idnits 2.17.1 draft-templin-6man-aero-04.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 (May 15, 2021) is 1077 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Informational ---------------------------------------------------------------------------- == Missing Reference: 'ULA' is mentioned on line 1087, but not defined == Missing Reference: 'ULA0' is mentioned on line 1087, but not defined == Missing Reference: 'ULA1' is mentioned on line 1087, but not defined == Missing Reference: 'ULA2' is mentioned on line 1087, but not defined == Unused Reference: 'RFC3971' is defined on line 3657, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 3662, but no explicit reference was found in the text == Unused Reference: 'RFC4193' is defined on line 3670, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 3719, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-comp-rtg-hdr' is defined on line 3722, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-crh-helper-opt' is defined on line 3728, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 3733, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 3756, but no explicit reference was found in the text == Unused Reference: 'RFC2003' is defined on line 3795, but no explicit reference was found in the text == Unused Reference: 'RFC2983' is defined on line 3816, but no explicit reference was found in the text == Unused Reference: 'RFC3168' is defined on line 3820, but no explicit reference was found in the text == Unused Reference: 'RFC4122' is defined on line 3834, but no explicit reference was found in the text == Unused Reference: 'RFC4982' is defined on line 3883, but no explicit reference was found in the text == Unused Reference: 'RFC6139' is defined on line 3925, but no explicit reference was found in the text == Unused Reference: 'RFC6273' is defined on line 3945, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 3954, but no explicit reference was found in the text == Unused Reference: 'RFC6438' is defined on line 3959, but no explicit reference was found in the text == Unused Reference: 'RFC6935' is defined on line 3968, but no explicit reference was found in the text == Unused Reference: 'RFC6936' is defined on line 3973, but no explicit reference was found in the text == Outdated reference: A later version (-74) exists of draft-templin-6man-omni-03 == Outdated reference: A later version (-31) exists of draft-bonica-6man-comp-rtg-hdr-24 == Outdated reference: A later version (-04) exists of draft-bonica-6man-crh-helper-opt-03 == Outdated reference: A later version (-13) exists of draft-ietf-intarea-tunnels-10 == Outdated reference: A later version (-30) exists of draft-ietf-ipwave-vehicular-networking-20 == Outdated reference: A later version (-26) exists of draft-ietf-rtgwg-atn-bgp-10 -- Obsolete informational reference (is this intentional?): RFC 3330 (Obsoleted by RFC 5735) -- Obsolete informational reference (is this intentional?): RFC 6106 (Obsoleted by RFC 8106) -- Obsolete informational reference (is this intentional?): RFC 6347 (Obsoleted by RFC 9147) Summary: 0 errors (**), 0 flaws (~~), 30 warnings (==), 5 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group F. Templin, Ed. 3 Internet-Draft Boeing Research & Technology 4 Intended status: Informational May 15, 2021 5 Expires: November 16, 2021 7 Asymmetric Extended Route Optimization (AERO) 8 draft-templin-6man-aero-04 10 Abstract 12 This document specifies an Asymmetric 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 (ND) protocol 16 and links ND to IP forwarding. Prefix delegation/registration 17 services are employed for network admission and to manage the routing 18 system. Secure multilink operation, mobility management, multicast, 19 quality of service (QoS) signaling and route optimization are 20 naturally supported through dynamic neighbor cache updates. AERO is 21 a widely-applicable mobile internetworking service especially well- 22 suited to aviation services, intelligent transportation systems, 23 mobile Virtual Private Networks (VPNs) 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 November 16, 2021. 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. Asymmetric Extended Route Optimization (AERO) . . . . . . . . 12 62 3.1. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 12 63 3.2. The AERO Service over OMNI Links . . . . . . . . . . . . 13 64 3.2.1. AERO/OMNI Reference Model . . . . . . . . . . . . . . 13 65 3.2.2. Addressing and Node Identification . . . . . . . . . 16 66 3.2.3. AERO Routing System . . . . . . . . . . . . . . . . . 17 67 3.2.4. OMNI Link Segment Routing . . . . . . . . . . . . . . 19 68 3.2.5. Segment Routing Topologies (SRTs) . . . . . . . . . . 23 69 3.2.6. Segment Routing For OMNI Link Selection . . . . . . . 24 70 3.2.7. Segment Routing Within the OMNI Link . . . . . . . . 24 71 3.3. OMNI Interface Characteristics . . . . . . . . . . . . . 25 72 3.4. OMNI Interface Initialization . . . . . . . . . . . . . . 27 73 3.4.1. AERO Proxy/Server and Relay Behavior . . . . . . . . 27 74 3.4.2. AERO Client Behavior . . . . . . . . . . . . . . . . 27 75 3.4.3. AERO Bridge Behavior . . . . . . . . . . . . . . . . 28 76 3.5. OMNI Interface Neighbor Cache Maintenance . . . . . . . . 28 77 3.5.1. OMNI Neighbor Interface Attributes . . . . . . . . . 30 78 3.5.2. OMNI Neighbor Advertisement Message Flags . . . . . . 30 79 3.6. OMNI Interface Encapsulation and Re-encapsulation . . . . 31 80 3.7. OMNI Interface Decapsulation . . . . . . . . . . . . . . 31 81 3.8. OMNI Interface Data Origin Authentication . . . . . . . . 31 82 3.9. OMNI Interface MTU . . . . . . . . . . . . . . . . . . . 32 83 3.10. OMNI Interface Forwarding Algorithm . . . . . . . . . . . 33 84 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 34 85 3.10.2. Proxy/Server and Relay Forwarding Algorithm . . . . 35 86 3.10.3. Bridge Forwarding Algorithm . . . . . . . . . . . . 38 87 3.11. OMNI Interface Error Handling . . . . . . . . . . . . . . 39 88 3.12. AERO Router Discovery, Prefix Delegation and 89 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 42 90 3.12.1. AERO Service Model . . . . . . . . . . . . . . . . . 42 91 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 42 92 3.12.3. AERO Proxy/Server Behavior . . . . . . . . . . . . . 45 93 3.13. The AERO Proxy Function . . . . . . . . . . . . . . . . . 48 94 3.13.1. Detecting and Responding to Proxy/Server Failures . 50 95 3.13.2. Point-to-Multipoint Proxy/Server Coordination . . . 51 96 3.14. AERO Route Optimization . . . . . . . . . . . . . . . . . 52 97 3.14.1. Route Optimization Initiation . . . . . . . . . . . 52 98 3.14.2. Relaying the NS(AR) *NET Packet(s) . . . . . . . . . 53 99 3.14.3. Processing the NS(AR) and Sending the NA(AR) . . . . 54 100 3.14.4. Relaying the NA(AR) . . . . . . . . . . . . . . . . 56 101 3.14.5. Processing the NA(AR) . . . . . . . . . . . . . . . 56 102 3.14.6. Route Optimization Maintenance . . . . . . . . . . . 57 103 3.15. Neighbor Unreachability Detection (NUD) . . . . . . . . . 58 104 3.16. Mobility Management and Quality of Service (QoS) . . . . 60 105 3.16.1. Mobility Update Messaging . . . . . . . . . . . . . 60 106 3.16.2. Announcing Link-Layer Address and/or QoS Preference 107 Changes . . . . . . . . . . . . . . . . . . . . . . 62 108 3.16.3. Bringing New Links Into Service . . . . . . . . . . 62 109 3.16.4. Deactivating Existing Links . . . . . . . . . . . . 62 110 3.16.5. Moving Between Proxy/Servers . . . . . . . . . . . . 63 111 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 64 112 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 64 113 3.17.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 66 114 3.17.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 66 115 3.18. Operation over Multiple OMNI Links . . . . . . . . . . . 67 116 3.19. DNS Considerations . . . . . . . . . . . . . . . . . . . 67 117 3.20. Transition/Coexistence Considerations . . . . . . . . . . 68 118 3.21. Detecting and Reacting to Proxy/Server and Bridge 119 Failures . . . . . . . . . . . . . . . . . . . . . . . . 68 120 3.22. AERO Clients on the Open Internet . . . . . . . . . . . . 69 121 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 72 122 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 72 123 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 73 124 6. Security Considerations . . . . . . . . . . . . . . . . . . . 73 125 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 75 126 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 77 127 8.1. Normative References . . . . . . . . . . . . . . . . . . 77 128 8.2. Informative References . . . . . . . . . . . . . . . . . 78 129 Appendix A. Non-Normative Considerations . . . . . . . . . . . . 85 130 A.1. Implementation Strategies for Route Optimization . . . . 85 131 A.2. Implicit Mobility Management . . . . . . . . . . . . . . 85 132 A.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 86 133 A.4. AERO Critical Infrastructure Considerations . . . . . . . 86 134 A.5. AERO Server Failure Implications . . . . . . . . . . . . 87 135 A.6. AERO Client / Server Architecture . . . . . . . . . . . . 87 136 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 90 137 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 91 139 1. Introduction 141 Asymmetric Extended Route Optimization (AERO) fulfills the 142 requirements of Distributed Mobility Management (DMM) [RFC7333] and 143 route optimization [RFC5522] for aeronautical networking and other 144 network mobility use cases including intelligent transportation 145 systems and enterprise mobile device users. AERO is a secure 146 internetworking and mobility management service that employs the 147 Overlay Multilink Network Interface (OMNI) [I-D.templin-6man-omni] 148 Non-Broadcast, Multiple Access (NBMA) virtual link model. The OMNI 149 link is a virtual overlay configured over one or more underlying 150 Internetworks, and nodes on the link can exchange original IP packets 151 as single-hop neighbors. The OMNI Adaptation Layer (OAL) supports 152 end system multilink operation for increased reliability, bandwidth 153 optimization and traffic path selection while performing 154 fragmentation and reassembly to support Internetwork segment routing 155 and Maximum Transmission Unit (MTU) diversity. 157 The AERO service comprises Clients, Proxy/Servers and Relays that are 158 seen as OMNI link neighbors as well as Bridges that interconnect 159 diverse Internetworks as OMNI link segments through OAL forwarding at 160 a layer below IP. Each node's OMNI interface uses an IPv6 link-local 161 address format that supports operation of the IPv6 Neighbor Discovery 162 (ND) protocol [RFC4861] and links ND to IP forwarding. A node's OMNI 163 interface can be configured over multiple underlying interfaces, and 164 therefore appears as a single interface with multiple link-layer 165 addresses. Each link-layer address is subject to change due to 166 mobility and/or QoS fluctuations, and link-layer address changes are 167 signaled by ND messaging the same as for any IPv6 link. 169 AERO provides a secure cloud-based service where mobile node Clients 170 may use any Proxy/Server acting as a Mobility Anchor Point (MAP) and 171 fixed nodes may use any Relay on the link for efficient 172 communications. Fixed nodes forward original IP packets destined to 173 other AERO nodes via the nearest Relay, which forwards them through 174 the cloud. A mobile node's initial packets are forwarded through the 175 Proxy/Server, and direct routing is supported through route 176 optimization while packets are flowing. Both unicast and multicast 177 communications are supported, and mobile nodes may efficiently move 178 between locations while maintaining continuous communications with 179 correspondents and without changing their IP Address. 181 AERO Bridges are interconnected in a secured private BGP overlay 182 routing instance to provide an OAL routing/bridging service that 183 joins the underlying Internetworks of multiple disjoint 184 administrative domains into a single unified OMNI link at a layer 185 below IP. Each OMNI link instance is characterized by the set of 186 Mobility Service Prefixes (MSPs) common to all mobile nodes. Relays 187 provide an optimal route from correspondent nodes on the underlying 188 Internetwork to nodes on the OMNI link. To the underlying 189 Internetwork, the Relay is the source of a route to the MSP, and 190 hence uplink traffic to the mobile node is naturally routed to the 191 nearest Relay. 193 AERO can be used with OMNI links that span private-use Internetworks 194 and/or public Internetworks such as the global Internet. In the 195 latter case, some end systems may be located behind global Internet 196 Network Address Translators (NATs). A means for robust traversal of 197 NATs while avoiding "triangle routing" is therefore provided. 199 AERO assumes the use of PIM Sparse Mode in support of multicast 200 communication. In support of Source Specific Multicast (SSM) when a 201 Mobile Node is the source, AERO route optimization ensures that a 202 shortest-path multicast tree is established with provisions for 203 mobility and multilink operation. In all other multicast scenarios 204 there are no AERO dependencies. 206 AERO was designed as a secure aeronautical internetworking service 207 for both manned and unmanned aircraft, where the aircraft is treated 208 as a mobile node that can connect an Internet of Things (IoT). AERO 209 is also applicable to a wide variety of other use cases. For 210 example, it can be used to coordinate the links of mobile nodes 211 (e.g., cellphones, tablets, laptop computers, etc.) that connect into 212 a home enterprise network via public access networks using tunneling 213 software such as OpenVPN [OVPN] with VPN or non-VPN services enabled 214 according to the appropriate security model. AERO can also be used 215 to facilitate terrestrial vehicular and urban air mobility (as well 216 as pedestrian communication services) for future intelligent 217 transportation systems 218 [I-D.ietf-ipwave-vehicular-networking][I-D.templin-ipwave-uam-its]. 219 Other applicable use cases are also in scope. 221 Along with OMNI, AERO provides secured optimal routing support for 222 the "6M's" of modern Internetworking, including: 224 1. Multilink - a mobile node's ability to coordinate multiple 225 diverse underlying data links as a single logical unit (i.e., the 226 OMNI interface) to achieve the required communications 227 performance and reliability objectives. 229 2. Multinet - the ability to span the OMNI link across multiple 230 diverse network administrative segments while maintaining 231 seamless end-to-end communications between mobile nodes and 232 correspondents such as air traffic controllers, fleet 233 administrators, etc. 235 3. Mobility - a mobile node's ability to change network points of 236 attachment (e.g., moving between wireless base stations) which 237 may result in an underlying interface address change, but without 238 disruptions to ongoing communication sessions with peers over the 239 OMNI link. 241 4. Multicast - the ability to send a single network transmission 242 that reaches multiple nodes belonging to the same interest group, 243 but without disturbing other nodes not subscribed to the interest 244 group. 246 5. Multihop - a mobile node vehicle-to-vehicle relaying capability 247 useful when multiple forwarding hops between vehicles may be 248 necessary to "reach back" to an infrastructure access point 249 connection to the OMNI link. 251 6. MTU assurance - the ability to deliver packets of various robust 252 sizes between peers without loss due to a link size restriction, 253 and to dynamically adjust packets sizes to achieve the optimal 254 performance for each independent traffic flow. 256 The following numbered sections present the AERO specification. The 257 appendices at the end of the document are non-normative. 259 2. Terminology 261 The terminology in the normative references applies; especially, the 262 terminology in the OMNI specification [I-D.templin-6man-omni] is used 263 extensively throughout. The following terms are defined within the 264 scope of this document: 266 IPv6 Neighbor Discovery (ND) 267 a control message service for coordinating neighbor relationships 268 between nodes connected to a common link. AERO uses the IPv6 ND 269 messaging service specified in [RFC4861]. 271 IPv6 Prefix Delegation 272 a networking service for delegating IPv6 prefixes to nodes on the 273 link. The nominal service is DHCPv6 [RFC8415], however alternate 274 services (e.g., based on ND messaging) are also in scope. Most 275 notably, a minimal form of prefix delegation known as "prefix 276 registration" can be used if the Client knows its prefix in 277 advance and can represent it in the IPv6 source address of an ND 278 message. 280 Access Network (ANET) 281 a node's first-hop data link service network (e.g., a radio access 282 network, cellular service provider network, corporate enterprise 283 network, etc.) that often provides link-layer security services 284 such as IEEE 802.1X and physical-layer security (e.g., "protected 285 spectrum") to prevent unauthorized access internally and with 286 border network-layer security services such as firewalls and 287 proxys that prevent unauthorized outside access. 289 ANET interface 290 a node's attachment to a link in an ANET. 292 Internetwork (INET) 293 a connected IP network topology with a coherent routing and 294 addressing plan and that provides a transit backbone service for 295 ANET end systems. INETs also provide an underlay service over 296 which the AERO virtual link is configured. Example INETs include 297 corporate enterprise networks, aviation networks, and the public 298 Internet itself. When there is no administrative boundary between 299 an ANET and the INET, the ANET and INET are one and the same. 301 INET interface 302 a node's attachment to a link in an INET. 304 *NET 305 a "wildcard" term referring to either ANET or INET when it is not 306 necessary to draw a distinction between the two. 308 *NET interface 309 a node's attachment to a link in a *NET. 311 *NET Partition 312 frequently, *NETs such as large corporate enterprise networks are 313 sub-divided internally into separate isolated partitions (a 314 technique also known as "network segmentation"). Each partition 315 is fully connected internally but disconnected from other 316 partitions, and there is no requirement that separate partitions 317 maintain consistent Internet Protocol and/or addressing plans. 318 (Each *NET partition is seen as a separate OMNI link segment as 319 discussed below.) 321 *NET address 322 an IP address assigned to a node's interface connection to a *NET. 324 *NET encapsulation 325 the encapsulation of a packet in an outer header or headers that 326 can be routed within the scope of the local *NET partition. 328 OMNI link 329 the same as defined in [I-D.templin-6man-omni], and manifested by 330 IPv6 encapsulation [RFC2473]. The OMNI link spans underlying *NET 331 segments joined by virtual bridges in a spanning tree the same as 332 a bridged campus LAN. AERO nodes on the OMNI link appear as 333 single-hop neighbors even though they may be separated by multiple 334 underlying *NET hops, and can use Segment Routing [RFC8402] to 335 cause packets to visit selected waypoints on the link. 337 OMNI Interface 338 a node's attachment to an OMNI link. Since the addresses assigned 339 to an OMNI interface are managed for uniqueness, OMNI interfaces 340 do not require Duplicate Address Detection (DAD) and therefore set 341 the administrative variable 'DupAddrDetectTransmits' to zero 342 [RFC4862]. 344 OMNI Adaptation Layer (OAL) 345 an OMNI interface process whereby original IP packets admitted 346 into the interface are wrapped in a mid-layer IPv6 header and 347 subject to fragmentation and reassembly. The OAL is also 348 responsible for generating MTU-related control messages as 349 necessary, and for providing addressing context for spanning 350 multiple segments of a bridged OMNI link. 352 original IP packet 353 a whole IP packet or fragment admitted into the OMNI interface by 354 the network layer prior to OAL encapsulation and fragmentation, or 355 an IP packet delivered to the network layer by the OMNI interface 356 following OAL decapsulation and reassembly. 358 OAL packet 359 an original IP packet encapsulated in OAL headers and trailers 360 before OAL fragmentation, or following OAL reassembly. 362 OAL fragment 363 a portion of an OAL packet following fragmentation but prior to 364 *NET encapsulation, or following *NET encapsulation but prior to 365 OAL reassembly. 367 (OAL) atomic fragment 368 an OAL packet that does not require fragmentation is always 369 encapsulated as an "atomic fragment" with a Fragment Header with 370 Fragment Offset and More Fragments both set to 0, but with a valid 371 Identification value. 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 may 378 perform re-encapsulation on carrier packets by removing the *NET 379 headers of the first hop network and replacing them with new *NET 380 headers for the next hop 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 on a first segment, 395 then re-encapsulates the carrier packets in new *NET headers and 396 forwards them into the next segment. OAL intermediate nodes 397 decrement the Hop Limit of the OAL IPv6 header during re- 398 encapsulation, and discard the packet if the Hop Limit reaches 0. 400 underlying interface 401 a *NET interface over which an OMNI interface is configured. 403 Mobility Service Prefix (MSP) 404 an aggregated IP Global Unicast Address (GUA) prefix (e.g., 405 2001:db8::/32, 192.0.2.0/24, etc.) assigned to the OMNI link and 406 from which more-specific Mobile Network Prefixes (MNPs) are 407 delegated. OMNI link administrators typically obtain MSPs from an 408 Internet address registry, however private-use prefixes can 409 alternatively be used subject to certain limitations (see: 410 [I-D.templin-6man-omni]). OMNI links that connect to the global 411 Internet advertise their MSPs to their interdomain routing peers. 413 Mobile Network Prefix (MNP) 414 a longer IP prefix delegated from an MSP (e.g., 415 2001:db8:1000:2000::/56, 192.0.2.8/30, etc.) and delegated to an 416 AERO Client or Relay. 418 Mobile Network Prefix Link Local Address (MNP-LLA) 419 an IPv6 Link Local Address that embeds the most significant 64 420 bits of an MNP in the lower 64 bits of fe80::/64, as specified in 421 [I-D.templin-6man-omni]. 423 Mobile Network Prefix Unique Local Address (MNP-ULA) 424 an IPv6 Unique-Local Address derived from an MNP-LLA. 426 Administrative Link Local Address (ADM-LLA) 427 an IPv6 Link Local Address that embeds a 32-bit administratively- 428 assigned identification value in the lower 32 bits of fe80::/96, 429 as specified in [I-D.templin-6man-omni]. 431 Administrative Unique Local Address (ADM-ULA) 432 an IPv6 Unique-Local Address derived from an ADM-LLA. 434 AERO node 435 a node that is connected to an OMNI link and participates in the 436 AERO internetworking and mobility service. 438 AERO Client ("Client") 439 an AERO node that connects over one or more underlying interfaces 440 and requests MNP delegation/registration service from AERO Proxy/ 441 Servers. The Client assigns an MNP-LLA to the OMNI interface for 442 use in ND exchanges with other AERO nodes and forwards original IP 443 packets to correspondents according to OMNI interface neighbor 444 cache state. 446 AERO Proxy/Server ("Proxy/Server") 447 a dual-function node that provides a proxying service between AERO 448 Clients and external peers on its Client-facing ANET interfaces 449 (i.e., in the same fashion as for an enterprise network proxy) as 450 well as default forwarding and Mobility Anchor Point (MAP) 451 services for coordination with correspondents on its INET-facing 452 interfaces (Proxy/Servers in the open Internetwork instead have a 453 single INET interface). The Proxy/Server configures an OMNI 454 interface and assigns an ADM-LLA to support the operation of IPv6 455 ND services, while advertising all of its associated MNPs via BGP 456 peerings with Bridges. Note that the Proxy and Server functions 457 can be considered logically separable, but since each Proxy/Server 458 must be informed of all of the Client's other multilink Proxy/ 459 Server affiliations the AERO service is best supported when the 460 two functions are coresident on the same physical or logical 461 platform. 463 AERO Relay ("Relay") 464 a Proxy/Server that provides forwarding services between nodes 465 reached via the OMNI link and correspondents on connected 466 downstream links. AERO Relays configure an OMNI interface and 467 assign an ADM-LLA the same as Proxy/Servers. AERO Relays also run 468 a dynamic routing protocol to discover any non-MNP IP GUA routes 469 in service on its connected downstream network links. In both 470 cases, the Relay advertises the MSP(s) to its downstream networks, 471 and distributes all of its associated non-MNP IP GUA routes via 472 BGP peerings with Bridges (i.e., the same as for Proxy/Servers). 474 AERO Bridge ("Bridge") 475 a node that provides hybrid routing/bridging services (as well as 476 a security trust anchor) for nodes on an OMNI link. The Bridge 477 forwards carrier packets between OMNI link segments as OAL 478 intermediate nodes while decrementing the OAL IPv6 header Hop 479 Limit but without decrementing the network layer IP TTL/Hop Limit. 480 AERO Bridges peer with Proxy/Servers and other Bridges over 481 secured tunnels to discover the full set of MNPs for the link as 482 well as any non-MNP IP GUA routes that are reachable via Relays. 484 link-layer address 485 an IP address used as an encapsulation header source or 486 destination address from the perspective of the OMNI interface. 487 When an upper layer protocol (e.g., UDP) is used as part of the 488 encapsulation, the port number is also considered as part of the 489 link-layer address. 491 network layer address 492 the source or destination address of an original IP packet 493 presented to the OMNI interface. 495 end user network (EUN) 496 an internal virtual or external edge IP network that an AERO 497 Client or Relay connects to the rest of the network via the OMNI 498 interface. The Client/Relay sees each EUN as a "downstream" 499 network, and sees the OMNI interface as the point of attachment to 500 the "upstream" network. 502 Mobile Node (MN) 503 an AERO Client and all of its downstream-attached networks that 504 move together as a single unit, i.e., an end system that connects 505 an Internet of Things. 507 Mobile Router (MR) 508 a MN's on-board router that forwards original IP packets between 509 any downstream-attached networks and the OMNI link. The MR is the 510 MN entity that hosts the AERO Client. 512 Route Optimization Source (ROS) 513 the AERO node nearest the source that initiates route 514 optimization. The ROS may be a Proxy/Server or Relay acting on 515 behalf of the source, or may be the source Client itself. 517 Route Optimization responder (ROR) 518 the AERO node nearest the target destination that responds to 519 route optimization requests. The ROR may be a Proxy/Server acting 520 on behalf of a target MNP Client, a Relay for a non-MNP 521 destination or may be the target Client itself. 523 MAP List 524 a geographically and/or topologically referenced list of addresses 525 of all Proxy/Servers within the same OMNI link. There is a single 526 MAP list for the entire OMNI link. 528 Distributed Mobility Management (DMM) 529 a BGP-based overlay routing service coordinated by Proxy/Servers 530 and Bridges that tracks all Proxy/Server-to-Client associations. 532 Mobility Service (MS) 533 the collective set of all Proxy/Servers, Bridges and Relays that 534 provide the AERO Service to Clients. 536 Mobility Service Endpoint MSE) 537 an individual Proxy/Server, Bridge or Relay in the Mobility 538 Service. 540 Throughout the document, the simple terms "Client", "Proxy/Server", 541 "Bridge" and "Relay" refer to "AERO Client", "AERO Proxy/Server", 542 "AERO Bridge" and "AERO Relay", respectively. Capitalization is used 543 to distinguish these terms from other common Internetworking uses in 544 which they appear without capitalization. 546 The terminology of DHCPv6 [RFC8415] and IPv6 ND [RFC4861] (including 547 the names of node variables, messages and protocol constants) is used 548 throughout this document. The terms "All-Routers multicast", "All- 549 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 550 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 551 generically refer to either Internet Protocol version, i.e., IPv4 552 [RFC0791] or IPv6 [RFC8200]. 554 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 555 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 556 "OPTIONAL" in this document are to be interpreted as described in BCP 557 14 [RFC2119][RFC8174] when, and only when, they appear in all 558 capitals, as shown here. 560 3. Asymmetric Extended Route Optimization (AERO) 562 The following sections specify the operation of IP over OMNI links 563 using the AERO service: 565 3.1. AERO Node Types 567 AERO Clients are Mobile Nodes (MNs) that configure OMNI interfaces 568 over underlying interfaces with addresses that may change when the 569 Client moves to a new network connection point. AERO Clients 570 register their Mobile Network Prefixes (MNPs) with the AERO service, 571 and distribute the MNPs to nodes on EUNs. AERO Bridges, Proxy/ 572 Servers and Relays are critical infrastructure elements in fixed 573 (i.e., non-mobile) INET deployments and hence have permanent and 574 unchanging INET addresses. Together, they constitute the AERO 575 service which provides an OMNI link virtual overlay for connecting 576 AERO Clients. 578 AERO Bridges provide hybrid routing/bridging services (as well as a 579 security trust anchor) for nodes on an OMNI link. Bridges use 580 standard IPv6 routing to forward carrier packets both within the same 581 *NET partition and between disjoint *NET partitions based on an IPv6 582 encapsulation mid-layer known as the OMNI Adaptation Layer (OAL) 583 [I-D.templin-6man-omni]. During forwarding, the inner IP layer 584 experiences a virtual bridging service since the inner IP TTL/Hop 585 Limit is not decremented. Each Bridge also peers with Proxy/Servers 586 and other Bridges in a dynamic routing protocol instance to provide a 587 Distributed Mobility Management (DMM) service for the list of active 588 MNPs (see Section 3.2.3). Bridges present the OMNI link as a set of 589 one or more Mobility Service Prefixes (MSPs) and configure secured 590 tunnels with Proxy/Servers, Relays and other Bridges; they further 591 maintain IP forwarding table entries for each MNP and any other 592 reachable non-MNP prefixes. 594 AERO Proxy/Servers in distributed *NET locations provide default 595 forwarding and mobility/multilink services for AERO Client Mobile 596 Nodes (MNs). Each Proxy/Server also peers with Bridges in a dynamic 597 routing protocol instance to advertise its list of associated MNPs 598 (see Section 3.2.3). Proxy/Servers facilitate prefix delegation/ 599 registration exchanges with Clients, where each delegated prefix 600 becomes an MNP taken from an MSP. Proxy/Servers forward carrier 601 packets between OMNI interface neighbors and track each Client's 602 mobility profiles. Proxy/Servers at ANET/INET boundaries provide a 603 conduit for ANET Clients to associate with peers reached through 604 external INETs. Proxy/Servers in the open INET support INET Clients 605 through authenticated IPv6 ND message exchanges. 607 AERO Relays are Proxy/Servers that provide forwarding services to 608 exchange original IP packets between the OMNI interface and INET/EUN 609 interfaces. Relays are provisioned with MNPs the same as for an AERO 610 Client, and also run a dynamic routing protocol to discover any non- 611 MNP IP routes. The Relay advertises the MSP(s) to its connected 612 networks, and distributes all of its associated MNP and non-MNP 613 routes via BGP peerings with Bridges 615 3.2. The AERO Service over OMNI Links 617 3.2.1. AERO/OMNI Reference Model 619 Figure 1 presents the basic OMNI link reference model: 621 +----------------+ 622 | AERO Bridge B1 | 623 | Nbr: S1, S2, P1| 624 |(X1->S1; X2->S2)| 625 | MSP M1 | 626 +-+------------+-+ 627 +--------------+ | Secured | +--------------+ 628 | AERO P/S S1 | | tunnels | | AERO P/S S2 | 629 | Nbr: C1, B1 +-----+ +-----+ Nbr: C2, B1 | 630 | default->B1 | | default->B1 | 631 | X1->C1 | | X2->C2 | 632 +-------+------+ +------+-------+ 633 | OMNI link | 634 X===+===+======================================+===+===X 635 | | 636 +-----+--------+ +--------+-----+ 637 |AERO Client C1| |AERO Client C2| 638 | Nbr: S1 | | Nbr: S2 | 639 | default->S1 | | default->S2 | 640 | MNP X1 | | MNP X2 | 641 +------+-------+ +-----+--------+ 642 | | 643 .-. .-. 644 ,-( _)-. ,-( _)-. 645 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 646 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 647 `-(______)-' +-------+ +-------+ `-(______)-' 649 Figure 1: AERO/OMNI Reference Model 651 In this model: 653 o the OMNI link is an overlay network service configured over one or 654 more underlying *NET partitions which may be managed by different 655 administrative authorities and have incompatible protocols and/or 656 addressing plans. 658 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 659 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 660 via BGP peerings over secured tunnels to Proxy/Servers (S1, S2). 661 Bridges connect the disjoint segments of a partitioned OMNI link. 663 o AERO Proxy/Servers S1 and S2 configure secured tunnels with Bridge 664 B1 and also provide mobility, multilink, multicast and default 665 router services for the MNPs of their associated Clients C1 and 666 C2. (AERO Proxy/Servers that act as Relays can also advertise 667 non-MNP routes for non-mobile correspondent nodes the same as for 668 MNP Clients.) 670 o AERO Clients C1 and C2 associate with Proxy/Servers S1 and S2, 671 respectively. They receive MNP delegations X1 and X2, and also 672 act as default routers for their associated physical or internal 673 virtual EUNs. Simple hosts H1 and H2 attach to the EUNs served by 674 Clients C1 and C2, respectively. 676 An OMNI link configured over a single *NET appears as a single 677 unified link with a consistent underlying network addressing plan. 678 In that case, all nodes on the link can exchange carrier packets via 679 simple *NET encapsulation, since the underlying *NET is connected. 680 In common practice, however, an OMNI link may be partitioned into 681 multiple "segments", where each segment is a distinct *NET 682 potentially managed under a different administrative authority (e.g., 683 as for worldwide aviation service providers such as ARINC, SITA, 684 Inmarsat, etc.). Individual *NETs may also themselves be partitioned 685 internally, in which case each internal partition is seen as a 686 separate segment. 688 The addressing plan of each segment is consistent internally but will 689 often bear no relation to the addressing plans of other segments. 690 Each segment is also likely to be separated from others by network 691 security devices (e.g., firewalls, proxys, packet filtering gateways, 692 etc.), and in many cases disjoint segments may not even have any 693 common physical link connections. Therefore, nodes can only be 694 assured of exchanging carrier packets directly with correspondents in 695 the same segment, and not with those in other segments. The only 696 means for joining the segments therefore is through inter-domain 697 peerings between AERO Bridges. 699 The same as for traditional campus LANs, multiple OMNI link segments 700 can be joined into a single unified link via a virtual bridging 701 service using the OMNI Adaptation Layer (OAL) [I-D.templin-6man-omni] 702 which inserts a mid-layer IPv6 encapsulation header that supports 703 inter-segment forwarding (i.e., bridging) without decrementing the 704 network-layer TTL/Hop Limit. This bridging of OMNI link segments is 705 shown in Figure 2: 707 . . . . . . . . . . . . . . . . . . . . . . . 708 . . 709 . .-(::::::::) . 710 . .-(::::::::::::)-. +-+ . 711 . (:::: Segment A :::)--|B|---+ . 712 . `-(::::::::::::)-' +-+ | . 713 . `-(::::::)-' | . 714 . | . 715 . .-(::::::::) | . 716 . .-(::::::::::::)-. +-+ | . 717 . (:::: Segment B :::)--|B|---+ . 718 . `-(::::::::::::)-' +-+ | . 719 . `-(::::::)-' | . 720 . | . 721 . .-(::::::::) | . 722 . .-(::::::::::::)-. +-+ | . 723 . (:::: Segment C :::)--|B|---+ . 724 . `-(::::::::::::)-' +-+ | . 725 . `-(::::::)-' | . 726 . | . 727 . ..(etc).. x . 728 . . 729 . . 730 . <- OMNI link Bridged by encapsulation -> . 731 . . . . . . . . . . . . . .. . . . . . . . . 733 Figure 2: Bridging OMNI Link Segments 735 Bridges, Proxy/Servers and Relay OMNI interfaces are configured over 736 both secured tunnels and open INET underlying interfaces over their 737 respective segments in a spanning tree topology rooted at the 738 Bridges. The "secured" spanning tree supports strong authentication 739 for control plane messages and may also be used to convey the initial 740 carrier packets in a flow. The "unsecured" spanning tree conveys 741 ordinary carrier packets without security codes and that must be 742 treated by destinations according to data origin authentication 743 procedures. Route optimization can be employed to cause carrier 744 packets to take more direct paths between OMNI link neighbors without 745 having to follow strict spanning tree paths. 747 3.2.2. Addressing and Node Identification 749 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 750 fe80::/64 [RFC4291] to assign LLAs used for network-layer addresses 751 in link-scoped IPv6 ND and data messages. AERO Clients use LLAs 752 constructed from MNPs (i.e., "MNP-LLAs") while other AERO nodes use 753 LLAs constructed from administrative identification values ("ADM- 754 LLAs") as specified in [I-D.templin-6man-omni]. Non-MNP routes are 755 also represented the same as for MNP-LLAs, but may include a prefix 756 that is not properly covered by the MSP. 758 AERO nodes also use the Unique Local Address (ULA) prefix fd00::/8 759 followed by a pseudo-random 40-bit OMNI domain identifier to form the 760 prefix [ULA]::/48, then include a 16-bit OMNI link identifier '*' to 761 form the prefix [ULA*]::/64 [RFC4291]. The AERO node then uses the 762 prefix [ULA*]::/64 to form "MNP-ULAs" or "ADM-ULA"s as specified in 763 [I-D.templin-6man-omni] to support OAL addressing. (The prefix 764 [ULA*]::/64 appearing alone and with no suffix represents "default".) 765 AERO Clients also use Temporary ULAs constructed per 766 [I-D.templin-6man-omni], where the addresses are typically used only 767 in initial control message exchanges until a stable MNP-LLA/ULA is 768 assigned. 770 AERO MSPs, MNPs and non-MNP routes are typically based on Global 771 Unicast Addresses (GUAs), but in some cases may be based on private- 772 use addresses. See [I-D.templin-6man-omni] for a full specification 773 of LLAs, ULAs and GUAs used by AERO nodes on OMNI links. 775 Finally, AERO Clients and Proxy/Servers configure node identification 776 values as specified in [I-D.templin-6man-omni]. 778 3.2.3. AERO Routing System 780 The AERO routing system comprises a private instance of the Border 781 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 782 and Proxy/Servers and does not interact with either the public 783 Internet BGP routing system or any underlying INET routing systems. 785 In a reference deployment, each Proxy/Server is configured as an 786 Autonomous System Border Router (ASBR) for a stub Autonomous System 787 (AS) using a 32-bit AS Number (ASN) [RFC4271] that is unique within 788 the BGP instance, and each Proxy/Server further uses eBGP to peer 789 with one or more Bridges but does not peer with other Proxy/Servers. 790 Each *NET of a multi-segment OMNI link must include one or more 791 Bridges, which peer with the Proxy/Servers within that *NET. All 792 Bridges within the same *NET are members of the same hub AS, and use 793 iBGP to maintain a consistent view of all active routes currently in 794 service. The Bridges of different *NETs peer with one another using 795 eBGP. 797 Bridges maintain forwarding table entries only for the MNP-ULAs 798 corresponding to MNP and non-MNP routes that are currently active, 799 and carrier packets destined to all other MNP-ULAs will correctly 800 incur Destination Unreachable messages due to the black-hole route. 801 In this way, Proxy/Servers and Relays have only partial topology 802 knowledge (i.e., they know only about the routes their directly 803 associated Clients and non-AERO links) and they forward all other 804 carrier packets to Bridges which have full topology knowledge. 806 Each OMNI link segment assigns a unique ADM-ULA sub-prefix of 807 [ULA*]::/96. For example, a first segment could assign 808 [ULA*]::1000/116, a second could assign [ULA*]::2000/116, a third 809 could assign [ULA*]::3000/116, etc. Within each segment, each Proxy/ 810 Server configures an ADM-ULA within the segment's prefix, e.g., the 811 Proxy/Servers within [ULA*]::2000/116 could assign the ADM-ULAs 812 [ULA*]::2011/116, [ULA*]::2026/116, [ULA*]::2003/116, etc. 814 The administrative authorities for each segment must therefore 815 coordinate to assure mutually-exclusive ADM-ULA prefix assignments, 816 but internal provisioning of ADM-ULAs an independent local 817 consideration for each administrative authority. For each ADM-ULA 818 prefix, the Bridge(s) that connect that segment assign the all-zero's 819 address of the prefix as a Subnet Router Anycast address. For 820 example, the Subnet Router Anycast address for [ULA*]::1023/116 is 821 simply [ULA*]::1000. 823 ADM-ULA prefixes are statically represented in Bridge forwarding 824 tables. Bridges join multiple segments into a unified OMNI link over 825 multiple diverse administrative domains. They support a bridging 826 function by first establishing forwarding table entries for their 827 ADM-ULA prefixes either via standard BGP routing or static routes. 828 For example, if three Bridges ('A', 'B' and 'C') from different 829 segments serviced [ULA*]::1000/116, [ULA*]::2000/116 and 830 [ULA*]::3000/116 respectively, then the forwarding tables in each 831 Bridge are as follows: 833 A: [ULA*]::1000/116->local, [ULA*]::2000/116->B, [ULA*]::3000/116->C 835 B: [ULA*]::1000/116->A, [ULA*]::2000/116->local, [ULA*]::3000/116->C 837 C: [ULA*]::1000/116->A, [ULA*]::2000/116->B, [ULA*]::3000/116->local 839 These forwarding table entries are permanent and never change, since 840 they correspond to fixed infrastructure elements in their respective 841 segments. 843 MNP ULAs are instead dynamically advertised in the AERO routing 844 system by Proxy/Servers and Relays that provide service for their 845 corresponding MNPs. For example, if three Proxy/Servers ('D', 'E' 846 and 'F') service the MNPs 2001:db8:1000:2000::/56, 847 2001:db8:3000:4000::/56 and 2001:db8:5000:6000::/56 then the routing 848 system would include: 850 D: [ULA*]:2001:db8:1000:2000/120 851 E: [ULA*]:2001:db8:3000:4000/120 853 F: [ULA*]:2001:db8:5000:6000/120 855 A full discussion of the BGP-based routing system used by AERO is 856 found in [I-D.ietf-rtgwg-atn-bgp]. 858 3.2.4. OMNI Link Segment Routing 860 With the Client and partition prefixes in place in Bridge forwarding 861 tables, the OMNI interface sends control and data messages toward 862 AERO destination nodes located in different OMNI link segments over 863 the spanning tree. The OMNI interface uses the OMNI Adaptation Layer 864 (OAL) encapsulation service [I-D.templin-6man-omni], and includes an 865 OMNI Routing Header (ORH) as an extension to the OAL header if final 866 segment forwarding information is available, e.g., in the neighbor 867 cache. (For nodes located in the same OMNI link segment, or when no 868 final segment forwarding information is available, the ORH may be 869 omitted.) The ORH is formatted as shown in Figure 3: 871 0 1 2 3 872 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 873 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 874 | Next Header | Hdr Ext Len | Routing Type | SRT | FMT | 875 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 876 | Last Hop Segment-id (LHS) | 877 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 878 ~ Link Layer Address (L2ADDR) ~ 879 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 880 ~ Destination Suffix (if necessary) ~ 881 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 882 ~ Null Padding (if necessary) ~ 883 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 885 Figure 3: OMNI Routing Header (ORH) Format 887 In this format: 889 o Next Header identifies the type of header immediately following 890 the ORH. 892 o Hdr Ext Len is the length of the Routing header in 8-octet units 893 (not including the first 8 octets), with trailing padding added if 894 necessary to produce an integral number of 8-octet units. 896 o Routing Type is set to TBD1 (see IANA Considerations). 898 o Segments Left is omitted, and replaced by a 5-bit SRT and 3-bit 899 FMT field. 901 o SRT - a 5-bit Segment Routing Topology prefix length value that 902 (when added to 96) determines the prefix length to apply to the 903 ADM-ULA formed from concatenating [ULA*]::/96 with the 32 bit LHS 904 value that follows (for example, the value 16 corresponds to the 905 prefix length 112). 907 o FMT - a 3-bit "Framework/Mode/Type" code corresponding to the 908 included Link Layer Address as follows: 910 * When the most significant bit (i.e., "Framework") is set to 1, 911 L2ADDR is the *NET encapsulation address for the target Client 912 itself; otherwise L2ADDR is the address of the Proxy/Server 913 named in the LHS. 915 * When the next most significant bit (i.e., "Mode") is set to 1, 916 the Framework node is (likely) located behind a *NET Network 917 Address Translator (NAT); otherwise, it is on the open *NET. 919 * When the least significant bit (i.e., "Type") is set to 0, 920 L2ADDR includes a UDP Port Number followed by an IPv4 address; 921 otherwise, it includes a UDP Port Number followed by an IPv6 922 address. 924 o LHS - the 32 bit ID of a node in the Last Hop Segment that 925 services the target. When SRT and LHS are both set to 0, the LHS 926 is considered unspecified. When SRT is set to 0 and LHS is non- 927 zero, the prefix length is set to 128. SRT and LHS provide 928 guidance to the OMNI interface forwarding algorithm. 929 Specifically, if SRT/LHS is located in the local OMNI link 930 segment, the OAL source can omit the ORH and (following any 931 necessary NAT traversal messaging) send directly to the OAL 932 destination according to FMT/L2ADDR. Otherwise, it includes the 933 ORH and forwards according to the OMNI link spanning tree. 935 o Link Layer Address (L2ADDR) - Formatted according to FMT, and 936 identifies the link-layer address (i.e., the encapsulation 937 address) of the target. The UDP Port Number appears in the first 938 two octets and the IP address appears in the next 4 octets for 939 IPv4 or 16 octets for IPv6. The Port Number and IP address are 940 recorded in network byte order, and in ones-compliment 941 "obfuscated" form per [RFC4380]. The OMNI interface forwarding 942 algorithm uses FMT/L2ADDR to determine the *NET encapsulation 943 address for local forwarding when SRT/LHS is located in the same 944 OMNI link segment. Note that if the target is behind a NAT, 945 L2ADDR will contain the mapped *NET address stored in the NAT; 946 otherwise, L2ADDR will contain the native *NET information of the 947 target itself. 949 o Destination Suffix is a 64-bit field included only for OAL non- 950 first-fragments. Present only when Hdr Ext Len indicates that at 951 least 8 bytes follow L2ADDR. When present, encodes the 64-bit 952 MNP-ULA suffix for the target Client. 954 o Null Padding contains zero-valued octets as necessary to pad the 955 ORH to an integral number of 8-octet units. 957 AERO neighbors use OAL encapsulation and fragmentation to exchange 958 OAL packets as specified in [I-D.templin-6man-omni]. When an AERO 959 node's OMNI interface (acting as an OAL source) uses OAL 960 encapsulation for an original IP packet with source address 961 2001:db8:1:2::1 and destination address 2001:db8:1234:5678::1, it 962 sets the OAL header source address to its own ULA (e.g., 963 [ULA*]::2001:db8:1:2), sets the destination address to the MNP-ULA 964 corresponding to the IP destination address (e.g., 965 [ULA*]::2001:db8:1234:5678), sets the Traffic Class, Flow Label, Hop 966 Limit and Payload Length as discussed in [I-D.templin-6man-omni], 967 then finally selects an Identification and appends an OAL checksum. 969 If the neighbor cache information indicates that the target is in a 970 different segment, the OAL source next inserts an ORH immediately 971 following the OAL header while including the correct SRT, FMT, LHS, 972 L2ADDR and Destination Suffix if fragmentation if needed (in this 973 case, the Destination Suffix is 2001:db8:1234:5678). Next, the OAL 974 source overwrites the OAL header destination address with the LHS 975 Subnet Router Anycast address (for example, for LHS 3000:4567 with 976 SRT 16, the Subnet Router Anycast address is [ULA*]::3000:0000). 977 (Note: if the ADM-ULA of the last-hop Proxy/Server is known but the 978 SRT, FMT, LHS and L2ADDR are not (yet) known, the OAL source instead 979 overwrites the OAL header destination address with the ADM-ULA.) 981 The OAL source then fragments the OAL packet, with each resulting OAL 982 fragment including the OAL/ORH headers while only the first fragment 983 includes the original IPv6 header. (Note that if no actual 984 fragmentation is required the OAL packet is still prepared as an 985 "atomic" fragment that includes a Fragment Header with Offset and 986 More Fragments both set to 0.) The OAL source finally encapsulates 987 each resulting OAL fragment in an *NET header to form an OAL carrier 988 packet, with source address set to its own *NET address (e.g., 989 192.0.2.100) and destination set to the *NET address of a Bridge 990 (e.g., 192.0.2.1). 992 The carrier packet encapsulation format in the above example is shown 993 in Figure 4: 995 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 996 | *NET Header | 997 | src = 192.0.2.100 | 998 | dst = 192.0.2.1 | 999 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1000 | OAL IPv6 Header | 1001 | src = [ULA*]::2001:db8:1:2 | 1002 | dst= [ULA*]::3000:0000 | 1003 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1004 | ORH (if necessary) | 1005 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1006 | OAL Fragment Header | 1007 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1008 | Original IP Header | 1009 | (first-fragment only) | 1010 | src = 2001:db8:1:2::1 | 1011 | dst = 2001:db8:1234:5678::1 | 1012 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1013 | | 1014 ~ ~ 1015 ~ Original Packet Body/Fragment ~ 1016 ~ ~ 1017 | | 1018 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1020 Figure 4: Carrier Packet Format 1022 In this format, the original IP header and packet body/fragment are 1023 from the original IP packet, the OAL header is an IPv6 header 1024 prepared according to [RFC2473], the ORH is a Routing Header 1025 extension of the OAL header, the Fragment Header identifies each 1026 fragment, and the INET header is prepared as discussed in 1027 Section 3.6. When the OAL source transmits the resulting carrier 1028 packets, they are forwarded over possibly multiple OAL intermediate 1029 nodes in the OMNI link spanning tree until they arrive at the OAL 1030 destination. 1032 This gives rise to a routing system that contains both Client MNP-ULA 1033 routes that may change dynamically due to regional node mobility and 1034 per-partition ADM-ULA routes that rarely if ever change. The Bridges 1035 can therefore provide link-layer bridging by sending carrier packets 1036 over the spanning tree instead of network-layer routing according to 1037 MNP routes. As a result, opportunities for loss due to node mobility 1038 between different segments are mitigated. 1040 Note: When the OAL source and destination are on the same INET 1041 segment, the ORH is not needed and OAL header compression can be used 1042 to significantly reduce encapsulation overhead 1043 [I-D.templin-6man-omni]. 1045 Note: When the OAL source has multiple original IP packets to send to 1046 the same OAL destination, it can perform "packing" to generate a 1047 "super-packet" [I-D.templin-6man-omni]. In that case, the OAL/ORH 1048 super-packet may include up to N original IP packets as long as the 1049 total length of the super-packet does not exceed the OMNI interface 1050 MTU. 1052 Note: Use of an IPv6 "minimal encapsulation" format (i.e., an IPv6 1053 variant of [RFC2004]) based on extensions to the ORH was considered 1054 and abandoned. In the approach, the ORH would be inserted as an 1055 extension header to the original IPv6 packet header. The IPv6 1056 destination address would then be written into the ORH, and the ULA 1057 corresponding to the destination would be overwritten in the IPv6 1058 destination address. This would seemingly convey enough forwarding 1059 information so that OAL encapsulation could be avoided. However, 1060 this "minimal encapsulation" IPv6 packet would then have a non-ULA 1061 source address and ULA destination address, an incorrect value in 1062 upper layer protocol checksums, and a Hop Limit that is decremented 1063 within the spanning tree when it should not be. The insertion and 1064 removal of the ORH would also entail rewriting the Payload Length and 1065 Next Header fields - again, invalidating upper layer checksums. 1066 These irregularities would result in implementation challenges and 1067 the potential for operational issues, e.g., since actionable ICMPv6 1068 error reports could not be delivered to the original source. In 1069 order to address the issues, still more information such as the 1070 original IPv6 source address could be written into the ORH. However, 1071 with the additional information the benefit of the "minimal 1072 encapsulation" savings quickly diminishes, and becomes overshadowed 1073 by the implementation and operational irregularities. 1075 3.2.5. Segment Routing Topologies (SRTs) 1077 The 64-bit sub-prefixes of [ULA]::/48 identify up to 2^16 distinct 1078 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 1079 OMNI link overlay instance using a distinct set of ULAs, and emulates 1080 a Virtual LAN (VLAN) service for the OMNI link. In some cases (e.g., 1081 when redundant topologies are needed for fault tolerance and 1082 reliability) it may be beneficial to deploy multiple SRTs that act as 1083 independent overlay instances. A communication failure in one 1084 instance therefore will not affect communications in other instances. 1086 Each SRT is identified by a distinct value in bits 48-63 of 1087 [ULA]::/48, i.e., as [ULA0]::/64, [ULA1]::/64, [ULA2]::/64, etc. 1088 Each OMNI interface is identified by a unique interface name (e.g., 1089 omni0, omni1, omni2, etc.) and assigns an anycast ADM-ULA 1090 corresponding to its SRT prefix length. The anycast ADM-ULA is used 1091 for OMNI interface determination in Safety-Based Multilink (SBM) as 1092 discussed in [I-D.templin-6man-omni]. Each OMNI interface further 1093 applies Performance-Based Multilink (PBM) internally. 1095 3.2.6. Segment Routing For OMNI Link Selection 1097 An original IPv6 source can direct an IPv6 packet to an AERO node by 1098 including a standard IPv6 Segment Routing Header (SRH) [RFC8754] with 1099 the anycast ADM-ULA for the selected SRT as either the IPv6 1100 destination or as an intermediate hop within the SRH. This allows 1101 the original source to determine the specific OMNI link topology an 1102 original IPv6 packet will traverse when there may be multiple 1103 alternatives. 1105 When the AERO node processes the SRH and forwards the original IPv6 1106 packet to the correct OMNI interface, the OMNI interface writes the 1107 next IPv6 address from the SRH into the IPv6 destination address and 1108 decrements Segments Left. If decrementing would cause Segments Left 1109 to become 0, the OMNI interface deletes the SRH before forwarding. 1110 This form of Segment Routing supports Safety-Based Multilink (SBM). 1112 3.2.7. Segment Routing Within the OMNI Link 1114 OAL sources can insert an ORH for Segment Routing within the OMNI 1115 link to influence the paths of OAL packets sent to OAL destinations 1116 in remote segments without requiring all carrier packets to traverse 1117 strict spanning tree paths. 1119 When an AERO node's OMNI interface has an original IP packet to send 1120 to a target discovered through route optimization located in the same 1121 OMNI link segment, it acts as an OAL source to perform OAL 1122 encapsulation and fragmentation. The node then uses the target's 1123 Link Layer Address (L2ADDR) information for *NET encapsulation. 1125 When an AERO node's OMNI interface has an original IP packet to send 1126 to a route optimization target located in a remote OMNI link segment, 1127 it acts as an OAL source the same as above but also includes an ORH 1128 while setting the OAL destination to the Subnet Router Anycast 1129 address for the final OMNI link segment, then forwards the resulting 1130 carrier packets to a Bridge. 1132 When a Bridge receives a carrier packet destined to its Subnet Router 1133 Anycast address with an ORH with SRT/LHS values corresponding to the 1134 local segment, it examines the L2ADDR according to FMT and removes 1135 the ORH from the carrier packet. The Bridge then writes the MNP-ULA 1136 corresponding to the ORH Destination Suffix into the OAL destination 1137 address, decrements the OAL IPv6 header Hop Limit (and discards the 1138 packet if the Hop Limit reaches 0), re-encapsulates the carrier 1139 packet according to L2ADDR and forwards the carrier packet either to 1140 the LHS Proxy/Server or directly to the target Client itself. In 1141 this way, the Bridge participates in route optimization to reduce 1142 traffic load and suboptimal routing through strict spanning tree 1143 paths. 1145 3.3. OMNI Interface Characteristics 1147 OMNI interfaces are virtual interfaces configured over one or more 1148 underlying interfaces classified as follows: 1150 o INET interfaces connect to an INET either natively or through one 1151 or more NATs. Native INET interfaces have global IP addresses 1152 that are reachable from any INET correspondent. The INET-facing 1153 interfaces of Proxy/Servers are native interfaces, as are Relay 1154 and Bridge interfaces. NATed INET interfaces connect to a private 1155 network behind one or more NATs that provide INET access. Clients 1156 that are behind a NAT are required to send periodic keepalive 1157 messages to keep NAT state alive when there are no carrier packets 1158 flowing. 1160 o ANET interfaces connect to an ANET that is separated from the open 1161 INET by a Proxy/Server. Proxy/Servers can actively issue control 1162 messages over the INET on behalf of the Client to reduce ANET 1163 congestion. 1165 o VPNed interfaces use security encapsulation over the INET to a 1166 Virtual Private Network (VPN) server that also acts as a Proxy/ 1167 Server. Other than the link-layer encapsulation format, VPNed 1168 interfaces behave the same as Direct interfaces. 1170 o Direct (i.e., single-hop point-to-point) interfaces connect a 1171 Client directly to a Proxy/Server without crossing any ANET/INET 1172 paths. An example is a line-of-sight link between a remote pilot 1173 and an unmanned aircraft. The same Client considerations apply as 1174 for VPNed interfaces. 1176 OMNI interfaces use OAL encapsulation and fragmentation as discussed 1177 in Section 3.2.4. OMNI interfaces use *NET encapsulation (see: 1178 Section 3.6) to exchange carrier packets with OMNI link neighbors 1179 over INET or VPNed interfaces as well as over ANET interfaces for 1180 which the Client and Proxy/Server may be multiple IP hops away. OMNI 1181 interfaces do not use link-layer encapsulation over Direct underlying 1182 interfaces or ANET interfaces when the Client and Proxy/Server are 1183 known to be on the same underlying link. 1185 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1186 state the same as for any interface. OMNI interfaces use ND messages 1187 including Router Solicitation (RS), Router Advertisement (RA), 1188 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1189 neighbor cache management. 1191 OMNI interfaces send ND messages with an OMNI option formatted as 1192 specified in [I-D.templin-6man-omni]. The OMNI option includes 1193 prefix registration information and Interface Attributes containing 1194 link information parameters for the OMNI interface's underlying 1195 interfaces. Each OMNI option may include multiple Interface 1196 Attributes sub-options, each identified by an omIndex value. 1198 A Client's OMNI interface may be configured over multiple underlying 1199 interface connections. For example, common mobile handheld devices 1200 have both wireless local area network ("WLAN") and cellular wireless 1201 links. These links are often used "one at a time" with low-cost WLAN 1202 preferred and highly-available cellular wireless as a standby, but a 1203 simultaneous-use capability could provide benefits. In a more 1204 complex example, aircraft frequently have many wireless data link 1205 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1206 directional, etc.) with diverse performance and cost properties. 1208 If a Client's multiple underlying interfaces are used "one at a time" 1209 (i.e., all other interfaces are in standby mode while one interface 1210 is active), then ND message OMNI options include only a single 1211 Interface Attributes sub-option set to constant values. In that 1212 case, the Client would appear to have a single interface but with a 1213 dynamically changing link-layer address. 1215 If the Client has multiple active underlying interfaces, then from 1216 the perspective of ND it would appear to have multiple link-layer 1217 addresses. In that case, ND message OMNI options MAY include 1218 multiple Interface Attributes sub-options - each with values that 1219 correspond to a specific interface. Every ND message need not 1220 include Interface Attributes for all underlying interfaces; for any 1221 attributes not included, the neighbor considers the status as 1222 unchanged. 1224 Bridge and Proxy/Server OMNI interfaces are configured over secured 1225 tunnel underlying interfaces for carrying IPv6 ND and BGP protocol 1226 control plane messages, plus open INET underlying interfaces for 1227 carrying unsecured messages. The OMNI interface configures both an 1228 ADM-LLA and its corresponding ADM-ULA, and acts as an OAL source to 1229 encapsulate and fragment original IP packets while presenting the 1230 resulting carrier packets to a secured or unsecured underlying 1231 interface. Note that Bridge and Proxy/Server BGP protocol TCP 1232 sessions are run directly over the OMNI interface using ADM-ULA 1233 source and destination addresses. The OMNI interface will 1234 encapsulate these as carrier packets even though the OAL header may 1235 use the same ADM-ULAs as the original IP header; these carrier 1236 packets must then be sent over a secured underlying interface. 1238 3.4. OMNI Interface Initialization 1240 AERO Proxy/Servers and Clients configure OMNI interfaces as their 1241 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1242 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1243 to ensure that original IP packets with destination addresses covered 1244 by an MNP not explicitly assigned to a non-OMNI interface are 1245 directed to the OMNI interface. 1247 OMNI interface initialization procedures for Proxy/Servers, Clients 1248 and Bridges are discussed in the following sections. 1250 3.4.1. AERO Proxy/Server and Relay Behavior 1252 When a Proxy/Server enables an OMNI interface, it assigns an 1253 ADM-{LLA,ULA} appropriate for the given OMNI link segment. The 1254 Proxy/Server also configures secured tunnels with one or more 1255 neighboring Bridges and engages in a BGP routing protocol session 1256 with each Bridge. 1258 The OMNI interface provides a single interface abstraction to the IP 1259 layer, but internally includes one or more secured tunnels as well as 1260 an NBMA nexus as underlying interfaces for sending carrier packets to 1261 OMNI interface neighbors. The Proxy/Server further configures a 1262 service to facilitate ND exchanges with AERO Clients and manages per- 1263 Client neighbor cache entries and IP forwarding table entries based 1264 on control message exchanges. 1266 Relays are simply Proxy/Servers that run a dynamic routing protocol 1267 to redistribute routes between the OMNI interface and INET/EUN 1268 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1269 networks on the INET/EUN interfaces (i.e., the same as a Client would 1270 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1271 interfaces. The Relay further provides an attachment point of the 1272 OMNI link to a non-MNP-based global topology. 1274 3.4.2. AERO Client Behavior 1276 When a Client enables an OMNI interface, it assigns either an 1277 MNP-{LLA, ULA} or a Temporary ULA and sends RS messages with ND 1278 parameters over its underlying interfaces to a Proxy/Server, which 1279 returns an RA message with corresponding parameters. The RS/RA 1280 messages may pass through one or more NATs in the case of a Client's 1281 INET interface. (Note: if the Client used a Temporary ULA in its 1282 initial RS message, it will discover an MNP-{LLA, ULA} in the 1283 corresponding RA that it receives from the Proxy/Server and begin 1284 using these new addresses. If the Client is operating outside the 1285 context of AERO infrastructure such as in a Mobile Ad-hoc Network 1286 (MANET), however, it may continue using Temporary ULAs for Client-to- 1287 Client communications until it encounters an infrastructure element 1288 that can provide an MNP.) 1290 3.4.3. AERO Bridge Behavior 1292 AERO Bridges configure an OMNI interface and assign the ADM-ULA 1293 Subnet Router Anycast address for each OMNI link segment they connect 1294 to. Bridges configure secured tunnels with Proxy/Servers and other 1295 Bridges, and engage in a BGP routing protocol session with neighbors 1296 on the spanning tree (see: Section 3.2.3). 1298 3.5. OMNI Interface Neighbor Cache Maintenance 1300 Each OMNI interface maintains a conceptual neighbor cache that 1301 includes an entry for each neighbor it communicates with on the OMNI 1302 link per [RFC4861]. In addition to ordinary neighbor cache entries, 1303 proxy neighbor cache entries are created and maintained by AERO 1304 Proxy/Servers when they proxy Client ND message exchanges [RFC4389]. 1305 AERO Proxy/Servers maintain proxy neighbor cache entries for each of 1306 their associated Clients. 1308 To the list of neighbor cache entry states in Section 7.3.2 of 1309 [RFC4861], Proxy/Server OMNI interfaces add an additional state 1310 DEPARTED that applies to Clients that have recently departed. The 1311 interface sets a "DepartTime" variable for the neighbor cache entry 1312 to "DEPART_TIME" seconds. DepartTime is decremented unless a new ND 1313 message causes the state to return to REACHABLE. While a neighbor 1314 cache entry is in the DEPARTED state, the Proxy/Server forwards 1315 carrier packets destined to the target Client to the Client's new 1316 location instead. When DepartTime decrements to 0, the neighbor 1317 cache entry is deleted. It is RECOMMENDED that DEPART_TIME be set to 1318 the default constant value REACHABLE_TIME plus 10 seconds (40 seconds 1319 by default) to allow a window for carrier packets in flight to be 1320 delivered while stale route optimization state may be present. 1322 Proxy/Servers can act as RORs on behalf of dependent Clients 1323 according to the Proxy Neighbor Advertisement specification in 1324 Section 7.2.8 of [RFC4861], while well-connected Clients can act as 1325 an ROR on their own behalf. When a Proxy/Server ROR receives an 1326 authentic NS message used for route optimization, it first searches 1327 for a proxy neighbor cache entry for the target Client and accepts 1328 the message only if there is an entry. The Proxy/Server then returns 1329 a solicited NA message while creating a neighbor cache entry for the 1330 ROS and caching the Identification value found in the NS message 1331 carrier packet as the starting window Identification value for this 1332 ROS. Proxy/Server RORs also create or update a "Report List" entry 1333 for the ROS in the target Client's neighbor cache entry with a 1334 "ReportTime" variable set to REPORT_TIME seconds. The ROR resets 1335 ReportTime when it receives a new authentic NS message, and otherwise 1336 decrements ReportTime while no authentic NS messages have been 1337 received. It is RECOMMENDED that REPORT_TIME be set to the default 1338 constant value REACHABLE_TIME plus 10 seconds (40 seconds by default) 1339 to allow a window for route optimization to converge before 1340 ReportTime decrements below REACHABLE_TIME. 1342 When the ROS receives a solicited NA message response to its NS 1343 message used for route optimization, it creates or updates a neighbor 1344 cache entry for the target network-layer and link-layer addresses. 1345 The ROS then (re)sets ReachableTime for the neighbor cache entry to 1346 REACHABLE_TIME seconds and uses this value to determine whether 1347 carrier packets can be forwarded directly to the target, i.e., 1348 instead of via a default route. The ROS also maintains a window 1349 start Identification value that is monotonically incremented for each 1350 OAL packet sent to this target, and sets new window start 1351 Identification values when it sends a new NS. The ROS otherwise 1352 decrements ReachableTime while no further solicited NA messages 1353 arrive. It is RECOMMENDED that REACHABLE_TIME be set to the default 1354 constant value 30 seconds as specified in [RFC4861]. 1356 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1357 of NS keepalives sent when a correspondent may have gone unreachable, 1358 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1359 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1360 to limit the number of unsolicited NAs that can be sent based on a 1361 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1362 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1363 same as specified in [RFC4861]. 1365 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1366 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1367 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1368 different values are chosen, all nodes on the link MUST consistently 1369 configure the same values. Most importantly, DEPART_TIME and 1370 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1371 REACHABLE_TIME to avoid packet loss due to stale route optimization 1372 state. 1374 3.5.1. OMNI Neighbor Interface Attributes 1376 OMNI interface IPv6 ND messages include OMNI options 1377 [I-D.templin-6man-omni] with Interface Attributes that provide Link- 1378 Layer Address and QoS Preference information for the neighbor's 1379 underlying interfaces. This information is stored in the neighbor 1380 cache and provides the basis for the forwarding algorithm specified 1381 in Section 3.10. The information is cumulative and reflects the 1382 union of the OMNI information from the most recent ND messages 1383 received from the neighbor; it is therefore not required that each ND 1384 message contain all neighbor information. 1386 The OMNI option Interface Attributes for each underlying interface 1387 includes a two-part "Link-Layer Address" consisting of a simple IP 1388 encapsulation address determined by the FMT and L2ADDR fields and an 1389 ADM-ULA determined by the SRT and LHS fields. Underlying interfaces 1390 are further selected based on their associated preference values 1391 "high", "medium", "low" or "disabled". 1393 Note: the OMNI option is distinct from any Source/Target Link-Layer 1394 Address Options (S/TLLAOs) that may appear in an ND message according 1395 to the appropriate IPv6 over specific link layer specification (e.g., 1396 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1397 pertains to encapsulation addresses while the latter pertains to the 1398 native L2 address format of the underlying media. 1400 3.5.2. OMNI Neighbor Advertisement Message Flags 1402 As discussed in Section 4.4 of [RFC4861] NA messages include three 1403 flag bits R, S and O. OMNI interface NA messages treat the flags as 1404 follows: 1406 o R: The R ("Router") flag is set to 1 in the NA messages sent by 1407 all AERO/OMNI node types. Simple hosts that would set R to 0 do 1408 not occur on the OMNI link itself, but may occur on the downstream 1409 links of Clients and Relays. 1411 o S: The S ("Solicited") flag is set exactly as specified in 1412 Section 4.4. of [RFC4861], i.e., it is set to 1 for Solicited NAs 1413 and set to 0 for Unsolicited NAs (both unicast and multicast). 1415 o O: The O ("Override") flag is set to 0 for solicited proxy NAs 1416 returned by a Proxy/Server ROR and set to 1 for all other 1417 solicited and unsolicited NAs. For further study is whether 1418 solicited NAs for anycast targets apply for OMNI links. Since 1419 MNP-LLAs must be uniquely assigned to Clients to support correct 1420 ND protocol operation, however, no role is currently seen for 1421 assigning the same MNP-LLA to multiple Clients. 1423 3.6. OMNI Interface Encapsulation and Re-encapsulation 1425 The OMNI interface admits original IP packets then (acting as an OAL 1426 source) performs OAL encapsulation and fragmentation as specified in 1427 [I-D.templin-6man-omni] while including an ORH if necessary as 1428 specified in Section 3.2.4. OAL encapsulation produces OAL packets, 1429 while OAL fragmentation turns them into OAL fragments which are then 1430 encapsulated in *NET headers as carrier packets. 1432 For carrier packets undergoing re-encapsulation at an OAL 1433 intermediate node, the OMNI interface decrements the OAL IPv6 header 1434 Hop Limit and discards the carrier packet if the Hop Limit reaches 0. 1435 The intermediate node next removes the *NET encapsulation headers 1436 from the first segment and re-encapsulates the packet in new *NET 1437 encapsulation headers for the next segment. 1439 When a Proxy/Server or Relay re-encapsulates a carrier packet 1440 received from a Client that includes an OAL but no ORH, it inserts an 1441 ORH immediately following the OAL header and adjusts the OAL payload 1442 length and destination address field. The inserted ORH will be 1443 removed by the final-hop Bridge, but its insertion and removal will 1444 not interfere with reassembly at the final destination. For this 1445 reason, Clients must reserve 40 bytes for a maximum-length ORH when 1446 they perform OAL encapsulation (see: Section 3.9). 1448 3.7. OMNI Interface Decapsulation 1450 OMNI interfaces (acting as OAL destinations) decapsulate and 1451 reassemble OAL packets into original IP packets destined either to 1452 the AERO node itself or to a destination reached via an interface 1453 other than the OMNI interface the original IP packet was received on. 1454 When carrier packets containing OAL fragments arrive, the OMNI 1455 interface reassembles as discussed in Section 3.9. 1457 3.8. OMNI Interface Data Origin Authentication 1459 AERO nodes employ simple data origin authentication procedures. In 1460 particular: 1462 o AERO Bridges and Proxy/Servers accept carrier packets received 1463 from secured underlying interfaces. 1465 o AERO Proxy/Servers and Clients accept carrier packets and original 1466 IP packets that originate from within the same secured ANET. 1468 o AERO Clients and Relays accept original IP packets from downstream 1469 network correspondents based on ingress filtering. 1471 o AERO Clients, Relays and Proxy/Servers verify carrier packet UDP/ 1472 IP encapsulation addresses according to [I-D.templin-6man-omni]. 1474 o AERO Clients (as well as Proxy/Servers and Relays when acting as 1475 OAL destinations) accept carrier packets with Identification 1476 values within the current window for the OAL source. 1478 AERO nodes silently drop any packets that do not satisfy the above 1479 data origin authentication procedures. Further security 1480 considerations are discussed in Section 6. 1482 3.9. OMNI Interface MTU 1484 The OMNI interface observes the link nature of tunnels, including the 1485 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1486 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 1487 The OMNI interface employs an OMNI Adaptation Layer (OAL) that 1488 accommodates multiple underlying links with diverse MTUs while 1489 observing both a minimum and per-path Maximum Payload Size (MPS). 1490 The functions of the OAL and the OMNI interface MTU/MRU/MPS are 1491 specified in [I-D.templin-6man-omni] with MTU/MRU both set to the 1492 constant value 9180 bytes, with minimum MPS set to 400 bytes, and 1493 with per-path MPS set to potentially larger values depending on the 1494 underlying path. 1496 When the network layer presents an original IP packet to the OMNI 1497 interface, the OAL source encapsulates and fragments the original IP 1498 packet if necessary. When the network layer presents the OMNI 1499 interface with multiple original IP packets bound to the same OAL 1500 destination, the OAL source can concatenate them together into a 1501 single OAL super-packet as discussed in [I-D.templin-6man-omni]. The 1502 OAL source then fragments the OAL packet if necessary according to 1503 the minimum/path MPS such that the OAL headers appear in each 1504 fragment while the original IP packet header appears only in the 1505 first fragment. The OAL source then encapsulates each OAL fragment 1506 in *NET headers for transmission as carrier packets over an 1507 underlying interface connected to either a physical link such as 1508 Ethernet, WiFi and the like or a virtual link such as an Internet or 1509 higher-layer tunnel (see the definition of link in [RFC8200]). 1511 Note: A Client that does not (yet) have neighbor cache state for a 1512 target may omit the ORH in carrier packets with the understanding 1513 that a Proxy/Server may insert an ORH on its behalf. For this 1514 reason, Clients reserve 40 bytes for the largest possible ORH in 1515 their OAL fragment size calculations. 1517 Note: Although the ORH may be removed by a Bridge on the path (see: 1518 Section 3.10.3), this does not interfere with the destination's 1519 ability to reassemble. This is due to the fact that the ORH is not 1520 included in the fragmentable part; therefore, its removal does not 1521 invalidate the offset values in any fragment headers. 1523 3.10. OMNI Interface Forwarding Algorithm 1525 Original IP packets enter a node's OMNI interface either from the 1526 network layer (i.e., from a local application or the IP forwarding 1527 system) while carrier packets enter from the link layer (i.e., from 1528 an OMNI interface neighbor). All original IP packets and carrier 1529 packets entering a node's OMNI interface first undergo data origin 1530 authentication as discussed in Section 3.8. Those that satisfy data 1531 origin authentication are processed further, while all others are 1532 dropped silently. 1534 Original IP packets that enter the OMNI interface from the network 1535 layer are forwarded to an OMNI interface neighbor using OAL 1536 encapsulation and fragmentation to produce carrier packets for 1537 transmission over underlying interfaces. (If routing indicates that 1538 the original IP packet should instead be forwarded back to the 1539 network layer, the packet is dropped to avoid looping). Carrier 1540 packets that enter the OMNI interface from the link layer are either 1541 re-encapsulated and re-admitted into the OMNI link, or reassembled 1542 and forwarded to the network layer where they are subject to either 1543 local delivery or IP forwarding. In all cases, the OAL MUST NOT 1544 decrement the network layer TTL/Hop-count since its forwarding 1545 actions occur below the network layer. 1547 OMNI interfaces may have multiple underlying interfaces and/or 1548 neighbor cache entries for neighbors with multiple underlying 1549 interfaces (see Section 3.3). The OAL uses interface attributes and/ 1550 or traffic classifiers (e.g., DSCP value, port number, flow 1551 specification, etc.) to select an outgoing underlying interface for 1552 each OAL packet based on the node's own QoS preferences, and also to 1553 select a destination link-layer address based on the neighbor's 1554 underlying interface with the highest preference. AERO 1555 implementations SHOULD allow for QoS preference values to be modified 1556 at runtime through network management. 1558 If multiple outgoing interfaces and/or neighbor interfaces have a 1559 preference of "high", the AERO node replicates the OAL packet and 1560 sends one copy via each of the (outgoing / neighbor) interface pairs; 1561 otherwise, the node sends a single copy of the OAL packet via an 1562 interface with the highest preference. (While not strictly required, 1563 successful delivery may be more likely when all OAL fragments of the 1564 same OAL packet are sent over the same underlying interface.) AERO 1565 nodes keep track of which underlying interfaces are currently 1566 "reachable" or "unreachable", and only use "reachable" interfaces for 1567 forwarding purposes. 1569 The following sections discuss the OMNI interface forwarding 1570 algorithms for Clients, Proxy/Servers and Bridges. In the following 1571 discussion, an original IP packet's destination address is said to 1572 "match" if it is the same as a cached address, or if it is covered by 1573 a cached prefix (which may be encoded in an MNP-LLA). 1575 3.10.1. Client Forwarding Algorithm 1577 When an original IP packet enters a Client's OMNI interface from the 1578 network layer the Client searches for a neighbor cache entry that 1579 matches the destination. If there is a match, the Client selects one 1580 or more "reachable" neighbor interfaces in the entry for forwarding 1581 purposes. If there is no neighbor cache entry, the Client instead 1582 either enqueues the original IP packet and invokes route optimization 1583 or forwards the original IP packet toward a Proxy/Server. The Client 1584 (acting as an OAL source) performs OAL encapsulation and sets the OAL 1585 destination address to the MNP-ULA if there is a matching neighbor 1586 cache entry; otherwise, it sets the OAL destination to the ADM-ULA of 1587 the Proxy/Server. If the Client has multiple original IP packets to 1588 send to the same neighbor, it can concatenate them in a single super- 1589 packet [I-D.templin-6man-omni]. The OAL source then performs 1590 fragmentation to create OAL fragments (see: Section 3.9), appends any 1591 *NET encapsulation, and sends the resulting carrier packets over 1592 underlying interfaces to the neighbor acting as an OAL destination. 1594 If the neighbor interface selected for forwarding is located on the 1595 same OMNI link segment and not behind a NAT, the Client forwards the 1596 carrier packets directly according to the L2ADDR information for the 1597 neighbor. If the neighbor interface is behind a NAT on the same OMNI 1598 link segment, the Client instead forwards the initial carrier packets 1599 to its Proxy/Server and initiates NAT traversal procedures. If the 1600 Client's intended source underlying interface is also behind a NAT 1601 and located on the same OMNI link segment, it sends a "direct bubble" 1602 over the interface per [RFC6081][RFC4380] to the L2ADDR found in the 1603 neighbor cache in order to establish state in its own NAT by 1604 generating traffic toward the neighbor (note that no response to the 1605 bubble is expected). 1607 The Client next sends an NS(NUD) message toward the MNP-ULA of the 1608 neighbor via its Proxy/Server as discussed in Section 3.15. If the 1609 Client receives an NA(NUD) from the neighbor over the underlying 1610 interface, it marks the neighbor interface as "trusted" and sends 1611 future carrier packets directly to the L2ADDR information for the 1612 neighbor instead of indirectly via the Proxy/Server. The Client must 1613 honor the neighbor cache maintenance procedure by sending additional 1614 direct bubbles and/or NS/NA(NUD) messages as discussed in 1615 [RFC6081][RFC4380] in order to keep NAT state alive as long as 1616 carrier packets are still flowing. 1618 When a carrier packet enters a Client's OMNI interface from the link- 1619 layer, if the OAL destination matches one of the Client's ULAs the 1620 Client (acting as an OAL destination) reassembles and decapsulates as 1621 necessary and delivers the original IP packet to the network layer. 1622 Otherwise, the Client drops the original IP packet and MAY return a 1623 network-layer ICMP Destination Unreachable message subject to rate 1624 limiting (see: Section 3.11). 1626 Note: Clients and their Proxy/Server (and other Client) peers can 1627 exchange original IP packets over ANET underlying interfaces without 1628 invoking the OAL, since the ANET is secured at the link and physical 1629 layers. By forwarding original IP packets without invoking the OAL, 1630 however, the ANET peers can engage only in classical path MTU 1631 discovery since the packets are subject to loss and/or corruption due 1632 to the various per-link MTU limitations that may occur within the 1633 ANET. Moreover, the original IP packets do not include per-packet 1634 Identification values that can be used for data origin authentication 1635 and link-layer retransmission purposes, nor the OAL integrity check. 1636 The tradeoff therefore involves an assessment of the per-packet 1637 encapsulation overhead saved by bypassing the OAL vs. inheritance of 1638 classical network "brittleness". (Note however that ANET peers can 1639 send small original IP packets without invoking the OAL, while 1640 invoking the OAL for larger packets. This presents the beneficial 1641 aspects of both small packet efficiency and large packet robustness.) 1643 3.10.2. Proxy/Server and Relay Forwarding Algorithm 1645 When the Proxy/Server receives an original IP packet from the network 1646 layer, it drops the packet if routing indicates that it should be 1647 forwarded back to the network layer to avoid looping. Otherwise, the 1648 Proxy/Server regards the original IP packet the same as if it had 1649 arrived as carrier packets with OAL destination set to its own ADM- 1650 ULA. When the Proxy/Server receives carrier packets on underlying 1651 interfaces with OAL destination set to its own ADM-ULA, it performs 1652 OAL reassembly if necessary to obtain the original IP packet. 1654 The Proxy/Server next searches for a neighbor cache entry that 1655 matches the original IP destination and proceeds as follows: 1657 o if the original IP packet destination matches a neighbor cache 1658 entry, the Proxy/Sever uses one or more "reachable" neighbor 1659 interfaces in the entry for packet forwarding using OAL 1660 encapsulation and fragmentation according to the cached link-layer 1661 address information. If the neighbor interface is in a different 1662 OMNI link segment, the Proxy/Server forwards the resulting carrier 1663 packets via the spanning tree to a Bridge; otherwise, it forwards 1664 the carrier packets directly to the neighbor. If the neighbor is 1665 behind a NAT, the Proxy/Server instead forwards initial carrier 1666 packets via a Bridge while sending an NS(NUD) to the neighbor. 1667 When the Proxy/Server receives the NA(NUD), it can begin 1668 forwarding carrier packets directly to the neighbor the same as 1669 discussed in Section 3.10.1 while sending additional NS(NUD) 1670 messages as necessary to maintain NAT state. Note that no direct 1671 bubbles are necessary since the Proxy/Server is by definition not 1672 located behind a NAT. 1674 o else, if the original IP destination matches a non-MNP route in 1675 the IP forwarding table or an ADM-LLA assigned to the Proxy/ 1676 Server's OMNI interface, the Proxy/Server acting as a Relay 1677 presents the original IP packet to the network layer for local 1678 delivery or IP forwarding. 1680 o else, the Proxy/Server initiates address resolution as discussed 1681 in Section 3.14, while retaining initial original IP packets in a 1682 small queue awaiting address resolution completion. 1684 When the Proxy/Server receives a carrier packet with OAL destination 1685 set to an MNP-ULA that does not match the MSP, it accepts the carrier 1686 packet only if data origin authentication succeeds and if there is a 1687 network layer routing table entry for a GUA route that matches the 1688 MNP-ULA. If there is no route, the Proxy/Server drops the carrier 1689 packet; otherwise, it reassembles and decapsulates to obtain the 1690 original IP packet and acts as a Relay to present it to the network 1691 layer where it will be delivered according to standard IP forwarding. 1693 When the Proxy/Server receives a carrier packet from one of its 1694 Client neighbors with OAL destination set to another node, it 1695 forwards the packets via a matching neighbor cache entry or via the 1696 spanning tree if there is no matching entry. When the Proxy/Server 1697 receives a carrier packet with OAL destination set to the MNP-ULA of 1698 one of its Client neighbors established through RS/RA exchanges, it 1699 accepts the carrier packet only if data origin authentication 1700 succeeds. If the neighbor cache entry state is DEPARTED, the Proxy/ 1701 Server inserts an ORH that encodes the MNP-ULA destination suffix and 1702 changes the OAL destination address to the ADM-ULA of the new Proxy/ 1703 Server, then re-encapsulates the carrier packet and forwards it to a 1704 Bridge which will eventually deliver it to the new Proxy/Server. 1706 If the neighbor cache state for the MNP-ULA is REACHABLE, the Proxy/ 1707 Server forwards the carrier packets to the Client which then must 1708 reassemble. (Note that the Proxy/Server does not reassemble carrier 1709 packets not explicitly addressed to its own ADM-ULA, since routing 1710 could direct some of the carrier packet of the same original IP 1711 packet through a different Proxy/Server.) In that case, the Client 1712 may receive fragments that are smaller than its link MTU but can 1713 still be reassembled; if this proves inefficient, the Client can in 1714 the future elect to employ the Proxy/Server as a ROR instead of 1715 serving in that role on its own behalf. 1717 Note: Clients and their Proxy/Server peers can exchange original IP 1718 packets over ANET underlying interfaces without invoking the OAL, 1719 since the ANET is secured at the link and physical layers. By 1720 forwarding original IP packets without invoking the OAL, however, the 1721 Client and Proxy/Server can engage only in classical path MTU 1722 discovery since the packets are subject to loss and/or corruption due 1723 to the various per-link MTU limitations that may occur within the 1724 ANET. Moreover, the original IP packets do not include per-packet 1725 Identification values that can be used for data origin authentication 1726 and link-layer retransmission purposes, nor the OAL integrity check. 1727 The tradeoff therefore involves an assessment of the per-packet 1728 encapsulation overhead saved by bypassing the OAL vs. inheritance of 1729 classical network "brittleness". (Note however that ANET peers can 1730 send small original IP packets without invoking the OAL, while 1731 invoking the OAL for larger packets. This presents the beneficial 1732 aspects of both small packet efficiency and large packet robustness.) 1734 Note: When a Proxy/Server receives a (non-OAL) original IP packet 1735 from an ANET Client, or a carrier packet with OAL destination set to 1736 its own ADM-ULA from any Client, the Proxy/Server reassembles if 1737 necessary then performs ROS functions on behalf of the Client. The 1738 Client may at some later time begin sending carrier packets to the 1739 OAL address of the actual target instead of the Proxy/Server, at 1740 which point it may begin functioning as an ROS on its own behalf and 1741 thereby "override" the Proxy/Server's ROS role. 1743 Note: Proxy/Servers forward secure control plane carrier packets via 1744 the secured spanning tree and forwards other carrier packets via the 1745 unsecured spanning tree. When a Proxy/Server receives a carrier 1746 packet from the secured spanning tree, it considers the message as 1747 authentic without having to verify upper layer authentication 1748 signatures. When a Proxy/Server receives a carrier packet from the 1749 unsecured spanning tree, it verifies any upper layer authenticaton 1750 signatures and/or forwards the unsecured message toward the 1751 destination which must apply data origin authentication. 1753 Note: If the Proxy/Server has multiple original IP packets to send to 1754 the same neighbor, it can concatenate them in a single OAL super- 1755 packet [I-D.templin-6man-omni]. 1757 3.10.3. Bridge Forwarding Algorithm 1759 Bridges forward carrier packets the same as any IPv6 router. Bridges 1760 convey carrier packets and original IP packets that encapsulate IPv6 1761 ND control messages or routing protocol control messages using 1762 security encapsulations, and may convey packets that encapsulate 1763 ordinary data without including security encapsulations. When the 1764 Bridge receives a carrier packet or an original IP packet, it removes 1765 the outer *NET header and searches for a forwarding table entry that 1766 matches the OAL destination address. The Bridge then processes the 1767 packet as follows: 1769 o if the packet is a carrier packet with a destination that matches 1770 its ADM-ULA or the corresponding ADM-ULA Subnet Router Anycast 1771 address and the OAL header is followed by an ORH, the Bridge 1772 removes and sets aside the ORH then processes the carrier packet 1773 locally before forwarding. If the encapsulated packet is an 1774 atomic fragment containing an NA(NUD) message, the Bridge replaces 1775 the OMNI option Interface Attributes sub-option with information 1776 for its own interface while retaining the omIndex value supplied 1777 by the NA(NUD) message source. The Bridge next examines the ORH 1778 FMT code. If the code indicates the destination is a Client on 1779 the open *NET (or, a Client behind a NAT for which NAT traversal 1780 procedures have already converged) the Bridge writes the MNP-ULA 1781 formed from the ORH Destination Suffix into the OAL destination. 1782 The Bridge then re-encapsulates the carrier packet and forwards it 1783 to the ORH L2ADDR. For all other encapsulated packet cases, the 1784 Bridge instead writes the ADM-ULA formed from the ORH SRT/LHS into 1785 the OAL destination address and forwards the carrier packet to the 1786 ADM-ULA Proxy/Server while invoking NAT traversal procedures the 1787 same as for Proxy/Servers if necessary, noting that no direct 1788 bubbles are necessary since only the target Client and not the 1789 Bridge is behind a NAT 1791 o else, if the packet is a carrier packet with a destination that 1792 matches it ADM-ULA or the corresponding ADM-ULA Subnet Router 1793 Anycast address and the OAL header is not followed by an ORH, the 1794 Bridge submits the carrier packet for reassembly. When reassembly 1795 is complete, the Bridge submits the original packet to the IP 1796 layer to support local applications such as BGP routing protocol 1797 sessions. 1799 o else, if the packet is a carrier packet with a destination that 1800 matches a forwarding table entry the Bridge forwards the carrier 1801 packet via a secured tunnel to the next hop. (If the destination 1802 matches an MSP without matching an MNP, however, the Bridge 1803 instead drops the packet and returns an ICMP Destination 1804 Unreachable message subject to rate limiting - see: Section 3.11). 1806 o else, the Bridge drops the packet and returns an ICMP Destination 1807 Unreachable as above. 1809 As for any IP router, the Bridge decrements the OAL IPv6 header Hop 1810 Limit when it forwards the carrier packet and drops the packet if the 1811 Hop Limit reaches 0. Therefore, only the Hop Limit in the OAL header 1812 is decremented and not the TTL/Hop Limit in the original IP packet 1813 header. Bridges do not insert OAL/ORH headers themselves; instead, 1814 they act as IPv6 routers and forward carrier packets based on their 1815 destination addresses. 1817 Bridges forward carrier packets received from a first segment via the 1818 unsecured spanning tree to the next segment also via the unsecured 1819 spanning tree. Bridges forward carrier packets received from a first 1820 segment via the secured spanning tree to the next segment also via 1821 the secured spanning tree. Bridges use a single IPv6 routing table 1822 that always determines the same next hop for a given OAL destination, 1823 where the secured/unsecured spanning tree is determined through the 1824 selection of the underlying interface to be used for transmission 1825 (i.e., a secured tunnel or an open INET interface). 1827 3.11. OMNI Interface Error Handling 1829 When an AERO node admits an original IP packet into the OMNI 1830 interface, it may receive link-layer or network-layer error 1831 indications. 1833 A link-layer error indication is an ICMP error message generated by a 1834 router in the INET on the path to the neighbor or by the neighbor 1835 itself. The message includes an IP header with the address of the 1836 node that generated the error as the source address and with the 1837 link-layer address of the AERO node as the destination address. 1839 The IP header is followed by an ICMP header that includes an error 1840 Type, Code and Checksum. Valid type values include "Destination 1841 Unreachable", "Time Exceeded" and "Parameter Problem" 1842 [RFC0792][RFC4443]. (OMNI interfaces ignore link-layer IPv4 1843 "Fragmentation Needed" and IPv6 "Packet Too Big" messages for carrier 1844 packets that are no larger than the minimum/path MPS as discussed in 1845 Section 3.9, however these messages may provide useful hints of probe 1846 failures during path MPS probing.) 1848 The ICMP header is followed by the leading portion of the carrier 1849 packet that generated the error, also known as the "packet-in-error". 1850 For ICMPv6, [RFC4443] specifies that the packet-in-error includes: 1851 "As much of invoking packet as possible without the ICMPv6 packet 1852 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 1853 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 1855 "Internet Header + 64 bits of Original Data Datagram", however 1856 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 1857 ICMP datagram SHOULD contain as much of the original datagram as 1858 possible without the length of the ICMP datagram exceeding 576 1859 bytes". 1861 The link-layer error message format is shown in Figure 5 (where, "L2" 1862 and "L3" refer to link-layer and network-layer, respectively): 1864 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1865 ~ ~ 1866 | L2 IP Header of | 1867 | error message | 1868 ~ ~ 1869 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1870 | L2 ICMP Header | 1871 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1872 ~ ~ P 1873 | carrier packet *NET and OAL | a 1874 | encapsulation headers | c 1875 ~ ~ k 1876 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 1877 ~ ~ t 1878 | original IP packet headers | 1879 | (first-fragment only) | i 1880 ~ ~ n 1881 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1882 ~ ~ e 1883 | Portion of the body of | r 1884 | the original IP packet | r 1885 | (all fragments) | o 1886 ~ ~ r 1887 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 1889 Figure 5: OMNI Interface Link-Layer Error Message Format 1891 The AERO node rules for processing these link-layer error messages 1892 are as follows: 1894 o When an AERO node receives a link-layer Parameter Problem message, 1895 it processes the message the same as described as for ordinary 1896 ICMP errors in the normative references [RFC0792][RFC4443]. 1898 o When an AERO node receives persistent link-layer Time Exceeded 1899 messages, the IP ID field may be wrapping before earlier fragments 1900 awaiting reassembly have been processed. In that case, the node 1901 should begin including integrity checks and/or institute rate 1902 limits for subsequent packets. 1904 o When an AERO node receives persistent link-layer Destination 1905 Unreachable messages in response to carrier packets that it sends 1906 to one of its neighbor correspondents, the node should process the 1907 message as an indication that a path may be failing, and 1908 optionally initiate NUD over that path. If it receives 1909 Destination Unreachable messages over multiple paths, the node 1910 should allow future carrier packets destined to the correspondent 1911 to flow through a default route and re-initiate route 1912 optimization. 1914 o When an AERO Client receives persistent link-layer Destination 1915 Unreachable messages in response to carrier packets that it sends 1916 to one of its neighbor Proxy/Servers, the Client should mark the 1917 path as unusable and use another path. If it receives Destination 1918 Unreachable messages on many or all paths, the Client should 1919 associate with a new Proxy/Server and release its association with 1920 the old Proxy/Server as specified in Section 3.16.5. 1922 o When an AERO Proxy/Server receives persistent link-layer 1923 Destination Unreachable messages in response to carrier packets 1924 that it sends to one of its neighbor Clients, the Proxy/Server 1925 should mark the underlying path as unusable and use another 1926 underlying path. 1928 o When an AERO Proxy/Server receives link-layer Destination 1929 Unreachable messages in response to a carrier packet that it sends 1930 to one of its permanent neighbors, it treats the messages as an 1931 indication that the path to the neighbor may be failing. However, 1932 the dynamic routing protocol should soon reconverge and correct 1933 the temporary outage. 1935 When an AERO Bridge receives a carrier packet for which the network- 1936 layer destination address is covered by an MSP, the Bridge drops the 1937 packet if there is no more-specific routing information for the 1938 destination and returns a network-layer Destination Unreachable 1939 message subject to rate limiting. The Bridge writes the network- 1940 layer source address of the original IP packet as the destination 1941 address and uses one of its non link-local addresses as the source 1942 address of the message. 1944 When an AERO node receives a carrier packet for which reassembly is 1945 currently congested, it returns a network-layer Packet Too Big (PTB) 1946 message as discussed in [I-D.templin-6man-omni] (note that the PTB 1947 messages could indicate either "hard" or "soft" errors). 1949 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 1951 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 1952 coordinated as discussed in the following Sections. 1954 3.12.1. AERO Service Model 1956 Each AERO Proxy/Server on the OMNI link is configured to facilitate 1957 Client prefix delegation/registration requests. Each Proxy/Server is 1958 provisioned with a database of MNP-to-Client ID mappings for all 1959 Clients enrolled in the AERO service, as well as any information 1960 necessary to authenticate each Client. The Client database is 1961 maintained by a central administrative authority for the OMNI link 1962 and securely distributed to all Proxy/Servers, e.g., via the 1963 Lightweight Directory Access Protocol (LDAP) [RFC4511], via static 1964 configuration, etc. Clients receive the same service regardless of 1965 the Proxy/Servers they select. 1967 AERO Clients and Proxy/Servers use ND messages to maintain neighbor 1968 cache entries. AERO Proxy/Servers configure their OMNI interfaces as 1969 advertising NBMA interfaces, and therefore send unicast RA messages 1970 with a short Router Lifetime value (e.g., ReachableTime seconds) in 1971 response to a Client's RS message. Thereafter, Clients send 1972 additional RS messages to keep Proxy/Server state alive. 1974 AERO Clients and Proxy/Servers include prefix delegation and/or 1975 registration parameters in RS/RA messages (see 1976 [I-D.templin-6man-omni]). The ND messages are exchanged between 1977 Client and Proxy/Server according to the prefix management schedule 1978 required by the service. If the Client knows its MNP in advance, it 1979 can employ prefix registration by including its MNP-LLA as the source 1980 address of an RS message and with an OMNI option with valid prefix 1981 registration information for the MNP. If the Proxy/Server accepts 1982 the Client's MNP assertion, it injects the MNP into the routing 1983 system and establishes the necessary neighbor cache state. If the 1984 Client does not have a pre-assigned MNP, it can instead employ prefix 1985 delegation by including the unspecified address (::) as the source 1986 address of an RS message and with an OMNI option with prefix 1987 delegation parameters to request an MNP. 1989 The following sections specify the Client and Proxy/Server behavior. 1991 3.12.2. AERO Client Behavior 1993 AERO Clients discover the addresses of Proxy/Servers in a similar 1994 manner as described in [RFC5214]. Discovery methods include static 1995 configuration (e.g., from a flat-file map of Proxy/Server addresses 1996 and locations), or through an automated means such as Domain Name 1997 System (DNS) name resolution [RFC1035]. Alternatively, the Client 1998 can discover Proxy/Server addresses through a layer 2 data link login 1999 exchange, or through a unicast RA response to a multicast/anycast RS 2000 as described below. In the absence of other information, the Client 2001 can resolve the DNS Fully-Qualified Domain Name (FQDN) 2002 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 2003 text string and "[domainname]" is a DNS suffix for the OMNI link 2004 (e.g., "example.com"). 2006 To associate with a Proxy/Server, the Client acts as a requesting 2007 router to request MNPs. The Client prepares an RS message with 2008 prefix management parameters and includes a Nonce and Timestamp 2009 option if the Client needs them to correlate RA replies. If the 2010 Client already knows the Proxy/Server's ADM-LLA, it includes the LLA 2011 as the network-layer destination address; otherwise, the Client 2012 includes the (link-local) All-Routers multicast as the network-layer 2013 destination. If the Client already knows its own MNP-LLA, it can use 2014 the MNP-LLA as the network-layer source address and include an OMNI 2015 option with prefix registration information. Otherwise, the Client 2016 uses the unspecified address (::) as the network-layer source address 2017 and includes prefix delegation parameters in the OMNI option (see: 2018 [I-D.templin-6man-omni]). The Client includes Interface Attributes 2019 corresponding to the underlying interface over which it will send the 2020 RS message, and MAY include additional Interface Attributes specific 2021 to other underlying interfaces. 2023 For INET Clients, the Client must ensure that the RS message is no 2024 larger than the minimum/path MPS for the chosen Proxy/Server and must 2025 include an authentication signature that the Proxy/Server can verify. 2026 The Client next applies OAL encapsulation such that the entire RS 2027 message fits within an OAL First Fragment (i.e., as an atomic 2028 fragment) while including an unpredictable OAL Identification number 2029 selected per [RFC7739] that will serve as the window start 2030 Identification value for future packets it will send/accept with its 2031 own MNP-ULA and the Proxy/Server's ADM-ULA as the OAL addresses. 2032 (The Proxy/Server in turn caches the Identification number as start 2033 value for future packets it will accept/send with its own ADM-ULA and 2034 the Client's MNP-ULA as the OAL addresses.) 2036 The Client then sends the RS message (either directly via Direct 2037 interfaces, via a VPN for VPNed interfaces, via an access router for 2038 ANET interfaces or via INET encapsulation for INET interfaces) while 2039 using OAL encapsulation/fragmentation, then waits for an RA message 2040 reply (see Section 3.12.3). The Client retries up to 2041 MAX_RTR_SOLICITATIONS times until an RA is received. If the Client 2042 receives no RAs, or if it receives an RA with Router Lifetime set to 2043 0, the Client SHOULD abandon attempts through the first Proxy/Server 2044 and try another Proxy/Server. Otherwise, the Client processes the 2045 prefix information found in the RA message. 2047 When the Client processes an RA, it first performs OAL reassembly and 2048 decapsulation then creates a neighbor cache entry with the Proxy/ 2049 Server's ADM-LLA as the network-layer address and the Proxy/Server's 2050 encapsulation and/or link-layer addresses as the link-layer address. 2051 The Client next records the RA Router Lifetime field value in the 2052 neighbor cache entry as the time for which the Proxy/Server has 2053 committed to maintaining the MNP in the routing system via this 2054 underlying interface, and caches the other RA configuration 2055 information including Cur Hop Limit, M and O flags, Reachable Time 2056 and Retrans Timer. The Client then autoconfigures MNP-LLAs for any 2057 delegated MNPs and assigns them to the OMNI interface. The Client 2058 also caches any MSPs included in Route Information Options (RIOs) 2059 [RFC4191] as MSPs to associate with the OMNI link, and assigns the 2060 MTU value in the MTU option to the underlying interface. 2062 The Client then registers additional underlying interfaces with the 2063 Proxy/Server by sending RS messages via each additional interface as 2064 described above. The RS messages include the same parameters as for 2065 the initial RS/RA exchange, but with destination address set to the 2066 Proxy/Server's ADM-LLA. The Client finally sub-delegates the MNPs to 2067 its attached EUNs and/or the Client's own internal virtual interfaces 2068 as described in [I-D.templin-v6ops-pdhost] to support the Client's 2069 downstream attached "Internet of Things (IoT)". The Client then 2070 sends additional RS messages over each underlying interface before 2071 the Router Lifetime received for that interface expires. 2073 After the Client registers its underlying interfaces, it may wish to 2074 change one or more registrations, e.g., if an interface changes 2075 address or becomes unavailable, if QoS preferences change, etc. To 2076 do so, the Client prepares an RS message to send over any available 2077 underlying interface as above. The RS includes an OMNI option with 2078 prefix registration/delegation information, with Interface Attributes 2079 specific to the selected underlying interface, and with any 2080 additional Interface Attributes specific to other underlying 2081 interfaces. When the Client receives the Proxy/Server's RA response, 2082 it has assurance that the Proxy/Server has been updated with the new 2083 information. 2085 If the Client wishes to discontinue use of a Proxy/Server it issues 2086 an RS message over any underlying interface with an OMNI option with 2087 a prefix release indication. When the Proxy/Server processes the 2088 message, it releases the MNP, sets the neighbor cache entry state for 2089 the Client to DEPARTED and returns an RA reply with Router Lifetime 2090 set to 0. After a short delay (e.g., 2 seconds), the Proxy/Server 2091 withdraws the MNP from the routing system. 2093 3.12.3. AERO Proxy/Server Behavior 2095 AERO Proxy/Servers act as IP routers and support a prefix delegation/ 2096 registration service for Clients. Proxy/Servers arrange to add their 2097 ADM-LLAs to a static map of Proxy/Server addresses for the link and/ 2098 or the DNS resource records for the FQDN 2099 "linkupnetworks.[domainname]" before entering service. Proxy/Server 2100 addresses should be geographically and/or topologically referenced, 2101 and made available for discovery by Clients on the OMNI link. 2103 When a Proxy/Server receives a prospective Client's RS message on its 2104 OMNI interface, it SHOULD return an immediate RA reply with Router 2105 Lifetime set to 0 if it is currently too busy or otherwise unable to 2106 service the Client. Otherwise, the Proxy/Server performs OAL 2107 reassembly and decapsulation, then authenticates the RS message and 2108 processes the prefix delegation/registration parameters. The Proxy/ 2109 Server first determines the correct MNPs to provide to the Client by 2110 processing the MNP-LLA prefix parameters and/or the DHCPv6 OMNI sub- 2111 option. When the Proxy/Server returns the MNPs, it also creates a 2112 forwarding table entry for the MNP-ULA corresponding to each MNP so 2113 that the MNPs are propagated into the routing system (see: 2114 Section 3.2.3). For IPv6, the Proxy/Server creates an IPv6 2115 forwarding table entry for each MNP. For IPv4, the Proxy/Server 2116 creates an IPv6 forwarding table entry with the IPv4-compatibility 2117 MNP-ULA prefix corresponding to the IPv4 address. 2119 The Proxy/Server next creates a neighbor cache entry for the Client 2120 using the base MNP-LLA as the network-layer address and with lifetime 2121 set to no more than the smallest prefix lifetime. Next, the Proxy/ 2122 Server updates the neighbor cache entry by recording the information 2123 in each Interface Attributes sub-option in the RS OMNI option. The 2124 Proxy/Server also records the actual OAL/*NET addresses in the 2125 neighbor cache entry. For RS messages encapsulated as carrier 2126 packets, the Proxy/Server also records the OAL Identification number 2127 as the starting value for the window of future packets it will send/ 2128 accept with its own ADM-ULA and the Client's MNP-ULA as the OAL 2129 addresses. (The Client in turn caches the Identification number as 2130 start value for future packets it will accept/send with its own MNP- 2131 ULA and the Proxy/Server's ADM-ULA as the OAL addresses.) 2133 Next, the Proxy/Server prepares an RA message using its ADM-LLA as 2134 the network-layer source address and the network-layer source address 2135 of the RS message as the network-layer destination address. The 2136 Proxy/Server sets the Router Lifetime to the time for which it will 2137 maintain both this underlying interface individually and the neighbor 2138 cache entry as a whole. The Proxy/Server also sets Cur Hop Limit, M 2139 and O flags, Reachable Time and Retrans Timer to values appropriate 2140 for the OMNI link. The Proxy/Server includes the MNPs, any other 2141 prefix management parameters and an OMNI option with no Interface 2142 Attributes but with an Origin Indication sub-option per 2143 [I-D.templin-6man-omni] with the mapped and obfuscated Port Number 2144 and IP address corresponding to the Client's own INET address in the 2145 case of INET Clients or to the Proxy/Server's INET-facing address for 2146 all other Clients. The Proxy/Server then includes one or more RIOs 2147 that encode the MSPs for the OMNI link, plus an MTU option (see 2148 Section 3.9). The Proxy/Server finally forwards the message to the 2149 Client using OAL encapsulation/fragmentation as necessary with an OAL 2150 Identification value that matches the RS. 2152 After the initial RS/RA exchange, the Proxy/Server maintains a 2153 ReachableTime timer for each of the Client's underlying interfaces 2154 individually (and for the Client's neighbor cache entry collectively) 2155 set to expire after ReachableTime seconds. If the Client (or Proxy) 2156 issues additional RS messages, the Proxy/Server sends an RA response 2157 and resets ReachableTime. If the Proxy/Server receives an ND message 2158 with a prefix release indication it sets the Client's neighbor cache 2159 entry to the DEPARTED state and withdraws the MNP from the routing 2160 system after a short delay (e.g., 2 seconds). If ReachableTime 2161 expires before a new RS is received on an individual underlying 2162 interface, the Proxy/Server marks the interface as DOWN. If 2163 ReachableTime expires before any new RS is received on any individual 2164 underlying interface, the Proxy/Server sets the neighbor cache entry 2165 state to STALE and sets a 10 second timer. If the Proxy/Server has 2166 not received a new RS or ND message with a prefix release indication 2167 before the 10 second timer expires, it deletes the neighbor cache 2168 entry and withdraws the MNP from the routing system. 2170 The Proxy/Server processes any ND messages pertaining to the Client 2171 and returns an NA/RA reply in response to solicitations. The Proxy/ 2172 Server may also issue unsolicited RA messages, e.g., with reconfigure 2173 parameters to cause the Client to renegotiate its prefix delegation/ 2174 registrations, with Router Lifetime set to 0 if it can no longer 2175 service this Client, etc. Finally, If the neighbor cache entry is in 2176 the DEPARTED state, the Proxy/Server deletes the entry after 2177 DepartTime expires. 2179 Note: Clients SHOULD notify former Proxy/Servers of their departures, 2180 but Proxy/Servers are responsible for expiring neighbor cache entries 2181 and withdrawing routes even if no departure notification is received 2182 (e.g., if the Client leaves the network unexpectedly). Proxy/Servers 2183 SHOULD therefore set Router Lifetime to ReachableTime seconds in 2184 solicited RA messages to minimize persistent stale cache information 2185 in the absence of Client departure notifications. A short Router 2186 Lifetime also ensures that proactive RS/RA messaging between Clients 2187 and Proxy/Servers will keep any NAT state alive (see above). 2189 Note: All Proxy/Servers on an OMNI link MUST advertise consistent 2190 values in the RA Cur Hop Limit, M and O flags, Reachable Time and 2191 Retrans Timer fields the same as for any link, since unpredictable 2192 behavior could result if different Proxy/Servers on the same link 2193 advertised different values. 2195 3.12.3.1. DHCPv6-Based Prefix Registration 2197 When a Client is not pre-provisioned with an MNP-LLA, it will need 2198 for the Proxy/Server to select one or more MNPs on its behalf and set 2199 up the correct state in the AERO routing service. (A Client with a 2200 pre-provisioned MNP may also request the Proxy/Server to select 2201 additional MNPs.) The DHCPv6 service [RFC8415] is used to support 2202 this requirement. 2204 When a Client needs to have the Proxy/Server select MNPs, it sends an 2205 RS message with source address set to the unspecified address (::) 2206 and with an OMNI option that includes a DHCPv6 message sub-option 2207 with DHCPv6 Prefix Delegation (DHCPv6-PD) parameters. When the 2208 Proxy/Server receives the RS message, it extracts the DHCPv6-PD 2209 message from the OMNI option. 2211 The Proxy/Server then acts as a "Proxy DHCPv6 Client" in a message 2212 exchange with the locally-resident DHCPv6 server, which delegates 2213 MNPs and returns a DHCPv6-PD Reply message. (If the Proxy/Server 2214 wishes to defer creation of MN state until the DHCPv6-PD Reply is 2215 received, it can instead act as a Lightweight DHCPv6 Relay Agent per 2216 [RFC6221] by encapsulating the DHCPv6-PD message in a Relay-forward/ 2217 reply exchange with Relay Message and Interface ID options.) 2219 When the Proxy/Server receives the DHCPv6-PD Reply, it adds a route 2220 to the routing system and creates an MNP-LLA based on the delegated 2221 MNP. The Proxy/Server then sends an RA back to the Client with the 2222 (newly-created) MNP-LLA as the destination address and with the 2223 DHCPv6-PD Reply message coded in the OMNI option. When the Client 2224 receives the RA, it creates a default route, assigns the Subnet 2225 Router Anycast address and sets its MNP-LLA based on the delegated 2226 MNP. 2228 Note: See [I-D.templin-6man-omni] for an MNP delegation alternative 2229 in which the Client can optionally avoid including a DHCPv6 message 2230 sub-option. Namely, when the Client requests a single MNP it can set 2231 the RS source to the unspecified address (::) and include a Node 2232 Identification sub-option and Preflen in the OMNI option (but with no 2233 DHCPv6 message sub-option). When the Proxy/Server receives the RS 2234 message, it forwards a self-generated DHCPv6 Solicit message to the 2235 DHCPv6 server on behalf of the Client. When the Proxy/Server 2236 receives the DHCPv6 Reply, it prepares an RA message with an OMNI 2237 option with Preflen information (but with no DHCPv6 message sub- 2238 option), then places the (newly-created) MNP-LLA in the RA 2239 destination address and returns the message to the Client. 2241 3.13. The AERO Proxy Function 2243 Clients connect to the OMNI link via Proxy/Servers, with one Proxy/ 2244 Server for each underlying interface. Each of the Client's Proxy/ 2245 Servers must be informed of all of the Client's additional underlying 2246 interfaces. For Clients on Direct and VPNed underlying interfaces 2247 the Proxy/Server "A" for that interface is directly connected, for 2248 Clients on ANET underlying interfaces Proxy/Server "A" is located on 2249 the ANET/INET boundary, and for Clients on INET underlying interfaces 2250 Proxy/Server "A" is located somewhere in the connected Internetwork. 2251 When the Client registers with Proxy/Server "A", it must also report 2252 the registration to any other Proxy/Servers for other underlying 2253 interfaces "B", "C", "D", etc. for which an underlying interface 2254 relationship has already been established. The Proxy/Server 2255 satisfies these requirements as follows: 2257 o when Proxy/Server "A" receives an RS message from a new Client, it 2258 first authenticates the message then examines the network-layer 2259 destination address. If the destination address is Proxy/Server 2260 "A"'s ADM-LLA or (link-local) All-Routers multicast, Proxy/Server 2261 "A" creates a proxy neighbor cache entry and caches the Client 2262 link-layer addresses along with the OMNI option information and 2263 any other identifying information including OAL Identification 2264 values, Client Identifiers, Nonce values, etc. If the RS message 2265 destination was the ADM-LLA of a different Proxy/Server "B" (or, 2266 if the OMNI option included an MS-Register sub-option with the 2267 ADM-LLA of a different Proxy/Server "B"), Proxy/Server "A" 2268 encapsulates a proxyed version of the RS message in an OAL header 2269 with source set to Proxy/Server "A"'s ADM-ULA and destination set 2270 to Proxy/Server "B"'s ADM-ULA. Proxy/Server "A" also includes an 2271 OMNI header with an Interface Attributes option that includes its 2272 own INET address plus a unique UDP Port Number for this Client, 2273 then forwards the message into the OMNI link spanning tree. 2274 (Note: including a unique Port Number allows Proxy/Server "B" to 2275 distinguish different Clients located behind the same Proxy/Server 2276 "A" at the link-layer, whereas the link-layer addresses would 2277 otherwise be indistinguishable.) 2279 o when the Proxy/Server "B" receives the RS, it authenticates the 2280 message then creates or updates a neighbor cache entry for the 2281 Client with Proxy/Server "A"'s ADM-ULA, INET address and UDP Port 2282 Number as the link-layer address information. Proxy/Server "B" 2283 then sends an RA message back to Proxy/Server "A" via the spanning 2284 tree. 2286 o when Proxy/Server "A" receives the RA, it authenticates the 2287 message and matches it with the proxy neighbor cache entry created 2288 by the RS. Proxy/Server "A" then caches the prefix information as 2289 a mapping from the Client's MNPs to the Client's link-layer 2290 address, caches the Proxy/Server's advertised Router Lifetime and 2291 sets the neighbor cache entry state to REACHABLE. The Proxy/ 2292 Server then optionally rewrites the Router Lifetime and forwards 2293 the (proxyed) message to the Client. The Proxy/Server finally 2294 includes an MTU option (if necessary) with an MTU to use for the 2295 underlying ANET interface. 2297 o The Client repeats this process with each Proxy/Server "B", "C", 2298 "D" for each of its additional underlying interfaces. 2300 After the initial RS/RA exchanges each Proxy/Server forwards any of 2301 the Client's carrier packets for which there is no matching neighbor 2302 cache entry to a Bridge using OAL encapsulation with its own ADM-ULA 2303 as the source and the MNP-ULA corresponding to the Client as the 2304 destination. The Proxy/Server instead forwards any carrier packets 2305 destined to a neighbor cache target directly to the target according 2306 to the OAL/link-layer information - the process of establishing 2307 neighbor cache entries is specified in Section 3.14. 2309 While the Client is still associated with each Proxy/Server "A", "A" 2310 can send NS, RS and/or unsolicited NA messages to update the neighbor 2311 cache entries of other AERO nodes on behalf of the Client and/or to 2312 convey QoS updates. This allows for higher-frequency Proxy-initiated 2313 RS/RA messaging over well-connected INET infrastructure supplemented 2314 by lower-frequency Client-initiated RS/RA messaging over constrained 2315 ANET data links. 2317 If any Proxy/Server "B", "C", "D" ceases to send solicited 2318 advertisements, Proxy/Server "A" sends unsolicited RAs to the Client 2319 with destination set to (link-local) All-Nodes multicast and with 2320 Router Lifetime set to zero to inform Clients that a Proxy/Server has 2321 failed. Although Proxy/Server "A" can engage in ND exchanges on 2322 behalf of the Client, the Client can also send ND messages on its own 2323 behalf, e.g., if it is in a better position than "A" to convey QoS 2324 changes, etc. The ND messages sent by the Client include the 2325 Client's MNP-LLA as the source in order to differentiate them from 2326 the ND messages sent by Proxy/Server "A". 2328 If the Client becomes unreachable over an underlying interface, 2329 Proxy/Server "A" sets the neighbor cache entry state to DEPARTED and 2330 retains the entry for DepartTime seconds. While the state is 2331 DEPARTED, Proxy/Server "A" forwards any carrier packets destined to 2332 the Client to a Bridge via OAL/ORH encapsulation. When DepartTime 2333 expires, Proxy/Server "A" deletes the neighbor cache entry and 2334 discards any further carrier packets destined to this (now forgotten) 2335 Client. 2337 In some ANETs that employ a Proxy/Server, the Client's MNP can be 2338 injected into the ANET routing system. In that case, the Client can 2339 send original IP packets without invoking the OAL so that the ANET 2340 routing system transports the original IP packets to the Proxy. This 2341 can be very beneficial, e.g., if the Client connects to the ANET via 2342 low-end data links such as some aviation wireless links. 2344 If the ANET first-hop access router is on the same underlying link as 2345 the Client and recognizes the AERO/OMNI protocol, the Client can 2346 avoid OAL encapsulation for both its control and data messages. When 2347 the Client connects to the link, it can send an unencapsulated RS 2348 message with source address set to its own MNP-LLA (or to a Temporary 2349 LLA), and with destination address set to the ADM-LLA of the Client's 2350 selected Proxy/Server or to (link-local) All-Routers multicast. The 2351 Client includes an OMNI option formatted as specified in 2352 [I-D.templin-6man-omni]. The Client then sends the unencapsulated RS 2353 message, which will be intercepted by the AERO-Aware access router. 2355 The ANET access router then performs OAL encapsulation on the RS 2356 message and forwards it to a Proxy/Server at the ANET/INET boundary. 2357 When the access router and Proxy/Server are one and the same node, 2358 the Proxy/Server would share and underlying link with the Client but 2359 its message exchanges with outside correspondents would need to pass 2360 through a security gateway at the ANET/INET border. The method for 2361 deploying access routers and Proxys (i.e. as a single node or 2362 multiple nodes) is an ANET-local administrative consideration. 2364 Note: The Proxy/Server can apply packing as discussed in 2365 [I-D.templin-6man-omni] if an opportunity arises to concatenate 2366 multiple original IP packets that will be destined to the same 2367 neighbor. 2369 3.13.1. Detecting and Responding to Proxy/Server Failures 2371 In environments where fast recovery from Proxy/Server failure is 2372 required, Proxy/Server "A" SHOULD use proactive Neighbor 2373 Unreachability Detection (NUD) to track peer Proxy/Server "B" 2374 reachability in a similar fashion as for Bidirectional Forwarding 2375 Detection (BFD) [RFC5880]. Proxy/Server "A" can then quickly detect 2376 and react to failures so that cached information is re-established 2377 through alternate paths. The NUD control messaging is carried only 2378 over well-connected ground domain networks (i.e., and not low-end 2379 aeronautical radio links) and can therefore be tuned for rapid 2380 response. 2382 Proxy/Server "A" performs proactive NUD with peer Proxy/Server "B" 2383 for which there are currently active Clients by sending continuous NS 2384 messages in rapid succession, e.g., one message per second. Proxy/ 2385 Server "A" sends the NS message via the spanning tree with its own 2386 ADM-LLA as the source and the ADM-LLA of the peer Proxy/Server "B" as 2387 the destination. When Proxy/Server "A" is also sending RS messages 2388 to the peer Proxy/Server "B" on behalf of ANET Clients, the resulting 2389 RA responses can be considered as equivalent hints of forward 2390 progress. This means that Proxy/Server "B" need not also send a 2391 periodic NS if it has already sent an RS within the same period. If 2392 the peer Proxy/Server "B" fails (i.e., if "A" ceases to receive 2393 advertisements), Proxy/Server "A" can quickly inform Clients by 2394 sending multicast RA messages on the ANET interface. 2396 Proxy/Server "A" sends RA messages on the ANET interface with source 2397 address set to Proxy/Server "B"'s address, destination address set to 2398 (link-local) All-Nodes multicast, and Router Lifetime set to 0. 2399 Proxy/Server "A" SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA messages 2400 separated by small delays [RFC4861]. Any Clients on the ANET that 2401 had been using the failed Proxy/Server "B" will receive the RA 2402 messages and associate with a new Proxy/Server. 2404 3.13.2. Point-to-Multipoint Proxy/Server Coordination 2406 In environments where Client messaging over ANETs is bandwidth- 2407 limited and/or expensive, Clients can enlist the services of Proxy/ 2408 Server "A" to coordinate with multiple Proxy/Servers "B", "C", "D" 2409 etc. in a single RS/RA message exchange. The Client can send a 2410 single RS message to (link-local) All-Routers multicast that includes 2411 the ID's of multiple Proxy/Servers in MS-Register sub-options of the 2412 OMNI option. 2414 When Proxy/Server "A" receives the RS and processes the OMNI option, 2415 it sends a separate RS to each MS-Register Proxy/Server ID. When 2416 Proxy/Server "A" receives an RA, it can optionally return an 2417 immediate "singleton" RA to the Client or record the Proxy/Server's 2418 ID for inclusion in a pending "aggregate" RA message. Proxy/Server 2419 "A" can then return aggregate RA messages to the Client including 2420 multiple Proxy/Server IDs in order to conserve bandwidth. Each RA 2421 includes a proper subset of the Proxy/Server IDs from the original RS 2422 message, and Proxy/Server "A" must ensure that the message contents 2423 of each RA are consistent with the information received from the 2424 (aggregated) additional Proxy/Servers. 2426 Clients can thereafter employ efficient point-to-multipoint Proxy/ 2427 Server coordination under the assistance of Proxy/Server "A" to 2428 reduce the number of messages sent over the ANET while enlisting the 2429 support of multiple Proxy/Servers for fault tolerance. Clients can 2430 further include MS-Release sub-options in IPv6 ND messages to request 2431 Proxy/Server "A" to release from former Proxy/Servers via the 2432 procedures discussed in Section 3.16.5. 2434 The OMNI interface specification [I-D.templin-6man-omni] provides 2435 further discussion of the RS/RA messaging involved in point-to- 2436 multipoint coordination. 2438 3.14. AERO Route Optimization 2440 While AERO nodes can always send data packets over strict spanning 2441 tree paths, route optimization should be performed while carrier 2442 packets are flowing between a source and target node. Route 2443 optimization is based on asymmetric IPv6 ND Address Resolution 2444 messaging between a Route Optimization Source (ROS) and Route 2445 Optimization Responder (ROR), and later extended to the target using 2446 IPv6 ND Neighbor Unreachability Detection messaging. Route 2447 optimization is initiated by the first eligible ROS closest to the 2448 source as follows: 2450 o For Clients on VPNed and Direct interfaces, the Proxy/Server is 2451 the ROS. 2453 o For Clients on ANET interfaces, either the Client or the Proxy/ 2454 Server may be the ROS. 2456 o For Clients on INET interfaces, the Client itself is the ROS. 2458 o For correspondent nodes on INET/EUN interfaces serviced by a 2459 Relay, the Relay is the ROS. 2461 The route optimization procedure is conducted between the ROS and the 2462 target Proxy/Server/Relay acting as an ROR (the target may either be 2463 a MNP Client serviced by a Proxy/Server or a non-MNP correspondent 2464 reachable via a Relay). Note that in this arrangement the ROS is 2465 always the Client or Proxy/Server/Relay nearest the source over the 2466 selected source underlying interface, while the ROR is always a 2467 Proxy/Server/Relay that services the target regardless of the target 2468 underlying interface. 2470 The procedures are specified in the following sections. 2472 3.14.1. Route Optimization Initiation 2474 When an original IP packet from a source node destined to a target 2475 node arrives, the ROS checks for a neighbor cache entry with an MNP- 2476 LLA that matches the target destination. If there is a neighbor 2477 cache entry in the REACHABLE state, the ROS invokes the OAL and 2478 forwards the resulting carrier packets according to the cached state 2479 and returns from processing. Otherwise, if there is no neighbor 2480 cache entry the ROS creates one in the INCOMPLETE state. 2482 The ROS next places the original IP packet on a short queue then 2483 sends an NS message for Address Resolution (NS(AR)) to receive a 2484 solicited NA(AR) message from a ROR. The NS(AR) message must be no 2485 larger than the minimum/path MPS so that its entire contents will fit 2486 in an OAL first fragment (i.e., as an atomic fragment). The ROS 2487 prepares an NS(AR) that includes: 2489 o the LLA of the ROS as the source address. 2491 o the MNP-LLA corresponding to the original IP packet's destination 2492 as the Target Address, e.g., for 2001:db8:1:2::10:2000 the Target 2493 Address is fe80::2001:db8:1:2. 2495 o the Solicited-Node multicast address [RFC4291] formed from the 2496 lower 24 bits of the original IP packet's destination as the 2497 destination address, e.g., for 2001:db8:1:2::10:2000 the NS 2498 destination address is ff02:0:0:0:0:1:ff10:2000. 2500 The NS(AR) message also includes an OMNI option with an Interface 2501 Attributes entry for the sending interface, with S/T-omIndex set to 0 2502 and with Preflen set to the prefix length associated with the NS(AR) 2503 source. The ROS then submits the NS(AR) message for OAL 2504 encapsulation as an atomic fragment, with OAL source set to its own 2505 ULA and OAL destination set to the ULA corresponding to the target, 2506 and with an unpredictable initial Identification value. The ROS 2507 caches the initial Identification value in the (newly-created) 2508 neighbor cache entry as the starting sequence number for the "send" 2509 window for future carrier packets sent to this target via the 2510 responding ROR. 2512 The ROS then sends the resulting carrier packet into the spanning 2513 tree without decrementing the network-layer TTL/Hop Limit field. 2514 (When the ROS is an INET Client, it instead must first sign the 2515 NS(AR) message and send the resulting carrier packet to the ADM-ULA 2516 of one of its current Proxy/Servers which then verifies the NS(AR) 2517 signature and forwards the carrier packet over the secured spanning 2518 tree on behalf of the Client.) 2520 3.14.2. Relaying the NS(AR) *NET Packet(s) 2522 When the Bridge receives the carrier packet containing the RS from 2523 the ROS, it discards the *NET headers and determines the next hop by 2524 consulting its standard IPv6 forwarding table for the OAL header 2525 destination address. The Bridge then decrements the OAL header Hop- 2526 Limit, re-encapsulates the carrier packet and forwards it via the 2527 secured spanning tree the same as for any IPv6 router, where it may 2528 traverse multiple OMNI link segments. The final-hop Bridge will 2529 deliver the carrier packet via the secured spanning tree to a Proxy/ 2530 Server or Relay that services the target. 2532 3.14.3. Processing the NS(AR) and Sending the NA(AR) 2534 When the target Proxy/Server (or Relay) receives the carrier packet, 2535 it examines the enclosed atomic fragment to determine that it 2536 contains an NS(AR) then examines the NS(AR) target to determine 2537 whether it has a matching neighbor cache entry and/or non-MNP route. 2538 If there is no match, the Proxy/Server drops the message. Otherwise, 2539 the Proxy/Server/Relay continues processing as follows: 2541 o if the NS(AR) target matches a Client neighbor cache entry in the 2542 DEPARTED state, the Proxy/Server inserts an ORH with destination 2543 prefix set to the lower 64 bits of the Client's MNP-ULA and sets 2544 the destination address to the ADM-ULA of the Client's new Proxy/ 2545 Server. The (old) Proxy/Server then re-encapsulates the carrier 2546 packet, forwards it over the secured spanning tree and returns 2547 from processing. 2549 o If the NS(AR) target matches the MNP-LLA of a Client neighbor 2550 cache entry in the REACHABLE state, the Proxy/Server makes note of 2551 whether the NS arrived from the secured or unsecured spanning 2552 tree, then acts as an ROR to provide route optimization 2553 information on behalf of the Client. (Note that if the message 2554 arrived via the secured spanning tree the ROR need not perform 2555 further authentication, but if it arrived over an open INET 2556 underlying interface it must check for and verify the message 2557 authentication signature before accepting.) 2559 o If the NS(AR) target matches one of its non-MNP routes, the Relay 2560 acts as an ROR since it serves as a router to forward IP packets 2561 toward the (fixed network) target at the network layer. 2563 The ROR next checks the target neighbor cache entry for a Report List 2564 entry that matches the NS(AR) source LLA/ULA of the ROS. If there is 2565 a Report List entry, the ROR accepts the NS(AR) only if the OAL 2566 Identification value is within the "accept" window for this ROS or if 2567 the NS(AR) was received over the secured spanning tree. If the 2568 NS(AR) is authentic and the OAL Identification is outside of the 2569 current "accept" window for this ROS, the ROR resets the current 2570 "accept" window start to the new OAL Identification value. If the 2571 NS(AR) is authentic but the target neighbor cache entry does not 2572 already include a Report List entry for this ROS, the ROR creates a 2573 new entry and caches the ROS information. The Report List cache 2574 entry therefore includes the LLA and ULA of the ROS, the new "accept" 2575 Identification number for the ROS and the previous "accept" 2576 Identification number in case any packets sent under the previous 2577 window are still in flight. 2579 The ROR then prepares a (solicited) NA(AR) message to send back to 2580 the ROS using the same Identification value received in the NS(AR) 2581 (unlike the NS(AR), the NA(AR) need not fit in a single OAL 2582 fragment). The ROR sets the NA(AR) source address to the target's 2583 MNP-LLA, sets the destination address to the NS(AR) LLA source 2584 address and sets the Target Address to the same value that appeared 2585 in the NS(AR). The ROR then includes an OMNI option with Preflen set 2586 to the prefix length associated with the NA(AR) source address. If 2587 the NS(AR) target was an MNP Client, the ROR next includes Interface 2588 Attributes in the OMNI option for each of the target's underlying 2589 interfaces with current information for each interface and includes 2590 an authentication signature if necessary. The ROR then sets the S/ 2591 T-omIndex field in the OMNI header to 0. 2593 For each Interface Attributes sub-option, the ROR sets the L2ADDR 2594 according to the Proxy/Server's INET address for VPNed or Direct 2595 interfaces, to the INET address of the Proxy/Server for proxyed 2596 interfaces or to the Client's INET address for INET interfaces. The 2597 ROR then includes the lower 32 bits of the Proxy/Server's ADM-ULA as 2598 the LHS, encodes the ADM-ULA prefix length code in the SRT field and 2599 sets the FMT code accordingly as specified in Section 3.3. 2601 The ROR then sets the NA(AR) message R flag to 1 (as a router) and S 2602 flag to 1 (as a response to a solicitation) and sets the O flag to 0 2603 (as a proxy). The ROR finally submits the NA(AR) for OAL 2604 encapsulation with source set to its own ULA and destination set to 2605 the ULA of the ROS, then performs OAL fragmentation using the same 2606 Identification value that appeared in the NS(AR) and forwards the 2607 resulting (*NET-encapsulated) carrier packets via the secured 2608 spanning tree without decrementing the network-layer TTL/Hop Limit 2609 field. 2611 While sending the NA(AR), the ROR also performs extended route 2612 optimization by sending an NS(NUD) to each of the target's underlying 2613 interfaces by changing the NS(AR) destination address to the address 2614 of the target and sending a distinct copy of the message according to 2615 each of the target's Link-Layer addresses (while setting S/T-omIndex 2616 to the index of the target underlying interface). The process for 2617 sending NS(NUD) messages is discussed in greater detail in 2618 Section 3.15. It is therefore often the case that the ROS may 2619 receive multiple NA responses to its single NS(AR) message - one from 2620 the ROR plus zero or more corresponding to each of the target's 2621 underlying interfaces. (Note: when sending the NS(NUD) on open INET 2622 interfaces, the ROR must include an OMNI authentication using its own 2623 security parameters and not those that it may have received from the 2624 ROS NS(AR) message. Therefore, the NS(NUD) must be prepared as a new 2625 IPv6 ND message from-scratch instead of simply copying the 2626 information received in the NS(AR).) 2628 3.14.4. Relaying the NA(AR) 2630 When the Bridge receives the NA(AR) carrier packets from the ROR, it 2631 discards the *NET header and determines the next hop by consulting 2632 its standard IPv6 forwarding table for the OAL header destination 2633 address. The Bridge then decrements the OAL header Hop-Limit, re- 2634 encapsulates the carrier packet and forwards it via a spanning tree 2635 secured tunnel the same as for any IPv6 router, where it may traverse 2636 multiple OMNI link segments. The final-hop Bridge will deliver the 2637 carrier packet via the secured spanning tree to a Proxy/Server for 2638 the ROS. 2640 3.14.5. Processing the NA(AR) 2642 When the ROS receives the NA(AR) message, it first searches for a 2643 neighbor cache entry that matches the NA(AR) LLA source address. If 2644 there is an entry in the INCOMPLETE or STALE state, the ROS matches 2645 the OAL Identification value with the value it had included in the 2646 corresponding NS(AR). If the values match, the ROS processes the 2647 message the same as for standard IPv6 Address Resolution [RFC4861]. 2648 In the process, it caches the NA(AR) LLA source address and all 2649 information found in the OMNI option in the neighbor cache entry for 2650 the target. The ROS finally sets the neighbor cache entry state to 2651 REACHABLE and sets its lifetime to ReachableTime seconds. (When the 2652 ROS is a Client, the solicited NA(AR) message will first be delivered 2653 via the secured spanning tree to one of its current Proxy/Servers, 2654 which then securely forwards the message to the Client. If the 2655 Client is on an ANET, ANET physical security and protected spectrum 2656 ensures security; if the Client is on the open ANET, the Proxy/Server 2657 must re-format the NA(AR) message to include its own authentication 2658 signature.) 2660 The ROS may also receive zero or more NA(NUD) messages appearing to 2661 be responses to NS(NUD) messages sent by itself, but these messages 2662 will have been triggered by the ROR's actions. Each NA(NUD) message 2663 includes information for only one of the target's underlying 2664 interfaces, and has the O flag set to 1 such that it will override 2665 any information previously received from the ROR's NA(AR). The 2666 Identification value in the message will match the value included in 2667 the original NS(AR) sent by the ROS. The ROS caches the NA(NUD) 2668 information, and marks each underlying interface as reachable. 2670 3.14.6. Route Optimization Maintenance 2672 Following route optimization, the ROS forwards future carrier packets 2673 with user data destined to the target via the addresses found in the 2674 cached link-layer information and with a monotonically-incrementing 2675 Identification value for each OAL packet. The route optimization is 2676 shared by all sources that send original IP packets to the target via 2677 the ROS, i.e., and not just the source on behalf of which the route 2678 optimization was initiated. Note that route optimization is 2679 performed only for original IP packets that contain user data, and 2680 not for those that contain other IPv6 ND control messages. 2682 While the ROS continues to forward additional original IP packets 2683 destined to the target, it sends additional NS(AR) messages to the 2684 ROR before ReachableTime expires to receive fresh NA messages the 2685 same as described in the previous sections (route optimization 2686 refreshment strategies are an implementation matter, with a non- 2687 normative example given in Appendix A.1). The ROS may supply a new 2688 unpredictable Identification value if it wishes to reset the 2689 neighbor's "accept" Identification window. If the ROS is an INET 2690 Client, it must include an authentication signature with the NS(AR) 2691 message so that the Proxy/Server can authenticate. 2693 The ROS uses its own ULA as the NS(AR) OAL source address and the ULA 2694 of the ROR as the NS(AR) OAL destination address, and sends up to 2695 MAX_MULTICAST_SOLICIT NS(AR) messages separated by 1 second until an 2696 NA(AR) is received. If no NA(AR) is received, the ROS assumes that 2697 the current ROR has become unreachable and deletes the target 2698 neighbor cache entry. Subsequent original IP packets will trigger a 2699 new route optimization event (see: Section 3.14.1) which may discover 2700 a different ROR that services the same target. 2702 If an NA(AR) is received, the ROS then updates the neighbor cache 2703 entry for the target to refresh ReachableTime, while (for MNP 2704 targets) the ROR adds or updates the ROS address to the target's 2705 Report List and with time set to ReportTime. While no carrier 2706 packets are flowing, the ROS instead allows ReachableTime for the 2707 target neighbor cache entry to expire. When ReachableTime expires, 2708 the ROS places the target neighbor cache entry back in the STALE 2709 state. Any future carrier packets flowing through the ROS will again 2710 trigger a new route optimization. 2712 The ROS may also receive unsolicited NA (uNA) messages from the ROR 2713 at any time (see: Section 3.16). If there is a neighbor cache entry 2714 for the target and the carrier packet(s) containing the uNA is 2715 received securely, the ROS updates the link-layer information but 2716 does not update ReachableTime since the receipt of a uNA does not 2717 confirm that any forward paths are working. If there is no neighbor 2718 cache entry or the message cannot be authenticated, the ROS simply 2719 discards the uNA. 2721 In this arrangement, the ROS holds a neighbor cache entry for the 2722 target in the REACHABLE state with a "send" Identification window 2723 starting value, while the ROR's target neighbor cache entry holds a 2724 Report List entry for the ROS with an "accept" Identification window 2725 starting value for the ULA of the ROS. The route optimization 2726 neighbor relationship between the ROS and ROR is therefore asymmetric 2727 and unidirectional. If the target node also has carrier packets to 2728 send back to the source node, then a separate route optimization 2729 procedure is performed in the reverse direction, but there is no 2730 requirement that the forward and reverse paths be symmetric. 2732 3.15. Neighbor Unreachability Detection (NUD) 2734 AERO nodes perform Neighbor Unreachability Detection (NUD) per 2735 [RFC4861] either reactively in response to persistent link-layer 2736 errors (see Section 3.11) or on-demand to confirm reachability and/or 2737 extend route optimizations from the ROS to the target. The NUD 2738 algorithm is based on periodic control message exchanges. The 2739 algorithm may further be seeded by ND hints of forward progress, but 2740 care must be taken to avoid inferring reachability based on spoofed 2741 information. For example, authentic IPv6 ND message exchanges may be 2742 considered as acceptable hints of forward progress, while spurious 2743 random packets should not be. 2745 AERO nodes can use standard NS/NA(NUD) exchanges sent over the OMNI 2746 link spanning tree to securely test reachability without risk of DoS 2747 attacks from nodes pretending to be a neighbor. These NS/NA(NUD) 2748 messages use the unicast LLAs and ULAs of the two parties involved in 2749 the NUD test the same as for standard IPv6 ND over the secured 2750 spanning tree, however a means for an ROS to test the unsecured 2751 target route optimized paths is also necessary. 2753 When an ROR directs an ROS to a target neighbor with one or more 2754 link-layer addresses, the ROS probes each unsecured target underlying 2755 interface either proactively or on-demand of carrier packets directed 2756 to the path by multilink forwarding to maintain the interface's state 2757 as reachable. Probing is performed through secured NS(NUD) messages 2758 over the spanning tree in the forward path that invoke an unsecured 2759 NA(NUD) reply over the target underlying interface on return path. 2760 (The NS(NUD) messages must therefore include Identification values 2761 (and optionally Nonce and Timestamp options) that will be echoed in 2762 the unsecured NA(NUD) replies.) While testing a target underlying 2763 interface, the ROS can optionally continue to send carrier packets 2764 via alternate interfaces and/or the ROR or maintain a small queue of 2765 carrier packets until target reachability is confirmed. 2767 When the ROS sends an NS(NUD) message, it sets the IPv6 source to its 2768 own LLA and sets both the destination and Target Address to the LLA 2769 of the target. The ROS also includes an OMNI option with S/T-omIndex 2770 set to the index of the target underlying interface and with a single 2771 Interface Attributes sub-option with the SRT, FMT, LHS and L2ADDR 2772 information for its own underlying interface. The ROS includes an 2773 Identification value within the current "send" window for this target 2774 neighbor (and optionally Nonce and Timestamp options), then 2775 encapsulates the message in OAL/INET headers with its own ULA as the 2776 source and the ULA formed from the SRT/LHS values for this target 2777 interface as the destination. The ROS then forwards the NS(NUD) 2778 message toward the destination ULA via the spanning tree. 2780 When the destination ULA node receives the NS(NUD) message, it 2781 examines the S/T-omIndex. If the corresponding underlying interface 2782 is a Direct or VPNed interface, the node acts as the target. If the 2783 underlying interface is a Proxyed interface, the node either acts as 2784 the target itself or forwards the message to the target Client. If 2785 the underlying interface is an INET interface, the node changes the 2786 OAL destination address to the ULA of the target Client and forwards 2787 the NS(NUD) over the underlying interface to the target Client while 2788 including an authentication signature. 2790 The target then creates a neighbor cache entry for the ROS LLA 2791 address if necessary and caches the Identification value as the start 2792 of the "accept" window for this ROS. The target then creates an 2793 NA(NUD) by reversing the OAL and IPv6 addresses from the NS(NUD) 2794 while copying the Identification value, and next including an 2795 Interface Attributes sub-option with attributes for its own interface 2796 identified by the NS(NUD) S/T-omIndex. The target sets the NA(NUD) 2797 S/T-omIndex to the index of the ROS, sets the Target Address to the 2798 same value that was in the NS(NUD), sets R flag to 1, the S flag to 0 2799 and the O flag to 1, and returns the message using its own underlying 2800 interface identified by NS(NUD) S/T-omIndex and destined to the ROS's 2801 interface identified by the original Interface Attributes sub-option. 2803 When the ROS receives the NA(NUD) message, it can determine from the 2804 Identification value and Target Address (and optionally the Nonce and 2805 Timestamp) that the message matched its NS(NUD) and that it transited 2806 the direct path using the selected underlying interface pair. The 2807 ROS marks route optimization target paths that pass these NUD tests 2808 as "reachable", and those that do not as "unreachable". These 2809 markings inform the OMNI interface forwarding algorithm specified in 2810 Section 3.10. Following the initial NA(NUD), the ROS sends 2811 additional NS(NUDs) as necessary, but may send them via the cached 2812 SRT, FMT, LHS and L2ADDR for the target via each underlying interface 2813 to be tested. 2815 Note: If the target determines that the OMNI option Interface 2816 Attributes in the NS(NUD) is located in a different OMNI link segment 2817 than its own interface named in the S/T-omIndex, it instead returns 2818 the NA(NUD) via the spanning tree while including an ORH and setting 2819 the OAL destination address to the Subnet Router Anycast address used 2820 by Bridges on the ROS segment. When a Bridge on the ROS segment 2821 receives the NA(NUD), it replaces the Interface Attributes with 2822 information for its own interface while using the omIndex value 2823 specific to the target. 2825 3.16. Mobility Management and Quality of Service (QoS) 2827 AERO is a Distributed Mobility Management (DMM) service. Each Proxy/ 2828 Server is responsible for only a subset of the Clients on the OMNI 2829 link, as opposed to a Centralized Mobility Management (CMM) service 2830 where there is a single network mobility collective entity for all 2831 Clients. Clients coordinate with their associated Proxy/Servers via 2832 RS/RA exchanges to maintain the DMM profile, and the AERO routing 2833 system tracks all current Client/Proxy/Server peering relationships. 2835 Proxy/Servers provide default routing and mobility/multilink services 2836 for their dependent Clients. Clients are responsible for maintaining 2837 neighbor relationships with their Proxy/Servers through periodic RS/ 2838 RA exchanges, which also serves to confirm neighbor reachability. 2839 When a Client's underlying interface address and/or QoS information 2840 changes, the Client is responsible for updating the Proxy/Server with 2841 this new information. Note that when there is a Proxy/Server in the 2842 path, the Proxy function can also perform some RS/RA exchanges on the 2843 Client's behalf. 2845 Mobility management messaging is based on the transmission and 2846 reception of unsolicited Neighbor Advertisement (uNA) messages. Each 2847 uNA message sets the IPv6 destination address to (link-local) All- 2848 Nodes multicast to convey a general update of Interface Attributes to 2849 (possibly) multiple recipients, or to a specific unicast LLA to 2850 announce a departure event to a specific recipient. Implementations 2851 must therefore examine the destination address to determine the 2852 nature of the mobility event (i.e., update vs departure). 2854 Mobility management considerations are specified in the following 2855 sections. 2857 3.16.1. Mobility Update Messaging 2859 RORs accommodate Client mobility, multilink and/or QoS change events 2860 by sending unsolicited NA (uNA) messages to each ROS in the target 2861 Client's Report List. When an ROR sends a uNA message, it sets the 2862 IPv6 source address to the its own LLA, sets the destination address 2863 to (link-local) All-Nodes multicast and sets the Target Address to 2864 the Client's MNP-LLA. The ROR also includes an OMNI option with 2865 Preflen set to the prefix length associated with the Client's MNP- 2866 LLA, with Interface Attributes for the target Client's underlying 2867 interfaces and with the OMNI header S/T-omIndex set to 0. The ROR 2868 then sets the NA R flag to 1, the S flag to 0 and the O flag to 1, 2869 then encapsulates the message in an OAL header with source set to its 2870 own ULA and destination set to the ULA of the ROS and sends the 2871 message into the spanning tree. 2873 As discussed in Section 7.2.6 of [RFC4861], the transmission and 2874 reception of uNA messages is unreliable but provides a useful 2875 optimization. In well-connected Internetworks with robust data links 2876 uNA messages will be delivered with high probability, but in any case 2877 the Proxy/Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT 2878 uNAs to each ROS to increase the likelihood that at least one will be 2879 received. 2881 When the ROS receives a uNA message prepared as above, it ignores the 2882 message if there is no existing neighbor cache entry for the target 2883 Client. Otherwise, it uses the included OMNI option information to 2884 update the neighbor cache entry, but does not reset ReachableTime 2885 since the receipt of an unsolicited NA message from the ROR does not 2886 provide confirmation that any forward paths to the target Client are 2887 working. 2889 If uNA messages are lost, the ROS may be left with stale address and/ 2890 or QoS information for the Client for up to ReachableTime seconds. 2891 During this time, the ROS can continue sending carrier packets 2892 according to its stale neighbor cache information. When 2893 ReachableTime is close to expiring, the ROS will re-initiate route 2894 optimization and receive fresh link-layer address information. 2896 In addition to sending uNA messages to the current set of ROSs for 2897 the Client, the ROR also sends uNAs to the MNP-ULA associated with 2898 the link-layer address for any underlying interface for which the 2899 link-layer address has changed. These uNA messages update an old 2900 Proxy/Server that cannot easily detect (e.g., without active probing) 2901 when a formerly-active Client has departed. When the ROR sends the 2902 uNA, it sets the IPv6 source address to its LLA, sets the destination 2903 address to the old Proxy/Server's ADM-LLA, and sets the Target 2904 Address to the Client's MNP-LLA. The ROR also includes an OMNI 2905 option with Preflen set to the prefix length associated with the 2906 Client's MNP-LLA, with Interface Attributes for the changed 2907 underlying interface, and with the OMNI header S/T-omIndex set to its 2908 own omIndex if the ROR is a Client or 0 if the ROR is a Proxy/Server. 2909 The ROR then sets the NA R flag to 1, the S flag to 0 and the O flag 2910 to 1, then encapsulates the message in an OAL header with source set 2911 to its own ULA and destination set to the ADM-ULA of the old Proxy/ 2912 Server and sends the message into the spanning tree. 2914 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 2916 When a Client needs to change its underlying interface addresses and/ 2917 or QoS preferences (e.g., due to a mobility event), the Client 2918 requests one of its Proxy/Servers to send RS messages to all of its 2919 other Proxy/Servers via the spanning tree with an OMNI option that 2920 includes Interface attributes with the new link quality and address 2921 information. 2923 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 2924 sending carrier packets containing user data in case one or more RAs 2925 are lost. If all RAs are lost, the Client SHOULD re-associate with a 2926 new Proxy/Server. 2928 When the Proxy/Server receives the Client's changes, it sends uNA 2929 messages to all nodes in the Report List the same as described in the 2930 previous section. 2932 3.16.3. Bringing New Links Into Service 2934 When a Client needs to bring new underlying interfaces into service 2935 (e.g., when it activates a new data link), it sends an RS message to 2936 the Proxy/Server via the underlying interface with an OMNI option 2937 that includes Interface Attributes with appropriate link quality 2938 values and with link-layer address information for the new link. 2940 3.16.4. Deactivating Existing Links 2942 When a Client needs to deactivate an existing underlying interface, 2943 it sends an RS or uNA message to its Proxy/Server with an OMNI option 2944 with appropriate Interface Attribute values - in particular, the link 2945 quality value 0 assures that neighbors will cease to use the link. 2947 If the Client needs to send RS/uNA messages over an underlying 2948 interface other than the one being deactivated, it MUST include 2949 Interface Attributes with appropriate link quality values for any 2950 underlying interfaces being deactivated. 2952 Note that when a Client deactivates an underlying interface, 2953 neighbors that have received the RS/uNA messages need not purge all 2954 references for the underlying interface from their neighbor cache 2955 entries. The Client may reactivate or reuse the underlying interface 2956 and/or its omIndex at a later point in time, when it will send RS/uNA 2957 messages with fresh Interface Attributes to update any neighbors. 2959 3.16.5. Moving Between Proxy/Servers 2961 The Client performs the procedures specified in Section 3.12.2 when 2962 it first associates with a new Proxy/Server or renews its association 2963 with an existing Proxy/Server. The Client also includes MS-Release 2964 identifiers in the RS message OMNI option per [I-D.templin-6man-omni] 2965 if it wants the new Proxy/Server to notify any old Proxy/Servers from 2966 which the Client is departing. 2968 When the new Proxy/Server receives the Client's RS message, it 2969 returns an RA as specified in Section 3.12.3 and sends up to 2970 MAX_NEIGHBOR_ADVERTIISEMENT uNA messages to any old Proxy/Servers 2971 listed in OMNI option MS-Release identifiers. When the new Proxy/ 2972 Server sends a uNA message, it sets the IPv6 source address to the 2973 Client's MNP-LLA, sets the destination address to the old Proxy/ 2974 Server's ADM-LLA, and sets the Target Address to the Client's LLA. 2975 The new Proxy/Server also includes an OMNI option with Preflen set to 2976 the prefix length associated with the Client's MNP-LLA, with 2977 Interface Attributes for its own underlying interface, and with the 2978 OMNI header S/T-omIndex set to 0. The new Proxy/Server then sets the 2979 NA R flag to 1, the S flag to 0 and the O flag to 1, then 2980 encapsulates the message in an OAL header with source set to its own 2981 ADM-ULA and destination set to the ADM-ULA of the old Proxy/Server 2982 and sends the message into the spanning tree. 2984 When an old Proxy/Server receives the uNA, it changes the Client's 2985 neighbor cache entry state to DEPARTED, sets the link-layer address 2986 of the Client to the new Proxy/Server's ADM-ULA, and resets 2987 DepartTime. After a short delay (e.g., 2 seconds) the old Proxy/ 2988 Server withdraws the Client's MNP from the routing system. After 2989 DepartTime expires, the old Proxy/Server deletes the Client's 2990 neighbor cache entry. 2992 The old Proxy/Server also iteratively forwards a copy of the uNA 2993 message to each ROS in the Client's Report List by changing the OAL 2994 destination address to the ULA of the ROS while leaving all other 2995 fields of the message unmodified. When the ROS receives the uNA, it 2996 examines the Target address to determine the correct neighbor cache 2997 entry and verifies that the IPv6 destination address matches the old 2998 Proxy/Server. The ROS then caches the IPv6 source address as the new 2999 Proxy/Server for the existing neighbor cache entry and marks the 3000 entry as STALE. While in the STALE state, the ROS allows new carrier 3001 packets to flow according to any existing cached link-layer 3002 information and sends new NS(AR) messages using its own ULA as the 3003 OAL source and the ADM-ULA of the new Proxy/Server as the OAL 3004 destination address to elicit NA messages that reset the neighbor 3005 cache entry state to REACHABLE. If no new NA message is received for 3006 10 seconds while in the STALE state, the ROS deletes the neighbor 3007 cache entry. 3009 Clients SHOULD NOT move rapidly between Proxy/Servers in order to 3010 avoid causing excessive oscillations in the AERO routing system. 3011 Examples of when a Client might wish to change to a different Proxy/ 3012 Server include a Proxy/Server that has gone unreachable, topological 3013 movements of significant distance, movement to a new geographic 3014 region, movement to a new OMNI link segment, etc. 3016 When a Client moves to a new Proxy/Server, some of the fragments of a 3017 multiple fragment OAL packet may have already arrived at the old 3018 Proxy/Server while others are en route to the new Proxy/Server, 3019 however no special attention in the reassembly algorithm is necessary 3020 since all fragments will eventually be delivered to the Client which 3021 can then reassemble. 3023 3.17. Multicast 3025 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 3026 [RFC3810] proxy service for its EUNs and/or hosted applications 3027 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 3028 underlying interfaces for which group membership is required. The 3029 IGMP/MLD messages may be further forwarded by a first-hop ANET access 3030 router acting as an IGMP/MLD-snooping switch [RFC4541], then 3031 ultimately delivered to an AERO Proxy/Server acting as a Protocol 3032 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 3033 Designated Router (DR) [RFC7761]. AERO Relays also act as PIM 3034 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 3035 INET/EUN networks. The behaviors identified in the following 3036 sections correspond to Source-Specific Multicast (SSM) and Any-Source 3037 Multicast (ASM) operational modes. 3039 3.17.1. Source-Specific Multicast (SSM) 3041 When an ROS "X" acting as PIM router receives a Join/Prune message 3042 from a node on its downstream interfaces containing one or more 3043 ((S)ource, (G)roup) pairs, it updates its Multicast Routing 3044 Information Base (MRIB) accordingly. For each S belonging to a 3045 prefix reachable via X's non-OMNI interfaces, X then forwards the (S, 3046 G) Join/Prune to any PIM routers on those interfaces per [RFC7761]. 3048 For each S belonging to a prefix reachable via X's OMNI interface, X 3049 includes a PIM Join/Prune for each (S,G) in the OMNI option of an 3050 NS(AR) message (see: Section 3.14) using its own LLA as the source 3051 address and ALL-PIM-ROUTERS as the destination address. X then 3052 encapsulates the NS(AR) in an OAL header with source address set to 3053 the ULA of X and destination address set to S then forwards the 3054 message into the spanning tree, which delivers it to ROR "Y" that 3055 services S. The resulting NA(AR) will return the LLA for the prefix 3056 that matches S as the network-layer source address and with an OMNI 3057 option with the ULA corresponding to any underlying interfaces that 3058 are currently servicing S. 3060 When Y processes the NS(AR) it examines the PIM Join/Prune message. 3061 If S is located behind any INET, Direct, or VPNed interfaces Y acts 3062 as a PIM router and updates its MRIB to list X as the next hop in the 3063 reverse path. If S is located behind any Proxys "Z"*, Y then sends 3064 an NS(NUD) message containing the PIM message to each Z* with 3065 addressing and encapsulation details the same as specified in 3066 Section 3.15. Each Z* then updates its MRIB accordingly and 3067 maintains the LLA of X as the next hop in the reverse path. Since 3068 the Bridges do not examine network layer control messages, this means 3069 that the (reverse) multicast tree path is simply from each Z* (and/or 3070 Y) to X with no other multicast-aware routers in the path. 3072 Following the initial combined Join/Prune and NS/NA messaging, X 3073 maintains a neighbor cache entry for each S the same as if X was 3074 sending unicast data traffic to S. In particular, X performs 3075 additional NS/NA exchanges to keep the neighbor cache entry alive for 3076 up to t_periodic seconds [RFC7761]. If no new Joins are received 3077 within t_periodic seconds, X allows the neighbor cache entry to 3078 expire. Finally, if X receives any additional Join/Prune messages 3079 for (S,G) it forwards the messages in NS/NA exchanges with each Y and 3080 Z* in the neighbor cache entry over the spanning tree. 3082 At some later time, Client C that holds an MNP for source S may 3083 depart from a first Proxy/Server Z1 and/or connect via a new Proxy/ 3084 Server Z2. In that case, Y sends a uNA message to X the same as 3085 specified for unicast mobility in Section 3.16. When X receives the 3086 uNA message, it updates its neighbor cache entry for the LLA for 3087 source S and sends new Join messages to any new Proxys Z2. There is 3088 no requirement to send any Prune messages to old Proxy/Server Z1 3089 since source S will no longer source any multicast data traffic via 3090 Z1. Instead, the multicast state for (S,G) in Proxy/Server Z1 will 3091 soon time out since no new Joins will arrive. 3093 After some later time, C may move to a new Proxy/Server Y2 and depart 3094 from old Sever Y1. In that case, Y1 sends Join messages for any of 3095 C's active (S,G) groups to Y2 while including its own LLA as the 3096 source address. This causes Y2 to include Y1 in the multicast 3097 forwarding tree during the interim time that Y1's neighbor cache 3098 entry for C is in the DEPARTED state. At the same time, Y1 sends a 3099 uNA message to X with an OMNI option with S/T-omIndex in the header 3100 set to 0 and a release indication to cause X to release its neighbor 3101 cache entry for S. X then sends a new Join message to S via the 3102 spanning tree and re-initiates route optimization the same as if it 3103 were receiving a fresh Join message from a node on a downstream link. 3105 3.17.2. Any-Source Multicast (ASM) 3107 When an ROS X acting as a PIM router receives a Join/Prune from a 3108 node on its downstream interfaces containing one or more (*,G) pairs, 3109 it updates its Multicast Routing Information Base (MRIB) accordingly. 3110 X then forwards a copy of the message within the OMNI option of an 3111 NS(AR) message to the Rendezvous Point (RP) R for each G over the 3112 spanning tree. X uses its own LLA as the source address and ALL-PIM- 3113 ROUTERS as the destination address, then encapsulates the NS(AR) 3114 message in an OAL header with source address set to the ULA of X and 3115 destination address set to R, then sends the message into the 3116 spanning tree. 3118 For each source S that sends multicast traffic to group G via R, the 3119 Proxy/Server Z* for the Client that aggregates S encapsulates the 3120 original IP packets in PIM Register messages and forwards them to R 3121 via the spanning tree, which may then elect to send a PIM Join to Z*. 3122 This will result in an (S,G) tree rooted at Z* with R as the next hop 3123 so that R will begin to receive two copies of the original IP packet; 3124 one native copy from the (S, G) tree and a second copy from the pre- 3125 existing (*, G) tree that still uses PIM Register encapsulation. R 3126 can then issue a PIM Register-stop message to suppress the Register- 3127 encapsulated stream. At some later time, if C moves to a new Proxy/ 3128 Server Z*, it resumes sending original IP packets via PIM Register 3129 encapsulation via the new Z*. 3131 At the same time, as multicast listeners discover individual S's for 3132 a given G, they can initiate an (S,G) Join for each S under the same 3133 procedures discussed in Section 3.17.1. Once the (S,G) tree is 3134 established, the listeners can send (S, G) Prune messages to R so 3135 that multicast original IP packets for group G sourced by S will only 3136 be delivered via the (S, G) tree and not from the (*, G) tree rooted 3137 at R. All mobility considerations discussed for SSM apply. 3139 3.17.3. Bi-Directional PIM (BIDIR-PIM) 3141 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 3142 approach to ASM that treats the Rendezvous Point (RP) as a Designated 3143 Forwarder (DF). Further considerations for BIDIR-PIM are out of 3144 scope. 3146 3.18. Operation over Multiple OMNI Links 3148 An AERO Client can connect to multiple OMNI links the same as for any 3149 data link service. In that case, the Client maintains a distinct 3150 OMNI interface for each link, e.g., 'omni0' for the first link, 3151 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 3152 would include its own distinct set of Bridges and Proxy/Servers, 3153 thereby providing redundancy in case of failures. 3155 Each OMNI link could utilize the same or different ANET connections. 3156 The links can be distinguished at the link-layer via the SRT prefix 3157 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 3158 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 3159 MSPs on each link. This gives rise to the opportunity for supporting 3160 multiple redundant networked paths, with each VLAN distinguished by a 3161 different SRT "color" (see: Section 3.2.5). 3163 The Client's IP layer can select the outgoing OMNI interface 3164 appropriate for a given traffic profile while (in the reverse 3165 direction) correspondent nodes must have some way of steering their 3166 original IP packets destined to a target via the correct OMNI link. 3168 In a first alternative, if each OMNI link services different MSPs, 3169 then the Client can receive a distinct MNP from each of the links. 3170 IP routing will therefore assure that the correct OMNI link is used 3171 for both outbound and inbound traffic. This can be accomplished 3172 using existing technologies and approaches, and without requiring any 3173 special supporting code in correspondent nodes or Bridges. 3175 In a second alternative, if each OMNI link services the same MSP(s) 3176 then each link could assign a distinct "OMNI link Anycast" address 3177 that is configured by all Bridges on the link. Correspondent nodes 3178 can then perform Segment Routing to select the correct SRT, which 3179 will then direct the original IP packet over multiple hops to the 3180 target. 3182 3.19. DNS Considerations 3184 AERO Client MNs and INET correspondent nodes consult the Domain Name 3185 System (DNS) the same as for any Internetworking node. When 3186 correspondent nodes and Client MNs use different IP protocol versions 3187 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 3188 A records for IPv4 address mappings to MNs which must then be 3189 populated in Relay NAT64 mapping caches. In that way, an IPv4 3190 correspondent node can send original IPv4 packets to the IPv4 address 3191 mapping of the target MN, and the Relay will translate the IPv4 3192 header and destination address into an IPv6 header and IPv6 3193 destination address of the MN. 3195 When an AERO Client registers with an AERO Proxy/Server, the Proxy/ 3196 Server can return the address(es) of DNS servers in RDNSS options 3197 [RFC6106]. The DNS server provides the IP addresses of other MNs and 3198 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 3200 3.20. Transition/Coexistence Considerations 3202 OAL encapsulation ensures that dissimilar INET partitions can be 3203 joined into a single unified OMNI link, even though the partitions 3204 themselves may have differing protocol versions and/or incompatible 3205 addressing plans. However, a commonality can be achieved by 3206 incrementally distributing globally routable (i.e., native) IP 3207 prefixes to eventually reach all nodes (both mobile and fixed) in all 3208 OMNI link segments. This can be accomplished by incrementally 3209 deploying AERO Bridges on each INET partition, with each Bridge 3210 distributing its MNPs and/or discovering non-MNP IP GUA prefixes on 3211 its INET links. 3213 This gives rise to the opportunity to eventually distribute native IP 3214 addresses to all nodes, and to present a unified OMNI link view even 3215 if the INET partitions remain in their current protocol and 3216 addressing plans. In that way, the OMNI link can serve the dual 3217 purpose of providing a mobility/multilink service and a transition/ 3218 coexistence service. Or, if an INET partition is transitioned to a 3219 native IP protocol version and addressing scheme that is compatible 3220 with the OMNI link MNP-based addressing scheme, the partition and 3221 OMNI link can be joined by Bridges. 3223 Relays that connect INETs/EUNs with dissimilar IP protocol versions 3224 may need to employ a network address and protocol translation 3225 function such as NAT64 [RFC6146]. 3227 3.21. Detecting and Reacting to Proxy/Server and Bridge Failures 3229 In environments where rapid failure recovery is required, Proxy/ 3230 Servers and Bridges SHOULD use Bidirectional Forwarding Detection 3231 (BFD) [RFC5880]. Nodes that use BFD can quickly detect and react to 3232 failures so that cached information is re-established through 3233 alternate nodes. BFD control messaging is carried only over well- 3234 connected ground domain networks (i.e., and not low-end radio links) 3235 and can therefore be tuned for rapid response. 3237 Proxy/Servers and Bridges maintain BFD sessions in parallel with 3238 their BGP peerings. If a Proxy/Server or Bridge fails, BGP peers 3239 will quickly re-establish routes through alternate paths the same as 3240 for common BGP deployments. Similarly, Proxys maintain BFD sessions 3241 with their associated Bridges even though they do not establish BGP 3242 peerings with them. 3244 3.22. AERO Clients on the Open Internet 3246 AERO Clients that connect to the open Internet via INET interfaces 3247 can establish a VPN or direct link to securely connect to a Proxy/ 3248 Server in a "tethered" arrangement with all of the Client's traffic 3249 transiting the Proxy/Server. Alternatively, the Client can associate 3250 with an INET Proxy/Server using UDP/IP encapsulation and control 3251 message securing services as discussed in the following sections. 3253 When a Client's OMNI interface enables an INET underlying interface, 3254 it first determines whether the interface is likely to be behind a 3255 NAT. For IPv4, the Client assumes it is on the open Internet if the 3256 INET address is not a special-use IPv4 address per [RFC3330]. 3257 Similarly for IPv6, the Client assumes it is on the open Internet if 3258 the INET address is a Global Unicast Address (GUA) [RFC4291]. 3259 Otherwise, the Client assumes it may be behind one or several NATs. 3261 The Client then prepares an RS message with IPv6 source address set 3262 to its MNP-LLA, with IPv6 destination set to (link-local) All-Routers 3263 multicast and with an OMNI option with underlying interface 3264 attributes. If the Client believes that it is on the open Internet, 3265 it SHOULD include an L2ADDR in the Interface Attributes sub-option 3266 corresponding to the underlying interface; otherwise, it MAY omit 3267 L2ADDR. If the underlying address is IPv4, the Client includes the 3268 Port Number and IPv4 address written in obfuscated form [RFC4380] as 3269 discussed in Section 3.3. If the underlying interface address is 3270 IPv6, the Client instead includes the Port Number and IPv6 address in 3271 obfuscated form. The Client finally includes an authentication 3272 signature sub-option in the OMNI option [I-D.templin-6man-omni] to 3273 provide message authentication and submits the RS for OAL 3274 encapsulation as an atomic fragment using an unpredictable 3275 Identification value to establish the start of the "send/accept" 3276 window for this Proxy/Server. The Client then encapsulates the OAL 3277 fragment in UDP/IP headers to form a carrier packet, sets the UDP/IP 3278 source to its INET address and UDP port, sets the UDP/IP destination 3279 to the Proxy/Server's INET address and the AERO service port number 3280 (8060), then sends the carrier packet to the Proxy/Server. 3282 When the Proxy/Server receives the RS, it discards the OAL 3283 encapsulation, authenticates the RS message, creates a neighbor cache 3284 entry and registers the Client's MNP, Identification and INET 3285 interface information according to the OMNI option parameters. If 3286 the RS message OMNI option includes Interface Attributes with an 3287 L2ADDR, the Proxy/Server compares the encapsulation IP address and 3288 UDP port number with the (unobfuscated) values. If the values are 3289 the same, the Proxy/Server caches the Client's information as "INET" 3290 addresses meaning that the Client is likely to accept direct messages 3291 without requiring NAT traversal exchanges. If the values are 3292 different (or, if the OMNI option did not include an L2ADDR) the 3293 Proxy/Server instead caches the Client's information as "mapped" 3294 addresses meaning that NAT traversal exchanges may be necessary. 3296 The Proxy/Server then prepares an RA message with IPv6 source and 3297 destination set corresponding to the addresses in the RS, and with an 3298 OMNI option with an Origin Indication sub-option per 3299 [I-D.templin-6man-omni] with the mapped and obfuscated Port Number 3300 and IP address observed in the encapsulation headers. The Proxy/ 3301 Server also includes an authentication signature sub-option per 3302 [I-D.templin-6man-omni] that contains an acknowledgement of the 3303 update sent by the Client. The Proxy/Server then performs OAL 3304 encapsulation and fragmentation if necessary using the same 3305 Identification value that appeared in the RS, and encapsulates each 3306 fragment in UDP/IP headers with addresses set per the L2ADDR 3307 information in the neighbor cache entry for the Client. 3309 When the Client receives the RA message, it verifies the OAL 3310 Identification value, performs OAL reassembly if necessary, 3311 authenticates the message, then compares the mapped Port Number and 3312 IP address from the Origin Indication sub-option with its own 3313 address. If the addresses are the same, the Client assumes the open 3314 Internet / Cone NAT principle; if the addresses are different, the 3315 Client instead assumes that further qualification procedures are 3316 necessary to detect the type of NAT and proceeds according to 3317 standard procedures [RFC6081][RFC4380]. 3319 After the Client has registered its INET interfaces in such RS/RA 3320 exchanges it sends periodic RS messages to receive fresh RA messages 3321 before the Router Lifetime received on each INET interface expires. 3322 The Client also maintains default routes via its Proxy/Servers, i.e., 3323 the same as described in earlier sections. 3325 When the Client sends messages to target IP addresses, it also 3326 invokes route optimization per Section 3.14 using IPv6 ND address 3327 resolution messaging. The Client first creates a neighbor cache 3328 entry for the target in the INCOMPLETE state, then sends the NS(AR) 3329 message to the Proxy/Server with an OMNI option with an 3330 authentication signature sub-option. The Client sets the NS source 3331 address to its own MNP-LLA, destination address to the target 3332 solicited node multicast address and target address to the LLA of the 3333 target. The Client then wraps the NS message in OAL headers (i.e., 3334 as an atomic fragment) with an unpredictable Identification value to 3335 establish the "send" window for this target, with source address set 3336 to its own MNP-ULA and destination address set to the target's MNP- 3337 ULA. The Client then wraps the atomic fragment in a UDP/IP header 3338 and sends the resulting carrier packet to the Proxy/Server. 3340 When the Client's Proxy/Server receives the OAL-encapsulated NS, it 3341 authenticates the message by processing the authentication signature 3342 sub-option and forwards the message over the spanning tree on behalf 3343 of the Client. When the ROR receives the NS(AR), it adds an entry 3344 for the ROS to the target neighbor cache entry Report List and caches 3345 the Identification value as the start of the "accept" window for 3346 packets originating from this ROS (if the ROR is a Proxy/Server, it 3347 also creates a Report List entry for this ROS in the target Client's 3348 neighbor cache entry). The ROR then returns an NA(AR) with OMNI 3349 option information for the target including all of the target's 3350 Interface Attributes. 3352 The ROR sets the NA(AR) source address to its own LLA, sets the 3353 destination address to the ROS LLA and sets the target address to the 3354 LLA of the target. The ROR then performs OAL encapsulation using the 3355 same Identification value that appeared in the NS(AR), then sets the 3356 OAL source address to the ROR's ULA and destination address to ULA 3357 source of the NS(AR). If the ROR is an INET Client, it includes an 3358 authentication signature and sends the NA(AR) to its Proxy/Sever 3359 which verifies the authentication signature and forwards the NA(AR) 3360 into the secured spanning tree. If the ROR is an ANET Client or a 3361 Proxy/Server, it simply forwards the NA(AR) into the secured spanning 3362 tree. 3364 When the Proxy/Sever for the ROS Client receives the NA(AR) message 3365 contained in one or more carrier packets, it verifies the OAL 3366 Identification matches the same value that was used in the NS(AR) 3367 then reassembles if necessary. When reassembly is complete, the 3368 Proxy/Server includes an authentication signature and forwards the 3369 NA(AR) to the ROS Client. The ROS Client then verifies the 3370 authentication signature and changes the neighbor cache entry state 3371 for this target to REACHABLE. 3373 Following route optimization for targets in the same OMNI link 3374 segment, if the target's L2ADDR is on the open INET, the Client 3375 forwards carrier packets directly to the target INET address. If the 3376 target is behind a NAT, the Client first establishes NAT state for 3377 the L2ADDR using the "direct bubble" and NUD mechanisms discussed in 3378 Section 3.10.1. The Client continues to send carrier packets via its 3379 Proxy/Server until NAT state is populated, then begins forwarding 3380 carrier packets via the direct path through the NAT to the target. 3381 For targets in different OMNI link segments, the Client uses OAL/ORH 3382 encapsulation and forwards carrier packets to the Bridge that 3383 returned the NA message. 3385 The ROR may return uNAs via the ROS Proxy/Server if the target moves, 3386 and the Proxy/Server will send corresponding uNAs to the Client with 3387 an OMNI authentication sub-option. The Client can also send NUD 3388 messages to test forward path reachability even though there is no 3389 security association between the Client and the target. 3391 The Client can send original IP packets to route-optimized neighbors 3392 in the same OMNI link segment no larger than the minimum/path MPS in 3393 one piece and with OAL encapsulation but without fragmentation. For 3394 larger original IP packets, the Client applies OAL encapsulation and 3395 fragmentation if necessary according to Section 3.9, with OAL header 3396 with source set to its own MNP-ULA and destination set to the MNP-ULA 3397 of the target. The Client then encapsulates each original IP packet 3398 or OAL fragment in UDP/IP *NET headers and sends them to the next 3399 hop. 3401 Note: The NAT traversal procedures specified in this document are 3402 applicable for Cone, Address-Restricted and Port-Restricted NATs 3403 only. While future updates to this document may specify procedures 3404 for other NAT variations (e.g., hairpinning and various forms of 3405 Symmetric NATs), it should be noted that continuous communications 3406 are always possible through forwarding via a Proxy/Server even if NAT 3407 traversal is not employed. 3409 3.23. Time-Varying MNPs 3411 In some use cases, it is desirable, beneficial and efficient for the 3412 Client to receive a constant MNP that travels with the Client 3413 wherever it moves. For example, this would allow air traffic 3414 controllers to easily track aircraft, etc. In other cases, however 3415 (e.g., intelligent transportation systems), the MN may be willing to 3416 sacrifice a modicum of efficiency in order to have time-varying MNPs 3417 that can be changed every so often to defeat adversarial tracking. 3419 The DHCPv6 service offers a way for Clients that desire time-varying 3420 MNPs to obtain short-lived prefixes (e.g., on the order of a small 3421 number of minutes). In that case, the identity of the Client would 3422 not be bound to the MNP but rather to a Node Identification value 3423 (see: [I-D.templin-6man-omni]) to be used as the Client ID seed for 3424 MNP prefix delegation. The Client would then be obligated to 3425 renumber its internal networks whenever its MNP (and therefore also 3426 its MNP-LLA) changes. This should not present a challenge for 3427 Clients with automated network renumbering services, however presents 3428 limits for the durations of ongoing sessions that would prefer to use 3429 a constant address. 3431 4. Implementation Status 3433 An early AERO implementation based on OpenVPN (https://openvpn.net/) 3434 was announced on the v6ops mailing list on January 10, 2018 and an 3435 initial public release of the AERO proof-of-concept source code was 3436 announced on the intarea mailing list on August 21, 2015. 3438 AERO Release-3.2 was tagged on March 30, 2021, and is undergoing 3439 internal testing. Additional internal releases expected within the 3440 coming months, with first public release expected end of 1H2021. 3442 5. IANA Considerations 3444 The IANA is instructed to assign a new type value TBD1 in the IPv6 3445 Routing Types registry (IANA registration procedure is IETF Review or 3446 IESG Approval). 3448 The IANA has assigned the UDP port number "8060" for an earlier 3449 experimental first version of AERO [RFC6706]. This document 3450 obsoletes [RFC6706], and together with [I-D.templin-6man-omni] 3451 reclaims the UDP port number "8060" for 'aero' as the service port 3452 for UDP/IP encapsulation. (Note that, although [RFC6706] was not 3453 widely implemented or deployed, any messages coded to that 3454 specification can be easily distinguished and ignored since they use 3455 the invalid ICMPv6 message type number '0'.) This document makes no 3456 request of IANA, since [I-D.templin-6man-omni] already provides 3457 instructions. 3459 No further IANA actions are required. 3461 6. Security Considerations 3463 AERO Bridges configure secured tunnels with AERO Proxy/Servers and 3464 Relays within their local OMNI link segments. Applicable secured 3465 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3466 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3467 segments in turn configure secured tunnels for their neighboring AERO 3468 Bridges in a spanning tree topology. Therefore, control messages 3469 exchanged between any pair of OMNI link neighbors on the spanning 3470 tree are already secured. 3472 AERO nodes acting as Route Optimization Responders (RORs) may also 3473 receive packets directly from arbitrary nodes in INET partitions 3474 instead of via the secured spanning tree. For INET partitions that 3475 apply effective ingress filtering to defeat source address spoofing, 3476 the simple data origin authentication procedures in Section 3.8 can 3477 be applied. 3479 For INET partitions that require strong security in the data plane, 3480 two options for securing communications include 1) disable route 3481 optimization so that all traffic is conveyed over secured tunnels, or 3482 2) enable on-demand secure tunnel creation between Client neighbors. 3484 Option 1) would result in longer routes than necessary and impose 3485 traffic concentration on critical infrastructure elements. Option 2) 3486 could be coordinated between Clients using NS/NA messages with OMNI 3487 Host Identity Protocol (HIP) "Initiator/Responder" message sub- 3488 options [RFC7401][I-D.templin-6man-omni] to create a secured tunnel 3489 on-demand. 3491 AERO Clients that connect to secured ANETs need not apply security to 3492 their ND messages, since the messages will be intercepted by a 3493 perimeter Proxy/Server that applies security on its INET-facing 3494 interface as part of the spanning tree (see above). AERO Clients 3495 connected to the open INET can use network and/or transport layer 3496 security services such as VPNs or can by some other means establish a 3497 direct link to a Proxy/Server. When a VPN or direct link may be 3498 impractical, however, INET Clients and Proxy/Servers SHOULD include 3499 and verify authentication signatures for their IPv6 ND messages as 3500 specified in [I-D.templin-6man-omni]. 3502 Application endpoints SHOULD use transport-layer (or higher-layer) 3503 security services such as TLS/SSL, DTLS or SSH [RFC4251] to assure 3504 the same level of protection as for critical secured Internet 3505 services. AERO Clients that require host-based VPN services SHOULD 3506 use network and/or transport layer security services such as IPsec, 3507 TLS/SSL, DTLS, etc. AERO Proxys and Proxy/Servers can also provide a 3508 network-based VPN service on behalf of the Client, e.g., if the 3509 Client is located within a secured enclave and cannot establish a VPN 3510 on its own behalf. 3512 AERO Proxy/Servers and Bridges present targets for traffic 3513 amplification Denial of Service (DoS) attacks. This concern is no 3514 different than for widely-deployed VPN security gateways in the 3515 Internet, where attackers could send spoofed packets to the gateways 3516 at high data rates. This can be mitigated through the AERO/OMNI data 3517 origin authentication procedures, as well as connecting Proxy/Servers 3518 and Bridges over dedicated links with no connections to the Internet 3519 and/or when connections to the Internet are only permitted through 3520 well-managed firewalls. Traffic amplification DoS attacks can also 3521 target an AERO Client's low data rate links. This is a concern not 3522 only for Clients located on the open Internet but also for Clients in 3523 secured enclaves. AERO Proxy/Servers and Proxys can institute rate 3524 limits that protect Clients from receiving packet floods that could 3525 DoS low data rate links. 3527 AERO Relays must implement ingress filtering to avoid a spoofing 3528 attack in which spurious messages with ULA addresses are injected 3529 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3530 that their connectivity is not used by unauthorized nodes on their 3531 EUNs to gain access to a protected network, i.e., AERO Clients that 3532 act as routers MUST NOT provide routing services for unauthorized 3533 nodes. (This concern is no different than for ordinary hosts that 3534 receive an IP address delegation but then "share" the address with 3535 other nodes via some form of Internet connection sharing such as 3536 tethering.) 3538 The MAP list MUST be well-managed and secured from unauthorized 3539 tampering, even though the list contains only public information. 3540 The MAP list can be conveyed to the Client in a similar fashion as in 3541 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3542 upload of a static file, DNS lookups, etc.). 3544 The AERO service for open INET Clients depends on a public key 3545 distribution service in which Client public keys and identities are 3546 maintained in a shared database accessible to all open INET Proxy/ 3547 Servers. Similarly, each Client must be able to determine the public 3548 key of each Proxy/Server, e.g. by consulting an online database. 3549 When AERO nodes register their public keys indexed by a unique Host 3550 Identity Tag (HIT) [RFC7401] in a distributed database such as the 3551 DNS, and use the HIT as an identity for applying IPv6 ND message 3552 authentication signatures, a means for determining public key 3553 attestation is available. 3555 SRH authentication facilities are specified in [RFC8754]. 3557 Security considerations for accepting link-layer ICMP messages and 3558 reflected packets are discussed throughout the document. 3560 Security considerations for IPv6 fragmentation and reassembly are 3561 discussed in [I-D.templin-6man-omni]. 3563 7. Acknowledgements 3565 Discussions in the IETF, aviation standards communities and private 3566 exchanges helped shape some of the concepts in this work. 3567 Individuals who contributed insights include Mikael Abrahamsson, Mark 3568 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3569 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3570 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3571 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3572 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3573 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3574 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3575 Wood and James Woodyatt. Members of the IESG also provided valuable 3576 input during their review process that greatly improved the document. 3577 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3578 for their shepherding guidance during the publication of the AERO 3579 first edition. 3581 This work has further been encouraged and supported by Boeing 3582 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3583 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3584 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3585 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3586 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3587 Gene MacLean III, Kyle Mikos, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3588 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3589 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3590 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3591 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3592 Benson, Madhuri Madhava Badgandi, Vijayasarathy Rajagopalan, Katie 3593 Tran and Eric Yeh are especially acknowledged for implementing the 3594 AERO functions as extensions to the public domain OpenVPN 3595 distribution. Chuck Klabunde is honored and remembered for his early 3596 leadership, and we mourn his untimely loss. 3598 Earlier works on NBMA tunneling approaches are found in 3599 [RFC2529][RFC5214][RFC5569]. 3601 Many of the constructs presented in this second edition of AERO are 3602 based on the author's earlier works, including: 3604 o The Internet Routing Overlay Network (IRON) 3605 [RFC6179][I-D.templin-ironbis] 3607 o Virtual Enterprise Traversal (VET) 3608 [RFC5558][I-D.templin-intarea-vet] 3610 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3611 [RFC5320][I-D.templin-intarea-seal] 3613 o AERO, First Edition [RFC6706] 3615 Note that these works cite numerous earlier efforts that are not also 3616 cited here due to space limitations. The authors of those earlier 3617 works are acknowledged for their insights. 3619 This work is aligned with the NASA Safe Autonomous Systems Operation 3620 (SASO) program under NASA contract number NNA16BD84C. 3622 This work is aligned with the FAA as per the SE2025 contract number 3623 DTFAWA-15-D-00030. 3625 This work is aligned with the Boeing Commercial Airplanes (BCA) 3626 Internet of Things (IoT) and autonomy programs. 3628 This work is aligned with the Boeing Information Technology (BIT) 3629 MobileNet program. 3631 8. References 3633 8.1. Normative References 3635 [I-D.templin-6man-omni] 3636 Templin, F. L. and T. Whyman, "Transmission of IP Packets 3637 over Overlay Multilink Network (OMNI) Interfaces", draft- 3638 templin-6man-omni-03 (work in progress), April 2021. 3640 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3641 DOI 10.17487/RFC0791, September 1981, 3642 . 3644 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3645 RFC 792, DOI 10.17487/RFC0792, September 1981, 3646 . 3648 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3649 Requirement Levels", BCP 14, RFC 2119, 3650 DOI 10.17487/RFC2119, March 1997, 3651 . 3653 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3654 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3655 December 1998, . 3657 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3658 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3659 DOI 10.17487/RFC3971, March 2005, 3660 . 3662 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3663 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3664 . 3666 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3667 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3668 November 2005, . 3670 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3671 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3672 . 3674 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3675 Network Address Translations (NATs)", RFC 4380, 3676 DOI 10.17487/RFC4380, February 2006, 3677 . 3679 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3680 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3681 DOI 10.17487/RFC4861, September 2007, 3682 . 3684 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3685 Address Autoconfiguration", RFC 4862, 3686 DOI 10.17487/RFC4862, September 2007, 3687 . 3689 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3690 DOI 10.17487/RFC6081, January 2011, 3691 . 3693 [RFC7401] Moskowitz, R., Ed., Heer, T., Jokela, P., and T. 3694 Henderson, "Host Identity Protocol Version 2 (HIPv2)", 3695 RFC 7401, DOI 10.17487/RFC7401, April 2015, 3696 . 3698 [RFC7739] Gont, F., "Security Implications of Predictable Fragment 3699 Identification Values", RFC 7739, DOI 10.17487/RFC7739, 3700 February 2016, . 3702 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3703 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3704 May 2017, . 3706 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3707 (IPv6) Specification", STD 86, RFC 8200, 3708 DOI 10.17487/RFC8200, July 2017, 3709 . 3711 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3712 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3713 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3714 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3715 . 3717 8.2. Informative References 3719 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3720 2016. 3722 [I-D.bonica-6man-comp-rtg-hdr] 3723 Bonica, R., Kamite, Y., Alston, A., Henriques, D., and L. 3724 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 3725 bonica-6man-comp-rtg-hdr-24 (work in progress), January 3726 2021. 3728 [I-D.bonica-6man-crh-helper-opt] 3729 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 3730 Routing Header (CRH) Helper Option", draft-bonica-6man- 3731 crh-helper-opt-03 (work in progress), April 2021. 3733 [I-D.ietf-intarea-frag-fragile] 3734 Bonica, R., Baker, F., Huston, G., Hinden, R. M., Troan, 3735 O., and F. Gont, "IP Fragmentation Considered Fragile", 3736 draft-ietf-intarea-frag-fragile-17 (work in progress), 3737 September 2019. 3739 [I-D.ietf-intarea-tunnels] 3740 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3741 Architecture", draft-ietf-intarea-tunnels-10 (work in 3742 progress), September 2019. 3744 [I-D.ietf-ipwave-vehicular-networking] 3745 (editor), J. (. J., "IPv6 Wireless Access in Vehicular 3746 Environments (IPWAVE): Problem Statement and Use Cases", 3747 draft-ietf-ipwave-vehicular-networking-20 (work in 3748 progress), March 2021. 3750 [I-D.ietf-rtgwg-atn-bgp] 3751 Templin, F. L., Saccone, G., Dawra, G., Lindem, A., and V. 3752 Moreno, "A Simple BGP-based Mobile Routing System for the 3753 Aeronautical Telecommunications Network", draft-ietf- 3754 rtgwg-atn-bgp-10 (work in progress), January 2021. 3756 [I-D.templin-6man-dhcpv6-ndopt] 3757 Templin, F. L., "A Unified Stateful/Stateless 3758 Configuration Service for IPv6", draft-templin-6man- 3759 dhcpv6-ndopt-11 (work in progress), January 2021. 3761 [I-D.templin-intarea-seal] 3762 Templin, F. L., "The Subnetwork Encapsulation and 3763 Adaptation Layer (SEAL)", draft-templin-intarea-seal-68 3764 (work in progress), January 2014. 3766 [I-D.templin-intarea-vet] 3767 Templin, F. L., "Virtual Enterprise Traversal (VET)", 3768 draft-templin-intarea-vet-40 (work in progress), May 2013. 3770 [I-D.templin-ipwave-uam-its] 3771 Templin, F. L., "Urban Air Mobility Implications for 3772 Intelligent Transportation Systems", draft-templin-ipwave- 3773 uam-its-04 (work in progress), January 2021. 3775 [I-D.templin-ironbis] 3776 Templin, F. L., "The Interior Routing Overlay Network 3777 (IRON)", draft-templin-ironbis-16 (work in progress), 3778 March 2014. 3780 [I-D.templin-v6ops-pdhost] 3781 Templin, F. L., "IPv6 Prefix Delegation and Multi- 3782 Addressing Models", draft-templin-v6ops-pdhost-27 (work in 3783 progress), January 2021. 3785 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 3787 [RFC1035] Mockapetris, P., "Domain names - implementation and 3788 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 3789 November 1987, . 3791 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 3792 RFC 1812, DOI 10.17487/RFC1812, June 1995, 3793 . 3795 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 3796 DOI 10.17487/RFC2003, October 1996, 3797 . 3799 [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, 3800 DOI 10.17487/RFC2004, October 1996, 3801 . 3803 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 3804 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 3805 . 3807 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 3808 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 3809 . 3811 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 3812 Domains without Explicit Tunnels", RFC 2529, 3813 DOI 10.17487/RFC2529, March 1999, 3814 . 3816 [RFC2983] Black, D., "Differentiated Services and Tunnels", 3817 RFC 2983, DOI 10.17487/RFC2983, October 2000, 3818 . 3820 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 3821 of Explicit Congestion Notification (ECN) to IP", 3822 RFC 3168, DOI 10.17487/RFC3168, September 2001, 3823 . 3825 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 3826 DOI 10.17487/RFC3330, September 2002, 3827 . 3829 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 3830 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 3831 DOI 10.17487/RFC3810, June 2004, 3832 . 3834 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 3835 Unique IDentifier (UUID) URN Namespace", RFC 4122, 3836 DOI 10.17487/RFC4122, July 2005, 3837 . 3839 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 3840 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 3841 January 2006, . 3843 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 3844 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 3845 DOI 10.17487/RFC4271, January 2006, 3846 . 3848 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 3849 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 3850 2006, . 3852 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 3853 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 3854 December 2005, . 3856 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 3857 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 3858 2006, . 3860 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 3861 Control Message Protocol (ICMPv6) for the Internet 3862 Protocol Version 6 (IPv6) Specification", STD 89, 3863 RFC 4443, DOI 10.17487/RFC4443, March 2006, 3864 . 3866 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 3867 Protocol (LDAP): The Protocol", RFC 4511, 3868 DOI 10.17487/RFC4511, June 2006, 3869 . 3871 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 3872 "Considerations for Internet Group Management Protocol 3873 (IGMP) and Multicast Listener Discovery (MLD) Snooping 3874 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 3875 . 3877 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 3878 "Internet Group Management Protocol (IGMP) / Multicast 3879 Listener Discovery (MLD)-Based Multicast Forwarding 3880 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 3881 August 2006, . 3883 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 3884 Algorithms in Cryptographically Generated Addresses 3885 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 3886 . 3888 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 3889 "Bidirectional Protocol Independent Multicast (BIDIR- 3890 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 3891 . 3893 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 3894 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 3895 DOI 10.17487/RFC5214, March 2008, 3896 . 3898 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 3899 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 3900 February 2010, . 3902 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 3903 Route Optimization Requirements for Operational Use in 3904 Aeronautics and Space Exploration Mobile Networks", 3905 RFC 5522, DOI 10.17487/RFC5522, October 2009, 3906 . 3908 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 3909 RFC 5558, DOI 10.17487/RFC5558, February 2010, 3910 . 3912 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 3913 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 3914 January 2010, . 3916 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 3917 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 3918 . 3920 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 3921 "IPv6 Router Advertisement Options for DNS Configuration", 3922 RFC 6106, DOI 10.17487/RFC6106, November 2010, 3923 . 3925 [RFC6139] Russert, S., Ed., Fleischman, E., Ed., and F. Templin, 3926 Ed., "Routing and Addressing in Networks with Global 3927 Enterprise Recursion (RANGER) Scenarios", RFC 6139, 3928 DOI 10.17487/RFC6139, February 2011, 3929 . 3931 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 3932 NAT64: Network Address and Protocol Translation from IPv6 3933 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 3934 April 2011, . 3936 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 3937 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 3938 . 3940 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 3941 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 3942 DOI 10.17487/RFC6221, May 2011, 3943 . 3945 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 3946 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 3947 DOI 10.17487/RFC6273, June 2011, 3948 . 3950 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 3951 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 3952 January 2012, . 3954 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 3955 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 3956 DOI 10.17487/RFC6355, August 2011, 3957 . 3959 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 3960 for Equal Cost Multipath Routing and Link Aggregation in 3961 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 3962 . 3964 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 3965 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 3966 . 3968 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 3969 UDP Checksums for Tunneled Packets", RFC 6935, 3970 DOI 10.17487/RFC6935, April 2013, 3971 . 3973 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 3974 for the Use of IPv6 UDP Datagrams with Zero Checksums", 3975 RFC 6936, DOI 10.17487/RFC6936, April 2013, 3976 . 3978 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 3979 Korhonen, "Requirements for Distributed Mobility 3980 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 3981 . 3983 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 3984 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 3985 Multicast - Sparse Mode (PIM-SM): Protocol Specification 3986 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 3987 2016, . 3989 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 3990 Decraene, B., Litkowski, S., and R. Shakir, "Segment 3991 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 3992 July 2018, . 3994 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 3995 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 3996 . 3998 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 3999 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 4000 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 4001 . 4003 [WG] Wireguard, "WireGuard, https://www.wireguard.com", August 4004 2020. 4006 Appendix A. Non-Normative Considerations 4008 AERO can be applied to a multitude of Internetworking scenarios, with 4009 each having its own adaptations. The following considerations are 4010 provided as non-normative guidance: 4012 A.1. Implementation Strategies for Route Optimization 4014 Route optimization as discussed in Section 3.14 results in the route 4015 optimization source (ROS) creating a neighbor cache entry for the 4016 target neighbor. The neighbor cache entry state is set to REACHABLE 4017 for at most ReachableTime seconds. In order to refresh the neighbor 4018 cache entry lifetime before the ReachableTime timer expires, the 4019 specification requires implementations to issue a new NS/NA exchange 4020 to reset ReachableTime while data packets are still flowing. 4021 However, the decision of when to initiate a new NS/NA exchange and to 4022 perpetuate the process is left as an implementation detail. 4024 One possible strategy may be to monitor the neighbor cache entry 4025 watching for data packets for (ReachableTime - 5) seconds. If any 4026 data packets have been sent to the neighbor within this timeframe, 4027 then send an NS to receive a new NA. If no data packets have been 4028 sent, wait for 5 additional seconds and send an immediate NS if any 4029 data packets are sent within this "expiration pending" 5 second 4030 window. If no additional data packets are sent within the 5 second 4031 window, reset the neighbor cache entry state to STALE. 4033 The monitoring of the neighbor data packet traffic therefore becomes 4034 an ongoing process during the neighbor cache entry lifetime. If the 4035 neighbor cache entry expires, future data packets will trigger a new 4036 NS/NA exchange while the packets themselves are delivered over a 4037 longer path until route optimization state is re-established. 4039 A.2. Implicit Mobility Management 4041 OMNI interface neighbors MAY provide a configuration option that 4042 allows them to perform implicit mobility management in which no ND 4043 messaging is used. In that case, the Client only transmits packets 4044 over a single interface at a time, and the neighbor always observes 4045 packets arriving from the Client from the same link-layer source 4046 address. 4048 If the Client's underlying interface address changes (either due to a 4049 readdressing of the original interface or switching to a new 4050 interface) the neighbor immediately updates the neighbor cache entry 4051 for the Client and begins accepting and sending packets according to 4052 the Client's new address. This implicit mobility method applies to 4053 use cases such as cellphones with both WiFi and Cellular interfaces 4054 where only one of the interfaces is active at a given time, and the 4055 Client automatically switches over to the backup interface if the 4056 primary interface fails. 4058 A.3. Direct Underlying Interfaces 4060 When a Client's OMNI interface is configured over a Direct interface, 4061 the neighbor at the other end of the Direct link can receive packets 4062 without any encapsulation. In that case, the Client sends packets 4063 over the Direct link according to QoS preferences. If the Direct 4064 interface has the highest QoS preference, then the Client's IP 4065 packets are transmitted directly to the peer without going through an 4066 ANET/INET. If other interfaces have higher QoS preferences, then the 4067 Client's IP packets are transmitted via a different interface, which 4068 may result in the inclusion of Proxy/Servers and Bridges in the 4069 communications path. Direct interfaces must be tested periodically 4070 for reachability, e.g., via NUD. 4072 A.4. AERO Critical Infrastructure Considerations 4074 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 4075 IP routers or virtual machines in the cloud. Bridges must be 4076 provisioned, supported and managed by the INET administrative 4077 authority, and connected to the Bridges of other INETs via inter- 4078 domain peerings. Cost for purchasing, configuring and managing 4079 Bridges is nominal even for very large OMNI links. 4081 AERO cloud Proxy/Servers can be standard dedicated server platforms, 4082 but most often will be deployed as virtual machines in the cloud. 4083 The only requirements for cloud Proxy/Servers are that they can run 4084 the AERO user-level code and have at least one network interface 4085 connection to the INET. Cloud Proxy/Servers must be provisioned, 4086 supported and managed by the INET administrative authority. Cost for 4087 purchasing, configuring and managing cloud Proxy/Servers is nominal 4088 especially for virtual machines. 4090 AERO ANET Proxy/Servers are most often standard dedicated server 4091 platforms with one underlying interface connected to the ANET and a 4092 second interface connected to an INET. As with cloud Proxy/Servers, 4093 the only requirements are that they can run the AERO user-level code 4094 and have at least one interface connection to the INET. ANET Proxy/ 4095 Servers must be provisioned, supported and managed by the ANET 4096 administrative authority. Cost for purchasing, configuring and 4097 managing Proxys is nominal, and borne by the ANET administrative 4098 authority. 4100 AERO Relays are simply Proxy/Servers connected to INETs and/or EUNs 4101 that provide forwarding services for non-MNP destinations. The Relay 4102 connects to the OMNI link and engages in eBGP peering with one or 4103 more Bridges as a stub AS. The Relay then injects its MNPs and/or 4104 non-MNP prefixes into the BGP routing system, and provisions the 4105 prefixes to its downstream-attached networks. The Relay can perform 4106 ROS/ROR services the same as for any Proxy/Server, and can route 4107 between the MNP and non-MNP address spaces. 4109 A.5. AERO Server Failure Implications 4111 AERO Proxy/Servers may appear as a single point of failure in the 4112 architecture, but such is not the case since all Proxy/Servers on the 4113 link provide identical services and loss of a Proxy/Server does not 4114 imply immediate and/or comprehensive communication failures. Proxy/ 4115 Server failure is quickly detected and conveyed by Bidirectional 4116 Forward Detection (BFD) and/or proactive NUD allowing Clients to 4117 migrate to new Proxy/Servers. 4119 If a Proxy/Server fails, ongoing packet forwarding to Clients will 4120 continue by virtue of the neighbor cache entries that have already 4121 been established in route optimization sources (ROSs). If a Client 4122 also experiences mobility events at roughly the same time the Proxy/ 4123 Server fails, unsolicited NA messages may be lost but neighbor cache 4124 entries in the DEPARTED state will ensure that packet forwarding to 4125 the Client's new locations will continue for up to DepartTime 4126 seconds. 4128 If a Client is left without a Proxy/Server for a considerable length 4129 of time (e.g., greater than ReachableTime seconds) then existing 4130 neighbor cache entries will eventually expire and both ongoing and 4131 new communications will fail. The original source will continue to 4132 retransmit until the Client has established a new Proxy/Server 4133 relationship, after which time continuous communications will resume. 4135 Therefore, providing many Proxy/Servers on the link with high 4136 availability profiles provides resilience against loss of individual 4137 Proxy/Servers and assurance that Clients can establish new Proxy/ 4138 Server relationships quickly in event of a Proxy/Server failure. 4140 A.6. AERO Client / Server Architecture 4142 The AERO architectural model is client / server in the control plane, 4143 with route optimization in the data plane. The same as for common 4144 Internet services, the AERO Client discovers the addresses of AERO 4145 Proxy/Servers and connects to one or more of them. The AERO service 4146 is analogous to common Internet services such as google.com, 4147 yahoo.com, cnn.com, etc. However, there is only one AERO service for 4148 the link and all Proxy/Servers provide identical services. 4150 Common Internet services provide differing strategies for advertising 4151 server addresses to clients. The strategy is conveyed through the 4152 DNS resource records returned in response to name resolution queries. 4153 As of January 2020 Internet-based 'nslookup' services were used to 4154 determine the following: 4156 o When a client resolves the domainname "google.com", the DNS always 4157 returns one A record (i.e., an IPv4 address) and one AAAA record 4158 (i.e., an IPv6 address). The client receives the same addresses 4159 each time it resolves the domainname via the same DNS resolver, 4160 but may receive different addresses when it resolves the 4161 domainname via different DNS resolvers. But, in each case, 4162 exactly one A and one AAAA record are returned. 4164 o When a client resolves the domainname "ietf.org", the DNS always 4165 returns one A record and one AAAA record with the same addresses 4166 regardless of which DNS resolver is used. 4168 o When a client resolves the domainname "yahoo.com", the DNS always 4169 returns a list of 4 A records and 4 AAAA records. Each time the 4170 client resolves the domainname via the same DNS resolver, the same 4171 list of addresses are returned but in randomized order (i.e., 4172 consistent with a DNS round-robin strategy). But, interestingly, 4173 the same addresses are returned (albeit in randomized order) when 4174 the domainname is resolved via different DNS resolvers. 4176 o When a client resolves the domainname "amazon.com", the DNS always 4177 returns a list of 3 A records and no AAAA records. As with 4178 "yahoo.com", the same three A records are returned from any 4179 worldwide Internet connection point in randomized order. 4181 The above example strategies show differing approaches to Internet 4182 resilience and service distribution offered by major Internet 4183 services. The Google approach exposes only a single IPv4 and a 4184 single IPv6 address to clients. Clients can then select whichever IP 4185 protocol version offers the best response, but will always use the 4186 same IP address according to the current Internet connection point. 4187 This means that the IP address offered by the network must lead to a 4188 highly-available server and/or service distribution point. In other 4189 words, resilience is predicated on high availability within the 4190 network and with no client-initiated failovers expected (i.e., it is 4191 all-or-nothing from the client's perspective). However, Google does 4192 provide for worldwide distributed service distribution by virtue of 4193 the fact that each Internet connection point responds with a 4194 different IPv6 and IPv4 address. The IETF approach is like google 4195 (all-or-nothing from the client's perspective), but provides only a 4196 single IPv4 or IPv6 address on a worldwide basis. This means that 4197 the addresses must be made highly-available at the network level with 4198 no client failover possibility, and if there is any worldwide service 4199 distribution it would need to be conducted by a network element that 4200 is reached via the IP address acting as a service distribution point. 4202 In contrast to the Google and IETF philosophies, Yahoo and Amazon 4203 both provide clients with a (short) list of IP addresses with Yahoo 4204 providing both IP protocol versions and Amazon as IPv4-only. The 4205 order of the list is randomized with each name service query 4206 response, with the effect of round-robin load balancing for service 4207 distribution. With a short list of addresses, there is still 4208 expectation that the network will implement high availability for 4209 each address but in case any single address fails the client can 4210 switch over to using a different address. The balance then becomes 4211 one of function in the network vs function in the end system. 4213 The same implications observed for common highly-available services 4214 in the Internet apply also to the AERO client/server architecture. 4215 When an AERO Client connects to one or more ANETs, it discovers one 4216 or more AERO Proxy/Server addresses through the mechanisms discussed 4217 in earlier sections. Each Proxy/Server address presumably leads to a 4218 fault-tolerant clustering arrangement such as supported by Linux-HA, 4219 Extended Virtual Synchrony or Paxos. Such an arrangement has 4220 precedence in common Internet service deployments in lightweight 4221 virtual machines without requiring expensive hardware deployment. 4222 Similarly, common Internet service deployments set service IP 4223 addresses on service distribution points that may relay requests to 4224 many different servers. 4226 For AERO, the expectation is that a combination of the Google/IETF 4227 and Yahoo/Amazon philosophies would be employed. The AERO Client 4228 connects to different ANET access points and can receive 1-2 Proxy/ 4229 Server ADM-LLAs at each point. It then selects one AERO Proxy/Server 4230 address, and engages in RS/RA exchanges with the same Proxy/Server 4231 from all ANET connections. The Client remains with this Proxy/Server 4232 unless or until the Proxy/Server fails, in which case it can switch 4233 over to an alternate Proxy/Server. The Client can likewise switch 4234 over to a different Proxy/Server at any time if there is some reason 4235 for it to do so. So, the AERO expectation is for a balance of 4236 function in the network and end system, with fault tolerance and 4237 resilience at both levels. 4239 Appendix B. Change Log 4241 << RFC Editor - remove prior to publication >> 4243 Changes from draft-templin-6man-aero-03 to draft-templin-6man-aero- 4244 04: 4246 o Removed documents from "Obsoletes" list. 4248 o Introduced the concept of "secured" and "unsecured" spanning tree. 4250 o Additional security considerations. 4252 o Additional route optimization considerations. 4254 Changes from draft-templin-6man-aero-02 to draft-templin-6man-aero- 4255 03: 4257 o Support for extended route optimization from ROR to target over 4258 target's underlying interfaces. 4260 Changes from draft-templin-6man-aero-01 to draft-templin-6man-aero- 4261 02: 4263 o Changed reference citations to "draft-templin-6man-omni". 4265 o Several important updates to IPv6 ND cache states and route 4266 optimization message addressing. 4268 o Included introductory description of the "6M's". 4270 o Updated Multicast specification. 4272 Changes from draft-templin-6man-aero-00 to draft-templin-6man-aero- 4273 01: 4275 o Changed category to "Informational". 4277 o Updated implementation status. 4279 Changes from earlier versions to draft-templin-6man-aero-00: 4281 o Established working baseline reference. 4283 Author's Address 4285 Fred L. Templin (editor) 4286 Boeing Research & Technology 4287 P.O. Box 3707 4288 Seattle, WA 98124 4289 USA 4291 Email: fltemplin@acm.org