idnits 2.17.1 draft-templin-6man-aero-11.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 (June 3, 2021) is 1058 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 1164, but not defined == Missing Reference: 'ULA0' is mentioned on line 1164, but not defined == Missing Reference: 'ULA1' is mentioned on line 1164, but not defined == Missing Reference: 'ULA2' is mentioned on line 1164, but not defined == Unused Reference: 'RFC3971' is defined on line 3871, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 3876, but no explicit reference was found in the text == Unused Reference: 'RFC4193' is defined on line 3884, but no explicit reference was found in the text == Unused Reference: 'RFC7739' is defined on line 3912, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 3933, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-comp-rtg-hdr' is defined on line 3936, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-crh-helper-opt' is defined on line 3942, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 3947, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 3970, but no explicit reference was found in the text == Unused Reference: 'OVPN' is defined on line 3999, but no explicit reference was found in the text == Unused Reference: 'RFC2003' is defined on line 4009, but no explicit reference was found in the text == Unused Reference: 'RFC2983' is defined on line 4030, but no explicit reference was found in the text == Unused Reference: 'RFC3168' is defined on line 4034, but no explicit reference was found in the text == Unused Reference: 'RFC4122' is defined on line 4048, but no explicit reference was found in the text == Unused Reference: 'RFC4982' is defined on line 4097, but no explicit reference was found in the text == Unused Reference: 'RFC6139' is defined on line 4139, but no explicit reference was found in the text == Unused Reference: 'RFC6273' is defined on line 4159, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 4168, but no explicit reference was found in the text == Unused Reference: 'RFC6438' is defined on line 4173, but no explicit reference was found in the text == Unused Reference: 'RFC6935' is defined on line 4182, but no explicit reference was found in the text == Unused Reference: 'RFC6936' is defined on line 4187, 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 (~~), 32 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 June 3, 2021 5 Expires: December 5, 2021 7 Asymmetric Extended Route Optimization (AERO) 8 draft-templin-6man-aero-11 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 traffic selector signaling and route optimization are naturally 20 supported through dynamic neighbor cache updates. AERO is a widely- 21 applicable mobile internetworking service especially well-suited to 22 aviation services, intelligent transportation systems, mobile Virtual 23 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 December 5, 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 . . . . . . . . . . . . . . 14 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) . . . . . . . . . . 25 69 3.2.6. Segment Routing For OMNI Link Selection . . . . . . . 25 70 3.2.7. Segment Routing Within the OMNI Link . . . . . . . . 26 71 3.3. OMNI Interface Characteristics . . . . . . . . . . . . . 28 72 3.4. OMNI Interface Initialization . . . . . . . . . . . . . . 30 73 3.4.1. AERO Proxy/Server and Relay Behavior . . . . . . . . 30 74 3.4.2. AERO Client Behavior . . . . . . . . . . . . . . . . 31 75 3.4.3. AERO Bridge Behavior . . . . . . . . . . . . . . . . 31 76 3.5. OMNI Interface Neighbor Cache Maintenance . . . . . . . . 31 77 3.5.1. OMNI ND Messages . . . . . . . . . . . . . . . . . . 33 78 3.5.2. OMNI Neighbor Advertisement Message Flags . . . . . . 34 79 3.5.3. OMNI Neighbor Window Synchronization . . . . . . . . 35 80 3.6. OMNI Interface Encapsulation and Re-encapsulation . . . . 35 81 3.7. OMNI Interface Decapsulation . . . . . . . . . . . . . . 36 82 3.8. OMNI Interface Data Origin Authentication . . . . . . . . 36 83 3.9. OMNI Interface MTU . . . . . . . . . . . . . . . . . . . 37 84 3.10. OMNI Interface Forwarding Algorithm . . . . . . . . . . . 37 85 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 39 86 3.10.2. Proxy/Server and Relay Forwarding Algorithm . . . . 40 87 3.10.3. Bridge Forwarding Algorithm . . . . . . . . . . . . 43 88 3.11. OMNI Interface Error Handling . . . . . . . . . . . . . . 44 89 3.12. AERO Router Discovery, Prefix Delegation and 90 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 47 91 3.12.1. AERO Service Model . . . . . . . . . . . . . . . . . 48 92 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 48 93 3.12.3. AERO Proxy/Server Behavior . . . . . . . . . . . . . 50 94 3.13. The AERO Proxy Function . . . . . . . . . . . . . . . . . 53 95 3.13.1. Detecting and Responding to Proxy/Server Failures . 56 96 3.13.2. Point-to-Multipoint Proxy/Server Coordination . . . 57 98 3.14. AERO Route Optimization . . . . . . . . . . . . . . . . . 58 99 3.14.1. Route Optimization Initiation . . . . . . . . . . . 58 100 3.14.2. Relaying the NS(AR) *NET Packet(s) . . . . . . . . . 59 101 3.14.3. Processing the NS(AR) and Sending the NA(AR) . . . . 60 102 3.14.4. Relaying the NA(AR) . . . . . . . . . . . . . . . . 61 103 3.14.5. Processing the NA(AR) . . . . . . . . . . . . . . . 61 104 3.14.6. Forwarding Packets to Route Optimized Targets . . . 62 105 3.15. Neighbor Unreachability Detection (NUD) . . . . . . . . . 64 106 3.16. Mobility Management and Quality of Service (QoS) . . . . 66 107 3.16.1. Mobility Update Messaging . . . . . . . . . . . . . 67 108 3.16.2. Announcing Link-Layer Address and/or QoS Preference 109 Changes . . . . . . . . . . . . . . . . . . . . . . 68 110 3.16.3. Bringing New Links Into Service . . . . . . . . . . 68 111 3.16.4. Deactivating Existing Links . . . . . . . . . . . . 68 112 3.16.5. Moving Between Proxy/Servers . . . . . . . . . . . . 69 113 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 70 114 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 70 115 3.17.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 72 116 3.17.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 72 117 3.18. Operation over Multiple OMNI Links . . . . . . . . . . . 73 118 3.19. DNS Considerations . . . . . . . . . . . . . . . . . . . 73 119 3.20. Transition/Coexistence Considerations . . . . . . . . . . 74 120 3.21. Detecting and Reacting to Proxy/Server and Bridge 121 Failures . . . . . . . . . . . . . . . . . . . . . . . . 74 122 3.22. AERO Clients on the Open Internet . . . . . . . . . . . . 75 123 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 77 124 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 77 125 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 78 126 6. Security Considerations . . . . . . . . . . . . . . . . . . . 78 127 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 80 128 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 82 129 8.1. Normative References . . . . . . . . . . . . . . . . . . 82 130 8.2. Informative References . . . . . . . . . . . . . . . . . 83 131 Appendix A. Non-Normative Considerations . . . . . . . . . . . . 90 132 A.1. Implementation Strategies for Route Optimization . . . . 90 133 A.2. Implicit Mobility Management . . . . . . . . . . . . . . 90 134 A.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 91 135 A.4. AERO Critical Infrastructure Considerations . . . . . . . 91 136 A.5. AERO Server Failure Implications . . . . . . . . . . . . 92 137 A.6. AERO Client / Server Architecture . . . . . . . . . . . . 92 138 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 95 139 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 96 141 1. Introduction 143 Asymmetric Extended Route Optimization (AERO) fulfills the 144 requirements of Distributed Mobility Management (DMM) [RFC7333] and 145 route optimization [RFC5522] for aeronautical networking and other 146 network mobility use cases including intelligent transportation 147 systems and enterprise mobile device users. AERO is a secure 148 internetworking and mobility management service that employs the 149 Overlay Multilink Network Interface (OMNI) [I-D.templin-6man-omni] 150 Non-Broadcast, Multiple Access (NBMA) virtual link model. The OMNI 151 link is a virtual overlay configured over one or more underlying 152 Internetworks, and nodes on the link can exchange original IP packets 153 as single-hop neighbors. The OMNI Adaptation Layer (OAL) supports 154 end system multilink operation for increased reliability, bandwidth 155 optimization and traffic path selection while performing 156 fragmentation and reassembly to support Internetwork segment routing 157 and Maximum Transmission Unit (MTU) diversity. 159 The AERO service comprises Clients, Proxy/Servers and Relays that are 160 seen as OMNI link neighbors as well as Bridges that interconnect 161 diverse Internetworks as OMNI link segments through OAL forwarding at 162 a layer below IP. Each node's OMNI interface uses an IPv6 link-local 163 address format that supports operation of the IPv6 Neighbor Discovery 164 (ND) protocol [RFC4861] and links ND to IP forwarding. A node's OMNI 165 interface can be configured over multiple underlying interfaces, and 166 therefore appears as a single interface with multiple link-layer 167 addresses. Each link-layer address is subject to change due to 168 mobility and/or multilink fluctuations, and link-layer address 169 changes are signaled by ND messaging the same as for any IPv6 link. 171 AERO provides a secure cloud-based service where mobile node Clients 172 may use any Proxy/Server acting as a Mobility Anchor Point (MAP) and 173 fixed nodes may use any Relay on the link for efficient 174 communications. Fixed nodes forward original IP packets destined to 175 other AERO nodes via the nearest Relay, which forwards them through 176 the cloud. A mobile node's initial packets are forwarded through the 177 Proxy/Server, and direct routing is supported through route 178 optimization while packets are flowing. Both unicast and multicast 179 communications are supported, and mobile nodes may efficiently move 180 between locations while maintaining continuous communications with 181 correspondents and without changing their IP Address. 183 AERO Bridges are interconnected in a secured private BGP overlay 184 routing instance to provide an OAL routing/bridging service that 185 joins the underlying Internetworks of multiple disjoint 186 administrative domains into a single unified OMNI link at a layer 187 below IP. Each OMNI link instance is characterized by the set of 188 Mobility Service Prefixes (MSPs) common to all mobile nodes. Relays 189 provide an optimal route from correspondent nodes on the underlying 190 Internetwork to nodes on the OMNI link. To the underlying 191 Internetwork, the Relay is the source of a route to the MSP, and 192 hence uplink traffic to the mobile node is naturally routed to the 193 nearest Relay. 195 AERO can be used with OMNI links that span private-use Internetworks 196 and/or public Internetworks such as the global Internet. In the 197 latter case, some end systems may be located behind global Internet 198 Network Address Translators (NATs). A means for robust traversal of 199 NATs while avoiding "triangle routing" is therefore provided. 201 AERO assumes the use of PIM Sparse Mode in support of multicast 202 communication. In support of Source Specific Multicast (SSM) when a 203 Mobile Node is the source, AERO route optimization ensures that a 204 shortest-path multicast tree is established with provisions for 205 mobility and multilink operation. In all other multicast scenarios 206 there are no AERO dependencies. 208 AERO was designed as a secure aeronautical internetworking service 209 for both manned and unmanned aircraft, where the aircraft is treated 210 as a mobile node that can connect an Internet of Things (IoT). AERO 211 is also applicable to a wide variety of other use cases. For 212 example, it can be used to coordinate the links of mobile nodes 213 (e.g., cellphones, tablets, laptop computers, etc.) that connect into 214 a home enterprise network via public access networks with VPN or non- 215 VPN services enabled according to the appropriate security model. 216 AERO can also be used to facilitate terrestrial vehicular and urban 217 air mobility (as well as pedestrian communication services) for 218 future intelligent transportation systems 219 [I-D.ietf-ipwave-vehicular-networking][I-D.templin-ipwave-uam-its]. 220 Other applicable use cases are also in scope. 222 Along with OMNI, AERO provides secured optimal routing support for 223 the "6M's" of modern Internetworking, including: 225 1. Multilink - a mobile node's ability to coordinate multiple 226 diverse underlying data links as a single logical unit (i.e., the 227 OMNI interface) to achieve the required communications 228 performance and reliability objectives. 230 2. Multinet - the ability to span the OMNI link across multiple 231 diverse network administrative segments while maintaining 232 seamless end-to-end communications between mobile nodes and 233 correspondents such as air traffic controllers, fleet 234 administrators, etc. 236 3. Mobility - a mobile node's ability to change network points of 237 attachment (e.g., moving between wireless base stations) which 238 may result in an underlying interface address change, but without 239 disruptions to ongoing communication sessions with peers over the 240 OMNI link. 242 4. Multicast - the ability to send a single network transmission 243 that reaches multiple nodes belonging to the same interest group, 244 but without disturbing other nodes not subscribed to the interest 245 group. 247 5. Multihop - a mobile node vehicle-to-vehicle relaying capability 248 useful when multiple forwarding hops between vehicles may be 249 necessary to "reach back" to an infrastructure access point 250 connection to the OMNI link. 252 6. MTU assurance - the ability to deliver packets of various robust 253 sizes between peers without loss due to a link size restriction, 254 and to dynamically adjust packets sizes to achieve the optimal 255 performance for each independent traffic flow. 257 The following numbered sections present the AERO specification. The 258 appendices at the end of the document are non-normative. 260 2. Terminology 262 The terminology in the normative references applies; especially, the 263 terminology in the OMNI specification [I-D.templin-6man-omni] is used 264 extensively throughout. The following terms are defined within the 265 scope of this document: 267 IPv6 Neighbor Discovery (ND) 268 a control message service for coordinating neighbor relationships 269 between nodes connected to a common link. AERO uses the IPv6 ND 270 messaging service specified in [RFC4861]. 272 IPv6 Prefix Delegation 273 a networking service for delegating IPv6 prefixes to nodes on the 274 link. The nominal service is DHCPv6 [RFC8415], however alternate 275 services (e.g., based on ND messaging) are also in scope. A 276 minimal form of prefix delegation known as "prefix registration" 277 can be used if the Client knows its prefix in advance and can 278 represent it in the IPv6 source address of an ND message. 280 Access Network (ANET) 281 a node's first-hop data link service network (e.g., a radio access 282 network, cellular service provider network, corporate enterprise 283 network, etc.) that often provides link-layer security services 284 such as IEEE 802.1X and physical-layer security (e.g., "protected 285 spectrum") to prevent unauthorized access internally and with 286 border network-layer security services such as firewalls and 287 proxys that prevent unauthorized outside access. 289 ANET interface 290 a node's attachment to a link in an ANET. 292 Internetwork (INET) 293 a 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 at the network layer even though they may be 334 separated by multiple underlying *NET hops, and can use Segment 335 Routing [RFC8402] to cause packets to visit selected waypoints on 336 the link. 338 OMNI Interface 339 a node's attachment to an OMNI link. Since OMNI interface 340 addresses are managed for uniqueness, OMNI interfaces do not 341 require Duplicate Address Detection (DAD) and therefore set the 342 administrative variable 'DupAddrDetectTransmits' to zero 343 [RFC4862]. 345 OMNI Adaptation Layer (OAL) 346 an OMNI interface process whereby original IP packets admitted 347 into the interface are wrapped in a mid-layer IPv6 header and 348 subject to fragmentation and reassembly. The OAL is also 349 responsible for generating MTU-related control messages as 350 necessary, and for providing addressing context for spanning 351 multiple segments of a bridged OMNI link. 353 original IP packet 354 a whole IP packet or fragment admitted into the OMNI interface by 355 the network layer prior to OAL encapsulation and fragmentation, or 356 an IP packet delivered to the network layer by the OMNI interface 357 following OAL decapsulation and reassembly. 359 OAL packet 360 an original IP packet encapsulated in OAL headers and trailers 361 before OAL fragmentation, or following OAL reassembly. 363 OAL fragment 364 a portion of an OAL packet following fragmentation but prior to 365 *NET encapsulation, or following *NET encapsulation but prior to 366 OAL reassembly. 368 (OAL) atomic fragment 369 an OAL packet that does not require fragmentation is always 370 encapsulated as an "atomic fragment" and includes a Fragment 371 Header with Fragment Offset and More Fragments both set to 0, but 372 with a valid Identification value. 374 (OAL) carrier packet 375 an encapsulated OAL fragment following *NET encapsulation or prior 376 to *NET decapsulation. OAL sources and destinations exchange 377 carrier packets over underlying interfaces, and may be separated 378 by one or more OAL intermediate nodes. OAL intermediate nodes re- 379 encapsulate carrier packets during forwarding by removing the *NET 380 headers of the previous hop underlying network and replacing them 381 with new *NET headers for the next hop underlying network. 383 OAL source 384 an OMNI interface acts as an OAL source when it encapsulates 385 original IP packets to form OAL packets, then performs OAL 386 fragmentation and *NET encapsulation to create carrier packets. 388 OAL destination 389 an OMNI interface acts as an OAL destination when it decapsulates 390 carrier packets, then performs OAL reassembly and decapsulation to 391 derive the original IP packet. 393 OAL intermediate node 394 an OMNI interface acts as an OAL intermediate node when it removes 395 the *NET headers of carrier packets received on a first segment, 396 then re-encapsulates the carrier packets in new *NET headers and 397 forwards them into the next segment. OAL intermediate nodes 398 decrement the Hop Limit of the OAL IPv6 header during re- 399 encapsulation, and discard the packet if the Hop Limit reaches 0. 400 OAL intermediate nodes do not decrement the Hop Limit/TTL of the 401 original IP packet. 403 underlying interface 404 a *NET interface over which an OMNI interface is configured. 406 Mobility Service Prefix (MSP) 407 an aggregated IP Global Unicast Address (GUA) prefix (e.g., 408 2001:db8::/32, 192.0.2.0/24, etc.) assigned to the OMNI link and 409 from which more-specific Mobile Network Prefixes (MNPs) are 410 delegated. OMNI link administrators typically obtain MSPs from an 411 Internet address registry, however private-use prefixes can 412 alternatively be used subject to certain limitations (see: 413 [I-D.templin-6man-omni]). OMNI links that connect to the global 414 Internet advertise their MSPs to their interdomain routing peers. 416 Mobile Network Prefix (MNP) 417 a longer IP prefix delegated from an MSP (e.g., 418 2001:db8:1000:2000::/56, 192.0.2.8/30, etc.) and delegated to an 419 AERO Client or Relay. 421 Mobile Network Prefix Link Local Address (MNP-LLA) 422 an IPv6 Link Local Address that embeds the most significant 64 423 bits of an MNP in the lower 64 bits of fe80::/64, as specified in 424 [I-D.templin-6man-omni]. 426 Mobile Network Prefix Unique Local Address (MNP-ULA) 427 an IPv6 Unique-Local Address derived from an MNP-LLA. 429 Administrative Link Local Address (ADM-LLA) 430 an IPv6 Link Local Address that embeds a 32-bit administratively- 431 assigned identification value in the lower 32 bits of fe80::/96, 432 as specified in [I-D.templin-6man-omni]. 434 Administrative Unique Local Address (ADM-ULA) 435 an IPv6 Unique-Local Address derived from an ADM-LLA. 437 AERO node 438 a node that is connected to an OMNI link and participates in the 439 AERO internetworking and mobility service. 441 AERO Client ("Client") 442 an AERO node that connects over one or more underlying interfaces 443 and requests MNP delegation/registration service from AERO Proxy/ 444 Servers. The Client assigns an MNP-LLA to the OMNI interface for 445 use in ND exchanges with other AERO nodes and forwards original IP 446 packets to correspondents according to OMNI interface neighbor 447 cache state. 449 AERO Proxy/Server ("Proxy/Server") 450 a dual-function node that provides a proxying service between AERO 451 Clients and external peers on its Client-facing ANET interfaces 452 (i.e., in the same fashion as for an enterprise network proxy) as 453 well as default forwarding and Mobility Anchor Point (MAP) 454 services for coordination with correspondents on its INET-facing 455 interfaces. (Proxy/Servers in the open INET instead configure 456 only an INET interface and no ANET interfaces.) The Proxy/Server 457 configures an OMNI interface and assigns an ADM-LLA to support the 458 operation of IPv6 ND services, while advertising all of its 459 associated MNPs via BGP peerings with Bridges. Note that the 460 Proxy and Server functions can be considered logically separable, 461 but since each Proxy/Server must be informed of all of the 462 Client's other multilink Proxy/Server affiliations the AERO 463 service is best supported when the two functions are coresident on 464 the same physical or logical platform. 466 AERO Relay ("Relay") 467 a Proxy/Server that provides forwarding services between nodes 468 reached via the OMNI link and correspondents on connected 469 downstream links. AERO Relays configure an OMNI interface and 470 assign an ADM-LLA the same as Proxy/Servers. AERO Relays also run 471 a dynamic routing protocol to discover any non-MNP IP GUA routes 472 in service on its connected downstream network links. In both 473 cases, the Relay advertises the MSP(s) to its downstream networks, 474 and distributes all of its associated non-MNP IP GUA routes via 475 BGP peerings with Bridges (i.e., the same as for Proxy/Servers). 477 AERO Bridge ("Bridge") 478 a node that provides hybrid routing/bridging services (as well as 479 a security trust anchor) for nodes on an OMNI link. The Bridge 480 forwards carrier packets between OMNI link segments as OAL 481 intermediate nodes while decrementing the OAL IPv6 header Hop 482 Limit but without decrementing the network layer IP TTL/Hop Limit. 483 AERO Bridges peer with Proxy/Servers and other Bridges over 484 secured tunnels to discover the full set of MNPs for the link as 485 well as any non-MNP IP GUA routes that are reachable via Relays. 487 link-layer address 488 an IP address used as an encapsulation header source or 489 destination address from the perspective of the OMNI interface. 490 When an upper layer protocol (e.g., UDP) is used as part of the 491 encapsulation, the port number is also considered as part of the 492 link-layer address. 494 network layer address 495 the source or destination address of an original IP packet 496 presented to the OMNI interface. 498 end user network (EUN) 499 an internal virtual or external edge IP network that an AERO 500 Client or Relay connects to the rest of the network via the OMNI 501 interface. The Client/Relay sees each EUN as a "downstream" 502 network, and sees the OMNI interface as the point of attachment to 503 the "upstream" network. 505 Mobile Node (MN) 506 an AERO Client and all of its downstream-attached networks that 507 move together as a single unit, i.e., an end system that connects 508 an Internet of Things. 510 Mobile Router (MR) 511 a MN's on-board router that forwards original IP packets between 512 any downstream-attached networks and the OMNI link. The MR is the 513 MN entity that hosts the AERO Client. 515 Route Optimization Source (ROS) 516 the AERO node nearest the source that initiates route 517 optimization. The ROS may be a Proxy/Server or Relay acting on 518 behalf of the source, or may be the source Client itself. 520 Route Optimization responder (ROR) 521 the AERO node that responds to route optimization requests on 522 behalf of the target. The ROR may be a Proxy/Server acting on 523 behalf of a target MNP Client or a Relay for a non-MNP target. 525 MAP List 526 a geographically and/or topologically referenced list of addresses 527 of all Proxy/Servers within the same OMNI link. Each OMNI link 528 has its own MAP list. 530 Distributed Mobility Management (DMM) 531 a BGP-based overlay routing service coordinated by Proxy/Servers 532 and Bridges that tracks all Proxy/Server-to-Client associations. 534 Mobility Service (MS) 535 the collective set of all Proxy/Servers, Bridges and Relays that 536 provide the AERO Service to Clients. 538 Mobility Service Endpoint MSE) 539 an individual Proxy/Server, Bridge or Relay in the Mobility 540 Service. 542 Throughout the document, the simple terms "Client", "Proxy/Server", 543 "Bridge" and "Relay" refer to "AERO Client", "AERO Proxy/Server", 544 "AERO Bridge" and "AERO Relay", respectively. Capitalization is used 545 to distinguish these terms from other common Internetworking uses in 546 which they appear without capitalization. 548 The terminology of IPv6 ND [RFC4861] and DHCPv6 [RFC8415] (including 549 the names of node variables, messages and protocol constants) is used 550 throughout this document. The terms "All-Routers multicast", "All- 551 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 552 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 553 generically refer to either Internet Protocol version, i.e., IPv4 554 [RFC0791] or IPv6 [RFC8200]. 556 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 557 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 558 "OPTIONAL" in this document are to be interpreted as described in BCP 559 14 [RFC2119][RFC8174] when, and only when, they appear in all 560 capitals, as shown here. 562 3. Asymmetric Extended Route Optimization (AERO) 564 The following sections specify the operation of IP over OMNI links 565 using the AERO service: 567 3.1. AERO Node Types 569 AERO Clients are Mobile Nodes (MNs) that configure OMNI interfaces 570 over underlying interfaces with addresses that may change when the 571 Client moves to a new network connection point. AERO Clients 572 register their Mobile Network Prefixes (MNPs) with the AERO service, 573 and distribute the MNPs to nodes on EUNs. AERO Bridges, Proxy/ 574 Servers and Relays are critical infrastructure elements in fixed 575 (i.e., non-mobile) INET deployments and hence have permanent and 576 unchanging INET addresses. Together, they constitute the AERO 577 service which provides an OMNI link virtual overlay for connecting 578 AERO Clients. 580 AERO Bridges provide hybrid routing/bridging services (as well as a 581 security trust anchor) for nodes on an OMNI link. Bridges use 582 standard IPv6 routing to forward carrier packets both within the same 583 *NET partition and between disjoint *NET partitions based on an IPv6 584 encapsulation mid-layer known as the OMNI Adaptation Layer (OAL) 585 [I-D.templin-6man-omni]. During forwarding, the inner IP layer 586 experiences a virtual bridging service since the inner IP TTL/Hop 587 Limit is not decremented. Each Bridge also peers with Proxy/Servers 588 and other Bridges in a dynamic routing protocol instance to provide a 589 Distributed Mobility Management (DMM) service for the list of active 590 MNPs (see Section 3.2.3). Bridges present the OMNI link as a set of 591 one or more Mobility Service Prefixes (MSPs) and configure secured 592 tunnels with Proxy/Servers, Relays and other Bridges; they further 593 maintain IP forwarding table entries for each MNP and any other 594 reachable non-MNP prefixes. 596 AERO Proxy/Servers in distributed *NET locations provide default 597 forwarding and mobility/multilink services for AERO Client Mobile 598 Nodes (MNs). Each Proxy/Server also peers with Bridges in a dynamic 599 routing protocol instance to advertise its list of associated MNPs 600 (see Section 3.2.3). Proxy/Servers facilitate prefix delegation/ 601 registration exchanges with Clients, where each delegated prefix 602 becomes an MNP taken from an MSP. Proxy/Servers forward carrier 603 packets between OMNI interface neighbors and track each Client's 604 mobility profiles. Proxy/Servers at ANET/INET boundaries provide a 605 conduit for ANET Clients to associate with peers reached through 606 external INETs. Proxy/Servers in the open INET support INET Clients 607 through authenticated IPv6 ND message exchanges. 609 AERO Relays are Proxy/Servers that provide forwarding services to 610 exchange original IP packets between the OMNI interface and INET/EUN 611 interfaces. Relays are provisioned with MNPs the same as for an AERO 612 Client, and also run a dynamic routing protocol to discover any non- 613 MNP IP routes. The Relay advertises the MSP(s) to its connected 614 networks, and distributes all of its associated MNP and non-MNP 615 routes via BGP peerings with Bridges 617 3.2. The AERO Service over OMNI Links 618 3.2.1. AERO/OMNI Reference Model 620 Figure 1 presents the basic OMNI link reference model: 622 +----------------+ 623 | AERO Bridge B1 | 624 | Nbr: S1, S2, P1| 625 |(X1->S1; X2->S2)| 626 | MSP M1 | 627 +-+------------+-+ 628 +--------------+ | Secured | +--------------+ 629 | AERO P/S S1 | | tunnels | | AERO P/S S2 | 630 | Nbr: C1, B1 +-----+ +-----+ Nbr: C2, B1 | 631 | default->B1 | | default->B1 | 632 | X1->C1 | | X2->C2 | 633 +-------+------+ +------+-------+ 634 | OMNI link | 635 X===+===+======================================+===+===X 636 | | 637 +-----+--------+ +--------+-----+ 638 |AERO Client C1| |AERO Client C2| 639 | Nbr: S1 | | Nbr: S2 | 640 | default->S1 | | default->S2 | 641 | MNP X1 | | MNP X2 | 642 +------+-------+ +-----+--------+ 643 | | 644 .-. .-. 645 ,-( _)-. ,-( _)-. 646 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 647 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 648 `-(______)-' +-------+ +-------+ `-(______)-' 650 Figure 1: AERO/OMNI Reference Model 652 In this model: 654 o the OMNI link is an overlay network service configured over one or 655 more underlying *NET partitions which may be managed by different 656 administrative authorities and have incompatible protocols and/or 657 addressing plans. 659 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 660 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 661 via BGP peerings over secured tunnels to Proxy/Servers (S1, S2). 662 Bridges connect the disjoint segments of a partitioned OMNI link. 664 o AERO Proxy/Servers S1 and S2 configure secured tunnels with Bridge 665 B1 and also provide mobility, multilink, multicast and default 666 router services for the MNPs of their associated Clients C1 and 667 C2. (AERO Proxy/Servers that act as Relays can also advertise 668 non-MNP routes for non-mobile correspondent nodes the same as for 669 MNP Clients.) 671 o AERO Clients C1 and C2 associate with Proxy/Servers S1 and S2, 672 respectively. They receive MNP delegations X1 and X2, and also 673 act as default routers for their associated physical or internal 674 virtual EUNs. Simple hosts H1 and H2 attach to the EUNs served by 675 Clients C1 and C2, respectively. 677 An OMNI link configured over a single *NET appears as a single 678 unified link with a consistent underlying network addressing plan. 679 In that case, all nodes on the link can exchange carrier packets via 680 simple *NET encapsulation (i.e., following any necessary NAT 681 traversal), since the underlying *NET is connected. In common 682 practice, however, an OMNI link may be partitioned into multiple 683 "segments", where each segment is a distinct *NET potentially managed 684 under a different administrative authority (e.g., as for worldwide 685 aviation service providers such as ARINC, SITA, Inmarsat, etc.). 686 Individual *NETs may also themselves be partitioned internally, in 687 which case each internal partition is seen as a separate segment. 689 The addressing plan of each segment is consistent internally but will 690 often bear no relation to the addressing plans of other segments. 691 Each segment is also likely to be separated from others by network 692 security devices (e.g., firewalls, proxys, packet filtering gateways, 693 etc.), and in many cases disjoint segments may not even have any 694 common physical link connections. Therefore, nodes can only be 695 assured of exchanging carrier packets directly with correspondents in 696 the same segment, and not with those in other segments. The only 697 means for joining the segments therefore is through inter-domain 698 peerings between AERO Bridges. 700 The same as for traditional campus LANs, multiple OMNI link segments 701 can be joined into a single unified link via a virtual bridging 702 service using the OMNI Adaptation Layer (OAL) [I-D.templin-6man-omni] 703 which inserts a mid-layer IPv6 encapsulation header that supports 704 inter-segment forwarding (i.e., bridging) without decrementing the 705 network-layer TTL/Hop Limit of the original IP packet. This bridging 706 of OMNI link segments is shown in Figure 2: 708 . . . . . . . . . . . . . . . . . . . . . . . 709 . . 710 . .-(::::::::) . 711 . .-(::::::::::::)-. +-+ . 712 . (:::: Segment A :::)--|B|---+ . 713 . `-(::::::::::::)-' +-+ | . 714 . `-(::::::)-' | . 715 . | . 716 . .-(::::::::) | . 717 . .-(::::::::::::)-. +-+ | . 718 . (:::: Segment B :::)--|B|---+ . 719 . `-(::::::::::::)-' +-+ | . 720 . `-(::::::)-' | . 721 . | . 722 . .-(::::::::) | . 723 . .-(::::::::::::)-. +-+ | . 724 . (:::: Segment C :::)--|B|---+ . 725 . `-(::::::::::::)-' +-+ | . 726 . `-(::::::)-' | . 727 . | . 728 . ..(etc).. x . 729 . . 730 . . 731 . <- OMNI link Bridged by encapsulation -> . 732 . . . . . . . . . . . . . .. . . . . . . . . 734 Figure 2: Bridging OMNI Link Segments 736 Bridges, Proxy/Servers and Relay OMNI interfaces are configured over 737 both secured tunnels and open INET underlying interfaces over their 738 respective segments in a spanning tree topology rooted at the 739 Bridges. The "secured" spanning tree supports strong authentication 740 for control plane messages and may also be used to convey the initial 741 carrier packets in a flow. The "unsecured" spanning tree conveys 742 ordinary carrier packets without security codes and that must be 743 treated by destinations according to data origin authentication 744 procedures. Route optimization can be employed to cause carrier 745 packets to take more direct paths between OMNI link neighbors without 746 having to follow strict spanning tree paths. 748 3.2.2. Addressing and Node Identification 750 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 751 fe80::/64 [RFC4291] to assign LLAs used for network-layer addresses 752 in link-scoped IPv6 ND and data messages. AERO Clients use LLAs 753 constructed from MNPs (i.e., "MNP-LLAs") while other AERO nodes use 754 LLAs constructed from administrative identification values ("ADM- 755 LLAs") as specified in [I-D.templin-6man-omni]. Non-MNP routes are 756 also represented the same as for MNP-LLAs, but may include a prefix 757 that is not properly covered by the MSP. 759 AERO nodes also use the Unique Local Address (ULA) prefix fd00::/8 760 followed by a pseudo-random 40-bit OMNI domain identifier to form the 761 prefix [ULA]::/48, then include a 16-bit OMNI link identifier '*' to 762 form the prefix [ULA*]::/64 [RFC4291]. The AERO node then uses the 763 prefix [ULA*]::/64 to form "MNP-ULAs" or "ADM-ULA"s as specified in 764 [I-D.templin-6man-omni] to support OAL addressing. (The prefix 765 [ULA*]::/64 appearing alone and with no suffix represents "default".) 766 AERO Clients also use Temporary ULAs constructed per 767 [I-D.templin-6man-omni], where the addresses are typically used only 768 in initial control message exchanges until a stable MNP-LLA/ULA is 769 assigned. 771 AERO MSPs, MNPs and non-MNP routes are typically based on Global 772 Unicast Addresses (GUAs), but in some cases may be based on private- 773 use addresses. See [I-D.templin-6man-omni] for a full specification 774 of LLAs, ULAs and GUAs used by AERO nodes on OMNI links. 776 Finally, AERO Clients and Proxy/Servers configure node identification 777 values as specified in [I-D.templin-6man-omni]. 779 3.2.3. AERO Routing System 781 The AERO routing system comprises a private instance of the Border 782 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 783 and Proxy/Servers and does not interact with either the public 784 Internet BGP routing system or any underlying INET routing systems. 786 In a reference deployment, each Proxy/Server is configured as an 787 Autonomous System Border Router (ASBR) for a stub Autonomous System 788 (AS) using a 32-bit AS Number (ASN) [RFC4271] that is unique within 789 the BGP instance, and each Proxy/Server further uses eBGP to peer 790 with one or more Bridges but does not peer with other Proxy/Servers. 791 Each *NET of a multi-segment OMNI link must include one or more 792 Bridges, which peer with the Proxy/Servers within that *NET. All 793 Bridges within the same *NET are members of the same hub AS, and use 794 iBGP to maintain a consistent view of all active routes currently in 795 service. The Bridges of different *NETs peer with one another using 796 eBGP. 798 Bridges maintain forwarding table entries only for the MNP-ULAs 799 corresponding to MNP and non-MNP routes that are currently active, 800 and carrier packets destined to all other MNP-ULAs will correctly 801 incur Destination Unreachable messages due to the black-hole route. 802 In this way, Proxy/Servers and Relays have only partial topology 803 knowledge (i.e., they only maintain routing information for their 804 directly associated Clients and non-AERO links) and they forward all 805 other carrier packets to Bridges which have full topology knowledge. 807 Each OMNI link segment assigns a unique ADM-ULA sub-prefix of 808 [ULA*]::/96. For example, a first segment could assign 809 [ULA*]::1000/116, a second could assign [ULA*]::2000/116, a third 810 could assign [ULA*]::3000/116, etc. Within each segment, each Proxy/ 811 Server configures an ADM-ULA within the segment's prefix, e.g., the 812 Proxy/Servers within [ULA*]::2000/116 could assign the ADM-ULAs 813 [ULA*]::2011/116, [ULA*]::2026/116, [ULA*]::2003/116, etc. 815 The administrative authorities for each segment must therefore 816 coordinate to assure mutually-exclusive ADM-ULA prefix assignments, 817 but internal provisioning of ADM-ULAs an independent local 818 consideration for each administrative authority. For each ADM-ULA 819 prefix, the Bridge(s) that connect that segment assign the all-zero's 820 address of the prefix as a Subnet Router Anycast address. For 821 example, the Subnet Router Anycast address for [ULA*]::1023/116 is 822 simply [ULA*]::1000. 824 ADM-ULA prefixes are statically represented in Bridge forwarding 825 tables. Bridges join multiple segments into a unified OMNI link over 826 multiple diverse administrative domains. They support a bridging 827 function by first establishing forwarding table entries for their 828 ADM-ULA prefixes either via standard BGP routing or static routes. 829 For example, if three Bridges ('A', 'B' and 'C') from different 830 segments serviced [ULA*]::1000/116, [ULA*]::2000/116 and 831 [ULA*]::3000/116 respectively, then the forwarding tables in each 832 Bridge are as follows: 834 A: [ULA*]::1000/116->local, [ULA*]::2000/116->B, [ULA*]::3000/116->C 836 B: [ULA*]::1000/116->A, [ULA*]::2000/116->local, [ULA*]::3000/116->C 838 C: [ULA*]::1000/116->A, [ULA*]::2000/116->B, [ULA*]::3000/116->local 840 These forwarding table entries are permanent and never change, since 841 they correspond to fixed infrastructure elements in their respective 842 segments. 844 MNP ULAs are instead dynamically advertised in the AERO routing 845 system by Proxy/Servers and Relays that provide service for their 846 corresponding MNPs. For example, if three Proxy/Servers ('D', 'E' 847 and 'F') service the MNPs 2001:db8:1000:2000::/56, 848 2001:db8:3000:4000::/56 and 2001:db8:5000:6000::/56 then the routing 849 system would include: 851 D: [ULA*]:2001:db8:1000:2000/120 852 E: [ULA*]:2001:db8:3000:4000/120 854 F: [ULA*]:2001:db8:5000:6000/120 856 A full discussion of the BGP-based routing system used by AERO is 857 found in [I-D.ietf-rtgwg-atn-bgp]. 859 3.2.4. OMNI Link Segment Routing 861 With the Client and partition prefixes in place in Bridge forwarding 862 tables, the OMNI interface sends control and data carrier packets 863 toward AERO destination nodes located in different OMNI link segments 864 over the spanning tree. The OMNI interface uses the OMNI Adaptation 865 Layer (OAL) encapsulation service [I-D.templin-6man-omni], and 866 includes an OMNI Routing Header (ORH) as an extension to the OAL 867 header. Each carrier packet includes at most one ORH in compressed 868 or uncompressed form, with the uncompressed form shown in Figure 3: 870 0 1 2 3 871 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 1 872 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 873 | Next Header | Hdr Ext Len | Routing Type | Segments Left | 874 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 875 | omIndex | FMT | SRT | LHS (bits 0 -15) | 876 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 877 | LHS (bits 0 -15) | ~ 878 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 879 ~ Link Layer Address (L2ADDR) ~ 880 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 881 | Null Padding (if necessary) | 882 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 883 ~ Destination Suffix ~ 884 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 886 Figure 3: OMNI Routing Header (ORH) Format 888 The ORH includes the following fields, in consecutive order: 890 o Next Header identifies the type of header immediately following 891 the ORH. 893 o Hdr Ext Len is the length of the Routing header in 8-octet units 894 (not including the first 8 octets). The field must encode a value 895 between 0 and 4 (all other values are treated as a parameter 896 problem). 898 o Routing Type is set to TBD1 (see IANA Considerations). 900 o Segments Left encodes the value 0 or 1 (all other values are 901 treated as a parameter problem). 903 o omIndex - a 1-octet field consulted only when Segments Left is 0; 904 identifies a specific target Client underlying interface serviced 905 by the LHS Proxy-Server when there are multiple alternatives. 906 When FMT-Forward is clear, omIndex determines the interface for 907 forwarding the ORH packet following reassembly; when FMT-Forward 908 is set, omIndex determines the interface for forwarding the raw 909 carrier packets without first reassembling. When omIndex is set 910 to 0 (or when no ORH is present), the LHS Proxy/Server selects 911 among any of the Client's available underlying interfaces that it 912 services locally (i.e., and not those serviced by another Proxy/ 913 Server). 915 o FMT - a 3-bit "Forward/Mode/Trailer" code corresponding to the 916 included Link Layer Address as follows: 918 * When the most significant bit (i.e., "FMT-Forward") is clear, 919 the Last Hop Segment (LHS) Proxy/Server must reassemble. When 920 FMT-Forward is set, the LHS Proxy/Server must forward the 921 fragments to the Client (while changing the OAL destination 922 address to the MNP-ULA of the Client if necessary) without 923 reassembling. 925 * When the next most significant bit (i.e., "FMT-Mode") is clear, 926 L2ADDR is the INET address of the LHS Proxy/Server and the 927 Client must be reached through the LHS Proxy/Server. When FMT- 928 Mode is set, the Client is eligible for route optimization over 929 the open INET where it may be located behind one or more NATs, 930 and L2ADDR is either the INET address of the LHS Proxy/Server 931 (when FMT-Forward is set) or the native INET address of the 932 Client itself (when FMT-Forward is clear). 934 * The least significant bit (i.e., "FMT-Type") is consulted only 935 when Hdr Ext Len is 1 and ignored otherwise. If FMT-Type is 936 clear, the remaining 10 ORH octets contain an LHS followed by 937 an IPv4 L2ADDR. If FMT-Type is set, the remainder instead 938 contains 2 null padding octets followed by an 8-octet (IPv6) 939 Destination Suffix. 941 o SRT - a 5-bit Segment Routing Topology prefix length consulted 942 only when Segments Left is 1, and encodes a value that (when added 943 to 96) determines the prefix length to apply to the ADM-ULA formed 944 from concatenating [ULA*]::/96 with the 32 bit LHS value (for 945 example, the value 16 corresponds to the prefix length 112). 947 o LHS - a 4-octet field present only when indicated by the ORH 948 length (see below) and consulted only when Segments Left is 1. 949 The field encodes the 32-bit ADM-ULA suffix of a Last Hop Segment 950 (LHS) Proxy/Server for the target. When SRT and LHS are both set 951 to 0, the LHS must be reached directly via INET encapsulation 952 instead of over the spanning tree. When SRT is set to 0 and LHS 953 is non-zero, the prefix length is set to 128. SRT and LHS 954 determine the ADM-ULA of the LHS Proxy/Server over the spanning 955 tree. 957 o Link Layer Address (L2ADDR) - an IP encapsulation address present 958 only when indicated by the ORH length (see below) and consulted 959 only when Segments Left is 1. The ORH length also determines the 960 L2ADDR IP version since the field will always contain exactly 6 961 octets for UDP/IPv4 or 18 octets for UDP/IPv6. When present, 962 provides the link-layer address (i.e., the encapsulation address) 963 of the Proxy/Server or the target Client itself. The UDP Port 964 Number appears in the first two octets and the IP address appears 965 in the remaining octets. The Port Number and IP address are 966 recorded in network byte order, and in ones-compliment 967 "obfuscated" form per [RFC4380]. The OMNI interface forwarding 968 algorithm uses L2ADDR as the INET encapsulation address for 969 forwarding when SRT/LHS is located in the same OMNI link segment. 970 If direct INET encapsulation is not permitted, L2ADDR is instead 971 set to all-zeros and the packet must be forwarded to the LHS 972 Proxy-Server via the spanning tree. 974 o Null Padding - zero-valued octets added as necessary to pad the 975 portion of the ORH included up to this point to an even 8-octet 976 boundary. 978 o Destination Suffix - a trailing 8-octet field present only when 979 indicated by the ORH length (see below). When ORH length is 1, 980 FMT-Type determines whether the option includes a Destination 981 Suffix or an LHS/L2ADDR for IPv4 since there is only enough space 982 available for one. When present, encodes the 64-bit MNP-ULA 983 suffix for the target Client. 985 The ORH Hdr Ext Len field value also serves as an implicit ORH 986 "Type", with 5 distinct Types specified (i.e., ORH-0 through ORH-4). 987 All ORH-* Types include the same 6-octet preamble beginning with Next 988 Header up to and including omIndex, followed by a Type-specific 989 remainder as follows: 991 o ORH-0 - The preamble Hdr Ext Len and Segments Left must both be 0. 992 Two null padding octets follow the preamble, and all other fields 993 are omitted. 995 o ORH-1 - The preamble Hdr Ext Len is set to 1. When FMT-Type is 996 clear, the LHS and L2ADDR for IPv4 fields are included and the 997 Destination Suffix is omitted. When FMT-Type is set, the LHS and 998 L2ADDR fields are omitted, the Destination Suffix field is 999 included and Segments Left must be 0. 1001 o ORH-2 - The preamble Hdr Ext Len is set to 2. The LHS, L2ADDR for 1002 IPv4 and Destination Suffix fields are all included. 1004 o ORH-3 - The preamble Hdr Ext Len is set to 3. The LHS and L2ADDR 1005 for IPv6 fields are included and the Destination Suffix field is 1006 omitted. 1008 o ORH-4 - The preamble Hdr Ext Len is set to 4. The LHS, L2ADDR for 1009 IPv6 and Destination Suffix fields are all included. 1011 AERO neighbors use OAL encapsulation and fragmentation to exchange 1012 OAL packets as specified in [I-D.templin-6man-omni]. When an AERO 1013 node's OMNI interface (acting as an OAL source) uses OAL 1014 encapsulation for an original IP packet with source address 1015 2001:db8:1:2::1 and destination address 2001:db8:1234:5678::1, it 1016 sets the OAL header source address to its own ULA (e.g., 1017 [ULA*]::2001:db8:1:2), sets the destination address to the MNP-ULA 1018 corresponding to the IP destination address (e.g., 1019 [ULA*]::2001:db8:1234:5678), sets the Traffic Class, Flow Label, Hop 1020 Limit and Payload Length as discussed in [I-D.templin-6man-omni], 1021 then finally selects an Identification and appends an OAL checksum. 1023 If the neighbor cache information indicates that the target is in a 1024 different segment, the OAL source next inserts an ORH immediately 1025 following the OAL header while including Destination Suffix for non- 1026 first-fragments only when necessary (in this case, the Destination 1027 Suffix is 2001:db8:1234:5678). Next, to direct the packet to a 1028 first-hop Proxy/Server or a Bridge, the source prepares an ORH with 1029 Segments Left set to 1 and with SRT/LHS/L2ADDR included, then 1030 overwrites the OAL header destination address with the LHS Subnet 1031 Router Anycast address (for example, for LHS 3000:4567 with SRT 16, 1032 the Subnet Router Anycast address is [ULA*]::3000:0000). To send the 1033 packet to the LHS Proxy/Server either directly or via the spanning 1034 tree, the OAL source instead includes an ORH (Type 0 or 1) with 1035 Segments Left set to 0 and LHS/L2ADDR omitted, and overwrites the OAL 1036 header destination address with either the LHS Proxy/Server ADM-ULA 1037 or the MNP-ULA of the Client itself. 1039 The OAL source then fragments the OAL packet, with each resulting OAL 1040 fragment including the OAL/ORH headers while only the first fragment 1041 includes the original IPv6 header. If FMT-Forward is set, the 1042 Identification used for fragmentation must be within the window for 1043 the Client and a Destination Suffix must be included with each non- 1044 first-fragment when necessary; otherwise the Identification must be 1045 within the window for the Client's Proxy/Server and no Destination 1046 Suffix is needed. (Note that if no actual fragmentation is required 1047 the OAL source still prepares the packet as an "atomic" fragment that 1048 includes a Fragment Header with Offset and More Fragments both set to 1049 0.) The OAL source finally encapsulates each resulting OAL fragment 1050 in an *NET header to form an OAL carrier packet, with source address 1051 set to its own *NET address (e.g., 192.0.2.100) and destination set 1052 to the *NET address of the last hop itself or the next hop in the 1053 spanning tree (e.g., 192.0.2.1). 1055 The carrier packet encapsulation format in the above example is shown 1056 in Figure 4: 1058 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1059 | *NET Header | 1060 | src = 192.0.2.100 | 1061 | dst = 192.0.2.1 | 1062 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1063 | OAL IPv6 Header | 1064 | src = [ULA*]::2001:db8:1:2 | 1065 | dst= [ULA*]::3000:0000 | 1066 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1067 | ORH (if necessary) | 1068 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1069 | OAL Fragment Header | 1070 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1071 | Original IP Header | 1072 | (first-fragment only) | 1073 | src = 2001:db8:1:2::1 | 1074 | dst = 2001:db8:1234:5678::1 | 1075 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1076 | | 1077 ~ ~ 1078 ~ Original Packet Body/Fragment ~ 1079 ~ ~ 1080 | | 1081 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1083 Figure 4: Carrier Packet Format 1085 In this format, the original IP header and packet body/fragment are 1086 from the original IP packet, the OAL header is an IPv6 header 1087 prepared according to [RFC2473], the ORH is a Routing Header 1088 extension of the OAL header, the Fragment Header identifies each 1089 fragment, and the INET header is prepared as discussed in 1090 Section 3.6. When the OAL source transmits the resulting carrier 1091 packets, they are forwarded over possibly multiple OAL intermediate 1092 nodes in the OMNI link spanning tree until they arrive at the OAL 1093 destination. 1095 This gives rise to a routing system that contains both Client MNP-ULA 1096 routes that may change dynamically due to regional node mobility and 1097 per-partition ADM-ULA routes that rarely if ever change. The 1098 spanning tree can therefore provide link-layer bridging by sending 1099 carrier packets over the spanning tree instead of network-layer 1100 routing according to MNP routes. As a result, opportunities for loss 1101 due to node mobility between different segments are mitigated. 1103 Note: The document recommends that AERO nodes transform ORHs with 1104 Segments Left set to 1 into ORH-0 or ORH-1 during forwarding. While 1105 this may yield encapsulation overhead savings in some cases, the AERO 1106 node may instead simply set Segments Left to 0 and leave the original 1107 ORH in place. The LHS Proxy/Server or target Client that processes 1108 the ORH will receive the same information in both cases. 1110 Note: When the OAL source sets a carrier packet OAL destination 1111 address to a target's MNP-ULA but does not assert a specific target 1112 underlying interface, it may omit the ORH whether forwarding to the 1113 LHS Proxy/Server or directly to the target itself. When the LHS 1114 Proxy/Server receives a carrier packet with OAL destination set to 1115 the target MNP-ULA but with no ORH, it forwards over any available 1116 underlying interface for the target that it services locally. 1118 Note: When the OAL source and destination are on the same INET 1119 segment, OAL header compression can be used to significantly reduce 1120 encapsulation overhead [I-D.templin-6man-omni]. 1122 Note: When the OAL source has multiple original IP packets to send to 1123 the same OAL destination, it can perform "packing" to generate a 1124 "super-packet" [I-D.templin-6man-omni]. In that case, the OAL/ORH 1125 super-packet may include up to N original IP packets as long as the 1126 total length of the super-packet does not exceed the OMNI interface 1127 MTU. 1129 Note: Use of an IPv6 "minimal encapsulation" format (i.e., an IPv6 1130 variant of [RFC2004]) based on extensions to the ORH was considered 1131 and abandoned. In the approach, the ORH would be inserted as an 1132 extension header to the original IPv6 packet header. The IPv6 1133 destination address would then be written into the ORH, and the ULA 1134 corresponding to the destination would be overwritten in the IPv6 1135 destination address. This would seemingly convey enough forwarding 1136 information so that OAL encapsulation could be avoided. However, 1137 this "minimal encapsulation" IPv6 packet would then have a non-ULA 1138 source address and ULA destination address, an incorrect value in 1139 upper layer protocol checksums, and a Hop Limit that is decremented 1140 within the spanning tree when it should not be. The insertion and 1141 removal of the ORH would also entail rewriting the Payload Length and 1142 Next Header fields - again, invalidating upper layer checksums. 1143 These irregularities would result in implementation challenges and 1144 the potential for operational issues, e.g., since actionable ICMPv6 1145 error reports could not be delivered to the original source. In 1146 order to address the issues, still more information such as the 1147 original IPv6 source address could be written into the ORH. However, 1148 with the additional information the benefit of the "minimal 1149 encapsulation" savings quickly diminishes, and becomes overshadowed 1150 by the implementation and operational irregularities. 1152 3.2.5. Segment Routing Topologies (SRTs) 1154 The 64-bit sub-prefixes of [ULA]::/48 identify up to 2^16 distinct 1155 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 1156 OMNI link overlay instance using a distinct set of ULAs, and emulates 1157 a Virtual LAN (VLAN) service for the OMNI link. In some cases (e.g., 1158 when redundant topologies are needed for fault tolerance and 1159 reliability) it may be beneficial to deploy multiple SRTs that act as 1160 independent overlay instances. A communication failure in one 1161 instance therefore will not affect communications in other instances. 1163 Each SRT is identified by a distinct value in bits 48-63 of 1164 [ULA]::/48, i.e., as [ULA0]::/64, [ULA1]::/64, [ULA2]::/64, etc. 1165 Each OMNI interface is identified by a unique interface name (e.g., 1166 omni0, omni1, omni2, etc.) and assigns an anycast ADM-ULA 1167 corresponding to its SRT prefix length. The anycast ADM-ULA is used 1168 for OMNI interface determination in Safety-Based Multilink (SBM) as 1169 discussed in [I-D.templin-6man-omni]. Each OMNI interface further 1170 applies Performance-Based Multilink (PBM) internally. 1172 3.2.6. Segment Routing For OMNI Link Selection 1174 An original IPv6 source can direct an IPv6 packet to an AERO node by 1175 including a standard IPv6 Segment Routing Header (SRH) [RFC8754] with 1176 the anycast ADM-ULA for the selected SRT as either the IPv6 1177 destination or as an intermediate hop within the SRH. This allows 1178 the original source to determine the specific OMNI link topology an 1179 original IPv6 packet will traverse when there may be multiple 1180 alternatives. 1182 When the AERO node processes the SRH and forwards the original IPv6 1183 packet to the correct OMNI interface, the OMNI interface writes the 1184 next IPv6 address from the SRH into the IPv6 destination address and 1185 decrements Segments Left. If decrementing would cause Segments Left 1186 to become 0, the OMNI interface deletes the SRH before forwarding. 1187 This form of Segment Routing supports Safety-Based Multilink (SBM). 1189 3.2.7. Segment Routing Within the OMNI Link 1191 OAL sources can insert an ORH for Segment Routing within the OMNI 1192 link to influence the paths of carrier packets sent to OAL 1193 destinations in remote segments without requiring all carrier packets 1194 to traverse strict spanning tree paths. (OAL sources can also insert 1195 an ORH in carrier packets sent to OAL destinations in the local 1196 segment if additional last-hop forwarding information is required.) 1198 When an AERO node's OMNI interface has an original IP packet to send 1199 to a target discovered through route optimization located in the same 1200 OMNI link segment, it acts as an OAL source to perform OAL 1201 encapsulation and fragmentation. The node then uses L2ADDR for INET 1202 encapsulation while including an ORH-0 when sending the resulting 1203 carrier packets to the ADM-ULA of the LHS Proxy/Server, or optionally 1204 omitting the ORH-0 when sending to the MNP-ULA of the target Client 1205 itself. When the node sends carrier packets with an ORH-0 to the LHS 1206 Proxy/Server, it sets the OAL destination to the ADM-ULA of the 1207 Proxy/Server if the Proxy/Server is responsible for reassembly; 1208 otherwise, it sets the OAL destination to the MNP-ULA of the target 1209 Client to cause the Proxy/Server to forward without reassembling. 1210 The node also sets omIndex to select a specific target Client 1211 underlying interface, or sets omIndex to 0 when no preference is 1212 selected. 1214 When an AERO node's OMNI interface has an original IP packet to send 1215 to a route optimization target located in a remote OMNI link segment, 1216 it acts as an OAL source the same as above but also includes an 1217 appropriate ORH type with Segments Left set to 1 and with SRT/LHS/ 1218 L2ADDR information while setting the OAL destination to the Subnet 1219 Router Anycast address for the LHS OMNI link segment. (The OAL 1220 source can alternatively include an ORH with Segments Left set to 0 1221 while setting the OAL destination to the ADM-ULA of the LHS Proxy/ 1222 Server, but this would cause the carrier packets to follow strict 1223 spanning tree paths.) The OMNI interface then forwards the resulting 1224 carrier packets into the spanning tree. 1226 When a Bridge receives a carrier packet destined to its Subnet Router 1227 Anycast address with any ORH type with Segments Left set to 1 and 1228 with SRT/LHS/L2ADDR values corresponding to the local segment, it 1229 examines FMT-Mode to determine whether the target Client can accept 1230 packets directly (i.e., following any NAT traversal procedures 1231 necessary) while bypassing the LHS Proxy/Server. If the Client can 1232 be reached directly and NAT traversal has converged, the Bridge then 1233 writes the MNP-ULA (found in the inner IPv6 header for first 1234 fragments or the ORH Destination Suffix for non-first fragments) into 1235 the OAL destination address, decrements the OAL IPv6 header Hop Limit 1236 (and discards the packet if Hop Limit reaches 0), removes the ORH, 1237 re-encapsulates the carrier packet according to L2ADDR then forwards 1238 the carrier packet directly to the target Client. If the Client 1239 cannot be reached directly (or if NAT traversal has not yet 1240 converged), the Bridge instead transforms the ORH into an ORH-0, re- 1241 encapsulates the packet according to L2ADDR, changes the OAL 1242 destination to the ADM-ULA of the LHS Proxy/Server if FMT-Forward is 1243 clear or the MNP-ULA of the Client if FMT-Forward is set and forwards 1244 the carrier packet to the LHS Proxy/Server. 1246 When a Bridge receives a carrier packet destined to its Subnet Router 1247 Anycast address with any ORH type with Segments Left set to 1 and 1248 L2ADDR set to 0, the Bridge instead forwards the packet toward the 1249 LHS Proxy/Server via the spanning tree. The Bridge changes the OAL 1250 destination to the ADM-ULA of the LHS Proxy/Server, transforms the 1251 ORH into an ORH-0 (or an ORH-1 with FMT-Type set and Segments Left 1252 0), then forwards the packet to the next hop in the spanning tree. 1253 The Bridge may also elect to forward via the spanning tree as above 1254 even when it receives a carrier packet with an ORH that includes a 1255 valid L2ADDR Port Number and IP address, however this may result in a 1256 longer path than necessary. If the carrier packet arrived via the 1257 secured spanning tree, the Bridge forwards to the next hop also via 1258 the secured spanning tree. If the carrier packet arrived via the 1259 unsecured spanning tree, the Bridge forwards to the next hop also via 1260 the unsecured spanning tree. 1262 When an LHS Proxy/Server receives carrier packets with any ORH type 1263 with Segments Left set to 0 and with OAL destination set to its own 1264 ADM-ULA, it proceeds according to FMT-Forward and omIndex. If FMT- 1265 Forward is set, the LHS Proxy/Server changes the OAL destination to 1266 the MNP-ULA of the target Client found in the IPv6 header for first 1267 fragments or in the ORH Destination Suffix for non-first-fragments, 1268 removes the ORH and forwards to the target Client interface 1269 identified by omIndex. If FMT-Forward is clear, the LHS Proxy/Server 1270 instead reassembles then re-encapsulates while refragmenting if 1271 necessary, removes the ORH and forwards to the target Client 1272 according to omIndex. 1274 When an LHS Proxy/Server receives carrier packets with any ORH type 1275 with Segments Left set to 0 and with OAL destination set to the MNP- 1276 ULA of the target Client, it removes the ORH and forwards to the 1277 target Client according to omIndex. During forwarding, the LHS 1278 Proxy/Server must first verify that the omIndex corresponds to a 1279 target underlying interface that it services locally and must not 1280 forward to other target underlying interfaces. If omIndex is 0 (or 1281 if no ORH is included) the LHS Proxy/Server instead selects among any 1282 of the target underlying interfaces it services. 1284 When a target Client receives carrier packets with OAL destination 1285 set to is MNP-ULA, it reassembles to obtain the OAL packet then 1286 decapsulates and delivers the original IP packet to upper layers. 1288 Note: Special handling procedures are employed for the exchange of 1289 IPv6 ND messages used to establish neighbor cache state as discussed 1290 in Section 3.14. The procedures call for hop-by-hop authentication 1291 and neighbor cache state establishment based on the encapsulation 1292 ULA, with next-hop determination based on the IPv6 ND message LLA. 1294 3.3. OMNI Interface Characteristics 1296 OMNI interfaces are virtual interfaces configured over one or more 1297 underlying interfaces classified as follows: 1299 o INET interfaces connect to an INET either natively or through one 1300 or more NATs. Native INET interfaces have global IP addresses 1301 that are reachable from any INET correspondent. The INET-facing 1302 interfaces of Proxy/Servers are native interfaces, as are Relay 1303 and Bridge interfaces. NATed INET interfaces connect to a private 1304 network behind one or more NATs that provide INET access. Clients 1305 that are behind a NAT are required to send periodic keepalive 1306 messages to keep NAT state alive when there are no carrier packets 1307 flowing. 1309 o ANET interfaces connect to an ANET that is separated from the open 1310 INET by a Proxy/Server. Clients can issue control messages over 1311 the ANET without including an authentication signature since the 1312 ANET is secured at the network layer or below. Proxy/Servers can 1313 actively issue control messages over the INET on behalf of ANET 1314 Clients to reduce ANET congestion. 1316 o VPNed interfaces use security encapsulation over the INET to a 1317 Virtual Private Network (VPN) server that also acts as a Proxy/ 1318 Server. Other than the link-layer encapsulation format, VPNed 1319 interfaces behave the same as Direct interfaces. 1321 o Direct (i.e., single-hop point-to-point) interfaces connect a 1322 Client directly to a Proxy/Server without crossing any ANET/INET 1323 paths. An example is a line-of-sight link between a remote pilot 1324 and an unmanned aircraft. The same Client considerations apply as 1325 for VPNed interfaces. 1327 OMNI interfaces use OAL encapsulation and fragmentation as discussed 1328 in Section 3.2.4. OMNI interfaces use *NET encapsulation (see: 1330 Section 3.6) to exchange carrier packets with OMNI link neighbors 1331 over INET or VPNed interfaces as well as over ANET interfaces for 1332 which the Client and Proxy/Server may be multiple IP hops away. OMNI 1333 interfaces do not use link-layer encapsulation over Direct underlying 1334 interfaces or ANET interfaces when the Client and Proxy/Server are 1335 known to be on the same underlying link. 1337 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1338 state the same as for any interface. OMNI interfaces use ND messages 1339 including Router Solicitation (RS), Router Advertisement (RA), 1340 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1341 neighbor cache management. In environments where spoofing may be a 1342 threat, OMNI neighbors should employ OAL Identification window 1343 synchronization in their ND message exchanges. 1345 OMNI interfaces send ND messages with an OMNI option formatted as 1346 specified in [I-D.templin-6man-omni]. The OMNI option includes 1347 prefix registration information, Interface Attributes containing link 1348 information parameters for the OMNI interface's underlying interfaces 1349 and any other per-neighbor information. Each OMNI option may include 1350 multiple Interface Attributes sub-options identified by non-zero 1351 omIndex values. 1353 A Client's OMNI interface may be configured over multiple underlying 1354 interface connections. For example, common mobile handheld devices 1355 have both wireless local area network ("WLAN") and cellular wireless 1356 links. These links are often used "one at a time" with low-cost WLAN 1357 preferred and highly-available cellular wireless as a standby, but a 1358 simultaneous-use capability could provide benefits. In a more 1359 complex example, aircraft frequently have many wireless data link 1360 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1361 directional, etc.) with diverse performance and cost properties. 1363 If a Client's multiple underlying interfaces are used "one at a time" 1364 (i.e., all other interfaces are in standby mode while one interface 1365 is active), then successive ND messages all include OMNI option 1366 Interface Attributes sub-options with the same underlying interface 1367 index. In that case, the Client would appear to have a single 1368 underlying interface but with a dynamically changing link-layer 1369 address. 1371 If the Client has multiple active underlying interfaces, then from 1372 the perspective of ND it would appear to have multiple link-layer 1373 addresses. In that case, ND message OMNI options MAY include 1374 Interface Attributes sub-options with different underlying interface 1375 indexes. Every ND message need not include Interface Attributes for 1376 all underlying interfaces; for any attributes not included, the 1377 neighbor considers the status as unchanged. 1379 Bridge and Proxy/Server OMNI interfaces are configured over secured 1380 tunnel underlying interfaces for carrying IPv6 ND and BGP protocol 1381 control plane messages, plus open INET underlying interfaces for 1382 carrying unsecured messages. The OMNI interface configures both an 1383 ADM-LLA and its corresponding ADM-ULA, and acts as an OAL source to 1384 encapsulate and fragment original IP packets while presenting the 1385 resulting carrier packets to a secured or unsecured underlying 1386 interface. Note that Bridge and Proxy/Server BGP protocol TCP 1387 sessions are run directly over the OMNI interface using ADM-ULA 1388 source and destination addresses. The OMNI interface encapsulates 1389 the original IP packets for these sessions as carrier packets (i.e., 1390 even though the OAL header may use the same ADM-ULAs as the original 1391 IP header) and forwards them over a secured underlying interface. 1393 3.4. OMNI Interface Initialization 1395 AERO Proxy/Servers and Clients configure OMNI interfaces as their 1396 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1397 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1398 to ensure that original IP packets with destination addresses covered 1399 by an MNP not explicitly assigned to a non-OMNI interface are 1400 directed to the OMNI interface. 1402 OMNI interface initialization procedures for Proxy/Servers, Clients 1403 and Bridges are discussed in the following sections. 1405 3.4.1. AERO Proxy/Server and Relay Behavior 1407 When a Proxy/Server enables an OMNI interface, it assigns an 1408 ADM-{LLA,ULA} appropriate for the given OMNI link segment. The 1409 Proxy/Server also configures secured tunnels with one or more 1410 neighboring Bridges and engages in a BGP routing protocol session 1411 with each Bridge. 1413 The OMNI interface provides a single interface abstraction to the IP 1414 layer, but internally includes an NBMA nexus for sending carrier 1415 packets to OMNI interface neighbors over underlying INET interfaces 1416 and secured tunnels. The Proxy/Server further configures a service 1417 to facilitate ND exchanges with AERO Clients and manages per-Client 1418 neighbor cache entries and IP forwarding table entries based on 1419 control message exchanges. 1421 Relays are simply Proxy/Servers that run a dynamic routing protocol 1422 to redistribute routes between the OMNI interface and INET/EUN 1423 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1424 networks on the INET/EUN interfaces (i.e., the same as a Client would 1425 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1426 interfaces. The Relay further provides an attachment point of the 1427 OMNI link to a non-MNP-based global topology. 1429 3.4.2. AERO Client Behavior 1431 When a Client enables an OMNI interface, it assigns either an 1432 MNP-{LLA, ULA} or a Temporary ULA and sends RS messages with ND 1433 parameters over its underlying interfaces to a Proxy/Server, which 1434 returns an RA message with corresponding parameters. The RS/RA 1435 messages may pass through one or more NATs in the case of a Client's 1436 INET interface. (Note: if the Client used a Temporary ULA in its 1437 initial RS message, it will discover an MNP-{LLA, ULA} in the 1438 corresponding RA that it receives from the Proxy/Server and begin 1439 using these new addresses. If the Client is operating outside the 1440 context of AERO infrastructure such as in a Mobile Ad-hoc Network 1441 (MANET), however, it may continue using Temporary ULAs for Client-to- 1442 Client communications until it encounters an infrastructure element 1443 that can provide an MNP.) 1445 3.4.3. AERO Bridge Behavior 1447 AERO Bridges configure an OMNI interface and assign the ADM-ULA 1448 Subnet Router Anycast address for each OMNI link segment they connect 1449 to. Bridges configure secured tunnels with Proxy/Servers and other 1450 Bridges, and engage in a BGP routing protocol session with neighbors 1451 on the spanning tree (see: Section 3.2.3). 1453 3.5. OMNI Interface Neighbor Cache Maintenance 1455 Each OMNI interface maintains a conceptual neighbor cache that 1456 includes a Neighbor Cache Entry (NCE) for each of its active 1457 neighbors on the OMNI link per [RFC4861]. Each route optimization 1458 source NCE is indexed by the LLA of the neighbor, while the OAL 1459 encapsulation ULA determines the context for Identification 1460 verification. In addition to ordinary neighbor cache entries, proxy 1461 neighbor cache entries are created and maintained by AERO Proxy/ 1462 Servers when they proxy Client ND message exchanges [RFC4389]. AERO 1463 Proxy/Servers maintain proxy neighbor cache entries for each of their 1464 associated Clients. 1466 To the list of NCE states in Section 7.3.2 of [RFC4861], Proxy/Server 1467 OMNI interfaces add an additional state DEPARTED that applies to 1468 Clients that have recently departed. The interface sets a 1469 "DepartTime" variable for the NCE to "DEPART_TIME" seconds. 1470 DepartTime is decremented unless a new ND message causes the state to 1471 return to REACHABLE. While a NCE is in the DEPARTED state, the 1472 Proxy/Server forwards carrier packets destined to the target Client 1473 to the Client's new location instead. When DepartTime decrements to 1474 0, the NCE is deleted. It is RECOMMENDED that DEPART_TIME be set to 1475 the default constant value REACHABLE_TIME plus 10 seconds (40 seconds 1476 by default) to allow a window for carrier packets in flight to be 1477 delivered while stale route optimization state may be present. 1479 Proxy/Servers can act as RORs on behalf of dependent Clients 1480 according to the Proxy Neighbor Advertisement specification in 1481 Section 7.2.8 of [RFC4861]. When a Proxy/Server ROR receives an 1482 authentic NS message used for route optimization, it first searches 1483 for a NCE for the target Client and accepts the message only if there 1484 is an entry. The Proxy/Server then returns a solicited NA message 1485 while creating or updating a "Report List" entry in the target 1486 Client's NCE that caches both the LLA and ULA of ROS with a 1487 "ReportTime" variable set to REPORT_TIME seconds. The ROR resets 1488 ReportTime when it receives a new authentic NS message, and otherwise 1489 decrements ReportTime while no authentic NS messages have been 1490 received. It is RECOMMENDED that REPORT_TIME be set to the default 1491 constant value REACHABLE_TIME plus 10 seconds (40 seconds by default) 1492 to allow a window for route optimization to converge before 1493 ReportTime decrements below REACHABLE_TIME. 1495 When the ROS receives a solicited NA message response to its NS 1496 message used for route optimization, it creates or updates a NCE for 1497 the target network-layer and link-layer addresses. The ROS then 1498 (re)sets ReachableTime for the NCE to REACHABLE_TIME seconds and uses 1499 this value to determine whether carrier packets can be forwarded 1500 directly to the target, i.e., instead of via a default route. The 1501 ROS otherwise decrements ReachableTime while no further solicited NA 1502 messages arrive. It is RECOMMENDED that REACHABLE_TIME be set to the 1503 default constant value 30 seconds as specified in [RFC4861]. 1505 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1506 of NS messages sent when a correspondent may have gone unreachable, 1507 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1508 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1509 to limit the number of unsolicited NAs that can be sent based on a 1510 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1511 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1512 same as specified in [RFC4861]. 1514 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1515 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1516 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1517 different values are chosen, all nodes on the link MUST consistently 1518 configure the same values. Most importantly, DEPART_TIME and 1519 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1520 REACHABLE_TIME to avoid packet loss due to stale route optimization 1521 state. 1523 3.5.1. OMNI ND Messages 1525 OMNI interface IPv6 ND messages include OMNI options 1526 [I-D.templin-6man-omni] with per-neighbor information including 1527 Interface Attributes that provide Link-Layer Address and traffic 1528 selector information for the neighbor's underlying interfaces. This 1529 information is stored in the neighbor cache and provides the basis 1530 for the forwarding algorithm specified in Section 3.10. The 1531 information is cumulative and reflects the union of the OMNI 1532 information from the most recent ND messages received from the 1533 neighbor; it is therefore not required that each ND message contain 1534 all neighbor information. 1536 The OMNI option Interface Attributes for each underlying interface 1537 includes a two-part "Link-Layer Address" consisting of an INET 1538 encapsulation address determined by the FMT and L2ADDR fields and an 1539 ADM-ULA determined by the SRT and LHS fields. Underlying interfaces 1540 are further selected based on their associated traffic selectors. 1542 The OMNI option is distinct from any Source/Target Link-Layer Address 1543 Options (S/TLLAOs) that may appear in an ND message according to the 1544 appropriate IPv6 over specific link layer specification (e.g., 1545 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1546 pertains to encapsulation addresses while the latter pertains to the 1547 native L2 address format of the underlying media 1549 OMNI interface IPv6 ND messages may also include other IPv6 ND 1550 options. In particular, solicitation messages may include Nonce and/ 1551 or Timestamp options if required for verification of advertisement 1552 replies. If an OMNI ND solicitation message includes a Nonce option, 1553 the advertisement reply must echo the same Nonce. If an OMNI ND 1554 solicitation message includes a Timestamp option, the advertisement 1555 reply should also include a Timestamp option. 1557 AERO Clients send RS messages to the All-Routers multicast address 1558 while using unicast link-layer addresses. AERO Proxy/Servers respond 1559 by returning unicast RA messages. During the RS/RA exchange, AERO 1560 Clients and Servers include state synchronization parameters to 1561 establish Identification windows and other state. 1563 AERO nodes use NS/NA messages for the following purposes: 1565 o NS/NA(AR) messages are used for address resolution only. The ROS 1566 sends an NS(AR) to the solicited-node multicast address of the 1567 target, and an ROR in the network with addressing information for 1568 the target returns a unicast NA(AR). The NA(AR) contains 1569 authentic and current target address resolution information, but 1570 only an implicit third-party assertion of target reachability. 1571 NS/NA(AR) messages must be secured. 1573 o NS/NA(WIN) messages are used for establishing and maintaining 1574 window synchronization (and any other) state. The source sends an 1575 NS(WIN) to the unicast address of the target, and the target 1576 returns a unicast NA(WIN). The NS/NA(WIN) exchange synchronizes 1577 sequence numbers the neighbors will include in subsequent packets, 1578 and asserts reachability for the target without necessarily 1579 testing a specific underlying interface pair. NS/NA(WIN) messages 1580 must be secured. 1582 o NS/NA(NUD) messages are used for determining target reachability. 1583 The source sends an NS(NUD) to the unicast address of the target 1584 while naming a specific underlying interface pair, and the target 1585 returns a unicast NA(NUD). NS/NA(NUD) messages that use an in- 1586 window sequence number and do not update any other state need not 1587 be secured. NS/NA(NUD) messages may also be used in combination 1588 with window synchronization (i.e., NUD+WIN), in which case the 1589 messages must be secured. 1591 o Unsolicited NA (uNA) messages are used to signal addressing and/or 1592 other neighbor state changes (e.g., due to mobility, signal 1593 degradation, traffic selector updates, etc.). uNA messages that 1594 include state update information must be secured. 1596 o NS/NA(DAD) messages are not used in AERO, since Duplicate Address 1597 Detection is not required. 1599 Additionally, nodes may send NA/RA messages with the OMNI option PNG 1600 flag set to receive a solicited NA response from the neighbor. The 1601 solicited NA response MUST set the ACK flag (without also setting the 1602 SYN or PNG flags) and include the Identification used in the PNG 1603 message in the Acknowledgement. 1605 3.5.2. OMNI Neighbor Advertisement Message Flags 1607 As discussed in Section 4.4 of [RFC4861] NA messages include three 1608 flag bits R, S and O. OMNI interface NA messages treat the flags as 1609 follows: 1611 o R: The R ("Router") flag is set to 1 in the NA messages sent by 1612 all AERO/OMNI node types. Simple hosts that would set R to 0 do 1613 not occur on the OMNI link itself, but may occur on the downstream 1614 links of Clients and Relays. 1616 o S: The S ("Solicited") flag is set exactly as specified in 1617 Section 4.4. of [RFC4861], i.e., it is set to 1 for Solicited NAs 1618 and set to 0 for uNAs (both unicast and multicast). 1620 o O: The O ("Override") flag is set to 0 for solicited NAs returned 1621 by a Proxy/Server ROR and set to 1 for all other solicited and 1622 unsolicited NAs. For further study is whether solicited NAs for 1623 anycast targets apply for OMNI links. Since MNP-LLAs must be 1624 uniquely assigned to Clients to support correct ND protocol 1625 operation, however, no role is currently seen for assigning the 1626 same MNP-LLA to multiple Clients. 1628 3.5.3. OMNI Neighbor Window Synchronization 1630 In secured environments (e.g., such as between nodes on the same 1631 secured ANET), OMNI interface neighbors can exchange OAL packets 1632 using randomly-initialized and monotonically-increasing 1633 Identification values (modulo 2*32) without window synchronization. 1634 In environments where spoofing is considered a threat, OMNI interface 1635 neighbors instead invoke window synchronization in ND message 1636 exchanges to maintain send/receive window state in their respective 1637 neighbor cache entries as specified in [I-D.templin-6man-omni]. 1639 In the asymmetric window synchronization case, the initial ND message 1640 exchange establishes only the initiator's send window and the 1641 responder's receive window such that a corresponding exchange would 1642 be needed to establish the reverse direction. In the symmetric case, 1643 the initiator and responder engage in a three-way handshake to 1644 symmetrically establish the send/receive windows of both parties. 1646 3.6. OMNI Interface Encapsulation and Re-encapsulation 1648 The OMNI interface admits original IP packets then (acting as an OAL 1649 source) performs OAL encapsulation and fragmentation as specified in 1650 [I-D.templin-6man-omni] while including an ORH if necessary as 1651 specified in Section 3.2.4. OAL encapsulation produces OAL packets 1652 subject to fragmentation, with the resulting fragments encapsulated 1653 in *NET headers as carrier packets. 1655 For carrier packets undergoing re-encapsulation at an OAL 1656 intermediate node, the OMNI interface decrements the OAL IPv6 header 1657 Hop Limit and discards the carrier packet if the Hop Limit reaches 0. 1658 The intermediate node next removes the *NET encapsulation headers 1659 from the first segment and re-encapsulates the packet in new *NET 1660 encapsulation headers for the next segment. 1662 When a Proxy/Server or Relay re-encapsulates a carrier packet 1663 received from a Client that includes an OAL but no ORH, it inserts an 1664 ORH immediately following the OAL header and adjusts the OAL payload 1665 length and destination address field. The ORH will be removed by the 1666 LHS Bridge or Proxy/Server, but its insertion and removal will not 1667 interfere with reassembly at the final destination. For this reason, 1668 Clients must reserve 40 bytes for a maximum-length ORH when they 1669 perform OAL encapsulation (see: Section 3.9). 1671 3.7. OMNI Interface Decapsulation 1673 OMNI interfaces (acting as OAL destinations) decapsulate and 1674 reassemble OAL packets into original IP packets destined either to 1675 the AERO node itself or to a destination reached via an interface 1676 other than the OMNI interface the original IP packet was received on. 1677 When carrier packets containing OAL fragments addressed to itself 1678 arrive, the OMNI interface discards the NET encapsulation headers and 1679 reassembles as discussed in Section 3.9. 1681 3.8. OMNI Interface Data Origin Authentication 1683 AERO nodes employ simple data origin authentication procedures. In 1684 particular: 1686 o AERO Bridges and Proxy/Servers accept carrier packets received 1687 from secured underlying interfaces. 1689 o AERO Proxy/Servers and Clients accept carrier packets and original 1690 IP packets that originate from within the same secured ANET. 1692 o AERO Clients and Relays accept original IP packets from downstream 1693 network correspondents based on ingress filtering. 1695 o AERO Clients, Relays and Proxy/Servers verify carrier packet UDP/ 1696 IP encapsulation addresses according to [I-D.templin-6man-omni]. 1698 o AERO nodes accept carrier packets addressed to themselves with 1699 Identification values within the current window for the OAL source 1700 neighbor (when window synchronization is used) and drop any 1701 carrier packets with out-of-window Identification values. (AERO 1702 nodes may forward carrier packets not addressed to themselves 1703 without verifying the Identification value.) 1705 AERO nodes silently drop any packets that do not satisfy the above 1706 data origin authentication procedures. Further security 1707 considerations are discussed in Section 6. 1709 3.9. OMNI Interface MTU 1711 The OMNI interface observes the link nature of tunnels, including the 1712 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1713 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 1714 The OMNI interface employs an OMNI Adaptation Layer (OAL) that 1715 accommodates multiple underlying links with diverse MTUs while 1716 observing both a minimum and per-path Maximum Payload Size (MPS). 1717 The functions of the OAL and the OMNI interface MTU/MRU/MPS are 1718 specified in [I-D.templin-6man-omni] with MTU/MRU both set to the 1719 constant value 9180 bytes, with minimum MPS set to 400 bytes, and 1720 with per-path MPS set to potentially larger values depending on the 1721 underlying path. 1723 When the network layer presents an original IP packet to the OMNI 1724 interface, the OAL source encapsulates and fragments the original IP 1725 packet if necessary. When the network layer presents the OMNI 1726 interface with multiple original IP packets bound to the same OAL 1727 destination, the OAL source can concatenate them together into a 1728 single OAL super-packet as discussed in [I-D.templin-6man-omni]. The 1729 OAL source then fragments the OAL packet if necessary according to 1730 the minimum/path MPS such that the OAL headers appear in each 1731 fragment while the original IP packet header appears only in the 1732 first fragment. The OAL source then encapsulates each OAL fragment 1733 in *NET headers for transmission as carrier packets over an 1734 underlying interface connected to either a physical link such as 1735 Ethernet, WiFi and the like or a virtual link such as an Internet or 1736 higher-layer tunnel (see the definition of link in [RFC8200]). 1738 Note: A Client that does not (yet) have neighbor cache state for a 1739 target may omit the ORH in carrier packets with the understanding 1740 that a Proxy/Server may insert an ORH on its behalf. For this 1741 reason, Clients reserve 40 bytes for the largest possible ORH in 1742 their OAL fragment size calculations. 1744 Note: Although the ORH may be removed or replaced by a Bridge or 1745 Proxy/Server on the path (see: Section 3.10.3), this does not 1746 interfere with the destination's ability to reassemble since the ORH 1747 is not included in the fragmentable part and its removal/ 1748 transformation does not invalidate fragment header information. 1750 3.10. OMNI Interface Forwarding Algorithm 1752 Original IP packets enter a node's OMNI interface either from the 1753 network layer (i.e., from a local application or the IP forwarding 1754 system) while carrier packets enter from the link layer (i.e., from 1755 an OMNI interface neighbor). All original IP packets and carrier 1756 packets entering a node's OMNI interface first undergo data origin 1757 authentication as discussed in Section 3.8. Those that satisfy data 1758 origin authentication are processed further, while all others are 1759 dropped silently. 1761 Original IP packets that enter the OMNI interface from the network 1762 layer are forwarded to an OMNI interface neighbor using OAL 1763 encapsulation and fragmentation to produce carrier packets for 1764 transmission over underlying interfaces. (If routing indicates that 1765 the original IP packet should instead be forwarded back to the 1766 network layer, the packet is dropped to avoid looping). Carrier 1767 packets that enter the OMNI interface from the link layer are either 1768 re-encapsulated and re-admitted into the OMNI link, or reassembled 1769 and forwarded to the network layer where they are subject to either 1770 local delivery or IP forwarding. In all cases, the OAL MUST NOT 1771 decrement the network layer TTL/Hop-count since its forwarding 1772 actions occur below the network layer. 1774 OMNI interfaces may have multiple underlying interfaces and/or 1775 neighbor cache entries for neighbors with multiple underlying 1776 interfaces (see Section 3.3). The OAL uses Interface Attribute 1777 traffic selectors (e.g., port number, flow specification, etc.) to 1778 select an outbound underlying interface for each OAL packet based on 1779 the node's own interface attributes, and also to select a destination 1780 link-layer address based on the neighbor's underlying interface 1781 attributes. AERO implementations SHOULD permit network management to 1782 dynamically adjust traffic selector values at runtime. 1784 If an OAL packet matches the traffic selectors of multiple outgoing 1785 interfaces and/or neighbor interfaces, the OMNI interface replicates 1786 the packet and sends one copy via each of the (outgoing / neighbor) 1787 interface pairs; otherwise, it sends a single copy of the OAL packet 1788 via an interface with the best matching traffic selector. (While not 1789 strictly required, the likelihood of successful reassembly may 1790 improve when the OMNI interface sends all fragments of the same 1791 fragmented OAL packet consecutively over the same underlying 1792 interface pair instead of spread across multiple underlying interface 1793 pairs.) AERO nodes keep track of which underlying interfaces are 1794 currently "reachable" or "unreachable", and only use "reachable" 1795 interfaces for forwarding purposes. 1797 The following sections discuss the OMNI interface forwarding 1798 algorithms for Clients, Proxy/Servers and Bridges. In the following 1799 discussion, an original IP packet's destination address is said to 1800 "match" if it is the same as a cached address, or if it is covered by 1801 a cached prefix (which may be encoded in an MNP-LLA). 1803 3.10.1. Client Forwarding Algorithm 1805 When an original IP packet enters a Client's OMNI interface from the 1806 network layer the Client searches for a NCE that matches the 1807 destination. If there is a match, the Client selects one or more 1808 "reachable" neighbor interfaces in the entry for forwarding purposes. 1809 If there is no NCE, the Client instead either enqueues the original 1810 IP packet and invokes route optimization or forwards the original IP 1811 packet toward a Proxy/Server. The Client (acting as an OAL source) 1812 performs OAL encapsulation and sets the OAL destination address to 1813 the MNP-ULA of the target if there is a matching NCE; otherwise, it 1814 sets the OAL destination to the ADM-ULA of the Proxy/Server. If the 1815 Client has multiple original IP packets to send to the same neighbor, 1816 it can concatenate them in a single super-packet 1817 [I-D.templin-6man-omni]. The OAL source then performs fragmentation 1818 to create OAL fragments (see: Section 3.9), appends any *NET 1819 encapsulation, and sends the resulting carrier packets over 1820 underlying interfaces to the neighbor acting as an OAL destination. 1822 If the neighbor interface selected for forwarding is located on the 1823 same OMNI link segment and not behind a NAT, the Client forwards the 1824 carrier packets directly according to the L2ADDR information for the 1825 neighbor. If the neighbor interface is behind a NAT on the same OMNI 1826 link segment, the Client instead forwards the initial carrier packets 1827 to the LHS Proxy/Server (while inserting an ORH-0 if necessary) and 1828 initiates NAT traversal procedures. If the Client's intended source 1829 underlying interface is also behind a NAT and located on the same 1830 OMNI link segment, it sends a "direct bubble" over the interface per 1831 [RFC6081][RFC4380] to the L2ADDR found in the neighbor cache in order 1832 to establish state in its own NAT by generating traffic toward the 1833 neighbor (note that no response to the bubble is expected). 1835 The Client next sends an NS(NUD) message toward the MNP-ULA of the 1836 neighbor via the LHS Proxy/Server as discussed in Section 3.15. If 1837 the Client receives an NA(NUD) from the neighbor over the underlying 1838 interface, it marks the neighbor interface as "trusted" and sends 1839 future carrier packets directly to the L2ADDR information for the 1840 neighbor instead of indirectly via the LHS Proxy/Server. The Client 1841 must honor the neighbor cache maintenance procedure by sending 1842 additional direct bubbles and/or NS/NA(NUD) messages as discussed in 1843 [RFC6081][RFC4380] in order to keep NAT state alive as long as 1844 carrier packets are still flowing. 1846 When a carrier packet enters a Client's OMNI interface from the link- 1847 layer, if the OAL destination matches one of the Client's ULAs the 1848 Client (acting as an OAL destination) verifies that the 1849 Identification is in-window for this OAL source, then reassembles and 1850 decapsulates as necessary and delivers the original IP packet to the 1851 network layer. Otherwise, the Client drops the original IP packet 1852 and MAY return a network-layer ICMP Destination Unreachable message 1853 subject to rate limiting (see: Section 3.11). 1855 Note: Clients and their Proxy/Server (and other Client) peers can 1856 exchange original IP packets over ANET underlying interfaces without 1857 invoking the OAL, since the ANET is secured at the link and physical 1858 layers. By forwarding original IP packets without invoking the OAL, 1859 however, the ANET peers can engage only in classical path MTU 1860 discovery since the packets are subject to loss and/or corruption due 1861 to the various per-link MTU limitations that may occur within the 1862 ANET. Moreover, the original IP packets do not include either the 1863 OAL integrity check or per-packet Identification values that can be 1864 used for data origin authentication and link-layer retransmissions. 1865 The tradeoff therefore involves an assessment of the per-packet 1866 encapsulation overhead saved by bypassing the OAL vs. inheritance of 1867 classical network "brittleness". (Note however that ANET peers can 1868 send small original IP packets without invoking the OAL, while 1869 invoking the OAL for larger packets. This presents the beneficial 1870 aspects of both small packet efficiency and large packet robustness.) 1872 3.10.2. Proxy/Server and Relay Forwarding Algorithm 1874 When the Proxy/Server receives an original IP packet from the network 1875 layer, it drops the packet if routing indicates that it should be 1876 forwarded back to the network layer to avoid looping. Otherwise, the 1877 Proxy/Server regards the original IP packet the same as if it had 1878 arrived as carrier packets with OAL destination set to its own ADM- 1879 ULA. When the Proxy/Server receives carrier packets on underlying 1880 interfaces with OAL destination set to its own ADM-ULA, it performs 1881 OAL reassembly if necessary to obtain the original IP packet. 1883 The Proxy/Server next searches for a NCE that matches the original IP 1884 destination and proceeds as follows: 1886 o if the original IP packet destination matches a NCE, the Proxy/ 1887 Sever uses one or more "reachable" neighbor interfaces in the 1888 entry for packet forwarding using OAL encapsulation and 1889 fragmentation according to the cached link-layer address 1890 information. If the neighbor interface is in a different OMNI 1891 link segment, the Proxy/Server performs OAL encapsulation and 1892 fragmentation, inserts an ORH and forwards the resulting carrier 1893 packets via the spanning tree to a Bridge; otherwise, it forwards 1894 the carrier packets directly to the neighbor. If the neighbor is 1895 behind a NAT, the Proxy/Server instead forwards initial carrier 1896 packets via a Bridge while sending an NS(NUD) to the neighbor. 1897 When the Proxy/Server receives the NA(NUD), it can begin 1898 forwarding carrier packets directly to the neighbor the same as 1899 discussed in Section 3.10.1 while sending additional NS(NUD) 1900 messages as necessary to maintain NAT state. Note that no direct 1901 bubbles are necessary since the Proxy/Server is by definition not 1902 located behind a NAT. 1904 o else, if the original IP destination matches a non-MNP route in 1905 the IP forwarding table or an ADM-LLA assigned to the Proxy/ 1906 Server's OMNI interface, the Proxy/Server acting as a Relay 1907 presents the original IP packet to the network layer for local 1908 delivery or IP forwarding. 1910 o else, the Proxy/Server initiates address resolution as discussed 1911 in Section 3.14, while retaining initial original IP packets in a 1912 small queue awaiting address resolution completion. 1914 When the Proxy/Server receives a carrier packet with OAL destination 1915 set to an MNP-ULA that does not match the MSP, it accepts the carrier 1916 packet only if data origin authentication succeeds and if there is a 1917 network layer routing table entry for a GUA route that matches the 1918 MNP-ULA. If there is no route, the Proxy/Server drops the carrier 1919 packet; otherwise, it reassembles and decapsulates to obtain the 1920 original IP packet and acts as a Relay to present it to the network 1921 layer where it will be delivered according to standard IP forwarding. 1923 When the Proxy/Server receives a carrier packet from one of its 1924 Client neighbors with OAL destination set to another node, it 1925 forwards the packets via a matching NCE or via the spanning tree if 1926 there is no matching entry. When the Proxy/Server receives a carrier 1927 packet with OAL destination set to the MNP-ULA of one of its Client 1928 neighbors established through RS/RA exchanges, it accepts the carrier 1929 packet only if data origin authentication succeeds. If the NCE state 1930 is DEPARTED, the Proxy/Server inserts an ORH that encodes the MNP-ULA 1931 destination suffix and changes the OAL destination address to the 1932 ADM-ULA of the new Proxy/Server, then re-encapsulates the carrier 1933 packet and forwards it to a Bridge which will eventually deliver it 1934 to the new Proxy/Server. 1936 If the neighbor cache state for the MNP-ULA is REACHABLE, the Proxy/ 1937 Server forwards the carrier packets to the Client which then must 1938 reassemble. (Note that the Proxy/Server does not reassemble carrier 1939 packets not explicitly addressed to its own ADM-ULA, since routing 1940 could direct some of the carrier packet of the same original IP 1941 packet through a different Proxy/Server.) In that case, the Client 1942 may receive fragments that are smaller than its link MTU but can 1943 still be reassembled. 1945 Note: Proxy/Servers may receive carrier packets with ORHs that 1946 include additional forwarding information. Proxy/Servers use the 1947 forwarding information to determine the correct interface for 1948 forwarding to the target destination Client, then remove the ORH and 1949 forward the carrier packet. If the ORH information instead indicates 1950 that the Proxy/Server is responsible for reassembly, the Proxy/Server 1951 reassembles first before re-encapsulating (and possibly also re- 1952 fragmenting) then forwards to the target Client. For a full 1953 discussion of cases when the Proxy/Server may receive carrier packets 1954 with ORHs, see: Section 3.14.6. 1956 Note: Clients and their Proxy/Server peers can exchange original IP 1957 packets over ANET underlying interfaces without invoking the OAL, 1958 since the ANET is secured at the link and physical layers. By 1959 forwarding original IP packets without invoking the OAL, however, the 1960 Client and Proxy/Server can engage only in classical path MTU 1961 discovery since the packets are subject to loss and/or corruption due 1962 to the various per-link MTU limitations that may occur within the 1963 ANET. Moreover, the original IP packets do not include either the 1964 OAL integrity check or per-packet Identification values that can be 1965 used for data origin authentication and link-layer retransmissions. 1966 The tradeoff therefore involves an assessment of the per-packet 1967 encapsulation overhead saved by bypassing the OAL vs. inheritance of 1968 classical network "brittleness". (Note however that ANET peers can 1969 send small original IP packets without invoking the OAL, while 1970 invoking the OAL for larger packets. This presents the beneficial 1971 aspects of both small packet efficiency and large packet robustness.) 1973 Note: When a Proxy/Server receives a (non-OAL) original IP packet 1974 from an ANET Client, or a carrier packet with OAL destination set to 1975 its own ADM-ULA from any Client, the Proxy/Server reassembles if 1976 necessary then performs ROS functions on behalf of the Client. The 1977 Client may at some later time begin sending carrier packets to the 1978 OAL address of the actual target instead of the Proxy/Server, at 1979 which point it may begin functioning as an ROS on its own behalf and 1980 thereby "override" the Proxy/Server's ROS role. 1982 Note; When a Proxy/Server receives an original IP packet (either 1983 directly from an ANET Client or following reassembly of carrier 1984 packets received from an ANET/INET Client), it drops the packet if 1985 the destination is covered by the Client's delegated MNP. This is 1986 necessary to prevent Clients from either accidentally or 1987 intentionally establishing an endless loop that could congest ANET/ 1988 INET interfaces. 1990 Note: Proxy/Servers forward secure control plane carrier packets via 1991 the secured spanning tree and forwards other carrier packets via the 1992 unsecured spanning tree. When a Proxy/Server receives a carrier 1993 packet from the secured spanning tree, it considers the message as 1994 authentic without having to verify upper layer authentication 1995 signatures. When a Proxy/Server receives a carrier packet from the 1996 unsecured spanning tree, it verifies any upper layer authentication 1997 signatures and/or forwards the unsecured message toward the 1998 destination which must apply data origin authentication. 2000 Note: If the Proxy/Server has multiple original IP packets to send to 2001 the same neighbor, it can concatenate them in a single OAL super- 2002 packet [I-D.templin-6man-omni]. 2004 3.10.3. Bridge Forwarding Algorithm 2006 Bridges forward carrier packets the same as any IPv6 router. Bridges 2007 convey carrier packets that encapsulate IPv6 ND control messages or 2008 routing protocol control messages via the secured spanning tree, and 2009 may convey carrier packets that encapsulate ordinary data via the 2010 unsecured spanning tree. When the Bridge receives a carrier packet, 2011 it removes the outer *NET header and searches for a forwarding table 2012 entry that matches the OAL destination address. The Bridge then 2013 processes the packet as follows: 2015 o if the carrier packet destination matches its ADM-ULA or the 2016 corresponding ADM-ULA Subnet Router Anycast address and the OAL 2017 header is followed by an ORH, the Bridge reassembles if necessary 2018 then sets aside the ORH and processes the carrier packet locally 2019 before forwarding. If the OAL packet contains an NA(NUD) message, 2020 the Bridge replaces the OMNI option Interface Attributes sub- 2021 option with information for its own interface while retaining the 2022 omIndex value supplied by the NA(NUD) message source. The Bridge 2023 next examines the ORH, and if FMT-Mode indicates the destination 2024 is a Client on the open *NET (or, a Client behind a NAT for which 2025 NAT traversal procedures have already converged) the Bridge writes 2026 the MNP-ULA formed from the ORH Destination Suffix into the OAL 2027 destination. The Bridge then removes the ORH and forwards the 2028 packet using encapsulation based on L2ADDR. If the LHS Proxy/ 2029 Server will forward to the Client without reassembly, the Bridge 2030 writes the MNP-ULA into the OAL destination then replaces the ORH 2031 with an ORH-0 and forwards the carrier packet to the LHS Proxy/ 2032 Server while also invoking NAT traversal procedures if necessary 2033 (noting that no direct bubbles are necessary since only the target 2034 Client and not the Bridge is behind a NAT). If the LHS Proxy/ 2035 Server must perform reassembly before forwarding to the Client, 2036 the Bridge instead writes the ADM-ULA formed from the ORH SRT/LHS 2037 into the OAL destination address, replaces the ORH with an ORH-0 2038 and forwards the carrier packet to the LHS Proxy/Server. 2040 o else, if the carrier packet destination matches its ADM-ULA or the 2041 corresponding ADM-ULA Subnet Router Anycast address and the OAL 2042 header is not followed by an ORH with Segments Left set to 1, the 2043 Bridge submits the packet for reassembly. When reassembly is 2044 complete, the Bridge submits the original packet to the IP layer 2045 to support local applications such as BGP routing protocol 2046 sessions. 2048 o else, if the carrier packet destination matches a forwarding table 2049 entry the Bridge forwards the carrier packet to the next hop. (If 2050 the destination matches an MSP without matching an MNP, however, 2051 the Bridge instead drops the packet and returns an ICMP 2052 Destination Unreachable message subject to rate limiting - see: 2053 Section 3.11). 2055 o else, the Bridge drops the packet and returns an ICMP Destination 2056 Unreachable as above. 2058 The Bridge decrements the OAL IPv6 header Hop Limit when it forwards 2059 the carrier packet (i.e., the same as for any IPv6 router) and drops 2060 the packet if the Hop Limit reaches 0. Therefore, only the Hop Limit 2061 in the OAL header is decremented and not the TTL/Hop Limit in the 2062 original IP packet header. Bridges do not insert OAL/ORH headers 2063 themselves; instead, they act as IPv6 routers and forward carrier 2064 packets based on their destination addresses while also possibly 2065 transforming larger ORHs into an ORH-0 (or removing the ORH 2066 altogether). 2068 Bridges forward carrier packets received from a first segment via the 2069 secured spanning tree to the next segment also via the secured 2070 spanning tree. Bridges forward carrier packets received from a first 2071 segment via the unsecured spanning tree to the next segment also via 2072 the unsecured spanning tree. Bridges use a single IPv6 routing table 2073 that always determines the same next hop for a given OAL destination, 2074 where the secured/unsecured spanning tree is determined through the 2075 selection of the underlying interface to be used for transmission 2076 (i.e., a secured tunnel or an open INET interface). 2078 3.11. OMNI Interface Error Handling 2080 When an AERO node admits an original IP packet into the OMNI 2081 interface, it may receive link-layer or network-layer error 2082 indications. The AERO node may also receive OMNI link error 2083 indications in OAL-encapsulated uNA messages that include 2084 authentication signatures. 2086 A link-layer error indication is an ICMP error message generated by a 2087 router in the INET on the path to the neighbor or by the neighbor 2088 itself. The message includes an IP header with the address of the 2089 node that generated the error as the source address and with the 2090 link-layer address of the AERO node as the destination address. 2092 The IP header is followed by an ICMP header that includes an error 2093 Type, Code and Checksum. Valid type values include "Destination 2094 Unreachable", "Time Exceeded" and "Parameter Problem" 2095 [RFC0792][RFC4443]. (OMNI interfaces ignore link-layer IPv4 2096 "Fragmentation Needed" and IPv6 "Packet Too Big" messages for carrier 2097 packets that are no larger than the minimum/path MPS as discussed in 2098 Section 3.9, however these messages may provide useful hints of probe 2099 failures during path MPS probing.) 2101 The ICMP header is followed by the leading portion of the carrier 2102 packet that generated the error, also known as the "packet-in-error". 2103 For ICMPv6, [RFC4443] specifies that the packet-in-error includes: 2104 "As much of invoking packet as possible without the ICMPv6 packet 2105 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 2106 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 2107 "Internet Header + 64 bits of Original Data Datagram", however 2108 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 2109 ICMP datagram SHOULD contain as much of the original datagram as 2110 possible without the length of the ICMP datagram exceeding 576 2111 bytes". 2113 The link-layer error message format is shown in Figure 5 (where, "L2" 2114 and "L3" refer to link-layer and network-layer, respectively): 2116 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2117 ~ ~ 2118 | L2 IP Header of | 2119 | error message | 2120 ~ ~ 2121 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2122 | L2 ICMP Header | 2123 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 2124 ~ ~ P 2125 | carrier packet *NET and OAL | a 2126 | encapsulation headers | c 2127 ~ ~ k 2128 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 2129 ~ ~ t 2130 | original IP packet headers | 2131 | (first-fragment only) | i 2132 ~ ~ n 2133 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2134 ~ ~ e 2135 | Portion of the body of | r 2136 | the original IP packet | r 2137 | (all fragments) | o 2138 ~ ~ r 2139 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 2141 Figure 5: OMNI Interface Link-Layer Error Message Format 2143 The AERO node rules for processing these link-layer error messages 2144 are as follows: 2146 o When an AERO node receives a link-layer Parameter Problem message, 2147 it processes the message the same as described as for ordinary 2148 ICMP errors in the normative references [RFC0792][RFC4443]. 2150 o When an AERO node receives persistent link-layer Time Exceeded 2151 messages, the IP ID field may be wrapping before earlier fragments 2152 awaiting reassembly have been processed. In that case, the node 2153 should begin including integrity checks and/or institute rate 2154 limits for subsequent packets. 2156 o When an AERO node receives persistent link-layer Destination 2157 Unreachable messages in response to carrier packets that it sends 2158 to one of its neighbor correspondents, the node should process the 2159 message as an indication that a path may be failing, and 2160 optionally initiate NUD over that path. If it receives 2161 Destination Unreachable messages over multiple paths, the node 2162 should allow future carrier packets destined to the correspondent 2163 to flow through a default route and re-initiate route 2164 optimization. 2166 o When an AERO Client receives persistent link-layer Destination 2167 Unreachable messages in response to carrier packets that it sends 2168 to one of its neighbor Proxy/Servers, the Client should mark the 2169 path as unusable and use another path. If it receives Destination 2170 Unreachable messages on many or all paths, the Client should 2171 associate with a new Proxy/Server and release its association with 2172 the old Proxy/Server as specified in Section 3.16.5. 2174 o When an AERO Proxy/Server receives persistent link-layer 2175 Destination Unreachable messages in response to carrier packets 2176 that it sends to one of its neighbor Clients, the Proxy/Server 2177 should mark the underlying path as unusable and use another 2178 underlying path. 2180 o When an AERO Proxy/Server receives link-layer Destination 2181 Unreachable messages in response to a carrier packet that it sends 2182 to one of its permanent neighbors, it treats the messages as an 2183 indication that the path to the neighbor may be failing. However, 2184 the dynamic routing protocol should soon reconverge and correct 2185 the temporary outage. 2187 When an AERO Bridge receives a carrier packet for which the network- 2188 layer destination address is covered by an MSP, the Bridge drops the 2189 packet if there is no more-specific routing information for the 2190 destination and returns an OMNI interface Destination Unreachable 2191 message subject to rate limiting. 2193 When an AERO node receives a carrier packet for which reassembly is 2194 currently congested, it returns an OMNI interface Packet Too Big 2195 (PTB) message as discussed in [I-D.templin-6man-omni] (note that the 2196 PTB messages could indicate either "hard" or "soft" errors). 2198 AERO nodes include ICMPv6 error messages intended for the OAL source 2199 as sub-options in the OMNI option of secured uNA messages. When the 2200 OAL source receives the uNA message, it can extract the ICMPv6 error 2201 message enclosed in the OMNI option and either process it locally or 2202 translate it into a network-layer error to return to the original 2203 source. 2205 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 2207 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 2208 coordinated as discussed in the following Sections. 2210 3.12.1. AERO Service Model 2212 Each AERO Proxy/Server on the OMNI link is configured to facilitate 2213 Client prefix delegation/registration requests. Each Proxy/Server is 2214 provisioned with a database of MNP-to-Client ID mappings for all 2215 Clients enrolled in the AERO service, as well as any information 2216 necessary to authenticate each Client. The Client database is 2217 maintained by a central administrative authority for the OMNI link 2218 and securely distributed to all Proxy/Servers, e.g., via the 2219 Lightweight Directory Access Protocol (LDAP) [RFC4511], via static 2220 configuration, etc. Clients receive the same service regardless of 2221 the Proxy/Servers they select. 2223 AERO Clients and Proxy/Servers use ND messages to maintain neighbor 2224 cache entries. AERO Proxy/Servers configure their OMNI interfaces as 2225 advertising NBMA interfaces, and therefore send unicast RA messages 2226 with a short Router Lifetime value (e.g., ReachableTime seconds) in 2227 response to a Client's RS message. Thereafter, Clients send 2228 additional RS messages to keep Proxy/Server state alive. 2230 AERO Clients and Proxy/Servers include prefix delegation and/or 2231 registration parameters in RS/RA messages (see 2232 [I-D.templin-6man-omni]). The ND messages are exchanged between 2233 Client and Proxy/Server according to the prefix management schedule 2234 required by the service. If the Client knows its MNP in advance, it 2235 can employ prefix registration by including its MNP-LLA as the source 2236 address of an RS message and with an OMNI option with valid prefix 2237 registration information for the MNP. If the Proxy/Server accepts 2238 the Client's MNP assertion, it injects the MNP into the routing 2239 system and establishes the necessary neighbor cache state. If the 2240 Client does not have a pre-assigned MNP, it can instead employ prefix 2241 delegation by including the unspecified address (::) as the source 2242 address of an RS message and with an OMNI option with prefix 2243 delegation parameters to request an MNP. 2245 The following sections specify the Client and Proxy/Server behavior. 2247 3.12.2. AERO Client Behavior 2249 AERO Clients discover the addresses of Proxy/Servers in a similar 2250 manner as described in [RFC5214]. Discovery methods include static 2251 configuration (e.g., from a flat-file map of Proxy/Server addresses 2252 and locations), or through an automated means such as Domain Name 2253 System (DNS) name resolution [RFC1035]. Alternatively, the Client 2254 can discover Proxy/Server addresses through a layer 2 data link login 2255 exchange, or through a unicast RA response to a multicast/anycast RS 2256 as described below. In the absence of other information, the Client 2257 can resolve the DNS Fully-Qualified Domain Name (FQDN) 2258 "linkupnetworks.[domainname]" where "linkupnetworks" is a constant 2259 text string and "[domainname]" is a DNS suffix for the OMNI link 2260 (e.g., "example.com"). 2262 To associate with a Proxy/Server, the Client acts as a requesting 2263 router to request MNPs by preparing an RS message with prefix 2264 management parameters. If the Client already knows the Proxy/ 2265 Server's ADM-LLA, it includes the LLA as the network-layer 2266 destination address; otherwise, the Client includes the (link-local) 2267 All-Routers multicast as the network-layer destination. If the 2268 Client already knows its own MNP-LLA, it can use the MNP-LLA as the 2269 network-layer source address and include an OMNI option with prefix 2270 registration information. Otherwise, the Client uses the unspecified 2271 address (::) as the network-layer source address and includes prefix 2272 delegation parameters in the OMNI option (see: 2273 [I-D.templin-6man-omni]). 2275 The Client next includes Interface Attributes corresponding to the 2276 underlying interface over which it will send the RS message, and MAY 2277 include additional Interface Attributes specific to other underlying 2278 interfaces. Next, the Client submits the RS for OAL encapsulation 2279 and fragmentation if necessary with its own MNP-ULA and the Proxy/ 2280 Server's ADM-ULA or (site-scoped) All-Routers multicast as the OAL 2281 addresses while selecting an Identification value and invoking window 2282 synchronization as specified in [I-D.templin-6man-omni]. 2284 The Client then sends the RS (either directly via Direct interfaces, 2285 via a VPN for VPNed interfaces, via an access router for ANET 2286 interfaces or via INET encapsulation for INET interfaces) then waits 2287 up to RetransTimer milliseconds for an RA message reply (see 2288 Section 3.12.3) (retrying up to MAX_RTR_SOLICITATIONS). If the 2289 Client receives no RAs, or if it receives an RA with Router Lifetime 2290 set to 0, the Client SHOULD abandon attempts through the first Proxy/ 2291 Server and try another Proxy/Server. Otherwise, the Client processes 2292 the prefix information found in the RA message. 2294 When the Client processes an RA, it first performs OAL reassembly and 2295 decapsulation if necessary then creates a NCE with the Proxy/Server's 2296 ADM-LLA as the network-layer address and the Proxy/Server's 2297 encapsulation and/or link-layer addresses as the link-layer address. 2298 The Client next records the RA Router Lifetime field value in the NCE 2299 as the time for which the Proxy/Server has committed to maintaining 2300 the MNP in the routing system via this underlying interface, and 2301 caches the other RA configuration information including Cur Hop 2302 Limit, M and O flags, Reachable Time and Retrans Timer. The Client 2303 then autoconfigures MNP-LLAs for any delegated MNPs and assigns them 2304 to the OMNI interface. The Client also caches any MSPs included in 2305 Route Information Options (RIOs) [RFC4191] as MSPs to associate with 2306 the OMNI link, and assigns the MTU value in the MTU option to the 2307 underlying interface. 2309 The Client then registers its additional underlying interfaces with 2310 each separate Proxy/Server it discovers by sending RS messages via 2311 each additional interface as described above. The RS messages 2312 include the same parameters as for the initial RS/RA exchange, but 2313 with destination address set to the Proxy/Server's ADM-LLA. The 2314 Client finally sub-delegates the MNPs to its attached EUNs and/or the 2315 Client's own internal virtual interfaces as described in 2316 [I-D.templin-v6ops-pdhost] to support the Client's downstream 2317 attached "Internet of Things (IoT)". The Client then sends 2318 additional RS messages over each underlying interface before the 2319 Router Lifetime received for that interface expires. 2321 After the Client registers its underlying interfaces, it may wish to 2322 change one or more registrations, e.g., if an interface changes 2323 address or becomes unavailable, if traffic selectors change, etc. To 2324 do so, the Client prepares an RS message to send over any available 2325 underlying interface as above. The RS includes an OMNI option with 2326 prefix registration/delegation information, with Interface Attributes 2327 specific to the selected underlying interface, and with any 2328 additional Interface Attributes specific to other underlying 2329 interfaces. When the Client receives the Proxy/Server's RA response, 2330 it has assurance that the Proxy/Server has been updated with the new 2331 information. 2333 If the Client wishes to discontinue use of a Proxy/Server it issues 2334 an RS message over any underlying interface with an OMNI option with 2335 a prefix release indication. When the Proxy/Server processes the 2336 message, it releases the MNP, sets the NCE state for the Client to 2337 DEPARTED and returns an RA reply with Router Lifetime set to 0. 2338 After a short delay (e.g., 2 seconds), the Proxy/Server withdraws the 2339 MNP from the routing system. 2341 3.12.3. AERO Proxy/Server Behavior 2343 AERO Proxy/Servers act as IP routers and support a prefix delegation/ 2344 registration service for Clients. Proxy/Servers arrange to add their 2345 ADM-LLAs to a static map of Proxy/Server addresses for the link and/ 2346 or the DNS resource records for the FQDN 2347 "linkupnetworks.[domainname]" before entering service. Proxy/Server 2348 addresses should be geographically and/or topologically referenced, 2349 and made available for discovery by Clients on the OMNI link. 2351 When a Proxy/Server receives a prospective Client's RS message on its 2352 OMNI interface, it SHOULD return an immediate RA reply with Router 2353 Lifetime set to 0 if it is currently too busy or otherwise unable to 2354 service the Client. Otherwise, the Proxy/Server performs OAL 2355 reassembly and decapsulation if necessary, then authenticates the RS 2356 message and processes the prefix delegation/registration parameters. 2357 The Proxy/Server first determines the correct MNPs to provide to the 2358 Client by processing the MNP-LLA prefix parameters and/or the DHCPv6 2359 OMNI sub-option. When the Proxy/Server returns the MNPs, it also 2360 creates a forwarding table entry for the MNP-ULA corresponding to 2361 each MNP so that the MNPs are propagated into the routing system 2362 (see: Section 3.2.3). For IPv6, the Proxy/Server creates an IPv6 2363 forwarding table entry for each MNP. For IPv4, the Proxy/Server 2364 creates an IPv6 forwarding table entry with the IPv4-compatibility 2365 MNP-ULA prefix corresponding to the IPv4 address. 2367 The Proxy/Server next creates a NCE for the Client using the base 2368 MNP-LLA as the network-layer address. Next, the Proxy/Server updates 2369 the NCE by recording the information in each Interface Attributes 2370 sub-option in the RS OMNI option. The Proxy/Server also records the 2371 actual OAL/*NET addresses and RS message window synchronization 2372 parameters (if any) in the NCE. 2374 Next, the Proxy/Server prepares an RA message using its ADM-LLA as 2375 the network-layer source address and the network-layer source address 2376 of the RS message as the network-layer destination address. The 2377 Proxy/Server sets the Router Lifetime to the time for which it will 2378 maintain both this underlying interface individually and the NCE as a 2379 whole. The Proxy/Server also sets Cur Hop Limit, M and O flags, 2380 Reachable Time and Retrans Timer to values appropriate for the OMNI 2381 link. The Proxy/Server includes the MNPs, any other prefix 2382 management parameters and an OMNI option with no Interface Attributes 2383 but with an Origin Indication sub-option per [I-D.templin-6man-omni] 2384 with the mapped and obfuscated Port Number and IP address 2385 corresponding to the Client's own INET address in the case of INET 2386 Clients or to the Proxy/Server's INET-facing address for all other 2387 Clients. The Proxy/Server should also include an Interface 2388 Attributes sub-option in the OMNI option with FMT/SRT/LHS information 2389 for its INET interface. The Proxy/Server then includes one or more 2390 RIOs that encode the MSPs for the OMNI link, plus an MTU option (see 2391 Section 3.9). The Proxy/Server finally forwards the message to the 2392 Client using OAL encapsulation/fragmentation if necessary while 2393 including an acknowledgement if the RS invoked window 2394 synchronization. 2396 After the initial RS/RA exchange, the Proxy/Server maintains a 2397 ReachableTime timer for each of the Client's underlying interfaces 2398 individually (and for the Client's NCE collectively) set to expire 2399 after ReachableTime seconds. If the Client (or Proxy) issues 2400 additional RS messages, the Proxy/Server sends an RA response and 2401 resets ReachableTime. If the Proxy/Server receives an ND message 2402 with a prefix release indication it sets the Client's NCE to the 2403 DEPARTED state and withdraws the MNP from the routing system after a 2404 short delay (e.g., 2 seconds). If ReachableTime expires before a new 2405 RS is received on an individual underlying interface, the Proxy/ 2406 Server marks the interface as DOWN. If ReachableTime expires before 2407 any new RS is received on any individual underlying interface, the 2408 Proxy/Server sets the NCE state to STALE and sets a 10 second timer. 2409 If the Proxy/Server has not received a new RS or ND message with a 2410 prefix release indication before the 10 second timer expires, it 2411 deletes the NCE and withdraws the MNP from the routing system. 2413 The Proxy/Server processes any ND messages pertaining to the Client 2414 and returns an NA/RA reply in response to solicitations. The Proxy/ 2415 Server may also issue unsolicited RA messages, e.g., with reconfigure 2416 parameters to cause the Client to renegotiate its prefix delegation/ 2417 registrations, with Router Lifetime set to 0 if it can no longer 2418 service this Client, etc. Finally, If the NCE is in the DEPARTED 2419 state, the Proxy/Server deletes the entry after DepartTime expires. 2421 Note: Clients SHOULD notify former Proxy/Servers of their departures, 2422 but Proxy/Servers are responsible for expiring neighbor cache entries 2423 and withdrawing routes even if no departure notification is received 2424 (e.g., if the Client leaves the network unexpectedly). Proxy/Servers 2425 SHOULD therefore set Router Lifetime to ReachableTime seconds in 2426 solicited RA messages to minimize persistent stale cache information 2427 in the absence of Client departure notifications. A short Router 2428 Lifetime also ensures that proactive RS/RA messaging between Clients 2429 and Proxy/Servers will keep any NAT state alive (see above). 2431 Note: All Proxy/Servers on an OMNI link MUST advertise consistent 2432 values in the RA Cur Hop Limit, M and O flags, Reachable Time and 2433 Retrans Timer fields the same as for any link, since unpredictable 2434 behavior could result if different Proxy/Servers on the same link 2435 advertised different values. 2437 3.12.3.1. DHCPv6-Based Prefix Registration 2439 When a Client is not pre-provisioned with an MNP-LLA, it will need 2440 for the Proxy/Server to select one or more MNPs on its behalf and set 2441 up the correct state in the AERO routing service. (A Client with a 2442 pre-provisioned MNP may also request the Proxy/Server to select 2443 additional MNPs.) The DHCPv6 service [RFC8415] is used to support 2444 this requirement. 2446 When a Client needs to have the Proxy/Server select MNPs, it sends an 2447 RS message with source address set to the unspecified address (::) 2448 and with an OMNI option that includes a DHCPv6 message sub-option 2449 with DHCPv6 Prefix Delegation (DHCPv6-PD) parameters. When the 2450 Proxy/Server receives the RS message, it extracts the DHCPv6-PD 2451 message from the OMNI option. 2453 The Proxy/Server then acts as a "Proxy DHCPv6 Client" in a message 2454 exchange with the locally-resident DHCPv6 server, which delegates 2455 MNPs and returns a DHCPv6-PD Reply message. (If the Proxy/Server 2456 wishes to defer creation of MN state until the DHCPv6-PD Reply is 2457 received, it can instead act as a Lightweight DHCPv6 Relay Agent per 2458 [RFC6221] by encapsulating the DHCPv6-PD message in a Relay-forward/ 2459 reply exchange with Relay Message and Interface ID options.) 2461 When the Proxy/Server receives the DHCPv6-PD Reply, it adds a route 2462 to the routing system and creates an MNP-LLA based on the delegated 2463 MNP. The Proxy/Server then sends an RA back to the Client with the 2464 (newly-created) MNP-LLA as the destination address and with the 2465 DHCPv6-PD Reply message coded in the OMNI option. When the Client 2466 receives the RA, it creates a default route, assigns the Subnet 2467 Router Anycast address and sets its MNP-LLA based on the delegated 2468 MNP. 2470 Note: See [I-D.templin-6man-omni] for an MNP delegation alternative 2471 that avoids including a DHCPv6 message sub-option in the RS. Namely, 2472 when the Client requests a single MNP it can set the RS source to the 2473 unspecified address (::) and include a Node Identification sub-option 2474 and Preflen in the OMNI option (but with no DHCPv6 message sub- 2475 option). When the Proxy/Server receives the RS message, it forwards 2476 a self-generated DHCPv6 Solicit message to the DHCPv6 server on 2477 behalf of the Client. When the Proxy/Server receives the DHCPv6 2478 Reply, it prepares an RA message with an OMNI option with Preflen 2479 information (but with no DHCPv6 message sub-option), then places the 2480 (newly-created) MNP-LLA in the RA destination address and returns the 2481 message to the Client. 2483 3.13. The AERO Proxy Function 2485 Clients connect to the OMNI link via Proxy/Servers, with one Proxy/ 2486 Server for each underlying interface. Each of the Client's Proxy/ 2487 Servers must be informed of all of the Client's additional underlying 2488 interfaces. For Clients on Direct and VPNed underlying interfaces 2489 the Proxy/Server "A" for that interface is directly connected, for 2490 Clients on ANET underlying interfaces Proxy/Server "A" is located on 2491 the ANET/INET boundary, and for Clients on INET underlying interfaces 2492 Proxy/Server "A" is located somewhere in the connected Internetwork. 2493 When the Client registers with Proxy/Server "A", it must also report 2494 the registration to any other Proxy/Servers for other underlying 2495 interfaces "B", "C", "D", etc. for which an underlying interface 2496 relationship has already been established. The Proxy/Server 2497 satisfies these requirements as follows: 2499 o when Proxy/Server "A" receives a Client RS message, it first 2500 verifies that the OAL Identification is within the window for the 2501 NCE that matches the MNP-ULA for this Client neighbor and 2502 authenticates the message. (If no NCE was found, Proxy/Server "A 2503 instead creates one in the STALE state and returns an RA message 2504 with an authentication signature and any window synchronization 2505 parameters.) Proxy/Server "A" then examines the network-layer 2506 destination address. If the destination address is the ADM-LLA of 2507 a different Proxy/Server "B" (or, if the OMNI option included an 2508 MS-Register sub-option with the ADM-LLAs of one or more different 2509 Proxy/Servers "B", "C", "D", etc.), Proxy/Server "A" prepares a 2510 separate proxyed version of the RS message with an OAL header with 2511 source set to its own ADM-ULA and destination set to the other 2512 Proxy/Server's ADM-ULA. Proxy/Server "A" also includes an OMNI 2513 header with an Interface Attributes option that includes its own 2514 INET address, a unique UDP Port Number for this Client, and 2515 FMT/SRT/LHS information. Proxy/Server "A" then sets the S/ 2516 T-omIndex to the value for this Client underlying interface, then 2517 forwards the message into the OMNI link secured spanning tree. 2518 (Note: including a unique Port Number allows Proxy/Server "B" to 2519 distinguish different Clients located behind the same Proxy/Server 2520 "A" at the link-layer, whereas the link-layer addresses would 2521 otherwise be indistinguishable.) 2523 o when Proxy/Server "B" receives the RS, it authenticates the 2524 message then creates or updates a NCE for the Client with Proxy/ 2525 Server "A"'s Interface Attributes as the link-layer address 2526 information for this S/T-omIndex and caches any window 2527 synchronization parameters supplied by the Client. Proxy/Server 2528 "B" then prepares an RA message with source set to its own LLA and 2529 destination set to the Client's MNP-LLA, and with any window 2530 synchronization acknowledgements. Proxy/Server "B" then 2531 encapsulates the RA in an OAL header with source set to its own 2532 ADM-ULA and destination set to the ADM-ULA of Proxy/Server "A, 2533 performs fragmentation if necessary, then sends the resulting 2534 carrier packets into the secured spanning tree. 2536 o when Proxy/Server "A" reassembles the RA, it locates the Client 2537 NCE based on the RA destination LLA. Proxy/Server "A" then re- 2538 encapsulates the RA message with OAL source set to its own ADM-ULA 2539 and OAL destination set to the MNP-ULA of the Client, includes an 2540 authentication signature if necessary, fragments if necessary and 2541 returns the fragments to the Client. 2543 o The Client repeats this process with each Proxy/Server "B", "C", 2544 "D" for each of its additional underlying interfaces. When the 2545 Client includes multiple Proxy/Server IDs in the MS-Register 2546 option, it may receive multiple RAs - each with identical window 2547 acknowledgements. The Client can then create an independent NCE 2548 for each responding Proxy/Server and complete the window 2549 synchronization even though all Proxy/Servers received the same 2550 ISS. 2552 After the initial RS/RA exchanges each Proxy/Server forwards any of 2553 the Client's carrier packets with OAL destinations for which there is 2554 no matching NCE to a Bridge using OAL encapsulation with its own ADM- 2555 ULA as the source and the destination determined by the ORH supplied 2556 by the Client. The Proxy/Server instead forwards any carrier packets 2557 destined to a neighbor cache target directly to the target according 2558 to the OAL/link-layer information - the process of establishing 2559 neighbor cache entries is specified in Section 3.14. 2561 While the Client is still associated with each Proxy/Server "A", "A" 2562 can send NS, RS and/or unsolicited NA messages to update the neighbor 2563 cache entries of other AERO nodes on behalf of the Client and/or to 2564 convey Interface Attribute updates. This allows for higher-frequency 2565 Proxy-initiated RS/RA messaging over well-connected INET 2566 infrastructure supplemented by lower-frequency Client-initiated RS/RA 2567 messaging over constrained ANET data links. 2569 If any Proxy/Server "B", "C", "D" ceases to send solicited 2570 advertisements, Proxy/Server "A" sends unsolicited RAs to the Client 2571 with destination set to (link-local) All-Nodes multicast and with 2572 Router Lifetime set to zero to inform Clients that a Proxy/Server has 2573 failed. Although Proxy/Server "A" can engage in ND exchanges on 2574 behalf of the Client, the Client can also send ND messages on its own 2575 behalf, e.g., if it is in a better position than "A" to convey 2576 Interface Attribute changes, etc. The ND messages sent by the Client 2577 include the Client's MNP-LLA as the source in order to differentiate 2578 them from the ND messages sent by Proxy/Server "A". 2580 If the Client becomes unreachable over an underlying interface, 2581 Proxy/Server "A" sets the NCE state to DEPARTED and retains the entry 2582 for DepartTime seconds. While the state is DEPARTED, Proxy/Server 2583 "A" forwards any carrier packets destined to the Client to a Bridge 2584 via OAL/ORH encapsulation. When DepartTime expires, Proxy/Server "A" 2585 deletes the NCE and discards any further carrier packets destined to 2586 the former Client. 2588 In some ANETs that employ a Proxy/Server, the Client's MNP can be 2589 injected into the ANET routing system. In that case, the Client can 2590 send original IP packets without invoking the OAL so that the ANET 2591 routing system transports the original IP packets to the Proxy. This 2592 can be very beneficial, e.g., if the Client connects to the ANET via 2593 low-end data links such as some aviation wireless links. 2595 If the ANET first-hop access router is on the same underlying link as 2596 the Client and recognizes the AERO/OMNI protocol, the Client can 2597 avoid OAL encapsulation for both its control and data messages. When 2598 the Client connects to the link, it can send an unencapsulated RS 2599 message with source address set to its own MNP-LLA (or to a Temporary 2600 LLA), and with destination address set to the ADM-LLA of the Client's 2601 selected Proxy/Server or to (link-local) All-Routers multicast. The 2602 Client includes an OMNI option formatted as specified in 2603 [I-D.templin-6man-omni]. The Client then sends the unencapsulated RS 2604 message, which will be intercepted by the AERO-Aware access router. 2606 The ANET access router then performs OAL encapsulation on the RS 2607 message and forwards it to a Proxy/Server at the ANET/INET boundary. 2608 When the access router and Proxy/Server are one and the same node, 2609 the Proxy/Server would share and underlying link with the Client but 2610 its message exchanges with outside correspondents would need to pass 2611 through a security gateway at the ANET/INET border. The method for 2612 deploying access routers and Proxys (i.e. as a single node or 2613 multiple nodes) is an ANET-local administrative consideration. 2615 Note: The Proxy/Server can apply packing as discussed in 2616 [I-D.templin-6man-omni] if an opportunity arises to concatenate 2617 multiple original IP packets destined to the same neighbor. 2619 3.13.1. Detecting and Responding to Proxy/Server Failures 2621 In environments where fast recovery from Proxy/Server failure is 2622 required, Proxy/Server "A" SHOULD use proactive Neighbor 2623 Unreachability Detection (NUD) to track peer Proxy/Server "B" 2624 reachability in a similar fashion as for Bidirectional Forwarding 2625 Detection (BFD) [RFC5880]. Proxy/Server "A" can then quickly detect 2626 and react to failures so that cached information is re-established 2627 through alternate paths. The NUD control messaging is carried only 2628 over well-connected ground domain networks (i.e., and not low-end 2629 aeronautical radio links) and can therefore be tuned for rapid 2630 response. 2632 Proxy/Server "A" performs proactive NUD with peer Proxy/Server "B" 2633 for which there are currently active Clients by sending continuous NS 2634 messages in rapid succession, e.g., one message per second. Proxy/ 2635 Server "A" sends the NS message via the spanning tree with its own 2636 ADM-LLA as the source and the ADM-LLA of the peer Proxy/Server "B" as 2637 the destination. When Proxy/Server "A" is also sending RS messages 2638 to the peer Proxy/Server "B" on behalf of ANET Clients, the resulting 2639 RA responses can be considered as equivalent hints of forward 2640 progress. This means that Proxy/Server "B" need not also send a 2641 periodic NS if it has already sent an RS within the same period. If 2642 the peer Proxy/Server "B" fails (i.e., if "A" ceases to receive 2643 advertisements), Proxy/Server "A" can quickly inform Clients by 2644 sending multicast RA messages on the ANET interface. 2646 Proxy/Server "A" sends RA messages on the ANET interface with source 2647 address set to Proxy/Server "B"'s address, destination address set to 2648 (link-local) All-Nodes multicast, and Router Lifetime set to 0. 2649 Proxy/Server "A" SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA messages 2650 separated by small delays [RFC4861]. Any Clients on the ANET that 2651 had been using the failed Proxy/Server "B" will receive the RA 2652 messages and associate with a new Proxy/Server. 2654 3.13.2. Point-to-Multipoint Proxy/Server Coordination 2656 In environments where Client messaging over ANETs is bandwidth- 2657 limited and/or expensive, Clients can enlist the services of Proxy/ 2658 Server "A" to coordinate with multiple Proxy/Servers "B", "C", "D" 2659 etc. in a single RS/RA message exchange. The Client can send a 2660 single RS message to (link-local) All-Routers multicast that includes 2661 the ID's of multiple Proxy/Servers in MS-Register sub-options of the 2662 OMNI option. 2664 When Proxy/Server "A" receives the RS and processes the OMNI option, 2665 it sends a separate RS to each MS-Register Proxy/Server ID. When 2666 Proxy/Server "A" receives an RA, it can optionally return an 2667 immediate "singleton" RA to the Client or record the Proxy/Server's 2668 ID for inclusion in a pending "aggregate" RA message. Proxy/Server 2669 "A" can then return aggregate RA messages to the Client including 2670 multiple Proxy/Server IDs in order to conserve bandwidth. Each RA 2671 includes a proper subset of the Proxy/Server IDs from the original RS 2672 message, and Proxy/Server "A" must ensure that the message contents 2673 of each RA are consistent with the information received from the 2674 (aggregated) additional Proxy/Servers. 2676 Clients can thereafter employ efficient point-to-multipoint Proxy/ 2677 Server coordination under the assistance of Proxy/Server "A" to 2678 reduce the number of messages sent over the ANET while enlisting the 2679 support of multiple Proxy/Servers for fault tolerance. Clients can 2680 further include MS-Release sub-options in IPv6 ND messages to request 2681 Proxy/Server "A" to release from former Proxy/Servers via the 2682 procedures discussed in Section 3.16.5. 2684 When the Client sends an RS with window synchronization parameters 2685 and with multiple MS-Register Proxy/Server IDs, Proxy/Server "A" may 2686 receive multiple RAs - each with their own window synchronization 2687 parameters. Proxy/Server "A" must then immediately forward these RAs 2688 to the Client as singletons instead of including them in an 2689 aggregate, and the Client will use each RA to establish a separate 2690 NCE and window for each individual Proxy/Server. 2692 The OMNI interface specification [I-D.templin-6man-omni] provides 2693 further discussion of the RS/RA messaging involved in point-to- 2694 multipoint coordination. 2696 3.14. AERO Route Optimization 2698 AERO nodes invoke route optimization when they need to forward 2699 packets to new target destinations. Route optimization is based on 2700 IPv6 ND Address Resolution messaging between a Route Optimization 2701 Source (ROS) and Route Optimization Responder (ROR). Route 2702 optimization is initiated by the first eligible ROS closest to the 2703 source as follows: 2705 o For Clients on VPNed and Direct interfaces, the Proxy/Server is 2706 the ROS. 2708 o For Clients on ANET interfaces, either the Client or the Proxy/ 2709 Server may be the ROS. 2711 o For Clients on INET interfaces, the Client itself is the ROS. 2713 o For correspondent nodes on INET/EUN interfaces serviced by a 2714 Relay, the Relay is the ROS. 2716 The route optimization procedure is conducted between the ROS and the 2717 nearest Proxy/Server/Relay for the target selected by routing as the 2718 ROR. In this arrangement, the ROS is always the Client or 2719 Proxy/Server/Relay nearest the source over the selected source 2720 underlying interface, while the ROR is always a Proxy/Server/Relay 2721 that services the target regardless of the target underlying 2722 interface. 2724 The AERO routing system directs a route optimization solicitation 2725 sent by the ROS to the nearest available ROR, which returns a route 2726 optimization reply. The exact ROR selected is unimportant; all that 2727 matters is that the route optimization information returned must be 2728 current and authentic. The ROS is responsible for periodically 2729 refreshing the route optimization, and the ROR is responsible for 2730 quickly informing the ROS of any changes. 2732 The procedures are specified in the following sections. 2734 3.14.1. Route Optimization Initiation 2736 When an original IP packet from a source node destined to a target 2737 node arrives, the ROS checks for a NCE with an MNP-LLA that matches 2738 the target destination. If there is a NCE in the REACHABLE state, 2739 the ROS invokes the OAL and forwards the resulting carrier packets 2740 according to the cached state then returns from processing. 2741 Otherwise, if there is no NCE the ROS creates one in the INCOMPLETE 2742 state. 2744 The ROS next places the original IP packet on a short queue then 2745 sends an NS message for Address Resolution (NS(AR)) to receive a 2746 solicited NA(AR) message from an ROR. The NS(AR) message must be 2747 sent securely, and includes: 2749 o the LLA of the ROS as the source address. 2751 o the MNP-LLA corresponding to the original IP packet's destination 2752 as the Target Address, e.g., for 2001:db8:1:2::10:2000 the Target 2753 Address is fe80::2001:db8:1:2. 2755 o the Solicited-Node multicast address [RFC4291] formed from the 2756 lower 24 bits of the original IP packet's destination as the 2757 destination address, e.g., for 2001:db8:1:2::10:2000 the NS(AR) 2758 destination address is ff02:0:0:0:0:1:ff10:2000. 2760 The NS(AR) message also includes an OMNI option with an Interface 2761 Attributes entry for the underlying interface, with S/T-omIndex set 2762 to the underlying interface index and with Preflen set to the prefix 2763 length associated with the NS(AR) source. The ROS then selects an 2764 Identification value submits the NS(AR) message for OAL encapsulation 2765 with OAL source set to its own ULA and OAL destination set to the ULA 2766 corresponding to the target. (The ROS does not include any window 2767 synchronization parameters, since it will never exchange other 2768 carrier packet types directly with the ROR). 2770 The ROS then sends the resulting carrier packet(s) into the secured 2771 spanning tree without decrementing the network-layer TTL/Hop Limit 2772 field. (When the ROS is an INET Client, it instead sends the 2773 resulting carrier packets to the ADM-ULA of one of its current Proxy/ 2774 Servers. The Proxy/Server then reassembles if necessary, verifies 2775 the NS(AR) signature, then re-encapsulates with the OAL source set to 2776 its own ADM-ULA and OAL destination set to the ULA corresponding to 2777 the target. The Proxy/Server then fragments if necessary and sends 2778 the resulting carrier packets into the secured spanning tree on 2779 behalf of the Client.) 2781 3.14.2. Relaying the NS(AR) *NET Packet(s) 2783 When the Bridge receives the carrier packet(s) containing the RS from 2784 the ROS, it discards the *NET headers and determines the next hop by 2785 consulting its standard IPv6 forwarding table for the OAL header 2786 destination address. The Bridge then decrements the OAL header Hop- 2787 Limit, then re-encapsulates and forwards the carrier packet(s) via 2788 the secured spanning tree the same as for any IPv6 router, where it 2789 may traverse multiple OMNI link segments. The final-hop Bridge will 2790 deliver the carrier packet(s) via the secured spanning tree to a 2791 Proxy/Server or Relay that services the target. 2793 3.14.3. Processing the NS(AR) and Sending the NA(AR) 2795 When the target Proxy/Server (or Relay) receives the secured carrier 2796 packet(s), it reassembles if necessary then examines the NS(AR) 2797 target to determine whether it has a matching NCE and/or non-MNP 2798 route. If there is no match, the Proxy/Server drops the message. 2799 Otherwise, the Proxy/Server/Relay continues processing as follows: 2801 o if the NS(AR) target matches a Client NCE in the DEPARTED state, 2802 the Proxy/Server re-encapsulates while setting the OAL source to 2803 the ULA of the ROS and OAL destination address to the ADM-ULA of 2804 the Client's new Proxy/Server. The (old) Proxy/Server then 2805 fragments if necessary and forwards the resulting carrier 2806 packet(s) over the secured spanning tree then returns from 2807 processing. 2809 o If the NS(AR) target matches the MNP-LLA of a Client NCE in the 2810 REACHABLE state, the Proxy/Server makes note of whether the NS 2811 (AR) arrived from the secured or unsecured spanning tree then acts 2812 as an ROR to provide route optimization information on behalf of 2813 the Client. (Note that if the message arrived via the secured 2814 spanning tree the ROR need not perform further authentication, but 2815 if it arrived over an open INET underlying interface it must 2816 verify the message authentication signature before accepting.) 2818 o If the NS(AR) target matches one of its non-MNP routes, the Relay 2819 acts as both an ROR and a route optimization target, since the 2820 Relay forwards IP packets toward the (fixed network) target at the 2821 network layer. 2823 The ROR next checks the target NCE for a Report List entry that 2824 matches the NS(AR) source LLA/ULA of the ROS. If there is a Report 2825 List entry, the ROR refreshes ReportTime for this ROR; otherwise, the 2826 ROR creates a new entry for the ROS and records both the LLA and ULA. 2828 The ROR then prepares a (solicited) NA(AR) message to return to the 2829 ROS with the source address set to the target's MNP-LLA, the 2830 destination address set to the NS(AR) LLA source address and the 2831 Target Address set to the same value that appeared in the NS(AR). 2832 The ROR then includes an OMNI option with Preflen set to the prefix 2833 length associated with the NA(AR) source address. The ROR next 2834 includes Interface Attributes in the OMNI option for all of the 2835 target's underlying interfaces with current information for each 2836 interface. 2838 For each Interface Attributes sub-option, the ROR sets the L2ADDR 2839 according to the Proxy/Server's INET address for VPNed or Direct 2840 interfaces, to the INET address of the Proxy/Server for proxyed 2841 interfaces or to the Client's INET address for INET interfaces. The 2842 ROR then includes the lower 32 bits of the Proxy/Server's ADM-ULA as 2843 the LHS, encodes the ADM-ULA prefix length code in the SRT field and 2844 sets FMT as specified in Section 3.3. 2846 The ROR then sets the NA(AR) message R flag to 1 (as a router) and S 2847 flag to 1 (as a response to a solicitation) and sets the O flag to 0 2848 (as a proxy) and sets the OMNI header S/T-omIndex to 0. The ROR 2849 finally submits the NA(AR) for OAL encapsulation with source set to 2850 its own ULA and destination set to the same ULA that appeared in the 2851 NS(AR) OAL source, then performs OAL encapsulation and fragmentation 2852 using the same Identification value that appeared in the NS(AR) and 2853 finally forwards the resulting (*NET-encapsulated) carrier packets 2854 via the secured spanning tree without decrementing the network-layer 2855 TTL/Hop Limit field. 2857 3.14.4. Relaying the NA(AR) 2859 When the Bridge receives NA(AR) carrier packets from the ROR, it 2860 discards the *NET header and determines the next hop by consulting 2861 its standard IPv6 forwarding table for the OAL header destination 2862 address. The Bridge then decrements the OAL header Hop-Limit, re- 2863 encapsulates the carrier packet and forwards it via the secured 2864 spanning tree the same as for any IPv6 router, where it may traverse 2865 multiple OMNI link segments. The final-hop Bridge will deliver the 2866 carrier packet via the secured spanning tree to a Proxy/Server for 2867 the ROS. 2869 3.14.5. Processing the NA(AR) 2871 When the ROS receives the NA(AR) message from the ROR, it first 2872 searches for a NCE that matches the NA(AR) LLA source address. The 2873 ROS then processes the message the same as for standard IPv6 Address 2874 Resolution [RFC4861]. In the process, it caches all OMNI option 2875 information in the target NCE (including all Interface Attributes). 2877 When the ROS is a Client, the solicited NA(AR) message will first be 2878 delivered via the secured spanning tree to the Proxy/Server that 2879 forwarded the NS(AR), which reassembles if necessary. The Proxy/ 2880 Server then forwards the message to the Client while re-encapsulating 2881 and re-fragmenting if necessary. If the Client is on an ANET, ANET 2882 physical security and protected spectrum ensures security for the 2883 unmodified NA(AR); if the Client is on the open INET the Proxy/Server 2884 must instead insert an authentication signature. The Proxy/Server 2885 uses its own ADM-ULA as the OAL source and the MNP-ULA of the Client 2886 as the OAL destination. 2888 3.14.6. Forwarding Packets to Route Optimized Targets 2890 After the ROS receives the route optimization NA(AR) and updates the 2891 target NCE, it can begin forwarding packets along the best paths 2892 based on the target's Interface Attributes. The ROS selects target 2893 underlying interfaces according to traffic selectors and/or any other 2894 traffic discriminators, however each underlying interface selected 2895 must first establish window synchronization state if necessary. 2897 To establish window synchronization state, the ROS performs a secured 2898 unicast NS/NA(WIN) exchange with window synchronization parameters 2899 according to the Interface Attribute FMT. If FMT-Forward is set, the 2900 ROS prepares an NS(WIN) with its own LLA as the source and the MNP- 2901 LLA of the target Client as the destination; otherwise, it sets the 2902 ADM-LLA of the LHS Proxy/Server as the destination. The ROS then 2903 encapsulates the NS(WIN) in an OAL header with its own ULA as the 2904 source. If the ROS is the Client, it sets the OAL destination to the 2905 ADM-ULA of its Proxy/Server, includes an authentication signature if 2906 necessary, and includes an ORH-1 with FMT-Type clear for the first 2907 fragment. The Client sets the ORH Segments Left to 1 and includes 2908 valid SRT/LHS information for the LHS Proxy/Server with L2ADDR set to 2909 0, then forwards the NS(WIN) to its own Proxy/Server which must 2910 reassemble and verify the authentication signature if necessary. The 2911 Proxy/Server then re-encapsulates, re-fragments and forwards the 2912 NS(WIN) carrier packets into the secured spanning tree with its own 2913 ADM-ULA as the OAL source and the ADM-ULA of the LHS Proxy/Server as 2914 the OAL destination while replacing the ORH-1 with an ORH-0. (If the 2915 ROS was the Proxy/Server itself, it instead includes an ORH-0, and 2916 forwards the carrier packets into the secured spanning tree.) 2918 When an LHS Proxy/Server receives the NS(WIN) it first reassembles if 2919 necessary. If the NS(WIN) destination is its own ADM-LLA, the LHS 2920 Proxy/Server creates an NCE based on the NS(WIN) source LLA, caches 2921 the window synchronization information, and prepares an NA(WIN) while 2922 using its own ADM-LLA as the source and the ROS LLA as the 2923 destination. The LHS Proxy/Server then encapsulates the NA(WIN) in 2924 an OAL header with source set to its own ADM-ULA and destination set 2925 to the NS(WIN) OAL source. The LHS Proxy/Server then fragments if 2926 necessary includes an ORH-0 with omIndex set to the S/T-omIndex value 2927 found in the NS(WIN) OMNI option, then forwards the resulting carrier 2928 packets into the secured spanning tree which will deliver them to the 2929 ROS Proxy/Server. 2931 If the NS(WIN) destination is the MNP-LLA of the target Client, the 2932 LHS Proxy/Server instead re-encapsulates using the same OAL source 2933 and the MNP-ULA of the target as the OAL destination and includes an 2934 authentication signature if necessary while removing the ORH-0. The 2935 LHS Proxy/Server then forwards the NS(WIN) to the target over the 2936 underlying interface identified by the ORH-0 omIndex (or, over any 2937 underlying interface if omIndex is 0). When the target receives the 2938 NS(WIN), it verifies the authentication signature if necessary then 2939 creates an NCE for the ROS LLA, caches the window synchronization 2940 information and prepares an NA(WIN) to return to the ROS with its 2941 MNP-LLA as the source and the LLA of the ROS as the destination, and 2942 with an authentication signature if necessary. The target Client 2943 then encapsulates the NA(WIN) in an OAL header with its own MNP-ULA 2944 as the source, the ADM-ULA of the LHS Proxy/Server as the 2945 destination, and with an ORH-1 with SRT/LHS information copied from 2946 the ADM-ULA of the ROS Proxy/Server found in the NS(WIN) OAL source 2947 address. The target Client then sets the ORH-1 omIndex to the S/ 2948 T-omIndex value found in the NS(WIN) OMNI option, then forward the 2949 message to the LHS Proxy/Server. 2951 When the LHS Proxy/Server receives the message, it reassembles if 2952 necessary, verifies the authentication signature if necessary then 2953 re-encapsulates using its own ADM-ULA as the source and the ADM-ULA 2954 of the ROS Proxy/Server as the destination The LHS Proxy/Server then 2955 re-fragments and forwards the NS(WIN) carrier packets into the 2956 spanning tree while replacing the ORH-1 with an ORH-0. When the ROS 2957 Proxy/Server receives the NA(WIN), it reassembles if necessary then 2958 updates the target NCE based on the message contents if the Proxy/ 2959 Server itself is the ROS. If the NS(WIN) source was the ADM-LLA of 2960 the LHS Proxy/Server, the ROS must create and maintain a NCE for the 2961 LHS Proxy/Server which it must regard as a companion to the existing 2962 MNP-LLA NCE for the target Client. (The NCE for the LHS Proxy/Server 2963 can also be shared by multiple target Client NCEs if the ROS 2964 communicates with multiple active targets located behind the same LHS 2965 Proxy/Server.) If the ROS is the Client, the Proxy/Server instead 2966 inserts an authentication signature if necessary, removes the ORH-0 2967 then re-encapsulates and re-fragments if necessary while changing the 2968 OAL destination to the MNP-ULA of the ROS Client. The Proxy/Server 2969 then forwards the NA(WIN) to the ROS Client over the underlying 2970 interface identified by the ORH-0 omIndex which then updates its own 2971 NCE based on the target Client MNP-LLA or LHS Proxy/Server ADM-LLA. 2972 The ROS (whether the Proxy/Server or the Client itself) finally 2973 arranges to return an acknowledgement if requested by the NA(WIN). 2975 After window synchronization state has been established, the ROS can 2976 begin forwarding carrier packets as specified in Section 3.2.7 while 2977 performing additional NS/NA(WIN) exchanges as above to update window 2978 state and/or test reachability. These forwarding procedures apply to 2979 the case where the selected target interface SRT/LHS codes indicate 2980 that the interface is located in a foreign OMNI link segment. In 2981 that case, the ROS must include ORHs and send the resulting carrier 2982 packets into the spanning tree. 2984 If the SRT/LHS codes indicate that the interface is in the local OMNI 2985 link segment, the ROS can instead forward carrier packets directly to 2986 the LHS Proxy/Server using the L2ADDR for encapsulation, or even to 2987 the target Client itself while invoking NAT traversal if necessary. 2988 When the ROS sends packets directly to the LHS Proxy/Server, it 2989 includes an ORH-0 if necessary to inform the Proxy/Server as to 2990 whether it must reassemble and/or the correct target Client interface 2991 for (re)forwarding. If the LHS Proxy/Server is required to 2992 reassemble, the ROS sets the OAL destination to the ADM-ULA of the 2993 LHS Proxy/Server; otherwise, it sets the OAL destination to the MNP- 2994 ULA of the target Client itself. When the ROS sends packets directly 2995 to the target Client, it need not include an ORH. The LHS Proxy/ 2996 Server (or target Client) then saves the L2ADDR and full OAL 2997 addresses in the ROS NCE, and the ROS can begin applying OAL header 2998 compression in subsequent carrier packets as specified in 2999 [I-D.templin-6man-omni] since the OAL header is not examined by any 3000 forwarding nodes in the path. 3002 While the ROS continues to actively forward packets to the target 3003 Client, it is responsible for updating window synchronization state 3004 and per-interface reachability before expiration. Window 3005 synchronization state is shared by all underlying interfaces in the 3006 ROS' NCE that use the same destination LLA so that a single NS/ 3007 NA(NUD) exchange applies for all interfaces regardless of the 3008 (single) interface used to conduct the exchange. However, the window 3009 synchronization exchange only confirms target Client reachability 3010 over the specific interface used to conduct the exchange. 3011 Reachability for other underlying interfaces that share the same 3012 window synchronization state must be determined individually using 3013 NS/NA(NUD) messages which need not be secured as long as they use in- 3014 window Identifications and do not update other state information. 3016 3.15. Neighbor Unreachability Detection (NUD) 3018 AERO nodes perform Neighbor Unreachability Detection (NUD) per 3019 [RFC4861] either reactively in response to persistent link-layer 3020 errors (see Section 3.11) or proactively to confirm reachability. 3021 The NUD algorithm is based on periodic control message exchanges and 3022 may further be seeded by ND hints of forward progress, but care must 3023 be taken to avoid inferring reachability based on spoofed 3024 information. For example, IPv6 ND message exchanges that include 3025 authentication codes and/or in-window Identifications may be 3026 considered as acceptable hints of forward progress, while spurious 3027 random carrier packets should be ignored. 3029 AERO nodes can use standard NS/NA(NUD) exchanges sent over the OMNI 3030 link secured spanning tree (i.e. the same as described above for NS/ 3031 NA(WIN)) to test reachability without risk of DoS attacks from nodes 3032 pretending to be a neighbor. These NS/NA(NUD) messages use the 3033 unicast LLAs and ULAs of the parties involved in the NUD test the 3034 same as for standard IPv6 ND over the secured spanning tree. When 3035 only reachability information is required without updating any other 3036 NCE state, unsecured NS/NA(NUD) messages may instead be exchanged 3037 directly between neighbors as long as they include in-window 3038 Identifications. 3040 When an ROR directs an ROS to a target neighbor with one or more 3041 link-layer addresses, the ROS probes each unsecured target underlying 3042 interface either proactively or on-demand of carrier packets directed 3043 to the path by multilink forwarding to maintain the interface's state 3044 as reachable. Probing is performed through NS(NUD) messages over 3045 either the secured or unsecured spanning tree, or through NS(NUD) 3046 messages sent directly to an underlying interface of the target 3047 itself. While testing a target underlying interface, the ROS can 3048 optionally continue to forward carrier packets via alternate 3049 interfaces and/or maintain a small queue of carrier packets until 3050 target reachability is confirmed. 3052 NS(NUD) messages are encapsulated, fragmented and transmitted as 3053 carrier packets the same as for ordinary original IP data packets, 3054 however the encapsulated destinations are the LLA of the ROS and 3055 either the ADM-LLA of the LHS Proxy/Server or the MNP-LLA of the 3056 target itself. The ROS encapsulates the NS(NUD) message the same as 3057 described in Section 3.2.7, however Destination Suffixes (if present) 3058 are set according to the LLA destination (i.e., and not a ULA/GUA 3059 destination). The ROS sets the NS(NUD) OMNI header S/T-omIndex to 3060 identify the underlying interface used for forwarding (or to 0 if any 3061 underlying interface can be used). The ROS also includes an ORH with 3062 SRT/LHS/LLADDR information the same as for ordinary data packets, but 3063 no authentication signatures are included. The ROS then fragments 3064 the OAL packet and forwards the resulting carrier packets into the 3065 unsecured spanning tree or directly to the target (or LHS Proxy/ 3066 Server) if it is in the local segment. 3068 When the target (or LHS Proxy/Server) receives the NS(NUD) carrier 3069 packets, it verifies that it has a NCE for this ROS and that the 3070 Identification is in-window, then submits the carrier packets for 3071 reassembly. The node then searches for Interface Attributes in its 3072 NCE for the ROS that match the NS(NUD) S/T-omIndex and uses the 3073 SRT/LHS/L2ADDR and FMT information to prepare an ORH for the NA(NUD) 3074 reply. The node then prepare the NA(NUD) with the source and 3075 destination LLAs reversed, encapsulates and sets the OAL source and 3076 destination, sets the NA(NUD) S/T-omIndex to the index of the 3077 underlying interface the NS(NUD) arrived on and sets the Target 3078 Address to the same value included in the NS(NUD). The target next 3079 sets the R flag to 1, the S flag to 1 and the O flag to 1, then 3080 selects an in-window Identification for the ROS and performs 3081 fragmentation. The node then forwards the carrier packets into the 3082 unsecured spanning tree, directly to the ROS if it is in the local 3083 segment or directly to a Bridge in the local segment. 3085 When the ROS receives the NA(NUD), it marks the target underlying 3086 interface tested as "reachable". Note that underlying interface 3087 states are maintained independently of the overall NCE REACHABLE 3088 state, and that a single NCE may have multiple target underlying 3089 interfaces in various states "reachable" and otherwise while the NCE 3090 state as a whole remains REACHABLE. 3092 Note also that the exchange of NS/NA(NUD) messages has the useful 3093 side-benefit of opening holes in NATs that may be useful for NAT 3094 traversal. 3096 3.16. Mobility Management and Quality of Service (QoS) 3098 AERO is a Distributed Mobility Management (DMM) service. Each Proxy/ 3099 Server is responsible for only a subset of the Clients on the OMNI 3100 link, as opposed to a Centralized Mobility Management (CMM) service 3101 where there is a single network mobility collective entity for all 3102 Clients. Clients coordinate with their associated Proxy/Servers via 3103 RS/RA exchanges to maintain the DMM profile, and the AERO routing 3104 system tracks all current Client/Proxy/Server peering relationships. 3106 Proxy/Servers provide default routing and mobility/multilink services 3107 for their dependent Clients. Clients are responsible for maintaining 3108 neighbor relationships with their Proxy/Servers through periodic RS/ 3109 RA exchanges, which also serves to confirm neighbor reachability. 3110 When a Client's underlying Interface Attributes change, the Client is 3111 responsible for updating the Proxy/Server with this new information. 3112 Note that when there is a Proxy/Server in the path, the Proxy 3113 function can also perform some RS/RA exchanges on the Client's 3114 behalf. 3116 Mobility management messaging is based on the transmission and 3117 reception of unsolicited Neighbor Advertisement (uNA) messages. Each 3118 uNA message sets the IPv6 source address to the LLA of the ROR and 3119 the destination address to the unicast LLA of the ROS. 3121 Mobility management considerations are specified in the following 3122 sections. 3124 3.16.1. Mobility Update Messaging 3126 RORs accommodate Client mobility and/or multilink change events by 3127 sending secured uNA messages to each ROS in the target Client's 3128 Report List. When an ROR sends a uNA message, it sets the IPv6 3129 source address to the its own LLA, sets the destination address to 3130 the ROS LLA (i.e., an MNP-LLA if the ROS is a Client and an ADM-LLA 3131 if the ROS is a Proxy/Server) and sets the Target Address to the 3132 Client's MNP-LLA. The ROR also includes an OMNI option with Preflen 3133 set to the prefix length associated with the Client's MNP-LLA, with 3134 Interface Attributes for the target Client's underlying interfaces 3135 and with the OMNI header S/T-omIndex set to 0. The ROR then sets the 3136 uNA R flag to 1, S flag to 0 and O flag to 1, then encapsulates the 3137 message in an OAL header with source set to its own ADM-ULA and 3138 destination set to the ROS ULA (i.e., the ADM-ULA of the ROS Proxy/ 3139 Server) and sends the message into the secured spanning tree. 3141 As discussed in Section 7.2.6 of [RFC4861], the transmission and 3142 reception of uNA messages is unreliable but provides a useful 3143 optimization. In well-connected Internetworks with robust data links 3144 uNA messages will be delivered with high probability, but in any case 3145 the Proxy/Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT 3146 uNAs to each ROS to increase the likelihood that at least one will be 3147 received. Alternatively, the Proxy/Server can set the PNG flag in 3148 the uNA OMNI option header to request a solicited NA acknowledgement 3149 as specified in [I-D.templin-6man-omni]. 3151 When the ROS Proxy/Server receives a uNA message prepared as above, 3152 it ignores the message if the destination is not its own ADM-ULA or 3153 the MNP-ULA of the Client ROS. In the former case, it uses the 3154 included OMNI option information to update its NCE for the target, 3155 but does not reset ReachableTime since the receipt of an unsolicited 3156 NA message from the ROR does not provide confirmation that any 3157 forward paths to the target Client are working. If the destination 3158 was the MNP-ULA of the ROS Client, the ROS Proxy/Server instead re- 3159 encapsulates with the OAL source set to its own ADM-ULA, OAL 3160 destination set to the MNP-ULA of the ROS Client with an 3161 authentication signature if necessary, and with an in-window 3162 Identification for this Client. Finally, if the uNA message PNG flag 3163 was set, the ROS returns a solicited NA acknowledgement as specified 3164 in [I-D.templin-6man-omni]. 3166 In addition to sending uNA messages to the current set of ROSs for 3167 the target Client, the ROR also sends uNAs to the MNP-ULA associated 3168 with the link-layer address for any underlying interface for which 3169 the link-layer address has changed. These uNA messages update an old 3170 Proxy/Server that cannot easily detect (e.g., without active probing) 3171 when a formerly-active Client has departed. When the ROR sends the 3172 uNA, it sets the IPv6 source address to its LLA, sets the destination 3173 address to the old Proxy/Server's ADM-LLA, and sets the Target 3174 Address to the Client's MNP-LLA. The ROR also includes an OMNI 3175 option with Preflen set to the prefix length associated with the 3176 Client's MNP-LLA, with Interface Attributes for the changed 3177 underlying interface, and with the OMNI header S/T-omIndex set to 0. 3178 The ROR then sets the uNA R flag to 1, S flag to 0 and O flag to 1, 3179 then encapsulates the message in an OAL header with source set to its 3180 own ULA and destination set to the ADM-ULA of the old Proxy/Server 3181 and sends the message into the secured spanning tree. 3183 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 3185 When a Client needs to change its underlying Interface Attributes 3186 (e.g., due to a mobility event), the Client requests one of its 3187 Proxy/Servers to send uNA or RS messages to all of its other Proxy/ 3188 Servers via the secured spanning tree with an OMNI option that 3189 includes Interface attributes with the new link quality and address 3190 information. 3192 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 3193 sending carrier packets containing user data in case one or more RAs 3194 are lost. If all RAs are lost, the Client SHOULD re-associate with a 3195 new Proxy/Server. 3197 When the Proxy/Server receives the Client's changes, it sends uNA 3198 messages to all nodes in the Report List the same as described in the 3199 previous section. 3201 3.16.3. Bringing New Links Into Service 3203 When a Client needs to bring new underlying interfaces into service 3204 (e.g., when it activates a new data link), it sends an RS message to 3205 the Proxy/Server via the underlying interface with an OMNI option 3206 that includes Interface Attributes with appropriate link quality 3207 values and with link-layer address information for the new link. 3209 3.16.4. Deactivating Existing Links 3211 When a Client needs to deactivate an existing underlying interface, 3212 it sends an RS or uNA message to its Proxy/Server with an OMNI option 3213 with appropriate Interface Attribute values - in particular, the link 3214 quality value 0 assures that neighbors will cease to use the link. 3216 If the Client needs to send RS/uNA messages over an underlying 3217 interface other than the one being deactivated, it MUST include 3218 Interface Attributes with appropriate link quality values for any 3219 underlying interfaces being deactivated. 3221 Note that when a Client deactivates an underlying interface, 3222 neighbors that have received the RS/uNA messages need not purge all 3223 references for the underlying interface from their neighbor cache 3224 entries. The Client may reactivate or reuse the underlying interface 3225 and/or its omIndex at a later point in time, when it will send RS/uNA 3226 messages with fresh Interface Attributes to update any neighbors. 3228 3.16.5. Moving Between Proxy/Servers 3230 The Client performs the procedures specified in Section 3.12.2 when 3231 it first associates with a new Proxy/Server or renews its association 3232 with an existing Proxy/Server. The Client also includes MS-Release 3233 identifiers in the RS message OMNI option per [I-D.templin-6man-omni] 3234 if it wants the new Proxy/Server to notify any old Proxy/Servers from 3235 which the Client is departing. 3237 When the new Proxy/Server receives the Client's RS message, it 3238 returns an RA as specified in Section 3.12.3 and sends uNA messages 3239 to any old Proxy/Servers listed in OMNI option MS-Release 3240 identifiers. When the new Proxy/Server sends a uNA message, it sets 3241 the IPv6 source address to the Client's MNP-LLA, sets the destination 3242 address to the old Proxy/Server's ADM-LLA, and sets the Target 3243 Address to 0. The new Proxy/Server also includes an OMNI option with 3244 Preflen set to the prefix length associated with the Client's MNP- 3245 LLA, with Interface Attributes for its own underlying interface, and 3246 with the OMNI header S/T-omIndex set to 0. The new Proxy/Server then 3247 sets the uNA R flag to 1, S flag to 0 and O flag to 1, then 3248 encapsulates the message in an OAL header with source set to its own 3249 ADM-ULA and destination set to the ADM-ULA of the old Proxy/Server 3250 and sends the message into the secured spanning tree. 3252 When an old Proxy/Server receives the uNA, it notices that the 3253 message appears to have originated from the Client's MNP-LLA but that 3254 the Target Address is 0. The old Proxy/Server then changes the 3255 Client's NCE state to DEPARTED, sets the link-layer address of the 3256 Client to the new Proxy/Server's ADM-ULA, and resets DepartTime. 3257 After a short delay (e.g., 2 seconds) the old Proxy/Server withdraws 3258 the Client's MNP from the routing system. After DepartTime expires, 3259 the old Proxy/Server deletes the Client's NCE. 3261 The old Proxy/Server also iteratively forwards a copy of the uNA 3262 message to each ROS in the Client's Report List by changing the OAL 3263 destination address to the ULA of the ROS while leaving all other 3264 fields of the message unmodified. When the ROS receives the uNA, it 3265 examines the source address to determine the target Client NCE and 3266 verifies that the destination address matches the old Proxy/Server. 3267 The ROS then caches the ULA source address as the new Proxy/Server 3268 for the existing NCE and marks the entry as STALE. While in the 3269 STALE state, the ROS allows new carrier packets to flow according to 3270 any alternate reachable underlying interfaces and sends new NS(AR) 3271 messages using its own ULA as the OAL source and the ADM-ULA of the 3272 new Proxy/Server as the OAL destination address to elicit NA(AR) 3273 messages that reset the NCE state to REACHABLE. 3275 Clients SHOULD NOT move rapidly between Proxy/Servers in order to 3276 avoid causing excessive oscillations in the AERO routing system. 3277 Examples of when a Client might wish to change to a different Proxy/ 3278 Server include a Proxy/Server that has gone unreachable, topological 3279 movements of significant distance, movement to a new geographic 3280 region, movement to a new OMNI link segment, etc. 3282 When a Client moves to a new Proxy/Server, some of the carrier 3283 packets of a multiple fragment OAL packet may have already arrived at 3284 the old Proxy/Server while others are en route to the new Proxy/ 3285 Server, however no special attention in the reassembly algorithm is 3286 necessary since all carrier packets will eventually arrive at the 3287 Client which can then reassemble. However, any carrier packets that 3288 are somehow lost can often be recovered through retransmissions. 3290 3.17. Multicast 3292 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 3293 [RFC3810] proxy service for its EUNs and/or hosted applications 3294 [RFC4605]. The Client forwards IGMP/MLD messages over any of its 3295 underlying interfaces for which group membership is required. The 3296 IGMP/MLD messages may be further forwarded by a first-hop ANET access 3297 router acting as an IGMP/MLD-snooping switch [RFC4541], then 3298 ultimately delivered to an AERO Proxy/Server acting as a Protocol 3299 Independent Multicast - Sparse-Mode (PIM-SM, or simply "PIM") 3300 Designated Router (DR) [RFC7761]. AERO Relays also act as PIM 3301 routers (i.e., the same as AERO Proxys/Servers) on behalf of nodes on 3302 INET/EUN networks. The behaviors identified in the following 3303 sections correspond to Source-Specific Multicast (SSM) and Any-Source 3304 Multicast (ASM) operational modes. 3306 3.17.1. Source-Specific Multicast (SSM) 3308 When an ROS "X" acting as PIM router receives a Join/Prune message 3309 from a node on its downstream interfaces containing one or more 3310 ((S)ource, (G)roup) pairs, it updates its Multicast Routing 3311 Information Base (MRIB) accordingly. For each S belonging to a 3312 prefix reachable via X's non-OMNI interfaces, X then forwards the (S, 3313 G) Join/Prune to any PIM routers on those interfaces per [RFC7761]. 3315 For each S belonging to a prefix reachable via X's OMNI interface, X 3316 sends an NS(AR) message (see: Section 3.14) using its own LLA as the 3317 source address and the LLA of S as the destination address. X then 3318 encapsulates the NS(AR) in an OAL header with source address set to 3319 the ULA of X and destination address set to the solicited node 3320 multicast address for S, then forwards the message into the secured 3321 spanning tree, which delivers it to ROR "Y" that services S. The 3322 resulting NA(AR) will return the LLA for the prefix that matches S as 3323 the network-layer source address and with an OMNI option with 3324 interface attributes for any underlying interfaces that are currently 3325 servicing S. 3327 When X processes the NA(AR) it selects one or more underlying 3328 interfaces for S and performs an NS/NA(WIN) exchange while including 3329 a PIM Join/Prune message for each multicast group of interest in the 3330 OMNI option. If S is located behind any Proxys "Z"*, each Z* then 3331 updates its MRIB accordingly and maintains the LLA of X as the next 3332 hop in the reverse path. Since the Bridges do not examine network 3333 layer control messages, this means that the (reverse) multicast tree 3334 path is simply from each Z* (and/or S) to X with no other multicast- 3335 aware routers in the path. 3337 Following the initial combined Join/Prune and NS/NA messaging, X 3338 maintains a NCE for each S the same as if X was sending unicast data 3339 traffic to S. In particular, X performs additional NS/NA exchanges 3340 to keep the NCE alive for up to t_periodic seconds [RFC7761]. If no 3341 new Joins are received within t_periodic seconds, X allows the NCE to 3342 expire. Finally, if X receives any additional Join/Prune messages 3343 for (S,G) it forwards the messages over the secured spanning tree. 3345 At some later time, Client C that holds an MNP for source S may 3346 depart from a first Proxy/Server Z1 and/or connect via a new Proxy/ 3347 Server Z2. In that case, Y sends a uNA message to X the same as 3348 specified for unicast mobility in Section 3.16. When X receives the 3349 uNA message, it updates its NCE for the LLA for source S and sends 3350 new Join messages to any new Proxys Z2. There is no requirement to 3351 send any Prune messages to old Proxy/Server Z1 since source S will no 3352 longer source any multicast data traffic via Z1. Instead, the 3353 multicast state for (S,G) in Proxy/Server Z1 will soon time out since 3354 no new Joins will arrive. 3356 After some later time, C may move to a new Proxy/Server Y2 and depart 3357 from old Sever Y1. In that case, Y1 sends Join messages for any of 3358 C's active (S,G) groups to Y2 while including its own LLA as the 3359 source address. This causes Y2 to include Y1 in the multicast 3360 forwarding tree during the interim time that Y1's NCE for C is in the 3361 DEPARTED state. At the same time, Y1 sends a uNA message to X with 3362 an OMNI option with S/T-omIndex set to 0 and a release indication to 3363 cause X to release its NCE for S. X then sends a new Join message to 3364 S via the secured spanning tree and re-initiates route optimization 3365 the same as if it were receiving a fresh Join message from a node on 3366 a downstream link. 3368 3.17.2. Any-Source Multicast (ASM) 3370 When an ROS X acting as a PIM router receives a Join/Prune from a 3371 node on its downstream interfaces containing one or more (*,G) pairs, 3372 it updates its Multicast Routing Information Base (MRIB) accordingly. 3373 X then forwards a copy of the message within the OMNI option of an 3374 NS(WIN) message to the Rendezvous Point (RP) R for each G over the 3375 secured spanning tree. X uses its own LLA as the source address and 3376 the LLA for R as the destination address, then encapsulates the 3377 NS(WIN) message in an OAL header with source address set to the ULA 3378 of X and destination address set to the ULA of R's Proxy/Server then 3379 sends the message into the secured spanning tree. 3381 For each source S that sends multicast traffic to group G via R, the 3382 Proxy/Server Z* for the Client that aggregates S encapsulates the 3383 original IP packets in PIM Register messages and forwards them to R 3384 via the secured spanning tree, which may then elect to send a PIM 3385 Join to Z*. This will result in an (S,G) tree rooted at Z* with R as 3386 the next hop so that R will begin to receive two copies of the 3387 original IP packet; one native copy from the (S, G) tree and a second 3388 copy from the pre-existing (*, G) tree that still uses PIM Register 3389 encapsulation. R can then issue a PIM Register-stop message to 3390 suppress the Register-encapsulated stream. At some later time, if C 3391 moves to a new Proxy/Server Z*, it resumes sending original IP 3392 packets via PIM Register encapsulation via the new Z*. 3394 At the same time, as multicast listeners discover individual S's for 3395 a given G, they can initiate an (S,G) Join for each S under the same 3396 procedures discussed in Section 3.17.1. Once the (S,G) tree is 3397 established, the listeners can send (S, G) Prune messages to R so 3398 that multicast original IP packets for group G sourced by S will only 3399 be delivered via the (S, G) tree and not from the (*, G) tree rooted 3400 at R. All mobility considerations discussed for SSM apply. 3402 3.17.3. Bi-Directional PIM (BIDIR-PIM) 3404 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 3405 approach to ASM that treats the Rendezvous Point (RP) as a Designated 3406 Forwarder (DF). Further considerations for BIDIR-PIM are out of 3407 scope. 3409 3.18. Operation over Multiple OMNI Links 3411 An AERO Client can connect to multiple OMNI links the same as for any 3412 data link service. In that case, the Client maintains a distinct 3413 OMNI interface for each link, e.g., 'omni0' for the first link, 3414 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 3415 would include its own distinct set of Bridges and Proxy/Servers, 3416 thereby providing redundancy in case of failures. 3418 Each OMNI link could utilize the same or different ANET connections. 3419 The links can be distinguished at the link-layer via the SRT prefix 3420 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 3421 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 3422 MSPs on each link. This gives rise to the opportunity for supporting 3423 multiple redundant networked paths, with each VLAN distinguished by a 3424 different SRT "color" (see: Section 3.2.5). 3426 The Client's IP layer can select the outgoing OMNI interface 3427 appropriate for a given traffic profile while (in the reverse 3428 direction) correspondent nodes must have some way of steering their 3429 original IP packets destined to a target via the correct OMNI link. 3431 In a first alternative, if each OMNI link services different MSPs, 3432 then the Client can receive a distinct MNP from each of the links. 3433 IP routing will therefore assure that the correct OMNI link is used 3434 for both outbound and inbound traffic. This can be accomplished 3435 using existing technologies and approaches, and without requiring any 3436 special supporting code in correspondent nodes or Bridges. 3438 In a second alternative, if each OMNI link services the same MSP(s) 3439 then each link could assign a distinct "OMNI link Anycast" address 3440 that is configured by all Bridges on the link. Correspondent nodes 3441 can then perform Segment Routing to select the correct SRT, which 3442 will then direct the original IP packet over multiple hops to the 3443 target. 3445 3.19. DNS Considerations 3447 AERO Client MNs and INET correspondent nodes consult the Domain Name 3448 System (DNS) the same as for any Internetworking node. When 3449 correspondent nodes and Client MNs use different IP protocol versions 3450 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 3451 A records for IPv4 address mappings to MNs which must then be 3452 populated in Relay NAT64 mapping caches. In that way, an IPv4 3453 correspondent node can send original IPv4 packets to the IPv4 address 3454 mapping of the target MN, and the Relay will translate the IPv4 3455 header and destination address into an IPv6 header and IPv6 3456 destination address of the MN. 3458 When an AERO Client registers with an AERO Proxy/Server, the Proxy/ 3459 Server can return the address(es) of DNS servers in RDNSS options 3460 [RFC6106]. The DNS server provides the IP addresses of other MNs and 3461 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 3463 3.20. Transition/Coexistence Considerations 3465 OAL encapsulation ensures that dissimilar INET partitions can be 3466 joined into a single unified OMNI link, even though the partitions 3467 themselves may have differing protocol versions and/or incompatible 3468 addressing plans. However, a commonality can be achieved by 3469 incrementally distributing globally routable (i.e., native) IP 3470 prefixes to eventually reach all nodes (both mobile and fixed) in all 3471 OMNI link segments. This can be accomplished by incrementally 3472 deploying AERO Bridges on each INET partition, with each Bridge 3473 distributing its MNPs and/or discovering non-MNP IP GUA prefixes on 3474 its INET links. 3476 This gives rise to the opportunity to eventually distribute native IP 3477 addresses to all nodes, and to present a unified OMNI link view even 3478 if the INET partitions remain in their current protocol and 3479 addressing plans. In that way, the OMNI link can serve the dual 3480 purpose of providing a mobility/multilink service and a transition/ 3481 coexistence service. Or, if an INET partition is transitioned to a 3482 native IP protocol version and addressing scheme that is compatible 3483 with the OMNI link MNP-based addressing scheme, the partition and 3484 OMNI link can be joined by Bridges. 3486 Relays that connect INETs/EUNs with dissimilar IP protocol versions 3487 may need to employ a network address and protocol translation 3488 function such as NAT64 [RFC6146]. 3490 3.21. Detecting and Reacting to Proxy/Server and Bridge Failures 3492 In environments where rapid failure recovery is required, Proxy/ 3493 Servers and Bridges SHOULD use Bidirectional Forwarding Detection 3494 (BFD) [RFC5880]. Nodes that use BFD can quickly detect and react to 3495 failures so that cached information is re-established through 3496 alternate nodes. BFD control messaging is carried only over well- 3497 connected ground domain networks (i.e., and not low-end radio links) 3498 and can therefore be tuned for rapid response. 3500 Proxy/Servers and Bridges maintain BFD sessions in parallel with 3501 their BGP peerings. If a Proxy/Server or Bridge fails, BGP peers 3502 will quickly re-establish routes through alternate paths the same as 3503 for common BGP deployments. Similarly, Proxys maintain BFD sessions 3504 with their associated Bridges even though they do not establish BGP 3505 peerings with them. 3507 3.22. AERO Clients on the Open Internet 3509 AERO Clients that connect to the open Internet via INET interfaces 3510 can establish a VPN or direct link to securely connect to a Proxy/ 3511 Server in a "tethered" arrangement with all of the Client's traffic 3512 transiting the Proxy/Server which acts as a router. Alternatively, 3513 the Client can associate with an INET Proxy/Server using UDP/IP 3514 encapsulation and control message securing services as discussed in 3515 the following sections. 3517 When a Client's OMNI interface enables an INET underlying interface, 3518 it first examines the INET address. For IPv4, the Client assumes it 3519 is on the open Internet if the INET address is not a special-use IPv4 3520 address per [RFC3330]. Similarly for IPv6, the Client assumes it is 3521 on the open Internet if the INET address is a Global Unicast Address 3522 (GUA) [RFC4291]. Otherwise, the Client should assume it is behind 3523 one or several NATs. 3525 The Client then prepares an RS message with IPv6 source address set 3526 to its MNP-LLA, with IPv6 destination set to (link-local) All-Routers 3527 multicast and with an OMNI option with underlying interface 3528 attributes. If the Client believes that it is on the open Internet, 3529 it SHOULD include an L2ADDR in the Interface Attributes sub-option 3530 corresponding to the underlying interface; otherwise, it MAY set 3531 L2ADDR to 0. If the underlying address is IPv4, the Client includes 3532 the Port Number and IPv4 address written in obfuscated form [RFC4380] 3533 as discussed in Section 3.3. If the underlying interface address is 3534 IPv6, the Client instead includes the Port Number and IPv6 address in 3535 obfuscated form. The Client finally includes an authentication 3536 signature sub-option in the OMNI option [I-D.templin-6man-omni] to 3537 provide message authentication, selects an Identification value and 3538 window synchronization parameters, and submits the RS for OAL 3539 encapsulation. The Client then encapsulates the OAL fragment in UDP/ 3540 IP headers to form a carrier packet, sets the UDP/IP source to its 3541 INET address and UDP port, sets the UDP/IP destination to the Proxy/ 3542 Server's INET address and the AERO service port number (8060), then 3543 sends the carrier packet to the Proxy/Server. 3545 When the Proxy/Server receives the RS, it discards the OAL 3546 encapsulation, authenticates the RS message, creates a NCE and 3547 registers the Client's MNP, window synchronization state and INET 3548 interface information according to the OMNI option parameters. If 3549 the RS message OMNI option includes Interface Attributes with an 3550 L2ADDR, the Proxy/Server compares the encapsulation IP address and 3551 UDP port number with the (unobfuscated) values. If the values are 3552 the same, the Proxy/Server caches the Client's information as "INET" 3553 addresses meaning that the Client is likely to accept direct messages 3554 without requiring NAT traversal exchanges. If the values are 3555 different (or, if the OMNI option did not include an L2ADDR) the 3556 Proxy/Server instead caches the Client's information as "mapped" 3557 addresses meaning that NAT traversal exchanges may be necessary. 3559 The Proxy/Server then prepares an RA message with IPv6 source and 3560 destination set corresponding to the addresses in the RS, and with an 3561 OMNI option with an Origin Indication sub-option per 3562 [I-D.templin-6man-omni] with the mapped and obfuscated Port Number 3563 and IP address observed in the encapsulation headers. The Proxy/ 3564 Server also includes an Interface Attributes sub-option for its 3565 underlying interface with FMT/SRT/LHS information appropriate for its 3566 INET interface, and with an authentication signature sub-option per 3567 [I-D.templin-6man-omni] and/or a symmetric window synchronization/ 3568 acknowledgement if necessary. The Proxy/Server then performs OAL 3569 encapsulation and fragmentation if necessary and encapsulates each 3570 fragment in UDP/IP headers with addresses set per the L2ADDR 3571 information in the NCE for the Client. 3573 When the Client receives the RA, it authenticates the message then 3574 process the window synchronization/acknowledgement and compares the 3575 mapped Port Number and IP address from the Origin Indication sub- 3576 option with its own address. If the addresses are the same, the 3577 Client assumes the open Internet / Cone NAT principle; if the 3578 addresses are different, the Client instead assumes that further 3579 qualification procedures are necessary to detect the type of NAT and 3580 proceeds according to standard procedures [RFC6081][RFC4380]. The 3581 Client also caches the RA Interface Attributes FMT/SRT/LHS 3582 information to discover the Proxy/Server's spanning tree orientation. 3583 The Client finally arranges to return an explicit/implicit 3584 acknowledgement, and sends periodic RS messages to receive fresh RA 3585 messages before the Router Lifetime received on each INET interface 3586 expires. 3588 When the Client sends messages to target IP addresses, it also 3589 invokes route optimization per Section 3.14. For route optimized 3590 targets in the same OMNI link segment, if the target's L2ADDR is on 3591 the open INET, the Client forwards carrier packets directly to the 3592 target INET address. If the target is behind a NAT, the Client first 3593 establishes NAT state for the L2ADDR using the "direct bubble" and 3594 NUD mechanisms discussed in Section 3.10.1. The Client continues to 3595 send carrier packets via its Proxy/Server until NAT state is 3596 populated, then begins forwarding carrier packets via the direct path 3597 through the NAT to the target. For targets in different OMNI link 3598 segments, the Client uses OAL/ORH encapsulation and forwards carrier 3599 packets to the Bridge that returned the NA(AR) message. 3601 The Client can send original IP packets to route-optimized neighbors 3602 in the same OMNI link segment no larger than the minimum/path MPS in 3603 one piece and with OAL encapsulation as atomic fragments. For larger 3604 original IP packets, the Client applies OAL encapsulation and 3605 fragmentation if necessary according to Section 3.9, with OAL header 3606 with source set to its own MNP-ULA and destination set to the MNP-ULA 3607 of the target, and with an in-window Identification value. The 3608 Client then encapsulates each resulting carrier packet in UDP/IP *NET 3609 headers and sends them to the next hop. 3611 Note: The NAT traversal procedures specified in this document are 3612 applicable for Cone, Address-Restricted and Port-Restricted NATs 3613 only. While future updates to this document may specify procedures 3614 for other NAT variations (e.g., hairpinning and various forms of 3615 Symmetric NATs), it should be noted that continuous communications 3616 are always possible through forwarding via a Proxy/Server even if NAT 3617 traversal is not employed. 3619 3.23. Time-Varying MNPs 3621 In some use cases, it is desirable, beneficial and efficient for the 3622 Client to receive a constant MNP that travels with the Client 3623 wherever it moves. For example, this would allow air traffic 3624 controllers to easily track aircraft, etc. In other cases, however 3625 (e.g., intelligent transportation systems), the MN may be willing to 3626 sacrifice a modicum of efficiency in order to have time-varying MNPs 3627 that can be changed every so often to defeat adversarial tracking. 3629 The DHCPv6 service offers a way for Clients that desire time-varying 3630 MNPs to obtain short-lived prefixes (e.g., on the order of a small 3631 number of minutes). In that case, the identity of the Client would 3632 not be bound to the MNP but rather to a Node Identification value 3633 (see: [I-D.templin-6man-omni]) to be used as the Client ID seed for 3634 MNP prefix delegation. The Client would then be obligated to 3635 renumber its internal networks whenever its MNP (and therefore also 3636 its MNP-LLA) changes. This should not present a challenge for 3637 Clients with automated network renumbering services, however presents 3638 limits for the durations of ongoing sessions that would prefer to use 3639 a constant address. 3641 4. Implementation Status 3643 An early AERO implementation based on OpenVPN (https://openvpn.net/) 3644 was announced on the v6ops mailing list on January 10, 2018 and an 3645 initial public release of the AERO proof-of-concept source code was 3646 announced on the intarea mailing list on August 21, 2015. 3648 AERO Release-3.2 was tagged on March 30, 2021, and is undergoing 3649 internal testing. Additional internal releases expected within the 3650 coming months, with first public release expected end of 1H2021. 3652 5. IANA Considerations 3654 The IANA is instructed to assign a new type value TBD1 in the IPv6 3655 Routing Types registry (IANA registration procedure is IETF Review or 3656 IESG Approval). 3658 The IANA has assigned the UDP port number "8060" for an earlier 3659 experimental first version of AERO [RFC6706]. This document 3660 obsoletes [RFC6706], and together with [I-D.templin-6man-omni] 3661 reclaims the UDP port number "8060" for 'aero' as the service port 3662 for UDP/IP encapsulation. (Note that, although [RFC6706] was not 3663 widely implemented or deployed, any messages coded to that 3664 specification can be easily distinguished and ignored since they use 3665 the invalid ICMPv6 message type number '0'.) This document makes no 3666 request of IANA, since [I-D.templin-6man-omni] already provides 3667 instructions. 3669 No further IANA actions are required. 3671 6. Security Considerations 3673 AERO Bridges configure secured tunnels with AERO Proxy/Servers and 3674 Relays within their local OMNI link segments. Applicable secured 3675 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3676 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3677 segments in turn configure secured tunnels for their neighboring AERO 3678 Bridges in a secured spanning tree topology. Therefore, control 3679 messages exchanged between any pair of OMNI link neighbors over the 3680 secured spanning tree are already protected. 3682 To prevent spoofing vectors, Proxy/Servers MUST discard without 3683 responding to any unsecured NS(AR) messages. Also, Proxy/Servers 3684 MUST discard without forwarding any original IP packets received from 3685 one of their own Clients (whether directly or following OAL 3686 reassembly) with a source address that does not match the Client's 3687 MNP and/or a destination address that does match the Client's MNP. 3688 Finally, Proxy/Servers MUST discard without forwarding any carrier 3689 packets with an OAL source and destination that both match the same 3690 MNP (i.e., after consulting the ORH if present). 3692 For INET partitions that require strong security in the data plane, 3693 two options for securing communications include 1) disable route 3694 optimization so that all traffic is conveyed over secured tunnels, or 3695 2) enable on-demand secure tunnel creation between Client neighbors. 3696 Option 1) would result in longer routes than necessary and impose 3697 traffic concentration on critical infrastructure elements. Option 2) 3698 could be coordinated between Clients using NS/NA messages with OMNI 3699 Host Identity Protocol (HIP) "Initiator/Responder" message sub- 3700 options [RFC7401][I-D.templin-6man-omni] to create a secured tunnel 3701 on-demand. 3703 AERO Clients that connect to secured ANETs need not apply security to 3704 their ND messages, since the messages will be authenticated and 3705 forwarded by a perimeter Proxy/Server that applies security on its 3706 INET-facing interface as part of the spanning tree (see above). AERO 3707 Clients connected to the open INET can use network and/or transport 3708 layer security services such as VPNs or can by some other means 3709 establish a direct link to a Proxy/Server. When a VPN or direct link 3710 may be impractical, however, INET Clients and Proxy/Servers SHOULD 3711 include and verify authentication signatures for their IPv6 ND 3712 messages as specified in [I-D.templin-6man-omni]. 3714 Application endpoints SHOULD use transport-layer (or higher-layer) 3715 security services such as TLS/SSL, DTLS or SSH [RFC4251] to assure 3716 the same level of protection as for critical secured Internet 3717 services. AERO Clients that require host-based VPN services SHOULD 3718 use network and/or transport layer security services such as IPsec, 3719 TLS/SSL, DTLS, etc. AERO Proxys and Proxy/Servers can also provide a 3720 network-based VPN service on behalf of the Client, e.g., if the 3721 Client is located within a secured enclave and cannot establish a VPN 3722 on its own behalf. 3724 AERO Proxy/Servers and Bridges present targets for traffic 3725 amplification Denial of Service (DoS) attacks. This concern is no 3726 different than for widely-deployed VPN security gateways in the 3727 Internet, where attackers could send spoofed packets to the gateways 3728 at high data rates. This can be mitigated through the AERO/OMNI data 3729 origin authentication procedures, as well as connecting Proxy/Servers 3730 and Bridges over dedicated links with no connections to the Internet 3731 and/or when connections to the Internet are only permitted through 3732 well-managed firewalls. Traffic amplification DoS attacks can also 3733 target an AERO Client's low data rate links. This is a concern not 3734 only for Clients located on the open Internet but also for Clients in 3735 secured enclaves. AERO Proxy/Servers and Proxys can institute rate 3736 limits that protect Clients from receiving packet floods that could 3737 DoS low data rate links. 3739 AERO Relays must implement ingress filtering to avoid a spoofing 3740 attack in which spurious messages with ULA addresses are injected 3741 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3742 that their connectivity is not used by unauthorized nodes on their 3743 EUNs to gain access to a protected network, i.e., AERO Clients that 3744 act as routers MUST NOT provide routing services for unauthorized 3745 nodes. (This concern is no different than for ordinary hosts that 3746 receive an IP address delegation but then "share" the address with 3747 other nodes via some form of Internet connection sharing such as 3748 tethering.) 3750 The MAP list MUST be well-managed and secured from unauthorized 3751 tampering, even though the list contains only public information. 3752 The MAP list can be conveyed to the Client in a similar fashion as in 3753 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3754 upload of a static file, DNS lookups, etc.). 3756 The AERO service for open INET Clients depends on a public key 3757 distribution service in which Client public keys and identities are 3758 maintained in a shared database accessible to all open INET Proxy/ 3759 Servers. Similarly, each Client must be able to determine the public 3760 key of each Proxy/Server, e.g. by consulting an online database. 3761 When AERO nodes register their public keys indexed by a unique Host 3762 Identity Tag (HIT) [RFC7401] in a distributed database such as the 3763 DNS, and use the HIT as an identity for applying IPv6 ND message 3764 authentication signatures, a means for determining public key 3765 attestation is available. 3767 Security considerations for IPv6 fragmentation and reassembly are 3768 discussed in [I-D.templin-6man-omni]. In environments where spoofing 3769 is considered a threat, OMNI nodes SHOULD employ Identification 3770 window synchronization and OAL destinations SHOULD configure an (end- 3771 system-based) firewall. 3773 SRH authentication facilities are specified in [RFC8754]. Security 3774 considerations for accepting link-layer ICMP messages and reflected 3775 packets are discussed throughout the document. 3777 7. Acknowledgements 3779 Discussions in the IETF, aviation standards communities and private 3780 exchanges helped shape some of the concepts in this work. 3781 Individuals who contributed insights include Mikael Abrahamsson, Mark 3782 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3783 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3784 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3785 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3786 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3787 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3788 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3789 Wood and James Woodyatt. Members of the IESG also provided valuable 3790 input during their review process that greatly improved the document. 3791 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3792 for their shepherding guidance during the publication of the AERO 3793 first edition. 3795 This work has further been encouraged and supported by Boeing 3796 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3797 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3798 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3799 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3800 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3801 Gene MacLean III, Kyle Mikos, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3802 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3803 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3804 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3805 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3806 Benson, Madhuri Madhava Badgandi, Vijayasarathy Rajagopalan, Katie 3807 Tran and Eric Yeh are especially acknowledged for implementing the 3808 AERO functions as extensions to the public domain OpenVPN 3809 distribution. Chuck Klabunde is honored and remembered for his early 3810 leadership, and we mourn his untimely loss. 3812 Earlier works on NBMA tunneling approaches are found in 3813 [RFC2529][RFC5214][RFC5569]. 3815 Many of the constructs presented in this second edition of AERO are 3816 based on the author's earlier works, including: 3818 o The Internet Routing Overlay Network (IRON) 3819 [RFC6179][I-D.templin-ironbis] 3821 o Virtual Enterprise Traversal (VET) 3822 [RFC5558][I-D.templin-intarea-vet] 3824 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3825 [RFC5320][I-D.templin-intarea-seal] 3827 o AERO, First Edition [RFC6706] 3829 Note that these works cite numerous earlier efforts that are not also 3830 cited here due to space limitations. The authors of those earlier 3831 works are acknowledged for their insights. 3833 This work is aligned with the NASA Safe Autonomous Systems Operation 3834 (SASO) program under NASA contract number NNA16BD84C. 3836 This work is aligned with the FAA as per the SE2025 contract number 3837 DTFAWA-15-D-00030. 3839 This work is aligned with the Boeing Commercial Airplanes (BCA) 3840 Internet of Things (IoT) and autonomy programs. 3842 This work is aligned with the Boeing Information Technology (BIT) 3843 MobileNet program. 3845 8. References 3847 8.1. Normative References 3849 [I-D.templin-6man-omni] 3850 Templin, F. L. and T. Whyman, "Transmission of IP Packets 3851 over Overlay Multilink Network (OMNI) Interfaces", draft- 3852 templin-6man-omni-03 (work in progress), April 2021. 3854 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3855 DOI 10.17487/RFC0791, September 1981, 3856 . 3858 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3859 RFC 792, DOI 10.17487/RFC0792, September 1981, 3860 . 3862 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3863 Requirement Levels", BCP 14, RFC 2119, 3864 DOI 10.17487/RFC2119, March 1997, 3865 . 3867 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3868 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3869 December 1998, . 3871 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3872 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3873 DOI 10.17487/RFC3971, March 2005, 3874 . 3876 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3877 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3878 . 3880 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3881 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3882 November 2005, . 3884 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3885 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3886 . 3888 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3889 Network Address Translations (NATs)", RFC 4380, 3890 DOI 10.17487/RFC4380, February 2006, 3891 . 3893 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3894 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3895 DOI 10.17487/RFC4861, September 2007, 3896 . 3898 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3899 Address Autoconfiguration", RFC 4862, 3900 DOI 10.17487/RFC4862, September 2007, 3901 . 3903 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3904 DOI 10.17487/RFC6081, January 2011, 3905 . 3907 [RFC7401] Moskowitz, R., Ed., Heer, T., Jokela, P., and T. 3908 Henderson, "Host Identity Protocol Version 2 (HIPv2)", 3909 RFC 7401, DOI 10.17487/RFC7401, April 2015, 3910 . 3912 [RFC7739] Gont, F., "Security Implications of Predictable Fragment 3913 Identification Values", RFC 7739, DOI 10.17487/RFC7739, 3914 February 2016, . 3916 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3917 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3918 May 2017, . 3920 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3921 (IPv6) Specification", STD 86, RFC 8200, 3922 DOI 10.17487/RFC8200, July 2017, 3923 . 3925 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3926 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 3927 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 3928 RFC 8415, DOI 10.17487/RFC8415, November 2018, 3929 . 3931 8.2. Informative References 3933 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 3934 2016. 3936 [I-D.bonica-6man-comp-rtg-hdr] 3937 Bonica, R., Kamite, Y., Alston, A., Henriques, D., and L. 3938 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 3939 bonica-6man-comp-rtg-hdr-24 (work in progress), January 3940 2021. 3942 [I-D.bonica-6man-crh-helper-opt] 3943 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 3944 Routing Header (CRH) Helper Option", draft-bonica-6man- 3945 crh-helper-opt-03 (work in progress), April 2021. 3947 [I-D.ietf-intarea-frag-fragile] 3948 Bonica, R., Baker, F., Huston, G., Hinden, R. M., Troan, 3949 O., and F. Gont, "IP Fragmentation Considered Fragile", 3950 draft-ietf-intarea-frag-fragile-17 (work in progress), 3951 September 2019. 3953 [I-D.ietf-intarea-tunnels] 3954 Touch, J. and M. Townsley, "IP Tunnels in the Internet 3955 Architecture", draft-ietf-intarea-tunnels-10 (work in 3956 progress), September 2019. 3958 [I-D.ietf-ipwave-vehicular-networking] 3959 (editor), J. (. J., "IPv6 Wireless Access in Vehicular 3960 Environments (IPWAVE): Problem Statement and Use Cases", 3961 draft-ietf-ipwave-vehicular-networking-20 (work in 3962 progress), March 2021. 3964 [I-D.ietf-rtgwg-atn-bgp] 3965 Templin, F. L., Saccone, G., Dawra, G., Lindem, A., and V. 3966 Moreno, "A Simple BGP-based Mobile Routing System for the 3967 Aeronautical Telecommunications Network", draft-ietf- 3968 rtgwg-atn-bgp-10 (work in progress), January 2021. 3970 [I-D.templin-6man-dhcpv6-ndopt] 3971 Templin, F. L., "A Unified Stateful/Stateless 3972 Configuration Service for IPv6", draft-templin-6man- 3973 dhcpv6-ndopt-11 (work in progress), January 2021. 3975 [I-D.templin-intarea-seal] 3976 Templin, F. L., "The Subnetwork Encapsulation and 3977 Adaptation Layer (SEAL)", draft-templin-intarea-seal-68 3978 (work in progress), January 2014. 3980 [I-D.templin-intarea-vet] 3981 Templin, F. L., "Virtual Enterprise Traversal (VET)", 3982 draft-templin-intarea-vet-40 (work in progress), May 2013. 3984 [I-D.templin-ipwave-uam-its] 3985 Templin, F. L., "Urban Air Mobility Implications for 3986 Intelligent Transportation Systems", draft-templin-ipwave- 3987 uam-its-04 (work in progress), January 2021. 3989 [I-D.templin-ironbis] 3990 Templin, F. L., "The Interior Routing Overlay Network 3991 (IRON)", draft-templin-ironbis-16 (work in progress), 3992 March 2014. 3994 [I-D.templin-v6ops-pdhost] 3995 Templin, F. L., "IPv6 Prefix Delegation and Multi- 3996 Addressing Models", draft-templin-v6ops-pdhost-27 (work in 3997 progress), January 2021. 3999 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 4001 [RFC1035] Mockapetris, P., "Domain names - implementation and 4002 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 4003 November 1987, . 4005 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 4006 RFC 1812, DOI 10.17487/RFC1812, June 1995, 4007 . 4009 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 4010 DOI 10.17487/RFC2003, October 1996, 4011 . 4013 [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, 4014 DOI 10.17487/RFC2004, October 1996, 4015 . 4017 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 4018 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 4019 . 4021 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 4022 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 4023 . 4025 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 4026 Domains without Explicit Tunnels", RFC 2529, 4027 DOI 10.17487/RFC2529, March 1999, 4028 . 4030 [RFC2983] Black, D., "Differentiated Services and Tunnels", 4031 RFC 2983, DOI 10.17487/RFC2983, October 2000, 4032 . 4034 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 4035 of Explicit Congestion Notification (ECN) to IP", 4036 RFC 3168, DOI 10.17487/RFC3168, September 2001, 4037 . 4039 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 4040 DOI 10.17487/RFC3330, September 2002, 4041 . 4043 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 4044 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 4045 DOI 10.17487/RFC3810, June 2004, 4046 . 4048 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 4049 Unique IDentifier (UUID) URN Namespace", RFC 4122, 4050 DOI 10.17487/RFC4122, July 2005, 4051 . 4053 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 4054 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 4055 January 2006, . 4057 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 4058 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 4059 DOI 10.17487/RFC4271, January 2006, 4060 . 4062 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 4063 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 4064 2006, . 4066 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 4067 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 4068 December 2005, . 4070 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 4071 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 4072 2006, . 4074 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 4075 Control Message Protocol (ICMPv6) for the Internet 4076 Protocol Version 6 (IPv6) Specification", STD 89, 4077 RFC 4443, DOI 10.17487/RFC4443, March 2006, 4078 . 4080 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 4081 Protocol (LDAP): The Protocol", RFC 4511, 4082 DOI 10.17487/RFC4511, June 2006, 4083 . 4085 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 4086 "Considerations for Internet Group Management Protocol 4087 (IGMP) and Multicast Listener Discovery (MLD) Snooping 4088 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 4089 . 4091 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 4092 "Internet Group Management Protocol (IGMP) / Multicast 4093 Listener Discovery (MLD)-Based Multicast Forwarding 4094 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 4095 August 2006, . 4097 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 4098 Algorithms in Cryptographically Generated Addresses 4099 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 4100 . 4102 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 4103 "Bidirectional Protocol Independent Multicast (BIDIR- 4104 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 4105 . 4107 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 4108 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 4109 DOI 10.17487/RFC5214, March 2008, 4110 . 4112 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 4113 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 4114 February 2010, . 4116 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 4117 Route Optimization Requirements for Operational Use in 4118 Aeronautics and Space Exploration Mobile Networks", 4119 RFC 5522, DOI 10.17487/RFC5522, October 2009, 4120 . 4122 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 4123 RFC 5558, DOI 10.17487/RFC5558, February 2010, 4124 . 4126 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 4127 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 4128 January 2010, . 4130 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 4131 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 4132 . 4134 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 4135 "IPv6 Router Advertisement Options for DNS Configuration", 4136 RFC 6106, DOI 10.17487/RFC6106, November 2010, 4137 . 4139 [RFC6139] Russert, S., Ed., Fleischman, E., Ed., and F. Templin, 4140 Ed., "Routing and Addressing in Networks with Global 4141 Enterprise Recursion (RANGER) Scenarios", RFC 6139, 4142 DOI 10.17487/RFC6139, February 2011, 4143 . 4145 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 4146 NAT64: Network Address and Protocol Translation from IPv6 4147 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 4148 April 2011, . 4150 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 4151 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 4152 . 4154 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 4155 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 4156 DOI 10.17487/RFC6221, May 2011, 4157 . 4159 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 4160 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 4161 DOI 10.17487/RFC6273, June 2011, 4162 . 4164 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 4165 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 4166 January 2012, . 4168 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 4169 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 4170 DOI 10.17487/RFC6355, August 2011, 4171 . 4173 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 4174 for Equal Cost Multipath Routing and Link Aggregation in 4175 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 4176 . 4178 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 4179 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 4180 . 4182 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 4183 UDP Checksums for Tunneled Packets", RFC 6935, 4184 DOI 10.17487/RFC6935, April 2013, 4185 . 4187 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 4188 for the Use of IPv6 UDP Datagrams with Zero Checksums", 4189 RFC 6936, DOI 10.17487/RFC6936, April 2013, 4190 . 4192 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 4193 Korhonen, "Requirements for Distributed Mobility 4194 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 4195 . 4197 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 4198 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 4199 Multicast - Sparse Mode (PIM-SM): Protocol Specification 4200 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 4201 2016, . 4203 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 4204 Decraene, B., Litkowski, S., and R. Shakir, "Segment 4205 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 4206 July 2018, . 4208 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 4209 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 4210 . 4212 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 4213 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 4214 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 4215 . 4217 [WG] Wireguard, "WireGuard, https://www.wireguard.com", August 4218 2020. 4220 Appendix A. Non-Normative Considerations 4222 AERO can be applied to a multitude of Internetworking scenarios, with 4223 each having its own adaptations. The following considerations are 4224 provided as non-normative guidance: 4226 A.1. Implementation Strategies for Route Optimization 4228 Route optimization as discussed in Section 3.14 results in the route 4229 optimization source (ROS) creating a NCE for the target neighbor. 4230 The NCE state is set to REACHABLE for at most ReachableTime seconds. 4231 In order to refresh the NCE lifetime before the ReachableTime timer 4232 expires, the specification requires implementations to issue a new 4233 NS/NA exchange to reset ReachableTime while data packets are still 4234 flowing. However, the decision of when to initiate a new NS/NA 4235 exchange and to perpetuate the process is left as an implementation 4236 detail. 4238 One possible strategy may be to monitor the NCE watching for data 4239 packets for (ReachableTime - 5) seconds. If any data packets have 4240 been sent to the neighbor within this timeframe, then send an NS to 4241 receive a new NA. If no data packets have been sent, wait for 5 4242 additional seconds and send an immediate NS if any data packets are 4243 sent within this "expiration pending" 5 second window. If no 4244 additional data packets are sent within the 5 second window, reset 4245 the NCE state to STALE. 4247 The monitoring of the neighbor data packet traffic therefore becomes 4248 an ongoing process during the NCE lifetime. If the NCE expires, 4249 future data packets will trigger a new NS/NA exchange while the 4250 packets themselves are delivered over a longer path until route 4251 optimization state is re-established. 4253 A.2. Implicit Mobility Management 4255 OMNI interface neighbors MAY provide a configuration option that 4256 allows them to perform implicit mobility management in which no ND 4257 messaging is used. In that case, the Client only transmits packets 4258 over a single interface at a time, and the neighbor always observes 4259 packets arriving from the Client from the same link-layer source 4260 address. 4262 If the Client's underlying interface address changes (either due to a 4263 readdressing of the original interface or switching to a new 4264 interface) the neighbor immediately updates the NCE for the Client 4265 and begins accepting and sending packets according to the Client's 4266 new address. This implicit mobility method applies to use cases such 4267 as cellphones with both WiFi and Cellular interfaces where only one 4268 of the interfaces is active at a given time, and the Client 4269 automatically switches over to the backup interface if the primary 4270 interface fails. 4272 A.3. Direct Underlying Interfaces 4274 When a Client's OMNI interface is configured over a Direct interface, 4275 the neighbor at the other end of the Direct link can receive packets 4276 without any encapsulation. In that case, the Client sends packets 4277 over the Direct link according to traffic selectors. If the Direct 4278 interface is selected, then the Client's IP packets are transmitted 4279 directly to the peer without going through an ANET/INET. If other 4280 interfaces are selected, then the Client's IP packets are transmitted 4281 via a different interface, which may result in the inclusion of 4282 Proxy/Servers and Bridges in the communications path. Direct 4283 interfaces must be tested periodically for reachability, e.g., via 4284 NUD. 4286 A.4. AERO Critical Infrastructure Considerations 4288 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 4289 IP routers or virtual machines in the cloud. Bridges must be 4290 provisioned, supported and managed by the INET administrative 4291 authority, and connected to the Bridges of other INETs via inter- 4292 domain peerings. Cost for purchasing, configuring and managing 4293 Bridges is nominal even for very large OMNI links. 4295 AERO cloud Proxy/Servers can be standard dedicated server platforms, 4296 but most often will be deployed as virtual machines in the cloud. 4297 The only requirements for cloud Proxy/Servers are that they can run 4298 the AERO user-level code and have at least one network interface 4299 connection to the INET. Cloud Proxy/Servers must be provisioned, 4300 supported and managed by the INET administrative authority. Cost for 4301 purchasing, configuring and managing cloud Proxy/Servers is nominal 4302 especially for virtual machines. 4304 AERO ANET Proxy/Servers are most often standard dedicated server 4305 platforms with one underlying interface connected to the ANET and a 4306 second interface connected to an INET. As with cloud Proxy/Servers, 4307 the only requirements are that they can run the AERO user-level code 4308 and have at least one interface connection to the INET. ANET Proxy/ 4309 Servers must be provisioned, supported and managed by the ANET 4310 administrative authority. Cost for purchasing, configuring and 4311 managing Proxys is nominal, and borne by the ANET administrative 4312 authority. 4314 AERO Relays are simply Proxy/Servers connected to INETs and/or EUNs 4315 that provide forwarding services for non-MNP destinations. The Relay 4316 connects to the OMNI link and engages in eBGP peering with one or 4317 more Bridges as a stub AS. The Relay then injects its MNPs and/or 4318 non-MNP prefixes into the BGP routing system, and provisions the 4319 prefixes to its downstream-attached networks. The Relay can perform 4320 ROS/ROR services the same as for any Proxy/Server, and can route 4321 between the MNP and non-MNP address spaces. 4323 A.5. AERO Server Failure Implications 4325 AERO Proxy/Servers may appear as a single point of failure in the 4326 architecture, but such is not the case since all Proxy/Servers on the 4327 link provide identical services and loss of a Proxy/Server does not 4328 imply immediate and/or comprehensive communication failures. Proxy/ 4329 Server failure is quickly detected and conveyed by Bidirectional 4330 Forward Detection (BFD) and/or proactive NUD allowing Clients to 4331 migrate to new Proxy/Servers. 4333 If a Proxy/Server fails, ongoing packet forwarding to Clients will 4334 continue by virtue of the neighbor cache entries that have already 4335 been established in route optimization sources (ROSs). If a Client 4336 also experiences mobility events at roughly the same time the Proxy/ 4337 Server fails, unsolicited NA messages may be lost but neighbor cache 4338 entries in the DEPARTED state will ensure that packet forwarding to 4339 the Client's new locations will continue for up to DepartTime 4340 seconds. 4342 If a Client is left without a Proxy/Server for a considerable length 4343 of time (e.g., greater than ReachableTime seconds) then existing 4344 neighbor cache entries will eventually expire and both ongoing and 4345 new communications will fail. The original source will continue to 4346 retransmit until the Client has established a new Proxy/Server 4347 relationship, after which time continuous communications will resume. 4349 Therefore, providing many Proxy/Servers on the link with high 4350 availability profiles provides resilience against loss of individual 4351 Proxy/Servers and assurance that Clients can establish new Proxy/ 4352 Server relationships quickly in event of a Proxy/Server failure. 4354 A.6. AERO Client / Server Architecture 4356 The AERO architectural model is client / server in the control plane, 4357 with route optimization in the data plane. The same as for common 4358 Internet services, the AERO Client discovers the addresses of AERO 4359 Proxy/Servers and connects to one or more of them. The AERO service 4360 is analogous to common Internet services such as google.com, 4361 yahoo.com, cnn.com, etc. However, there is only one AERO service for 4362 the link and all Proxy/Servers provide identical services. 4364 Common Internet services provide differing strategies for advertising 4365 server addresses to clients. The strategy is conveyed through the 4366 DNS resource records returned in response to name resolution queries. 4367 As of January 2020 Internet-based 'nslookup' services were used to 4368 determine the following: 4370 o When a client resolves the domainname "google.com", the DNS always 4371 returns one A record (i.e., an IPv4 address) and one AAAA record 4372 (i.e., an IPv6 address). The client receives the same addresses 4373 each time it resolves the domainname via the same DNS resolver, 4374 but may receive different addresses when it resolves the 4375 domainname via different DNS resolvers. But, in each case, 4376 exactly one A and one AAAA record are returned. 4378 o When a client resolves the domainname "ietf.org", the DNS always 4379 returns one A record and one AAAA record with the same addresses 4380 regardless of which DNS resolver is used. 4382 o When a client resolves the domainname "yahoo.com", the DNS always 4383 returns a list of 4 A records and 4 AAAA records. Each time the 4384 client resolves the domainname via the same DNS resolver, the same 4385 list of addresses are returned but in randomized order (i.e., 4386 consistent with a DNS round-robin strategy). But, interestingly, 4387 the same addresses are returned (albeit in randomized order) when 4388 the domainname is resolved via different DNS resolvers. 4390 o When a client resolves the domainname "amazon.com", the DNS always 4391 returns a list of 3 A records and no AAAA records. As with 4392 "yahoo.com", the same three A records are returned from any 4393 worldwide Internet connection point in randomized order. 4395 The above example strategies show differing approaches to Internet 4396 resilience and service distribution offered by major Internet 4397 services. The Google approach exposes only a single IPv4 and a 4398 single IPv6 address to clients. Clients can then select whichever IP 4399 protocol version offers the best response, but will always use the 4400 same IP address according to the current Internet connection point. 4401 This means that the IP address offered by the network must lead to a 4402 highly-available server and/or service distribution point. In other 4403 words, resilience is predicated on high availability within the 4404 network and with no client-initiated failovers expected (i.e., it is 4405 all-or-nothing from the client's perspective). However, Google does 4406 provide for worldwide distributed service distribution by virtue of 4407 the fact that each Internet connection point responds with a 4408 different IPv6 and IPv4 address. The IETF approach is like google 4409 (all-or-nothing from the client's perspective), but provides only a 4410 single IPv4 or IPv6 address on a worldwide basis. This means that 4411 the addresses must be made highly-available at the network level with 4412 no client failover possibility, and if there is any worldwide service 4413 distribution it would need to be conducted by a network element that 4414 is reached via the IP address acting as a service distribution point. 4416 In contrast to the Google and IETF philosophies, Yahoo and Amazon 4417 both provide clients with a (short) list of IP addresses with Yahoo 4418 providing both IP protocol versions and Amazon as IPv4-only. The 4419 order of the list is randomized with each name service query 4420 response, with the effect of round-robin load balancing for service 4421 distribution. With a short list of addresses, there is still 4422 expectation that the network will implement high availability for 4423 each address but in case any single address fails the client can 4424 switch over to using a different address. The balance then becomes 4425 one of function in the network vs function in the end system. 4427 The same implications observed for common highly-available services 4428 in the Internet apply also to the AERO client/server architecture. 4429 When an AERO Client connects to one or more ANETs, it discovers one 4430 or more AERO Proxy/Server addresses through the mechanisms discussed 4431 in earlier sections. Each Proxy/Server address presumably leads to a 4432 fault-tolerant clustering arrangement such as supported by Linux-HA, 4433 Extended Virtual Synchrony or Paxos. Such an arrangement has 4434 precedence in common Internet service deployments in lightweight 4435 virtual machines without requiring expensive hardware deployment. 4436 Similarly, common Internet service deployments set service IP 4437 addresses on service distribution points that may relay requests to 4438 many different servers. 4440 For AERO, the expectation is that a combination of the Google/IETF 4441 and Yahoo/Amazon philosophies would be employed. The AERO Client 4442 connects to different ANET access points and can receive 1-2 Proxy/ 4443 Server ADM-LLAs at each point. It then selects one AERO Proxy/Server 4444 address, and engages in RS/RA exchanges with the same Proxy/Server 4445 from all ANET connections. The Client remains with this Proxy/Server 4446 unless or until the Proxy/Server fails, in which case it can switch 4447 over to an alternate Proxy/Server. The Client can likewise switch 4448 over to a different Proxy/Server at any time if there is some reason 4449 for it to do so. So, the AERO expectation is for a balance of 4450 function in the network and end system, with fault tolerance and 4451 resilience at both levels. 4453 Appendix B. Change Log 4455 << RFC Editor - remove prior to publication >> 4457 Changes from draft-templin-6man-aero-10 to draft-templin-6man-aero- 4458 11: 4460 o Final editorial review pass resulting in multiple changes. 4461 Document now submit for final approval (with reference to rfcdiff 4462 from previous version). 4464 Changes from draft-templin-6man-aero-09 to draft-templin-6man-aero- 4465 10: 4467 o Final editorial review pass resulting in multiple changes. 4468 Document now submit for final approval (with reference to rfcdiff 4469 from previous version). 4471 Changes from draft-templin-6man-aero-08 to draft-templin-6man-aero- 4472 09: 4474 o Final editorial review pass resulting in multiple changes. 4475 Document now submit for final approval (with reference to rfcdiff 4476 from previous version). 4478 Changes from draft-templin-6man-aero-07 to draft-templin-6man-aero- 4479 08: 4481 o Final editorial review pass resulting in multiple changes. 4482 Document now submit for final approval (with reference to rfcdiff 4483 from previous version). 4485 Changes from draft-templin-6man-aero-06 to draft-templin-6man-aero- 4486 07: 4488 o Final editorial review pass resulting in multiple changes. 4489 Document now submit for final approval (with reference to rfcdiff 4490 from previous version). 4492 Changes from draft-templin-6man-aero-05 to draft-templin-6man-aero- 4493 06: 4495 o Final editorial review pass resulting in multiple changes. 4496 Document now submit for final approval. 4498 Changes from draft-templin-6man-aero-04 to draft-templin-6man-aero- 4499 05: 4501 o Changed to use traffic selectors instead of the former multilink 4502 selection strategy. 4504 Changes from draft-templin-6man-aero-03 to draft-templin-6man-aero- 4505 04: 4507 o Removed documents from "Obsoletes" list. 4509 o Introduced the concept of "secured" and "unsecured" spanning tree. 4511 o Additional security considerations. 4513 o Additional route optimization considerations. 4515 Changes from draft-templin-6man-aero-02 to draft-templin-6man-aero- 4516 03: 4518 o Support for extended route optimization from ROR to target over 4519 target's underlying interfaces. 4521 Changes from draft-templin-6man-aero-01 to draft-templin-6man-aero- 4522 02: 4524 o Changed reference citations to "draft-templin-6man-omni". 4526 o Several important updates to IPv6 ND cache states and route 4527 optimization message addressing. 4529 o Included introductory description of the "6M's". 4531 o Updated Multicast specification. 4533 Changes from draft-templin-6man-aero-00 to draft-templin-6man-aero- 4534 01: 4536 o Changed category to "Informational". 4538 o Updated implementation status. 4540 Changes from earlier versions to draft-templin-6man-aero-00: 4542 o Established working baseline reference. 4544 Author's Address 4545 Fred L. Templin (editor) 4546 Boeing Research & Technology 4547 P.O. Box 3707 4548 Seattle, WA 98124 4549 USA 4551 Email: fltemplin@acm.org