . . . . . . . . . 33 83 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 34 84 3.10.2. Proxy/Server and Relay Forwarding Algorithm . . . . 35 85 3.10.3. Bridge Forwarding Algorithm . . . . . . . . . . . . 37 86 3.11. OMNI Interface Error Handling . . . . . . . . . . . . . . 38 87 3.12. AERO Mobility Service Coordination . . . . . . . . . . . 41 88 3.12.1. AERO Service Model . . . . . . . . . . . . . . . . . 42 89 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 43 90 3.12.3. AERO Proxy/Server Behavior . . . . . . . . . . . . . 44 91 3.13. AERO Route Optimization . . . . . . . . . . . . . . . . . 50 92 3.13.1. Multilink Address Resolution . . . . . . . . . . . . 51 93 3.13.2. Multilink Route Optimization . . . . . . . . . . . . 55 94 3.13.3. Rapid Commit Route Optimization . . . . . . . . . . 67 95 3.13.4. Client/Bridge Route Optimization . . . . . . . . . . 67 96 3.13.5. Client/Client Route Optimization . . . . . . . . . . 69 98 3.14. Neighbor Unreachability Detection (NUD) . . . . . . . . . 71 99 3.15. Mobility Management and Quality of Service (QoS) . . . . 72 100 3.15.1. Mobility Update Messaging . . . . . . . . . . . . . 73 101 3.15.2. Announcing Link-Layer Information Changes . . . . . 74 102 3.15.3. Bringing New Links Into Service . . . . . . . . . . 74 103 3.15.4. Deactivating Existing Links . . . . . . . . . . . . 74 104 3.15.5. Moving Between Proxy/Servers . . . . . . . . . . . . 75 105 3.16. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 76 106 3.16.1. Source-Specific Multicast (SSM) . . . . . . . . . . 76 107 3.16.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 77 108 3.16.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 78 109 3.17. Operation over Multiple OMNI Links . . . . . . . . . . . 78 110 3.18. DNS Considerations . . . . . . . . . . . . . . . . . . . 79 111 3.19. Transition/Coexistence Considerations . . . . . . . . . . 80 112 3.20. Proxy/Server-Bridge Bidirectional Forwarding Detection . 80 113 3.21. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 81 114 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 81 115 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 81 116 6. Security Considerations . . . . . . . . . . . . . . . . . . . 82 117 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 84 118 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 86 119 8.1. Normative References . . . . . . . . . . . . . . . . . . 86 120 8.2. Informative References . . . . . . . . . . . . . . . . . 87 121 Appendix A. Non-Normative Considerations . . . . . . . . . . . . 94 122 A.1. Implementation Strategies for Route Optimization . . . . 94 123 A.2. Implicit Mobility Management . . . . . . . . . . . . . . 95 124 A.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 95 125 A.4. AERO Critical Infrastructure Considerations . . . . . . . 95 126 A.5. AERO Server Failure Implications . . . . . . . . . . . . 96 127 A.6. AERO Client / Server Architecture . . . . . . . . . . . . 97 128 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 99 129 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 99 131 1. Introduction 133 Automatic Extended Route Optimization (AERO) fulfills the 134 requirements of Distributed Mobility Management (DMM) [RFC7333] and 135 route optimization [RFC5522] for aeronautical networking and other 136 network mobility use cases including intelligent transportation 137 systems and enterprise mobile device users. AERO is a secure 138 internetworking and mobility management service that employs the 139 Overlay Multilink Network Interface (OMNI) [I-D.templin-6man-omni] 140 Non-Broadcast, Multiple Access (NBMA) virtual link model. The OMNI 141 link is a virtual overlay configured over one or more underlying 142 Internetworks, and nodes on the link can exchange original IP packets 143 as single-hop neighbors. The OMNI Adaptation Layer (OAL) supports 144 multilink operation for increased reliability and path optimization 145 while providing fragmentation and reassembly services to support 146 Maximum Transmission Unit (MTU) diversity. In terms of precedence, 147 this specification may provide first-principle insights into a 148 representative mobility service architecture as context for 149 understanding the OMNI specification. 151 The AERO service connects Clients over Proxy/Servers and Relays that 152 are seen as OMNI link neighbors, and includes Bridges that 153 interconnect diverse Internetworks as OMNI link segments through OAL 154 forwarding at a layer below IP. Each node's OMNI interface uses an 155 IPv6 link-local address format that supports operation of the IPv6 156 Neighbor Discovery (IPv6 ND) protocol [RFC4861]. A node's OMNI 157 interface can be configured over multiple underlying interfaces, and 158 therefore appears as a single interface with multiple link-layer 159 addresses. Each link-layer address is subject to change due to 160 mobility and/or multilink fluctuations, and link-layer address 161 changes are signaled by ND messaging the same as for any IPv6 link. 163 AERO provides a secure cloud-based service where mobile node Clients 164 may use Proxy/Servers acting as proxys and/or designated routers 165 while fixed nodes may use any Relay on the link for efficient 166 communications. Fixed nodes forward original IP packets destined to 167 other AERO nodes via the nearest Relay, which forwards them through 168 the cloud. Mobile node Clients discover shortest paths to OMNI link 169 neighbors through AERO route optimization. Both unicast and 170 multicast communications are supported, and Clients may efficiently 171 move between locations while maintaining continuous communications 172 with correspondents and without changing their IP Address. 174 AERO Bridges peer with Proxy/Servers in a secured private BGP overlay 175 routing instance to establish a Segment Routing Topology (SRT) 176 spanning tree over the underlying Internetworks of one or more 177 disjoint administrative domains as a single unified OMNI link. Each 178 OMNI link instance is characterized by the set of Mobility Service 179 Prefixes (MSPs) common to all mobile nodes. Relays provide an 180 optimal route from (fixed) correspondent nodes on the underlying 181 Internetwork to (mobile or fixed) nodes on the OMNI link. To the 182 underlying Internetwork, the Relay is the source of a route to the 183 MSP; hence uplink traffic to mobile nodes is naturally routed to the 184 nearest Relay. 186 AERO can be used with OMNI links that span private-use Internetworks 187 and/or public Internetworks such as the global Internet. In both 188 cases, Clients may be located behind Network Address Translators 189 (NATs) on the path to their associated Proxy/Servers. A means for 190 robust traversal of NATs while avoiding "triangle routing" and 191 critical infrastructure traffic concentration is therefore provided. 193 AERO assumes the use of PIM Sparse Mode in support of multicast 194 communication. In support of Source Specific Multicast (SSM) when a 195 Mobile Node is the source, AERO route optimization ensures that a 196 shortest-path multicast tree is established with provisions for 197 mobility and multilink operation. In all other multicast scenarios 198 there are no AERO dependencies. 200 AERO provides a secure aeronautical internetworking service for both 201 manned and unmanned aircraft, where the aircraft is treated as a 202 mobile node that can connect an Internet of Things (IoT). AERO is 203 also applicable to a wide variety of other use cases. For example, 204 it can be used to coordinate the links of mobile nodes (e.g., 205 cellphones, tablets, laptop computers, etc.) that connect into a home 206 enterprise network via public access networks with VPN or non-VPN 207 services enabled according to the appropriate security model. AERO 208 can also be used to facilitate terrestrial vehicular and urban air 209 mobility (as well as pedestrian communication services) for future 210 intelligent transportation systems 211 [I-D.ietf-ipwave-vehicular-networking][I-D.templin-ipwave-uam-its]. 212 Other applicable use cases are also in scope. 214 Along with OMNI, AERO provides secured optimal routing support for 215 the "6M's" of modern Internetworking, including: 217 1. Multilink - a mobile node's ability to coordinate multiple 218 diverse underlying data links as a single logical unit (i.e., the 219 OMNI interface) to achieve the required communications 220 performance and reliability objectives. 222 2. Multinet - the ability to span the OMNI link over a segment 223 routing topology with multiple diverse administrative domain 224 network segments while maintaining seamless end-to-end 225 communications between mobile Clients and correspondents such as 226 air traffic controllers, fleet administrators, etc. 228 3. Mobility - a mobile node's ability to change network points of 229 attachment (e.g., moving between wireless base stations) which 230 may result in an underlying interface address change, but without 231 disruptions to ongoing communication sessions with peers over the 232 OMNI link. 234 4. Multicast - the ability to send a single network transmission 235 that reaches multiple nodes belonging to the same interest group, 236 but without disturbing other nodes not subscribed to the interest 237 group. 239 5. Multihop - a mobile node vehicle-to-vehicle relaying capability 240 useful when multiple forwarding hops between vehicles may be 241 necessary to "reach back" to an infrastructure access point 242 connection to the OMNI link. 244 6. MTU assurance - the ability to deliver packets of various robust 245 sizes between peers without loss due to a link size restriction, 246 and to dynamically adjust packets sizes to achieve the optimal 247 performance for each independent traffic flow. 249 The following numbered sections present the AERO specification. The 250 appendices at the end of the document are non-normative. 252 2. Terminology 254 The terminology in the normative references applies; especially, the 255 terminology in the OMNI specification [I-D.templin-6man-omni] is used 256 extensively throughout. The following terms are defined within the 257 scope of this document: 259 IPv6 Neighbor Discovery (IPv6 ND) 260 a control message service for coordinating neighbor relationships 261 between nodes connected to a common link. AERO uses the IPv6 ND 262 messaging service specified in [RFC4861] while including the OMNI 263 option extensions specified in [I-D.templin-6man-omni]. 265 IPv6 Prefix Delegation 266 a networking service for delegating IPv6 prefixes to nodes on the 267 link. The nominal service is DHCPv6 [RFC8415], however alternate 268 services (e.g., based on IPv6 ND messaging) are also in scope. A 269 minimal form of prefix delegation known as "prefix registration" 270 can be used if the Client knows its prefix in advance and can 271 represent it in the source address of an IPv6 ND message. 273 Access Network (ANET) 274 a node's first-hop data link service network (e.g., a radio access 275 network, cellular service provider network, corporate enterprise 276 network, etc.) that often provides link-layer security services 277 such as IEEE 802.1X and physical-layer security (e.g., "protected 278 spectrum") to prevent unauthorized access internally and with 279 border network-layer security services such as firewalls and 280 proxys that prevent unauthorized outside access. 282 ANET interface 283 a node's attachment to a link in an ANET. 285 Internetwork (INET) 286 a network topology with a coherent IP routing and addressing plan 287 and that provides a transit backbone service for its connected end 288 systems. INETs also provide an underlay service over which the 289 AERO virtual link is configured. Example INETs include corporate 290 enterprise networks, aviation networks, and the public Internet 291 itself. When there is no administrative boundary between an ANET 292 and the INET, the ANET and INET are one and the same. 294 INET interface 295 a node's attachment to a link in an INET. 297 *NET 298 a "wildcard" term referring to either ANET or INET when it is not 299 necessary to draw a distinction between the two. 301 *NET interface 302 a node's attachment to a link in a *NET. 304 *NET Partition 305 frequently, *NETs such as large corporate enterprise networks are 306 sub-divided internally into separate isolated partitions (a 307 technique also known as "network segmentation"). Each partition 308 is fully connected internally but disconnected from other 309 partitions, and there is no requirement that separate partitions 310 maintain consistent Internet Protocol and/or addressing plans. 311 (Each *NET partition is seen as a separate OMNI link segment as 312 discussed throughout this document.) 314 *NET encapsulation 315 the encapsulation of a packet in an outer header or headers that 316 can be routed within the scope of the local *NET partition. 318 *NET address 319 the IP address (and also UDP port number when UDP is used) that 320 appears in *NET encapsulations sent over a node's interface 321 connection to a *NET. 323 INADDR 324 the same as defined for "*NET" address above, with both terms used 325 interchangeably throughout the document. 327 OMNI link 328 the same as defined in [I-D.templin-6man-omni]. The OMNI link 329 employs IPv6 encapsulation [RFC2473] to traverse intermediate 330 nodes in a spanning tree over underlying *NET segments the same as 331 a bridged campus LAN. AERO nodes on the OMNI link appear as 332 single-hop neighbors at the network layer even though they may be 333 separated by many underlying *NET hops; AERO nodes can employ 334 Segment Routing [RFC8402] to navigate between different OMNI 335 links, and/or to cause packets to visit selected waypoints within 336 the same OMNI link. 338 OMNI Interface 339 a node's attachment to an OMNI link. Since OMNI interface 340 addresses are managed for uniqueness, OMNI interfaces do not 341 require Duplicate Address Detection (DAD) and therefore set the 342 administrative variable 'DupAddrDetectTransmits' to zero 343 [RFC4862]. 345 OMNI Adaptation Layer (OAL) 346 an OMNI interface service that subjects original IP packets 347 admitted into the interface to mid-layer IPv6 header encapsulation 348 followed by fragmentation and reassembly. The OAL is also 349 responsible for generating MTU-related control messages as 350 necessary, and for providing addressing context for spanning 351 multiple segments of a bridged OMNI link. 353 original IP packet 354 a whole IP packet or fragment admitted into the OMNI interface by 355 the network layer prior to OAL encapsulation and fragmentation, or 356 an IP packet delivered to the network layer by the OMNI interface 357 following OAL decapsulation and reassembly. 359 OAL packet 360 an original IP packet encapsulated in OAL headers and trailers 361 before OAL fragmentation, or following OAL reassembly. 363 OAL fragment 364 a portion of an OAL packet following fragmentation but prior to 365 *NET encapsulation, or following *NET decapsulation but prior to 366 OAL reassembly. 368 (OAL) atomic fragment 369 an OAL packet that can be forwarded without fragmentation, but 370 still includes a Fragment Header with a valid Identification value 371 and with Fragment Offset and More Fragments both set to 0. 373 (OAL) carrier packet 374 an encapsulated OAL fragment following *NET encapsulation or prior 375 to *NET decapsulation. OAL sources and destinations exchange 376 carrier packets over underlying interfaces, and may be separated 377 by one or more OAL intermediate nodes. OAL intermediate nodes re- 378 encapsulate carrier packets during forwarding by removing the *NET 379 headers of the previous hop underlying network and replacing them 380 with new *NET headers for the next hop underlying network. 382 OAL source 383 an OMNI interface acts as an OAL source when it encapsulates 384 original IP packets to form OAL packets, then performs OAL 385 fragmentation and *NET encapsulation to create carrier packets. 387 OAL destination 388 an OMNI interface acts as an OAL destination when it decapsulates 389 carrier packets, then performs OAL reassembly and decapsulation to 390 derive the original IP packet. 392 OAL intermediate node 393 an OMNI interface acts as an OAL intermediate node when it removes 394 the *NET headers of carrier packets received from a previous hop, 395 then re-encapsulates the carrier packets in new *NET headers and 396 forwards them to the next hop. OAL intermediate nodes decrement 397 the Hop Limit of the OAL IPv6 header during re-encapsulation, and 398 discard the packet if the Hop Limit reaches 0. OAL intermediate 399 nodes do not decrement the Hop Limit/TTL of the original IP 400 packet. 402 underlying interface 403 a *NET interface over which an OMNI interface is configured. 405 Mobility Service Prefix (MSP) 406 an aggregated IP Global Unicast Address (GUA) prefix (e.g., 407 2001:db8::/32, 192.0.2.0/24, etc.) assigned to the OMNI link and 408 from which more-specific Mobile Network Prefixes (MNPs) are 409 delegated. OMNI link administrators typically obtain MSPs from an 410 Internet address registry, however private-use prefixes can 411 alternatively be used subject to certain limitations (see: 412 [I-D.templin-6man-omni]). OMNI links that connect to the global 413 Internet advertise their MSPs to their interdomain routing peers. 415 Mobile Network Prefix (MNP) 416 a longer IP prefix delegated from an MSP (e.g., 417 2001:db8:1000:2000::/56, 192.0.2.8/30, etc.) and delegated to an 418 AERO Client or Relay. 420 Mobile Network Prefix Link Local Address (MNP-LLA) 421 an IPv6 Link Local Address that embeds the most significant 64 422 bits of an MNP in the lower 64 bits of fe80::/64, as specified in 423 [I-D.templin-6man-omni]. 425 Mobile Network Prefix Unique Local Address (MNP-ULA) 426 an IPv6 Unique-Local Address derived from an MNP-LLA. 428 Administrative Link Local Address (ADM-LLA) 429 an IPv6 Link Local Address that embeds a 32-bit administratively- 430 assigned identification value in the lower 32 bits of fe80::/96, 431 as specified in [I-D.templin-6man-omni]. 433 Administrative Unique Local Address (ADM-ULA) 434 an IPv6 Unique-Local Address derived from an ADM-LLA. 436 AERO node 437 a node that is connected to an OMNI link and participates in the 438 AERO internetworking and mobility service. 440 AERO Client ("Client") 441 an AERO node that connects over one or more underlying interfaces 442 and requests MNP delegation/registration service from AERO Proxy/ 443 Servers. The Client assigns an MNP-LLA to the OMNI interface for 444 use in IPv6 ND exchanges with other AERO nodes and forwards 445 original IP packets to correspondents according to OMNI interface 446 neighbor cache state. 448 AERO Proxy/Server ("Proxy/Server") 449 a node that provides a proxying service between AERO Clients and 450 external peers on its Client-facing ANET interfaces (i.e., in the 451 same fashion as for an enterprise network proxy) as well as 452 designated router services for coordination with correspondents on 453 its INET-facing interfaces. (Proxy/Servers in the open INET 454 instead configure only a single INET interface and no ANET 455 interfaces.) The Proxy/Server configures an OMNI interface and 456 assigns an ADM-LLA to support the operation of IPv6 ND services, 457 while advertising any associated MNPs for which it is acting as a 458 hub via BGP peerings with Bridges. 460 AERO Relay ("Relay") 461 a Proxy/Server that provides forwarding services between nodes 462 reached via the OMNI link and correspondents on other links/ 463 networks. AERO Relays configure an OMNI interface, assign an ADM- 464 LLA and maintain BGP peerings with Bridges the same as Proxy/ 465 Servers and run a dynamic routing protocol to discover any non-MNP 466 IP GUA routes in service on other links/networks. The Relay 467 advertises the MSP(s) to its other links/networks, and 468 redistributes routes discovered on other links/networks into the 469 OMNI link BGP routing system the same as for Proxy/Servers. 470 (Relays that connect to major Internetworks such as the global 471 IPv6 or IPv4 Internet can also be configured to advertise 472 "default" routes into the OMNI link BGP routing system.) 474 AERO Bridge ("Bridge") 475 a BGP hub autonomous system node that also provides OAL forwarding 476 services for nodes on an OMNI link. Bridges forwards carrier 477 packets between OMNI link segments as OAL intermediate nodes while 478 decrementing the OAL IPv6 header Hop Limit but without 479 decrementing the network layer IP TTL/Hop Limit. Bridges peer 480 with Proxy/Servers and other Bridges to form a spanning tree over 481 all OMNI link segments and to discover the set of all MNP and non- 482 MNP prefixes in service. Bridges process carrier packets received 483 over the secured spanning tree that are addressed to themselves, 484 while forwarding all other carrier packets to the next hop also 485 via the secured spanning tree. Bridges forward carrier packets 486 received over the unsecured spanning tree to the next hop either 487 via the unsecured spanning tree or via direct encapsulation if the 488 next hop is on the same OMNI link segment. 490 First-Hop Segment (FHS) Proxy/Server 491 a Proxy/Server for a source Client's underlying interface that 492 forwards the Client's packets into the segment routing topology. 493 FHS Proxy/Servers also act as intermediate forwarding nodes to 494 facilitate RS/RA exchanges between a Client and its Hub Proxy/ 495 Server. 497 Hub Proxy/Server 498 a single Proxy/Server selected by a Client that provides a 499 designated router service for all of the Client's underlying 500 interfaces. Clients often select the first FHS Proxy/Server they 501 coordinate with to serve in the Hub role (as all FHS Proxy/Servers 502 are equally capable candidates to serve in that capacity), however 503 the Client can also select any available Proxy/Server for the OMNI 504 link (as there is no requirement that the Hub must also be one of 505 the Client's FHS Proxy/Servers). 507 Last-Hop Segment (LHS) Proxy/Server 508 a Proxy/Server for an underlying interface of the target Client 509 that forwards packets received from the segment routing topology 510 to the target Client over that interface. 512 Segment Routing Topology (SRT) 513 a multinet OMNI link forwarding region between FHS and LHS Proxy/ 514 Servers. FHS/LHS Proxy/Servers and SRT Bridges span the OMNI link 515 on behalf of source/target Client pairs. The SRT maintains a 516 spanning tree established through BGP peerings between Bridges and 517 Proxy/Servers. Each SRT segment includes Bridges in a "hub" and 518 Proxy/Servers in "spokes", while adjacent segments are 519 interconnected by Bridge-Bridge peerings. The BGP peerings are 520 configured over both secured and unsecured underlying network 521 paths such that a secured spanning tree is available for critical 522 control messages while other messages can use the unsecured 523 spanning tree. 525 link-layer address 526 an IP address used as an encapsulation header source or 527 destination address from the perspective of the OMNI interface. 528 When an upper layer protocol (e.g., UDP) is used as part of the 529 encapsulation, the port number is also considered as part of the 530 link-layer address. 532 network layer address 533 the source or destination address of an original IP packet 534 presented to the OMNI interface. 536 end user network (EUN) 537 an internal virtual or external edge IP network that an AERO 538 Client or Relay connects to the rest of the network via the OMNI 539 interface. The Client/Relay sees each EUN as a "downstream" 540 network, and sees the OMNI interface as the point of attachment to 541 the "upstream" network. 543 Mobile Node (MN) 544 an AERO Client and all of its downstream-attached networks that 545 move together as a single unit, i.e., an end system that connects 546 an Internet of Things. 548 Mobile Router (MR) 549 a MN's on-board router that forwards original IP packets between 550 any downstream-attached networks and the OMNI link. The MR is the 551 MN entity that hosts the AERO Client. 553 Route Optimization Source (ROS) 554 the AERO node nearest the source that initiates route 555 optimization. The ROS may be a FHS Proxy/Server or Relay for the 556 source, or may be the source Client itself. 558 Route Optimization responder (ROR) 559 the AERO node that responds to route optimization requests on 560 behalf of the target. The ROR may be either the target MNP Client 561 itself, the Client's current Hub Proxy/Server or a Relay for a 562 non-MNP target. 564 Potential Router List (PRL) 565 a geographically and/or topologically referenced list of addresses 566 of all Proxy/Servers within the same OMNI link. Each OMNI link 567 has its own PRL. 569 Distributed Mobility Management (DMM) 570 a BGP-based overlay routing service coordinated by Proxy/Servers 571 and Bridges that tracks all Proxy/Server-to-Client associations. 573 Mobility Service (MS) 574 the collective set of all Proxy/Servers, Bridges and Relays that 575 provide the AERO Service to Clients. 577 Multilink Forwarding Information Base (MFIB) 578 A forwarding table on each AERO/OMNI source, destination and 579 intermediate node that includes Multilink Forwarding Vectors (MFV) 580 with both next hop forwarding instructions and context for 581 reconstructing compressed headers for specific underlying 582 interface pairs used to communicate with peers. 584 Multilink Forwarding Vector (MFV) 585 An MFIB entry that includes soft state for each underlying 586 interface pairwise communication session between peer OMNI nodes. 587 MFVs are identified by both a next-hop and previous-hop MFV Index 588 (MFVI), with the next-hop established based on an IPv6 ND 589 solicitation and the previous hop established based on the 590 solicited IPv6 ND advertisement response. 592 Multilink Forwarding Vector Index (MVFI) 593 A 4 octet value selected by an AERO/OMNI node when it creates an 594 MFV, then advertises to either a next-hop or previous-hop. AERO/ 595 OMNI intermediate nodes assign two distinct local MFVIs for each 596 MFV and advertise one to the next-hop and the other to the 597 previous-hop. AERO/OMNI end systems assign and advertise a single 598 MFVI. AERO/OMNI nodes also discover the remote MFVIs advertised 599 by other nodes that indicate a value the remote node is willing to 600 accept. 602 Throughout the document, the simple terms "Client", "Proxy/Server", 603 "Bridge" and "Relay" refer to "AERO Client", "AERO Proxy/Server", 604 "AERO Bridge" and "AERO Relay", respectively. Capitalization is used 605 to distinguish these terms from other common Internetworking uses in 606 which they appear without capitalization. 608 The terminology of IPv6 ND [RFC4861] and DHCPv6 [RFC8415] (including 609 the names of node variables, messages and protocol constants) is used 610 throughout this document. The terms "All-Routers multicast", "All- 611 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 612 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 613 generically refer to either Internet Protocol version, i.e., IPv4 614 [RFC0791] or IPv6 [RFC8200]. 616 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 617 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 618 "OPTIONAL" in this document are to be interpreted as described in BCP 619 14 [RFC2119][RFC8174] when, and only when, they appear in all 620 capitals, as shown here. 622 3. Automatic Extended Route Optimization (AERO) 624 The following sections specify the operation of IP over OMNI links 625 using the AERO service: 627 3.1. AERO Node Types 629 AERO Clients can be deployed as fixed infrastructure nodes close to 630 end systems, or as Mobile Nodes (MNs) that can change their network 631 attachment points dynamically. AERO Clients configure OMNI 632 interfaces over underlying interfaces with addresses that may change 633 due to mobility. AERO Clients register their Mobile Network Prefixes 634 (MNPs) with the AERO service, and distribute the MNPs to nodes on 635 EUNs. AERO Bridges, Proxy/Servers and Relays are critical 636 infrastructure elements in fixed (i.e., non-mobile) INET deployments 637 and hence have permanent and unchanging INET addresses. Together, 638 they constitute the AERO service which provides an OMNI link virtual 639 overlay for connecting AERO Clients. 641 AERO Bridges (together with Proxy/Servers) provide the secured 642 backbone supporting infrastructure for a Segment Routing Topology 643 (SRT) spanning tree for the OMNI link. Bridges forward carrier 644 packets both within the same SRT segment and between disjoint SRT 645 segments based on an IPv6 encapsulation mid-layer known as the OMNI 646 Adaptation Layer (OAL) [I-D.templin-6man-omni]. The OMNI interface 647 and OAL provide a virtual bridging service, since the inner IP TTL/ 648 Hop Limit is not decremented. Each Bridge also peers with Proxy/ 649 Servers and other Bridges in a dynamic routing protocol instance to 650 provide a Distributed Mobility Management (DMM) service for the list 651 of active MNPs (see Section 3.2.3). Bridges assign one or more 652 Mobility Service Prefixes (MSPs) to the OMNI link and configure 653 secured tunnels with Proxy/Servers, Relays and other Bridges; they 654 further maintain forwarding table entries for each MNP or non-MNP 655 prefix in service on the OMNI link. 657 AERO Proxy/Servers distributed across one or more SRT segments 658 provide default forwarding and mobility/multilink services for AERO 659 Client mobile nodes. Each Proxy/Server also peers with Bridges in a 660 dynamic routing protocol instance to advertise its list of associated 661 MNPs (see Section 3.2.3). Hub Proxy/Servers provide prefix 662 delegation/registration services and track the mobility/multilink 663 profiles of each of their associated Clients, where each delegated 664 prefix becomes an MNP taken from an MSP. Proxy/Servers at ANET/INET 665 boundaries provide a forwarding service for ANET Clients to 666 communicate with peers in external INETs, while Proxy/Servers in the 667 open INET provide an authentication service for INET Client IPv6 ND 668 messages but only a secondary forwarding service when the Client 669 cannot forward directly to a peer or Bridge. Source Clients securely 670 coordinate with target Clients by sending control messages via a 671 First-Hop Segment (FHS) Proxy/Server which forwards them over the SRT 672 spanning tree to a Last-Hop Segment (LHS) Proxy/Server which finally 673 forwards them to the target. 675 AERO Relays are Proxy/Servers that provide forwarding services to 676 exchange original IP packets between the OMNI link and nodes on other 677 links/networks. Relays run a dynamic routing protocol to discover 678 any non-MNP prefixes in service on other links/networks, and Relays 679 that connect to larger Internetworks (such as the Internet) may 680 originate default routes. The Relay redistributes OMNI link MSP(s) 681 into other links/networks, and redistributes non-MNP prefixes via 682 OMNI link Bridge BGP peerings. 684 3.2. The AERO Service over OMNI Links 686 3.2.1. AERO/OMNI Reference Model 688 Figure 1 presents the basic OMNI link reference model: 690 +----------------+ 691 | AERO Bridge B1 | 692 | Nbr: S1, S2, P1| 693 |(X1->S1; X2->S2)| 694 | MSP M1 | 695 +-------+--------+ 696 +--------------+ | +--------------+ 697 | AERO P/S S1 | | | AERO P/S S2 | 698 | Nbr: C1, B1 | | | Nbr: C2, B1 | 699 | default->B1 | | | default->B1 | 700 | X1->C1 | | | X2->C2 | 701 +-------+------+ | +------+-------+ 702 | OMNI link | | 703 X===+===+==================+===================+===+===X 704 | | 705 +-----+--------+ +--------+-----+ 706 |AERO Client C1| |AERO Client C2| 707 | Nbr: S1 | | Nbr: S2 | 708 | default->S1 | | default->S2 | 709 | MNP X1 | | MNP X2 | 710 +------+-------+ +-----+--------+ 711 | | 712 .-. .-. 713 ,-( _)-. ,-( _)-. 714 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 715 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 716 `-(______)-' +-------+ +-------+ `-(______)-' 718 Figure 1: AERO/OMNI Reference Model 720 In this model: 722 o the OMNI link is an overlay network service configured over one or 723 more underlying SRT segments which may be managed by different 724 administrative authorities and have incompatible protocols and/or 725 addressing plans. 727 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 728 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 729 via BGP peerings over secured tunnels to Proxy/Servers (S1, S2). 730 Bridges provide the backbone for an SRT spanning tree for the OMNI 731 link. 733 o AERO Proxy/Servers S1 and S2 configure secured tunnels with Bridge 734 B1 and also provide mobility, multilink, multicast and default 735 router services for the MNPs of their associated Clients C1 and 736 C2. (Proxy/Servers that act as Relays can also advertise non-MNP 737 routes for non-mobile correspondent nodes the same as for MNP 738 Clients.) 740 o AERO Clients C1 and C2 associate with Proxy/Servers S1 and S2, 741 respectively. They receive MNP delegations X1 and X2, and also 742 act as default routers for their associated physical or internal 743 virtual EUNs. Simple hosts H1 and H2 attach to the EUNs served by 744 Clients C1 and C2, respectively. 746 An OMNI link configured over a single *NET appears as a single 747 unified link with a consistent underlying network addressing plan; 748 all nodes on the link can exchange carrier packets via simple *NET 749 encapsulation (i.e., following any necessary NAT traversal) since the 750 underlying *NET is connected. In common practice, however, OMNI 751 links are often configured over an SRT spanning tree that bridges 752 multiple distinct *NET segments managed under different 753 administrative authorities (e.g., as for worldwide aviation service 754 providers such as ARINC, SITA, Inmarsat, etc.). Individual *NETs may 755 also be partitioned internally, in which case each internal partition 756 appears as a separate segment. 758 The addressing plan of each SRT segment is consistent internally but 759 will often bear no relation to the addressing plans of other 760 segments. Each segment is also likely to be separated from others by 761 network security devices (e.g., firewalls, proxys, packet filtering 762 gateways, etc.), and disjoint segments often have no common physical 763 link connections. Therefore, nodes can only be assured of exchanging 764 carrier packets directly with correspondents in the same segment, and 765 not with those in other segments. The only means for joining the 766 segments therefore is through inter-domain peerings between AERO 767 Bridges. 769 The OMNI link spans multiple SRT segments using the OMNI Adaptation 770 Layer (OAL) [I-D.templin-6man-omni] to provide the network layer with 771 a virtual abstraction similar to a bridged campus LAN. The OAL is an 772 OMNI interface sublayer that inserts a mid-layer IPv6 encapsulation 773 header for inter-segment forwarding (i.e., bridging) without 774 decrementing the network-layer TTL/Hop Limit of the original IP 775 packet. An example OMNI link SRT is shown in Figure 2: 777 . . . . . . . . . . . . . . . . . . . . . . . 778 . . 779 . .-(::::::::) . 780 . .-(::::::::::::)-. +-+ . 781 . (:::: Segment A :::)--|B|---+ . 782 . `-(::::::::::::)-' +-+ | . 783 . `-(::::::)-' | . 784 . | . 785 . .-(::::::::) | . 786 . .-(::::::::::::)-. +-+ | . 787 . (:::: Segment B :::)--|B|---+ . 788 . `-(::::::::::::)-' +-+ | . 789 . `-(::::::)-' | . 790 . | . 791 . .-(::::::::) | . 792 . .-(::::::::::::)-. +-+ | . 793 . (:::: Segment C :::)--|B|---+ . 794 . `-(::::::::::::)-' +-+ | . 795 . `-(::::::)-' | . 796 . | . 797 . ..(etc).. x . 798 . . 799 . . 800 . <- Segment Routing Topology (SRT) -> . 801 . (Spanned by OMNI Link) . 802 . . . . . . . . . . . . . .. . . . . . . . . 804 Figure 2: OMNI Link Segment Routing Topology (SRT) 806 Bridge, Proxy/Server and Relay OMNI interfaces are configured over 807 both secured tunnels and open INET underlying interfaces within their 808 respective SRT segments. Within each segment, Bridges configure 809 "hub-and-spokes" BGP peerings with Proxy/Server/Relays as "spokes". 810 Adjacent SRT segments are joined by Bridge-to-Bridge peerings to 811 collectively form a spanning tree over the entire SRT. The "secured" 812 spanning tree supports authentication and integrity for critical 813 control plane messages. The "unsecured" spanning tree conveys 814 ordinary carrier packets without security codes and that must be 815 treated by destinations according to data origin authentication 816 procedures. AERO nodes can employ route optimization to cause 817 carrier packets to take more direct paths between OMNI link neighbors 818 without having to follow strict spanning tree paths. 820 3.2.2. Addressing and Node Identification 822 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 823 fe80::/64 [RFC4291] to assign LLAs used for network-layer addresses 824 in link-scoped IPv6 ND and data messages. AERO Clients use LLAs 825 constructed from MNPs (i.e., "MNP-LLAs") while other AERO nodes use 826 LLAs constructed based on 32-bit Mobility Service ID (MSID) values 827 ("ADM-LLAs") per [I-D.templin-6man-omni]. Non-MNP routes are also 828 represented the same as for MNP-LLAs, but may include a prefix that 829 is not properly covered by an MSP. 831 AERO nodes also use the Unique Local Address (ULA) prefix fd00::/8 832 followed by a pseudo-random 40-bit OMNI domain identifier to form the 833 prefix [ULA]::/48, then include a 16-bit OMNI link identifier '*' to 834 form the prefix [ULA*]::/64 [RFC4291]. The AERO node then uses the 835 prefix [ULA*]::/64 to form "MNP-ULAs" or "ADM-ULA"s as specified in 836 [I-D.templin-6man-omni] to support OAL addressing. (The prefix 837 [ULA*]::/64 appearing alone and with no suffix represents "default".) 838 AERO Clients also use Temporary ULAs constructed per 839 [I-D.templin-6man-omni], where the addresses are typically used only 840 in initial control message exchanges until a stable MNP-LLA/ULA is 841 assigned. 843 AERO MSPs, MNPs and non-MNP routes are typically based on Global 844 Unicast Addresses (GUAs), but in some cases may be based on private- 845 use addresses. A GUA block is also reserved for OMNI link anycast 846 purposes. See [I-D.templin-6man-omni] for a full specification of 847 LLAs, ULAs and GUAs used by AERO nodes on OMNI links. 849 Finally, AERO Clients and Proxy/Servers configure node identification 850 values as specified in [I-D.templin-6man-omni]. 852 3.2.3. AERO Routing System 854 The AERO routing system comprises a private Border Gateway Protocol 855 (BGP) [RFC4271] service coordinated between Bridges and Proxy/ 856 Servers. The service supports carrier packet forwarding at a layer 857 below IP and does not interact with the public Internet BGP routing 858 system, but supports redistribution of information for other links 859 and networks discovered by Relays. 861 In a reference deployment, each Proxy/Server is configured as an 862 Autonomous System Border Router (ASBR) for a stub Autonomous System 863 (AS) using a 32-bit AS Number (ASN) [RFC4271] that is unique within 864 the BGP instance, and each Proxy/Server further uses eBGP to peer 865 with one or more Bridges but does not peer with other Proxy/Servers. 866 Each SRT segment in the OMNI link must include one or more Bridges in 867 a "hub" AS, which peer with the Proxy/Servers within that segment as 868 "spoke" ASes. All Bridges within the same segment are members of the 869 same hub AS, and use iBGP to maintain a consistent view of all active 870 routes currently in service. The Bridges of different segments peer 871 with one another using eBGP. 873 Bridges maintain forwarding table entries only for the MNP-ULAs 874 corresponding to MNP and non-MNP routes that are currently active, 875 and also maintain black-hole routes for the OMNI link MSPs so that 876 carrier packets destined to non-existent MNP-ULAs are dropped with a 877 Destination Unreachable message returned. In this way, Proxy/Servers 878 and Relays have only partial topology knowledge (i.e., they only 879 maintain routing information for their directly associated Clients 880 and non-AERO links) and they forward all other carrier packets to 881 Bridges which have full topology knowledge. 883 Each OMNI link segment assigns a unique ADM-ULA sub-prefix of 884 [ULA*]::/96 known as the "SRT prefix". For example, a first segment 885 could assign [ULA*]::1000/116, a second could assign 886 [ULA*]::2000/116, a third could assign [ULA*]::3000/116, etc. Within 887 each segment, each Proxy/Server configures an ADM-ULA within the 888 segment's SRT prefix, e.g., the Proxy/Servers within [ULA*]::2000/116 889 could assign the ADM-ULAs [ULA*]::2011/116, [ULA*]::2026/116, 890 [ULA*]::2003/116, etc. 892 The administrative authorities for each segment must therefore 893 coordinate to assure mutually-exclusive ADM-ULA prefix assignments, 894 but internal provisioning of ADM-ULAs an independent local 895 consideration for each administrative authority. For each ADM-ULA 896 prefix, the Bridge(s) that connect that segment assign the all-zero's 897 address of the prefix as a Subnet Router Anycast address. For 898 example, the Subnet Router Anycast address for [ULA*]::1023/116 is 899 simply [ULA*]::1000. 901 ADM-ULA prefixes are statically represented in Bridge forwarding 902 tables. Bridges join multiple SRT segments into a unified OMNI link 903 over multiple diverse network administrative domains. They support a 904 virtual bridging service by first establishing forwarding table 905 entries for their ADM-ULA prefixes either via standard BGP routing or 906 static routes. For example, if three Bridges ('A', 'B' and 'C') from 907 different segments serviced [ULA*]::1000/116, [ULA*]::2000/116 and 908 [ULA*]::3000/116 respectively, then the forwarding tables in each 909 Bridge appear as follows: 911 A: [ULA*]::1000/116->local, [ULA*]::2000/116->B, [ULA*]::3000/116->C 912 B: [ULA*]::1000/116->A, [ULA*]::2000/116->local, [ULA*]::3000/116->C 914 C: [ULA*]::1000/116->A, [ULA*]::2000/116->B, [ULA*]::3000/116->local 916 These forwarding table entries rarely change, since they correspond 917 to fixed infrastructure elements in their respective segments. 919 MNP (and non-MNP) ULAs are instead dynamically advertised in the AERO 920 routing system by Proxy/Servers and Relays that provide service for 921 their corresponding MNPs. For example, if three Proxy/Servers ('D', 922 'E' and 'F') service the MNPs 2001:db8:1000:2000::/56, 923 2001:db8:3000:4000::/56 and 2001:db8:5000:6000::/56 then the routing 924 system would include: 926 D: [ULA*]:2001:db8:1000:2000/120 928 E: [ULA*]:2001:db8:3000:4000/120 930 F: [ULA*]:2001:db8:5000:6000/120 932 A full discussion of the BGP-based routing system used by AERO is 933 found in [I-D.ietf-rtgwg-atn-bgp]. 935 3.2.4. Segment Routing Topologies (SRTs) 937 The 64-bit sub-prefixes of [ULA]::/48 identify up to 2^16 distinct 938 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 939 OMNI link overlay instance using a distinct set of ULAs, and emulates 940 a bridged campus LAN service for the OMNI link. In some cases (e.g., 941 when redundant topologies are needed for fault tolerance and 942 reliability) it may be beneficial to deploy multiple SRTs that act as 943 independent overlay instances. A communication failure in one 944 instance therefore will not affect communications in other instances. 946 Each SRT is identified by a distinct value in bits 48-63 of 947 [ULA]::/48, i.e., as [ULA0]::/64, [ULA1]::/64, [ULA2]::/64, etc. 948 Each OMNI interface is identified by a unique interface name (e.g., 949 omni0, omni1, omni2, etc.) and assigns an OMNI IPv6 anycast address 950 used for OMNI interface determination in Safety-Based Multilink (SBM) 951 as discussed in [I-D.templin-6man-omni]. Each OMNI interface further 952 applies Performance-Based Multilink (PBM) internally. 954 The Bridges and Proxy/Servers of each independent SRT engage in BGP 955 peerings to form a spanning tree with the Bridges in non-leaf nodes 956 and the Proxy/Servers in leaf nodes. The spanning tree is configured 957 over both secured and unsecured underlying network paths. The 958 secured spanning tree is used to convey secured control messages 959 between Proxy/Servers and Bridges, while the unsecured spanning tree 960 forwards data messages and/or unsecured control messages. 962 Each SRT segment is identified by a unique ADM-ULA prefix used by all 963 Proxy/Servers and Bridges in the segment. Each AERO node must 964 therefore discover an SRT prefix that correspondents can use to 965 determine the correct segment, and must publish the SRT prefix in 966 IPv6 ND messages. 968 3.2.5. Segment Routing For OMNI Link Selection 970 Original IPv6 source can direct IPv6 packets to an AERO node by 971 including a standard IPv6 Segment Routing Header (SRH) [RFC8754] with 972 the OMNI IPv6 anycast address for the selected OMNI link as either 973 the IPv6 destination or as an intermediate hop within the SRH. This 974 allows the original source to determine the specific OMNI link SRT an 975 original IPv6 packet will traverse when there may be multiple 976 alternatives. 978 When an AERO node processes the SRH and forwards the original IPv6 979 packet to the correct OMNI interface, the OMNI interface writes the 980 next IPv6 address from the SRH into the IPv6 destination address and 981 decrements Segments Left. If decrementing would cause Segments Left 982 to become 0, the OMNI interface deletes the SRH before forwarding. 983 This form of Segment Routing supports Safety-Based Multilink (SBM). 985 3.3. OMNI Interface Characteristics 987 OMNI interfaces are virtual interfaces configured over one or more 988 underlying interfaces classified as follows: 990 o INET interfaces connect to an INET either natively or through one 991 or more NATs. Native INET interfaces have global IP addresses 992 that are reachable from any INET correspondent. The INET-facing 993 interfaces of Proxy/Servers are native interfaces, as are Relay 994 and Bridge interfaces. NATed INET interfaces connect to a private 995 network behind one or more NATs with the outermost NAT providing 996 INET access. Clients that are behind a NAT are required to send 997 periodic keepalive messages to keep NAT state alive when there are 998 no carrier packets flowing. 1000 o ANET interfaces connect to an ANET that is separated from the open 1001 INET by an FHS Proxy/Server. Clients can issue control messages 1002 over the ANET without including an authentication signature since 1003 the ANET is secured at the network layer or below. Proxy/Servers 1004 can actively issue control messages over the INET on behalf of 1005 ANET Clients to reduce ANET congestion. The same as for INET 1006 interfaces, there may be NATs on the path from the Client to its 1007 FHS Proxy/Server. 1009 o VPNed interfaces use security encapsulation over the INET to a 1010 Virtual Private Network (VPN) server that also acts as an FHS 1011 Proxy/Server. Other than the link-layer encapsulation format, 1012 VPNed interfaces behave the same as Direct interfaces. 1014 o Direct (i.e., single-hop point-to-point) interfaces connect a 1015 Client directly to an FHS Proxy/Server without crossing any ANET/ 1016 INET paths. An example is a line-of-sight link between a remote 1017 pilot and an unmanned aircraft. The same Client considerations 1018 apply as for VPNed interfaces. 1020 OMNI interfaces use OAL encapsulation and fragmentation as discussed 1021 in Section 3.6. OMNI interfaces use *NET encapsulation (see: 1022 Section 3.6) to exchange carrier packets with OMNI link neighbors 1023 over INET or VPNed interfaces as well as over ANET interfaces for 1024 which the Client and FHS Proxy/Server may be multiple IP hops away. 1025 OMNI interfaces do not use link-layer encapsulation over Direct 1026 underlying interfaces or ANET interfaces when the Client and FHS 1027 Proxy/Server are known to be on the same underlying link. 1029 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1030 state the same as for any interface. OMNI interfaces use IPv6 ND 1031 messages including Router Solicitation (RS), Router Advertisement 1032 (RA), Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1033 neighbor cache management. In environments where spoofing may be a 1034 threat, OMNI neighbors should invoke OAL Identification window 1035 synchronization in their IPv6 ND message exchanges. 1037 OMNI interfaces send IPv6 ND messages with an OMNI option formatted 1038 as specified in [I-D.templin-6man-omni]. The OMNI option includes 1039 prefix registration information, Interface Attributes and/or 1040 Multilink Forwarding Parameters containing link information 1041 parameters for the OMNI interface's underlying interfaces and any 1042 other per-neighbor information. 1044 A Client's OMNI interface may be configured over multiple underlying 1045 interfaces. For example, common mobile handheld devices have both 1046 wireless local area network ("WLAN") and cellular wireless links. 1047 These links are often used "one at a time" with low-cost WLAN 1048 preferred and highly-available cellular wireless as a standby, but a 1049 simultaneous-use capability could provide benefits. In a more 1050 complex example, aircraft frequently have many wireless data link 1051 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1052 directional, etc.) with diverse performance and cost properties. 1054 If a Client's multiple underlying interfaces are used "one at a time" 1055 (i.e., all other interfaces are in standby mode while one interface 1056 is active), then successive IPv6 ND messages all include OMNI option 1057 Multilink Forwarding Parameters sub-options with the same underlying 1058 interface index. In that case, the Client would appear to have a 1059 single underlying interface but with a dynamically changing link- 1060 layer address. 1062 If the Client has multiple active underlying interfaces, then from 1063 the perspective of IPv6 ND it would appear to have multiple link- 1064 layer addresses. In that case, IPv6 ND message OMNI options MAY 1065 include Interface Attributes and/or Multilink Forwarding Parameters 1066 sub-options with different underlying interface indexes. 1068 Proxy/Servers on the open Internet include only a single INET 1069 underlying interface. INET Clients therefore discover only the 1070 INADDR information for the Proxy/Server's INET interface. Proxy/ 1071 Servers on an ANET/INET boundary include both an ANET and INET 1072 underlying interface. ANET Clients therefore must discover both the 1073 ANET and INET INADDR information for the Proxy/Server. 1075 Bridge and Proxy/Server OMNI interfaces are configured over 1076 underlying interfaces that provide both secured tunnels for carrying 1077 IPv6 ND and BGP protocol control plane messages and open INET access 1078 for carrying unsecured messages. The OMNI interface configures both 1079 an ADM-LLA and its corresponding ADM-ULA, and acts as an OAL source 1080 to encapsulate and fragment original IP packets while presenting the 1081 resulting carrier packets over the secured or unsecured underlying 1082 paths. Note that Bridge and Proxy/Server end-to-end transport 1083 protocol sessions used by the BGP are run directly over the OMNI 1084 interface and use ADM-ULA source and destination addresses. The OMNI 1085 interface employs the OAL to encapsulate the original IP packets for 1086 these sessions as carrier packets (i.e., even though the OAL header 1087 may use the same ADM-ULAs as the original IP header) and forwards 1088 them over the secured underlying path. 1090 3.4. OMNI Interface Initialization 1092 AERO Proxy/Servers and Clients configure OMNI interfaces as their 1093 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1094 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1095 to ensure that original IP packets with destination addresses covered 1096 by an MNP not explicitly associated with another interface are 1097 directed to an OMNI interface. 1099 OMNI interface initialization procedures for Proxy/Servers, Clients 1100 and Bridges are discussed in the following sections. 1102 3.4.1. AERO Proxy/Server and Relay Behavior 1104 When a Proxy/Server enables an OMNI interface, it assigns an 1105 ADM-{LLA,ULA} appropriate for the given OMNI link SRT segment. The 1106 Proxy/Server also configures secured tunnels with one or more 1107 neighboring Bridges and engages in BGP routing protocol sessions with 1108 one or more Bridges. 1110 The OMNI interface provides a single interface abstraction to the IP 1111 layer, but internally includes an NBMA nexus for sending carrier 1112 packets to OMNI interface neighbors over underlying INET interfaces 1113 and secured tunnels. The Proxy/Server further configures a service 1114 to facilitate IPv6 ND exchanges with AERO Clients and manages per- 1115 Client neighbor cache entries and IP forwarding table entries based 1116 on control message exchanges. 1118 Relays are simply Proxy/Servers that run a dynamic routing protocol 1119 to redistribute routes between the OMNI interface and INET/EUN 1120 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1121 networks on the INET/EUN interfaces (i.e., the same as a Client would 1122 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1123 interfaces. The Relay further provides an attachment point of the 1124 OMNI link to a non-MNP-based global topology. 1126 3.4.2. AERO Client Behavior 1128 When a Client enables an OMNI interface, it assigns either an 1129 MNP-{LLA, ULA} or a Temporary ULA and sends RS messages over its 1130 underlying interfaces to an FHS Proxy/Server, which coordinates with 1131 a Hub Proxy/Server that returns an RA message with corresponding 1132 parameters. The RS/RA messages may pass through one or more NATs in 1133 the path between the Client and FHS Proxy/Server. (Note: if the 1134 Client used a Temporary ULA in its initial RS message, it will 1135 discover an MNP-{LLA,ULA} in the corresponding RA that it receives 1136 from the FHS Proxy/Server and begin using these new addresses. If 1137 the Client is operating outside the context of AERO infrastructure 1138 such as in a Mobile Ad-hoc Network (MANET), however, it may continue 1139 using Temporary ULAs for Client-to-Client communications until it 1140 encounters an infrastructure element that can delegate an MNP.) 1142 3.4.3. AERO Bridge Behavior 1144 AERO Bridges configure an OMNI interface and assign an ADM-ULA and 1145 corresponding Subnet Router Anycast address for each OMNI link SRT 1146 segment they connect to. Bridges configure secured tunnels with 1147 Proxy/Servers in the same SRT segment and other Bridges in the same 1148 (or an adjacent) SRT segment. Bridges then engage in a BGP routing 1149 protocol session with neighbors over the secured spanning tree (see: 1150 Section 3.2.3). 1152 3.5. OMNI Interface Neighbor Cache Maintenance 1154 Each OMNI interface maintains a conceptual neighbor cache that 1155 includes a Neighbor Cache Entry (NCE) for each of its active 1156 neighbors on the OMNI link per [RFC4861]. Each NCE is indexed by the 1157 LLA of the neighbor, while the OAL encapsulation ULA determines the 1158 context for Identification verification. Clients and Proxy/Servers 1159 maintain NCEs through RS/RA exchanges, and also maintain NCEs for any 1160 active correspondent peers through NS/NA exchanges. 1162 Bridges also maintain NCEs for Clients within their local segments 1163 based on NS/NA route optimization messaging (see: Section 3.13.4). 1164 When a Bridge creates/updates a NCE for a local segment Client based 1165 on NS/NA route optimization, it also maintains MVFI and INADDR state 1166 for messages destined to this local segment Client. 1168 Proxy/Servers add an additional state DEPARTED to the list of NCE 1169 states found in Section 7.3.2 of [RFC4861]. When a Client terminates 1170 its association, the Proxy/Server OMNI interface sets a "DepartTime" 1171 variable for the NCE to "DEPART_TIME" seconds. DepartTime is 1172 decremented unless a new IPv6 ND message causes the state to return 1173 to REACHABLE. While a NCE is in the DEPARTED state, the Proxy/Server 1174 forwards carrier packets destined to the target Client to the 1175 Client's new FHS/Hub Proxy/Server instead. It is RECOMMENDED that 1176 DEPART_TIME be set to the default constant value 10 seconds to accept 1177 any carrier packets that may be in flight. When DepartTime 1178 decrements to 0, the NCE is deleted. 1180 Clients determine the service profiles for their FHS and Hub Proxy/ 1181 Servers by setting the N/A/U flags in the first OMNI option in RS 1182 messages. When the N/A/U flags are clear, Proxy/Servers forward all 1183 NS/NA messages to the Client, while the Client performs mobility 1184 update signaling through the transmission of uNA messages to all 1185 active neighbors following a mobility event. However, in some 1186 environments this may result in excessive NS/NA control message 1187 overhead especially for Clients connected to low-end data links. 1189 To minimize NS/NA message overhead, Clients can set the N/A/U flags 1190 in the first OMNI option of RS messages they send. If the N flag is 1191 set, the FHS Proxy/Server that forwards the RS message assumes the 1192 role of responding to NS(NUD) messages and maintains peer NCEs 1193 associated with the NCE for this Client. If the A flag is set, the 1194 Hub Proxy/Server that processes the RS message assumes the role of 1195 responding to NS(AR) messages on behalf of this Client NCE. If the U 1196 flag is set, the Hub Proxy/Server that processes the RS message 1197 becomes responsible for maintaining a "Report List" of sources from 1198 which it has received an NS(AR) for this Client NCE. The Hub Proxy/ 1199 Server maintains each Report List entry for REPORT_TIME seconds, and 1200 sends uNA messages to each member of the Report List when it receives 1201 a Client mobility update indication (e.g., through receipt of an RS 1202 with updated Interface Attributes, Traffic Selectors, etc.). 1204 Clients and their Hub Proxy/Servers have full knowledge of the 1205 Client's current underlying Interface Attributes, while FHS Proxy/ 1206 Servers acting in "proxy" mode have knowledge of only the individual 1207 Client underlying interfaces they service. Clients determine their 1208 FHS and Hub Proxy/Server service models by setting the N/A/U flags in 1209 the RS messages they send as discussed above. 1211 Clients act as RORs on their own behalf when they receive an NS(AR) 1212 from an ROS via their Hub Proxy/Server (Relays instead act as RORs on 1213 behalf of non-MNP targets specific to other links/networks that the 1214 Relay services and/or "default"). The ROR returns and NA(AR) 1215 response to the ROS, which creates or updates a NCE for the target 1216 network-layer and link-layer addresses. The ROS then (re)sets 1217 ReachableTime for the NCE to REACHABLE_TIME seconds and performs 1218 reachability tests over specific underlying interface pairs to 1219 determine paths for forwarding carrier packets directly to the 1220 target. The ROS otherwise decrements ReachableTime while no further 1221 solicited NA messages arrive. It is RECOMMENDED that REACHABLE_TIME 1222 be set to the default constant value 30 seconds as specified in 1223 [RFC4861]. 1225 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1226 of NS messages sent when a correspondent may have gone unreachable, 1227 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1228 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1229 to limit the number of unsolicited NAs that can be sent based on a 1230 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1231 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1232 same as specified in [RFC4861]. 1234 Different values for the above constants MAY be administratively set; 1235 however, if different values are chosen, all nodes on the link MUST 1236 consistently configure the same values. Most importantly, 1237 DEPART_TIME and REPORT_TIME SHOULD be set to a value that is 1238 sufficiently longer than REACHABLE_TIME to avoid packet loss due to 1239 stale route optimization state. 1241 3.5.1. OMNI ND Messages 1243 OMNI interfaces prepare IPv6 ND messages the same as for standard 1244 IPv6 ND, but also include a new option type termed the OMNI option 1245 [I-D.templin-6man-omni]. For each IPv6 ND message, OMNI interfaces 1246 include one or more OMNI options (and any other ND message options) 1247 then completely populate all option information. If the OMNI 1248 interface includes an authentication signature, it sets the IPv6 ND 1249 message Checksum field to 0 and calculates the authentication 1250 signature over the entire length of the OAL packet or super-packet 1251 (beginning with a pseudo-header of the IPv6 header up to but not 1252 including the trailing OAL checksum) but does not then proceed to 1253 calculate the IPv6 ND message checksum itself. Otherwise, the OMNI 1254 interface calculates the standard IPv6 ND message checksum over the 1255 OAL packet or super-packet and writes the value in the Checksum 1256 field. OMNI interfaces verify authentication and integrity of each 1257 IPv6 ND message received according to the specific check(s) included, 1258 and process the message further only following verification. 1260 OMNI options include per-neighbor information that provides multilink 1261 forwarding, link-layer address and traffic selector information for 1262 the neighbor's underlying interfaces. This information is stored in 1263 the neighbor cache and provides the basis for the forwarding 1264 algorithm specified in Section 3.10. The information is cumulative 1265 and reflects the union of the OMNI information from the most recent 1266 IPv6 ND messages received from the neighbor; it is therefore not 1267 required that each IPv6 ND message contain all neighbor information. 1269 The OMNI option is distinct from any Source/Target Link-Layer Address 1270 Options (S/TLLAOs) that may appear in an IPv6 ND message according to 1271 the appropriate IPv6 over specific link layer specification (e.g., 1272 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1273 pertains to encapsulation addresses while the latter pertains to the 1274 native L2 address format of the underlying media 1276 OMNI interface IPv6 ND messages may also include other IPv6 ND 1277 options. In particular, solicitation messages may include a Nonce 1278 option if required for verification of advertisement replies. If an 1279 OMNI IPv6 ND solicitation message includes a Nonce option, the 1280 advertisement reply must echo the same Nonce. If an OMNI IPv6 ND 1281 advertisement message includes a Timestamp option, the recipient 1282 should check the Timestamp to determine if the message is current. 1284 AERO Clients send RS messages to the link-scoped All-Routers 1285 multicast address or an ADM-LLA while using unicast link-layer 1286 addresses. AERO Proxy/Servers respond by returning unicast RA 1287 messages. During the RS/RA exchange, AERO Clients and Proxy/Servers 1288 include state synchronization parameters to establish Identification 1289 windows and other state. 1291 AERO nodes use NS/NA messages for the following purposes: 1293 o NS/NA(AR) messages are used for address resolution and optionally 1294 to establish sequence number windows. The ROS sends an NS(AR) to 1295 the solicited-node multicast address of the target, and an ROR 1296 with addressing information for the target returns a unicast 1297 NA(AR) that contains current, consistent and authentic target 1298 address resolution information. NS/NA(AR) messages must be 1299 secured. 1301 o NS/NA(NUD) messages are used to establish multilink forwarding 1302 state and determine target reachability. The source sends an 1303 NS(NUD) to the unicast address of the target while naming a 1304 specific underlying interface pair, and the target returns a 1305 unicast NA(NUD). NS/NA(NUD) messages that use an in-window 1306 sequence number and do not update any other state need not include 1307 an authentication signature but instead must include an IPv6 ND 1308 message checksum. NS/NA(NUD) messages may also be used to 1309 establish window synchronization and/or MFIB state, in which case 1310 the messages must be secured. 1312 o Unsolicited NA (uNA) messages are used to signal addressing and/or 1313 other neighbor state changes (e.g., address changes due to 1314 mobility, signal degradation, traffic selector updates, etc.). uNA 1315 messages that update state information must be secured. 1317 o NS/NA(DAD) messages are not used in AERO, since Duplicate Address 1318 Detection is not required. 1320 Additionally, nodes may sent the OMNI option PNG flag in NA/RA 1321 messages to receive a uNA response from the neighbor. The uNA 1322 response MUST set the ACK flag (without also setting the SYN or PNG 1323 flags) with the Acknowledgement field set to the Identification used 1324 in the PNG message. 1326 3.5.2. OMNI Neighbor Advertisement Message Flags 1328 As discussed in Section 4.4 of [RFC4861] NA messages include three 1329 flag bits R, S and O. OMNI interface NA messages treat the flags as 1330 follows: 1332 o R: The R ("Router") flag is set to 1 in the NA messages sent by 1333 all AERO/OMNI node types. Simple hosts that would set R to 0 do 1334 not occur on the OMNI link itself, but may occur on the downstream 1335 links of Clients and Relays. 1337 o S: The S ("Solicited") flag is set exactly as specified in 1338 Section 4.4. of [RFC4861], i.e., it is set to 1 for Solicited NAs 1339 and set to 0 for uNAs (both unicast and multicast). 1341 o O: The O ("Override") flag is set to 0 for solicited NAs returned 1342 by a Proxy/Server ROR and set to 1 for all other solicited and 1343 unsolicited NAs. For further study is whether solicited NAs for 1344 anycast targets apply for OMNI links. Since MNP-LLAs must be 1345 uniquely assigned to Clients to support correct IPv6 ND protocol 1346 operation, however, no role is currently seen for assigning the 1347 same MNP-LLA to multiple Clients. 1349 3.5.3. OMNI Neighbor Window Synchronization 1351 In secured environments (e.g., between secured spanning tree 1352 neighbors, between neighbors on the same secured ANET, etc.), OMNI 1353 interface neighbors can exchange OAL packets using randomly- 1354 initialized and monotonically-increasing Identification values 1355 (modulo 2**32) without window synchronization. In environments where 1356 spoofing is considered a threat, OMNI interface neighbors instead 1357 invoke window synchronization in NS/NA message exchanges to maintain 1358 send/receive window state in their respective neighbor cache entries 1359 as specified in [I-D.templin-6man-omni]. 1361 3.6. OMNI Interface Encapsulation and Fragmentation 1363 When the network layer forwards an original IP packet into an OMNI 1364 interface, the interface locates or creates a Neighbor Cache Entry 1365 (NCE) that matches the destination. The OMNI interface then invokes 1366 the OMNI Adaptation Layer (OAL) as discussed in 1367 [I-D.templin-6man-omni] which encapsulates the packet in an IPv6 1368 header, and IPv6 Fragment Header and 2-octet Checksum trailer to 1369 produce an OAL packet. This OAL source then calculates the checksum 1370 and fragments the OAL packet (and invokes OAL header compression when 1371 appropriate) while including an identical Identification value for 1372 each fragment that must be within the window for the LHS Proxy/Server 1373 or the target Client itself. 1375 The OAL source next includes an identical Compressed Routing Header 1376 with 32-bit ID fields (CRH-32) [I-D.bonica-6man-comp-rtg-hdr] with 1377 each fragment if necessary containing one or more Multilink 1378 Forwarding Vector Indices (MFVIs) as discussed in Section 3.13. (The 1379 OAL source can then optionally invoke OAL header compression by 1380 replacing the OAL IPv6 header and CRH-32 with an OAL Compressed 1381 Header (OCH), Type 0 or 1.) 1383 The OAL source finally encapsulates each resulting OAL fragment in 1384 *NET headers to form an OAL carrier packet, with source address set 1385 to its own *NET address (e.g., 192.0.2.100) and destination set to 1386 the *NET address of the next hop OAL intermediate node or destination 1387 (e.g., 192.0.2.1). The carrier packet encapsulation format in the 1388 above example is shown in Figure 3: 1390 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1391 | *NET Header | 1392 | src = 192.0.2.100 | 1393 | dst = 192.0.2.1 | 1394 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1395 | OAL IPv6 Header | 1396 | src = [ULA*]::2001:db8:1:2 | 1397 | dst= [ULA*]::3000:0000 | 1398 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1399 | CRH-32 (if necessary) | 1400 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1401 | OAL Fragment Header | 1402 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1403 | Original IP Header | 1404 | (first-fragment only) | 1405 | src = 2001:db8:1:2::1 | 1406 | dst = 2001:db8:1234:5678::1 | 1407 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1408 | | 1409 ~ ~ 1410 ~ Original Packet Body/Fragment ~ 1411 ~ ~ 1412 | | 1413 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1414 | OAL Trailing Checksum | 1415 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1417 Figure 3: Carrier Packet Format 1419 In this format, the OAL source encapsulates the original IP header 1420 and packet body/fragment in an OAL IPv6 header prepared according to 1421 [RFC2473], the CRH-32 is a Routing Header extension of the OAL 1422 header, the Fragment Header identifies each fragment, and the *NET 1423 header is prepared as discussed in [I-D.templin-6man-omni]. The OAL 1424 source transmits each such carrier packet into the SRT spanning tree, 1425 where they are forwarded over possibly multiple OAL intermediate 1426 nodes until they arrive at the OAL destination. 1428 The OMNI link control plane service distributes Client MNP-ULA prefix 1429 information that may change dynamically due to regional node mobility 1430 as well as Relay non-MNP-ULA and per-segment ADM-ULA prefix 1431 information that rarely changes. OMNI link Bridges and Proxy/Servers 1432 use the information to establish and maintain a forwarding plane 1433 spanning tree that connects all nodes on the link. The spanning tree 1434 supports a carrier packet virtual bridging service according to link- 1435 layer (instead of network-layer) information, but may often include 1436 longer paths than necessary. 1438 Each OMNI interface therefore also includes a Multilink Forwarding 1439 Information Base (MFIB) with Multilink Forwarding Vectors (MFVs) that 1440 can often provide more direct forwarding "shortcuts" that avoid 1441 strict spanning tree paths. As a result, the spanning tree is always 1442 available but OMNI interfaces can often use the MFIB to greatly 1443 improve performance and reduce load on critical infrastructure 1444 elements. 1446 For carrier packets undergoing re-encapsulation at an OAL 1447 intermediate node, the OMNI interface decrements the OAL IPv6 header 1448 Hop Limit and discards the carrier packet if the Hop Limit reaches 0. 1449 The intermediate node next removes the *NET encapsulation headers 1450 from the first segment and re-encapsulates the packet in new *NET 1451 encapsulation headers for the next segment. 1453 When an FHS Bridge receives a carrier packet with an OCH-0/1 header 1454 that must be forwarded to an LHS Bridge over the unsecured spanning 1455 tree, it reconstructs the headers based on MFV state, inserts a 1456 CRH-32 immediately following the OAL header and adjusts the OAL 1457 payload length and destination address field. The FHS Bridge 1458 includes a single MFVI in the CRH-32 that will be meaningful to the 1459 LHS Bridge. When the LHS Bridge receives the carrier packet, it 1460 locates the MFV for the next hop based on the CRH-32 MFVI then re- 1461 applies header compression (resulting in the removal of the CRH-32) 1462 and forwards the carrier packet to the next hop. 1464 3.7. OMNI Interface Decapsulation 1466 OMNI interfaces (acting as OAL destinations) decapsulate and 1467 reassemble OAL packets into original IP packets destined either to 1468 the AERO node itself or to a destination reached via an interface 1469 other than the OMNI interface the original IP packet was received on. 1470 When carrier packets containing OAL fragments addressed to itself 1471 arrive, this OAL destination discards the NET encapsulation headers 1472 and reassembles to obtain the OAL packet or super-packet (see: 1473 [I-D.templin-6man-omni]). The OAL destination then verifies the OAL 1474 checksum, discards the OAL encapsulations to obtain the original IP 1475 packet(s) and finally forwards them to either the network layer or a 1476 next-hop on the OMNI link. 1478 3.8. OMNI Interface Data Origin Authentication 1480 AERO nodes employ simple data origin authentication procedures. In 1481 particular: 1483 o AERO Bridges and Proxy/Servers accept carrier packets received 1484 from the secured spanning tree. 1486 o AERO Proxy/Servers and Clients accept carrier packets and original 1487 IP packets that originate from within the same secured ANET. 1489 o AERO Clients and Relays accept original IP packets from downstream 1490 network correspondents based on ingress filtering. 1492 o AERO Clients, Relays, Proxy/Servers and Bridges verify carrier 1493 packet UDP/IP encapsulation addresses according to 1494 [I-D.templin-6man-omni]. 1496 o AERO nodes accept carrier packets addressed to themselves with 1497 Identification values within the current window for the OAL source 1498 neighbor (when window synchronization is used) and drop any 1499 carrier packets with out-of-window Identification values. (AERO 1500 nodes may forward carrier packets not addressed to themselves 1501 without verifying the Identification value.) 1503 AERO nodes silently drop any packets that do not satisfy the above 1504 data origin authentication procedures. Further security 1505 considerations are discussed in Section 6. 1507 3.9. OMNI Interface MTU 1509 The OMNI interface observes the link nature of tunnels, including the 1510 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1511 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 1512 The OMNI interface employs an OMNI Adaptation Layer (OAL) that 1513 accommodates multiple underlying links with diverse MTUs while 1514 observing both a minimum and per-path Maximum Payload Size (MPS). 1515 The functions of the OAL and the OMNI interface MTU/MRU/MPS are 1516 specified in [I-D.templin-6man-omni] with MTU/MRU both set to the 1517 constant value 9180 bytes, with minimum MPS set to 400 bytes, and 1518 with potentially larger per-path MPS values depending on the 1519 underlying path. 1521 When the network layer presents an original IP packet to the OMNI 1522 interface, the OAL source encapsulates and fragments the original IP 1523 packet if necessary. When the network layer presents the OMNI 1524 interface with multiple original IP packets bound to the same OAL 1525 destination, the OAL source can concatenate them together into a 1526 single OAL super-packet as discussed in [I-D.templin-6man-omni]. (If 1527 the super-packet begins with an IPv6 ND message that includes and 1528 authentication signature, the signature is calculated over the entire 1529 length of the super-packet up to but not including the trailing 1530 Checksum.) 1532 The OAL source then fragments the OAL packet if necessary according 1533 to the minimum/path MPS such that the OAL headers appear in each 1534 fragment while the original IP packet header appears only in the 1535 first fragment. The OAL source then encapsulates each OAL fragment 1536 in *NET headers for transmission as carrier packets over an 1537 underlying interface connected to either a physical link (such as 1538 Ethernet, WiFi and the like) or a virtual link such as an Internet or 1539 higher-layer tunnel (see the definition of link in [RFC8200]). 1541 Note: Although a CRH-32 may be inserted or removed by a Bridge in the 1542 path (see: Section 3.10.3), this does not interfere with the 1543 destination's ability to reassemble since the CRH-32 is not included 1544 in the fragmentable part and its removal/transformation does not 1545 invalidate fragment header information. 1547 3.10. OMNI Interface Forwarding Algorithm 1549 Original IP packets enter a node's OMNI interface either from the 1550 network layer (i.e., from a local application or the IP forwarding 1551 system) while carrier packets enter from the link layer (i.e., from 1552 an OMNI interface neighbor). All original IP packets and carrier 1553 packets entering a node's OMNI interface first undergo data origin 1554 authentication as discussed in Section 3.8. Those that satisfy data 1555 origin authentication are processed further, while all others are 1556 dropped silently. 1558 Original IP packets that enter the OMNI interface from the network 1559 layer are forwarded to an OMNI interface neighbor using OAL 1560 encapsulation and fragmentation to produce carrier packets for 1561 transmission over underlying interfaces. (If routing indicates that 1562 the original IP packet should instead be forwarded back to the 1563 network layer, the packet is dropped to avoid looping). Carrier 1564 packets that enter the OMNI interface from the link layer are either 1565 re-encapsulated and re-admitted into the OMNI link, or reassembled 1566 and forwarded to the network layer where they are subject to either 1567 local delivery or IP forwarding. In all cases, the OAL MUST NOT 1568 decrement the original IP packet TTL/Hop-count since its forwarding 1569 actions occur below the network layer. 1571 OMNI interfaces may have multiple underlying interfaces and/or 1572 neighbor cache entries for neighbors with multiple underlying 1573 interfaces (see Section 3.3). The OAL uses Interface Attributes and/ 1574 or Traffic Selectors (e.g., port number, flow specification, etc.) to 1575 select an outbound underlying interface for each OAL packet and also 1576 to select segment routing and/or link-layer destination addresses 1577 based on the neighbor's underlying interfaces. AERO implementations 1578 SHOULD permit network management to dynamically adjust Traffic 1579 Selector values at runtime. 1581 If an OAL packet matches the Traffic Selectors of multiple outgoing 1582 interfaces and/or neighbor interfaces, the OMNI interface replicates 1583 the packet and sends one copy via each of the (outgoing / neighbor) 1584 interface pairs; otherwise, it sends a single copy of the OAL packet 1585 via an interface with the best matching Traffic Selector. (While not 1586 strictly required, the likelihood of successful reassembly may 1587 improve when the OMNI interface sends all fragments of the same 1588 fragmented OAL packet consecutively over the same underlying 1589 interface pair to avoid complicating factors such as delay variance 1590 and reordering.) AERO nodes keep track of which underlying 1591 interfaces are currently "reachable" or "unreachable", and only use 1592 "reachable" interfaces for forwarding purposes. 1594 The following sections discuss the OMNI interface forwarding 1595 algorithms for Clients, Proxy/Servers and Bridges. In the following 1596 discussion, an original IP packet's destination address is said to 1597 "match" if it is the same as a cached address, or if it is covered by 1598 a cached prefix (which may be encoded in an MNP-LLA). 1600 3.10.1. Client Forwarding Algorithm 1602 When an original IP packet enters a Client's OMNI interface from the 1603 network layer the Client searches for a NCE that matches the 1604 destination. If there is a matching NCE, the Client selects one or 1605 more "reachable" neighbor interfaces in the entry for forwarding 1606 purposes. Otherwise, the Client invokes route optimization per 1607 Section 3.13 and follows the multilink forwarding procedures outlined 1608 there. 1610 When a carrier packet enters a Client's OMNI interface from the link- 1611 layer, if the OAL destination matches one of the Client's ULAs the 1612 Client (acting as an OAL destination) verifies that the 1613 Identification is in-window for this OAL source, then reassembles and 1614 decapsulates as necessary and delivers the original IP packet to the 1615 network layer. If the OAL destination does not match, the Client 1616 drops the original IP packet and MAY return a network-layer ICMP 1617 Destination Unreachable message subject to rate limiting (see: 1618 Section 3.11). 1620 Note: Clients and their FHS Proxy/Server (and other Client) peers can 1621 exchange original IP packets over ANET underlying interfaces without 1622 invoking the OAL, since the ANET is secured at the link and physical 1623 layers. By forwarding original IP packets without invoking the OAL, 1624 however, the ANET peers can engage only in classical path MTU 1625 discovery since the packets are subject to loss and/or corruption due 1626 to the various per-link MTU limitations that may occur within the 1627 ANET. Moreover, the original IP packets do not include either the 1628 OAL integrity check or per-packet Identification values that can be 1629 used for data origin authentication and link-layer retransmissions. 1630 The tradeoff therefore involves an assessment of the per-packet 1631 encapsulation overhead saved by bypassing the OAL vs. inheritance of 1632 classical network "brittleness". (Note however that ANET peers can 1633 send small original IP packets without invoking the OAL, while 1634 invoking the OAL for larger packets. This presents the beneficial 1635 aspects of both small packet efficiency and large packet robustness, 1636 with delay variance and reordering as possible side effects.) 1638 3.10.2. Proxy/Server and Relay Forwarding Algorithm 1640 When a Proxy/Server receives an original IP packet from the network 1641 layer, it drops the packet if routing indicates that it should be 1642 forwarded back to the network layer to avoid looping. Otherwise, the 1643 Proxy/Server regards the original IP packet the same as if it had 1644 arrived as carrier packets with OAL destination set to its own ADM- 1645 ULA. When the Proxy/Server receives carrier packets on underlying 1646 interfaces with OAL destination set to its own ADM-ULA, it performs 1647 OAL reassembly if necessary to obtain the original IP packet. The 1648 Proxy/Server then supports multilink forwarding procedures as 1649 specified in Section 3.13.2 and/or acts as an ROS to initiate route 1650 optimization as specified in Section 3.13. 1652 When the Proxy/Server receives a carrier packet with OAL destination 1653 set to an MNP-ULA that does not match the MSP, it accepts the carrier 1654 packet only if data origin authentication succeeds and if there is a 1655 network layer routing table entry for a GUA route that matches the 1656 MNP-ULA. If there is no route, the Proxy/Server drops the carrier 1657 packet; otherwise, it reassembles and decapsulates to obtain the 1658 original IP packet then acts as a Relay to present it to the network 1659 layer where it will be delivered according to standard IP forwarding. 1661 When a Proxy/Server receives a carrier packet from one of its Client 1662 neighbors with OAL destination set to another node, it forwards the 1663 packets via a matching NCE or via the spanning tree if there is no 1664 matching entry. When the Proxy/Server receives a carrier packet with 1665 OAL destination set to the MNP-ULA of one of its Client neighbors 1666 established through RS/RA exchanges, it accepts the carrier packet 1667 only if data origin authentication succeeds. If the NCE state is 1668 DEPARTED, the Proxy/Server changes the OAL destination address to the 1669 ADM-ULA of the new Proxy/Server, then re-encapsulates the carrier 1670 packet and forwards it to a Bridge which will eventually deliver it 1671 to the new Proxy/Server. If the neighbor cache state for the MNP-ULA 1672 is REACHABLE, the Proxy/Server forwards the carrier packets to the 1673 Client which then must reassemble. (Note that the Proxy/Server does 1674 not reassemble carrier packets not explicitly addressed to its own 1675 ADM-ULA, since some of the carrier packets of the same original IP 1676 packet could be forwarded through a different Proxy/Server.) In that 1677 case, the Client may receive fragments that are smaller than its link 1678 MTU but that can still be reassembled. 1680 Proxy/Servers process carrier packets with OAL destinations that do 1681 not match their ADM-ULA in the same manner as for traditional IP 1682 forwarding within the OAL, i.e., nodes use IP forwarding to forward 1683 packets not explicitly addressed to themselves. (Proxy/Servers 1684 include a special case that accepts and reassembles carrier packets 1685 destined to the MNP-ULA of one of their Clients received over the 1686 secured spanning tree.) Proxy/Servers process carrier packets with 1687 their ADM-ULA as the destination by first examining the packet for a 1688 CRH-32 header or an OCH-0/1 header. In that case, the Proxy/Server 1689 examines the next MFVI in the carrier packet to locate the MFV entry 1690 in the MFIB for next hop forwarding (i.e., without examining IP 1691 addresses). When the Proxy/Server forwards the carrier packet, it 1692 changes the destination address according to the MFVI value for the 1693 next hop found either in the CRH-32 header or in the node's own MFIB. 1694 Proxy/Servers must verify that the *NET addresses of carrier packets 1695 not received from the secured spanning tree are "trusted" before 1696 forwarding according to an MFV (otherwise, the carrier packet must be 1697 dropped). 1699 Note: Proxy/Servers may receive carrier packets addressed to their 1700 own ADM-ULA with CRH-32s that include additional forwarding 1701 information. Proxy/Servers use the forwarding information to 1702 determine the correct NCE and underlying interface for forwarding to 1703 the target Client, then remove the CRH-32 and forward the carrier 1704 packet. If necessary, the Proxy/Server reassembles first before re- 1705 encapsulating (and possibly also re-fragmenting) then forwards to the 1706 target Client. 1708 Note: Clients and their FHS Proxy/Server peers can exchange original 1709 IP packets over ANET underlying interfaces without invoking the OAL, 1710 since the ANET is secured at the link and physical layers. By 1711 forwarding original IP packets without invoking the OAL, however, the 1712 Client and Proxy/Server can engage only in classical path MTU 1713 discovery since the packets are subject to loss and/or corruption due 1714 to the various per-link MTU limitations that may occur within the 1715 ANET. Moreover, the original IP packets do not include either the 1716 OAL integrity check or per-packet Identification values that can be 1717 used for data origin authentication and link-layer retransmissions. 1719 The tradeoff therefore involves an assessment of the per-packet 1720 encapsulation overhead saved by bypassing the OAL vs. inheritance of 1721 classical network "brittleness". (Note however that ANET peers can 1722 send small original IP packets without invoking the OAL, while 1723 invoking the OAL for larger packets. This presents the beneficial 1724 aspects of both small packet efficiency and large packet robustness.) 1726 Note: When a Proxy/Server receives a (non-OAL) original IP packet 1727 from an ANET Client, or a carrier packet with OAL destination set to 1728 its own ADM-ULA from any Client, the Proxy/Server reassembles if 1729 necessary then performs ROS functions on behalf of the Client. The 1730 Client may at some later time begin sending carrier packets to the 1731 OAL address of the actual target instead of the Proxy/Server, at 1732 which point it may begin functioning as an ROS on its own behalf and 1733 thereby "override" the Proxy/Server's ROS role. 1735 Note; Proxy/Servers drop any original IP packets (received either 1736 directly from an ANET Client or following reassembly of carrier 1737 packets received from an ANET/INET Client) with a destination that 1738 corresponds to the Client's delegated MNP. Similarly, Proxy/Servers 1739 drop any carrier packet received with both a source and destination 1740 that correspond to the Client's delegated MNP regardless of their 1741 OMNI link point of origin. These checks are necessary to prevent 1742 Clients from either accidentally or intentionally establishing 1743 endless loops that could congest Proxy/Servers and/or ANET/INET 1744 links. 1746 Note: Proxy/Servers forward secure control plane carrier packets via 1747 the SRT secured spanning tree and forward other carrier packets via 1748 the unsecured spanning tree. When a Proxy/Server receives a carrier 1749 packet from the secured spanning tree, it considers the message as 1750 authentic without having to verify upper layer authentication 1751 signatures. When a Proxy/Server receives a carrier packet from the 1752 unsecured spanning tree, it applies data origin authentication itself 1753 and/or forwards the unsecured message toward the destination which 1754 must apply data origin authentication on its own behalf. 1756 Note: If the Proxy/Server has multiple original IP packets to send to 1757 the same neighbor, it can concatenate them in a single OAL super- 1758 packet [I-D.templin-6man-omni]. 1760 3.10.3. Bridge Forwarding Algorithm 1762 Bridges forward spanning tree carrier packets while decrementing the 1763 OAL header Hop Count but not the original IP header Hop Count/TTL. 1764 Bridges convey carrier packets that encapsulate critical IPv6 ND 1765 control messages or routing protocol control messages via the SRT 1766 secured spanning tree, and may convey other carrier packets via the 1767 secured/unsecured spanning tree or via more direct paths according to 1768 MFIB information. When the Bridge receives a carrier packet, it 1769 removes the outer *NET header and searches for an MFIB entry that 1770 matches an MFVI or an IP forwarding table entry that matches the OAL 1771 destination address. 1773 Bridges process carrier packets with OAL destinations that do not 1774 match their ADM-ULA or the SRT Subnet Router Anycast address in the 1775 same manner as for traditional IP forwarding within the OAL, i.e., 1776 nodes use IP forwarding to forward packets not explicitly addressed 1777 to themselves. Bridges process carrier packets with their ADM-ULA or 1778 the SRT Subnet Router Anycast address as the destination by first 1779 examining the packet for a full OAL header with a CRH-32 extension or 1780 an OCH-0/1 header. In that case, the Bridge examines the next MFVI 1781 in the carrier packet to locate the MFV entry in the MFIB for next 1782 hop forwarding (i.e., without examining IP addresses). When the 1783 Bridge forwards the carrier packet, it changes the destination 1784 address according to the MFVI value for the next hop found either in 1785 the CRH-32 header or in the node's own MFIB. If the Bridge has a NCE 1786 for the target Client with an entry for the target underlying 1787 interface and current *NET addresses, the Bridge instead forwards 1788 directly to the target Client while using the final hop MFVI instead 1789 of the next hop (see: Section 3.13.4). 1791 Bridges forward carrier packets received from a first segment via the 1792 secured spanning tree to the next segment also via the secured 1793 spanning tree. Bridges forward carrier packets received from a first 1794 segment via the unsecured spanning tree to the next segment also via 1795 the unsecured spanning tree. Bridges use a single IPv6 routing table 1796 that always determines the same next hop for a given OAL destination, 1797 where the secured/unsecured spanning tree is determined through the 1798 selection of the underlying interface to be used for transmission 1799 (i.e., a secured tunnel or an open INET interface). 1801 As for Proxy/Servers, Bridges must verify that the *NET addresses of 1802 carrier packets not received from the secured spanning tree are 1803 "trusted" before forwarding according to an MFV (otherwise, the 1804 carrier packet must be dropped). 1806 3.11. OMNI Interface Error Handling 1808 When an AERO node admits an original IP packet into the OMNI 1809 interface, it may receive link-layer or network-layer error 1810 indications. The AERO node may also receive OMNI link error 1811 indications in OAL-encapsulated uNA messages that include 1812 authentication signatures. 1814 A link-layer error indication is an ICMP error message generated by a 1815 router in the *NET on the path to the neighbor or by the neighbor 1816 itself. The message includes an IP header with the address of the 1817 node that generated the error as the source address and with the 1818 link-layer address of the AERO node as the destination address. 1820 The IP header is followed by an ICMP header that includes an error 1821 Type, Code and Checksum. Valid type values include "Destination 1822 Unreachable", "Time Exceeded" and "Parameter Problem" 1823 [RFC0792][RFC4443]. (OMNI interfaces ignore link-layer IPv4 1824 "Fragmentation Needed" and IPv6 "Packet Too Big" messages for carrier 1825 packets that are no larger than the minimum/path MPS as discussed in 1826 Section 3.9, however these messages may provide useful hints of probe 1827 failures during path MPS probing.) 1829 The ICMP header is followed by the leading portion of the carrier 1830 packet that generated the error, also known as the "packet-in-error". 1831 For ICMPv6, [RFC4443] specifies that the packet-in-error includes: 1832 "As much of invoking packet as possible without the ICMPv6 packet 1833 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1834 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1835 "Internet Header + 64 bits of Original Data Datagram", however 1836 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1837 ICMP datagram SHOULD contain as much of the original datagram as 1838 possible without the length of the ICMP datagram exceeding 576 1839 bytes". 1841 The link-layer error message format is shown in Figure 4: 1843 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1844 ~ ~ 1845 | IP Header of link layer | 1846 | error message | 1847 ~ ~ 1848 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1849 | ICMP Header | 1850 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1851 ~ ~ P 1852 | carrier packet *NET and OAL | a 1853 | encapsulation headers | c 1854 ~ ~ k 1855 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1856 ~ ~ t 1857 | original IP packet headers | 1858 | (first-fragment only) | i 1859 ~ ~ n 1860 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1861 ~ ~ e 1862 | Portion of the body of | r 1863 | the original IP packet | r 1864 | (all fragments) | o 1865 ~ ~ r 1866 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1868 Figure 4: OMNI Interface Link-Layer Error Message Format 1870 The AERO node rules for processing these link-layer error messages 1871 are as follows: 1873 o When an AERO node receives a link-layer Parameter Problem message, 1874 it processes the message the same as described as for ordinary 1875 ICMP errors in the normative references [RFC0792][RFC4443]. 1877 o When an AERO node receives persistent link-layer Time Exceeded 1878 messages, the IP ID field may be wrapping before earlier fragments 1879 awaiting reassembly have been processed. In that case, the node 1880 should begin including integrity checks and/or institute rate 1881 limits for subsequent packets. 1883 o When an AERO node receives persistent link-layer Destination 1884 Unreachable messages in response to carrier packets that it sends 1885 to one of its neighbor correspondents, the node should process the 1886 message as an indication that a path may be failing, and 1887 optionally initiate NUD over that path. If it receives 1888 Destination Unreachable messages over multiple paths, the node 1889 should allow future carrier packets destined to the correspondent 1890 to flow through a default route and re-initiate route 1891 optimization. 1893 o When an AERO Client receives persistent link-layer Destination 1894 Unreachable messages in response to carrier packets that it sends 1895 to one of its neighbor Proxy/Servers, the Client should mark the 1896 path as unusable and use another path. If it receives Destination 1897 Unreachable messages on many or all paths, the Client should 1898 associate with a new Proxy/Server and release its association with 1899 the old Proxy/Server as specified in Section 3.15.5. 1901 o When an AERO Proxy/Server receives persistent link-layer 1902 Destination Unreachable messages in response to carrier packets 1903 that it sends to one of its neighbor Clients, the Proxy/Server 1904 should mark the underlying path as unusable and use another 1905 underlying path. 1907 o When an AERO Proxy/Server receives link-layer Destination 1908 Unreachable messages in response to a carrier packet that it sends 1909 to one of its permanent neighbors, it treats the messages as an 1910 indication that the path to the neighbor may be failing. However, 1911 the dynamic routing protocol should soon reconverge and correct 1912 the temporary outage. 1914 When an AERO Bridge receives a carrier packet for which the network- 1915 layer destination address is covered by an MSP assigned to a black- 1916 hole route, the Bridge drops the packet if there is no more-specific 1917 routing information for the destination and returns an OMNI interface 1918 Destination Unreachable message subject to rate limiting. 1920 When an AERO node receives a carrier packet for which reassembly is 1921 currently congested, it returns an OMNI interface Packet Too Big 1922 (PTB) message as discussed in [I-D.templin-6man-omni] (note that the 1923 PTB messages could indicate either "hard" or "soft" errors). 1925 AERO nodes include ICMPv6 error messages intended for the OAL source 1926 as sub-options in the OMNI option of secured uNA messages. When the 1927 OAL source receives the uNA message, it can extract the ICMPv6 error 1928 message enclosed in the OMNI option and either process it locally or 1929 translate it into a network-layer error to return to the original 1930 source. 1932 3.12. AERO Mobility Service Coordination 1934 AERO nodes observes the Router Discovery and Prefix Registration 1935 specifications found in Section 15 of [I-D.templin-6man-omni]. AERO 1936 nodes further coordinate their autoconfiguration actions with the 1937 mobility service as discussed in the following sections. 1939 3.12.1. AERO Service Model 1941 Each AERO Proxy/Server on the OMNI link is configured to facilitate 1942 Client prefix delegation/registration requests. Each Proxy/Server is 1943 provisioned with a database of MNP-to-Client ID mappings for all 1944 Clients enrolled in the AERO service, as well as any information 1945 necessary to authenticate each Client. The Client database is 1946 maintained by a central administrative authority for the OMNI link 1947 and securely distributed to all Proxy/Servers, e.g., via the 1948 Lightweight Directory Access Protocol (LDAP) [RFC4511], via static 1949 configuration, etc. Clients receive the same service regardless of 1950 the Proxy/Servers they select. 1952 Clients associate each of their underlying interfaces with a FHS 1953 Proxy/Server. Each FHS Proxy/Server locally services one or more of 1954 the Client's underlying interfaces, and the Client typically selects 1955 one among them to serve as the Hub Proxy/Server (the Client may 1956 instead select a "third-party" Hub Proxy/Server that does not 1957 directly service any of its underlying interfaces). All of the 1958 Client's other FHS Proxy/Servers forward proxyed copies of RS/RA 1959 messages between the Hub Proxy/Server and Client without assuming the 1960 Hub role functions themselves. 1962 Each Client associates with a single Hub Proxy/Server at a time, 1963 while all other Proxy/Servers are candidates for providing the Hub 1964 role for other Clients. An FHS Proxy/Server assumes the Hub role 1965 when it receives an RS message with its own ADM-LLA or link-scoped 1966 All-Routers multicast as the destination. An FHS Proxy/Server 1967 assumes the proxy role when it receives an RS message with the ADM- 1968 LLA of another Proxy/Server as the destination. (An FHS Proxy/Server 1969 can also assume the proxy role when it receives an RS message 1970 addressed to link-scoped All-Routers multicast if it can determine 1971 the ADM-LLA of another Proxy/Server to serve as a Hub.) 1973 AERO Clients and Proxy/Servers use IPv6 ND messages to maintain 1974 neighbor cache entries. AERO Proxy/Servers configure their OMNI 1975 interfaces as advertising NBMA interfaces, and therefore send unicast 1976 RA messages with a short Router Lifetime value (e.g., ReachableTime 1977 seconds) in response to a Client's RS message. Thereafter, Clients 1978 send additional RS messages to keep Proxy/Server state alive. 1980 AERO Clients and Hub Proxy/Servers include prefix delegation and/or 1981 registration parameters in RS/RA messages. The IPv6 ND messages are 1982 exchanged between the Client and Hub Proxy/Server (via any FHS Proxy/ 1983 Servers acting as proxys) according to the prefix management schedule 1984 required by the service. If the Client knows its MNP in advance, it 1985 can employ prefix registration by including its MNP-LLA as the source 1986 address of an RS message and with an OMNI option with valid prefix 1987 registration information for the MNP. If the Hub Proxy/Server 1988 accepts the Client's MNP assertion, it injects the MNP into the 1989 routing system and establishes the necessary neighbor cache state. 1990 If the Client does not have a pre-assigned MNP, it can instead employ 1991 prefix delegation by including the unspecified address (::) as the 1992 source address of an RS message and with an OMNI option with prefix 1993 delegation parameters to request an MNP. 1995 The following sections outlines Client and Proxy/Server behaviors 1996 based on the Router Discovery and Prefix Registration specifications 1997 found in Section 15 of [I-D.templin-6man-omni]. These sections 1998 observe all of the OMNI specifications, and include additional 1999 specifications of the interactions of Client-Proxy/Server RS/RA 2000 exchanges with the AERO mobility service. 2002 3.12.2. AERO Client Behavior 2004 AERO Clients discover the addresses of candidate Proxy/Servers by 2005 resolving the Potential Router List (PRL) in a similar manner as 2006 described in [RFC5214]. Discovery methods include static 2007 configuration (e.g., a flat-file map of Proxy/Server addresses and 2008 locations), or through an automated means such as Domain Name System 2009 (DNS) name resolution [RFC1035]. Alternatively, the Client can 2010 discover Proxy/Server addresses through a layer 2 data link login 2011 exchange, or through an RA response to a multicast/anycast RS as 2012 described below. In the absence of other information, the Client can 2013 resolve the DNS Fully-Qualified Domain Name (FQDN) 2014 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 2015 text string and "[domainname]" is a DNS suffix for the OMNI link 2016 (e.g., "example.com"). 2018 The Client then performs RS/RA exchanges over each of its underlying 2019 interfaces to associate with (possibly multiple) FHS Proxy/Serves and 2020 a single Hub Proxy/Server as specified in Section 15 of 2021 [I-D.templin-6man-omni]. The Client then sends each RS (either 2022 directly via Direct interfaces, via a VPN for VPNed interfaces, via 2023 an access router for ANET interfaces or via INET encapsulation for 2024 INET interfaces) and waits up to RetransTimer milliseconds for an RA 2025 message reply (see Section 3.12.3) while retrying up to 2026 MAX_RTR_SOLICITATIONS if necessary. If the Client receives no RAs, 2027 or if it receives an RA with Router Lifetime set to 0, the Client 2028 SHOULD abandon attempts through the first candidate Proxy/Server and 2029 try another Proxy/Server. 2031 After the Client registers its underlying interfaces, it may wish to 2032 change one or more registrations, e.g., if an interface changes 2033 address or becomes unavailable, if traffic selectors change, etc. To 2034 do so, the Client prepares an RS message to send over any available 2035 underlying interface as above. The RS includes an OMNI option with 2036 prefix registration/delegation information and with an Interface 2037 Attributes sub-option specific to the selected underlying interface. 2038 When the Client receives the Hub Proxy/Server's RA response, it has 2039 assurance that both the Hub and FHS Proxy/Servers have been updated 2040 with the new information. 2042 If the Client wishes to discontinue use of a Hub Proxy/Server it 2043 issues an RS message over any underlying interface with an OMNI 2044 option with a prefix release indication (i.e., by setting the OMNI 2045 extension header Preflen to 0). When the Hub Proxy/Server processes 2046 the message, it releases the MNP, sets the NCE state for the Client 2047 to DEPARTED and returns an RA reply with Router Lifetime set to 0. 2048 After a short delay (e.g., 2 seconds), the Hub Proxy/Server withdraws 2049 the MNP from the routing system. (Alternatively, when the Client 2050 associates with a new FHS/Hub Proxy/Server it can include an OMNI 2051 "Proxy/Server Departure" sub-option in RS messages with the MSIDs of 2052 the Old FHS/Hub Proxy/Server.) 2054 3.12.3. AERO Proxy/Server Behavior 2056 AERO Proxy/Servers act as both IP routers and IPv6 ND proxys, and 2057 support a prefix delegation/registration service for Clients. Proxy/ 2058 Servers arrange to add their ADM-LLAs to the PRL maintained in a 2059 static map of Proxy/Server addresses for the link, the DNS resource 2060 records for the FQDN "linkupnetworks.[domainname]", etc. before 2061 entering service. The PRL should be arranged such that Clients can 2062 discover the addresses of Proxy/Servers that are geographically and/ 2063 or topologically "close" to their underlying network connections. 2065 When a FHS/Hub Proxy/Server receives a prospective Client's RS 2066 message, it SHOULD return an immediate RA reply with Router Lifetime 2067 set to 0 if it is currently too busy or otherwise unable to service 2068 the Client; otherwise, it processes the RS as specified in Section 15 2069 of [I-D.templin-6man-omni]. When the Hub Proxy/Server receives the 2070 RS, it determines the correct MNPs to provide to the Client by 2071 processing the MNP-LLA prefix parameters and/or the DHCPv6 OMNI sub- 2072 option. When the Hub Proxy/Server returns the MNPs, it also creates 2073 a forwarding table entry for the MNP-ULA corresponding to each MNP 2074 resulting in a BGP update (see: Section 3.2.3). For IPv6, the Hub 2075 Proxy/Server creates an IPv6 forwarding table entry for each MNP-ULA. 2076 For IPv4, the Hub Proxy/Server creates an IPv6 forwarding table entry 2077 with the IPv4-compatibility MNP-ULA prefix corresponding to the IPv4 2078 address. The Hub Proxy/Server then returns an RA to the Client via 2079 an FHS Proxy/Server if necessary. 2081 After the initial RS/RA exchange, the Hub Proxy/Server maintains a 2082 ReachableTime timer for each of the Client's underlying interfaces 2083 individually (and for the Client's NCE collectively) set to expire 2084 after ReachableTime seconds. If the Client (or an FHS Proxy/Server) 2085 issues additional RS messages, the Hub Proxy/Server sends an RA 2086 response and resets ReachableTime. If the Hub Proxy/Server receives 2087 an IPv6 ND message with a prefix release indication it sets the 2088 Client's NCE to the DEPARTED state and withdraws the MNP-ULA route 2089 from the routing system after a short delay (e.g., 2 seconds). If 2090 ReachableTime expires before a new RS is received on an individual 2091 underlying interface, the Hub Proxy/Server marks the interface as 2092 DOWN. If ReachableTime expires before any new RS is received on any 2093 individual underlying interface, the Hub Proxy/Server sets the NCE 2094 state to STALE and sets a 10 second timer. If the Hub Proxy/Server 2095 has not received a new RS or uNA message with a prefix release 2096 indication before the 10 second timer expires, it deletes the NCE and 2097 withdraws the MNP from the routing system. 2099 The Hub Proxy/Server processes any IPv6 ND messages pertaining to the 2100 Client while forwarding to the Client or responding on the Client's 2101 behalf as necessary. The Hub Proxy/Server may also issue unsolicited 2102 RA messages, e.g., with reconfigure parameters to cause the Client to 2103 renegotiate its prefix delegation/registrations, with Router Lifetime 2104 set to 0 if it can no longer service this Client, etc. The Hub 2105 Proxy/Server may also receive carrier packets via the secured 2106 spanning tree that contain initial data packets sent while route 2107 optimization is in progress. The Hub Proxy/Server reassembles, then 2108 re-encapsulates/re-fragments and forwards the packets to the target 2109 Client via an FHS Proxy/Server if necessary. Finally, If the NCE is 2110 in the DEPARTED state, the old Hub Proxy/Server forwards any carrier 2111 packets it receives from the secure spanning tree and destined to the 2112 Client to the new Hub Proxy/Server, then deletes the entry after 2113 DepartTime expires. 2115 Note: Clients SHOULD arrange to notify former Hub Proxy/Servers of 2116 their departures, but Hub Proxy/Servers are responsible for expiring 2117 neighbor cache entries and withdrawing routes even if no departure 2118 notification is received (e.g., if the Client leaves the network 2119 unexpectedly). Hub Proxy/Servers SHOULD therefore set Router 2120 Lifetime to ReachableTime seconds in solicited RA messages to 2121 minimize persistent stale cache information in the absence of Client 2122 departure notifications. A short Router Lifetime also ensures that 2123 proactive RS/RA messaging between Clients and FHS Proxy/Servers will 2124 keep any NAT state alive (see above). 2126 Note: All Proxy/Servers on an OMNI link MUST advertise consistent 2127 values in the RA Cur Hop Limit, M and O flags, Reachable Time and 2128 Retrans Timer fields the same as for any link, since unpredictable 2129 behavior could result if different Proxy/Servers on the same link 2130 advertised different values. 2132 3.12.3.1. Additional Proxy/Server Considerations 2134 AERO Clients register with FHS Proxy/Servers for each underlying 2135 interface. Each of the Client's FHS Proxy/Servers must inform a 2136 single Hub Proxy/Server of the Client's underlying interface(s) that 2137 it services. For Clients on Direct and VPNed underlying interfaces, 2138 the FHS Proxy/Server for each interface is directly connected, for 2139 Clients on ANET underlying interfaces the FHS Proxy/Server is located 2140 on the ANET/INET boundary, and for Clients on INET underlying 2141 interfaces the FHS Proxy/Server is located somewhere in the connected 2142 Internetwork. When FHS Proxy/Server "B" processes a Client 2143 registration, it must either assume the Hub role or forward a proxyed 2144 registration to another Proxy/Server "A" acting as the Hub. Proxy/ 2145 Servers satisfy these requirements as follows: 2147 o when FHS Proxy/Server "B" receives a Client RS message, it first 2148 verifies that the OAL Identification is within the window for the 2149 NCE that matches the MNP-ULA for this Client neighbor and 2150 authenticates the message. If no NCE was found, Proxy/Server "B" 2151 instead creates one in the STALE state and caches the Client- 2152 supplied Interface Attributes, Origin Indication and OMNI 2153 extension header window synchronization parameters as well as the 2154 Client's observed *NET addresses (noting that they may differ from 2155 the Origin addresses if there were NATs on the path). Proxy/ 2156 Server "B" then examines the network-layer destination address. 2157 If the destination address is the ADM-LLA of a different Proxy/ 2158 Server "A", Proxy/Server "B" prepares a separate proxyed version 2159 of the RS message with an OAL header with source set to its own 2160 ADM-ULA and destination set to Proxy/Server B's ADM-ULA. Proxy/ 2161 Server "B" also writes its own information over the Interface 2162 Attributes sub-option supplied by the Client, omits or zeros the 2163 Origin Indication sub-option then forwards the message into the 2164 OMNI link secured spanning tree. 2166 o when Hub Proxy/Server "A" receives the RS, it assume the Hub role 2167 and creates or updates a NCE for the Client with FHS Proxy/Server 2168 "B"'s Interface Attributes as the link-layer address information 2169 for this FHS omIndex. Hub Proxy/Server "A" then prepares an RA 2170 message with source set to its own LLA and destination set to the 2171 Client's MNP-LLA, then encapsulates the RA in an OAL header with 2172 source set to its own ADM-ULA and destination set to the ADM-ULA 2173 of FHS Proxy/Server "B". Hub Proxy/Server "A" then performs 2174 fragmentation if necessary and sends the resulting carrier packets 2175 into the secured spanning tree. 2177 o when FHS Proxy/Server "B" reassembles the RA, it locates the 2178 Client NCE based on the RA destination LLA. If the RA message 2179 includes an OMNI "Proxy/Server Departure" sub-option, Proxy/Server 2180 "B" first sends a uNA to the old FHS/Hub Proxy/Servers named in 2181 the sub-option. Proxy/Server "B" then re-encapsulates the RA 2182 message with OAL source set to its own ADM-ULA and OAL destination 2183 set to the MNP-ULA of the Client, with an appropriate 2184 Identification value, with an authentication signature if 2185 necessary, with the Client's Interface Attributes sub-option 2186 echoed and with the cached observed *NET addresses written into an 2187 Origin Indication sub-option. Proxy/Server "B" sets the P flag in 2188 the RA flags field to indicate that the message has passed through 2189 a proxy [RFC4389], includes responsive window synchronization 2190 parameters, then fragments the RA if necessary and returns the 2191 fragments to the Client. 2193 o The Client repeats this process over each of its additional 2194 underlying interfaces while treating each additional FHS Proxy/ 2195 Server "C", "D", "E", etc. as a proxy to facilitate RS/RA 2196 exchanges between the Hub and the Client. 2198 After the initial RS/RA exchanges each FHS Proxy/Server forwards any 2199 of the Client's carrier packets with OAL destinations for which there 2200 is no matching NCE to a Bridge using OAL encapsulation with its own 2201 ADM-ULA as the source and with destination determined by the Client. 2202 The Proxy/Server instead forwards any carrier packets destined to a 2203 neighbor cache target directly to the target according to the OAL/ 2204 link-layer information - the process of establishing neighbor cache 2205 entries is specified in Section 3.13. 2207 While the Client is still associated with FHS Proxy/Servers "B", "C", 2208 "D", etc., each FHS Proxy/Server can send NS, RS and/or unsolicited 2209 NA messages to update the neighbor cache entries of other AERO nodes 2210 on behalf of the Client based on changes in Interface Attributes, 2211 Traffic Selectors, etc. This allows for higher-frequency Proxy- 2212 initiated RS/RA messaging over well-connected INET infrastructure 2213 supplemented by lower-frequency Client-initiated RS/RA messaging over 2214 constrained ANET data links. 2216 If the Hub Proxy/Server "A" ceases to send solicited RAs, FHS Proxy/ 2217 Servers "B", "C", "D" can send unsolicited RAs over the Client's 2218 underlying interface with destination set to (link-local) All-Nodes 2219 multicast and with Router Lifetime set to zero to inform Clients that 2220 the Hub Proxy/Server has failed. Although FHS Proxy/Servers "B", "C" 2221 and "D" can engage in IPv6 ND exchanges on behalf of the Client, the 2222 Client can also send IPv6 ND messages on its own behalf, e.g., if it 2223 is in a better position to convey state changes. The IPv6 ND 2224 messages sent by the Client include the Client's MNP-LLA as the 2225 source in order to differentiate them from the IPv6 ND messages sent 2226 by a FHS Proxy/Server. 2228 If the Client becomes unreachable over all underlying interface it 2229 serves, the Hub Proxy/Server sets the NCE state to DEPARTED and 2230 retains the entry for DepartTime seconds. While the state is 2231 DEPARTED, the Hub Proxy/Server forwards any carrier packets destined 2232 to the Client to a Bridge via OAL encapsulation. When DepartTime 2233 expires, the Hub Proxy/Server deletes the NCE and discards any 2234 further carrier packets destined to the former Client. 2236 In some ANETs that employ a Proxy/Server, the Client's MNP can be 2237 injected into the ANET routing system. In that case, the Client can 2238 send original IP packets without invoking the OAL so that the ANET 2239 routing system transports the original IP packets to the Proxy/ 2240 Server. This can be beneficial, e.g., if the Client connects to the 2241 ANET via low-end data links such as some aviation wireless links. 2243 If the ANET first-hop access router is on the same underlying link as 2244 the Client and recognizes the AERO/OMNI protocol, the Client can 2245 avoid OAL encapsulation for both its control and data messages. When 2246 the Client connects to the link, it can send an unencapsulated RS 2247 message with source address set to its own MNP-LLA (or to a Temporary 2248 LLA), and with destination address set to the ADM-LLA of the Client's 2249 selected Proxy/Server or to link-scoped All-Routers multicast. The 2250 Client includes an OMNI option formatted as specified in 2251 [I-D.templin-6man-omni]. The Client then sends the unencapsulated RS 2252 message, which will be intercepted by the AERO-aware ANET access 2253 router. 2255 The ANET access router then performs OAL encapsulation on the RS 2256 message and forwards it to a Proxy/Server at the ANET/INET boundary. 2257 When the access router and Proxy/Server are one and the same node, 2258 the Proxy/Server would share and underlying link with the Client but 2259 its message exchanges with outside correspondents would need to pass 2260 through a security gateway at the ANET/INET border. The method for 2261 deploying access routers and Proxys (i.e. as a single node or 2262 multiple nodes) is an ANET-local administrative consideration. 2264 Note: When a Proxy/Server alters the IPv6 ND message contents before 2265 forwarding (e.g., such as altering the OMNI option contents), the 2266 original IPv6 ND message checksum or authentication signature is 2267 invalidated, and a new checksum or authentication signature must be 2268 calculated and included. 2270 Note: When a Proxy/Server receives a secured Client NS message, it 2271 performs the same proxying procedures as for described for RS 2272 messages above. The proxying procedures for NS/NA message exchanges 2273 is specified in Section 3.13. 2275 3.12.3.2. Detecting and Responding to Proxy/Server Failures 2277 In environments where fast recovery from Proxy/Server failure is 2278 required, FHS Proxy/Servers SHOULD use proactive Neighbor 2279 Unreachability Detection (NUD) to track Hub Proxy/Server reachability 2280 in a similar fashion as for Bidirectional Forwarding Detection (BFD) 2281 [RFC5880]. Each FHS Proxy/Server can then quickly detect and react 2282 to failures so that cached information is re-established through 2283 alternate paths. The NS/NA(NUD) control messaging is carried only 2284 over well-connected ground domain networks (i.e., and not low-end 2285 aeronautical radio links) and can therefore be tuned for rapid 2286 response. 2288 FHS Proxy/Servers perform continuous NS/NA(NUD) exchanges with the 2289 Hub Proxy/Server, e.g., one exchange per second. The FHS Proxy/ 2290 Server sends the NS(NUD) message via the spanning tree with its own 2291 ADM-LLA as the source and the ADM-LLA of the Hub Proxy/Server as the 2292 destination, and the Hub Proxy/Server responds with an NA(NUD). When 2293 the FHS Proxy/Server is also sending RS messages to a Hub Proxy/ 2294 Server on behalf of Clients, the resulting RA responses can be 2295 considered as equivalent hints of forward progress. This means that 2296 the FHS Proxy/Server need not also send a periodic NS(NUD) if it has 2297 already sent an RS within the same period. If the Hub Proxy/Server 2298 fails (i.e., if the FHS Proxy/Server ceases to receive 2299 advertisements), the FHS Proxy/Server can quickly inform Clients by 2300 sending unsolicited RA messages 2302 The FHS Proxy/Server sends unsolicited RA messages with source 2303 address set to the Hub Proxy/Server's address, destination address 2304 set to (link-local) All-Nodes multicast, and Router Lifetime set to 2305 0. The FHS Proxy/Server SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA 2306 messages separated by small delays [RFC4861]. Any Clients that had 2307 been using the failed Hub Proxy/Server will receive the RA messages 2308 and select one of its other FHS Proxy/Servers to assume the Hub role 2309 (i.e., by sending an RS with destination set to the ADM-LLA of the 2310 new Hub). 2312 3.12.3.3. DHCPv6-Based Prefix Registration 2314 When a Client is not pre-provisioned with an MNP-LLA, it will need 2315 for the Hub Proxy/Server to select one or more MNPs on its behalf and 2316 set up the correct state in the AERO routing service. (A Client with 2317 a pre-provisioned MNP may also request the Hub Proxy/Server to select 2318 additional MNPs.) The DHCPv6 service [RFC8415] is used to support 2319 this requirement. 2321 When a Client needs to have the Hub Proxy/Server select MNPs, it 2322 sends an RS message with source address set to the unspecified 2323 address (::) and with an OMNI option that includes a DHCPv6 message 2324 sub-option with DHCPv6 Prefix Delegation (DHCPv6-PD) parameters. 2325 When the Hub Proxy/Server receives the RS message, it extracts the 2326 DHCPv6-PD message from the OMNI option. 2328 The Hub Proxy/Server then acts as a "Proxy DHCPv6 Client" in a 2329 message exchange with the locally-resident DHCPv6 server, which 2330 delegates MNPs and returns a DHCPv6-PD Reply message. (If the Hub 2331 Proxy/Server wishes to defer creation of MN state until the DHCPv6-PD 2332 Reply is received, it can instead act as a Lightweight DHCPv6 Relay 2333 Agent per [RFC6221] by encapsulating the DHCPv6-PD message in a 2334 Relay-forward/reply exchange with Relay Message and Interface ID 2335 options.) 2337 When the Hub Proxy/Server receives the DHCPv6-PD Reply, it adds a 2338 route to the routing system and creates an MNP-LLA based on the 2339 delegated MNP. The Hub Proxy/Server then sends an RA back to the 2340 Client with the (newly-created) MNP-LLA as the destination address 2341 and with the DHCPv6-PD Reply message and OMNI extension header 2342 Preflen coded in the OMNI option. When the Client receives the RA, 2343 it creates a default route, assigns the Subnet Router Anycast address 2344 and sets its MNP-LLA based on the delegated MNP. 2346 Note: Further details of the DHCPv6-PD based MNP registration (as 2347 well as a minimal MNP delegation alternative that avoids including a 2348 DHCPv6 message sub-option in the RS) are found in 2349 [I-D.templin-6man-omni]. 2351 3.13. AERO Route Optimization 2353 AERO nodes invoke route optimization when they need to forward 2354 initial packets to new target destinations and for ongoing multilink 2355 forwarding for current destinations. Route optimization is based on 2356 IPv6 ND Address Resolution messaging between a Route Optimization 2357 Source (ROS) and a Relay or the target Client itself (reached via the 2358 current Hub Proxy/Server) acting as a Route Optimization Responder 2359 (ROR). Route optimization is initiated by the first eligible ROS 2360 closest to the source as follows: 2362 o For Clients on VPNed and Direct interfaces, the Client's FHS 2363 Proxy/Server is the ROS. 2365 o For Clients on ANET interfaces, either the Client or the FHS 2366 Proxy/Server may be the ROS. 2368 o For Clients on INET interfaces, the Client itself is the ROS. 2370 o For correspondent nodes on INET/EUN interfaces serviced by a 2371 Relay, the Relay is the ROS. 2373 o For Clients that engage the Hub Proxy/Server in "mobility anchor" 2374 mode, the Hub Proxy/Server is the ROS. 2376 The AERO routing system directs a route optimization request sent by 2377 the ROS to the ROR, which returns a route optimization reply which 2378 must include information that is current, consistent and authentic. 2379 The ROS is responsible for periodically refreshing the route 2380 optimization, and the ROR is responsible for quickly informing the 2381 ROS of any changes. Following address resolution, the ROS and ROR 2382 perform ongoing multilink route optimizations to maintain optimal 2383 forwarding profiles. 2385 The route optimization procedures are specified in the following 2386 sections. 2388 3.13.1. Multilink Address Resolution 2390 When one or more original IP packets from a source node destined to a 2391 target node arrives, the ROS checks for a NCE with an MNP-LLA that 2392 matches the target destination. If there is a NCE in the REACHABLE 2393 state, the ROS invokes the OAL and forwards the resulting carrier 2394 packets according to the cached state then returns from processing. 2395 Otherwise, if there is no NCE the ROS creates one in the INCOMPLETE 2396 state. 2398 The ROS next prepares an NS message for Address Resolution (NS(AR)) 2399 to send toward an ROR while including the original IP packet(s) as 2400 trailing data following the NS(AR) in an OAL super-packet 2401 [I-D.templin-6man-omni]. The resulting NS(AR) message must be sent 2402 securely, and includes: 2404 o the LLA of the ROS as the source address. 2406 o the MNP-LLA corresponding to the original IP packet's destination 2407 as the Target Address, e.g., for 2001:db8:1:2::10:2000 the Target 2408 Address is fe80::2001:db8:1:2. 2410 o the Solicited-Node multicast address [RFC4291] formed from the 2411 lower 24 bits of the original IP packet's destination as the 2412 destination address, e.g., for 2001:db8:1:2::10:2000 the NS(AR) 2413 destination address is ff02:0:0:0:0:1:ff10:2000. 2415 The NS(AR) message also includes an OMNI option with an 2416 authentication sub-option if necessary and with OMNI extension header 2417 Preflen set to the prefix length associated with the NS(AR) source. 2419 The ROS also includes Interface Attributes and Traffic Selectors for 2420 all of the source Client's underlying interfaces, calculates the 2421 authentication signature or checksum, then selects an Identification 2422 value and submits the NS(AR) message for OAL encapsulation with OAL 2423 source set to its own {ADM,MNP}-ULA and OAL destination set to the 2424 MNP-ULA corresponding to the target and with window synchronization 2425 parameters. The ROS then inserts a fragment header, performs 2426 fragmentation and *NET encapsulation, then sends the resulting 2427 carrier packets into the SRT secured spanning tree without 2428 decrementing the network-layer TTL/Hop Limit field. 2430 When the ROS is a Client, it must instead use the ADM-ULA of one of 2431 its FHS Proxy/Servers as the destination. The ROS Client then 2432 fragments, performs *NET encapsulation and forwards the carrier 2433 packets to the FHS Proxy/Server. The FHS Proxy/Server then 2434 reassembles, verifies the NS(AR) authentication signature or 2435 checksum, changes the OAL source to its own ADM-ULA, changes the OAL 2436 destination to the MNP-ULA corresponding to the target, selects an 2437 appropriate Identification, then re-fragments and forwards the 2438 resulting carrier packets into the secured spanning tree on behalf of 2439 the Client. 2441 Note: both the target Client and its Hub Proxy/Server include current 2442 and accurate information for the Client's multilink Interface 2443 Attributes profile. The Hub Proxy/Server can be trusted to provide 2444 an authoritative response on behalf of the Client should the need 2445 arise. While the Client has no such trust basis, any attempt by the 2446 Client to mount an attack by providing false Interface Attributes 2447 information would only result in black-holing of return traffic, 2448 i.e., the "attack" could only result in denial of service to the 2449 Client itself. Therefore, the Client's asserted Interface Attributes 2450 need not be validated by the Hub Proxy/Server. 2452 3.13.1.1. Relaying the NS(AR) *NET Packet(s) 2454 When the Bridge receives carrier packets containing the NS(AR), it 2455 discards the *NET headers and determines the next hop by consulting 2456 its standard IPv6 forwarding table for the OAL header destination 2457 address. The Bridge then decrements the OAL header Hop-Limit, then 2458 re-encapsulates and forwards the carrier packet(s) via the secured 2459 spanning tree the same as for any IPv6 router, where they may 2460 traverse multiple OMNI link segments. The final-hop Bridge will 2461 deliver the carrier packet via the secured spanning tree to the Hub 2462 Proxy/Server (or Relay) that services the target. 2464 3.13.1.2. Processing and Responding to the NS(AR) 2466 When the Hub Proxy/Server for the target receives the NS(AR) secured 2467 carrier packets with the MNP-ULA of the target as the OAL 2468 destination, it reassembles then forwards the message to the target 2469 Client (while including an authentication signature and encapsulation 2470 if necessary) or processes the NS(AR) locally if it is acting as a 2471 Relay/IP router or the Client's designated ROR. The Hub Proxy/Server 2472 processes the message as follows: 2474 o if the NS(AR) target matches a Client NCE in the DEPARTED state, 2475 the (old) Hub Proxy/Server re-encapsulates by setting the OAL 2476 destination address to the ADM-ULA of the Client's new Hub Proxy/ 2477 Server. The old Hub Proxy/Server then re-fragments and re- 2478 encapsulates, then forwards the resulting carrier packets over the 2479 secured spanning tree. 2481 o If the NS(AR) target matches the MNP-LLA of a Client NCE in the 2482 REACHABLE state, the Hub Proxy/Server notes whether the NS(AR) 2483 arrived from the secured spanning tree then sets the OAL 2484 destination address to the MNP-ULA of the Client or the ADM-ULA of 2485 the selected FHS Proxy/Server for the Client. If the message 2486 arrived via the secured spanning tree the Hub Proxy/Server 2487 verifies the checksum; otherwise, it must verify the message 2488 authentication signature before forwarding. When the Hub Proxy/ 2489 Server determines the underlying interface for the target Client, 2490 it then changes the OAL destination to the ADM-ULA of the target 2491 Client's FHS Proxy/Server, re-fragments and forwards the resulting 2492 carrier packets into the secured spanning tree. When the FHS 2493 Proxy/Server receives the carrier packets, it reassembles and 2494 verifies the checksum, then includes an authentication signature 2495 if necessary, changes the OAL source to its own ADM-ULA and 2496 destination to the MNP-ULA of the target Client, includes an 2497 Identification value within the current window, then re-fragments 2498 and forwards the resulting carrier packets to the target Client 2499 ROR. (Note that if the Hub and FHS Proxy/Server are one and the 2500 same the Hub itself will perform the FHS procedures.) 2502 o If the NS(AR) target matches one of its non-MNP routes, the Hub 2503 Proxy/Server serves as both a Relay and a ROR, since the Relay 2504 forwards IP packets toward the (fixed network) target at the 2505 network layer. 2507 The ROR then creates a NCE for the NS(AR) LLA source address if 2508 necessary, processes the window synchronization parameters, caches 2509 all Interface Attributes and Traffic Selector information, and 2510 prepares a (solicited) NA(AR) message to return to the ROS with the 2511 source address set to its own MNP-LLA, the destination address set to 2512 the NS(AR) LLA source address and the Target Address set to the same 2513 value that appeared in the NS(AR) Target Address. The ROR includes 2514 an OMNI option with OMNI extension header Preflen set to the prefix 2515 length associated with the NA(AR) source address. 2517 The ROR then sets the NA(AR) message R flag to 1 (as a router) and S 2518 flag to 1 (as a response to a solicitation) and sets the O flag to 1 2519 (as an authoritative responder). The ROR finally submits the NA(AR) 2520 for OAL encapsulation with source set to its own ULA and destination 2521 set to either the ULA corresponding to the NS(AR) source or the ADM- 2522 ULA of its FHS Proxy/Server, selects an appropriate Identification, 2523 and includes window synchronization parameters and authentication 2524 signature or checksum. The ROR then includes Interface Attributes 2525 and Traffic Selector sub-options for all of the target's underlying 2526 interfaces with current information for each interface, fragments and 2527 encapsulates each fragment in appropriate *NET headers, then forwards 2528 the resulting (*NET-encapsulated) carrier packets to the FHS Proxy/ 2529 Server. 2531 When the FHS Proxy/Server receives the carrier packets, it 2532 reassembles if necessary and verifies the authentication signature or 2533 checksum. The FHS Proxy/Server then changes the OAL source address 2534 to its own ADM-ULA, changes the destination to the {ADM,MNP}-ULA 2535 corresponding to the NA(AR) LLA destination, includes an appropriate 2536 Identification, then fragments and forwards the carrier packets into 2537 the secured spanning tree. 2539 Note: If the Hub Proxy/Server is acting as the Client's ROR but not 2540 as a Relay/IP router (i.e., by virtue of receipt of an RS message 2541 with the A flag set), it prepares the NS(AR) with the R flag set to 0 2542 but without setting the SYN flag in the OMNI extension header window 2543 synchronization parameters. This informs the ROS that it must 2544 initiate multilink route optimization to synchronize with the Client 2545 either directly or via a FHS Proxy/Server (see: Section 3.13.2). 2547 3.13.1.3. Relaying the NA(AR) 2549 When the Bridge receives NA(AR) carrier packets, it discards the *NET 2550 header and determines the next hop by consulting its standard IPv6 2551 forwarding table for the OAL header destination address. The Bridge 2552 then decrements the OAL header Hop-Limit, re-encapsulates the carrier 2553 packet and forwards it via the SRT secured spanning tree, where it 2554 may traverse multiple OMNI link segments. The final-hop Bridge will 2555 deliver the carrier packet via the secured spanning tree to a Proxy/ 2556 Server for the ROS. 2558 3.13.1.4. Processing the NA(AR) 2560 When the ROS receives the NA(AR) message, it first searches for a NCE 2561 that matches the NA(AR) target address. The ROS then processes the 2562 message the same as for standard IPv6 Address Resolution [RFC4861]. 2563 In the process, it caches all OMNI option information in the target 2564 NCE (including all Interface Attributes), and caches the NA(AR) MNP- 2565 LLA source address as the address of the target Client. 2567 When the ROS is a Client, the SRT secured spanning tree will first 2568 deliver the solicited NA(AR) message to the FHS Proxy/Server, which 2569 re-encapsulates and forwards the message to the Client. If the 2570 Client is on a well-managed ANET, physical security and protected 2571 spectrum ensures security for the NA(AR) without needing an 2572 additional authentication signature; if the Client is on the open 2573 INET the Proxy/Server must instead include an authentication 2574 signature (while adjusting the OMNI option size, if necessary). The 2575 Proxy/Server uses its own ADM-ULA as the OAL source and the MNP-ULA 2576 of the Client as the OAL destination. 2578 3.13.2. Multilink Route Optimization 2580 Following address resolution, the ROS and ROR can assert multilink 2581 paths through underlying interface pairs serviced by the same source/ 2582 destination LLAs by sending unicast NS/NA messages with Multilink 2583 Forwarding Parameters and OMNI extension header window 2584 synchronization parameters when necessary. The unicast NS/NA 2585 messages establish multilink forwarding state in intermediate nodes 2586 in the path between the ROS and ROR. 2588 To support multilink route optimization, OMNI interfaces include an 2589 additional forwarding table termed the Multilink Forwarding 2590 Information Base (MFIB) that supports carrier packet forwarding based 2591 on OMNI neighbor underlying interface pairs. The MFIB contains 2592 Multilink Forwarding Vectors (MFVs) indexed by 4-octet values known 2593 as MFV Indexes (MFVIs). 2595 OAL source, intermediate and destination nodes create MFVs/MFVIs when 2596 they process an NS message with a Multilink Forwarding Parameters 2597 sub-option with Job code "00" (Initialize; Build B) or a solicited NA 2598 with Job code "01" (Follow B; Build A) (see: 2599 [I-D.templin-6man-omni]). The OAL source of the NS (and OAL 2600 destination of the solicited NA) are considered to reside in the 2601 "First Hop Segment (FHS)", while the OAL destination of the NS (and 2602 OAL source of the solicited NA) are considered to reside in the "Last 2603 Hop Segment (LHS)". 2605 When an OAL node processes an NS with Job code "00", it creates an 2606 MFV, records the NS source and destination ULAs and assigns a "B" 2607 MFVI. When the "B" MVFI is referenced, the MVF retains the ULAs in 2608 (dst,src) order the opposite of how they appeared in the original NS 2609 to support full header reconstruction. (If the NS message included a 2610 nested OAL encapsulation, the ULAs of both OAL headers are retained.) 2612 When an OAL node processes a solicited NA with Job code "01", it 2613 locates the MFV created by the NS and assigns an "A" MFVI. When the 2614 "A" MFVI is referenced, the MFV retains the ULAs in (src,dst) order 2615 the same as they appeared in the original NS to support full header 2616 reconstruction. (If the NS message included a nested OAL 2617 encapsulation, the ULAs of both OAL headers are retained.) 2619 OAL nodes generate random 32-bit values as candidate A/B MFVIs which 2620 must first be tested for local uniqueness. If a candidate MFVI s 2621 already in use (or if the value is 0), the OAL node repeats the 2622 random generation process until it obtains a unique non-zero value. 2623 (Since the number of MFVs in service at each OAL node is likely to be 2624 much smaller than 2**32, the process will generate a unique value 2625 after a small number of tries; also, an MFVI generated by a first OAL 2626 node is never tested for uniqueness on other OAL nodes, since the 2627 uniqueness property is node-local only.) 2629 OAL nodes maintain A/B MFVIs as follows: 2631 o "B1" - a locally-unique MFVI maintained independently by each OAL 2632 node on the path from the FHS OAL source to the last OAL 2633 intermediate node before the LHS OAL destination. The OAL node 2634 generates and assigns a "B1" MFVI to a newly-created MFV when it 2635 processes an NS message with Job code "00". When the OAL node 2636 receives future carrier packets that include this value, it can 2637 unambiguously locate the correct MFV and determine directionality 2638 without examining addresses. 2640 o "A1" - a locally unique MFVI maintained independently by each OAL 2641 node on the path from the LHS OAL source to the last OAL 2642 intermediate node before the FHS OAL destination. The OAL node 2643 generates and assigns an "A1" MFVI to the MVF that configures the 2644 corresponding "B1" MFVI when it processes a solicited NA message 2645 with Job code "01". When the OAL node receives future carrier 2646 packets that include this value, it can unambiguously locate the 2647 correct MFV and determine directionality without examining 2648 addresses. 2650 o "A2" - the A1 MFVI of a remote OAL node discovered by an FHS OAL 2651 source or OAL intermediate node when it processes an NA message 2652 with Job code "01" that originated from an LHS OAL source. A2 2653 values MUST NOT be tested for uniqueness within the OAL node's 2654 local context. 2656 o "B2" - the B1 MFVI of a remote OAL node discovered by an LHS OAL 2657 source or OAL intermediate node when it processes an NS message 2658 with Job code "00" that originated from an FHS OAL source. B2 2659 values MUST NOT be tested for uniqueness within the OAL node's 2660 local context. 2662 When an FHS OAL source has an original IP packet to send to an LHS 2663 OAL destination discovered via multilink address resolution, it first 2664 selects a source and target underlying interface pair. The OAL 2665 source uses its cached information for the target underlying 2666 interface as LHS information then prepares an NS message with an OMNI 2667 Multilink Forwarding Parameters sub-option with Job code "00" and 2668 with source set to its own {ADM,MNP}-LLA. If the LHS FMT-Forward and 2669 FMT-Mode bits are both clear, the OAL source sets the destination to 2670 the ADM-LLA of the LHS Proxy/Server; otherwise, it sets the 2671 destination to the MNP-LLA of the target Client. The OAL source then 2672 sets window synchronization information in the OMNI extension header 2673 and updates/creates a NCE for the selected destination {ADM,MNP}-LLA 2674 in the INCOMPLETE state. The OAL source next creates an MFV based on 2675 the NS source and destination LLAs, then generates a "B1" MFVI and 2676 assigns it to the MFV while also including it as the first B entry in 2677 the MFVI List. The OAL source then populates the NS Multilink 2678 Forwarding Parameters based on any FHS/LHS information it knows 2679 locally. OAL intermediate nodes on the path to the OAL destination 2680 may populate additional FHS/LHS information on a hop-by-hop basis. 2682 If the OAL source is the FHS Proxy/Server, it then performs OAL 2683 encapsulation/fragmentation while setting the source to its own ADM- 2684 ULA and setting the destination to the FHS Subnet Router Anycast ULA 2685 determined by applying the FHS SRT prefix length to its ADM-ULA. The 2686 FHS Proxy/Server next examines the LHS FMT code. If FMT-Forward is 2687 clear and FMT-Mode is set, the FHS Proxy/Server checks for a NCE for 2688 the ADM-LLA of the LHS Proxy/Server. If there is no NCE, the FHS 2689 Proxy/Server creates one in the INCOMPLETE state. If a new NCE was 2690 created (or if the existing NCE requires fresh window 2691 synchronization), the FHS Proxy/Server then writes window 2692 synchronization parameters into the OMNI Multilink Forwarding 2693 Parameters Tunnel Window Synchronization fields. The FHS Proxy/ 2694 Server then selects an appropriate Identification value and *NET 2695 headers and forwards the resulting carrier packets into the secured 2696 spanning tree which will deliver them to a Bridge interface that 2697 assigns the FHS Subnet Router Anycast ULA. 2699 If the OAL source is the FHS Client, it instead includes an 2700 authentication signature if necessary, performs OAL encapsulation, 2701 sets the source to its own MNP-ULA, sets the destination to 2702 {ADM,MNP}-ULA of the FHS Proxy/Server and selects an appropriate 2703 Identification value for the FHS Proxy/Server. If FHS FMT-Forward is 2704 set and LHS FMT-Forward is clear, the FHS Client creates/updates a 2705 NCE for the ADM-LLA of the LHS Proxy/Server as above and includes 2706 Tunnel Window Synchronization parameters. The FHS Client then 2707 fragments and encapsulates in appropriate *NET headers then forwards 2708 the carrier packets to the FHS Proxy/Server. When the FHS Proxy/ 2709 Server receives the carrier packets, it verifies the Identification, 2710 reassembles/decapsulates to obtain the NS then verifies the 2711 authentication signature or checksum. The FHS Proxy/Server then 2712 creates an MFV (i.e., the same as the FHS Client had done) while 2713 assigning the current B entry in the MFVI List (i.e., the one 2714 included by the FHS Client) as the "B2" MFVI for this MVF. The FHS 2715 Proxy/Server next generates a new unique "B1" MFVI, then both assigns 2716 it to the MFV and writes it as the next B entry in the OMNI Multilink 2717 Forwarding Parameters MFVI List (while also writing any FHS Client 2718 and Proxy/Server addressing information). The FHS Proxy/Server then 2719 checks FHS/LHS FMT-Forward/Mode to determine whether to create a NCE 2720 for the LHS Proxy/Server ADM-LLA and include Tunnel Window 2721 Synchronization parameters the same as above. The FHS Proxy/Server 2722 then calculates the checksum, re-fragments while setting the OAL 2723 source address to its own ADM-ULA and destination address to the FHS 2724 Subnet Router Anycast ULA, and includes an Identification appropriate 2725 for the secured spanning tree. The FHS Proxy/Server finally includes 2726 appropriate *NET headers and forwards the carrier packets into the 2727 secured spanning tree the same as above. 2729 Bridges in the spanning tree forward carrier packets not explicitly 2730 addressed to themselves, while forwarding those that arrived via the 2731 secured spanning tree to the next hop also via the secured spanning 2732 tree and forwarding all others via the unsecured spanning tree. When 2733 an FHS Bridge receives a carrier packet over the secured spanning 2734 tree addressed to its ADM-ULA or the FHS Subnet Router Anycast ULA, 2735 it instead reassembles/decapsulates to obtain the NS then verifies 2736 the checksum. The FHS Bridge next creates an MFV (i.e., the same as 2737 the FHS Proxy/Server had done) while assigning the current B entry in 2738 the MFVI List as the MFV "B2" index. The FHS Bridge also caches the 2739 NS Multilink Forwarding Parameters FHS information in the MFV, and 2740 also caches the first B entry in the MFVI List as "FHS-Client" when 2741 FHS FMT-Forward/Mode are both set to enable future direct forwarding 2742 to this FHS Client. The FHS Bridge then generates a "B1" MFVI for 2743 the MFV and also writes it as the next B entry in the NS's MFVI List. 2745 The FHS Bridge then examines the SRT prefixes corresponding to both 2746 FHS and LHS. If the FHS Bridge has a local interface connection to 2747 both the FHS and LHS (whether they are the same or different 2748 segments), the FHS/LHS Bridge caches the NS LHS information, writes 2749 its ADM-ULA suffix and LHS INADDR into the NS OMNI Multilink 2750 Forwarding Parameters LHS fields, then sets its own ADM-ULA as the 2751 source and the ADM-ULA of the LHS Proxy/Server as the destination 2752 while selecting an appropriate identification. If the FHS and LHS 2753 prefixes are different, the FHS Bridge instead sets the LHS Subnet 2754 Router Anycast ULA as the destination. The FHS Bridge then 2755 recalculates the NS checksum, selects an appropriate Identification 2756 and *NET headers as above then forwards the carrier packets into the 2757 secured spanning tree. 2759 When the FHS and LHS Bridges are different, the LHS Bridge will 2760 receive carrier packets over the secured spanning tree from the FHS 2761 Bridge. The LHS Bridge reassembles/decapsulates to obtain the NS 2762 then verifies the checksum and creates an MFV (i.e., the same as the 2763 FHS Bridge had done) while assigning the current B entry in the MFVI 2764 List as the MFV "B2" index. The LHS Bridge also caches the ADM-ULA 2765 of the FHS Bridge found in the Multilink Forwarding Parameters as the 2766 spanning tree address for "B2", caches the NS Multilink Forwarding 2767 Parameters LHS information then generates a "B1" MFVI for the MFV 2768 while also writing it as the next B entry in the MFVI List. The LHS 2769 Bridge also writes its own ADM-ULA suffix and LHS INADDR into the 2770 OMNI Multilink Forwarding Parameters. The LHS Bridge then sets the 2771 its own ADM-ULA as the source and the ADM-ULA of the LHS Proxy/Server 2772 as the OAL destination, recalculates the checksum, selects an 2773 appropriate Identification, then fragments while including 2774 appropriate *NET headers and forwards the carrier packets into the 2775 secured spanning tree. 2777 When the LHS Proxy/Server receives the carrier packets from the 2778 secured spanning tree, it reassembles/decapsulates to obtain the NS, 2779 verifies the checksum then verifies that the LHS information supplied 2780 by the FHS source is consistent with its own cached information. If 2781 the information is consistent, the LHS Proxy/Server then creates an 2782 MFV and assigns the current B entry in the MFVI List as the "B2" MFVI 2783 the same as for the prior hop. If the NS destination is the MNP-LLA 2784 of the target Client, the LHS Proxy/Server also generates a "B1" MFVI 2785 and assigns it both to the MFVI and as the next B entry in the MFVI 2786 List. The LHS Proxy/Server then examines FHS FMT; if FMT-Forward is 2787 clear and FMT-Mode is set, the LHS Proxy/Server creates a NCE for the 2788 ADM-LLA of the FHS Proxy/Server (if necessary) and sets the state to 2789 STALE, then caches any Tunnel Window Synchronization parameters. 2791 If the NS destination is its own ADM-LLA, the LHS Proxy/Server next 2792 prepares to return a solicited NA with Job code "01". If the NS 2793 source was the MNP-LLA of the FHS Client, the LHS Proxy/Server first 2794 creates or updates an NCE for the MNP-LLA with state set to STALE. 2795 The LHS Proxy/Server next caches the NS OMNI extension header window 2796 synchronization parameters and Multilink Forwarding Parameters 2797 information (including the MFVI List) in the NCE corresponding to the 2798 LLA source. When the LHS Proxy/Server forwards future carrier 2799 packets based on the NCE, it can populate reverse-path forwarding 2800 information in a CRH-32 routing header to enable forwarding based on 2801 the cached MFVI List B entries instead of ULA addresses. 2803 The LHS Proxy/Server then creates an NA with Job code "01" while 2804 copying the NS OMNI Multilink Forwarding Parameters FHS/LHS 2805 information into the corresponding fields in the NA. The LHS Proxy/ 2806 Server then generates an "A1" MFVI and both assigns it to the MFV and 2807 includes it as the first A entry in NA's MFVI List (see: 2808 [I-D.templin-6man-omni] for details on MFVI List A/B processing). 2809 The LHS Proxy/Server then includes end-to-end window synchronization 2810 parameters in the OMIN extension header (if necessary) and also 2811 tunnel window synchronization parameters in the Multilink Forwarding 2812 Parameters (if necessary). The LHS Proxy/Server then encapsulates 2813 the NA, calculates the checksum, sets the source to its own ADM-ULA, 2814 sets the destination to the ADM-ULA of the LHS Bridge, selects an 2815 appropriate Identification value and *NET headers then forwards the 2816 carrier packets into the secured spanning tree. 2818 If the NS destination was the MNP-LLA of the LHS Client, the LHS 2819 Proxy/Server instead includes an authentication signature in the NS 2820 if necessary (otherwise recalculates the checksum), then changes the 2821 OAL source to its own ADM-ULA and changes the destination to the MNP- 2822 ULA of the LHS Client. The LHS Proxy/Server then selects an 2823 appropriate Identification value, fragments if necessary, includes 2824 appropriate *NET headers and forwards the carrier packets to the LHS 2825 Client. When the LHS Client receives the carrier packets, it 2826 verifies the Identification and reassembles/decapsulates to obtain 2827 the NS then verifies the authentication signature or checksum. The 2828 LHS Client then creates a NCE for the NS LLA source address in the 2829 STALE state. If LHS FMT-Forward is set, FHS FMT-Forward is clear and 2830 the NS source was an MNP-LLA, the Client also creates a NCE for the 2831 ADM-LLA of the FHS Proxy/Server in the STALE state and caches any 2832 Tunnel Window Synchronization parameters. The Client then caches the 2833 NS OMNI extension header window synchronization parameters and 2834 Multilink Forwarding Parameters in the NCE corresponding to the NS 2835 LLA source, then creates an MFV and assigns both the current MFVI 2836 List B entry as "B2" and a locally generated "A1" MFVI the same as 2837 for previous hops (the LHS Client also includes the "A1" value in the 2838 solicited NA - see above and below). The LHS Client also caches the 2839 previous MFVI List B entry as "LHS-Bridge" since it can include this 2840 value when it sends future carrier packets directly to the Bridge 2841 (following appropriate neighbor coordination). 2843 The LHS Client then prepares an NA using exactly the same procedures 2844 as for the LHS Proxy/Server above, except that it uses its MNP-LLA as 2845 the source and the {ADM,MNP}-LLA of the FHS correspondent as the 2846 destination. The LHS Client also includes an authentication 2847 signature if necessary (otherwise calculates the checksum), then 2848 encapsulates the NA with OAL source set to its own MNP-ULA and 2849 destination set to the ADM-ULA of the LHS Proxy/Server, includes an 2850 appropriate Identification and *NET headers and forwards the carrier 2851 packets to the LHS Proxy/Server. When the LHS Proxy/Server receives 2852 the carrier packets, it verifies the Identifications, reassembles/ 2853 decapsulates to obtain the NA, verifies the authentication signature 2854 or checksum, then uses the current MVFI List B entry to locate the 2855 MFV. The LHS Proxy/Server then writes the current MFVI List A entry 2856 as the "A2" value for the MVF, generates an "A1" MFVI and both 2857 assigns it to the MFV and writes it as the next MFVI List A entry. 2858 The LHS Proxy/Server then examines the FHS/LHS FMT codes to determine 2859 if it needs to include Tunnel Window Synchronization parameters. The 2860 LHS Proxy/Server then recalculates the checksum, re-fragments the NA 2861 while setting the OAL source to its own ADM-ULA and destination to 2862 the ADM-ULA of the LHS Bridge, includes an appropriate Identification 2863 and *NET headers and forwards the carrier packets into the secured 2864 spanning tree. 2866 When the LHS Bridge receives the carrier packets, it reassembles/ 2867 decapsulates to obtain the NA while verifying the checksum then uses 2868 the current MFVI List B entry to locate the MFV. The LHS Bridge then 2869 writes the current MFVI List A entry as the MFV "A2" index and 2870 generates a new "A1" value which it both assigns the MFV and writes 2871 as the next MFVI List A entry. (The LHS Bridge also caches the first 2872 A entry in the MFVI List as "LHS-Client" when LHS FMT-Forward/Mode 2873 are both set to enable future direct forwarding to this LHS Client.) 2874 If the LHS Bridge is connected directly to both the FHS and LHS 2875 segments (whether the segments are the same or different), the FHS/ 2876 LHS Bridge will have already cached the FHS/LHS information based on 2877 the original NS. The FHS/LHS Bridge recalculates the checksum then 2878 re-fragments the NA while setting the OAL source to its own ADM-ULA 2879 and destination to the ADM-ULA of the FHS Proxy/Server. If the FHS 2880 and LHS prefixes are different, the FHS Bridge instead re-fragments 2881 while setting the destination to the ADM-ULA of the FHS Bridge. The 2882 LHS Bridge selects an appropriate Identification and *NET headers 2883 then forwards the carrier packets into the secured spanning tree. 2885 When the FHS and LHS Bridges are different, the FHS Bridge will 2886 receive the carrier packets from the LHS Bridge over the secured 2887 spanning tree. The FHS Bridge reassembles/decapsulates to obtain the 2888 NA while verifying the checksum, then locates the MFV based on the 2889 current MFVI List B entry. The FHS Bridge then assigns the current 2890 MFVI List A entry as the MFV "A2" index and caches the ADM-ULA of the 2891 LHS Bridge as the spanning tree address for "A2". The FHS Bridge 2892 then generates an "A1" MVFI and both assigns it to the MVF and writes 2893 it as the next MFVI List A entry while also writing its ADM-ULA and 2894 INADDR in the NA FHS Bridge fields. The FHS Bridge then recalculates 2895 the checksum, re-encapsulates/re-fragments with its own ADM-ULA as 2896 the source, with the ADM-ULA of the FHS Proxy/Server as the 2897 destination, then selects an appropriate Identification value and 2898 *NET headers and forwards the carrier packets into the secured 2899 spanning tree. 2901 When the FHS Proxy/Server receives the carrier packets from the 2902 secured spanning tree, it reassembles/decapsulates to obtain the NA 2903 while verifying the checksum then locates the MFV based on the 2904 current MFVI List B entry. The FHS Proxy/Server then assigns the 2905 current MFVI List A entry as the "A2" MFVI the same as for the prior 2906 hop. If the NA destination is its own ADM-LLA, the FHS Proxy/Server 2907 then caches the NA Multilink Forwarding Parameters with the MFV and 2908 examines LHS FMT. If FMT-Forward is clear, the FHS Proxy/Server 2909 locates the NCE for the ADM-LLA of the LHS Proxy/Server and sets the 2910 state to REACHABLE then caches any Tunnel Window Synchronization 2911 parameters. If the NA source is the MNP-LLA of the LHS Client, the 2912 FHS Proxy/Server then locates the LHS Client NCE and sets the state 2913 to REACHABLE then caches the OMNI extension header window 2914 synchronization parameters and prepares to return an NA 2915 acknowledgement, if necessary. 2917 If the NA destination is the MNP-LLA of the FHS Client, the FHS 2918 Proxy/Server also searches for and updates the NCE for the ADM-LLA of 2919 the LHS Proxy/Server if necessary the same as above. The FHS Proxy/ 2920 Server then generates an "A1" MFVI and assigns it both to the MFVI 2921 and as the next MFVI List A entry, then includes an authentication 2922 signature or checksum in the NA message. The FHS Proxy/Server then 2923 sets the OAL source to its own ADM-LA and sets the destination to the 2924 MNP-ULA of the FHS Client, then selects an appropriate Identification 2925 value and *NET headers and forwards the carrier packets to the FHS 2926 Client. 2928 When the FHS Client receives the carrier packets, it verifies the 2929 Identification, reassembles/decapsulates to obtain the NA, verifies 2930 the authentication signature or checksum, then locates the MFV based 2931 on the current MFVI List B entry. The FHS Client then assigns the 2932 current MFVI List A entry as the "A2" MFVI the same as for the prior 2933 hop. The FHS Client then caches the NA Multilink Forwarding 2934 Parameters (including the MFVI List) with the MFV and examines LHS 2935 FMT. If FMT-Forward is clear, the FHS Client locates the NCE for the 2936 ADM-LLA of the LHS Proxy/Server and sets the state to REACHABLE then 2937 caches any Tunnel Window Synchronization parameters. If the NA 2938 source is the MNP-LLA of the LHS Client, the FHS Proxy/Server then 2939 locates the LHS Client NCE and sets the state to REACHABLE then 2940 caches the OMNI extension header window synchronization parameters 2941 and prepares to return an NA acknowledgement, if necessary. The FHS 2942 Client also caches the previous MFVI List A entry as "FHS-Bridge" 2943 since it can include this value when it sends future carrier packets 2944 directly to the Bridge (following appropriate neighbor coordination). 2946 If either the FHS Client or FHS Proxy/Server needs to return an 2947 acknowledgement to complete window synchronization, it prepares a uNA 2948 message with an OMNI Multilink Forwarding Parameters sub-option with 2949 Job code set to "10" (Follow A; Record B) (note that this step is 2950 unnecessary when Rapid Commit route optimization is used per 2951 Section 3.13.3). The FHS node sets the source to its own {ADM,MNP}- 2952 LLA, sets the destination to the {ADM,MNP}-LLA of the LHS node then 2953 includes Tunnel Window Synchronization parameters if necessary. The 2954 FHS node next sets the MFVI List to the cached list of A entries 2955 received in the Job code "01" NA, but need not set any other FHS/LHS 2956 information. The FHS node then encapsulates the uNA message in an 2957 OAL header with its own {ADM,MNP}-ULA as the source. If the FHS node 2958 is the Client, it next sets the ADM-ULA of the FHS Proxy/Server as 2959 the OAL destination, includes an authentication signature or 2960 checksum, selects an appropriate Identification value and *NET 2961 headers and forwards the carrier packets to the FHS Proxy/Server. 2962 The FHS Proxy/Server then verifies the Identification, reassembles/ 2963 decapsulates, verifies the authentication signature or checksum, then 2964 uses the current MFVI List A entry to locate the MFV. The FHS Proxy/ 2965 Server then writes its "B1" MFVI as the next MFVI List B entry and 2966 determines whether it needs to include Tunnel Window Synchronization 2967 parameters the same as it had done when it forwarded the original NS. 2969 The FHS Proxy/Server recalculates the uNA checksum then re-fragments 2970 while setting its own ADM-ULA as the source and the ADM-ULA of the 2971 FHS Bridge as the destination, then selects an appropriate 2972 Identification and *NET headers and forwards the carrier packets into 2973 the secured spanning tree. When the FHS Bridge receives the carrier 2974 packets, it reassembles/decapsulates to obtain the uNA while 2975 verifying the checksum then uses the current MFVI List A entry to 2976 locate the MFV. The FHS Bridge then writes its "B1" MFVI as the next 2977 MFVI List B entry, then re-fragments while setting the OAL source and 2978 destination. If the FHS Bridge is also the LHS Bridge, it sets the 2979 ADM-ULA of the LHS Proxy/Server as the destination; otherwise it sets 2980 the ADM-ULA of the LHS Bridge. The FHS Bridge recalculates the 2981 checksum then selects an appropriate Identification and *NET headers, 2982 re-fragments/forwards the carrier packets into the secured spanning 2983 tree. If an LHS Bridge receives the carrier packets, it processes 2984 them exactly the same as the FHS Bridge had done while setting the 2985 carrier packet destination to the ADM-ULA of the LHS Proxy/Server. 2987 When the LHS Proxy/Server receives the carrier packets, it 2988 reassembles/decapsulates to obtain the uNA message while verifying 2989 the checksum. The LHS Proxy/Server then locates the MFV based on the 2990 current MFVI List A entry then determines whether it is a tunnel 2991 ingress the same as for the original NS. If it is a tunnel ingress, 2992 the LHS Proxy/Server updates the NCE for the tunnel far-end based on 2993 the Tunnel Window Synchronization parameters. If the uNA destination 2994 is its own ADM-LLA, the LHS Proxy/Server next updates the NCE for the 2995 source LLA based on the OMNI extension header window synchronization 2996 parameters and MAY compare the MVFI List to the version it had cached 2997 in the MFV based on the original NS. 2999 If the uNA destination is the MNP-LLA of the LHS Client, the LHS 3000 Proxy/Server instead writes its "B1" MFV as the next MFVI List B 3001 entry, includes an authentication signature or checksum, writes its 3002 own ADM-ULA as the source and the MNP-ULA of the Client as the 3003 destination then selects an appropriate Identification and *NET 3004 headers and forwards the resulting carrier packets to the LHS Client. 3005 When the LHS Client receives the carrier packets, it verifies the 3006 Identification, reassembles/decapsulates to obtain the uNA, verifies 3007 the authentication signature or checksum then processes the message 3008 exactly the same as for the LHS Proxy/Server case above. 3010 Following the NS/NA exchange with Multilink Forwarding Parameters, 3011 OAL end systems and tunnel endpoints can begin exchanging ordinary 3012 carrier packets with Identification values within their respective 3013 send/receive windows without requiring security signatures and/or 3014 secured spanning tree traversal. Either peer can refresh window 3015 synchronization parameters and/or send other carrier packets 3016 requiring security at any time using the same secured procedures 3017 described above. OAL end systems and intermediate nodes can also use 3018 their own A1/B1 MFVIs when they receive carrier packets to 3019 unambiguously locate the correct MFV and determine directionality and 3020 can use any discovered A2/B2 MFVIs to forward carrier packets to 3021 other OAL nodes that configure the corresponding A1/B1 MFVIs. When 3022 an OAL node uses an MFVI included in a carrier packet to locate an 3023 MFV, it need not also examine the carrier packet addresses. 3025 OAL sources can also begin including CRH-32s in carrier packets with 3026 a list of A/B MFVIs that OAL intermediate nodes can use for shortest- 3027 path carrier packet forwarding based on MFVIs instead of spanning 3028 tree addresses. OAL sources and intermediate nodes can also begin 3029 forwarding carrier packets with OAL compressed headers termed "OCH- 3030 0/1" (Type 0 or 1) (see: [I-D.templin-6man-omni]) that include only a 3031 single A/B MFVI meaningful to the next hop, since all nodes in the 3032 path up to (and sometimes including) the OAL destination have already 3033 established MFV forwarding information. Note that when an FHS OAL 3034 source receives a solicited NA with Job code "01', the message will 3035 contain an MFVI List with A entries populated in the reverse order 3036 needed for populating a CRH-32 routing header. The FHS OAL source 3037 must therefore write the MFVI List A entries last-to-first when it 3038 populates a CRH-32, or must select the correct A entry to include in 3039 an OCH-0/1 header based on the intended OAL intermediate node or 3040 destination. 3042 When a Bridge receives unsecured carrier packets destined to a local 3043 segment Client that has asserted direct reachability, the Bridge 3044 performs direct carrier packet forwarding while bypassing the local 3045 Proxy/Server based on the Client's advertised MFVIs and discovered 3046 NATed INADDR information (see: Section 3.13.4). If the Client cannot 3047 be reached directly (or if NAT traversal has not yet converged), the 3048 Bridge instead forwards carrier packets directly to the local Proxy/ 3049 Server. 3051 When a Proxy/Server receives carrier packets destined to a local 3052 Client or forwards carrier packets received from a local Client, it 3053 first locates the correct MFV. If the carrier packets include a 3054 secured IPv6 ND message, the Proxy/Server uses the Client's NCE 3055 established through RS/RA exchanges to re-encapsulate/re-fragment 3056 while forwarding outbound secured carrier packets via the secured 3057 spanning tree and forwarding inbound secured carrier packets while 3058 including an authentication signature or checksum. For ordinary 3059 carrier packets, the Proxy/Server uses the same MFV if directed by 3060 MFVI and/or OAL addressing. Otherwise it locates an MFV established 3061 through an NS/NA exchange between the Client and the remote peer, and 3062 forwards the carrier packets without first reassembling/ 3063 decapsulating. 3065 When a Proxy/Server or Client configured as a tunnel ingress receives 3066 a carrier packet with a full OAL header with an MNP-ULA source and 3067 CRH-32 routing header, or an OCH-0/1 header with an MFVI that matches 3068 an MFV, the ingress encapsulates the carrier packet in a new full OAL 3069 header or an OCH-0/1 header containing the next hop MVFI and an 3070 Identification value appropriate for the end-to-end window and the 3071 outer header containing an Identification value appropriate for the 3072 tunnel endpoints. When a Proxy/Server or Client configured as a 3073 tunnel egress receives an encapsulated carrier packet, it verifies 3074 the Identification in the outer header, then discards the outer 3075 header and forwards the inner carrier packet to the final 3076 destination. 3078 When a Proxy/Server with FMT-Forward/Mode set to 0/1 for a source 3079 Client receives carrier packets from the source Client, it first 3080 reassembles to obtain the original OAL packet then re-fragments if 3081 necessary to cause the Client's packets to match the MPS on the path 3082 from the Proxy/Server as a tunnel ingress to the tunnel egress. The 3083 Proxy/Server then performs OAL-in-OAL encapsulation and forwards the 3084 resulting carrier packets to the tunnel egress. When a Proxy/Server 3085 with FMT-Forward/Mode set to 0/1 for a target Client receives carrier 3086 packets from a tunnel ingress, it first decapsulates to obtain the 3087 original fragments then reassembles to obtain the original OAL 3088 packet. The Proxy/Server then re-fragments if necessary to cause the 3089 fragments to match the target Client's underlying interface (Path) 3090 MTU and forwards the resulting carrier packets to the target Client. 3092 When a source Client forwards carrier packets it can employ header 3093 compression according to the MFVIs established through an NS/NA 3094 exchange with a remote or local peer. When the source Client 3095 forwards to a remote peer, it can forward carrier packets to a local 3096 SRT Bridge (following the establishment of INADDR information) while 3097 bypassing the Proxy/Server (see: Section 3.13.4). When a target 3098 Client receives carrier packets that match a local MFV, the Client 3099 first verifies the Identification then decompresses the headers if 3100 necessary, reassembles if necessary to obtain the OAL packet then 3101 decapsulates and delivers the IP packet to upper layers. 3103 When synchronized peer Clients in the same SRT segment with FMT- 3104 Forward and FMT-Mode set discover each other's NATed INADDR 3105 addresses, they can exchange carrier packets directly with header 3106 compression using MFVIs discovered as above (see: Section 3.13.5). 3107 The FHS Client will have cached the A MFVI for the LHS Client, which 3108 will have cached the B MVFI for the FHS Client. 3110 After window synchronization state has been established, the ROS and 3111 ROR can begin forwarding carrier packets while performing additional 3112 NS/NA exchanges as above to update window state, register new 3113 interface pairs for optimized multilink forwarding and/or confirm 3114 reachability. The ROS sends carrier packets to the FHS Bridge 3115 discovered through the NS/NA exchange. The FHS Bridge then forwards 3116 the carrier packets over the unsecured spanning tree to the LHS 3117 Bridge, which forwards them via LHS encapsulation to the LHS Proxy/ 3118 Server or directly to the target Client itself. The target Client in 3119 turn sends packets to the ROS in the reverse direction while 3120 forwarding through the Bridges to minimize Proxy/Server load whenever 3121 possible. 3123 While the ROS continues to actively forward packets to the target 3124 Client, it is responsible for updating window synchronization state 3125 and per-interface reachability before expiration. Window 3126 synchronization state is shared by all underlying interfaces in the 3127 ROS' NCE that use the same destination LLA so that a single NS/NA 3128 exchange applies for all interfaces regardless of the specific 3129 interface used to conduct the exchange. However, the window 3130 synchronization exchange only confirms target Client reachability 3131 over the specific underlying interface pair. Reachability for other 3132 underlying interfaces that share the same window synchronization 3133 state must be determined individually using additional NS/NA 3134 messages. 3136 3.13.3. Rapid Commit Route Optimization 3138 When the ROR receives an NS(AR) with a set of Interface Attributes 3139 for the source Client, it can perform "rapid commit" by immediately 3140 invoking multilink route optimization as above instead of returning 3141 an NA(AR). In order to perform rapid commit, the ROR prepares a 3142 unicast NS message with an OMNI option with window synchronization 3143 information responsive to the NS(AR), with a Multilink Forwarding 3144 Parameters sub-option selected for a specific underlying interface 3145 pair and with Interface Attributes for all of the ROR's other 3146 underlying interfaces. The ROR can also include ordinary IP packets 3147 as OAL super-packet trailers to the NS message if it has immediate 3148 data to send to the ROS. The ROR then returns the NS to the ROS the 3149 same as for the NA(AR) case. 3151 When the NS message traverses the return path to the ROR, all 3152 intermediate nodes in the path establish state exactly the same as 3153 for an ordinary NS/NA multilink route optimization exchange. When 3154 the NS message arrives at the ROS, the window synchronization 3155 parameters confirm that the NS is taking the place of the NA(AR), 3156 thereby eliminating an extraneous message transmission and associated 3157 delay. The ROS then completes the route optimization by returning a 3158 responsive NA. 3160 Note: The ROS must accept unicast NS messages with an ACK matching 3161 the SYN included in the NS(AR) as an equivalent message replacement 3162 for the NA(AR). Address resolution and multilink forwarding 3163 coordination can therefore be coordinated in a single three-way 3164 handshake connection with minimal messaging and delay (i.e., as 3165 opposed to a four-message exchange). 3167 3.13.4. Client/Bridge Route Optimization 3169 Following multilink route optimization for specific underlying 3170 interface pairs, ROS/ROR Clients located on open INETs can invoke 3171 Client/Bridge route optimization to improve performance and reduce 3172 load and congestion on their respective FHS/LHS Proxy/Servers. To 3173 initiate Client/Bridge route optimization, the Client prepares an NS 3174 message with its own MNP-LLA address as the source and the ADM-LLA of 3175 its Bridge as the destination while creating a NCE for the Bridge if 3176 necessary. The NS message must be no larger than the minimum MPS and 3177 encapsulated as an atomic fragment. 3179 The Client then includes an Interface Attributes sub-option for its 3180 underlying interface as well as an authentication signature but does 3181 not include window synchronization parameters. The Client then 3182 performs OAL encapsulation with its own MNP-ULA as the source and the 3183 ADM-ULA of the Bridge as the destination while including a randomly- 3184 chosen Identification value, then performs *NET encapsulation on the 3185 atomic fragment and sends the resulting carrier packet directly to 3186 the Bridge. 3188 When the Bridge receives the carrier packet, it verifies the 3189 authentication signature then creates a NCE for the Client. The 3190 Bridge then caches the *NET encapsulation addresses (which may have 3191 been altered by one or more NATs on the path) as well as the 3192 Interface Attributes for this Client omIndex, and marks this Client 3193 underlying interface as "trusted". The Bridge then prepares an NA 3194 reply with its own ADM-LLA as the source and the MNP-LLA of the 3195 Client as the destination where the NA again must be no larger than 3196 the minimum MPS. 3198 The Bridge then echoes the Client's Interface Attributes, includes an 3199 Origin Indication with the Client's observed *NET addresses and 3200 includes an authentication signature. The Bridge then performs OAL 3201 encapsulation with its own ADM-ULA as the source and the MNP-ULA of 3202 the Client as the destination while using the same Identification 3203 value that appeared in the NS, then performs *NET encapsulation on 3204 the atomic fragment and sends the resulting carrier packet directly 3205 to the Client. 3207 When the Client receives the NA reply, it caches the carrier packet 3208 *NET source address information as the Bridge target address via this 3209 underlying interface while marking the interface as "trusted". The 3210 Client also caches the Origin Indication *NET address information as 3211 its own (external) source address for this underlying interface. 3213 After the Client and Bridge have established NCEs as well as 3214 "trusted" status for a particular underlying interface pair, each 3215 node can begin forwarding ordinary carrier packets intended for this 3216 multilink route optimization directly to one another while omitting 3217 the Proxy/Server from the forwarding path while the status is 3218 "trusted". The NS/NA messaging will have established the correct 3219 state in any NATs in the path so that NAT traversal is naturally 3220 supported. The Client and Bridge must maintain a timer that watches 3221 for activity on the path; if no carrier packets and/or NS/NA messages 3222 are sent or received over the path before NAT state is likely to have 3223 expired, the underlying interface pair status becomes "untrusted". 3225 Thereafter, when the Client forwards a carrier packet with an MFVI 3226 toward the Bridge as the next hop, the Client uses the MFVI for the 3227 Bridge (discovered during multilink route optimization) instead of 3228 the MFVI for its Proxy/Server; the Bridge will accept the packet from 3229 the Client if and only if the underlying interface status is trusted 3230 and if the MFVI is correct for the next hop toward the final 3231 destination. (The same is true in the reverse direction when the 3232 Bridge sends carrier packets directly to the Client.) 3234 Note that the Client and Bridge each maintain a single NCE, but that 3235 the NCE may aggregate multiple underlying interface pairs. Each 3236 underlying interface pair may use differing source and target *NET 3237 addresses according to NAT mappings, and the "trusted/untrusted" 3238 status of each pair must be tested independently. When no "trusted" 3239 pairs remain, the NCE is deleted. 3241 Note that the above method requires Bridges to participate in NS/NA 3242 message authentication signature application and verification. In an 3243 alternate approach, the Client could instead exchange NS/NA messages 3244 with authentication signatures via its Proxy/Server but addressed to 3245 the ADM-LLA of the Bridge, and the Proxy/Server and Bridge could 3246 relay the messages over the secured spanning tree. However, this 3247 would still require the Client to send additional messages toward the 3248 *NET address of the Bridge to populate NAT state; hence the savings 3249 in complexity for Bridges would result in increased message overhead 3250 for Clients. 3252 3.13.5. Client/Client Route Optimization 3254 When the ROS/ROR Clients are both located on the same SRT segment, 3255 Client-to-Client route optimization is possible following the 3256 establishment of any necessary state in NATs in the path. Both 3257 Clients will have already established state via their respective 3258 shared segment Proxy/Servers (and possibly also the shared segment 3259 Bridge) and can begin forwarding packets directly via NAT traversal 3260 while avoiding any Proxy/Server and/or Bridge hops. 3262 When the ROR/ROS Clients on the same SRT segment perform the initial 3263 NS/NA exchange to establish Multilink Forwarding state, they also 3264 include an Origin Indication (i.e., in addition to Multilink 3265 Forwarding Parameters) with the mapped addresses discovered during 3266 the RS/RA exchanges with their respective Proxy/Servers. After the 3267 MFV paths have been established, both Clients can begin sending 3268 packets via strict MFV paths while establishing a direct path for 3269 Client-to-Client route optimization. 3271 To establish the direct path, either Client (acting as the source) 3272 transmits a bubble to the mapped *NET address for the target Client 3273 which primes its local chain of NATs for reception of future packets 3274 from that *NET address (see: [RFC4380] and [I-D.templin-6man-omni]). 3275 The source Client then prepares an NS message with its own MNP-LLA as 3276 the source, with the MNP-LLA of the target as the destination and 3277 with an OMNI option with an Interface Attributes sub-option. The 3278 source Client then encapsulates the NS in an OAL header with its own 3279 MNP-ULA as the source, with the MNP-ULA of the target Client as the 3280 destination and with an in-window Identification for the target. The 3281 source Client then fragments and encapsulates in *NET headers 3282 addressed to its FHS Proxy/Server then forwards the resulting carrier 3283 packets to the Proxy/Server. 3285 When the FHS Proxy/Server receives the carrier packets, it re- 3286 encapsulates and forwards them as unsecured carrier packets according 3287 to MFV state where they will eventually arrive at the target Client 3288 which can verify that the identifications are within the acceptable 3289 window and reassemble if necessary. Following reassembly, the target 3290 Client prepares an NA message with its own MNP-LLA as the source, 3291 with the MNP-LLA of the source Client as the destination and with an 3292 OMNI option with an Interface Attributes sub-option. The target 3293 Client then encapsulates the NA in an OAL header with its own MNP-ULA 3294 as the source, with the MNP-ULA of the source Client as the 3295 destination and with an in-window Identification for the source 3296 Client. The target Client then fragments and encapsulates in *NET 3297 headers addressed to the source Client's Origin addresses then 3298 forwards the resulting carrier packets directly to the source Client. 3300 Following the initial NS/NA exchange, both Clients mark their 3301 respective (source, target) underlying interface pairs as "trusted" 3302 for no more than ReachableTime seconds. While the Clients continue 3303 to exchange carrier packets via the direct path avoiding all Proxy/ 3304 Servers and Bridges, they should perform additional NS/NA exchanges 3305 via their local Proxy/Servers to refresh NCE state as well as send 3306 additional bubbles to the peer's Origin address information if 3307 necessary to refresh NAT state. 3309 Note that these procedures are suitable for a widely-deployed but 3310 basic class of NATs. Procedures for advanced NAT classes are 3311 outlined in [RFC6081], which provides mechanisms that can be employed 3312 equally for AERO using the corresponding sub-options specified by 3313 OMNI. 3315 Note also that each communicating pair of Clients may need to 3316 maintain NAT state for peer to peer communications via multiple 3317 underlying interface pairs. It is therefore important that Origin 3318 Indications are maintained with the correct peer interface and that 3319 the NCE may cache information for multiple peer interfaces. 3321 Note that the source and target Client exchange Origin information 3322 during the secured NS/NA multilink route optimization exchange. This 3323 allows for subsequent NS/NA exchanges to proceed using only the 3324 Identification value as a data origin confirmation. However, Client- 3325 to-Client peerings that require stronger security may also include 3326 authentication signatures for mutual authentication. 3328 3.14. Neighbor Unreachability Detection (NUD) 3330 AERO nodes perform Neighbor Unreachability Detection (NUD) per 3331 [RFC4861] either reactively in response to persistent link-layer 3332 errors (see Section 3.11) or proactively to confirm reachability. 3333 The NUD algorithm is based on periodic control message exchanges and 3334 may further be seeded by IPv6 ND hints of forward progress, but care 3335 must be taken to avoid inferring reachability based on spoofed 3336 information. For example, IPv6 ND message exchanges that include 3337 authentication codes and/or in-window Identifications may be 3338 considered as acceptable hints of forward progress, while spurious 3339 random carrier packets should be ignored. 3341 AERO nodes can perform NS/NA(NUD) exchanges over the OMNI link 3342 secured spanning tree (i.e. the same as described above) to test 3343 reachability without risk of DoS attacks from nodes pretending to be 3344 a neighbor. These NS/NA(NUD) messages use the unicast LLAs and ULAs 3345 of the parties involved in the NUD test. When only reachability 3346 information is required without updating any other NCE state, AERO 3347 nodes can instead perform NS/NA(NUD) exchanges directly between 3348 neighbors without employing the secured spanning tree as long as they 3349 include in-window Identifications and either an authentication 3350 signature or checksum. 3352 After an ROR directs an ROS to a target neighbor with one or more 3353 link-layer addresses, either node may invoke multilink forwarding 3354 state initialization to establish authentic intermediate node state 3355 between specific underlying interface pairs which also tests their 3356 reachability. Thereafter, either node acting as the source may 3357 perform additional reachability probing through NS(NUD) messages over 3358 the SRT secured or unsecured spanning tree, or through NS(NUD) 3359 messages sent directly to an underlying interface of the target 3360 itself. While testing a target underlying interface, the source can 3361 optionally continue to forward carrier packets via alternate 3362 interfaces, maintain a small queue of carrier packets until target 3363 reachability is confirmed or include them as trailing data with the 3364 NS(NUD) in an OAL super-packet [I-D.templin-6man-omni]. 3366 NS(NUD) messages are encapsulated, fragmented and transmitted as 3367 carrier packets the same as for ordinary original IP data packets, 3368 however the encapsulated destinations are the LLA of the source and 3369 either the ADM-LLA of the LHS Proxy/Server or the MNP-LLA of the 3370 target itself. The source encapsulates the NS(NUD) message the same 3371 as described in Section 3.13.2 and includes an Interface Attributes 3372 sub-option with omIndex set to identify its underlying interface used 3373 for forwarding. The source then includes an in-window 3374 Identification, fragments the OAL packet and forwards the resulting 3375 carrier packets into the unsecured spanning tree, directly to the 3376 target if it is in the local segment or directly to a Bridge in the 3377 local segment. 3379 When the target receives the NS(NUD) carrier packets, it verifies 3380 that it has a NCE for this source and that the Identification is in- 3381 window, then submits the carrier packets for reassembly. The target 3382 then verifies the authentication signature or checksum, then searches 3383 for Interface Attributes in its NCE for the source that match the 3384 NS(NUD) for the NA(NUD) reply. The target then prepares the NA(NUD) 3385 with the source and destination LLAs reversed, encapsulates and sets 3386 the OAL source and destination, includes an Interface Attributes sub- 3387 option in the NA(NUD) to identify the omIndex of the underlying 3388 interface the NS(NUD) arrived on and sets the Target Address to the 3389 same value included in the NS(NUD). The target next sets the R flag 3390 to 1, the S flag to 1 and the O flag to 1, then selects an in-window 3391 Identification for the source and performs fragmentation. The node 3392 then forwards the carrier packets into the unsecured spanning tree, 3393 directly to the source if it is in the local segment or directly to a 3394 Bridge in the local segment. 3396 When the source receives the NA(NUD), it marks the target underlying 3397 interface tested as "trusted". Note that underlying interface states 3398 are maintained independently of the overall NCE REACHABLE state, and 3399 that a single NCE may have multiple target underlying interfaces in 3400 various "trusted/untrusted" states while the NCE state as a whole 3401 remains REACHABLE. 3403 3.15. Mobility Management and Quality of Service (QoS) 3405 AERO is a fully Distributed Mobility Management (DMM) service in 3406 which each Proxy/Server is responsible for only a small subset of the 3407 Clients on the OMNI link. This is in contrast to a Centralized 3408 Mobility Management (CMM) service where there are only one or a few 3409 network mobility collective entities for large Client populations. 3410 Clients coordinate with their associated FHS and Hub Proxy/Servers 3411 via RS/RA exchanges to maintain the DMM profile, and the AERO routing 3412 system tracks all current Client/Proxy/Server peering relationships. 3414 Hub Proxy/Servers provide a designated router service for their 3415 dependent Clients, while FHS Proxy/Servers provide a proxy conduit 3416 between the Client and both the Hub and OMNI link in general. 3417 Clients are responsible for maintaining neighbor relationships with 3418 their Proxy/Servers through periodic RS/RA exchanges, which also 3419 serves to confirm neighbor reachability. When a Client's underlying 3420 interface attributes change, the Client is responsible for updating 3421 the Hub Proxy/Server through new RS/RA exchanges using the FHS Proxy/ 3422 Server as a first-hop conduit. The FHS Proxy/Server can also act as 3423 a proxy to perform some IPv6 ND exchanges on the Client's behalf 3424 without consuming bandwidth on the Client underlying interface. 3426 Mobility management considerations are specified in the following 3427 sections. 3429 3.15.1. Mobility Update Messaging 3431 RORs and ROSs accommodate Client mobility and/or multilink change 3432 events by sending secured uNA messages to each active neighbor. When 3433 an ROR/ROS sends a uNA message, it sets the IPv6 source address to 3434 the its own LLA, sets the destination address to the neighbor's 3435 {ADM,MNP}-LLA and sets the Target Address to the Client's MNP-LLA. 3436 The ROR/ROS also includes an OMNI option with OMNI extension header 3437 Preflen set to the prefix length associated with the Client's MNP- 3438 LLA, includes Interface Attributes and Traffic Selectors for the 3439 Client's underlying interfaces and includes an authentication 3440 signature if necessary. The ROR then sets the uNA R flag to 1, S 3441 flag to 0 and O flag to 1, then encapsulates the message in an OAL 3442 header with source set to its own ULA and destination set to its FHS 3443 Proxy/Server's ADM-ULA. When the FHS Proxy/Server receives the uNA, 3444 it reassembles, verifies the authentication signature, then changes 3445 the destination to the ULA corresponding to the LLA destination and 3446 forwards the uNA into the secured spanning tree. 3448 As discussed in Section 7.2.6 of [RFC4861], the transmission and 3449 reception of uNA messages is unreliable but provides a useful 3450 optimization. In well-connected Internetworks with robust data links 3451 uNA messages will be delivered with high probability, but in any case 3452 the ROR/ROS can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT uNAs 3453 to each neighbor to increase the likelihood that at least one will be 3454 received. Alternatively, the ROR/ROS can set the PNG flag in the uNA 3455 OMNI option header to request a uNA acknowledgement as specified in 3456 [I-D.templin-6man-omni]. 3458 When the ROR/ROS Proxy/Server receives a uNA message prepared as 3459 above, if the uNA destination was its own ADM-LLA the Proxy/Server 3460 uses the included OMNI option information to update its NCE for the 3461 target but does not reset ReachableTime since the receipt of a uNA 3462 message does not provide confirmation that any forward paths to the 3463 target Client are working. If the destination was the MNP-LLA of the 3464 ROR/ROS Client, the Proxy/Server instead changes the OAL source to 3465 its own ADM-ULA, includes an authentication signature if necessary, 3466 and includes an in-window Identification for this Client. Finally, 3467 if the uNA message PNG flag was set, the node that processes the uNA 3468 returns a uNA acknowledgement as specified in 3469 [I-D.templin-6man-omni]. 3471 3.15.2. Announcing Link-Layer Information Changes 3473 When a Client needs to change its underlying Interface Attributes 3474 and/or Traffic Selectors (e.g., due to a mobility event), the Client 3475 sends an RS message to its Hub Proxy/Server via a first-hop FHS 3476 Proxy/Server, if necessary. The RS includes an OMNI option with an 3477 Interface Attributes sub-option with the omIndex and with new link 3478 quality and any other information. 3480 Note that the first FHS Proxy/Server may change due to the underlying 3481 interface change. If the Client supplies the address of the former 3482 FHS Proxy/Server, the new FHS Proxy/Server can send a departure 3483 indication (see below); otherwise, any stale state in the former FHS 3484 Proxy/Server will simply expire after ReachableTime expires with no 3485 effect on the Hub Proxy/Server. 3487 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 3488 sending carrier packets containing user data in case one or more RAs 3489 are lost. If all RAs are lost, the Client SHOULD re-associate with a 3490 new Proxy/Server. 3492 After performing the RS/RA exchange, the Client sends uNA messages to 3493 all neighbors the same as described in the previous section. 3495 3.15.3. Bringing New Links Into Service 3497 When a Client needs to bring new underlying interfaces into service 3498 (e.g., when it activates a new data link), it sends an RS message to 3499 the Hub Proxy/Server via a FHS Proxy/Server for the underlying 3500 interface (if necessary) with an OMNI option that includes an 3501 Interface Attributes sub-option with appropriate link quality values 3502 and with link-layer address information for the new link. The Client 3503 then again sends uNA messages to all neighbors the same as described 3504 above. 3506 3.15.4. Deactivating Existing Links 3508 When a Client needs to deactivate an existing underlying interface, 3509 it sends a uNA message toward the Hub Proxy/Server via an FHS Proxy/ 3510 Server with an OMNI option with appropriate Interface Attributes 3511 values for the deactivated link - in particular, the link quality 3512 value 0 assures that neighbors will cease to use the link. 3514 If the Client needs to send uNA messages over an underlying interface 3515 other than the one being deactivated, it MUST include Interface 3516 Attributes with appropriate link quality values for any underlying 3517 interfaces being deactivated. The Client then again sends uNA 3518 messages to all neighbors the same as described above. 3520 Note that when a Client deactivates an underlying interface, 3521 neighbors that receive the ensuing uNA messages need not purge all 3522 references for the underlying interface from their neighbor cache 3523 entries. The Client may reactivate or reuse the underlying interface 3524 and/or its omIndex at a later point in time, when it will send new RS 3525 messages to an FHS Proxy/Server with fresh interface parameters to 3526 update any neighbors. 3528 3.15.5. Moving Between Proxy/Servers 3530 The Client performs the procedures specified in Section 3.12.2 when 3531 it first associates with a new Hub Proxy/Server or renews its 3532 association with an existing Hub Proxy/Server. 3534 When a Client associates with a new Hub Proxy/Server, it sends RS 3535 messages to register its underlying interfaces with the new Hub while 3536 including the 32 least significant bits of the old Hub's ADM-LLA in 3537 the "Old Hub Proxy/Server MSID" field of a Proxy/Server Departure 3538 OMNI sub-option. When the new Hub Proxy/Server returns the RA 3539 message via the FHS Proxy/Server (acting as a Proxy), the FHS Proxy/ 3540 Server sends a uNA to the old Hub Proxy/Server (i.e., if the MSID is 3541 non-zero and different from its own). The uNA has the MNP-LLA of the 3542 Client as the source and the ADM-LLA of the old hub as the 3543 destination and with OMNI extension header Preflen set to 0. The FHS 3544 Proxy/Server encapsulates the uNA in an OAL header with the ADM-ULA 3545 of the new Hub as the source and the ADM-ULA of the old Hub as the 3546 destination, the fragments and sends the carrier packets via the 3547 secured spanning tree. 3549 When the old Hub Proxy/Server receives the uNA, it changes the 3550 Client's NCE state to DEPARTED, resets DepartTime and caches the new 3551 Hub Proxy/Server ADM-ULA. After a short delay (e.g., 2 seconds) the 3552 old Hub Proxy/Server withdraws the Client's MNP from the routing 3553 system. While in the DEPARTED state, the old Hub Proxy/Server 3554 forwards any carrier packets received via the secured spanning tree 3555 destined to the Client's MNP-ULA to the new Hub Proxy/Server's ADM- 3556 ULA. After DepartTime expires, the old Hub Proxy/Server deletes the 3557 Client's NCE. 3559 Mobility events may also cause a Client to change to a new FHS Proxy/ 3560 Server over a specific underlying interface at any time such that a 3561 Client RS/RA exchange over the underlying interface will engage the 3562 new FHS Proxy/Server instead of the old. The Client can arrange to 3563 inform the old FHS Proxy/Server of the departure by including a 3564 Proxy/Server Departure sub-option with an MSID for the "Old FHS 3565 Proxy/Server MSID", and the new FHS Proxy/Server will issue a uNA 3566 using the same procedures as outlined for the Hub above while using 3567 its own ADM-ULA as the source address. This can often result in 3568 successful delivery of packets that would otherwise be lost due to 3569 the mobility event. 3571 Clients SHOULD NOT move rapidly between Hub Proxy/Servers in order to 3572 avoid causing excessive oscillations in the AERO routing system. 3573 Examples of when a Client might wish to change to a different Hub 3574 Proxy/Server include a Hub Proxy/Server that has gone unreachable, 3575 topological movements of significant distance, movement to a new 3576 geographic region, movement to a new OMNI link segment, etc. 3578 3.16. Multicast 3580 Clients provide an IGMP (IPv4) [RFC2236] or MLD (IPv6) [RFC3810] 3581 proxy service for its EUNs and/or hosted applications [RFC4605] and 3582 act as a Protocol Independent Multicast - Sparse-Mode (PIM-SM, or 3583 simply "PIM") Designated Router (DR) [RFC7761] on the OMNI link. 3584 Proxy/Servers act as OMNI link PIM routers for Clients on ANET, VPNed 3585 or Direct interfaces, and Relays also act as OMNI link PIM routers on 3586 behalf of nodes on other links/networks. 3588 Clients on VPNed, Direct or ANET underlying interfaces for which the 3589 ANET has deployed native multicast services forward IGMP/MLD messages 3590 into the ANET. The IGMP/MLD messages may be further forwarded by a 3591 first-hop ANET access router acting as an IGMP/MLD-snooping switch 3592 [RFC4541], then ultimately delivered to an ANET Proxy/Server. The 3593 FHS Proxy/Server then acts as an ROS to send NS(AR) messages to an 3594 ROR for the multicast source. Clients on INET and ANET underlying 3595 interfaces without native multicast services instead send NS(AR) 3596 messages as an ROS to cause their FHS Proxy/Server forward the 3597 message to an ROR. When the ROR receives an NA(AR) response, it 3598 initiates PIM protocol messaging according to the Source-Specific 3599 Multicast (SSM) and Any-Source Multicast (ASM) operational modes as 3600 discussed in the following sections. 3602 3.16.1. Source-Specific Multicast (SSM) 3604 When an ROS "X" (i.e., either a Client or Proxy/Server) acting as PIM 3605 router receives a Join/Prune message from a node on its downstream 3606 interfaces containing one or more ((S)ource, (G)roup) pairs, it 3607 updates its Multicast Routing Information Base (MRIB) accordingly. 3608 For each S belonging to a prefix reachable via X's non-OMNI 3609 interfaces, X then forwards the (S, G) Join/Prune to any PIM routers 3610 on those interfaces per [RFC7761]. 3612 For each S belonging to a prefix reachable via X's OMNI interface, X 3613 sends an NS(AR) message (see: Section 3.13) using its own LLA as the 3614 source address, the solicited node multicast address corresponding to 3615 S as the destination and the LLA of S as the target address. X then 3616 encapsulates the NS(AR) in an OAL header with source address set to 3617 its own ULA and destination address set to the ULA for S, then 3618 forwards the message into the secured spanning tree which delivers it 3619 to ROR "Y" that services S. The resulting NA(AR) will return an OMNI 3620 option with Interface Attributes for any underlying interfaces that 3621 are currently servicing S. 3623 When X processes the NA(AR) it selects one or more underlying 3624 interfaces for S and performs an NS/NA multilink route optimization 3625 exchange over the secured spanning tree while including a PIM Join/ 3626 Prune message for each multicast group of interest in the OMNI 3627 option. If S is located behind any Proxys "Z"*, each Z* then updates 3628 its MRIB accordingly and maintains the LLA of X as the next hop in 3629 the reverse path. Since Bridges forward messages not addressed to 3630 themselves without examining them, this means that the (reverse) 3631 multicast tree path is simply from each Z* (and/or S) to X with no 3632 other multicast-aware routers in the path. 3634 Following the initial combined Join/Prune and NS/NA messaging, X 3635 maintains a NCE for each S the same as if X was sending unicast data 3636 traffic to S. In particular, X performs additional NS/NA exchanges 3637 to keep the NCE alive for up to t_periodic seconds [RFC7761]. If no 3638 new Joins are received within t_periodic seconds, X allows the NCE to 3639 expire. Finally, if X receives any additional Join/Prune messages 3640 for (S,G) it forwards the messages over the secured spanning tree. 3642 Client C that holds an MNP for source S may later depart from a first 3643 Proxy/Server Z1 and/or connect via a new Proxy/Server Z2. In that 3644 case, Y sends a uNA message to X the same as specified for unicast 3645 mobility in Section 3.15. When X receives the uNA message, it 3646 updates its NCE for the LLA for source S and sends new Join messages 3647 in NS/NA exchanges addressed to the new target Client underlying 3648 interface connection for S. There is no requirement to send any 3649 Prune messages to old Proxy/Server Z1 since source S will no longer 3650 source any multicast data traffic via Z1. Instead, the multicast 3651 state for (S,G) in Proxy/Server Z1 will soon expire since no new 3652 Joins will arrive. 3654 3.16.2. Any-Source Multicast (ASM) 3656 When an ROS X acting as a PIM router receives Join/Prune messages 3657 from a node on its downstream interfaces containing one or more (*,G) 3658 pairs, it updates its Multicast Routing Information Base (MRIB) 3659 accordingly. X first performs an NS/NA(AR) exchange to receive route 3660 optimization information for Rendezvous Point (RP) R for each G. X 3661 then includes a copy of each Join/Prune message in the OMNI option of 3662 an NS message with its own LLA as the source address and the LLA for 3663 R as the destination address, then encapsulates the NS message in an 3664 OAL header with its own ULA as the source and the ADM-ULA of R's 3665 Proxy/Server as the destination then sends the message into the 3666 secured spanning tree. 3668 For each source S that sends multicast traffic to group G via R, 3669 Client S* that aggregates S (or its Proxy/Server) encapsulates the 3670 original IP packets in PIM Register messages, includes the PIM 3671 Register messages in the OMNI options of uNA messages, performs OAL 3672 encapsulation and fragmentation then forwards the resulting carrier 3673 packets with Identification values within the receive window for 3674 Client R* that aggregates R. Client R* may then elect to send a PIM 3675 Join to S* in the OMNI option of a uNA over the secured spanning 3676 tree. This will result in an (S,G) tree rooted at S* with R as the 3677 next hop so that R will begin to receive two copies of the original 3678 IP packet; one native copy from the (S, G) tree and a second copy 3679 from the pre-existing (*, G) tree that still uses uNA PIM Register 3680 encapsulation. R can then issue a uNA PIM Register-stop message over 3681 the secured spanning tree to suppress the Register-encapsulated 3682 stream. At some later time, if Client S* moves to a new Proxy/ 3683 Server, it resumes sending original IP packets via uNA PIM Register 3684 encapsulation via the new Proxy/Server. 3686 At the same time, as multicast listeners discover individual S's for 3687 a given G, they can initiate an (S,G) Join for each S under the same 3688 procedures discussed in Section 3.16.1. Once the (S,G) tree is 3689 established, the listeners can send (S, G) Prune messages to R so 3690 that multicast original IP packets for group G sourced by S will only 3691 be delivered via the (S, G) tree and not from the (*, G) tree rooted 3692 at R. All mobility considerations discussed for SSM apply. 3694 3.16.3. Bi-Directional PIM (BIDIR-PIM) 3696 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 3697 approach to ASM that treats the Rendezvous Point (RP) as a Designated 3698 Forwarder (DF). Further considerations for BIDIR-PIM are out of 3699 scope. 3701 3.17. Operation over Multiple OMNI Links 3703 An AERO Client can connect to multiple OMNI links the same as for any 3704 data link service. In that case, the Client maintains a distinct 3705 OMNI interface for each link, e.g., 'omni0' for the first link, 3706 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 3707 would include its own distinct set of Bridges and Proxy/Servers, 3708 thereby providing redundancy in case of failures. 3710 Each OMNI link could utilize the same or different ANET connections. 3711 The links can be distinguished at the link-layer via the SRT prefix 3712 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 3713 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 3714 MSPs on each link. This gives rise to the opportunity for supporting 3715 multiple redundant networked paths (see: Section 3.2.4). 3717 The Client's IP layer can select the outgoing OMNI interface 3718 appropriate for a given traffic profile while (in the reverse 3719 direction) correspondent nodes must have some way of steering their 3720 original IP packets destined to a target via the correct OMNI link. 3722 In a first alternative, if each OMNI link services different MSPs the 3723 Client can receive a distinct MNP from each of the links. IP routing 3724 will therefore assure that the correct OMNI link is used for both 3725 outbound and inbound traffic. This can be accomplished using 3726 existing technologies and approaches, and without requiring any 3727 special supporting code in correspondent nodes or Bridges. 3729 In a second alternative, if each OMNI link services the same MSP(s) 3730 then each link could assign a distinct "OMNI link Anycast" address 3731 that is configured by all Bridges on the link. Correspondent nodes 3732 can then perform Segment Routing to select the correct SRT, which 3733 will then direct the original IP packet over multiple hops to the 3734 target. 3736 3.18. DNS Considerations 3738 AERO Client MNs and INET correspondent nodes consult the Domain Name 3739 System (DNS) the same as for any Internetworking node. When 3740 correspondent nodes and Client MNs use different IP protocol versions 3741 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 3742 A records for IPv4 address mappings to MNs which must then be 3743 populated in Relay NAT64 mapping caches. In that way, an IPv4 3744 correspondent node can send original IPv4 packets to the IPv4 address 3745 mapping of the target MN, and the Relay will translate the IPv4 3746 header and destination address into an IPv6 header and IPv6 3747 destination address of the MN. 3749 When an AERO Client registers with an AERO Proxy/Server, the Proxy/ 3750 Server can return the address(es) of DNS servers in RDNSS options 3751 [RFC6106]. The DNS server provides the IP addresses of other MNs and 3752 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 3754 3.19. Transition/Coexistence Considerations 3756 OAL encapsulation ensures that dissimilar INET partitions can be 3757 joined into a single unified OMNI link, even though the partitions 3758 themselves may have differing protocol versions and/or incompatible 3759 addressing plans. However, a commonality can be achieved by 3760 incrementally distributing globally routable (i.e., native) IP 3761 prefixes to eventually reach all nodes (both mobile and fixed) in all 3762 OMNI link segments. This can be accomplished by incrementally 3763 deploying AERO Bridges on each INET partition, with each Bridge 3764 distributing its MNPs and/or discovering non-MNP IP GUA prefixes on 3765 its INET links. 3767 This gives rise to the opportunity to eventually distribute native IP 3768 addresses to all nodes, and to present a unified OMNI link view even 3769 if the INET partitions remain in their current protocol and 3770 addressing plans. In that way, the OMNI link can serve the dual 3771 purpose of providing a mobility/multilink service and a transition/ 3772 coexistence service. Or, if an INET partition is transitioned to a 3773 native IP protocol version and addressing scheme that is compatible 3774 with the OMNI link MNP-based addressing scheme, the partition and 3775 OMNI link can be joined by Bridges. 3777 Relays that connect INETs/EUNs with dissimilar IP protocol versions 3778 may need to employ a network address and protocol translation 3779 function such as NAT64 [RFC6146]. 3781 3.20. Proxy/Server-Bridge Bidirectional Forwarding Detection 3783 In environments where rapid failure recovery is required, Proxy/ 3784 Servers and Bridges SHOULD use Bidirectional Forwarding Detection 3785 (BFD) [RFC5880]. Nodes that use BFD can quickly detect and react to 3786 failures so that cached information is re-established through 3787 alternate nodes. BFD control messaging is carried only over well- 3788 connected ground domain networks (i.e., and not low-end radio links) 3789 and can therefore be tuned for rapid response. 3791 Proxy/Servers and Bridges maintain BFD sessions in parallel with 3792 their BGP peerings. If a Proxy/Server or Bridge fails, BGP peers 3793 will quickly re-establish routes through alternate paths the same as 3794 for common BGP deployments. Similarly, Proxys maintain BFD sessions 3795 with their associated Bridges even though they do not establish BGP 3796 peerings with them. 3798 3.21. Time-Varying MNPs 3800 In some use cases, it is desirable, beneficial and efficient for the 3801 Client to receive a constant MNP that travels with the Client 3802 wherever it moves. For example, this would allow air traffic 3803 controllers to easily track aircraft, etc. In other cases, however 3804 (e.g., intelligent transportation systems), the MN may be willing to 3805 sacrifice a modicum of efficiency in order to have time-varying MNPs 3806 that can be changed every so often to defeat adversarial tracking. 3808 The DHCPv6 service offers a way for Clients that desire time-varying 3809 MNPs to obtain short-lived prefixes (e.g., on the order of a small 3810 number of minutes). In that case, the identity of the Client would 3811 not be bound to the MNP but rather to a Node Identification value 3812 (see: [I-D.templin-6man-omni]) to be used as the Client ID seed for 3813 MNP prefix delegation. The Client would then be obligated to 3814 renumber its internal networks whenever its MNP (and therefore also 3815 its MNP-LLA) changes. This should not present a challenge for 3816 Clients with automated network renumbering services, however presents 3817 limits for the durations of ongoing sessions that would prefer to use 3818 a constant address. 3820 4. Implementation Status 3822 An early AERO implementation based on OpenVPN (https://openvpn.net/) 3823 was announced on the v6ops mailing list on January 10, 2018 and an 3824 initial public release of the AERO proof-of-concept source code was 3825 announced on the intarea mailing list on August 21, 2015. 3827 Many AERO/OMNI functions are implemented and undergoing final 3828 integration. OAL fragmentation/reassembly buffer management code has 3829 been cleared for public release. 3831 5. IANA Considerations 3833 The IANA has assigned the UDP port number "8060" for an earlier 3834 experimental first version of AERO [RFC6706]. This document together 3835 with [I-D.templin-6man-omni] reclaims UDP port number "8060" as the 3836 service port for UDP/IP encapsulation. This document makes no 3837 request of IANA, since [I-D.templin-6man-omni] already provides 3838 instructions. (Note: although [RFC6706] was not widely implemented 3839 or deployed, it need not be obsoleted since its messages use the 3840 invalid ICMPv6 message type number '0' which implementations of this 3841 specification can easily distinguish and ignore.) 3843 No further IANA actions are required. 3845 6. Security Considerations 3847 AERO Bridges configure secured tunnels with AERO Proxy/Servers and 3848 Relays within their local OMNI link segments. Applicable secured 3849 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3850 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3851 segments in turn configure secured tunnels for their neighboring AERO 3852 Bridges in a secured spanning tree topology. Therefore, control 3853 messages exchanged between any pair of OMNI link neighbors over the 3854 secured spanning tree are already protected. 3856 To prevent spoofing vectors, Proxy/Servers MUST discard without 3857 responding to any unsecured NS/NA(AR) messages. Also, Proxy/Servers 3858 MUST discard without forwarding any original IP packets received from 3859 one of their own Clients (whether directly or following OAL 3860 reassembly) with a source address that does not match the Client's 3861 MNP and/or a destination address that does match the Client's MNP. 3862 Finally, Proxy/Servers MUST discard without forwarding any carrier 3863 packets with an OAL source and destination that both match the same 3864 MNP. 3866 For INET partitions that require strong security in the data plane, 3867 two options for securing communications include 1) disable route 3868 optimization so that all traffic is conveyed over secured tunnels, or 3869 2) enable on-demand secure tunnel creation between Client neighbors. 3870 Option 1) would result in longer routes than necessary and impose 3871 traffic concentration on critical infrastructure elements. Option 2) 3872 could be coordinated between Clients using NS/NA messages with OMNI 3873 Host Identity Protocol (HIP) "Initiator/Responder" message sub- 3874 options [RFC7401][I-D.templin-6man-omni] to create a secured tunnel 3875 on-demand, or to use the QUIC-TLS protocol to establish a secured 3876 connection [RFC9000][RFC9001][RFC9002]. 3878 AERO Clients that connect to secured ANETs need not apply security to 3879 their IPv6 ND messages, since the messages will be authenticated and 3880 forwarded by a perimeter Proxy/Server that applies security on its 3881 INET-facing interface as part of the secured spanning tree (see 3882 above). AERO Clients connected to the open INET can use network and/ 3883 or transport layer security services such as VPNs or can by some 3884 other means establish a direct link to a Proxy/Server. When a VPN or 3885 direct link may be impractical, however, INET Clients and Proxy/ 3886 Servers SHOULD include and verify authentication signatures for their 3887 IPv6 ND messages as specified in [I-D.templin-6man-omni]. 3889 Application endpoints SHOULD use transport-layer (or higher-layer) 3890 security services such as QUIC-TLS, TLS/SSL, DTLS or SSH [RFC4251] to 3891 assure the same level of protection as for critical secured Internet 3892 services. AERO Clients that require host-based VPN services SHOULD 3893 use network and/or transport layer security services such as IPsec, 3894 TLS/SSL, DTLS, etc. AERO Proxys and Proxy/Servers can also provide a 3895 network-based VPN service on behalf of the Client, e.g., if the 3896 Client is located within a secured enclave and cannot establish a VPN 3897 on its own behalf. 3899 AERO Proxy/Servers and Bridges present targets for traffic 3900 amplification Denial of Service (DoS) attacks. This concern is no 3901 different than for widely-deployed VPN security gateways in the 3902 Internet, where attackers could send spoofed packets to the gateways 3903 at high data rates. This can be mitigated through the AERO/OMNI data 3904 origin authentication procedures, as well as connecting Proxy/Servers 3905 and Bridges over dedicated links with no connections to the Internet 3906 and/or when connections to the Internet are only permitted through 3907 well-managed firewalls. Traffic amplification DoS attacks can also 3908 target an AERO Client's low data rate links. This is a concern not 3909 only for Clients located on the open Internet but also for Clients in 3910 secured enclaves. AERO Proxy/Servers and Proxys can institute rate 3911 limits that protect Clients from receiving packet floods that could 3912 DoS low data rate links. 3914 AERO Relays must implement ingress filtering to avoid a spoofing 3915 attack in which spurious messages with ULA addresses are injected 3916 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3917 that their connectivity is not used by unauthorized nodes on their 3918 EUNs to gain access to a protected network, i.e., AERO Clients that 3919 act as routers MUST NOT provide routing services for unauthorized 3920 nodes. (This concern is no different than for ordinary hosts that 3921 receive an IP address delegation but then "share" the address with 3922 other nodes via some form of Internet connection sharing such as 3923 tethering.) 3925 The PRL MUST be well-managed and secured from unauthorized tampering, 3926 even though the list contains only public information. The PRL can 3927 be conveyed to the Client in a similar fashion as in [RFC5214] (e.g., 3928 through layer 2 data link login messaging, secure upload of a static 3929 file, DNS lookups, etc.). 3931 The AERO service for open INET Clients depends on a public key 3932 distribution service in which Client public keys and identities are 3933 maintained in a shared database accessible to all open INET Proxy/ 3934 Servers. Similarly, each Client must be able to determine the public 3935 key of each Proxy/Server, e.g. by consulting an online database. 3936 When AERO nodes register their public keys indexed by a unique Host 3937 Identity Tag (HIT) [RFC7401] in a distributed database such as the 3938 DNS, and use the HIT as an identity for applying IPv6 ND message 3939 authentication signatures, a means for determining public key 3940 attestation is available. 3942 Security considerations for IPv6 fragmentation and reassembly are 3943 discussed in [I-D.templin-6man-omni]. In environments where spoofing 3944 is considered a threat, OMNI nodes SHOULD employ Identification 3945 window synchronization and OAL destinations SHOULD configure an (end- 3946 system-based) firewall. 3948 SRH authentication facilities are specified in [RFC8754]. Security 3949 considerations for accepting link-layer ICMP messages and reflected 3950 packets are discussed throughout the document. 3952 7. Acknowledgements 3954 Discussions in the IETF, aviation standards communities and private 3955 exchanges helped shape some of the concepts in this work. 3956 Individuals who contributed insights include Mikael Abrahamsson, Mark 3957 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Scott Burleigh, 3958 Brian Carpenter, Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian 3959 Farrel, Nick Green, Sri Gundavelli, Brian Haberman, Bernhard Haindl, 3960 Joel Halpern, Tom Herbert, Bob Hinden, Sascha Hlusiak, Lee Howard, 3961 Christian Huitema, Zdenek Jaron, Andre Kostur, Hubert Kuenig, Ted 3962 Lemon, Andy Malis, Satoru Matsushima, Tomek Mrugalski, Thomas Narten, 3963 Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal Skorepa, 3964 Dave Thaler, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, 3965 Lloyd Wood and James Woodyatt. Members of the IESG also provided 3966 valuable input during their review process that greatly improved the 3967 document. Special thanks go to Stewart Bryant, Joel Halpern and 3968 Brian Haberman for their shepherding guidance during the publication 3969 of the AERO first edition. 3971 This work has further been encouraged and supported by Boeing 3972 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3973 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3974 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3975 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3976 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3977 Gene MacLean III, Kyle Mikos, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3978 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3979 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3980 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3981 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3982 Benson, Madhuri Madhava Badgandi, Vijayasarathy Rajagopalan, Katie 3983 Tran and Eric Yeh are especially acknowledged for their work on the 3984 AERO implementation. Chuck Klabunde is honored and remembered for 3985 his early leadership, and we mourn his untimely loss. 3987 This work was inspired by the support and encouragement of countless 3988 outstanding colleagues, managers and program directors over the span 3989 of many decades. Beginning in the late 1980s,' the Digital Equipment 3990 Corporation (DEC) Ultrix Engineering and DECnet Architects groups 3991 identified early issues with fragmentation and bridging links with 3992 diverse MTUs. In the early 1990s, engagements at DEC Project Sequoia 3993 at UC Berkeley and the DEC Western Research Lab in Palo Alto included 3994 investigations into large-scale networked filesystems, ATM vs 3995 Internet and network security proxys. In the mid-1990s to early 3996 2000s employment at the NASA Ames Research Center (Sterling Software) 3997 and SRI International supported early investigations of IPv6, ONR UAV 3998 Communications and the IETF. An employment at Nokia where important 3999 IETF documents were published gave way to a present-day engagement 4000 with The Boeing Company. The work matured at Boeing through major 4001 programs including Future Combat Systems, Advanced Airplane Program, 4002 DTN for the International Space Station, Mobility Vision Lab, CAST, 4003 Caravan, Airplane Internet of Things, the NASA UAS/CNS program, the 4004 FAA/ICAO ATN/IPS program and many others. An attempt to name all who 4005 gave support and encouragement would double the current document size 4006 and result in many unintentional omissions - but to all a humble 4007 thanks. 4009 Earlier works on NBMA tunneling approaches are found in 4010 [RFC2529][RFC5214][RFC5569]. 4012 Many of the constructs presented in this second edition of AERO are 4013 based on the author's earlier works, including: 4015 o The Internet Routing Overlay Network (IRON) 4016 [RFC6179][I-D.templin-ironbis] 4018 o Virtual Enterprise Traversal (VET) 4019 [RFC5558][I-D.templin-intarea-vet] 4021 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 4022 [RFC5320][I-D.templin-intarea-seal] 4024 o AERO, First Edition [RFC6706] 4026 Note that these works cite numerous earlier efforts that are not also 4027 cited here due to space limitations. The authors of those earlier 4028 works are acknowledged for their insights. 4030 This work is aligned with the NASA Safe Autonomous Systems Operation 4031 (SASO) program under NASA contract number NNA16BD84C. 4033 This work is aligned with the FAA as per the SE2025 contract number 4034 DTFAWA-15-D-00030. 4036 This work is aligned with the Boeing Commercial Airplanes (BCA) 4037 Internet of Things (IoT) and autonomy programs. 4039 This work is aligned with the Boeing Information Technology (BIT) 4040 MobileNet program. 4042 8. References 4044 8.1. Normative References 4046 [I-D.templin-6man-omni] 4047 Templin, F. L. and T. Whyman, "Transmission of IP Packets 4048 over Overlay Multilink Network (OMNI) Interfaces", draft- 4049 templin-6man-omni-48 (work in progress), October 2021. 4051 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 4052 DOI 10.17487/RFC0791, September 1981, 4053 . 4055 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 4056 RFC 792, DOI 10.17487/RFC0792, September 1981, 4057 . 4059 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 4060 Requirement Levels", BCP 14, RFC 2119, 4061 DOI 10.17487/RFC2119, March 1997, 4062 . 4064 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 4065 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 4066 December 1998, . 4068 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 4069 "SEcure Neighbor Discovery (SEND)", RFC 3971, 4070 DOI 10.17487/RFC3971, March 2005, 4071 . 4073 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 4074 RFC 3972, DOI 10.17487/RFC3972, March 2005, 4075 . 4077 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 4078 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 4079 November 2005, . 4081 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 4082 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 4083 . 4085 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 4086 Network Address Translations (NATs)", RFC 4380, 4087 DOI 10.17487/RFC4380, February 2006, 4088 . 4090 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 4091 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 4092 DOI 10.17487/RFC4861, September 2007, 4093 . 4095 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 4096 Address Autoconfiguration", RFC 4862, 4097 DOI 10.17487/RFC4862, September 2007, 4098 . 4100 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 4101 DOI 10.17487/RFC6081, January 2011, 4102 . 4104 [RFC7401] Moskowitz, R., Ed., Heer, T., Jokela, P., and T. 4105 Henderson, "Host Identity Protocol Version 2 (HIPv2)", 4106 RFC 7401, DOI 10.17487/RFC7401, April 2015, 4107 . 4109 [RFC7739] Gont, F., "Security Implications of Predictable Fragment 4110 Identification Values", RFC 7739, DOI 10.17487/RFC7739, 4111 February 2016, . 4113 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 4114 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 4115 May 2017, . 4117 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 4118 (IPv6) Specification", STD 86, RFC 8200, 4119 DOI 10.17487/RFC8200, July 2017, 4120 . 4122 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 4123 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 4124 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 4125 RFC 8415, DOI 10.17487/RFC8415, November 2018, 4126 . 4128 8.2. Informative References 4130 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 4131 2016. 4133 [I-D.bonica-6man-comp-rtg-hdr] 4134 Bonica, R., Kamite, Y., Alston, A., Henriques, D., and L. 4135 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 4136 bonica-6man-comp-rtg-hdr-26 (work in progress), May 2021. 4138 [I-D.bonica-6man-crh-helper-opt] 4139 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 4140 Routing Header (CRH) Helper Option", draft-bonica-6man- 4141 crh-helper-opt-04 (work in progress), October 2021. 4143 [I-D.ietf-intarea-frag-fragile] 4144 Bonica, R., Baker, F., Huston, G., Hinden, R. M., Troan, 4145 O., and F. Gont, "IP Fragmentation Considered Fragile", 4146 draft-ietf-intarea-frag-fragile-17 (work in progress), 4147 September 2019. 4149 [I-D.ietf-intarea-tunnels] 4150 Touch, J. and M. Townsley, "IP Tunnels in the Internet 4151 Architecture", draft-ietf-intarea-tunnels-10 (work in 4152 progress), September 2019. 4154 [I-D.ietf-ipwave-vehicular-networking] 4155 (editor), J. (. J., "IPv6 Wireless Access in Vehicular 4156 Environments (IPWAVE): Problem Statement and Use Cases", 4157 draft-ietf-ipwave-vehicular-networking-24 (work in 4158 progress), October 2021. 4160 [I-D.ietf-rtgwg-atn-bgp] 4161 Templin, F. L., Saccone, G., Dawra, G., Lindem, A., and V. 4162 Moreno, "A Simple BGP-based Mobile Routing System for the 4163 Aeronautical Telecommunications Network", draft-ietf- 4164 rtgwg-atn-bgp-11 (work in progress), July 2021. 4166 [I-D.templin-6man-dhcpv6-ndopt] 4167 Templin, F. L., "A Unified Stateful/Stateless 4168 Configuration Service for IPv6", draft-templin-6man- 4169 dhcpv6-ndopt-11 (work in progress), January 2021. 4171 [I-D.templin-intarea-seal] 4172 Templin, F. L., "The Subnetwork Encapsulation and 4173 Adaptation Layer (SEAL)", draft-templin-intarea-seal-68 4174 (work in progress), January 2014. 4176 [I-D.templin-intarea-vet] 4177 Templin, F. L., "Virtual Enterprise Traversal (VET)", 4178 draft-templin-intarea-vet-40 (work in progress), May 2013. 4180 [I-D.templin-ipwave-uam-its] 4181 Templin, F. L., "Urban Air Mobility Implications for 4182 Intelligent Transportation Systems", draft-templin-ipwave- 4183 uam-its-04 (work in progress), January 2021. 4185 [I-D.templin-ironbis] 4186 Templin, F. L., "The Interior Routing Overlay Network 4187 (IRON)", draft-templin-ironbis-16 (work in progress), 4188 March 2014. 4190 [I-D.templin-v6ops-pdhost] 4191 Templin, F. L., "IPv6 Prefix Delegation and Multi- 4192 Addressing Models", draft-templin-v6ops-pdhost-27 (work in 4193 progress), January 2021. 4195 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 4197 [RFC1035] Mockapetris, P., "Domain names - implementation and 4198 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 4199 November 1987, . 4201 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 4202 RFC 1812, DOI 10.17487/RFC1812, June 1995, 4203 . 4205 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 4206 DOI 10.17487/RFC2003, October 1996, 4207 . 4209 [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, 4210 DOI 10.17487/RFC2004, October 1996, 4211 . 4213 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 4214 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 4215 . 4217 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 4218 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 4219 . 4221 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 4222 Domains without Explicit Tunnels", RFC 2529, 4223 DOI 10.17487/RFC2529, March 1999, 4224 . 4226 [RFC2983] Black, D., "Differentiated Services and Tunnels", 4227 RFC 2983, DOI 10.17487/RFC2983, October 2000, 4228 . 4230 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 4231 of Explicit Congestion Notification (ECN) to IP", 4232 RFC 3168, DOI 10.17487/RFC3168, September 2001, 4233 . 4235 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 4236 DOI 10.17487/RFC3330, September 2002, 4237 . 4239 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 4240 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 4241 DOI 10.17487/RFC3810, June 2004, 4242 . 4244 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 4245 Unique IDentifier (UUID) URN Namespace", RFC 4122, 4246 DOI 10.17487/RFC4122, July 2005, 4247 . 4249 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 4250 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 4251 January 2006, . 4253 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 4254 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 4255 DOI 10.17487/RFC4271, January 2006, 4256 . 4258 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 4259 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 4260 2006, . 4262 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 4263 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 4264 December 2005, . 4266 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 4267 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 4268 2006, . 4270 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 4271 Control Message Protocol (ICMPv6) for the Internet 4272 Protocol Version 6 (IPv6) Specification", STD 89, 4273 RFC 4443, DOI 10.17487/RFC4443, March 2006, 4274 . 4276 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 4277 Protocol (LDAP): The Protocol", RFC 4511, 4278 DOI 10.17487/RFC4511, June 2006, 4279 . 4281 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 4282 "Considerations for Internet Group Management Protocol 4283 (IGMP) and Multicast Listener Discovery (MLD) Snooping 4284 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 4285 . 4287 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 4288 "Internet Group Management Protocol (IGMP) / Multicast 4289 Listener Discovery (MLD)-Based Multicast Forwarding 4290 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 4291 August 2006, . 4293 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 4294 Algorithms in Cryptographically Generated Addresses 4295 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 4296 . 4298 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 4299 "Bidirectional Protocol Independent Multicast (BIDIR- 4300 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 4301 . 4303 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 4304 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 4305 DOI 10.17487/RFC5214, March 2008, 4306 . 4308 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 4309 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 4310 February 2010, . 4312 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 4313 Route Optimization Requirements for Operational Use in 4314 Aeronautics and Space Exploration Mobile Networks", 4315 RFC 5522, DOI 10.17487/RFC5522, October 2009, 4316 . 4318 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 4319 RFC 5558, DOI 10.17487/RFC5558, February 2010, 4320 . 4322 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 4323 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 4324 January 2010, . 4326 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 4327 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 4328 . 4330 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 4331 "IPv6 Router Advertisement Options for DNS Configuration", 4332 RFC 6106, DOI 10.17487/RFC6106, November 2010, 4333 . 4335 [RFC6139] Russert, S., Ed., Fleischman, E., Ed., and F. Templin, 4336 Ed., "Routing and Addressing in Networks with Global 4337 Enterprise Recursion (RANGER) Scenarios", RFC 6139, 4338 DOI 10.17487/RFC6139, February 2011, 4339 . 4341 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 4342 NAT64: Network Address and Protocol Translation from IPv6 4343 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 4344 April 2011, . 4346 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 4347 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 4348 . 4350 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 4351 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 4352 DOI 10.17487/RFC6221, May 2011, 4353 . 4355 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 4356 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 4357 DOI 10.17487/RFC6273, June 2011, 4358 . 4360 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 4361 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 4362 January 2012, . 4364 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 4365 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 4366 DOI 10.17487/RFC6355, August 2011, 4367 . 4369 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 4370 for Equal Cost Multipath Routing and Link Aggregation in 4371 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 4372 . 4374 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 4375 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 4376 . 4378 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 4379 UDP Checksums for Tunneled Packets", RFC 6935, 4380 DOI 10.17487/RFC6935, April 2013, 4381 . 4383 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 4384 for the Use of IPv6 UDP Datagrams with Zero Checksums", 4385 RFC 6936, DOI 10.17487/RFC6936, April 2013, 4386 . 4388 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 4389 Korhonen, "Requirements for Distributed Mobility 4390 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 4391 . 4393 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 4394 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 4395 Multicast - Sparse Mode (PIM-SM): Protocol Specification 4396 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 4397 2016, . 4399 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 4400 Decraene, B., Litkowski, S., and R. Shakir, "Segment 4401 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 4402 July 2018, . 4404 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 4405 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 4406 . 4408 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 4409 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 4410 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 4411 . 4413 [RFC9000] Iyengar, J., Ed. and M. Thomson, Ed., "QUIC: A UDP-Based 4414 Multiplexed and Secure Transport", RFC 9000, 4415 DOI 10.17487/RFC9000, May 2021, 4416 . 4418 [RFC9001] Thomson, M., Ed. and S. Turner, Ed., "Using TLS to Secure 4419 QUIC", RFC 9001, DOI 10.17487/RFC9001, May 2021, 4420 . 4422 [RFC9002] Iyengar, J., Ed. and I. Swett, Ed., "QUIC Loss Detection 4423 and Congestion Control", RFC 9002, DOI 10.17487/RFC9002, 4424 May 2021, . 4426 [WG] Wireguard, "WireGuard, https://www.wireguard.com", August 4427 2020. 4429 Appendix A. Non-Normative Considerations 4431 AERO can be applied to a multitude of Internetworking scenarios, with 4432 each having its own adaptations. The following considerations are 4433 provided as non-normative guidance: 4435 A.1. Implementation Strategies for Route Optimization 4437 Route optimization as discussed in Section 3.13 results in the 4438 creation of NCEs. The NCE state is set to REACHABLE for at most 4439 ReachableTime seconds. In order to refresh the NCE lifetime before 4440 the ReachableTime timer expires, the specification requires 4441 implementations to issue a new NS/NA(AR) exchange to reset 4442 ReachableTime while data packets are still flowing. However, the 4443 decision of when to initiate a new NS/NA(AR) exchange and to 4444 perpetuate the process is left as an implementation detail. 4446 One possible strategy may be to monitor the NCE watching for data 4447 packets for (ReachableTime - 5) seconds. If any data packets have 4448 been sent to the neighbor within this timeframe, then send an NS(AR) 4449 to receive a new NA(AR). If no data packets have been sent, wait for 4450 5 additional seconds and send an immediate NS(AR) if any data packets 4451 are sent within this "expiration pending" 5 second window. If no 4452 additional data packets are sent within the 5 second window, reset 4453 the NCE state to STALE. 4455 The monitoring of the neighbor data packet traffic therefore becomes 4456 an ongoing process during the NCE lifetime. If the NCE expires, 4457 future data packets will trigger a new NS/NA(AR) exchange while the 4458 packets themselves are delivered over a longer path until route 4459 optimization state is re-established. 4461 A.2. Implicit Mobility Management 4463 OMNI interface neighbors MAY provide a configuration option that 4464 allows them to perform implicit mobility management in which no IPv6 4465 ND messaging is used. In that case, the Client only transmits 4466 packets over a single interface at a time, and the neighbor always 4467 observes packets arriving from the Client from the same link-layer 4468 source address. 4470 If the Client's underlying interface address changes (either due to a 4471 readdressing of the original interface or switching to a new 4472 interface) the neighbor immediately updates the NCE for the Client 4473 and begins accepting and sending packets according to the Client's 4474 new address. This implicit mobility method applies to use cases such 4475 as cellphones with both WiFi and Cellular interfaces where only one 4476 of the interfaces is active at a given time, and the Client 4477 automatically switches over to the backup interface if the primary 4478 interface fails. 4480 A.3. Direct Underlying Interfaces 4482 When a Client's OMNI interface is configured over a Direct interface, 4483 the neighbor at the other end of the Direct link can receive packets 4484 without any encapsulation. In that case, the Client sends packets 4485 over the Direct link according to traffic selectors. If the Direct 4486 interface is selected, then the Client's IP packets are transmitted 4487 directly to the peer without going through an ANET/INET. If other 4488 interfaces are selected, then the Client's IP packets are transmitted 4489 via a different interface, which may result in the inclusion of 4490 Proxy/Servers and Bridges in the communications path. Direct 4491 interfaces must be tested periodically for reachability, e.g., via 4492 NUD. 4494 A.4. AERO Critical Infrastructure Considerations 4496 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 4497 IP routers or virtual machines in the cloud. Bridges must be 4498 provisioned, supported and managed by the INET administrative 4499 authority, and connected to the Bridges of other INETs via inter- 4500 domain peerings. Cost for purchasing, configuring and managing 4501 Bridges is nominal even for very large OMNI links. 4503 AERO INET Proxy/Servers can be standard dedicated server platforms, 4504 but most often will be deployed as virtual machines in the cloud. 4505 The only requirements for INET Proxy/Servers are that they can run 4506 the AERO/OMNI code and have at least one network interface connection 4507 to the INET. INET Proxy/Servers must be provisioned, supported and 4508 managed by the INET administrative authority. Cost for purchasing, 4509 configuring and managing cloud Proxy/Servers is nominal especially 4510 for virtual machines. 4512 AERO ANET Proxy/Servers are most often standard dedicated server 4513 platforms with one underlying interface connected to the ANET and a 4514 second interface connected to an INET. As with INET Proxy/Servers, 4515 the only requirements are that they can run the AERO/OMNI code and 4516 have at least one interface connection to the INET. ANET Proxy/ 4517 Servers must be provisioned, supported and managed by the ANET 4518 administrative authority. Cost for purchasing, configuring and 4519 managing Proxys is nominal, and borne by the ANET administrative 4520 authority. 4522 AERO Relays are simply Proxy/Servers connected to INETs and/or EUNs 4523 that provide forwarding services for non-MNP destinations. The Relay 4524 connects to the OMNI link and engages in eBGP peering with one or 4525 more Bridges as a stub AS. The Relay then injects its MNPs and/or 4526 non-MNP prefixes into the BGP routing system, and provisions the 4527 prefixes to its downstream-attached networks. The Relay can perform 4528 ROS/ROR services the same as for any Proxy/Server, and can route 4529 between the MNP and non-MNP address spaces. 4531 A.5. AERO Server Failure Implications 4533 AERO Proxy/Servers may appear as a single point of failure in the 4534 architecture, but such is not the case since all Proxy/Servers on the 4535 link provide identical services and loss of a Proxy/Server does not 4536 imply immediate and/or comprehensive communication failures. Proxy/ 4537 Server failure is quickly detected and conveyed by Bidirectional 4538 Forward Detection (BFD) and/or proactive NUD allowing Clients to 4539 migrate to new Proxy/Servers. 4541 If a Proxy/Server fails, ongoing packet forwarding to Clients will 4542 continue by virtue of the neighbor cache entries that have already 4543 been established in route optimization sources (ROSs). If a Client 4544 also experiences mobility events at roughly the same time the Proxy/ 4545 Server fails, uNA messages may be lost but neighbor cache entries in 4546 the DEPARTED state will ensure that packet forwarding to the Client's 4547 new locations will continue for up to DepartTime seconds. 4549 If a Client is left without a Proxy/Server for a considerable length 4550 of time (e.g., greater than ReachableTime seconds) then existing 4551 neighbor cache entries will eventually expire and both ongoing and 4552 new communications will fail. The original source will continue to 4553 retransmit until the Client has established a new Proxy/Server 4554 relationship, after which time continuous communications will resume. 4556 Therefore, providing many Proxy/Servers on the link with high 4557 availability profiles provides resilience against loss of individual 4558 Proxy/Servers and assurance that Clients can establish new Proxy/ 4559 Server relationships quickly in event of a Proxy/Server failure. 4561 A.6. AERO Client / Server Architecture 4563 The AERO architectural model is client / server in the control plane, 4564 with route optimization in the data plane. The same as for common 4565 Internet services, the AERO Client discovers the addresses of AERO 4566 Proxy/Servers and connects to one or more of them. The AERO service 4567 is analogous to common Internet services such as google.com, 4568 yahoo.com, cnn.com, etc. However, there is only one AERO service for 4569 the link and all Proxy/Servers provide identical services. 4571 Common Internet services provide differing strategies for advertising 4572 server addresses to clients. The strategy is conveyed through the 4573 DNS resource records returned in response to name resolution queries. 4574 As of January 2020 Internet-based 'nslookup' services were used to 4575 determine the following: 4577 o When a client resolves the domainname "google.com", the DNS always 4578 returns one A record (i.e., an IPv4 address) and one AAAA record 4579 (i.e., an IPv6 address). The client receives the same addresses 4580 each time it resolves the domainname via the same DNS resolver, 4581 but may receive different addresses when it resolves the 4582 domainname via different DNS resolvers. But, in each case, 4583 exactly one A and one AAAA record are returned. 4585 o When a client resolves the domainname "ietf.org", the DNS always 4586 returns one A record and one AAAA record with the same addresses 4587 regardless of which DNS resolver is used. 4589 o When a client resolves the domainname "yahoo.com", the DNS always 4590 returns a list of 4 A records and 4 AAAA records. Each time the 4591 client resolves the domainname via the same DNS resolver, the same 4592 list of addresses are returned but in randomized order (i.e., 4593 consistent with a DNS round-robin strategy). But, interestingly, 4594 the same addresses are returned (albeit in randomized order) when 4595 the domainname is resolved via different DNS resolvers. 4597 o When a client resolves the domainname "amazon.com", the DNS always 4598 returns a list of 3 A records and no AAAA records. As with 4599 "yahoo.com", the same three A records are returned from any 4600 worldwide Internet connection point in randomized order. 4602 The above example strategies show differing approaches to Internet 4603 resilience and service distribution offered by major Internet 4604 services. The Google approach exposes only a single IPv4 and a 4605 single IPv6 address to clients. Clients can then select whichever IP 4606 protocol version offers the best response, but will always use the 4607 same IP address according to the current Internet connection point. 4608 This means that the IP address offered by the network must lead to a 4609 highly-available server and/or service distribution point. In other 4610 words, resilience is predicated on high availability within the 4611 network and with no client-initiated failovers expected (i.e., it is 4612 all-or-nothing from the client's perspective). However, Google does 4613 provide for worldwide distributed service distribution by virtue of 4614 the fact that each Internet connection point responds with a 4615 different IPv6 and IPv4 address. The IETF approach is like google 4616 (all-or-nothing from the client's perspective), but provides only a 4617 single IPv4 or IPv6 address on a worldwide basis. This means that 4618 the addresses must be made highly-available at the network level with 4619 no client failover possibility, and if there is any worldwide service 4620 distribution it would need to be conducted by a network element that 4621 is reached via the IP address acting as a service distribution point. 4623 In contrast to the Google and IETF philosophies, Yahoo and Amazon 4624 both provide clients with a (short) list of IP addresses with Yahoo 4625 providing both IP protocol versions and Amazon as IPv4-only. The 4626 order of the list is randomized with each name service query 4627 response, with the effect of round-robin load balancing for service 4628 distribution. With a short list of addresses, there is still 4629 expectation that the network will implement high availability for 4630 each address but in case any single address fails the client can 4631 switch over to using a different address. The balance then becomes 4632 one of function in the network vs function in the end system. 4634 The same implications observed for common highly-available services 4635 in the Internet apply also to the AERO client/server architecture. 4636 When an AERO Client connects to one or more ANETs, it discovers one 4637 or more AERO Proxy/Server addresses through the mechanisms discussed 4638 in earlier sections. Each Proxy/Server address presumably leads to a 4639 fault-tolerant clustering arrangement such as supported by Linux-HA, 4640 Extended Virtual Synchrony or Paxos. Such an arrangement has 4641 precedence in common Internet service deployments in lightweight 4642 virtual machines without requiring expensive hardware deployment. 4643 Similarly, common Internet service deployments set service IP 4644 addresses on service distribution points that may relay requests to 4645 many different servers. 4647 For AERO, the expectation is that a combination of the Google/IETF 4648 and Yahoo/Amazon philosophies would be employed. The AERO Client 4649 connects to different ANET access points and can receive 1-2 Proxy/ 4650 Server ADM-LLAs at each point. It then selects one AERO Proxy/Server 4651 address, and engages in RS/RA exchanges with the same Proxy/Server 4652 from all ANET connections. The Client remains with this Proxy/Server 4653 unless or until the Proxy/Server fails, in which case it can switch 4654 over to an alternate Proxy/Server. The Client can likewise switch 4655 over to a different Proxy/Server at any time if there is some reason 4656 for it to do so. So, the AERO expectation is for a balance of 4657 function in the network and end system, with fault tolerance and 4658 resilience at both levels. 4660 Appendix B. Change Log 4662 << RFC Editor - remove prior to publication >> 4664 Changes from earlier versions to draft-templin-6man-aero-33: 4666 o New baseline version with corrections and section reorganizations 4667 to improve document flow. 4669 Author's Address 4671 Fred L. Templin (editor) 4672 Boeing Research & Technology 4673 P.O. Box 3707 4674 Seattle, WA 98124 4675 USA 4677 Email: fltemplin@acm.org