idnits 2.17.1 draft-templin-6man-aero-13.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 8, 2021) is 1052 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 1192, but not defined == Missing Reference: 'ULA0' is mentioned on line 1192, but not defined == Missing Reference: 'ULA1' is mentioned on line 1192, but not defined == Missing Reference: 'ULA2' is mentioned on line 1192, but not defined == Unused Reference: 'RFC3971' is defined on line 3944, but no explicit reference was found in the text == Unused Reference: 'RFC3972' is defined on line 3949, but no explicit reference was found in the text == Unused Reference: 'RFC4193' is defined on line 3957, but no explicit reference was found in the text == Unused Reference: 'RFC7739' is defined on line 3985, but no explicit reference was found in the text == Unused Reference: 'BGP' is defined on line 4006, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-comp-rtg-hdr' is defined on line 4009, but no explicit reference was found in the text == Unused Reference: 'I-D.bonica-6man-crh-helper-opt' is defined on line 4015, but no explicit reference was found in the text == Unused Reference: 'I-D.ietf-intarea-frag-fragile' is defined on line 4020, but no explicit reference was found in the text == Unused Reference: 'I-D.templin-6man-dhcpv6-ndopt' is defined on line 4043, but no explicit reference was found in the text == Unused Reference: 'OVPN' is defined on line 4072, but no explicit reference was found in the text == Unused Reference: 'RFC2003' is defined on line 4082, but no explicit reference was found in the text == Unused Reference: 'RFC2983' is defined on line 4103, but no explicit reference was found in the text == Unused Reference: 'RFC3168' is defined on line 4107, but no explicit reference was found in the text == Unused Reference: 'RFC4122' is defined on line 4121, but no explicit reference was found in the text == Unused Reference: 'RFC4982' is defined on line 4170, but no explicit reference was found in the text == Unused Reference: 'RFC6139' is defined on line 4212, but no explicit reference was found in the text == Unused Reference: 'RFC6273' is defined on line 4232, but no explicit reference was found in the text == Unused Reference: 'RFC6355' is defined on line 4241, but no explicit reference was found in the text == Unused Reference: 'RFC6438' is defined on line 4246, but no explicit reference was found in the text == Unused Reference: 'RFC6935' is defined on line 4255, but no explicit reference was found in the text == Unused Reference: 'RFC6936' is defined on line 4260, 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 8, 2021 5 Expires: December 10, 2021 7 Asymmetric Extended Route Optimization (AERO) 8 draft-templin-6man-aero-13 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 10, 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) . . . . . . . . 13 62 3.1. AERO Node Types . . . . . . . . . . . . . . . . . . . . . 13 63 3.2. The AERO Service over OMNI Links . . . . . . . . . . . . 14 64 3.2.1. AERO/OMNI Reference Model . . . . . . . . . . . . . . 14 65 3.2.2. Addressing and Node Identification . . . . . . . . . 17 66 3.2.3. AERO Routing System . . . . . . . . . . . . . . . . . 18 67 3.2.4. OMNI Link Segment Routing . . . . . . . . . . . . . . 20 68 3.2.5. Segment Routing Topologies (SRTs) . . . . . . . . . . 26 69 3.2.6. Segment Routing For OMNI Link Selection . . . . . . . 26 70 3.2.7. Segment Routing Within the OMNI Link . . . . . . . . 27 71 3.3. OMNI Interface Characteristics . . . . . . . . . . . . . 29 72 3.4. OMNI Interface Initialization . . . . . . . . . . . . . . 31 73 3.4.1. AERO Proxy/Server and Relay Behavior . . . . . . . . 31 74 3.4.2. AERO Client Behavior . . . . . . . . . . . . . . . . 32 75 3.4.3. AERO Bridge Behavior . . . . . . . . . . . . . . . . 32 76 3.5. OMNI Interface Neighbor Cache Maintenance . . . . . . . . 32 77 3.5.1. OMNI ND Messages . . . . . . . . . . . . . . . . . . 34 78 3.5.2. OMNI Neighbor Advertisement Message Flags . . . . . . 36 79 3.5.3. OMNI Neighbor Window Synchronization . . . . . . . . 36 80 3.6. OMNI Interface Encapsulation and Re-encapsulation . . . . 36 81 3.7. OMNI Interface Decapsulation . . . . . . . . . . . . . . 37 82 3.8. OMNI Interface Data Origin Authentication . . . . . . . . 37 83 3.9. OMNI Interface MTU . . . . . . . . . . . . . . . . . . . 38 84 3.10. OMNI Interface Forwarding Algorithm . . . . . . . . . . . 39 85 3.10.1. Client Forwarding Algorithm . . . . . . . . . . . . 40 86 3.10.2. Proxy/Server and Relay Forwarding Algorithm . . . . 41 87 3.10.3. Bridge Forwarding Algorithm . . . . . . . . . . . . 44 88 3.11. OMNI Interface Error Handling . . . . . . . . . . . . . . 46 89 3.12. AERO Router Discovery, Prefix Delegation and 90 Autoconfiguration . . . . . . . . . . . . . . . . . . . . 48 91 3.12.1. AERO Service Model . . . . . . . . . . . . . . . . . 49 92 3.12.2. AERO Client Behavior . . . . . . . . . . . . . . . . 49 93 3.12.3. AERO Proxy/Server Behavior . . . . . . . . . . . . . 51 94 3.13. The AERO Proxy Function . . . . . . . . . . . . . . . . . 54 95 3.13.1. Detecting and Responding to Proxy/Server Failures . 57 96 3.13.2. Point-to-Multipoint Proxy/Server Coordination . . . 58 98 3.14. AERO Route Optimization . . . . . . . . . . . . . . . . . 59 99 3.14.1. Route Optimization Initiation . . . . . . . . . . . 60 100 3.14.2. Relaying the NS(AR) *NET Packet(s) . . . . . . . . . 61 101 3.14.3. Processing the NS(AR) and Sending the NA(AR) . . . . 61 102 3.14.4. Relaying the NA(AR) . . . . . . . . . . . . . . . . 62 103 3.14.5. Processing the NA(AR) . . . . . . . . . . . . . . . 62 104 3.14.6. Forwarding Packets to Route Optimized Targets . . . 63 105 3.15. Neighbor Unreachability Detection (NUD) . . . . . . . . . 66 106 3.16. Mobility Management and Quality of Service (QoS) . . . . 67 107 3.16.1. Mobility Update Messaging . . . . . . . . . . . . . 68 108 3.16.2. Announcing Link-Layer Address and/or QoS Preference 109 Changes . . . . . . . . . . . . . . . . . . . . . . 69 110 3.16.3. Bringing New Links Into Service . . . . . . . . . . 69 111 3.16.4. Deactivating Existing Links . . . . . . . . . . . . 70 112 3.16.5. Moving Between Proxy/Servers . . . . . . . . . . . . 70 113 3.17. Multicast . . . . . . . . . . . . . . . . . . . . . . . . 71 114 3.17.1. Source-Specific Multicast (SSM) . . . . . . . . . . 72 115 3.17.2. Any-Source Multicast (ASM) . . . . . . . . . . . . . 73 116 3.17.3. Bi-Directional PIM (BIDIR-PIM) . . . . . . . . . . . 74 117 3.18. Operation over Multiple OMNI Links . . . . . . . . . . . 74 118 3.19. DNS Considerations . . . . . . . . . . . . . . . . . . . 75 119 3.20. Transition/Coexistence Considerations . . . . . . . . . . 75 120 3.21. Detecting and Reacting to Proxy/Server and Bridge 121 Failures . . . . . . . . . . . . . . . . . . . . . . . . 76 122 3.22. AERO Clients on the Open Internet . . . . . . . . . . . . 76 123 3.23. Time-Varying MNPs . . . . . . . . . . . . . . . . . . . . 78 124 4. Implementation Status . . . . . . . . . . . . . . . . . . . . 79 125 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 79 126 6. Security Considerations . . . . . . . . . . . . . . . . . . . 80 127 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 82 128 8. References . . . . . . . . . . . . . . . . . . . . . . . . . 83 129 8.1. Normative References . . . . . . . . . . . . . . . . . . 83 130 8.2. Informative References . . . . . . . . . . . . . . . . . 85 131 Appendix A. Non-Normative Considerations . . . . . . . . . . . . 91 132 A.1. Implementation Strategies for Route Optimization . . . . 91 133 A.2. Implicit Mobility Management . . . . . . . . . . . . . . 92 134 A.3. Direct Underlying Interfaces . . . . . . . . . . . . . . 92 135 A.4. AERO Critical Infrastructure Considerations . . . . . . . 93 136 A.5. AERO Server Failure Implications . . . . . . . . . . . . 93 137 A.6. AERO Client / Server Architecture . . . . . . . . . . . . 94 138 Appendix B. Change Log . . . . . . . . . . . . . . . . . . . . . 96 139 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 98 141 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. In terms of 158 precedence, readers may appreciate reading the AERO specification 159 first to gain an understanding of the overall architecture and 160 mobility services then return to the OMNI specification for a deeper 161 analysis of the NBMA link model. 163 The AERO service comprises Clients, Proxy/Servers and Relays that are 164 seen as OMNI link neighbors as well as Bridges that interconnect 165 diverse Internetworks as OMNI link segments through OAL forwarding at 166 a layer below IP. Each node's OMNI interface uses an IPv6 link-local 167 address format that supports operation of the IPv6 Neighbor Discovery 168 (ND) protocol [RFC4861] and links ND to IP forwarding. A node's OMNI 169 interface can be configured over multiple underlying interfaces, and 170 therefore appears as a single interface with multiple link-layer 171 addresses. Each link-layer address is subject to change due to 172 mobility and/or multilink fluctuations, and link-layer address 173 changes are signaled by ND messaging the same as for any IPv6 link. 175 AERO provides a secure cloud-based service where mobile node Clients 176 may use any Proxy/Server acting as a Mobility Anchor Point (MAP) and 177 fixed nodes may use any Relay on the link for efficient 178 communications. Fixed nodes forward original IP packets destined to 179 other AERO nodes via the nearest Relay, which forwards them through 180 the cloud. A mobile node's initial packets are forwarded through the 181 Proxy/Server, and direct routing is supported through route 182 optimization while packets are flowing. Both unicast and multicast 183 communications are supported, and mobile nodes may efficiently move 184 between locations while maintaining continuous communications with 185 correspondents and without changing their IP Address. 187 AERO Bridges peer with Proxy/Servers in a secured private BGP overlay 188 routing instance to provide a Segment Routing Topology (SRT) that 189 allows the OAL to span the underlying Internetworks of multiple 190 disjoint administrative domains as a single unified OMNI link at a 191 layer below IP. Each OMNI link instance is characterized by the set 192 of Mobility Service Prefixes (MSPs) common to all mobile nodes. 193 Relays provide an optimal route from correspondent nodes on the 194 underlying Internetwork to nodes on the OMNI link. To the underlying 195 Internetwork, the Relay is the source of a route to the MSP, and 196 hence uplink traffic to the mobile node is naturally routed to the 197 nearest Relay. A Relay can be considered as a simple case of a 198 Proxy/Server that provides only forwarding and not proxying services. 200 AERO can be used with OMNI links that span private-use Internetworks 201 and/or public Internetworks such as the global Internet. In the 202 latter case, some end systems may be located behind global Internet 203 Network Address Translators (NATs). A means for robust traversal of 204 NATs while avoiding "triangle routing" is therefore provided. 206 AERO assumes the use of PIM Sparse Mode in support of multicast 207 communication. In support of Source Specific Multicast (SSM) when a 208 Mobile Node is the source, AERO route optimization ensures that a 209 shortest-path multicast tree is established with provisions for 210 mobility and multilink operation. In all other multicast scenarios 211 there are no AERO dependencies. 213 AERO was designed as a secure aeronautical internetworking service 214 for both manned and unmanned aircraft, where the aircraft is treated 215 as a mobile node that can connect an Internet of Things (IoT). AERO 216 is also applicable to a wide variety of other use cases. For 217 example, it can be used to coordinate the links of mobile nodes 218 (e.g., cellphones, tablets, laptop computers, etc.) that connect into 219 a home enterprise network via public access networks with VPN or non- 220 VPN services enabled according to the appropriate security model. 221 AERO can also be used to facilitate terrestrial vehicular and urban 222 air mobility (as well as pedestrian communication services) for 223 future intelligent transportation systems 224 [I-D.ietf-ipwave-vehicular-networking][I-D.templin-ipwave-uam-its]. 225 Other applicable use cases are also in scope. 227 Along with OMNI, AERO provides secured optimal routing support for 228 the "6M's" of modern Internetworking, including: 230 1. Multilink - a mobile node's ability to coordinate multiple 231 diverse underlying data links as a single logical unit (i.e., the 232 OMNI interface) to achieve the required communications 233 performance and reliability objectives. 235 2. Multinet - the ability to span the OMNI link over a segment 236 routing topology with multiple diverse network administrative 237 domains while maintaining seamless end-to-end communications 238 between mobile Clients and correspondents such as air traffic 239 controllers, fleet administrators, etc. 241 3. Mobility - a mobile node's ability to change network points of 242 attachment (e.g., moving between wireless base stations) which 243 may result in an underlying interface address change, but without 244 disruptions to ongoing communication sessions with peers over the 245 OMNI link. 247 4. Multicast - the ability to send a single network transmission 248 that reaches multiple nodes belonging to the same interest group, 249 but without disturbing other nodes not subscribed to the interest 250 group. 252 5. Multihop - a mobile node vehicle-to-vehicle relaying capability 253 useful when multiple forwarding hops between vehicles may be 254 necessary to "reach back" to an infrastructure access point 255 connection to the OMNI link. 257 6. MTU assurance - the ability to deliver packets of various robust 258 sizes between peers without loss due to a link size restriction, 259 and to dynamically adjust packets sizes to achieve the optimal 260 performance for each independent traffic flow. 262 The following numbered sections present the AERO specification. The 263 appendices at the end of the document are non-normative. 265 2. Terminology 267 The terminology in the normative references applies; especially, the 268 terminology in the OMNI specification [I-D.templin-6man-omni] is used 269 extensively throughout. The following terms are defined within the 270 scope of this document: 272 IPv6 Neighbor Discovery (ND) 273 a control message service for coordinating neighbor relationships 274 between nodes connected to a common link. AERO uses the IPv6 ND 275 messaging service specified in [RFC4861]. 277 IPv6 Prefix Delegation 278 a networking service for delegating IPv6 prefixes to nodes on the 279 link. The nominal service is DHCPv6 [RFC8415], however alternate 280 services (e.g., based on ND messaging) are also in scope. A 281 minimal form of prefix delegation known as "prefix registration" 282 can be used if the Client knows its prefix in advance and can 283 represent it in the IPv6 source address of an ND message. 285 Access Network (ANET) 286 a node's first-hop data link service network (e.g., a radio access 287 network, cellular service provider network, corporate enterprise 288 network, etc.) that often provides link-layer security services 289 such as IEEE 802.1X and physical-layer security (e.g., "protected 290 spectrum") to prevent unauthorized access internally and with 291 border network-layer security services such as firewalls and 292 proxys that prevent unauthorized outside access. 294 ANET interface 295 a node's attachment to a link in an ANET. 297 Internetwork (INET) 298 a connected IP network topology with a coherent routing and 299 addressing plan and that provides a transit backbone service for 300 ANET end systems. INETs also provide an underlay service over 301 which the AERO virtual link is configured. Example INETs include 302 corporate enterprise networks, aviation networks, and the public 303 Internet itself. When there is no administrative boundary between 304 an ANET and the INET, the ANET and INET are one and the same. 306 INET interface 307 a node's attachment to a link in an INET. 309 *NET 310 a "wildcard" term referring to either ANET or INET when it is not 311 necessary to draw a distinction between the two. 313 *NET interface 314 a node's attachment to a link in a *NET. 316 *NET Partition 317 frequently, *NETs such as large corporate enterprise networks are 318 sub-divided internally into separate isolated partitions (a 319 technique also known as "network segmentation"). Each partition 320 is fully connected internally but disconnected from other 321 partitions, and there is no requirement that separate partitions 322 maintain consistent Internet Protocol and/or addressing plans. 323 (Each *NET partition is seen as a separate OMNI link segment as 324 discussed below.) 326 *NET address 327 an IP address assigned to a node's interface connection to a *NET. 329 *NET encapsulation 330 the encapsulation of a packet in an outer header or headers that 331 can be routed within the scope of the local *NET partition. 333 OMNI link 334 the same as defined in [I-D.templin-6man-omni], and manifested by 335 IPv6 encapsulation [RFC2473]. The OMNI link spans underlying *NET 336 segments joined by virtual bridges in a spanning tree the same as 337 a bridged campus LAN. AERO nodes on the OMNI link appear as 338 single-hop neighbors at the network layer even though they may be 339 separated by multiple underlying *NET hops, and can use Segment 340 Routing [RFC8402] to cause packets to visit selected waypoints on 341 the link. 343 OMNI Interface 344 a node's attachment to an OMNI link. Since OMNI interface 345 addresses are managed for uniqueness, OMNI interfaces do not 346 require Duplicate Address Detection (DAD) and therefore set the 347 administrative variable 'DupAddrDetectTransmits' to zero 348 [RFC4862]. 350 OMNI Adaptation Layer (OAL) 351 an OMNI interface process whereby original IP packets admitted 352 into the interface are wrapped in a mid-layer IPv6 header and 353 subject to fragmentation and reassembly. The OAL is also 354 responsible for generating MTU-related control messages as 355 necessary, and for providing addressing context for spanning 356 multiple segments of a bridged OMNI link. 358 original IP packet 359 a whole IP packet or fragment admitted into the OMNI interface by 360 the network layer prior to OAL encapsulation and fragmentation, or 361 an IP packet delivered to the network layer by the OMNI interface 362 following OAL decapsulation and reassembly. 364 OAL packet 365 an original IP packet encapsulated in OAL headers and trailers 366 before OAL fragmentation, or following OAL reassembly. 368 OAL fragment 369 a portion of an OAL packet following fragmentation but prior to 370 *NET encapsulation, or following *NET encapsulation but prior to 371 OAL reassembly. 373 (OAL) atomic fragment 374 an OAL packet that does not require fragmentation is always 375 encapsulated as an "atomic fragment" and includes a Fragment 376 Header with Fragment Offset and More Fragments both set to 0, but 377 with a valid Identification value. 379 (OAL) carrier packet 380 an encapsulated OAL fragment following *NET encapsulation or prior 381 to *NET decapsulation. OAL sources and destinations exchange 382 carrier packets over underlying interfaces, and may be separated 383 by one or more OAL intermediate nodes. OAL intermediate nodes re- 384 encapsulate carrier packets during forwarding by removing the *NET 385 headers of the previous hop underlying network and replacing them 386 with new *NET headers for the next hop underlying network. 388 OAL source 389 an OMNI interface acts as an OAL source when it encapsulates 390 original IP packets to form OAL packets, then performs OAL 391 fragmentation and *NET encapsulation to create carrier packets. 393 OAL destination 394 an OMNI interface acts as an OAL destination when it decapsulates 395 carrier packets, then performs OAL reassembly and decapsulation to 396 derive the original IP packet. 398 OAL intermediate node 399 an OMNI interface acts as an OAL intermediate node when it removes 400 the *NET headers of carrier packets received on a first segment, 401 then re-encapsulates the carrier packets in new *NET headers and 402 forwards them into the next segment. OAL intermediate nodes 403 decrement the Hop Limit of the OAL IPv6 header during re- 404 encapsulation, and discard the packet if the Hop Limit reaches 0. 405 OAL intermediate nodes do not decrement the Hop Limit/TTL of the 406 original IP packet. 408 underlying interface 409 a *NET interface over which an OMNI interface is configured. 411 Mobility Service Prefix (MSP) 412 an aggregated IP Global Unicast Address (GUA) prefix (e.g., 413 2001:db8::/32, 192.0.2.0/24, etc.) assigned to the OMNI link and 414 from which more-specific Mobile Network Prefixes (MNPs) are 415 delegated. OMNI link administrators typically obtain MSPs from an 416 Internet address registry, however private-use prefixes can 417 alternatively be used subject to certain limitations (see: 418 [I-D.templin-6man-omni]). OMNI links that connect to the global 419 Internet advertise their MSPs to their interdomain routing peers. 421 Mobile Network Prefix (MNP) 422 a longer IP prefix delegated from an MSP (e.g., 423 2001:db8:1000:2000::/56, 192.0.2.8/30, etc.) and delegated to an 424 AERO Client or Relay. 426 Mobile Network Prefix Link Local Address (MNP-LLA) 427 an IPv6 Link Local Address that embeds the most significant 64 428 bits of an MNP in the lower 64 bits of fe80::/64, as specified in 429 [I-D.templin-6man-omni]. 431 Mobile Network Prefix Unique Local Address (MNP-ULA) 432 an IPv6 Unique-Local Address derived from an MNP-LLA. 434 Administrative Link Local Address (ADM-LLA) 435 an IPv6 Link Local Address that embeds a 32-bit administratively- 436 assigned identification value in the lower 32 bits of fe80::/96, 437 as specified in [I-D.templin-6man-omni]. 439 Administrative Unique Local Address (ADM-ULA) 440 an IPv6 Unique-Local Address derived from an ADM-LLA. 442 AERO node 443 a node that is connected to an OMNI link and participates in the 444 AERO internetworking and mobility service. 446 AERO Client ("Client") 447 an AERO node that connects over one or more underlying interfaces 448 and requests MNP delegation/registration service from AERO Proxy/ 449 Servers. The Client assigns an MNP-LLA to the OMNI interface for 450 use in ND exchanges with other AERO nodes and forwards original IP 451 packets to correspondents according to OMNI interface neighbor 452 cache state. 454 AERO Proxy/Server ("Proxy/Server") 455 a dual-function node that provides a proxying service between AERO 456 Clients and external peers on its Client-facing ANET interfaces 457 (i.e., in the same fashion as for an enterprise network proxy) as 458 well as default forwarding and Mobility Anchor Point (MAP) 459 services for coordination with correspondents on its INET-facing 460 interfaces. (Proxy/Servers in the open INET instead configure 461 only an INET interface and no ANET interfaces.) The Proxy/Server 462 configures an OMNI interface and assigns an ADM-LLA to support the 463 operation of IPv6 ND services, while advertising all of its 464 associated MNPs via BGP peerings with Bridges. Note that the 465 Proxy and Server functions can be considered logically separable, 466 but since each Proxy/Server must be informed of all of the 467 Client's other multilink Proxy/Server affiliations the AERO 468 service is best supported when the two functions are coresident on 469 the same physical or logical platform. 471 AERO Relay ("Relay") 472 a Proxy/Server that provides forwarding services between nodes 473 reached via the OMNI link and correspondents on connected 474 downstream links. AERO Relays configure an OMNI interface and 475 assign an ADM-LLA the same as Proxy/Servers. AERO Relays also run 476 a dynamic routing protocol to discover any non-MNP IP GUA routes 477 in service on its connected downstream network links. In both 478 cases, the Relay advertises the MSP(s) to its downstream networks, 479 and distributes all of its associated non-MNP IP GUA routes via 480 BGP peerings with Bridges (i.e., the same as for Proxy/Servers). 482 AERO Bridge ("Bridge") 483 a node that provides OAL forwarding services (as well as a 484 security trust anchor) for nodes on an OMNI link. The Bridge 485 forwards carrier packets between OMNI link segments as OAL 486 intermediate nodes while decrementing the OAL IPv6 header Hop 487 Limit but without decrementing the network layer IP TTL/Hop Limit. 488 AERO Bridges peer with Proxy/Servers and other Bridges over 489 secured tunnels to discover the full set of MNPs for the link as 490 well as any non-MNP IP GUA routes that are reachable via Relays. 492 First-Hop Segment (FHS) Proxy/Server 493 a Proxy/Server for an underlying interface of the source Client 494 that forwards packets sent by the source Client over that 495 interface into the segment routing topology. 497 Last-Hop Segment (LHS) Proxy/Server 498 a Proxy/Server for an underlying interface of the target Client 499 that forwards packets received from the segment routing topology 500 to the target Client over that interface. 502 Segment Routing Topology (SRT) 503 a multinet forwarding region between the FHS Proxy/Server and LHS 504 Proxy/Server. FHS/LHS Proxy/Servers and SRT Bridges span the OMNI 505 link on behalf of source/target Client pairs. The SRT maintains a 506 spanning tree established through BGP peerings between Bridges and 507 Proxy/Servers. Each SRT segment includes Bridges in a "hub" and 508 Proxy/Servers in "spokes", while adjacent segments are connected 509 by Bridge-Bridge peerings. The BGP peerings are configured over 510 both secured and unsecured underlying network paths such that a 511 secured spanning tree is available for critical control messages 512 while other messages can use the unsecured spanning tree. 514 link-layer address 515 an IP address used as an encapsulation header source or 516 destination address from the perspective of the OMNI interface. 517 When an upper layer protocol (e.g., UDP) is used as part of the 518 encapsulation, the port number is also considered as part of the 519 link-layer address. 521 network layer address 522 the source or destination address of an original IP packet 523 presented to the OMNI interface. 525 end user network (EUN) 526 an internal virtual or external edge IP network that an AERO 527 Client or Relay connects to the rest of the network via the OMNI 528 interface. The Client/Relay sees each EUN as a "downstream" 529 network, and sees the OMNI interface as the point of attachment to 530 the "upstream" network. 532 Mobile Node (MN) 533 an AERO Client and all of its downstream-attached networks that 534 move together as a single unit, i.e., an end system that connects 535 an Internet of Things. 537 Mobile Router (MR) 538 a MN's on-board router that forwards original IP packets between 539 any downstream-attached networks and the OMNI link. The MR is the 540 MN entity that hosts the AERO Client. 542 Route Optimization Source (ROS) 543 the AERO node nearest the source that initiates route 544 optimization. The ROS may be a FHS Proxy/Server or Relay for the 545 source, or may be the source Client itself. 547 Route Optimization responder (ROR) 548 the AERO node that responds to route optimization requests on 549 behalf of the target. The ROR may be an LHS Proxy/Server for a 550 target MNP Client or an LHS Relay for a non-MNP target. 552 MAP List 553 a geographically and/or topologically referenced list of addresses 554 of all Proxy/Servers within the same OMNI link. Each OMNI link 555 has its own MAP list. 557 Distributed Mobility Management (DMM) 558 a BGP-based overlay routing service coordinated by Proxy/Servers 559 and Bridges that tracks all Proxy/Server-to-Client associations. 561 Mobility Service (MS) 562 the collective set of all Proxy/Servers, Bridges and Relays that 563 provide the AERO Service to Clients. 565 Mobility Service Endpoint MSE) 566 an individual Proxy/Server, Bridge or Relay in the Mobility 567 Service. 569 Throughout the document, the simple terms "Client", "Proxy/Server", 570 "Bridge" and "Relay" refer to "AERO Client", "AERO Proxy/Server", 571 "AERO Bridge" and "AERO Relay", respectively. Capitalization is used 572 to distinguish these terms from other common Internetworking uses in 573 which they appear without capitalization. 575 The terminology of IPv6 ND [RFC4861] and DHCPv6 [RFC8415] (including 576 the names of node variables, messages and protocol constants) is used 577 throughout this document. The terms "All-Routers multicast", "All- 578 Nodes multicast", "Solicited-Node multicast" and "Subnet-Router 579 anycast" are defined in [RFC4291]. Also, the term "IP" is used to 580 generically refer to either Internet Protocol version, i.e., IPv4 581 [RFC0791] or IPv6 [RFC8200]. 583 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 584 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 585 "OPTIONAL" in this document are to be interpreted as described in BCP 586 14 [RFC2119][RFC8174] when, and only when, they appear in all 587 capitals, as shown here. 589 3. Asymmetric Extended Route Optimization (AERO) 591 The following sections specify the operation of IP over OMNI links 592 using the AERO service: 594 3.1. AERO Node Types 596 AERO Clients are Mobile Nodes (MNs) that configure OMNI interfaces 597 over underlying interfaces with addresses that may change when the 598 Client moves to a new network connection point. AERO Clients 599 register their Mobile Network Prefixes (MNPs) with the AERO service, 600 and distribute the MNPs to nodes on EUNs. AERO Bridges, Proxy/ 601 Servers and Relays are critical infrastructure elements in fixed 602 (i.e., non-mobile) INET deployments and hence have permanent and 603 unchanging INET addresses. Together, they constitute the AERO 604 service which provides an OMNI link virtual overlay for connecting 605 AERO Clients. 607 AERO Bridges (together with Proxy/Servers) provide the secured 608 backbone supporting infrastucture for a Segment Routing Topology 609 (SRT) that spans the OMNI link. Bridges forward carrier packets both 610 within the same SRT partition and between disjoint SRT partitions 611 based on an IPv6 encapsulation mid-layer known as the OMNI Adaptation 612 Layer (OAL) [I-D.templin-6man-omni]. During forwarding, the inner IP 613 layer experiences a virtual bridging service since the inner IP TTL/ 614 Hop Limit is not decremented. Each Bridge also peers with Proxy/ 615 Servers and other Bridges in a dynamic routing protocol instance to 616 provide a Distributed Mobility Management (DMM) service for the list 617 of active MNPs (see Section 3.2.3). Bridges present the OMNI link as 618 a set of one or more Mobility Service Prefixes (MSPs) and configure 619 secured tunnels with Proxy/Servers, Relays and other Bridges; they 620 further maintain IP forwarding table entries for each MNP and any 621 other reachable non-MNP prefixes. 623 AERO Proxy/Servers in distributed SRT segments provide default 624 forwarding and mobility/multilink services for AERO Client Mobile 625 Nodes (MNs). Each Proxy/Server also peers with Bridges in a dynamic 626 routing protocol instance to advertise its list of associated MNPs 627 (see Section 3.2.3). Proxy/Servers facilitate prefix delegation/ 628 registration exchanges with Clients, where each delegated prefix 629 becomes an MNP taken from an MSP. Proxy/Servers forward carrier 630 packets between OMNI interface neighbors and track each Client's 631 mobility profiles. Proxy/Servers at ANET/INET boundaries provide a 632 conduit for ANET Clients to associate with peers reached through 633 external INETs. Proxy/Servers in the open INET support INET Clients 634 through authenticated IPv6 ND message exchanges. Source Clients 635 employ First-Hop Segment (FHS) Proxy/Servers to forward packets over 636 the SRT to Last-Hop Segment (LHS) Proxy/Servers which finally forward 637 to target Clients. 639 AERO Relays are Proxy/Servers that provide forwarding services to 640 exchange original IP packets between the OMNI interface and INET/EUN 641 interfaces. Relays are provisioned with MNPs the same as for an AERO 642 Client, and also run a dynamic routing protocol to discover any non- 643 MNP IP routes. The Relay advertises the MSP(s) to its connected 644 networks, and distributes all of its associated MNP and non-MNP 645 routes via BGP peerings with Bridges 647 3.2. The AERO Service over OMNI Links 649 3.2.1. AERO/OMNI Reference Model 651 Figure 1 presents the basic OMNI link reference model: 653 +----------------+ 654 | AERO Bridge B1 | 655 | Nbr: S1, S2, P1| 656 |(X1->S1; X2->S2)| 657 | MSP M1 | 658 +-------+--------+ 659 +--------------+ | +--------------+ 660 | AERO P/S S1 | | | AERO P/S S2 | 661 | Nbr: C1, B1 | | | Nbr: C2, B1 | 662 | default->B1 | | | default->B1 | 663 | X1->C1 | | | X2->C2 | 664 +-------+------+ | +------+-------+ 665 | OMNI link | | 666 X===+===+==================+===================+===+===X 667 | | 668 +-----+--------+ +--------+-----+ 669 |AERO Client C1| |AERO Client C2| 670 | Nbr: S1 | | Nbr: S2 | 671 | default->S1 | | default->S2 | 672 | MNP X1 | | MNP X2 | 673 +------+-------+ +-----+--------+ 674 | | 675 .-. .-. 676 ,-( _)-. ,-( _)-. 677 .-(_ IP )-. +-------+ +-------+ .-(_ IP )-. 678 (__ EUN )--|Host H1| |Host H2|--(__ EUN ) 679 `-(______)-' +-------+ +-------+ `-(______)-' 681 Figure 1: AERO/OMNI Reference Model 683 In this model: 685 o the OMNI link is an overlay network service configured over one or 686 more underlying SRT partitions which may be managed by different 687 administrative authorities and have incompatible protocols and/or 688 addressing plans. 690 o AERO Bridge B1 aggregates Mobility Service Prefix (MSP) M1, 691 discovers Mobile Network Prefixes (MNPs) X* and advertises the MSP 692 via BGP peerings over secured tunnels to Proxy/Servers (S1, S2). 693 Bridges provide the backbone for an SRT that spans the OMNI link. 695 o AERO Proxy/Servers S1 and S2 configure secured tunnels with Bridge 696 B1 and also provide mobility, multilink, multicast and default 697 router services for the MNPs of their associated Clients C1 and 698 C2. (Proxy/Servers that act as Relays can also advertise non-MNP 699 routes for non-mobile correspondent nodes the same as for MNP 700 Clients.) 702 o AERO Clients C1 and C2 associate with Proxy/Servers S1 and S2, 703 respectively. They receive MNP delegations X1 and X2, and also 704 act as default routers for their associated physical or internal 705 virtual EUNs. Simple hosts H1 and H2 attach to the EUNs served by 706 Clients C1 and C2, respectively. 708 An OMNI link configured over a single *NET appears as a single 709 unified link with a consistent underlying network addressing plan. 710 In that case, all nodes on the link can exchange carrier packets via 711 simple *NET encapsulation (i.e., following any necessary NAT 712 traversal), since the underlying *NET is connected. In common 713 practice, however, OMNI links are traversed by a Segment Routing 714 Topology (SRT) spanning tree, where each segment is a distinct *NET 715 potentially managed under a different administrative authority (e.g., 716 as for worldwide aviation service providers such as ARINC, SITA, 717 Inmarsat, etc.). Individual *NETs may also themselves be partitioned 718 internally, in which case each internal partition is seen as a 719 separate segment. 721 The addressing plan of each SRT segment is consistent internally but 722 will often bear no relation to the addressing plans of other 723 segments. Each segment is also likely to be separated from others by 724 network security devices (e.g., firewalls, proxys, packet filtering 725 gateways, etc.), and in many cases disjoint segments may not even 726 have any common physical link connections. Therefore, nodes can only 727 be assured of exchanging carrier packets directly with correspondents 728 in the same segment, and not with those in other segments. The only 729 means for joining the segments therefore is through inter-domain 730 peerings between AERO Bridges. 732 The same as for traditional campus LANs, the OMNI link SRT spans 733 multiple segments that can be joined into a single unified link using 734 the OMNI Adaptation Layer (OAL) [I-D.templin-6man-omni] which inserts 735 a mid-layer IPv6 encapsulation header that supports inter-segment 736 forwarding (i.e., bridging) without decrementing the network-layer 737 TTL/Hop Limit of the original IP packet. An example OMNI link SRT is 738 shown in Figure 2: 740 . . . . . . . . . . . . . . . . . . . . . . . 741 . . 742 . .-(::::::::) . 743 . .-(::::::::::::)-. +-+ . 744 . (:::: Segment A :::)--|B|---+ . 745 . `-(::::::::::::)-' +-+ | . 746 . `-(::::::)-' | . 747 . | . 748 . .-(::::::::) | . 749 . .-(::::::::::::)-. +-+ | . 750 . (:::: Segment B :::)--|B|---+ . 751 . `-(::::::::::::)-' +-+ | . 752 . `-(::::::)-' | . 753 . | . 754 . .-(::::::::) | . 755 . .-(::::::::::::)-. +-+ | . 756 . (:::: Segment C :::)--|B|---+ . 757 . `-(::::::::::::)-' +-+ | . 758 . `-(::::::)-' | . 759 . | . 760 . ..(etc).. x . 761 . . 762 . . 763 . <- Segment Routing Topology (SRT) -> . 764 . . . . . . . . . . . . . .. . . . . . . . . 766 Figure 2: OMNI Link Segment Routing Topology (SRT) 768 Bridges, Proxy/Servers and Relay OMNI interfaces are configured over 769 both secured tunnels and open INET underlying interfaces within their 770 respective SRT segments. Within each segment, Bridges configure 771 "hub-and-spokes" BGP peerings with Proxy/Server/Relays as "spokes". 772 Adjacent SRT segments are joined by Bridge-to-Bridge peerings to 773 collectively form a spanning tree over the entire SRT. The "secured" 774 spanning tree supports strong authentication for control plane 775 messages. The "unsecured" spanning tree conveys ordinary carrier 776 packets without security codes and that must be treated by 777 destinations according to data origin authentication procedures. 778 Route optimization can be employed to cause carrier packets to take 779 more direct paths between OMNI link neighbors without having to 780 follow strict SRT spanning tree paths. 782 3.2.2. Addressing and Node Identification 784 AERO nodes on OMNI links use the Link-Local Address (LLA) prefix 785 fe80::/64 [RFC4291] to assign LLAs used for network-layer addresses 786 in link-scoped IPv6 ND and data messages. AERO Clients use LLAs 787 constructed from MNPs (i.e., "MNP-LLAs") while other AERO nodes use 788 LLAs constructed from administrative identification values ("ADM- 789 LLAs") as specified in [I-D.templin-6man-omni]. Non-MNP routes are 790 also represented the same as for MNP-LLAs, but may include a prefix 791 that is not properly covered by the MSP. 793 AERO nodes also use the Unique Local Address (ULA) prefix fd00::/8 794 followed by a pseudo-random 40-bit OMNI domain identifier to form the 795 prefix [ULA]::/48, then include a 16-bit OMNI link identifier '*' to 796 form the prefix [ULA*]::/64 [RFC4291]. The AERO node then uses the 797 prefix [ULA*]::/64 to form "MNP-ULAs" or "ADM-ULA"s as specified in 798 [I-D.templin-6man-omni] to support OAL addressing. (The prefix 799 [ULA*]::/64 appearing alone and with no suffix represents "default".) 800 AERO Clients also use Temporary ULAs constructed per 801 [I-D.templin-6man-omni], where the addresses are typically used only 802 in initial control message exchanges until a stable MNP-LLA/ULA is 803 assigned. 805 AERO MSPs, MNPs and non-MNP routes are typically based on Global 806 Unicast Addresses (GUAs), but in some cases may be based on private- 807 use addresses. See [I-D.templin-6man-omni] for a full specification 808 of LLAs, ULAs and GUAs used by AERO nodes on OMNI links. 810 Finally, AERO Clients and Proxy/Servers configure node identification 811 values as specified in [I-D.templin-6man-omni]. 813 3.2.3. AERO Routing System 815 The AERO routing system comprises a private instance of the Border 816 Gateway Protocol (BGP) [RFC4271] that is coordinated between Bridges 817 and Proxy/Servers and does not interact with either the public 818 Internet BGP routing system or any underlying INET routing systems. 820 In a reference deployment, each Proxy/Server is configured as an 821 Autonomous System Border Router (ASBR) for a stub Autonomous System 822 (AS) using a 32-bit AS Number (ASN) [RFC4271] that is unique within 823 the BGP instance, and each Proxy/Server further uses eBGP to peer 824 with one or more Bridges but does not peer with other Proxy/Servers. 825 Each SRT segment in the OMNI link must include one or more Bridges, 826 which peer with the Proxy/Servers within that segment. All Bridges 827 within the same segment are members of the same hub AS, and use iBGP 828 to maintain a consistent view of all active routes currently in 829 service. The Bridges of different segments peer with one another 830 using eBGP. 832 Bridges maintain forwarding table entries only for the MNP-ULAs 833 corresponding to MNP and non-MNP routes that are currently active, 834 and carrier packets destined to all other MNP-ULAs will correctly 835 incur Destination Unreachable messages due to the black-hole route. 837 In this way, Proxy/Servers and Relays have only partial topology 838 knowledge (i.e., they only maintain routing information for their 839 directly associated Clients and non-AERO links) and they forward all 840 other carrier packets to Bridges which have full topology knowledge. 842 Each OMNI link SRT segment assigns a unique ADM-ULA sub-prefix of 843 [ULA*]::/96. For example, a first segment could assign 844 [ULA*]::1000/116, a second could assign [ULA*]::2000/116, a third 845 could assign [ULA*]::3000/116, etc. Within each segment, each Proxy/ 846 Server configures an ADM-ULA within the segment's prefix, e.g., the 847 Proxy/Servers within [ULA*]::2000/116 could assign the ADM-ULAs 848 [ULA*]::2011/116, [ULA*]::2026/116, [ULA*]::2003/116, etc. 850 The administrative authorities for each segment must therefore 851 coordinate to assure mutually-exclusive ADM-ULA prefix assignments, 852 but internal provisioning of ADM-ULAs an independent local 853 consideration for each administrative authority. For each ADM-ULA 854 prefix, the Bridge(s) that connect that segment assign the all-zero's 855 address of the prefix as a Subnet Router Anycast address. For 856 example, the Subnet Router Anycast address for [ULA*]::1023/116 is 857 simply [ULA*]::1000. 859 ADM-ULA prefixes are statically represented in Bridge forwarding 860 tables. Bridges join multiple SRT segments into a unified OMNI link 861 over multiple diverse network administrative domains. They support a 862 bridging function by first establishing forwarding table entries for 863 their ADM-ULA prefixes either via standard BGP routing or static 864 routes. For example, if three Bridges ('A', 'B' and 'C') from 865 different segments serviced [ULA*]::1000/116, [ULA*]::2000/116 and 866 [ULA*]::3000/116 respectively, then the forwarding tables in each 867 Bridge are as follows: 869 A: [ULA*]::1000/116->local, [ULA*]::2000/116->B, [ULA*]::3000/116->C 871 B: [ULA*]::1000/116->A, [ULA*]::2000/116->local, [ULA*]::3000/116->C 873 C: [ULA*]::1000/116->A, [ULA*]::2000/116->B, [ULA*]::3000/116->local 875 These forwarding table entries are permanent and never change, since 876 they correspond to fixed infrastructure elements in their respective 877 segments. 879 MNP ULAs are instead dynamically advertised in the AERO routing 880 system by Proxy/Servers and Relays that provide service for their 881 corresponding MNPs. For example, if three Proxy/Servers ('D', 'E' 882 and 'F') service the MNPs 2001:db8:1000:2000::/56, 883 2001:db8:3000:4000::/56 and 2001:db8:5000:6000::/56 then the routing 884 system would include: 886 D: [ULA*]:2001:db8:1000:2000/120 888 E: [ULA*]:2001:db8:3000:4000/120 890 F: [ULA*]:2001:db8:5000:6000/120 892 A full discussion of the BGP-based routing system used by AERO is 893 found in [I-D.ietf-rtgwg-atn-bgp]. 895 3.2.4. OMNI Link Segment Routing 897 With the Client and partition prefixes in place in Bridge forwarding 898 tables, the OMNI interface sends control and data carrier packets 899 toward AERO destination nodes located in different OMNI link segments 900 over the SRT spanning tree. The OMNI interface uses the OMNI 901 Adaptation Layer (OAL) encapsulation service [I-D.templin-6man-omni], 902 and includes an OMNI Routing Header (ORH) as an extension to the OAL 903 header. Each carrier packet includes at most one ORH in compressed 904 or uncompressed form, with the uncompressed form shown in Figure 3: 906 0 1 2 3 907 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 1 908 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 909 | Next Header | Hdr Ext Len | Routing Type | Segments Left | 910 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 911 | omIndex | FMT | SRT | LHS (bits 0 -15) | 912 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 913 | LHS (bits 0 -15) | ~ 914 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ 915 ~ Link Layer Address (L2ADDR) ~ 916 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 917 | Null Padding (if necessary) | 918 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 919 ~ Destination Suffix ~ 920 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 922 Figure 3: OMNI Routing Header (ORH) Format 924 The ORH includes the following fields, in consecutive order: 926 o Next Header identifies the type of header immediately following 927 the ORH. 929 o Hdr Ext Len is the length of the Routing header in 8-octet units 930 (not including the first 8 octets). The field must encode a value 931 between 0 and 4 (all other values indicate a parameter problem). 933 o Routing Type is set to TBD1 (see IANA Considerations). 935 o Segments Left encodes the value 0 or 1 (all other values indicate 936 a parameter problem). 938 o omIndex - a 1-octet field consulted only when Segments Left is 0; 939 identifies a specific target Client underlying interface serviced 940 by the LHS Proxy-Server when there are multiple alternatives. 941 When FMT-Forward is clear, omIndex determines the interface for 942 forwarding the ORH packet following reassembly; when FMT-Forward 943 is set, omIndex determines the interface for forwarding the raw 944 carrier packets without first reassembling. When omIndex is set 945 to 0 (or when no ORH is present), the LHS Proxy/Server selects 946 among any of the Client's available underlying interfaces that it 947 services locally (i.e., and not those serviced by another Proxy/ 948 Server). 950 o FMT - a 3-bit "Forward/Mode/Trailer" code corresponding to the 951 included Link Layer Address as follows: 953 * When the most significant bit (i.e., "FMT-Forward") is clear, 954 the LHS Proxy/Server must reassemble. When FMT-Forward is set, 955 the LHS Proxy/Server must forward the fragments to the Client 956 (while changing the OAL destination address to the MNP-ULA of 957 the Client if necessary) without reassembling. 959 * When the next most significant bit (i.e., "FMT-Mode") is clear, 960 L2ADDR is the INET address of the LHS Proxy/Server and the 961 Client must be reached through the LHS Proxy/Server. When FMT- 962 Mode is set, the Client is eligible for route optimization over 963 the open INET where it may be located behind one or more NATs, 964 and L2ADDR is either the INET address of the LHS Proxy/Server 965 (when FMT-Forward is set) or the native INET address of the 966 Client itself (when FMT-Forward is clear). 968 * The least significant bit (i.e., "FMT-Type") is consulted only 969 when Hdr Ext Len is 1 and ignored otherwise. If FMT-Type is 970 clear, the remaining 10 ORH octets contain an LHS followed by 971 an IPv4 L2ADDR. If FMT-Type is set, the remainder instead 972 contains 2 null padding octets followed by an 8-octet (IPv6) 973 Destination Suffix. 975 o SRT - a 5-bit Segment Routing Topology prefix length consulted 976 only when Segments Left is 1, and encodes a value that (when added 977 to 96) determines the prefix length to apply to the ADM-ULA formed 978 from concatenating [ULA*]::/96 with the 32 bit LHS value (for 979 example, the value 16 corresponds to the prefix length 112). 981 o LHS - a 4-octet field present only when indicated by the ORH 982 length (see below) and consulted only when Segments Left is 1. 983 The field encodes the 32-bit ADM-ULA suffix of an LHS Proxy/Server 984 for the target. When SRT and LHS are both set to 0, the LHS 985 Proxy/Server must be reached directly via INET encapsulation 986 instead of over the spanning tree. When SRT is set to 0 and LHS 987 is non-zero, the prefix length is set to 128. SRT and LHS 988 determine the ADM-ULA of the LHS Proxy/Server over the spanning 989 tree. 991 o Link Layer Address (L2ADDR) - an IP encapsulation address present 992 only when indicated by the ORH length (see below) and consulted 993 only when Segments Left is 1. The ORH length also determines the 994 L2ADDR IP version since the field will always contain exactly 6 995 octets for UDP/IPv4 or 18 octets for UDP/IPv6. When present, 996 provides the link-layer address (i.e., the encapsulation address) 997 of the LHS Proxy/Server or the target Client itself. The UDP Port 998 Number appears in the first two octets and the IP address appears 999 in the remaining octets. The Port Number and IP address are 1000 recorded in network byte order, and in ones-compliment 1001 "obfuscated" form per [RFC4380]. The OMNI interface forwarding 1002 algorithm uses L2ADDR as the INET encapsulation address for 1003 forwarding when SRT/LHS is located in the same OMNI link segment. 1004 If direct INET encapsulation is not permitted, L2ADDR is instead 1005 set to all-zeros and the packet must be forwarded to the LHS 1006 Proxy-Server via the spanning tree. 1008 o Null Padding - zero-valued octets added as necessary to pad the 1009 portion of the ORH included up to this point to an even 8-octet 1010 boundary. 1012 o Destination Suffix - a trailing 8-octet field present only when 1013 indicated by the ORH length (see below). When ORH length is 1, 1014 FMT-Type determines whether the option includes a Destination 1015 Suffix or an LHS/L2ADDR for IPv4 since there is only enough space 1016 available for one. When present, encodes the 64-bit MNP-ULA 1017 suffix for the target Client. 1019 The ORH Hdr Ext Len field value also serves as an implicit ORH 1020 "Type", with 5 distinct Types specified (i.e., ORH-0 through ORH-4). 1021 All ORH-* Types include the same 6-octet preamble beginning with Next 1022 Header up to and including omIndex, followed by a Type-specific 1023 remainder as follows: 1025 o ORH-0 - The preamble Hdr Ext Len and Segments Left must both be 0. 1026 Two null padding octets follow the preamble, and all other fields 1027 are omitted. 1029 o ORH-1 - The preamble Hdr Ext Len is set to 1. When FMT-Type is 1030 clear, the LHS and L2ADDR for IPv4 fields are included and the 1031 Destination Suffix is omitted. When FMT-Type is set, the LHS and 1032 L2ADDR fields are omitted, the Destination Suffix field is 1033 included and Segments Left must be 0. 1035 o ORH-2 - The preamble Hdr Ext Len is set to 2. The LHS, L2ADDR for 1036 IPv4 and Destination Suffix fields are all included. 1038 o ORH-3 - The preamble Hdr Ext Len is set to 3. The LHS and L2ADDR 1039 for IPv6 fields are included and the Destination Suffix field is 1040 omitted. 1042 o ORH-4 - The preamble Hdr Ext Len is set to 4. The LHS, L2ADDR for 1043 IPv6 and Destination Suffix fields are all included. 1045 AERO neighbors use OAL encapsulation and fragmentation to exchange 1046 OAL packets as specified in [I-D.templin-6man-omni]. When an AERO 1047 node's OMNI interface (acting as an OAL source) uses OAL 1048 encapsulation for an original IP packet with source address 1049 2001:db8:1:2::1 and destination address 2001:db8:1234:5678::1, it 1050 sets the OAL header source address to its own ULA (e.g., 1051 [ULA*]::2001:db8:1:2), sets the destination address to the MNP-ULA 1052 corresponding to the IP destination address (e.g., 1053 [ULA*]::2001:db8:1234:5678), sets the Traffic Class, Flow Label, Hop 1054 Limit and Payload Length as discussed in [I-D.templin-6man-omni], 1055 then finally selects an Identification and appends an OAL checksum. 1057 If the neighbor cache information indicates that the target is in a 1058 different segment, the OAL source next inserts an ORH immediately 1059 following the OAL header while including Destination Suffix for non- 1060 first-fragments only when necessary (in this case, the Destination 1061 Suffix is 2001:db8:1234:5678). Next, to direct the packet to a 1062 first-hop Proxy/Server or a Bridge, the source prepares an ORH with 1063 Segments Left set to 1 and with SRT/LHS/L2ADDR included, then 1064 overwrites the OAL header destination address with the LHS Subnet 1065 Router Anycast address (for example, for LHS 3000:4567 with SRT 16, 1066 the Subnet Router Anycast address is [ULA*]::3000:0000). To send the 1067 packet to the LHS Proxy/Server either directly or via the spanning 1068 tree, the OAL source instead includes an ORH (Type 0 or 1) with 1069 Segments Left set to 0 and LHS/L2ADDR omitted, and overwrites the OAL 1070 header destination address with either the LHS Proxy/Server ADM-ULA 1071 or the MNP-ULA of the Client itself. 1073 The OAL source then fragments the OAL packet, with each resulting OAL 1074 fragment including the OAL/ORH headers while only the first fragment 1075 includes the original IPv6 header. If FMT-Forward is set, the 1076 Identification used for fragmentation must be within the window for 1077 the Client and a Destination Suffix must be included with each non- 1078 first-fragment when necessary; otherwise the Identification must be 1079 within the window for the Client's Proxy/Server and no Destination 1080 Suffix is needed. (Note that if no actual fragmentation is required 1081 the OAL source still prepares the packet as an "atomic" fragment that 1082 includes a Fragment Header with Offset and More Fragments both set to 1083 0.) The OAL source finally encapsulates each resulting OAL fragment 1084 in an *NET header to form an OAL carrier packet, with source address 1085 set to its own *NET address (e.g., 192.0.2.100) and destination set 1086 to the *NET address of the last hop itself or the next hop in the 1087 spanning tree (e.g., 192.0.2.1). 1089 The carrier packet encapsulation format in the above example is shown 1090 in Figure 4: 1092 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1093 | *NET Header | 1094 | src = 192.0.2.100 | 1095 | dst = 192.0.2.1 | 1096 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1097 | OAL IPv6 Header | 1098 | src = [ULA*]::2001:db8:1:2 | 1099 | dst= [ULA*]::3000:0000 | 1100 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1101 | ORH (if necessary) | 1102 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1103 | OAL Fragment Header | 1104 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1105 | Original IP Header | 1106 | (first-fragment only) | 1107 | src = 2001:db8:1:2::1 | 1108 | dst = 2001:db8:1234:5678::1 | 1109 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1110 | | 1111 ~ ~ 1112 ~ Original Packet Body/Fragment ~ 1113 ~ ~ 1114 | | 1115 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1117 Figure 4: Carrier Packet Format 1119 In this format, the original IP header and packet body/fragment are 1120 from the original IP packet, the OAL header is an IPv6 header 1121 prepared according to [RFC2473], the ORH is a Routing Header 1122 extension of the OAL header, the Fragment Header identifies each 1123 fragment, and the INET header is prepared as discussed in 1124 Section 3.6. When the OAL source transmits the resulting carrier 1125 packets, they are forwarded over possibly multiple OAL intermediate 1126 nodes in the OMNI link spanning tree until they arrive at the OAL 1127 destination. 1129 This gives rise to an SRT routing system that contains both Client 1130 MNP-ULA routes that may change dynamically due to regional node 1131 mobility and per-partition ADM-ULA routes that rarely if ever change. 1132 The spanning tree can therefore provide link-layer bridging by 1133 sending carrier packets over the spanning tree instead of network- 1134 layer routing according to MNP routes. As a result, opportunities 1135 for loss due to node mobility between different segments are 1136 mitigated. 1138 Note: The document recommends that AERO nodes transform ORHs with 1139 Segments Left set to 1 into ORH-0 or ORH-1 during forwarding. While 1140 this may yield encapsulation overhead savings in some cases, the AERO 1141 node may instead simply set Segments Left to 0 and leave the original 1142 ORH in place. The LHS Proxy/Server or target Client that processes 1143 the ORH will receive the same information in both cases. 1145 Note: When the OAL source sets a carrier packet OAL destination 1146 address to a target's MNP-ULA but does not assert a specific target 1147 underlying interface, it may omit the ORH whether forwarding to the 1148 LHS Proxy/Server or directly to the target itself. When the LHS 1149 Proxy/Server receives a carrier packet with OAL destination set to 1150 the target MNP-ULA but with no ORH, it forwards over any available 1151 underlying interface for the target that it services locally. 1153 Note: When the OAL source and destination are on the same INET 1154 segment, OAL header compression can be used to significantly reduce 1155 encapsulation overhead as discussed in [I-D.templin-6man-omni]. 1157 Note: Use of an IPv6 "minimal encapsulation" format (i.e., an IPv6 1158 variant of [RFC2004]) based on extensions to the ORH was considered 1159 and abandoned. In the approach, the ORH would be inserted as an 1160 extension header to the original IPv6 packet header. The IPv6 1161 destination address would then be written into the ORH, and the ULA 1162 corresponding to the destination would be overwritten in the IPv6 1163 destination address. This would seemingly convey enough forwarding 1164 information so that OAL encapsulation could be avoided. However, 1165 this "minimal encapsulation" IPv6 packet would then have a non-ULA 1166 source address and ULA destination address, an incorrect value in 1167 upper layer protocol checksums, and a Hop Limit that is decremented 1168 within the spanning tree when it should not be. The insertion and 1169 removal of the ORH would also entail rewriting the Payload Length and 1170 Next Header fields - again, invalidating upper layer checksums. 1171 These irregularities would result in implementation challenges and 1172 the potential for operational issues, e.g., since actionable ICMPv6 1173 error reports could not be delivered to the original source. In 1174 order to address the issues, still more information such as the 1175 original IPv6 source address could be written into the ORH. However, 1176 with the additional information the benefit of the "minimal 1177 encapsulation" savings quickly diminishes, and becomes overshadowed 1178 by the implementation and operational irregularities. 1180 3.2.5. Segment Routing Topologies (SRTs) 1182 The 64-bit sub-prefixes of [ULA]::/48 identify up to 2^16 distinct 1183 Segment Routing Topologies (SRTs). Each SRT is a mutually-exclusive 1184 OMNI link overlay instance using a distinct set of ULAs, and emulates 1185 a Virtual LAN (VLAN) service for the OMNI link. In some cases (e.g., 1186 when redundant topologies are needed for fault tolerance and 1187 reliability) it may be beneficial to deploy multiple SRTs that act as 1188 independent overlay instances. A communication failure in one 1189 instance therefore will not affect communications in other instances. 1191 Each SRT is identified by a distinct value in bits 48-63 of 1192 [ULA]::/48, i.e., as [ULA0]::/64, [ULA1]::/64, [ULA2]::/64, etc. 1193 Each OMNI interface is identified by a unique interface name (e.g., 1194 omni0, omni1, omni2, etc.) and assigns an anycast ADM-ULA 1195 corresponding to its SRT prefix length. The anycast ADM-ULA is used 1196 for OMNI interface determination in Safety-Based Multilink (SBM) as 1197 discussed in [I-D.templin-6man-omni]. Each OMNI interface further 1198 applies Performance-Based Multilink (PBM) internally. 1200 The Bridges and Proxy/Servers of each independent SRT engage in BGP 1201 peerings to form a spanning tree with the Bridges in non-leaf nodes 1202 and the Proxy/Servers in leaf nodes. The spanning tree is configured 1203 over both secured and unsecured underlying network paths. The 1204 secured spanning tree is used to convey secured control messages 1205 between FHS and LHS Proxy/Servers, while the unsecured spanning tree 1206 forwards data messages and/or unsecured control messages. 1208 3.2.6. Segment Routing For OMNI Link Selection 1210 Original IPv6 sourcse can direct IPv6 packets to an AERO node by 1211 including a standard IPv6 Segment Routing Header (SRH) [RFC8754] with 1212 the anycast ADM-ULA for the selected OMNI link as either the IPv6 1213 destination or as an intermediate hop within the SRH. This allows 1214 the original source to determine the specific OMNI link SRT an 1215 original IPv6 packet will traverse when there may be multiple 1216 alternatives. 1218 When an AERO node processes the SRH and forwards the original IPv6 1219 packet to the correct OMNI interface, the OMNI interface writes the 1220 next IPv6 address from the SRH into the IPv6 destination address and 1221 decrements Segments Left. If decrementing would cause Segments Left 1222 to become 0, the OMNI interface deletes the SRH before forwarding. 1223 This form of Segment Routing supports Safety-Based Multilink (SBM). 1225 3.2.7. Segment Routing Within the OMNI Link 1227 OAL sources can insert an ORH for Segment Routing within the same 1228 OMNI link to influence the paths of carrier packets sent to OAL 1229 destinations in remote SRT segments without requiring all carrier 1230 packets to traverse strict SRT spanning tree paths. (OAL sources can 1231 also insert an ORH in carrier packets sent to OAL destinations in the 1232 local segment if additional last-hop forwarding information is 1233 required.) 1235 When an AERO node's OMNI interface has an original IP packet to send 1236 to a target discovered through route optimization located in the same 1237 SRT segment, it acts as an OAL source to perform OAL encapsulation 1238 and fragmentation. The node then uses L2ADDR for INET encapsulation 1239 while including an ORH-0 when sending the resulting carrier packets 1240 to the ADM-ULA of the LHS Proxy/Server, or optionally omitting the 1241 ORH-0 when sending to the MNP-ULA of the target Client itself. When 1242 the node sends carrier packets with an ORH-0 to the LHS Proxy/Server, 1243 it sets the OAL destination to the ADM-ULA of the Proxy/Server if the 1244 Proxy/Server is responsible for reassembly; otherwise, it sets the 1245 OAL destination to the MNP-ULA of the target Client to cause the 1246 Proxy/Server to forward without reassembling. The node also sets 1247 omIndex to select a specific target Client underlying interface, or 1248 sets omIndex to 0 when no preference is selected. 1250 When an AERO node's OMNI interface has an original IP packet to send 1251 to a route optimization target located in a remote OMNI link segment, 1252 it acts as an OAL source the same as above but also includes an 1253 appropriate ORH type with Segments Left set to 1 and with SRT/LHS/ 1254 L2ADDR information while setting the OAL destination to the Subnet 1255 Router Anycast address for the LHS OMNI link segment. (The OAL 1256 source can alternatively include an ORH with Segments Left set to 0 1257 while setting the OAL destination to the ADM-ULA of the LHS Proxy/ 1258 Server, but this would cause the carrier packets to follow strict 1259 spanning tree paths.) The OMNI interface then forwards the resulting 1260 carrier packets into the spanning tree. 1262 When a Bridge receives a carrier packet destined to its Subnet Router 1263 Anycast address with any ORH type with Segments Left set to 1 and 1264 with SRT/LHS/L2ADDR values corresponding to the local segment, it 1265 examines FMT-Mode to determine whether the target Client can accept 1266 packets directly (i.e., following any NAT traversal procedures 1267 necessary) while bypassing the LHS Proxy/Server. If the Client can 1268 be reached directly and NAT traversal has converged, the Bridge then 1269 writes the MNP-ULA (found in the inner IPv6 header for first 1270 fragments or the ORH Destination Suffix for non-first fragments) into 1271 the OAL destination address, decrements the OAL IPv6 header Hop Limit 1272 (and discards the packet if Hop Limit reaches 0), removes the ORH, 1273 re-encapsulates the carrier packet according to L2ADDR then forwards 1274 the carrier packet directly to the target Client. If the Client 1275 cannot be reached directly (or if NAT traversal has not yet 1276 converged), the Bridge instead transforms the ORH into an ORH-0, re- 1277 encapsulates the packet according to L2ADDR, changes the OAL 1278 destination to the ADM-ULA of the LHS Proxy/Server if FMT-Forward is 1279 clear or the MNP-ULA of the Client if FMT-Forward is set and forwards 1280 the carrier packet to the LHS Proxy/Server. 1282 When a Bridge receives a carrier packet destined to its Subnet Router 1283 Anycast address with any ORH type with Segments Left set to 1 and 1284 L2ADDR set to 0, the Bridge instead forwards the packet toward the 1285 LHS Proxy/Server via the spanning tree. The Bridge changes the OAL 1286 destination to the ADM-ULA of the LHS Proxy/Server, transforms the 1287 ORH into an ORH-0 (or an ORH-1 with FMT-Type set and Segments Left 1288 0), then forwards the packet to the next hop in the spanning tree. 1289 The Bridge may also elect to forward via the spanning tree as above 1290 even when it receives a carrier packet with an ORH that includes a 1291 valid L2ADDR Port Number and IP address, however this may result in a 1292 longer path than necessary. If the carrier packet arrived via the 1293 secured spanning tree, the Bridge forwards to the next hop also via 1294 the secured spanning tree. If the carrier packet arrived via the 1295 unsecured spanning tree, the Bridge forwards to the next hop also via 1296 the unsecured spanning tree. 1298 When an LHS Proxy/Server receives carrier packets with any ORH type 1299 with Segments Left set to 0 and with OAL destination set to its own 1300 ADM-ULA, it proceeds according to FMT-Forward and omIndex. If FMT- 1301 Forward is set, the LHS Proxy/Server changes the OAL destination to 1302 the MNP-ULA of the target Client found in the IPv6 header for first 1303 fragments or in the ORH Destination Suffix for non-first-fragments, 1304 removes the ORH and forwards to the target Client interface 1305 identified by omIndex. If FMT-Forward is clear, the LHS Proxy/Server 1306 instead reassembles then re-encapsulates while refragmenting if 1307 necessary, removes the ORH and forwards to the target Client 1308 according to omIndex. 1310 When an LHS Proxy/Server receives carrier packets with any ORH type 1311 with Segments Left set to 0 and with OAL destination set to the MNP- 1312 ULA of the target Client, it removes the ORH and forwards to the 1313 target Client according to omIndex. During forwarding, the LHS 1314 Proxy/Server must first verify that the omIndex corresponds to a 1315 target underlying interface that it services locally and must not 1316 forward to other target underlying interfaces. If omIndex is 0 (or 1317 if no ORH is included) the LHS Proxy/Server instead selects among any 1318 of the target underlying interfaces it services. 1320 When a target Client receives carrier packets with OAL destination 1321 set to is MNP-ULA, it reassembles to obtain the OAL packet then 1322 decapsulates and delivers the original IP packet to upper layers. 1324 Note: Special handling procedures are employed for the exchange of 1325 IPv6 ND messages used to establish neighbor cache state as discussed 1326 in Section 3.14. The procedures call for hop-by-hop authentication 1327 and neighbor cache state establishment based on the encapsulation 1328 ULA, with next-hop determination based on the IPv6 ND message LLA. 1330 3.3. OMNI Interface Characteristics 1332 OMNI interfaces are virtual interfaces configured over one or more 1333 underlying interfaces classified as follows: 1335 o INET interfaces connect to an INET either natively or through one 1336 or more NATs. Native INET interfaces have global IP addresses 1337 that are reachable from any INET correspondent. The INET-facing 1338 interfaces of Proxy/Servers are native interfaces, as are Relay 1339 and Bridge interfaces. NATed INET interfaces connect to a private 1340 network behind one or more NATs that provide INET access. Clients 1341 that are behind a NAT are required to send periodic keepalive 1342 messages to keep NAT state alive when there are no carrier packets 1343 flowing. 1345 o ANET interfaces connect to an ANET that is separated from the open 1346 INET by an FHS Proxy/Server. Clients can issue control messages 1347 over the ANET without including an authentication signature since 1348 the ANET is secured at the network layer or below. Proxy/Servers 1349 can actively issue control messages over the INET on behalf of 1350 ANET Clients to reduce ANET congestion. 1352 o VPNed interfaces use security encapsulation over the INET to a 1353 Virtual Private Network (VPN) server that also acts as an FHS 1354 Proxy/Server. Other than the link-layer encapsulation format, 1355 VPNed interfaces behave the same as Direct interfaces. 1357 o Direct (i.e., single-hop point-to-point) interfaces connect a 1358 Client directly to an FHS Proxy/Server without crossing any ANET/ 1359 INET paths. An example is a line-of-sight link between a remote 1360 pilot and an unmanned aircraft. The same Client considerations 1361 apply as for VPNed interfaces. 1363 OMNI interfaces use OAL encapsulation and fragmentation as discussed 1364 in Section 3.2.4. OMNI interfaces use *NET encapsulation (see: 1366 Section 3.6) to exchange carrier packets with OMNI link neighbors 1367 over INET or VPNed interfaces as well as over ANET interfaces for 1368 which the Client and FHS Proxy/Server may be multiple IP hops away. 1369 OMNI interfaces do not use link-layer encapsulation over Direct 1370 underlying interfaces or ANET interfaces when the Client and FHS 1371 Proxy/Server are known to be on the same underlying link. 1373 OMNI interfaces maintain a neighbor cache for tracking per-neighbor 1374 state the same as for any interface. OMNI interfaces use ND messages 1375 including Router Solicitation (RS), Router Advertisement (RA), 1376 Neighbor Solicitation (NS) and Neighbor Advertisement (NA) for 1377 neighbor cache management. In environments where spoofing may be a 1378 threat, OMNI neighbors should employ OAL Identification window 1379 synchronization in their ND message exchanges. 1381 OMNI interfaces send ND messages with an OMNI option formatted as 1382 specified in [I-D.templin-6man-omni]. The OMNI option includes 1383 prefix registration information, Interface Attributes containing link 1384 information parameters for the OMNI interface's underlying interfaces 1385 and any other per-neighbor information. Each OMNI option may include 1386 multiple Interface Attributes sub-options identified by non-zero 1387 omIndex values. 1389 A Client's OMNI interface may be configured over multiple underlying 1390 interface connections. For example, common mobile handheld devices 1391 have both wireless local area network ("WLAN") and cellular wireless 1392 links. These links are often used "one at a time" with low-cost WLAN 1393 preferred and highly-available cellular wireless as a standby, but a 1394 simultaneous-use capability could provide benefits. In a more 1395 complex example, aircraft frequently have many wireless data link 1396 types (e.g. satellite-based, cellular, terrestrial, air-to-air 1397 directional, etc.) with diverse performance and cost properties. 1399 If a Client's multiple underlying interfaces are used "one at a time" 1400 (i.e., all other interfaces are in standby mode while one interface 1401 is active), then successive ND messages all include OMNI option 1402 Interface Attributes sub-options with the same underlying interface 1403 index. In that case, the Client would appear to have a single 1404 underlying interface but with a dynamically changing link-layer 1405 address. 1407 If the Client has multiple active underlying interfaces, then from 1408 the perspective of ND it would appear to have multiple link-layer 1409 addresses. In that case, ND message OMNI options MAY include 1410 Interface Attributes sub-options with different underlying interface 1411 indexes. Every ND message need not include Interface Attributes for 1412 all underlying interfaces; for any attributes not included, the 1413 neighbor considers the status as unchanged. 1415 Bridge and Proxy/Server OMNI interfaces are configured over secured 1416 tunnel underlying interfaces for carrying IPv6 ND and BGP protocol 1417 control plane messages, plus open INET underlying interfaces for 1418 carrying unsecured messages. The OMNI interface configures both an 1419 ADM-LLA and its corresponding ADM-ULA, and acts as an OAL source to 1420 encapsulate and fragment original IP packets while presenting the 1421 resulting carrier packets to a secured or unsecured underlying 1422 interface. Note that Bridge and Proxy/Server BGP protocol TCP 1423 sessions are run directly over the OMNI interface using ADM-ULA 1424 source and destination addresses. The OMNI interface encapsulates 1425 the original IP packets for these sessions as carrier packets (i.e., 1426 even though the OAL header may use the same ADM-ULAs as the original 1427 IP header) and forwards them over a secured underlying interface. 1429 3.4. OMNI Interface Initialization 1431 AERO Proxy/Servers and Clients configure OMNI interfaces as their 1432 point of attachment to the OMNI link. AERO nodes assign the MSPs for 1433 the link to their OMNI interfaces (i.e., as a "route-to-interface") 1434 to ensure that original IP packets with destination addresses covered 1435 by an MNP not explicitly assigned to a non-OMNI interface are 1436 directed to the OMNI interface. 1438 OMNI interface initialization procedures for Proxy/Servers, Clients 1439 and Bridges are discussed in the following sections. 1441 3.4.1. AERO Proxy/Server and Relay Behavior 1443 When a Proxy/Server enables an OMNI interface, it assigns an 1444 ADM-{LLA,ULA} appropriate for the given OMNI link segment. The 1445 Proxy/Server also configures secured tunnels with one or more 1446 neighboring Bridges and engages in a BGP routing protocol session 1447 with each Bridge. 1449 The OMNI interface provides a single interface abstraction to the IP 1450 layer, but internally includes an NBMA nexus for sending carrier 1451 packets to OMNI interface neighbors over underlying INET interfaces 1452 and secured tunnels. The Proxy/Server further configures a service 1453 to facilitate ND exchanges with AERO Clients and manages per-Client 1454 neighbor cache entries and IP forwarding table entries based on 1455 control message exchanges. 1457 Relays are simply Proxy/Servers that run a dynamic routing protocol 1458 to redistribute routes between the OMNI interface and INET/EUN 1459 interfaces (see: Section 3.2.3). The Relay provisions MNPs to 1460 networks on the INET/EUN interfaces (i.e., the same as a Client would 1461 do) and advertises the MSP(s) for the OMNI link over the INET/EUN 1462 interfaces. The Relay further provides an attachment point of the 1463 OMNI link to a non-MNP-based global topology. 1465 3.4.2. AERO Client Behavior 1467 When a Client enables an OMNI interface, it assigns either an 1468 MNP-{LLA, ULA} or a Temporary ULA and sends RS messages with ND 1469 parameters over its underlying interfaces to an FHS Proxy/Server, 1470 which returns an RA message with corresponding parameters. The RS/RA 1471 messages may pass through one or more NATs in the case of a Client's 1472 INET interface. (Note: if the Client used a Temporary ULA in its 1473 initial RS message, it will discover an MNP-{LLA, ULA} in the 1474 corresponding RA that it receives from the FHS Proxy/Server and begin 1475 using these new addresses. If the Client is operating outside the 1476 context of AERO infrastructure such as in a Mobile Ad-hoc Network 1477 (MANET), however, it may continue using Temporary ULAs for Client-to- 1478 Client communications until it encounters an infrastructure element 1479 that can provide an MNP.) 1481 3.4.3. AERO Bridge Behavior 1483 AERO Bridges configure an OMNI interface and assign the ADM-ULA 1484 Subnet Router Anycast address for each OMNI link segment they connect 1485 to. Bridges configure secured tunnels with Proxy/Servers and other 1486 Bridges, and engage in a BGP routing protocol session with neighbors 1487 on the spanning tree (see: Section 3.2.3). 1489 3.5. OMNI Interface Neighbor Cache Maintenance 1491 Each OMNI interface maintains a conceptual neighbor cache that 1492 includes a Neighbor Cache Entry (NCE) for each of its active 1493 neighbors on the OMNI link per [RFC4861]. Each route optimization 1494 source NCE is indexed by the LLA of the neighbor, while the OAL 1495 encapsulation ULA determines the context for Identification 1496 verification. In addition to ordinary neighbor cache entries, proxy 1497 neighbor cache entries are created and maintained by AERO Proxy/ 1498 Servers when they proxy Client ND message exchanges [RFC4389]. AERO 1499 Proxy/Servers maintain proxy neighbor cache entries for each of their 1500 associated Clients. 1502 To the list of NCE states in Section 7.3.2 of [RFC4861], Proxy/Server 1503 OMNI interfaces add an additional state DEPARTED that applies to 1504 Clients that have recently departed. The interface sets a 1505 "DepartTime" variable for the NCE to "DEPART_TIME" seconds. 1506 DepartTime is decremented unless a new ND message causes the state to 1507 return to REACHABLE. While a NCE is in the DEPARTED state, the 1508 Proxy/Server forwards carrier packets destined to the target Client 1509 to the Client's new location instead. When DepartTime decrements to 1510 0, the NCE is deleted. It is RECOMMENDED that DEPART_TIME be set to 1511 the default constant value REACHABLE_TIME plus 10 seconds (40 seconds 1512 by default) to allow a window for carrier packets in flight to be 1513 delivered while stale route optimization state may be present. 1515 Proxy/Servers can act as RORs on behalf of dependent Clients 1516 according to the Proxy Neighbor Advertisement specification in 1517 Section 7.2.8 of [RFC4861]. When a Proxy/Server ROR receives an 1518 authentic NS message used for route optimization, it first searches 1519 for a NCE for the target Client and accepts the message only if there 1520 is an entry. The Proxy/Server then returns a solicited NA message 1521 while creating or updating a "Report List" entry in the target 1522 Client's NCE that caches both the LLA and ULA of ROS with a 1523 "ReportTime" variable set to REPORT_TIME seconds. The ROR resets 1524 ReportTime when it receives a new authentic NS message, and otherwise 1525 decrements ReportTime while no authentic NS messages have been 1526 received. It is RECOMMENDED that REPORT_TIME be set to the default 1527 constant value REACHABLE_TIME plus 10 seconds (40 seconds by default) 1528 to allow a window for route optimization to converge before 1529 ReportTime decrements below REACHABLE_TIME. 1531 When the ROS receives a solicited NA message response to its NS 1532 message used for route optimization, it creates or updates a NCE for 1533 the target network-layer and link-layer addresses. The ROS then 1534 (re)sets ReachableTime for the NCE to REACHABLE_TIME seconds and uses 1535 this value to determine whether carrier packets can be forwarded 1536 directly to the target, i.e., instead of via a default route. The 1537 ROS otherwise decrements ReachableTime while no further solicited NA 1538 messages arrive. It is RECOMMENDED that REACHABLE_TIME be set to the 1539 default constant value 30 seconds as specified in [RFC4861]. 1541 AERO nodes also use the value MAX_UNICAST_SOLICIT to limit the number 1542 of NS messages sent when a correspondent may have gone unreachable, 1543 the value MAX_RTR_SOLICITATIONS to limit the number of RS messages 1544 sent without receiving an RA and the value MAX_NEIGHBOR_ADVERTISEMENT 1545 to limit the number of unsolicited NAs that can be sent based on a 1546 single event. It is RECOMMENDED that MAX_UNICAST_SOLICIT, 1547 MAX_RTR_SOLICITATIONS and MAX_NEIGHBOR_ADVERTISEMENT be set to 3 the 1548 same as specified in [RFC4861]. 1550 Different values for DEPART_TIME, REPORT_TIME, REACHABLE_TIME, 1551 MAX_UNICAST_SOLICIT, MAX_RTR_SOLCITATIONS and 1552 MAX_NEIGHBOR_ADVERTISEMENT MAY be administratively set; however, if 1553 different values are chosen, all nodes on the link MUST consistently 1554 configure the same values. Most importantly, DEPART_TIME and 1555 REPORT_TIME SHOULD be set to a value that is sufficiently longer than 1556 REACHABLE_TIME to avoid packet loss due to stale route optimization 1557 state. 1559 3.5.1. OMNI ND Messages 1561 OMNI interfaces prepare IPv6 ND messages the same as for standard 1562 IPv6 ND, but also include a new option type termed the OMNI option 1563 [I-D.templin-6man-omni]. When an OMNI interface prepares an IPv6 ND 1564 message, it includes any necessary options including one or more OMNI 1565 options and completely populates all option information. The OMNI 1566 interface then calculates the checksum over the IPv6 ND message and 1567 writes the checksum in the IPv6 ND message header. If an 1568 intermediate node alters an IPv6 ND message OMNI option in-transit, 1569 it must re-calculate the IPv6 ND message checksum before forwarding 1570 further. OMNI interfaces verify the checksum of each IPv6 ND message 1571 received, and process the message further only if the checksum is 1572 correct. 1574 OMNI options include per-neighbor information such as Interface 1575 Attributes that provide Link-Layer Address and traffic selector 1576 information for the neighbor's underlying interfaces. This 1577 information is stored in the neighbor cache and provides the basis 1578 for the forwarding algorithm specified in Section 3.10. The 1579 information is cumulative and reflects the union of the OMNI 1580 information from the most recent ND messages received from the 1581 neighbor; it is therefore not required that each ND message contain 1582 all neighbor information. 1584 The OMNI option Interface Attributes for each underlying interface 1585 includes a two-part "Link-Layer Address" consisting of an INET 1586 encapsulation address determined by the FMT and L2ADDR fields and an 1587 ADM-ULA determined by the SRT and LHS fields. Underlying interfaces 1588 are further selected based on their associated traffic selectors. 1590 The OMNI option is distinct from any Source/Target Link-Layer Address 1591 Options (S/TLLAOs) that may appear in an ND message according to the 1592 appropriate IPv6 over specific link layer specification (e.g., 1593 [RFC2464]). If both an OMNI option and S/TLLAO appear, the former 1594 pertains to encapsulation addresses while the latter pertains to the 1595 native L2 address format of the underlying media 1597 OMNI interface IPv6 ND messages may also include other IPv6 ND 1598 options. In particular, solicitation messages may include Nonce and/ 1599 or Timestamp options if required for verification of advertisement 1600 replies. If an OMNI ND solicitation message includes a Nonce option, 1601 the advertisement reply must echo the same Nonce. If an OMNI ND 1602 solicitation message includes a Timestamp option, the advertisement 1603 reply should also include a Timestamp option. 1605 AERO Clients send RS messages to the All-Routers multicast address 1606 while using unicast link-layer addresses. AERO Proxy/Servers respond 1607 by returning unicast RA messages. During the RS/RA exchange, AERO 1608 Clients and Servers include state synchronization parameters to 1609 establish Identification windows and other state. 1611 AERO nodes use NS/NA messages for the following purposes: 1613 o NS/NA(AR) messages are used for address resolution only. The ROS 1614 sends an NS(AR) to the solicited-node multicast address of the 1615 target, and an ROR in the network with addressing information for 1616 the target returns a unicast NA(AR). The NA(AR) contains 1617 authentic and current target address resolution information, but 1618 only an implicit third-party assertion of target reachability. 1619 NS/NA(AR) messages must be secured. 1621 o NS/NA(WIN) messages are used for establishing and maintaining 1622 window synchronization (and any other) state. The source sends an 1623 NS(WIN) to the unicast address of the target, and the target 1624 returns a unicast NA(WIN). The NS/NA(WIN) exchange synchronizes 1625 sequence numbers the neighbors will include in subsequent packets, 1626 and asserts reachability for the target without necessarily 1627 testing a specific underlying interface pair. NS/NA(WIN) messages 1628 must be secured. 1630 o NS/NA(NUD) messages are used for determining target reachability. 1631 The source sends an NS(NUD) to the unicast address of the target 1632 while naming a specific underlying interface pair, and the target 1633 returns a unicast NA(NUD). NS/NA(NUD) messages that use an in- 1634 window sequence number and do not update any other state need not 1635 be secured. NS/NA(NUD) messages may also be used in combination 1636 with window synchronization (i.e., NUD+WIN), in which case the 1637 messages must be secured. 1639 o Unsolicited NA (uNA) messages are used to signal addressing and/or 1640 other neighbor state changes (e.g., due to mobility, signal 1641 degradation, traffic selector updates, etc.). uNA messages that 1642 include state update information must be secured. 1644 o NS/NA(DAD) messages are not used in AERO, since Duplicate Address 1645 Detection is not required. 1647 Additionally, nodes may send NA/RA messages with the OMNI option PNG 1648 flag set to receive a solicited NA response from the neighbor. The 1649 solicited NA response MUST set the ACK flag (without also setting the 1650 SYN or PNG flags) and include the Identification used in the PNG 1651 message in the Acknowledgement. 1653 3.5.2. OMNI Neighbor Advertisement Message Flags 1655 As discussed in Section 4.4 of [RFC4861] NA messages include three 1656 flag bits R, S and O. OMNI interface NA messages treat the flags as 1657 follows: 1659 o R: The R ("Router") flag is set to 1 in the NA messages sent by 1660 all AERO/OMNI node types. Simple hosts that would set R to 0 do 1661 not occur on the OMNI link itself, but may occur on the downstream 1662 links of Clients and Relays. 1664 o S: The S ("Solicited") flag is set exactly as specified in 1665 Section 4.4. of [RFC4861], i.e., it is set to 1 for Solicited NAs 1666 and set to 0 for uNAs (both unicast and multicast). 1668 o O: The O ("Override") flag is set to 0 for solicited NAs returned 1669 by a Proxy/Server ROR and set to 1 for all other solicited and 1670 unsolicited NAs. For further study is whether solicited NAs for 1671 anycast targets apply for OMNI links. Since MNP-LLAs must be 1672 uniquely assigned to Clients to support correct ND protocol 1673 operation, however, no role is currently seen for assigning the 1674 same MNP-LLA to multiple Clients. 1676 3.5.3. OMNI Neighbor Window Synchronization 1678 In secured environments (e.g., such as between nodes on the same 1679 secured ANET), OMNI interface neighbors can exchange OAL packets 1680 using randomly-initialized and monotonically-increasing 1681 Identification values (modulo 2*32) without window synchronization. 1682 In environments where spoofing is considered a threat, OMNI interface 1683 neighbors instead invoke window synchronization in ND message 1684 exchanges to maintain send/receive window state in their respective 1685 neighbor cache entries as specified in [I-D.templin-6man-omni]. 1687 In the asymmetric window synchronization case, the initial ND message 1688 exchange establishes only the initiator's send window and the 1689 responder's receive window such that a corresponding exchange would 1690 be needed to establish the reverse direction. In the symmetric case, 1691 the initiator and responder engage in a three-way handshake to 1692 symmetrically establish the send/receive windows of both parties. 1694 3.6. OMNI Interface Encapsulation and Re-encapsulation 1696 The OMNI interface admits original IP packets then (acting as an OAL 1697 source) performs OAL encapsulation and fragmentation as specified in 1698 [I-D.templin-6man-omni] while including an ORH if necessary as 1699 specified in Section 3.2.4. OAL encapsulation produces OAL packets 1700 subject to fragmentation, with the resulting fragments encapsulated 1701 in *NET headers as carrier packets. 1703 For carrier packets undergoing re-encapsulation at an OAL 1704 intermediate node, the OMNI interface decrements the OAL IPv6 header 1705 Hop Limit and discards the carrier packet if the Hop Limit reaches 0. 1706 The intermediate node next removes the *NET encapsulation headers 1707 from the first segment and re-encapsulates the packet in new *NET 1708 encapsulation headers for the next segment. 1710 When an FHS Proxy/Server re-encapsulates a carrier packet received 1711 from a Client that includes an OAL but no ORH, it inserts an ORH 1712 immediately following the OAL header and adjusts the OAL payload 1713 length and destination address field. The ORH will be removed by an 1714 LHS Bridge or Proxy/Server, but its insertion and removal will not 1715 interfere with reassembly at the final destination. For this reason, 1716 Clients must reserve 40 bytes for a maximum-length ORH when they 1717 perform OAL encapsulation (see: Section 3.9). 1719 3.7. OMNI Interface Decapsulation 1721 OMNI interfaces (acting as OAL destinations) decapsulate and 1722 reassemble OAL packets into original IP packets destined either to 1723 the AERO node itself or to a destination reached via an interface 1724 other than the OMNI interface the original IP packet was received on. 1725 When carrier packets containing OAL fragments addressed to itself 1726 arrive, the OMNI interface discards the NET encapsulation headers and 1727 reassembles as discussed in Section 3.9. 1729 3.8. OMNI Interface Data Origin Authentication 1731 AERO nodes employ simple data origin authentication procedures. In 1732 particular: 1734 o AERO Bridges and Proxy/Servers accept carrier packets received 1735 from secured underlying interfaces. 1737 o AERO Proxy/Servers and Clients accept carrier packets and original 1738 IP packets that originate from within the same secured ANET. 1740 o AERO Clients and Relays accept original IP packets from downstream 1741 network correspondents based on ingress filtering. 1743 o AERO Clients, Relays and Proxy/Servers verify carrier packet UDP/ 1744 IP encapsulation addresses according to [I-D.templin-6man-omni]. 1746 o AERO nodes accept carrier packets addressed to themselves with 1747 Identification values within the current window for the OAL source 1748 neighbor (when window synchronization is used) and drop any 1749 carrier packets with out-of-window Identification values. (AERO 1750 nodes may forward carrier packets not addressed to themselves 1751 without verifying the Identification value.) 1753 AERO nodes silently drop any packets that do not satisfy the above 1754 data origin authentication procedures. Further security 1755 considerations are discussed in Section 6. 1757 3.9. OMNI Interface MTU 1759 The OMNI interface observes the link nature of tunnels, including the 1760 Maximum Transmission Unit (MTU), Maximum Reassembly Unit (MRU) and 1761 the role of fragmentation and reassembly [I-D.ietf-intarea-tunnels]. 1762 The OMNI interface employs an OMNI Adaptation Layer (OAL) that 1763 accommodates multiple underlying links with diverse MTUs while 1764 observing both a minimum and per-path Maximum Payload Size (MPS). 1765 The functions of the OAL and the OMNI interface MTU/MRU/MPS are 1766 specified in [I-D.templin-6man-omni] with MTU/MRU both set to the 1767 constant value 9180 bytes, with minimum MPS set to 400 bytes, and 1768 with per-path MPS set to potentially larger values depending on the 1769 underlying path. 1771 When the network layer presents an original IP packet to the OMNI 1772 interface, the OAL source encapsulates and fragments the original IP 1773 packet if necessary. When the network layer presents the OMNI 1774 interface with multiple original IP packets bound to the same OAL 1775 destination, the OAL source can concatenate them together into a 1776 single OAL super-packet as discussed in [I-D.templin-6man-omni]. The 1777 OAL source then fragments the OAL packet if necessary according to 1778 the minimum/path MPS such that the OAL headers appear in each 1779 fragment while the original IP packet header appears only in the 1780 first fragment. The OAL source then encapsulates each OAL fragment 1781 in *NET headers for transmission as carrier packets over an 1782 underlying interface connected to either a physical link such as 1783 Ethernet, WiFi and the like or a virtual link such as an Internet or 1784 higher-layer tunnel (see the definition of link in [RFC8200]). 1786 Note: A Client that does not (yet) have neighbor cache state for a 1787 target may omit the ORH in carrier packets with the understanding 1788 that a Proxy/Server may insert an ORH on its behalf. For this 1789 reason, Clients reserve 40 bytes for the largest possible ORH in 1790 their OAL fragment size calculations. 1792 Note: Although the ORH may be removed or replaced by a Bridge or 1793 Proxy/Server on the path (see: Section 3.10.3), this does not 1794 interfere with the destination's ability to reassemble since the ORH 1795 is not included in the fragmentable part and its removal/ 1796 transformation does not invalidate fragment header information. 1798 3.10. OMNI Interface Forwarding Algorithm 1800 Original IP packets enter a node's OMNI interface either from the 1801 network layer (i.e., from a local application or the IP forwarding 1802 system) while carrier packets enter from the link layer (i.e., from 1803 an OMNI interface neighbor). All original IP packets and carrier 1804 packets entering a node's OMNI interface first undergo data origin 1805 authentication as discussed in Section 3.8. Those that satisfy data 1806 origin authentication are processed further, while all others are 1807 dropped silently. 1809 Original IP packets that enter the OMNI interface from the network 1810 layer are forwarded to an OMNI interface neighbor using OAL 1811 encapsulation and fragmentation to produce carrier packets for 1812 transmission over underlying interfaces. (If routing indicates that 1813 the original IP packet should instead be forwarded back to the 1814 network layer, the packet is dropped to avoid looping). Carrier 1815 packets that enter the OMNI interface from the link layer are either 1816 re-encapsulated and re-admitted into the OMNI link, or reassembled 1817 and forwarded to the network layer where they are subject to either 1818 local delivery or IP forwarding. In all cases, the OAL MUST NOT 1819 decrement the network layer TTL/Hop-count since its forwarding 1820 actions occur below the network layer. 1822 OMNI interfaces may have multiple underlying interfaces and/or 1823 neighbor cache entries for neighbors with multiple underlying 1824 interfaces (see Section 3.3). The OAL uses Interface Attribute 1825 traffic selectors (e.g., port number, flow specification, etc.) to 1826 select an outbound underlying interface for each OAL packet based on 1827 the node's own interface attributes, and also to select a destination 1828 link-layer address based on the neighbor's underlying interface 1829 attributes. AERO implementations SHOULD permit network management to 1830 dynamically adjust traffic selector values at runtime. 1832 If an OAL packet matches the traffic selectors of multiple outgoing 1833 interfaces and/or neighbor interfaces, the OMNI interface replicates 1834 the packet and sends one copy via each of the (outgoing / neighbor) 1835 interface pairs; otherwise, it sends a single copy of the OAL packet 1836 via an interface with the best matching traffic selector. (While not 1837 strictly required, the likelihood of successful reassembly may 1838 improve when the OMNI interface sends all fragments of the same 1839 fragmented OAL packet consecutively over the same underlying 1840 interface pair to avoid complicating factors such as delay variance 1841 and reordering.) AERO nodes keep track of which underlying 1842 interfaces are currently "reachable" or "unreachable", and only use 1843 "reachable" interfaces for forwarding purposes. 1845 The following sections discuss the OMNI interface forwarding 1846 algorithms for Clients, Proxy/Servers and Bridges. In the following 1847 discussion, an original IP packet's destination address is said to 1848 "match" if it is the same as a cached address, or if it is covered by 1849 a cached prefix (which may be encoded in an MNP-LLA). 1851 3.10.1. Client Forwarding Algorithm 1853 When an original IP packet enters a Client's OMNI interface from the 1854 network layer the Client searches for a NCE that matches the 1855 destination. If there is a match, the Client selects one or more 1856 "reachable" neighbor interfaces in the entry for forwarding purposes. 1857 If there is no NCE, the Client instead either enqueues the original 1858 IP packet and invokes route optimization or forwards the original IP 1859 packet toward a Proxy/Server. The Client (acting as an OAL source) 1860 performs OAL encapsulation and sets the OAL destination address to 1861 the MNP-ULA of the target if there is a matching NCE; otherwise, it 1862 sets the OAL destination to the ADM-ULA of the Proxy/Server. If the 1863 Client has multiple original IP packets to send to the same neighbor, 1864 it can concatenate them in a single super-packet 1865 [I-D.templin-6man-omni]. The OAL source then performs fragmentation 1866 to create OAL fragments (see: Section 3.9), appends any *NET 1867 encapsulation, and sends the resulting carrier packets over 1868 underlying interfaces to the neighbor acting as an OAL destination. 1870 If the neighbor interface selected for forwarding is located on the 1871 same OMNI link segment and not behind a NAT, the Client forwards the 1872 carrier packets directly according to the L2ADDR information for the 1873 neighbor. If the neighbor interface is behind a NAT on the same OMNI 1874 link segment, the Client instead forwards the initial carrier packets 1875 to the LHS Proxy/Server (while inserting an ORH-0 if necessary) and 1876 initiates NAT traversal procedures. If the Client's intended source 1877 underlying interface is also behind a NAT and located on the same 1878 OMNI link segment, it sends a "direct bubble" over the interface per 1879 [RFC6081][RFC4380] to the L2ADDR found in the neighbor cache in order 1880 to establish state in its own NAT by generating traffic toward the 1881 neighbor (note that no response to the bubble is expected). 1883 The Client next sends an NS(NUD) message toward the MNP-ULA of the 1884 neighbor via the LHS Proxy/Server as discussed in Section 3.15. If 1885 the Client receives an NA(NUD) from the neighbor over the underlying 1886 interface, it marks the neighbor interface as "trusted" and sends 1887 future carrier packets directly to the L2ADDR information for the 1888 neighbor instead of indirectly via the LHS Proxy/Server. The Client 1889 must honor the neighbor cache maintenance procedure by sending 1890 additional direct bubbles and/or NS/NA(NUD) messages as discussed in 1891 [RFC6081][RFC4380] in order to keep NAT state alive as long as 1892 carrier packets are still flowing. 1894 When a carrier packet enters a Client's OMNI interface from the link- 1895 layer, if the OAL destination matches one of the Client's ULAs the 1896 Client (acting as an OAL destination) verifies that the 1897 Identification is in-window for this OAL source, then reassembles and 1898 decapsulates as necessary and delivers the original IP packet to the 1899 network layer. Otherwise, the Client drops the original IP packet 1900 and MAY return a network-layer ICMP Destination Unreachable message 1901 subject to rate limiting (see: Section 3.11). 1903 Note: Clients and their FHS Proxy/Server (and other Client) peers can 1904 exchange original IP packets over ANET underlying interfaces without 1905 invoking the OAL, since the ANET is secured at the link and physical 1906 layers. By forwarding original IP packets without invoking the OAL, 1907 however, the ANET peers can engage only in classical path MTU 1908 discovery since the packets are subject to loss and/or corruption due 1909 to the various per-link MTU limitations that may occur within the 1910 ANET. Moreover, the original IP packets do not include either the 1911 OAL integrity check or per-packet Identification values that can be 1912 used for data origin authentication and link-layer retransmissions. 1913 The tradeoff therefore involves an assessment of the per-packet 1914 encapsulation overhead saved by bypassing the OAL vs. inheritance of 1915 classical network "brittleness". (Note however that ANET peers can 1916 send small original IP packets without invoking the OAL, while 1917 invoking the OAL for larger packets. This presents the beneficial 1918 aspects of both small packet efficiency and large packet robustness, 1919 with delay variance and reordering as possible side effects.) 1921 3.10.2. Proxy/Server and Relay Forwarding Algorithm 1923 When the Proxy/Server receives an original IP packet from the network 1924 layer, it drops the packet if routing indicates that it should be 1925 forwarded back to the network layer to avoid looping. Otherwise, the 1926 Proxy/Server regards the original IP packet the same as if it had 1927 arrived as carrier packets with OAL destination set to its own ADM- 1928 ULA. When the Proxy/Server receives carrier packets on underlying 1929 interfaces with OAL destination set to its own ADM-ULA, it performs 1930 OAL reassembly if necessary to obtain the original IP packet. 1932 The Proxy/Server next searches for a NCE that matches the original IP 1933 destination and proceeds as follows: 1935 o if the original IP packet destination matches a NCE, the Proxy/ 1936 Sever uses one or more "reachable" neighbor interfaces in the 1937 entry for packet forwarding using OAL encapsulation and 1938 fragmentation according to the cached link-layer address 1939 information. If the neighbor interface is in a different OMNI 1940 link segment, the Proxy/Server performs OAL encapsulation and 1941 fragmentation, inserts an ORH and forwards the resulting carrier 1942 packets via the spanning tree to a Bridge; otherwise, it forwards 1943 the carrier packets directly to the neighbor. If the neighbor is 1944 behind a NAT, the Proxy/Server instead forwards initial carrier 1945 packets via a Bridge while sending an NS(NUD) to the neighbor. 1946 When the Proxy/Server receives the NA(NUD), it can begin 1947 forwarding carrier packets directly to the neighbor the same as 1948 discussed in Section 3.10.1 while sending additional NS(NUD) 1949 messages as necessary to maintain NAT state. Note that no direct 1950 bubbles are necessary since the Proxy/Server is by definition not 1951 located behind a NAT. 1953 o else, if the original IP destination matches a non-MNP route in 1954 the IP forwarding table or an ADM-LLA assigned to the Proxy/ 1955 Server's OMNI interface, the Proxy/Server acting as a Relay 1956 presents the original IP packet to the network layer for local 1957 delivery or IP forwarding. 1959 o else, the Proxy/Server initiates address resolution as discussed 1960 in Section 3.14, while retaining initial original IP packets in a 1961 small queue awaiting address resolution completion. 1963 When the Proxy/Server receives a carrier packet with OAL destination 1964 set to an MNP-ULA that does not match the MSP, it accepts the carrier 1965 packet only if data origin authentication succeeds and if there is a 1966 network layer routing table entry for a GUA route that matches the 1967 MNP-ULA. If there is no route, the Proxy/Server drops the carrier 1968 packet; otherwise, it reassembles and decapsulates to obtain the 1969 original IP packet and acts as a Relay to present it to the network 1970 layer where it will be delivered according to standard IP forwarding. 1972 When a Proxy/Server receives a carrier packet from one of its Client 1973 neighbors with OAL destination set to another node, it forwards the 1974 packets via a matching NCE or via the spanning tree if there is no 1975 matching entry. When the Proxy/Server receives a carrier packet with 1976 OAL destination set to the MNP-ULA of one of its Client neighbors 1977 established through RS/RA exchanges, it accepts the carrier packet 1978 only if data origin authentication succeeds. If the NCE state is 1979 DEPARTED, the Proxy/Server inserts an ORH that encodes the MNP-ULA 1980 destination suffix and changes the OAL destination address to the 1981 ADM-ULA of the new Proxy/Server, then re-encapsulates the carrier 1982 packet and forwards it to a Bridge which will eventually deliver it 1983 to the new Proxy/Server. 1985 If the neighbor cache state for the MNP-ULA is REACHABLE, the Proxy/ 1986 Server forwards the carrier packets to the Client which then must 1987 reassemble. (Note that the Proxy/Server does not reassemble carrier 1988 packets not explicitly addressed to its own ADM-ULA, since routing 1989 could direct some of the carrier packet of the same original IP 1990 packet through a different Proxy/Server.) In that case, the Client 1991 may receive fragments that are smaller than its link MTU but can 1992 still be reassembled. 1994 Note: Proxy/Servers may receive carrier packets with ORHs that 1995 include additional forwarding information. Proxy/Servers use the 1996 forwarding information to determine the correct interface for 1997 forwarding to the target destination Client, then remove the ORH and 1998 forward the carrier packet. If the ORH information instead indicates 1999 that the Proxy/Server is responsible for reassembly, the Proxy/Server 2000 reassembles first before re-encapsulating (and possibly also re- 2001 fragmenting) then forwards to the target Client. For a full 2002 discussion of cases when the Proxy/Server may receive carrier packets 2003 with ORHs, see: Section 3.14.6. 2005 Note: Clients and their FHS Proxy/Server peers can exchange original 2006 IP packets over ANET underlying interfaces without invoking the OAL, 2007 since the ANET is secured at the link and physical layers. By 2008 forwarding original IP packets without invoking the OAL, however, the 2009 Client and Proxy/Server can engage only in classical path MTU 2010 discovery since the packets are subject to loss and/or corruption due 2011 to the various per-link MTU limitations that may occur within the 2012 ANET. Moreover, the original IP packets do not include either the 2013 OAL integrity check or per-packet Identification values that can be 2014 used for data origin authentication and link-layer retransmissions. 2015 The tradeoff therefore involves an assessment of the per-packet 2016 encapsulation overhead saved by bypassing the OAL vs. inheritance of 2017 classical network "brittleness". (Note however that ANET peers can 2018 send small original IP packets without invoking the OAL, while 2019 invoking the OAL for larger packets. This presents the beneficial 2020 aspects of both small packet efficiency and large packet robustness.) 2022 Note: When a Proxy/Server receives a (non-OAL) original IP packet 2023 from an ANET Client, or a carrier packet with OAL destination set to 2024 its own ADM-ULA from any Client, the Proxy/Server reassembles if 2025 necessary then performs ROS functions on behalf of the Client. The 2026 Client may at some later time begin sending carrier packets to the 2027 OAL address of the actual target instead of the Proxy/Server, at 2028 which point it may begin functioning as an ROS on its own behalf and 2029 thereby "override" the Proxy/Server's ROS role. 2031 Note; Proxy/Servers drop any original IP packets (received either 2032 directly from an ANET Client or following ANET/INET Client carrier 2033 packet reassembly) with a destination that corresponds to the 2034 Client's delegated MNP. Similiarly, Proxy/Servers drop any carrier 2035 packet received with both a source and destination that correspond to 2036 the Client's delegated MNP. These checks are necessary to prevent 2037 Clients from either accidentally or intentionally establishing an 2038 endless loop that could congest Proxy/Servers and/or ANET/INET links. 2040 Note: Proxy/Servers forward secure control plane carrier packets via 2041 the SRT secured spanning tree and forwards other carrier packets via 2042 the unsecured spanning tree. When a Proxy/Server receives a carrier 2043 packet from the secured spanning tree, it considers the message as 2044 authentic without having to verify upper layer authentication 2045 signatures. When a Proxy/Server receives a carrier packet from the 2046 unsecured spanning tree, it verifies any upper layer authentication 2047 signatures and/or forwards the unsecured message toward the 2048 destination which must apply data origin authentication. 2050 Note: If the Proxy/Server has multiple original IP packets to send to 2051 the same neighbor, it can concatenate them in a single OAL super- 2052 packet [I-D.templin-6man-omni]. 2054 3.10.3. Bridge Forwarding Algorithm 2056 Bridges forward carrier packets while decrementing the OAL header Hop 2057 Count but not the original IP header Hop Count/TTL. Bridges convey 2058 carrier packets that encapsulate IPv6 ND control messages or routing 2059 protocol control messages via the secured spanning tree, and may 2060 convey carrier packets that encapsulate ordinary data via the 2061 unsecured spanning tree. When the Bridge receives a carrier packet, 2062 it removes the outer *NET header and searches for a forwarding table 2063 entry that matches the OAL destination address. The Bridge then 2064 processes the packet as follows: 2066 o if the carrier packet destination matches its ADM-ULA or the 2067 corresponding ADM-ULA Subnet Router Anycast address and the OAL 2068 header is followed by an ORH, the Bridge reassembles if necessary 2069 then sets aside the ORH and processes the carrier packet locally 2070 before forwarding. If the OAL packet contains an NA(NUD) message, 2071 the Bridge replaces the OMNI option Interface Attributes sub- 2072 option with information for its own interface while retaining the 2073 omIndex value supplied by the NA(NUD) message source. The Bridge 2074 next examines the ORH, and if FMT-Mode indicates the destination 2075 is a Client on the open *NET (or, a Client behind a NAT for which 2076 NAT traversal procedures have already converged) the Bridge writes 2077 the MNP-ULA formed from the ORH Destination Suffix into the OAL 2078 destination. The Bridge then removes the ORH and forwards the 2079 packet using encapsulation based on L2ADDR. If the LHS Proxy/ 2080 Server will forward to the Client without reassembly, the Bridge 2081 writes the MNP-ULA into the OAL destination then replaces the ORH 2082 with an ORH-0 and forwards the carrier packet to the LHS Proxy/ 2083 Server while also invoking NAT traversal procedures if necessary 2084 (noting that no direct bubbles are necessary since only the target 2085 Client and not the Bridge is behind a NAT). If the LHS Proxy/ 2086 Server must perform reassembly before forwarding to the Client, 2087 the Bridge instead writes the ADM-ULA formed from the ORH SRT/LHS 2088 into the OAL destination address, replaces the ORH with an ORH-0 2089 and forwards the carrier packet to the LHS Proxy/Server. 2091 o else, if the carrier packet destination matches its ADM-ULA or the 2092 corresponding ADM-ULA Subnet Router Anycast address and the OAL 2093 header is not followed by an ORH with Segments Left set to 1, the 2094 Bridge submits the packet for reassembly. When reassembly is 2095 complete, the Bridge submits the original packet to the IP layer 2096 to support local applications such as BGP routing protocol 2097 sessions. 2099 o else, if the carrier packet destination matches a forwarding table 2100 entry the Bridge forwards the carrier packet to the next hop. (If 2101 the destination matches an MSP without matching an MNP, however, 2102 the Bridge instead drops the packet and returns an ICMP 2103 Destination Unreachable message subject to rate limiting - see: 2104 Section 3.11). 2106 o else, the Bridge drops the packet and returns an ICMP Destination 2107 Unreachable as above. 2109 The Bridge decrements the OAL IPv6 header Hop Limit when it forwards 2110 the carrier packet (i.e., the same as for any IPv6 router) and drops 2111 the packet if the Hop Limit reaches 0. Therefore, only the Hop Limit 2112 in the OAL header is decremented and not the TTL/Hop Limit in the 2113 original IP packet header. Bridges do not insert OAL/ORH headers 2114 themselves; instead, they act as IPv6 routers and forward carrier 2115 packets based on their destination addresses while also possibly 2116 transforming larger ORHs into an ORH-0 (or removing the ORH 2117 altogether). 2119 Bridges forward carrier packets received from a first segment via the 2120 SRT secured spanning tree to the next segment also via the secured 2121 spanning tree. Bridges forward carrier packets received from a first 2122 segment via the unsecured spanning tree to the next segment also via 2123 the unsecured spanning tree. Bridges use a single IPv6 routing table 2124 that always determines the same next hop for a given OAL destination, 2125 where the secured/unsecured spanning tree is determined through the 2126 selection of the underlying interface to be used for transmission 2127 (i.e., a secured tunnel or an open INET interface). 2129 3.11. OMNI Interface Error Handling 2131 When an AERO node admits an original IP packet into the OMNI 2132 interface, it may receive link-layer or network-layer error 2133 indications. The AERO node may also receive OMNI link error 2134 indications in OAL-encapsulated uNA messages that include 2135 authentication signatures. 2137 A link-layer error indication is an ICMP error message generated by a 2138 router in the INET on the path to the neighbor or by the neighbor 2139 itself. The message includes an IP header with the address of the 2140 node that generated the error as the source address and with the 2141 link-layer address of the AERO node as the destination address. 2143 The IP header is followed by an ICMP header that includes an error 2144 Type, Code and Checksum. Valid type values include "Destination 2145 Unreachable", "Time Exceeded" and "Parameter Problem" 2146 [RFC0792][RFC4443]. (OMNI interfaces ignore link-layer IPv4 2147 "Fragmentation Needed" and IPv6 "Packet Too Big" messages for carrier 2148 packets that are no larger than the minimum/path MPS as discussed in 2149 Section 3.9, however these messages may provide useful hints of probe 2150 failures during path MPS probing.) 2152 The ICMP header is followed by the leading portion of the carrier 2153 packet that generated the error, also known as the "packet-in-error". 2154 For ICMPv6, [RFC4443] specifies that the packet-in-error includes: 2155 "As much of invoking packet as possible without the ICMPv6 packet 2156 exceeding the minimum IPv6 MTU" (i.e., no more than 1280 bytes). For 2157 ICMPv4, [RFC0792] specifies that the packet-in-error includes: 2158 "Internet Header + 64 bits of Original Data Datagram", however 2159 [RFC1812] Section 4.3.2.3 updates this specification by stating: "the 2160 ICMP datagram SHOULD contain as much of the original datagram as 2161 possible without the length of the ICMP datagram exceeding 576 2162 bytes". 2164 The link-layer error message format is shown in Figure 5: 2166 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2167 ~ ~ 2168 | IP Header of link layer | 2169 | error message | 2170 ~ ~ 2171 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2172 | ICMP Header | 2173 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 2174 ~ ~ P 2175 | carrier packet *NET and OAL | a 2176 | encapsulation headers | c 2177 ~ ~ k 2178 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ e 2179 ~ ~ t 2180 | original IP packet headers | 2181 | (first-fragment only) | i 2182 ~ ~ n 2183 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2184 ~ ~ e 2185 | Portion of the body of | r 2186 | the original IP packet | r 2187 | (all fragments) | o 2188 ~ ~ r 2189 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ --- 2191 Figure 5: OMNI Interface Link-Layer Error Message Format 2193 The AERO node rules for processing these link-layer error messages 2194 are as follows: 2196 o When an AERO node receives a link-layer Parameter Problem message, 2197 it processes the message the same as described as for ordinary 2198 ICMP errors in the normative references [RFC0792][RFC4443]. 2200 o When an AERO node receives persistent link-layer Time Exceeded 2201 messages, the IP ID field may be wrapping before earlier fragments 2202 awaiting reassembly have been processed. In that case, the node 2203 should begin including integrity checks and/or institute rate 2204 limits for subsequent packets. 2206 o When an AERO node receives persistent link-layer Destination 2207 Unreachable messages in response to carrier packets that it sends 2208 to one of its neighbor correspondents, the node should process the 2209 message as an indication that a path may be failing, and 2210 optionally initiate NUD over that path. If it receives 2211 Destination Unreachable messages over multiple paths, the node 2212 should allow future carrier packets destined to the correspondent 2213 to flow through a default route and re-initiate route 2214 optimization. 2216 o When an AERO Client receives persistent link-layer Destination 2217 Unreachable messages in response to carrier packets that it sends 2218 to one of its neighbor Proxy/Servers, the Client should mark the 2219 path as unusable and use another path. If it receives Destination 2220 Unreachable messages on many or all paths, the Client should 2221 associate with a new Proxy/Server and release its association with 2222 the old Proxy/Server as specified in Section 3.16.5. 2224 o When an AERO Proxy/Server receives persistent link-layer 2225 Destination Unreachable messages in response to carrier packets 2226 that it sends to one of its neighbor Clients, the Proxy/Server 2227 should mark the underlying path as unusable and use another 2228 underlying path. 2230 o When an AERO Proxy/Server receives link-layer Destination 2231 Unreachable messages in response to a carrier packet that it sends 2232 to one of its permanent neighbors, it treats the messages as an 2233 indication that the path to the neighbor may be failing. However, 2234 the dynamic routing protocol should soon reconverge and correct 2235 the temporary outage. 2237 When an AERO Bridge receives a carrier packet for which the network- 2238 layer destination address is covered by an MSP, the Bridge drops the 2239 packet if there is no more-specific routing information for the 2240 destination and returns an OMNI interface Destination Unreachable 2241 message subject to rate limiting. 2243 When an AERO node receives a carrier packet for which reassembly is 2244 currently congested, it returns an OMNI interface Packet Too Big 2245 (PTB) message as discussed in [I-D.templin-6man-omni] (note that the 2246 PTB messages could indicate either "hard" or "soft" errors). 2248 AERO nodes include ICMPv6 error messages intended for the OAL source 2249 as sub-options in the OMNI option of secured uNA messages. When the 2250 OAL source receives the uNA message, it can extract the ICMPv6 error 2251 message enclosed in the OMNI option and either process it locally or 2252 translate it into a network-layer error to return to the original 2253 source. 2255 3.12. AERO Router Discovery, Prefix Delegation and Autoconfiguration 2257 AERO Router Discovery, Prefix Delegation and Autoconfiguration are 2258 coordinated as discussed in the following Sections. 2260 3.12.1. AERO Service Model 2262 Each AERO Proxy/Server on the OMNI link is configured to facilitate 2263 Client prefix delegation/registration requests. Each Proxy/Server is 2264 provisioned with a database of MNP-to-Client ID mappings for all 2265 Clients enrolled in the AERO service, as well as any information 2266 necessary to authenticate each Client. The Client database is 2267 maintained by a central administrative authority for the OMNI link 2268 and securely distributed to all Proxy/Servers, e.g., via the 2269 Lightweight Directory Access Protocol (LDAP) [RFC4511], via static 2270 configuration, etc. Clients receive the same service regardless of 2271 the Proxy/Servers they select. 2273 AERO Clients and Proxy/Servers use ND messages to maintain neighbor 2274 cache entries. AERO Proxy/Servers configure their OMNI interfaces as 2275 advertising NBMA interfaces, and therefore send unicast RA messages 2276 with a short Router Lifetime value (e.g., ReachableTime seconds) in 2277 response to a Client's RS message. Thereafter, Clients send 2278 additional RS messages to keep Proxy/Server state alive. 2280 AERO Clients and Proxy/Servers include prefix delegation and/or 2281 registration parameters in RS/RA messages (see 2282 [I-D.templin-6man-omni]). The ND messages are exchanged between 2283 Client and FHS Proxy/Servers according to the prefix management 2284 schedule required by the service. If the Client knows its MNP in 2285 advance, it can employ prefix registration by including its MNP-LLA 2286 as the source address of an RS message and with an OMNI option with 2287 valid prefix registration information for the MNP. If the Proxy/ 2288 Server accepts the Client's MNP assertion, it injects the MNP into 2289 the routing system and establishes the necessary neighbor cache 2290 state. If the Client does not have a pre-assigned MNP, it can 2291 instead employ prefix delegation by including the unspecified address 2292 (::) as the source address of an RS message and with an OMNI option 2293 with prefix delegation parameters to request an MNP. 2295 The following sections specify the Client and Proxy/Server behavior. 2297 3.12.2. AERO Client Behavior 2299 AERO Clients discover the addresses of candidate FHS Proxy/Servers in 2300 a similar manner as described in [RFC5214]. Discovery methods 2301 include static configuration (e.g., from a flat-file map of Proxy/ 2302 Server addresses and locations), or through an automated means such 2303 as Domain Name System (DNS) name resolution [RFC1035]. 2304 Alternatively, the Client can discover Proxy/Server addresses through 2305 a layer 2 data link login exchange, or through a unicast RA response 2306 to a multicast/anycast RS as described below. In the absence of 2307 other information, the Client can resolve the DNS Fully-Qualified 2308 Domain Name (FQDN) "linkupnetworks.[domainname]" where 2309 "linkupnetworks" is a constant text string and "[domainname]" is a 2310 DNS suffix for the OMNI link (e.g., "example.com"). 2312 To associate with a FHS Proxy/Server over an underlying interface, 2313 the Client acts as a requesting router to request MNPs by preparing 2314 an RS message with prefix management parameters. If the Client 2315 already knows the Proxy/Server's ADM-LLA, it includes the LLA as the 2316 network-layer destination address; otherwise, the Client includes the 2317 (link-local) All-Routers multicast as the network-layer destination. 2318 If the Client already knows its own MNP-LLA, it can use the MNP-LLA 2319 as the network-layer source address and include an OMNI option with 2320 prefix registration information. Otherwise, the Client uses the 2321 unspecified address (::) as the network-layer source address and 2322 includes prefix delegation parameters in the OMNI option (see: 2323 [I-D.templin-6man-omni]). 2325 The Client next includes Interface Attributes corresponding to the 2326 underlying interface over which it will send the RS message, and MAY 2327 include additional Interface Attributes specific to other underlying 2328 interfaces. Next, the Client submits the RS for OAL encapsulation 2329 and fragmentation if necessary with its own MNP-ULA and the Proxy/ 2330 Server's ADM-ULA or (site-scoped) All-Routers multicast as the OAL 2331 addresses while selecting an Identification value and invoking window 2332 synchronization as specified in [I-D.templin-6man-omni]. 2334 The Client then sends the RS (either directly via Direct interfaces, 2335 via a VPN for VPNed interfaces, via an access router for ANET 2336 interfaces or via INET encapsulation for INET interfaces) then waits 2337 up to RetransTimer milliseconds for an RA message reply (see 2338 Section 3.12.3) (retrying up to MAX_RTR_SOLICITATIONS). If the 2339 Client receives no RAs, or if it receives an RA with Router Lifetime 2340 set to 0, the Client SHOULD abandon attempts through the first 2341 candidate FHS Proxy/Server and try another Proxy/Server. Otherwise, 2342 the Client processes the prefix information found in the RA message. 2344 When the Client processes an RA, it first performs OAL reassembly and 2345 decapsulation if necessary then creates a NCE with the Proxy/Server's 2346 ADM-LLA as the network-layer address and the Proxy/Server's 2347 encapsulation and/or link-layer addresses as the link-layer address. 2348 The Client next records the RA Router Lifetime field value in the NCE 2349 as the time for which the Proxy/Server has committed to maintaining 2350 the MNP in the routing system via this underlying interface, and 2351 caches the other RA configuration information including Cur Hop 2352 Limit, M and O flags, Reachable Time and Retrans Timer. The Client 2353 then autoconfigures MNP-LLAs for any delegated MNPs and assigns them 2354 to the OMNI interface. The Client also caches any MSPs included in 2355 Route Information Options (RIOs) [RFC4191] as MSPs to associate with 2356 the OMNI link, and assigns the MTU value in the MTU option to the 2357 underlying interface. 2359 The Client then registers its additional underlying interfaces with 2360 FHS Proxy/Servers for those interfaces discovered by sending RS 2361 messages via each additional interface as described above. The RS 2362 messages include the same parameters as for the initial RS/RA 2363 exchange, but with destination address set to the Proxy/Server's ADM- 2364 LLA. The Client finally sub-delegates the MNPs to its attached EUNs 2365 and/or the Client's own internal virtual interfaces as described in 2366 [I-D.templin-v6ops-pdhost] to support the Client's downstream 2367 attached "Internet of Things (IoT)". The Client then sends 2368 additional RS messages over each underlying interface before the 2369 Router Lifetime received for that interface expires. 2371 After the Client registers its underlying interfaces, it may wish to 2372 change one or more registrations, e.g., if an interface changes 2373 address or becomes unavailable, if traffic selectors change, etc. To 2374 do so, the Client prepares an RS message to send over any available 2375 underlying interface as above. The RS includes an OMNI option with 2376 prefix registration/delegation information, with Interface Attributes 2377 specific to the selected underlying interface, and with any 2378 additional Interface Attributes specific to other underlying 2379 interfaces. When the Client receives the Proxy/Server's RA response, 2380 it has assurance that the Proxy/Server has been updated with the new 2381 information. 2383 If the Client wishes to discontinue use of a Proxy/Server it issues 2384 an RS message over any underlying interface with an OMNI option with 2385 a prefix release indication. When the Proxy/Server processes the 2386 message, it releases the MNP, sets the NCE state for the Client to 2387 DEPARTED and returns an RA reply with Router Lifetime set to 0. 2388 After a short delay (e.g., 2 seconds), the Proxy/Server withdraws the 2389 MNP from the routing system. 2391 3.12.3. AERO Proxy/Server Behavior 2393 AERO Proxy/Servers act as both IP routers and ND proxies, and support 2394 a prefix delegation/registration service for Clients. Proxy/Servers 2395 arrange to add their ADM-LLAs to a static map of Proxy/Server 2396 addresses for the link and/or the DNS resource records for the FQDN 2397 "linkupnetworks.[domainname]" before entering service. The static 2398 map and/or DNS resource records should be arranged such that Clients 2399 can discover the addresses of Proxy/Servers that are geographically 2400 and/or topologically "close" to their underlying network connections. 2402 When an FHS Proxy/Server receives a prospective Client's RS message 2403 on its OMNI interface, it SHOULD return an immediate RA reply with 2404 Router Lifetime set to 0 if it is currently too busy or otherwise 2405 unable to service the Client. Otherwise, the Proxy/Server performs 2406 OAL reassembly and decapsulation if necessary, then authenticates the 2407 RS message and processes the prefix delegation/registration 2408 parameters. The Proxy/Server first determines the correct MNPs to 2409 provide to the Client by processing the MNP-LLA prefix parameters 2410 and/or the DHCPv6 OMNI sub-option. When the Proxy/Server returns the 2411 MNPs, it also creates a forwarding table entry for the MNP-ULA 2412 corresponding to each MNP so that the MNPs are propagated into the 2413 routing system (see: Section 3.2.3). For IPv6, the Proxy/Server 2414 creates an IPv6 forwarding table entry for each MNP. For IPv4, the 2415 Proxy/Server creates an IPv6 forwarding table entry with the 2416 IPv4-compatibility MNP-ULA prefix corresponding to the IPv4 address. 2418 The Proxy/Server next creates a NCE for the Client using the base 2419 MNP-LLA as the network-layer address. Next, the Proxy/Server updates 2420 the NCE by recording the information in each Interface Attributes 2421 sub-option in the RS OMNI option. The Proxy/Server also records the 2422 actual OAL/*NET addresses and RS message window synchronization 2423 parameters (if any) in the NCE. 2425 Next, the Proxy/Server prepares an RA message using its ADM-LLA as 2426 the network-layer source address and the network-layer source address 2427 of the RS message as the network-layer destination address. The 2428 Proxy/Server sets the Router Lifetime to the time for which it will 2429 maintain both this underlying interface individually and the NCE as a 2430 whole. The Proxy/Server also sets Cur Hop Limit, M and O flags, 2431 Reachable Time and Retrans Timer to values appropriate for the OMNI 2432 link. The Proxy/Server includes the MNPs, any other prefix 2433 management parameters and an OMNI option with no Interface Attributes 2434 but with an Origin Indication sub-option per [I-D.templin-6man-omni] 2435 with the mapped and obfuscated Port Number and IP address 2436 corresponding to the Client's own INET address in the case of INET 2437 Clients or to the Proxy/Server's INET-facing address for all other 2438 Clients. The Proxy/Server should also include an Interface 2439 Attributes sub-option in the OMNI option with FMT/SRT/LHS information 2440 for its INET interface. The Proxy/Server then includes one or more 2441 RIOs that encode the MSPs for the OMNI link, plus an MTU option (see 2442 Section 3.9). The Proxy/Server finally forwards the message to the 2443 Client using OAL encapsulation/fragmentation if necessary while 2444 including an acknowledgement if the RS invoked window 2445 synchronization. 2447 After the initial RS/RA exchange, the Proxy/Server maintains a 2448 ReachableTime timer for each of the Client's underlying interfaces 2449 individually (and for the Client's NCE collectively) set to expire 2450 after ReachableTime seconds. If the Client (or Proxy) issues 2451 additional RS messages, the Proxy/Server sends an RA response and 2452 resets ReachableTime. If the Proxy/Server receives an ND message 2453 with a prefix release indication it sets the Client's NCE to the 2454 DEPARTED state and withdraws the MNP from the routing system after a 2455 short delay (e.g., 2 seconds). If ReachableTime expires before a new 2456 RS is received on an individual underlying interface, the Proxy/ 2457 Server marks the interface as DOWN. If ReachableTime expires before 2458 any new RS is received on any individual underlying interface, the 2459 Proxy/Server sets the NCE state to STALE and sets a 10 second timer. 2460 If the Proxy/Server has not received a new RS or ND message with a 2461 prefix release indication before the 10 second timer expires, it 2462 deletes the NCE and withdraws the MNP from the routing system. 2464 The Proxy/Server processes any ND messages pertaining to the Client 2465 and returns an NA/RA reply in response to solicitations. The Proxy/ 2466 Server may also issue unsolicited RA messages, e.g., with reconfigure 2467 parameters to cause the Client to renegotiate its prefix delegation/ 2468 registrations, with Router Lifetime set to 0 if it can no longer 2469 service this Client, etc. Finally, If the NCE is in the DEPARTED 2470 state, the Proxy/Server deletes the entry after DepartTime expires. 2472 Note: Clients SHOULD notify former Proxy/Servers of their departures, 2473 but Proxy/Servers are responsible for expiring neighbor cache entries 2474 and withdrawing routes even if no departure notification is received 2475 (e.g., if the Client leaves the network unexpectedly). Proxy/Servers 2476 SHOULD therefore set Router Lifetime to ReachableTime seconds in 2477 solicited RA messages to minimize persistent stale cache information 2478 in the absence of Client departure notifications. A short Router 2479 Lifetime also ensures that proactive RS/RA messaging between Clients 2480 and Proxy/Servers will keep any NAT state alive (see above). 2482 Note: All Proxy/Servers on an OMNI link MUST advertise consistent 2483 values in the RA Cur Hop Limit, M and O flags, Reachable Time and 2484 Retrans Timer fields the same as for any link, since unpredictable 2485 behavior could result if different Proxy/Servers on the same link 2486 advertised different values. 2488 3.12.3.1. DHCPv6-Based Prefix Registration 2490 When a Client is not pre-provisioned with an MNP-LLA, it will need 2491 for the FHS Proxy/Server to select one or more MNPs on its behalf and 2492 set up the correct state in the AERO routing service. (A Client with 2493 a pre-provisioned MNP may also request the Proxy/Server to select 2494 additional MNPs.) The DHCPv6 service [RFC8415] is used to support 2495 this requirement. 2497 When a Client needs to have the FHS Proxy/Server select MNPs, it 2498 sends an RS message with source address set to the unspecified 2499 address (::) and with an OMNI option that includes a DHCPv6 message 2500 sub-option with DHCPv6 Prefix Delegation (DHCPv6-PD) parameters. 2501 When the Proxy/Server receives the RS message, it extracts the 2502 DHCPv6-PD message from the OMNI option. 2504 The Proxy/Server then acts as a "Proxy DHCPv6 Client" in a message 2505 exchange with the locally-resident DHCPv6 server, which delegates 2506 MNPs and returns a DHCPv6-PD Reply message. (If the Proxy/Server 2507 wishes to defer creation of MN state until the DHCPv6-PD Reply is 2508 received, it can instead act as a Lightweight DHCPv6 Relay Agent per 2509 [RFC6221] by encapsulating the DHCPv6-PD message in a Relay-forward/ 2510 reply exchange with Relay Message and Interface ID options.) 2512 When the Proxy/Server receives the DHCPv6-PD Reply, it adds a route 2513 to the routing system and creates an MNP-LLA based on the delegated 2514 MNP. The Proxy/Server then sends an RA back to the Client with the 2515 (newly-created) MNP-LLA as the destination address and with the 2516 DHCPv6-PD Reply message coded in the OMNI option. When the Client 2517 receives the RA, it creates a default route, assigns the Subnet 2518 Router Anycast address and sets its MNP-LLA based on the delegated 2519 MNP. 2521 Note: See [I-D.templin-6man-omni] for an MNP delegation alternative 2522 that avoids including a DHCPv6 message sub-option in the RS. Namely, 2523 when the Client requests a single MNP it can set the RS source to the 2524 unspecified address (::) and include a Node Identification sub-option 2525 and Preflen in the OMNI option (but with no DHCPv6 message sub- 2526 option). When the Proxy/Server receives the RS message, it forwards 2527 a self-generated DHCPv6 Solicit message to the DHCPv6 server on 2528 behalf of the Client. When the Proxy/Server receives the DHCPv6 2529 Reply, it prepares an RA message with an OMNI option with Preflen 2530 information (but with no DHCPv6 message sub-option), then places the 2531 (newly-created) MNP-LLA in the RA destination address and returns the 2532 message to the Client. 2534 3.13. The AERO Proxy Function 2536 Clients connect to the OMNI link via FHS Proxy/Servers, with one or 2537 more FHS Proxy/Servers for each underlying interface. Each of the 2538 Client's FHS Proxy/Servers must be informed of all of the Client's 2539 additional underlying interfaces. For Clients on Direct and VPNed 2540 underlying interfaces the Proxy/Server "A" for that interface is 2541 directly connected, for Clients on ANET underlying interfaces Proxy/ 2542 Server "A" is located on the ANET/INET boundary, and for Clients on 2543 INET underlying interfaces Proxy/Server "A" is located somewhere in 2544 the connected Internetwork. When the Client registers with Proxy/ 2545 Server "A", it must also report the registration to any other Proxy/ 2546 Servers for other underlying interfaces "B", "C", "D", etc. for which 2547 an underlying interface relationship has already been established. 2548 The Proxy/Server satisfies these requirements as follows: 2550 o when FHS Proxy/Server "A" receives a Client RS message, it first 2551 verifies that the OAL Identification is within the window for the 2552 NCE that matches the MNP-ULA for this Client neighbor and 2553 authenticates the message. (If no NCE was found, Proxy/Server "A 2554 instead creates one in the STALE state and returns an RA message 2555 with an authentication signature and any window synchronization 2556 parameters.) Proxy/Server "A" then examines the network-layer 2557 destination address. If the destination address is the ADM-LLA of 2558 a different Proxy/Server "B" (or, if the OMNI option included an 2559 MS-Register sub-option with the ADM-LLAs of one or more different 2560 "LHS" Proxy/Servers "B", "C", "D", etc.), Proxy/Server "A" 2561 prepares a separate proxyed version of the RS message with an OAL 2562 header with source set to its own ADM-ULA and destination set to 2563 the LHS Proxy/Server's ADM-ULA. Proxy/Server "A" also includes an 2564 OMNI header with an Interface Attributes option that includes its 2565 own INET address, a unique UDP Port Number for this Client, and 2566 FMT/SRT/LHS information. Proxy/Server "A" then sets the S/ 2567 T-omIndex to the value for this Client underlying interface, then 2568 forwards the message into the OMNI link secured spanning tree. 2569 (Note: including a unique Port Number allows LHS Proxy/Server "B" 2570 to distinguish different Clients located behind the same FHS 2571 Proxy/Server "A" at the link-layer, whereas the link-layer 2572 addresses would otherwise be indistinguishable.) 2574 o when LHS Proxy/Server "B" receives the RS, it authenticates the 2575 message then creates or updates a NCE for the Client with FHS 2576 Proxy/Server "A"'s Interface Attributes as the link-layer address 2577 information for this S/T-omIndex and caches any window 2578 synchronization parameters supplied by the Client. LHS Proxy/ 2579 Server "B" then prepares an RA message with source set to its own 2580 LLA and destination set to the Client's MNP-LLA, and with any 2581 window synchronization acknowledgements (i.e., only if the RS 2582 destination was its own ADM-LLA). Proxy/Server "B" then 2583 encapsulates the RA in an OAL header with source set to its own 2584 ADM-ULA and destination set to the ADM-ULA of Proxy/Server "A, 2585 performs fragmentation if necessary, then sends the resulting 2586 carrier packets into the secured spanning tree. 2588 o when Proxy/Server "A" reassembles the RA, it locates the Client 2589 NCE based on the RA destination LLA. Proxy/Server "A" then re- 2590 encapsulates the RA message with OAL source set to its own ADM-ULA 2591 and OAL destination set to the MNP-ULA of the Client, includes an 2592 authentication signature if necessary, fragments if necessary and 2593 returns the fragments to the Client. 2595 o The Client repeats this process over each of its additional 2596 underlying interfaces while treating each Proxy/Server "B", "C", 2597 "D" as an FHS while providing MS-Register information for other 2598 Proxy/Servers as an LHS. 2600 After the initial RS/RA exchanges each Proxy/Server forwards any of 2601 the Client's carrier packets with OAL destinations for which there is 2602 no matching NCE to a Bridge using OAL encapsulation with its own ADM- 2603 ULA as the source and the destination determined by the ORH supplied 2604 by the Client. The Proxy/Server instead forwards any carrier packets 2605 destined to a neighbor cache target directly to the target according 2606 to the OAL/link-layer information - the process of establishing 2607 neighbor cache entries is specified in Section 3.14. 2609 While the Client is still associated with each Proxy/Server "A", "A" 2610 can send NS, RS and/or unsolicited NA messages to update the neighbor 2611 cache entries of other AERO nodes on behalf of the Client and/or to 2612 convey Interface Attribute updates. This allows for higher-frequency 2613 Proxy-initiated RS/RA messaging over well-connected INET 2614 infrastructure supplemented by lower-frequency Client-initiated RS/RA 2615 messaging over constrained ANET data links. 2617 If any Proxy/Server "B", "C", "D" ceases to send solicited 2618 advertisements, Proxy/Server "A" sends unsolicited RAs to the Client 2619 with destination set to (link-local) All-Nodes multicast and with 2620 Router Lifetime set to zero to inform Clients that a Proxy/Server has 2621 failed. Although Proxy/Server "A" can engage in ND exchanges on 2622 behalf of the Client, the Client can also send ND messages on its own 2623 behalf, e.g., if it is in a better position than "A" to convey 2624 Interface Attribute changes, etc. The ND messages sent by the Client 2625 include the Client's MNP-LLA as the source in order to differentiate 2626 them from the ND messages sent by Proxy/Server "A". 2628 If the Client becomes unreachable over an underlying interface, 2629 Proxy/Server "A" sets the NCE state to DEPARTED and retains the entry 2630 for DepartTime seconds. While the state is DEPARTED, Proxy/Server 2631 "A" forwards any carrier packets destined to the Client to a Bridge 2632 via OAL/ORH encapsulation. When DepartTime expires, Proxy/Server "A" 2633 deletes the NCE and discards any further carrier packets destined to 2634 the former Client. 2636 In some ANETs that employ a Proxy/Server, the Client's MNP can be 2637 injected into the ANET routing system. In that case, the Client can 2638 send original IP packets without invoking the OAL so that the ANET 2639 routing system transports the original IP packets to the Proxy. This 2640 can be very beneficial, e.g., if the Client connects to the ANET via 2641 low-end data links such as some aviation wireless links. 2643 If the ANET first-hop access router is on the same underlying link as 2644 the Client and recognizes the AERO/OMNI protocol, the Client can 2645 avoid OAL encapsulation for both its control and data messages. When 2646 the Client connects to the link, it can send an unencapsulated RS 2647 message with source address set to its own MNP-LLA (or to a Temporary 2648 LLA), and with destination address set to the ADM-LLA of the Client's 2649 selected Proxy/Server or to (link-local) All-Routers multicast. The 2650 Client includes an OMNI option formatted as specified in 2651 [I-D.templin-6man-omni]. The Client then sends the unencapsulated RS 2652 message, which will be intercepted by the AERO-Aware access router. 2654 The ANET access router then performs OAL encapsulation on the RS 2655 message and forwards it to a Proxy/Server at the ANET/INET boundary. 2656 When the access router and Proxy/Server are one and the same node, 2657 the Proxy/Server would share and underlying link with the Client but 2658 its message exchanges with outside correspondents would need to pass 2659 through a security gateway at the ANET/INET border. The method for 2660 deploying access routers and Proxys (i.e. as a single node or 2661 multiple nodes) is an ANET-local administrative consideration. 2663 Note: When a Proxy/Server alters the IPv6 ND message contents before 2664 forwarding (e.g., such as altering the OMNI option contents), it must 2665 re-calculate the IPv6 ND message checksum. Altering the OMNI option 2666 contents also invalidates any OMNI option authentication signatures, 2667 which need not be re-calculated if they will not be examined by other 2668 nodes following forwarding. 2670 Note: The Proxy/Server can apply packing as discussed in 2671 [I-D.templin-6man-omni] if an opportunity arises to concatenate 2672 multiple original IP packets destined to the same neighbor. 2674 3.13.1. Detecting and Responding to Proxy/Server Failures 2676 In environments where fast recovery from Proxy/Server failure is 2677 required, Proxy/Server "A" SHOULD use proactive Neighbor 2678 Unreachability Detection (NUD) to track each peer Proxy/Server "B" 2679 reachability in a similar fashion as for Bidirectional Forwarding 2680 Detection (BFD) [RFC5880]. Proxy/Server "A" can then quickly detect 2681 and react to failures so that cached information is re-established 2682 through alternate paths. The NUD control messaging is carried only 2683 over well-connected ground domain networks (i.e., and not low-end 2684 aeronautical radio links) and can therefore be tuned for rapid 2685 response. 2687 Proxy/Server "A" performs proactive NUD with peer Proxy/Server "B" 2688 for which there are currently active Clients by sending continuous NS 2689 messages in rapid succession, e.g., one message per second. Proxy/ 2690 Server "A" sends the NS message via the spanning tree with its own 2691 ADM-LLA as the source and the ADM-LLA of the peer Proxy/Server "B" as 2692 the destination. When Proxy/Server "A" is also sending RS messages 2693 to the peer Proxy/Server "B" on behalf of ANET Clients, the resulting 2694 RA responses can be considered as equivalent hints of forward 2695 progress. This means that Proxy/Server "B" need not also send a 2696 periodic NS if it has already sent an RS within the same period. If 2697 the peer Proxy/Server "B" fails (i.e., if "A" ceases to receive 2698 advertisements), Proxy/Server "A" can quickly inform Clients by 2699 sending multicast RA messages on the ANET interface. 2701 Proxy/Server "A" sends RA messages on the ANET interface with source 2702 address set to Proxy/Server "B"'s address, destination address set to 2703 (link-local) All-Nodes multicast, and Router Lifetime set to 0. 2704 Proxy/Server "A" SHOULD send MAX_FINAL_RTR_ADVERTISEMENTS RA messages 2705 separated by small delays [RFC4861]. Any Clients on the ANET that 2706 had been using the failed Proxy/Server "B" will receive the RA 2707 messages and associate with a new Proxy/Server. 2709 3.13.2. Point-to-Multipoint Proxy/Server Coordination 2711 In environments where Client messaging over ANETs is bandwidth- 2712 limited and/or expensive, Clients can enlist the services of Proxy/ 2713 Server "A" to coordinate with multiple Proxy/Servers "B", "C", "D" 2714 etc. in a single RS/RA message exchange. The Client can send a 2715 single RS message to (link-local) All-Routers multicast that includes 2716 the ID's of multiple Proxy/Servers in MS-Register sub-options of the 2717 OMNI option. 2719 When Proxy/Server "A" receives the RS and processes the OMNI option, 2720 it sends a separate RS to each MS-Register Proxy/Server ID. When 2721 Proxy/Server "A" receives an RA, it can optionally return an 2722 immediate "singleton" RA to the Client or record the Proxy/Server's 2723 ID for inclusion in a pending "aggregate" RA message. Proxy/Server 2724 "A" can then return aggregate RA messages to the Client including 2725 multiple Proxy/Server IDs in order to conserve bandwidth. Each RA 2726 includes a proper subset of the Proxy/Server IDs from the original RS 2727 message, and Proxy/Server "A" must ensure that the message contents 2728 of each RA are consistent with the information received from the 2729 (aggregated) additional Proxy/Servers. 2731 Clients can thereafter employ efficient point-to-multipoint Proxy/ 2732 Server coordination under the assistance of Proxy/Server "A" to 2733 reduce the number of messages sent over the ANET while enlisting the 2734 support of multiple Proxy/Servers for fault tolerance. Clients can 2735 further include MS-Release sub-options in IPv6 ND messages to request 2736 Proxy/Server "A" to release from former Proxy/Servers via the 2737 procedures discussed in Section 3.16.5. 2739 When the Client sends an RS with window synchronization parameters 2740 and with multiple MS-Register Proxy/Server IDs, Proxy/Server "A" may 2741 receive multiple RAs - each with their own window synchronization 2742 parameters. Proxy/Server "A" must then immediately forward these RAs 2743 to the Client as singletons instead of including them in an 2744 aggregate, and the Client will use each RA to establish a separate 2745 NCE and window for each individual Proxy/Server. 2747 The OMNI interface specification [I-D.templin-6man-omni] provides 2748 further discussion of the RS/RA messaging involved in point-to- 2749 multipoint coordination. 2751 3.14. AERO Route Optimization 2753 AERO nodes invoke route optimization when they need to forward 2754 packets to new target destinations. Route optimization is based on 2755 IPv6 ND Address Resolution messaging between a Route Optimization 2756 Source (ROS) and Route Optimization Responder (ROR). Route 2757 optimization is initiated by the first eligible ROS closest to the 2758 source as follows: 2760 o For Clients on VPNed and Direct interfaces, the Client's FHS 2761 Proxy/Server is the ROS. 2763 o For Clients on ANET interfaces, either the Client or the FHS 2764 Proxy/Server may be the ROS. 2766 o For Clients on INET interfaces, the Client itself is the ROS. 2768 o For correspondent nodes on INET/EUN interfaces serviced by a 2769 Relay, the Relay is the ROS. 2771 The route optimization procedure is conducted between the ROS and an 2772 LHS Proxy/Server/Relay for the target selected by routing as the ROR. 2773 In this arrangement, the ROS is always the Client or Proxy/Server/ 2774 Relay nearest the source over the selected source underlying 2775 interface, while the ROR is always an LHS Proxy/Server/Relay for the 2776 target regardless of the target underlying interface. 2778 The AERO routing system directs a route optimization solicitation 2779 sent by the ROS to the nearest available ROR, which returns a route 2780 optimization reply. The exact ROR selected is unimportant; all that 2781 matters is that the route optimization information returned must be 2782 current and authentic. The ROS is responsible for periodically 2783 refreshing the route optimization, and the ROR is responsible for 2784 quickly informing the ROS of any changes. 2786 The procedures are specified in the following sections. 2788 3.14.1. Route Optimization Initiation 2790 When an original IP packet from a source node destined to a target 2791 node arrives, the ROS checks for a NCE with an MNP-LLA that matches 2792 the target destination. If there is a NCE in the REACHABLE state, 2793 the ROS invokes the OAL and forwards the resulting carrier packets 2794 according to the cached state then returns from processing. 2795 Otherwise, if there is no NCE the ROS creates one in the INCOMPLETE 2796 state. 2798 The ROS next places the original IP packet on a short queue then 2799 sends an NS message for Address Resolution (NS(AR)) to receive a 2800 solicited NA(AR) message from an ROR. The NS(AR) message must be 2801 sent securely, and includes: 2803 o the LLA of the ROS as the source address. 2805 o the MNP-LLA corresponding to the original IP packet's destination 2806 as the Target Address, e.g., for 2001:db8:1:2::10:2000 the Target 2807 Address is fe80::2001:db8:1:2. 2809 o the Solicited-Node multicast address [RFC4291] formed from the 2810 lower 24 bits of the original IP packet's destination as the 2811 destination address, e.g., for 2001:db8:1:2::10:2000 the NS(AR) 2812 destination address is ff02:0:0:0:0:1:ff10:2000. 2814 The NS(AR) message also includes an OMNI option with an Interface 2815 Attributes entry for the underlying interface, with S/T-omIndex set 2816 to the underlying interface index and with Preflen set to the prefix 2817 length associated with the NS(AR) source. The ROS then selects an 2818 Identification value submits the NS(AR) message for OAL encapsulation 2819 with OAL source set to its own ULA and OAL destination set to the ULA 2820 corresponding to the target. (The ROS does not include any window 2821 synchronization parameters, since it will never exchange other 2822 carrier packet types directly with the ROR). 2824 The ROS then sends the resulting carrier packet(s) into the SRT 2825 secured spanning tree without decrementing the network-layer TTL/Hop 2826 Limit field. (When the ROS is an INET Client, it instead sends the 2827 resulting carrier packets to the ADM-ULA of one of its current Proxy/ 2828 Servers. The Proxy/Server then reassembles if necessary, verifies 2829 the NS(AR) signature, then re-encapsulates with the OAL source set to 2830 its own ADM-ULA and OAL destination set to the ULA corresponding to 2831 the target. The Proxy/Server then fragments if necessary and sends 2832 the resulting carrier packets into the secured spanning tree on 2833 behalf of the Client.) 2835 3.14.2. Relaying the NS(AR) *NET Packet(s) 2837 When the Bridge receives the carrier packet(s) containing the RS from 2838 the ROS, it discards the *NET headers and determines the next hop by 2839 consulting its standard IPv6 forwarding table for the OAL header 2840 destination address. The Bridge then decrements the OAL header Hop- 2841 Limit, then re-encapsulates and forwards the carrier packet(s) via 2842 the secured spanning tree the same as for any IPv6 router, where it 2843 may traverse multiple OMNI link segments. The final-hop Bridge will 2844 deliver the carrier packet(s) via the secured spanning tree to a 2845 Proxy/Server or Relay that services the target. 2847 3.14.3. Processing the NS(AR) and Sending the NA(AR) 2849 When an LHS Proxy/Server (or Relay) for the target receives the 2850 secured carrier packet(s), it reassembles if necessary then examines 2851 the NS(AR) target to determine whether it has a matching NCE and/or 2852 non-MNP route. If there is no match, the Proxy/Server drops the 2853 message. Otherwise, the LHS Proxy/Server/Relay continues processing 2854 as follows: 2856 o if the NS(AR) target matches a Client NCE in the DEPARTED state, 2857 the Proxy/Server re-encapsulates while setting the OAL source to 2858 the ULA of the ROS and OAL destination address to the ADM-ULA of 2859 the Client's new Proxy/Server. The (old) Proxy/Server then 2860 fragments if necessary and forwards the resulting carrier 2861 packet(s) over the secured spanning tree then returns from 2862 processing. 2864 o If the NS(AR) target matches the MNP-LLA of a Client NCE in the 2865 REACHABLE state, the Proxy/Server makes note of whether the NS 2866 (AR) arrived from the secured or unsecured spanning tree then acts 2867 as an ROR to provide route optimization information on behalf of 2868 the Client. (Note that if the message arrived via the secured 2869 spanning tree the ROR need not perform further authentication, but 2870 if it arrived over an open INET underlying interface it must 2871 verify the message authentication signature before accepting.) 2873 o If the NS(AR) target matches one of its non-MNP routes, the Relay 2874 acts as both an ROR and a route optimization target, since the 2875 Relay forwards IP packets toward the (fixed network) target at the 2876 network layer. 2878 The ROR next checks the target NCE for a Report List entry that 2879 matches the NS(AR) source LLA/ULA of the ROS. If there is a Report 2880 List entry, the ROR refreshes ReportTime for this ROR; otherwise, the 2881 ROR creates a new entry for the ROS and records both the LLA and ULA. 2883 The ROR then prepares a (solicited) NA(AR) message to return to the 2884 ROS with the source address set to the target's MNP-LLA, the 2885 destination address set to the NS(AR) LLA source address and the 2886 Target Address set to the same value that appeared in the NS(AR). 2887 The ROR then includes an OMNI option with Preflen set to the prefix 2888 length associated with the NA(AR) source address. The ROR next 2889 includes Interface Attributes in the OMNI option for all of the 2890 target's underlying interfaces with current information for each 2891 interface. 2893 For each Interface Attributes sub-option, the ROR sets the L2ADDR 2894 according to its own INET address for VPNed, Direct or ANET 2895 interfaces, to its own INET address for NATed Client interfaces, or 2896 to the Client's INET address for native Client interfaces. The ROR 2897 then includes the lower 32 bits of the Proxy/Server's ADM-ULA as the 2898 LHS, encodes the ADM-ULA prefix length code in the SRT field and sets 2899 FMT as specified in Section 3.3. 2901 The ROR then sets the NA(AR) message R flag to 1 (as a router) and S 2902 flag to 1 (as a response to a solicitation) and sets the O flag to 0 2903 (as a proxy) and sets the OMNI header S/T-omIndex to 0. The ROR 2904 finally submits the NA(AR) for OAL encapsulation with source set to 2905 its own ULA and destination set to the same ULA that appeared in the 2906 NS(AR) OAL source, then performs OAL encapsulation and fragmentation 2907 using the same Identification value that appeared in the NS(AR) and 2908 finally forwards the resulting (*NET-encapsulated) carrier packets 2909 via the secured spanning tree without decrementing the network-layer 2910 TTL/Hop Limit field. 2912 3.14.4. Relaying the NA(AR) 2914 When the Bridge receives NA(AR) carrier packets from the ROR, it 2915 discards the *NET header and determines the next hop by consulting 2916 its standard IPv6 forwarding table for the OAL header destination 2917 address. The Bridge then decrements the OAL header Hop-Limit, re- 2918 encapsulates the carrier packet and forwards it via the SRT secured 2919 spanning tree the same as for any IPv6 router, where it may traverse 2920 multiple OMNI link segments. The final-hop Bridge will deliver the 2921 carrier packet via the secured spanning tree to a Proxy/Server for 2922 the ROS. 2924 3.14.5. Processing the NA(AR) 2926 When the ROS receives the NA(AR) message from the ROR, it first 2927 searches for a NCE that matches the NA(AR) LLA source address. The 2928 ROS then processes the message the same as for standard IPv6 Address 2929 Resolution [RFC4861]. In the process, it caches all OMNI option 2930 information in the target NCE (including all Interface Attributes). 2932 When the ROS is a Client, the solicited NA(AR) message will first be 2933 delivered via the SRT secured spanning tree to the Proxy/Server that 2934 forwarded the NS(AR), which reassembles if necessary. The Proxy/ 2935 Server then forwards the message to the Client while re-encapsulating 2936 and re-fragmenting if necessary. If the Client is on an ANET, ANET 2937 physical security and protected spectrum ensures security for the 2938 unmodified NA(AR); if the Client is on the open INET the Proxy/Server 2939 must instead insert an authentication signature. The Proxy/Server 2940 uses its own ADM-ULA as the OAL source and the MNP-ULA of the Client 2941 as the OAL destination. 2943 3.14.6. Forwarding Packets to Route Optimized Targets 2945 After the ROS receives the route optimization NA(AR) and updates the 2946 target NCE, it can begin forwarding packets along the best paths 2947 based on the target's Interface Attributes. The ROS selects target 2948 underlying interfaces according to traffic selectors and/or any other 2949 traffic discriminators, however each underlying interface selected 2950 must first establish window synchronization state if necessary. 2952 To establish window synchronization state, the ROS performs a secured 2953 unicast NS/NA(WIN) exchange with window synchronization parameters 2954 according to the Interface Attribute FMT. If FMT-Forward is set, the 2955 ROS prepares an NS(WIN) with its own LLA as the source and the MNP- 2956 LLA of the target Client as the destination; otherwise, it sets the 2957 ADM-LLA of the LHS Proxy/Server as the destination. The ROS then 2958 encapsulates the NS(WIN) in an OAL header with its own ULA as the 2959 source. If the ROS is the Client, it sets the OAL destination to the 2960 ADM-ULA of its FHS Proxy/Server, includes an authentication signature 2961 if necessary, and includes an ORH-1 with FMT-Type clear for the first 2962 fragment. The Client sets the ORH Segments Left to 1 and includes 2963 valid SRT/LHS information for the LHS Proxy/Server with L2ADDR set to 2964 0, then forwards the NS(WIN) to its FHS Proxy/Server which must 2965 reassemble and verify the authentication signature if necessary. The 2966 FHS Proxy/Server then re-encapsulates, re-fragments and forwards the 2967 NS(WIN) carrier packets into the secured spanning tree with its own 2968 ADM-ULA as the OAL source and the ADM-ULA of the LHS Proxy/Server as 2969 the OAL destination while replacing the ORH-1 with an ORH-0. (If the 2970 ROS was the FHS Proxy/Server itself, it instead includes an ORH-0, 2971 and forwards the carrier packets into the secured spanning tree.) 2973 When an LHS Proxy/Server receives the NS(WIN) it first reassembles if 2974 necessary. If the NS(WIN) destination is its own ADM-LLA, the LHS 2975 Proxy/Server creates an NCE based on the NS(WIN) source LLA, caches 2976 the window synchronization information, and prepares an NA(WIN) while 2977 using its own ADM-LLA as the source and the ROS LLA as the 2978 destination. The LHS Proxy/Server then encapsulates the NA(WIN) in 2979 an OAL header with source set to its own ADM-ULA and destination set 2980 to the NS(WIN) OAL source. The LHS Proxy/Server then fragments if 2981 necessary includes an ORH-0 with omIndex set to the S/T-omIndex value 2982 found in the NS(WIN) OMNI option, then forwards the resulting carrier 2983 packets into the secured spanning tree which will deliver them to the 2984 ROS Proxy/Server. 2986 If the NS(WIN) destination is the MNP-LLA of the target Client, the 2987 LHS Proxy/Server instead re-encapsulates using the same OAL source 2988 and the MNP-ULA of the target as the OAL destination and includes an 2989 authentication signature if necessary while removing the ORH-0. The 2990 LHS Proxy/Server then forwards the NS(WIN) to the target over the 2991 underlying interface identified by the ORH-0 omIndex (or, over any 2992 underlying interface if omIndex is 0). When the target receives the 2993 NS(WIN), it verifies the authentication signature if necessary then 2994 creates an NCE for the ROS LLA, caches the window synchronization 2995 information and prepares an NA(WIN) to return to the ROS with its 2996 MNP-LLA as the source and the LLA of the ROS as the destination, and 2997 with an authentication signature if necessary. The target Client 2998 then encapsulates the NA(WIN) in an OAL header with its own MNP-ULA 2999 as the source, the ADM-ULA of the LHS Proxy/Server as the 3000 destination, and with an ORH-1 with SRT/LHS information copied from 3001 the ADM-ULA of the FHS Proxy/Server found in the NS(WIN) OAL source 3002 address. The target Client then sets the ORH-1 omIndex to the S/ 3003 T-omIndex value found in the NS(WIN) OMNI option, then forward the 3004 message to the LHS Proxy/Server. 3006 When the LHS Proxy/Server receives the message, it reassembles if 3007 necessary, verifies the authentication signature if necessary then 3008 re-encapsulates using its own ADM-ULA as the source and the ADM-ULA 3009 of the FHS Proxy/Server as the destination The LHS Proxy/Server then 3010 re-fragments and forwards the NS(WIN) carrier packets into the 3011 spanning tree while replacing the ORH-1 with an ORH-0. When the FHS 3012 Proxy/Server receives the NA(WIN), it reassembles if necessary then 3013 updates the target NCE based on the message contents if the Proxy/ 3014 Server itself is the ROS. If the NS(WIN) source was the ADM-LLA of 3015 the LHS Proxy/Server, the ROS must create and maintain a NCE for the 3016 LHS Proxy/Server which it must regard as a companion to the existing 3017 MNP-LLA NCE for the target Client. (The NCE for the LHS Proxy/Server 3018 can also be shared by multiple target Client NCEs if the ROS 3019 communicates with multiple active targets located behind the same LHS 3020 Proxy/Server.) If the Client is the ROS, the FHS Proxy/Server 3021 instead inserts an authentication signature if necessary, removes the 3022 ORH-0 then re-encapsulates and re-fragments if necessary while 3023 changing the OAL destination to the MNP-ULA of the Client. The FHS 3024 Proxy/Server then forwards the NA(WIN) to the Client over the 3025 underlying interface identified by the ORH-0 omIndex which then 3026 updates its own NCE based on the target Client MNP-LLA or LHS Proxy/ 3027 Server ADM-LLA. The ROS (whether the Proxy/Server or the Client 3028 itself) finally arranges to return an acknowledgement if requested by 3029 the NA(WIN). 3031 After window synchronization state has been established, the ROS can 3032 begin forwarding carrier packets as specified in Section 3.2.7 while 3033 performing additional NS/NA(WIN) exchanges as above to update window 3034 state and/or test reachability. These forwarding procedures apply to 3035 the case where the selected target interface SRT/LHS codes indicate 3036 that the interface is located in a foreign OMNI link segment. In 3037 that case, the ROS must include ORHs and send the resulting carrier 3038 packets into the spanning tree. 3040 If the SRT/LHS codes indicate that the interface is in the local OMNI 3041 link segment, the ROS can instead forward carrier packets directly to 3042 the LHS Proxy/Server using the L2ADDR for encapsulation, or even to 3043 the target Client itself while invoking NAT traversal if necessary. 3044 When the ROS sends packets directly to the LHS Proxy/Server, it 3045 includes an ORH-0 if necessary to inform the Proxy/Server as to 3046 whether it must reassemble and/or the correct target Client interface 3047 for (re)forwarding. If the LHS Proxy/Server is required to 3048 reassemble, the ROS sets the OAL destination to the ADM-ULA of the 3049 LHS Proxy/Server; otherwise, it sets the OAL destination to the MNP- 3050 ULA of the target Client itself. When the ROS sends packets directly 3051 to the target Client, it need not include an ORH. The LHS Proxy/ 3052 Server (or target Client) then saves the L2ADDR and full OAL 3053 addresses in the ROS NCE, and the ROS can begin applying OAL header 3054 compression in subsequent carrier packets as specified in 3055 [I-D.templin-6man-omni] since the OAL header is not examined by any 3056 forwarding nodes in the path. 3058 While the ROS continues to actively forward packets to the target 3059 Client, it is responsible for updating window synchronization state 3060 and per-interface reachability before expiration. Window 3061 synchronization state is shared by all underlying interfaces in the 3062 ROS' NCE that use the same destination LLA so that a single NS/ 3063 NA(NUD) exchange applies for all interfaces regardless of the 3064 (single) interface used to conduct the exchange. However, the window 3065 synchronization exchange only confirms target Client reachability 3066 over the specific interface used to conduct the exchange. 3067 Reachability for other underlying interfaces that share the same 3068 window synchronization state must be determined individually using 3069 NS/NA(NUD) messages which need not be secured as long as they use in- 3070 window Identifications and do not update other state information. 3072 3.15. Neighbor Unreachability Detection (NUD) 3074 AERO nodes perform Neighbor Unreachability Detection (NUD) per 3075 [RFC4861] either reactively in response to persistent link-layer 3076 errors (see Section 3.11) or proactively to confirm reachability. 3077 The NUD algorithm is based on periodic control message exchanges and 3078 may further be seeded by ND hints of forward progress, but care must 3079 be taken to avoid inferring reachability based on spoofed 3080 information. For example, IPv6 ND message exchanges that include 3081 authentication codes and/or in-window Identifications may be 3082 considered as acceptable hints of forward progress, while spurious 3083 random carrier packets should be ignored. 3085 AERO nodes can use standard NS/NA(NUD) exchanges sent over the OMNI 3086 link secured spanning tree (i.e. the same as described above for NS/ 3087 NA(WIN)) to test reachability without risk of DoS attacks from nodes 3088 pretending to be a neighbor. These NS/NA(NUD) messages use the 3089 unicast LLAs and ULAs of the parties involved in the NUD test the 3090 same as for standard IPv6 ND over the secured spanning tree. When 3091 only reachability information is required without updating any other 3092 NCE state, unsecured NS/NA(NUD) messages may instead be exchanged 3093 directly between neighbors as long as they include in-window 3094 Identifications. 3096 When an ROR directs an ROS to a target neighbor with one or more 3097 link-layer addresses, the ROS probes each unsecured target underlying 3098 interface either proactively or on-demand of carrier packets directed 3099 to the path by multilink forwarding to maintain the interface's state 3100 as reachable. Probing is performed through NS(NUD) messages over the 3101 SRT secured or unsecured spanning tree, or through NS(NUD) messages 3102 sent directly to an underlying interface of the target itself. While 3103 testing a target underlying interface, the ROS can optionally 3104 continue to forward carrier packets via alternate interfaces and/or 3105 maintain a small queue of carrier packets until target reachability 3106 is confirmed. 3108 NS(NUD) messages are encapsulated, fragmented and transmitted as 3109 carrier packets the same as for ordinary original IP data packets, 3110 however the encapsulated destinations are the LLA of the ROS and 3111 either the ADM-LLA of the LHS Proxy/Server or the MNP-LLA of the 3112 target itself. The ROS encapsulates the NS(NUD) message the same as 3113 described in Section 3.2.7, however Destination Suffixes (if present) 3114 are set according to the LLA destination (i.e., and not a ULA/GUA 3115 destination). The ROS sets the NS(NUD) OMNI header S/T-omIndex to 3116 identify the underlying interface used for forwarding (or to 0 if any 3117 underlying interface can be used). The ROS also includes an ORH with 3118 SRT/LHS/LLADDR information the same as for ordinary data packets, but 3119 no authentication signatures are included. The ROS then fragments 3120 the OAL packet and forwards the resulting carrier packets into the 3121 unsecured spanning tree or directly to the target (or LHS Proxy/ 3122 Server) if it is in the local segment. 3124 When the target (or LHS Proxy/Server) receives the NS(NUD) carrier 3125 packets, it verifies that it has a NCE for this ROS and that the 3126 Identification is in-window, then submits the carrier packets for 3127 reassembly. The node then searches for Interface Attributes in its 3128 NCE for the ROS that match the NS(NUD) S/T-omIndex and uses the 3129 SRT/LHS/L2ADDR and FMT information to prepare an ORH for the NA(NUD) 3130 reply. The node then prepares the NA(NUD) with the source and 3131 destination LLAs reversed, encapsulates and sets the OAL source and 3132 destination, sets the NA(NUD) S/T-omIndex to the index of the 3133 underlying interface the NS(NUD) arrived on and sets the Target 3134 Address to the same value included in the NS(NUD). The target next 3135 sets the R flag to 1, the S flag to 1 and the O flag to 1, then 3136 selects an in-window Identification for the ROS and performs 3137 fragmentation. The node then forwards the carrier packets into the 3138 unsecured spanning tree, directly to the ROS if it is in the local 3139 segment or directly to a Bridge in the local segment. 3141 When the ROS receives the NA(NUD), it marks the target underlying 3142 interface tested as "reachable". Note that underlying interface 3143 states are maintained independently of the overall NCE REACHABLE 3144 state, and that a single NCE may have multiple target underlying 3145 interfaces in various states "reachable" and otherwise while the NCE 3146 state as a whole remains REACHABLE. 3148 Note also that the exchange of NS/NA(NUD) messages has the useful 3149 side-benefit of opening holes in NATs that may be useful for NAT 3150 traversal. 3152 3.16. Mobility Management and Quality of Service (QoS) 3154 AERO is a Distributed Mobility Management (DMM) service. Each Proxy/ 3155 Server is responsible for only a subset of the Clients on the OMNI 3156 link, as opposed to a Centralized Mobility Management (CMM) service 3157 where there is a single network mobility collective entity for all 3158 Clients. Clients coordinate with their associated Proxy/Servers via 3159 RS/RA exchanges to maintain the DMM profile, and the AERO routing 3160 system tracks all current Client/Proxy/Server peering relationships. 3162 Proxy/Servers provide default routing and mobility/multilink services 3163 for their dependent Clients. Clients are responsible for maintaining 3164 neighbor relationships with their Proxy/Servers through periodic RS/ 3165 RA exchanges, which also serves to confirm neighbor reachability. 3166 When a Client's underlying Interface Attributes change, the Client is 3167 responsible for updating the Proxy/Server with this new information. 3169 Note that when there is a Proxy/Server in the path, the Proxy 3170 function can also perform some RS/RA exchanges on the Client's 3171 behalf. 3173 Mobility management messaging is based on the transmission and 3174 reception of unsolicited Neighbor Advertisement (uNA) messages. Each 3175 uNA message sets the IPv6 source address to the LLA of the ROR and 3176 the destination address to the unicast LLA of the ROS. 3178 Mobility management considerations are specified in the following 3179 sections. 3181 3.16.1. Mobility Update Messaging 3183 RORs accommodate Client mobility and/or multilink change events by 3184 sending secured uNA messages to each ROS in the target Client's 3185 Report List. When an ROR sends a uNA message, it sets the IPv6 3186 source address to the its own LLA, sets the destination address to 3187 the ROS LLA (i.e., an MNP-LLA if the ROS is a Client and an ADM-LLA 3188 if the ROS is a Proxy/Server) and sets the Target Address to the 3189 Client's MNP-LLA. The ROR also includes an OMNI option with Preflen 3190 set to the prefix length associated with the Client's MNP-LLA, with 3191 Interface Attributes for the target Client's underlying interfaces 3192 and with the OMNI header S/T-omIndex set to 0. The ROR then sets the 3193 uNA R flag to 1, S flag to 0 and O flag to 1, then encapsulates the 3194 message in an OAL header with source set to its own ADM-ULA and 3195 destination set to the ROS ULA (i.e., the ADM-ULA of the ROS Proxy/ 3196 Server) and sends the message into the secured spanning tree. 3198 As discussed in Section 7.2.6 of [RFC4861], the transmission and 3199 reception of uNA messages is unreliable but provides a useful 3200 optimization. In well-connected Internetworks with robust data links 3201 uNA messages will be delivered with high probability, but in any case 3202 the Proxy/Server can optionally send up to MAX_NEIGHBOR_ADVERTISEMENT 3203 uNAs to each ROS to increase the likelihood that at least one will be 3204 received. Alternatively, the Proxy/Server can set the PNG flag in 3205 the uNA OMNI option header to request a solicited NA acknowledgement 3206 as specified in [I-D.templin-6man-omni]. 3208 When the ROS Proxy/Server receives a uNA message prepared as above, 3209 it ignores the message if the destination is not its own ADM-ULA or 3210 the MNP-ULA of the ROS Client. In the former case, it uses the 3211 included OMNI option information to update its NCE for the target, 3212 but does not reset ReachableTime since the receipt of an unsolicited 3213 NA message from the ROR does not provide confirmation that any 3214 forward paths to the target Client are working. If the destination 3215 was the MNP-ULA of the ROS Client, the ROS Proxy/Server instead re- 3216 encapsulates with the OAL source set to its own ADM-ULA, OAL 3217 destination set to the MNP-ULA of the ROS Client with an 3218 authentication signature if necessary, and with an in-window 3219 Identification for this Client. Finally, if the uNA message PNG flag 3220 was set, the ROS returns a solicited NA acknowledgement as specified 3221 in [I-D.templin-6man-omni]. 3223 In addition to sending uNA messages to the current set of ROSs for 3224 the target Client, the ROR also sends uNAs to the MNP-ULA associated 3225 with the link-layer address for any underlying interface for which 3226 the link-layer address has changed. These uNA messages update an old 3227 Proxy/Server that cannot easily detect (e.g., without active probing) 3228 when a formerly-active Client has departed. When the ROR sends the 3229 uNA, it sets the IPv6 source address to its LLA, sets the destination 3230 address to the old Proxy/Server's ADM-LLA, and sets the Target 3231 Address to the Client's MNP-LLA. The ROR also includes an OMNI 3232 option with Preflen set to the prefix length associated with the 3233 Client's MNP-LLA, with Interface Attributes for the changed 3234 underlying interface, and with the OMNI header S/T-omIndex set to 0. 3235 The ROR then sets the uNA R flag to 1, S flag to 0 and O flag to 1, 3236 then encapsulates the message in an OAL header with source set to its 3237 own ULA and destination set to the ADM-ULA of the old Proxy/Server 3238 and sends the message into the secured spanning tree. 3240 3.16.2. Announcing Link-Layer Address and/or QoS Preference Changes 3242 When a Client needs to change its underlying Interface Attributes 3243 (e.g., due to a mobility event), the Client requests one of its 3244 Proxy/Servers to send uNA or RS messages to all of its other Proxy/ 3245 Servers via the secured spanning tree with an OMNI option that 3246 includes Interface attributes with the new link quality and address 3247 information. 3249 Up to MAX_RTR_SOLICITATIONS RS messages MAY be sent in parallel with 3250 sending carrier packets containing user data in case one or more RAs 3251 are lost. If all RAs are lost, the Client SHOULD re-associate with a 3252 new Proxy/Server. 3254 When the Proxy/Server receives the Client's changes, it sends uNA 3255 messages to all nodes in the Report List the same as described in the 3256 previous section. 3258 3.16.3. Bringing New Links Into Service 3260 When a Client needs to bring new underlying interfaces into service 3261 (e.g., when it activates a new data link), it sends an RS message to 3262 the Proxy/Server via the underlying interface with an OMNI option 3263 that includes Interface Attributes with appropriate link quality 3264 values and with link-layer address information for the new link. 3266 3.16.4. Deactivating Existing Links 3268 When a Client needs to deactivate an existing underlying interface, 3269 it sends an RS or uNA message to its Proxy/Server with an OMNI option 3270 with appropriate Interface Attribute values - in particular, the link 3271 quality value 0 assures that neighbors will cease to use the link. 3273 If the Client needs to send RS/uNA messages over an underlying 3274 interface other than the one being deactivated, it MUST include 3275 Interface Attributes with appropriate link quality values for any 3276 underlying interfaces being deactivated. 3278 Note that when a Client deactivates an underlying interface, 3279 neighbors that have received the RS/uNA messages need not purge all 3280 references for the underlying interface from their neighbor cache 3281 entries. The Client may reactivate or reuse the underlying interface 3282 and/or its omIndex at a later point in time, when it will send RS/uNA 3283 messages with fresh Interface Attributes to update any neighbors. 3285 3.16.5. Moving Between Proxy/Servers 3287 The Client performs the procedures specified in Section 3.12.2 when 3288 it first associates with a new FHS Proxy/Server or renews its 3289 association with an existing Proxy/Server. The Client also includes 3290 MS-Release identifiers in the RS message OMNI option per 3291 [I-D.templin-6man-omni] if it wants the new Proxy/Server to notify 3292 any old Proxy/Servers from which the Client is departing. 3294 When the new FHS Proxy/Server receives the Client's RS message, it 3295 returns an RA as specified in Section 3.12.3 and sends RS messages to 3296 any old Proxy/Servers listed in OMNI option MS-Release identifiers. 3297 When the new Proxy/Server sends an RS message, it sets the source to 3298 the MNP-LLA of the Client and sets the destination to the ADM-LLA of 3299 the old Proxy/Server. The new Proxy/Server also includes an OMNI 3300 option with Preflen set to the prefix length associated with the 3301 Client's MNP-LLA, with Interface Attributes for its own underlying 3302 interface, and with the OMNI header S/T-omIndex set to 0. The new 3303 Proxy/Server then encapsulates the message in an OAL header with 3304 source set to its own ADM-ULA and destination set to the ADM-ULA of 3305 the old Proxy/Server and sends the message into the secured spanning 3306 tree. 3308 When an old Proxy/Server receives the RS, it notices that the message 3309 appears to have originated from the Client's MNP-LLA but that the S/ 3310 T-omIndex is 0. The old Proxy/Server then changes the Client's NCE 3311 state to DEPARTED, sets the link-layer address of the Client to the 3312 new Proxy/Server's ADM-ULA, and resets DepartTime. The old Proxy/ 3313 Server then returns an RA message via the secured spanning tree by 3314 reversing the LLA and ULA addresses found in the RS message. After a 3315 short delay (e.g., 2 seconds) the old Proxy/Server withdraws the 3316 Client's MNP from the routing system. After DepartTime expires, the 3317 old Proxy/Server deletes the Client's NCE. 3319 The old Proxy/Server also iteratively forwards a copy of the uNA 3320 message to each ROS in the Client's Report List by changing the OAL 3321 destination address to the ULA of the ROS while leaving all other 3322 fields of the message unmodified. When the ROS receives the uNA, it 3323 examines the source address to determine the target Client NCE and 3324 verifies that the destination address matches the old Proxy/Server. 3325 The ROS then caches the ULA source address as the new Proxy/Server 3326 for the existing NCE and marks the entry as STALE. While in the 3327 STALE state, the ROS allows new carrier packets to flow according to 3328 any alternate reachable underlying interfaces and sends new NS(AR) 3329 messages using its own ULA as the OAL source and the ADM-ULA of the 3330 new Proxy/Server as the OAL destination address to elicit NA(AR) 3331 messages that reset the NCE state to REACHABLE. 3333 Clients SHOULD NOT move rapidly between Proxy/Servers in order to 3334 avoid causing excessive oscillations in the AERO routing system. 3335 Examples of when a Client might wish to change to a different Proxy/ 3336 Server include a Proxy/Server that has gone unreachable, topological 3337 movements of significant distance, movement to a new geographic 3338 region, movement to a new OMNI link segment, etc. 3340 When a Client moves to a new Proxy/Server, some of the carrier 3341 packets of a multiple fragment OAL packet may have already arrived at 3342 the old Proxy/Server while others are en route to the new Proxy/ 3343 Server, however no special attention in the reassembly algorithm is 3344 necessary since all carrier packets will eventually arrive at the 3345 Client which can then reassemble. However, any carrier packets that 3346 are somehow lost can often be recovered through retransmissions. 3348 3.17. Multicast 3350 The AERO Client provides an IGMP (IPv4) [RFC2236] or MLD (IPv6) 3351 [RFC3810] proxy service for its EUNs and/or hosted applications 3352 [RFC4605]. Proxy/Servers act as a Protocol Independent Multicast - 3353 Sparse-Mode (PIM-SM, or simply "PIM") Designated Router (DR) 3354 [RFC7761]. AERO Relays also act as PIM routers (i.e., the same as 3355 AERO Proxys/Servers) on behalf of nodes on INET/EUN networks. 3357 Clients on ANET underlying interfaces for which the ANET has deployed 3358 native multicast services forward IGMP/MLD messages into the ANET. 3359 The IGMP/MLD messages may be further forwarded by a first-hop ANET 3360 access router acting as an IGMP/MLD-snooping switch [RFC4541], then 3361 ultimately delivered to an ANET FHS Proxy/Server. 3363 Clients on ANET underlying interfaces without native multicast 3364 services instead send NS(AR) messages to cause their FHS Proxy/Server 3365 to act as an ROS and forward the message to an LHS Proxy/Server ROR. 3366 Clients on INET interfaces act as an ROS on their own behalf and 3367 forward NS(AR) messages directly to the LHS Proxy/Server ROR (i.e., 3368 via the FHS Proxy/Server as a proxy). When the Client receives an 3369 NA(AR) response, it initiates PIM protocol messaging according to the 3370 Source-Specific Multicast (SSM) and Any-Source Multicast (ASM) 3371 operational modes as discussed in the following sections. 3373 3.17.1. Source-Specific Multicast (SSM) 3375 When an ROS "X" (i.e., either a ROS Client or its FHS Proxy Server) 3376 acting as PIM router receives a Join/Prune message from a node on its 3377 downstream interfaces containing one or more ((S)ource, (G)roup) 3378 pairs, it updates its Multicast Routing Information Base (MRIB) 3379 accordingly. For each S belonging to a prefix reachable via X's non- 3380 OMNI interfaces, X then forwards the (S, G) Join/Prune to any PIM 3381 routers on those interfaces per [RFC7761]. 3383 For each S belonging to a prefix reachable via X's OMNI interface, X 3384 sends an NS(AR) message (see: Section 3.14) using its own LLA as the 3385 source address and the LLA of S as the destination address. X then 3386 encapsulates the NS(AR) in an OAL header with source address set to 3387 the ULA of X and destination address set to the solicited node 3388 multicast address for S, then forwards the message into the secured 3389 spanning tree, which delivers it to ROR "Y" that services S. The 3390 resulting NA(AR) will return the LLA for the prefix that matches S as 3391 the network-layer source address and with an OMNI option with 3392 interface attributes for any underlying interfaces that are currently 3393 servicing S. 3395 When X processes the NA(AR) it selects one or more underlying 3396 interfaces for S and performs an NS/NA(WIN) exchange while including 3397 a PIM Join/Prune message for each multicast group of interest in the 3398 OMNI option. If S is located behind any Proxys "Z"*, each Z* then 3399 updates its MRIB accordingly and maintains the LLA of X as the next 3400 hop in the reverse path. Since the Bridges do not examine network 3401 layer control messages, this means that the (reverse) multicast tree 3402 path is simply from each Z* (and/or S) to X with no other multicast- 3403 aware routers in the path. 3405 Following the initial combined Join/Prune and NS/NA messaging, X 3406 maintains a NCE for each S the same as if X was sending unicast data 3407 traffic to S. In particular, X performs additional NS/NA exchanges 3408 to keep the NCE alive for up to t_periodic seconds [RFC7761]. If no 3409 new Joins are received within t_periodic seconds, X allows the NCE to 3410 expire. Finally, if X receives any additional Join/Prune messages 3411 for (S,G) it forwards the messages over the secured spanning tree. 3413 At some later time, Client C that holds an MNP for source S may 3414 depart from a first Proxy/Server Z1 and/or connect via a new Proxy/ 3415 Server Z2. In that case, Y sends a uNA message to X the same as 3416 specified for unicast mobility in Section 3.16. When X receives the 3417 uNA message, it updates its NCE for the LLA for source S and sends 3418 new Join messages to any new Proxys Z2. There is no requirement to 3419 send any Prune messages to old Proxy/Server Z1 since source S will no 3420 longer source any multicast data traffic via Z1. Instead, the 3421 multicast state for (S,G) in Proxy/Server Z1 will soon time out since 3422 no new Joins will arrive. 3424 After some later time, C may move to a new Proxy/Server Y2 and depart 3425 from old Sever Y1. In that case, Y1 sends Join messages for any of 3426 C's active (S,G) groups to Y2 while including its own LLA as the 3427 source address. This causes Y2 to include Y1 in the multicast 3428 forwarding tree during the interim time that Y1's NCE for C is in the 3429 DEPARTED state. At the same time, Y1 sends a uNA message to X with 3430 an OMNI option with S/T-omIndex set to 0 and a release indication to 3431 cause X to release its NCE for S. X then sends a new Join message to 3432 S via the secured spanning tree and re-initiates route optimization 3433 the same as if it were receiving a fresh Join message from a node on 3434 a downstream link. 3436 3.17.2. Any-Source Multicast (ASM) 3438 When an ROS X acting as a PIM router receives a Join/Prune from a 3439 node on its downstream interfaces containing one or more (*,G) pairs, 3440 it updates its Multicast Routing Information Base (MRIB) accordingly. 3441 X then forwards a copy of the message within the OMNI option of an 3442 NS(WIN) message to the Rendezvous Point (RP) R for each G over the 3443 secured spanning tree. X uses its own LLA as the source address and 3444 the LLA for R as the destination address, then encapsulates the 3445 NS(WIN) message in an OAL header with source address set to the ULA 3446 of X and destination address set to the ULA of R's Proxy/Server then 3447 sends the message into the secured spanning tree. 3449 For each source S that sends multicast traffic to group G via R, the 3450 Proxy/Server Z* for the Client that aggregates S encapsulates the 3451 original IP packets in PIM Register messages and forwards them to R 3452 via the secured spanning tree, which may then elect to send a PIM 3453 Join to Z*. This will result in an (S,G) tree rooted at Z* with R as 3454 the next hop so that R will begin to receive two copies of the 3455 original IP packet; one native copy from the (S, G) tree and a second 3456 copy from the pre-existing (*, G) tree that still uses PIM Register 3457 encapsulation. R can then issue a PIM Register-stop message to 3458 suppress the Register-encapsulated stream. At some later time, if C 3459 moves to a new Proxy/Server Z*, it resumes sending original IP 3460 packets via PIM Register encapsulation via the new Z*. 3462 At the same time, as multicast listeners discover individual S's for 3463 a given G, they can initiate an (S,G) Join for each S under the same 3464 procedures discussed in Section 3.17.1. Once the (S,G) tree is 3465 established, the listeners can send (S, G) Prune messages to R so 3466 that multicast original IP packets for group G sourced by S will only 3467 be delivered via the (S, G) tree and not from the (*, G) tree rooted 3468 at R. All mobility considerations discussed for SSM apply. 3470 3.17.3. Bi-Directional PIM (BIDIR-PIM) 3472 Bi-Directional PIM (BIDIR-PIM) [RFC5015] provides an alternate 3473 approach to ASM that treats the Rendezvous Point (RP) as a Designated 3474 Forwarder (DF). Further considerations for BIDIR-PIM are out of 3475 scope. 3477 3.18. Operation over Multiple OMNI Links 3479 An AERO Client can connect to multiple OMNI links the same as for any 3480 data link service. In that case, the Client maintains a distinct 3481 OMNI interface for each link, e.g., 'omni0' for the first link, 3482 'omni1' for the second, 'omni2' for the third, etc. Each OMNI link 3483 would include its own distinct set of Bridges and Proxy/Servers, 3484 thereby providing redundancy in case of failures. 3486 Each OMNI link could utilize the same or different ANET connections. 3487 The links can be distinguished at the link-layer via the SRT prefix 3488 in a similar fashion as for Virtual Local Area Network (VLAN) tagging 3489 (e.g., IEEE 802.1Q) and/or through assignment of distinct sets of 3490 MSPs on each link. This gives rise to the opportunity for supporting 3491 multiple redundant networked paths, with each VLAN distinguished by a 3492 different SRT "color" (see: Section 3.2.5). 3494 The Client's IP layer can select the outgoing OMNI interface 3495 appropriate for a given traffic profile while (in the reverse 3496 direction) correspondent nodes must have some way of steering their 3497 original IP packets destined to a target via the correct OMNI link. 3499 In a first alternative, if each OMNI link services different MSPs, 3500 then the Client can receive a distinct MNP from each of the links. 3501 IP routing will therefore assure that the correct OMNI link is used 3502 for both outbound and inbound traffic. This can be accomplished 3503 using existing technologies and approaches, and without requiring any 3504 special supporting code in correspondent nodes or Bridges. 3506 In a second alternative, if each OMNI link services the same MSP(s) 3507 then each link could assign a distinct "OMNI link Anycast" address 3508 that is configured by all Bridges on the link. Correspondent nodes 3509 can then perform Segment Routing to select the correct SRT, which 3510 will then direct the original IP packet over multiple hops to the 3511 target. 3513 3.19. DNS Considerations 3515 AERO Client MNs and INET correspondent nodes consult the Domain Name 3516 System (DNS) the same as for any Internetworking node. When 3517 correspondent nodes and Client MNs use different IP protocol versions 3518 (e.g., IPv4 correspondents and IPv6 MNs), the INET DNS must maintain 3519 A records for IPv4 address mappings to MNs which must then be 3520 populated in Relay NAT64 mapping caches. In that way, an IPv4 3521 correspondent node can send original IPv4 packets to the IPv4 address 3522 mapping of the target MN, and the Relay will translate the IPv4 3523 header and destination address into an IPv6 header and IPv6 3524 destination address of the MN. 3526 When an AERO Client registers with an AERO Proxy/Server, the Proxy/ 3527 Server can return the address(es) of DNS servers in RDNSS options 3528 [RFC6106]. The DNS server provides the IP addresses of other MNs and 3529 correspondent nodes in AAAA records for IPv6 or A records for IPv4. 3531 3.20. Transition/Coexistence Considerations 3533 OAL encapsulation ensures that dissimilar INET partitions can be 3534 joined into a single unified OMNI link, even though the partitions 3535 themselves may have differing protocol versions and/or incompatible 3536 addressing plans. However, a commonality can be achieved by 3537 incrementally distributing globally routable (i.e., native) IP 3538 prefixes to eventually reach all nodes (both mobile and fixed) in all 3539 OMNI link segments. This can be accomplished by incrementally 3540 deploying AERO Bridges on each INET partition, with each Bridge 3541 distributing its MNPs and/or discovering non-MNP IP GUA prefixes on 3542 its INET links. 3544 This gives rise to the opportunity to eventually distribute native IP 3545 addresses to all nodes, and to present a unified OMNI link view even 3546 if the INET partitions remain in their current protocol and 3547 addressing plans. In that way, the OMNI link can serve the dual 3548 purpose of providing a mobility/multilink service and a transition/ 3549 coexistence service. Or, if an INET partition is transitioned to a 3550 native IP protocol version and addressing scheme that is compatible 3551 with the OMNI link MNP-based addressing scheme, the partition and 3552 OMNI link can be joined by Bridges. 3554 Relays that connect INETs/EUNs with dissimilar IP protocol versions 3555 may need to employ a network address and protocol translation 3556 function such as NAT64 [RFC6146]. 3558 3.21. Detecting and Reacting to Proxy/Server and Bridge Failures 3560 In environments where rapid failure recovery is required, Proxy/ 3561 Servers and Bridges SHOULD use Bidirectional Forwarding Detection 3562 (BFD) [RFC5880]. Nodes that use BFD can quickly detect and react to 3563 failures so that cached information is re-established through 3564 alternate nodes. BFD control messaging is carried only over well- 3565 connected ground domain networks (i.e., and not low-end radio links) 3566 and can therefore be tuned for rapid response. 3568 Proxy/Servers and Bridges maintain BFD sessions in parallel with 3569 their BGP peerings. If a Proxy/Server or Bridge fails, BGP peers 3570 will quickly re-establish routes through alternate paths the same as 3571 for common BGP deployments. Similarly, Proxys maintain BFD sessions 3572 with their associated Bridges even though they do not establish BGP 3573 peerings with them. 3575 3.22. AERO Clients on the Open Internet 3577 AERO Clients that connect to the open Internet via INET interfaces 3578 can establish a VPN or direct link to securely connect to a FHS 3579 Proxy/Server in a "tethered" arrangement with all of the Client's 3580 traffic transiting the Proxy/Server which acts as a router. 3581 Alternatively, the Client can associate with an INET FHS Proxy/Server 3582 using UDP/IP encapsulation and control message securing services as 3583 discussed in the following sections. 3585 When a Client's OMNI interface enables an INET underlying interface, 3586 it first examines the INET address. For IPv4, the Client assumes it 3587 is on the open Internet if the INET address is not a special-use IPv4 3588 address per [RFC3330]. Similarly for IPv6, the Client assumes it is 3589 on the open Internet if the INET address is a Global Unicast Address 3590 (GUA) [RFC4291]. Otherwise, the Client should assume it is behind 3591 one or several NATs. 3593 The Client then prepares an RS message with IPv6 source address set 3594 to its MNP-LLA, with IPv6 destination set to (link-local) All-Routers 3595 multicast and with an OMNI option with underlying interface 3596 attributes. If the Client believes that it is on the open Internet, 3597 it SHOULD include an L2ADDR in the Interface Attributes sub-option 3598 corresponding to the underlying interface; otherwise, it MAY set 3599 L2ADDR to 0. If the underlying address is IPv4, the Client includes 3600 the Port Number and IPv4 address written in obfuscated form [RFC4380] 3601 as discussed in Section 3.3. If the underlying interface address is 3602 IPv6, the Client instead includes the Port Number and IPv6 address in 3603 obfuscated form. The Client finally includes an authentication 3604 signature sub-option in the OMNI option [I-D.templin-6man-omni] to 3605 provide message authentication, selects an Identification value and 3606 window synchronization parameters, and submits the RS for OAL 3607 encapsulation. The Client then encapsulates the OAL fragment in UDP/ 3608 IP headers to form a carrier packet, sets the UDP/IP source to its 3609 INET address and UDP port, sets the UDP/IP destination to the FHS 3610 Proxy/Server's INET address and the AERO service port number (8060), 3611 then sends the carrier packet to the Proxy/Server. 3613 When the FHS Proxy/Server receives the RS, it discards the OAL 3614 encapsulation, authenticates the RS message, creates a NCE and 3615 registers the Client's MNP, window synchronization state and INET 3616 interface information according to the OMNI option parameters. If 3617 the RS message OMNI option includes Interface Attributes with an 3618 L2ADDR, the Proxy/Server compares the encapsulation IP address and 3619 UDP port number with the (unobfuscated) values. If the values are 3620 the same, the Proxy/Server caches the Client's information as "INET" 3621 addresses meaning that the Client is likely to accept direct messages 3622 without requiring NAT traversal exchanges. If the values are 3623 different (or, if the OMNI option did not include an L2ADDR) the 3624 Proxy/Server instead caches the Client's information as "mapped" 3625 addresses meaning that NAT traversal exchanges may be necessary. 3627 The FHS Proxy/Server then prepares an RA message with IPv6 source and 3628 destination set corresponding to the addresses in the RS, and with an 3629 OMNI option with an Origin Indication sub-option per 3630 [I-D.templin-6man-omni] with the mapped and obfuscated Port Number 3631 and IP address observed in the encapsulation headers. The Proxy/ 3632 Server also includes an Interface Attributes sub-option for its 3633 underlying interface with FMT/SRT/LHS information appropriate for its 3634 INET interface, and with an authentication signature sub-option per 3635 [I-D.templin-6man-omni] and/or a symmetric window synchronization/ 3636 acknowledgement if necessary. The Proxy/Server then performs OAL 3637 encapsulation and fragmentation if necessary and encapsulates each 3638 fragment in UDP/IP headers with addresses set per the L2ADDR 3639 information in the NCE for the Client. 3641 When the Client receives the RA, it authenticates the message then 3642 process the window synchronization/acknowledgement and compares the 3643 mapped Port Number and IP address from the Origin Indication sub- 3644 option with its own address. If the addresses are the same, the 3645 Client assumes the open Internet / Cone NAT principle; if the 3646 addresses are different, the Client instead assumes that further 3647 qualification procedures are necessary to detect the type of NAT and 3648 proceeds according to standard procedures [RFC6081][RFC4380]. The 3649 Client also caches the RA Interface Attributes FMT/SRT/LHS 3650 information to discover the Proxy/Server's spanning tree orientation. 3651 The Client finally arranges to return an explicit/implicit 3652 acknowledgement, and sends periodic RS messages to receive fresh RA 3653 messages before the Router Lifetime received on each INET interface 3654 expires. 3656 When the Client sends messages to target IP addresses, it also 3657 invokes route optimization per Section 3.14. For route optimized 3658 targets in the same OMNI link segment, if the target's L2ADDR is on 3659 the open INET, the Client forwards carrier packets directly to the 3660 target INET address. If the target is behind a NAT, the Client first 3661 establishes NAT state for the L2ADDR using the "direct bubble" and 3662 NUD mechanisms discussed in Section 3.10.1. The Client continues to 3663 send carrier packets via its Proxy/Server until NAT state is 3664 populated, then begins forwarding carrier packets via the direct path 3665 through the NAT to the target. For targets in different OMNI link 3666 segments, the Client uses OAL/ORH encapsulation and forwards carrier 3667 packets to the Bridge that returned the NA(AR) message. 3669 The Client can send original IP packets to route-optimized neighbors 3670 in the same OMNI link segment no larger than the minimum/path MPS in 3671 one piece and with OAL encapsulation as atomic fragments. For larger 3672 original IP packets, the Client applies OAL encapsulation then 3673 fragments if necessary according to Section 3.9, with OAL header with 3674 source set to its own MNP-ULA and destination set to the MNP-ULA of 3675 the target, and with an in-window Identification value. The Client 3676 then encapsulates each resulting carrier packet in UDP/IP *NET 3677 headers and sends them to the next hop. 3679 Note: The NAT traversal procedures specified in this document are 3680 applicable for Cone, Address-Restricted and Port-Restricted NATs 3681 only. While future updates to this document may specify procedures 3682 for other NAT variations (e.g., hairpinning and various forms of 3683 Symmetric NATs), it should be noted that continuous communications 3684 are always possible through forwarding via a Proxy/Server even if NAT 3685 traversal is not employed. 3687 3.23. Time-Varying MNPs 3689 In some use cases, it is desirable, beneficial and efficient for the 3690 Client to receive a constant MNP that travels with the Client 3691 wherever it moves. For example, this would allow air traffic 3692 controllers to easily track aircraft, etc. In other cases, however 3693 (e.g., intelligent transportation systems), the MN may be willing to 3694 sacrifice a modicum of efficiency in order to have time-varying MNPs 3695 that can be changed every so often to defeat adversarial tracking. 3697 The DHCPv6 service offers a way for Clients that desire time-varying 3698 MNPs to obtain short-lived prefixes (e.g., on the order of a small 3699 number of minutes). In that case, the identity of the Client would 3700 not be bound to the MNP but rather to a Node Identification value 3701 (see: [I-D.templin-6man-omni]) to be used as the Client ID seed for 3702 MNP prefix delegation. The Client would then be obligated to 3703 renumber its internal networks whenever its MNP (and therefore also 3704 its MNP-LLA) changes. This should not present a challenge for 3705 Clients with automated network renumbering services, however presents 3706 limits for the durations of ongoing sessions that would prefer to use 3707 a constant address. 3709 4. Implementation Status 3711 An early AERO implementation based on OpenVPN (https://openvpn.net/) 3712 was announced on the v6ops mailing list on January 10, 2018 and an 3713 initial public release of the AERO proof-of-concept source code was 3714 announced on the intarea mailing list on August 21, 2015. 3716 AERO Release-3.2 was tagged on March 30, 2021, and is undergoing 3717 internal testing. Additional internal releases expected within the 3718 coming months, with first public release expected end of 1H2021. 3720 Many AERO/OMNI functions are implemented and undergoing final 3721 integration. OAL fragmentation/reassembly buffer management code has 3722 been cleared for public release and will be presented at the June 3723 2021 ICAO mobility subgroup meeting. 3725 5. IANA Considerations 3727 The IANA is instructed to assign a new type value TBD1 in the IPv6 3728 Routing Types registry (IANA registration procedure is IETF Review or 3729 IESG Approval). 3731 The IANA has assigned the UDP port number "8060" for an earlier 3732 experimental first version of AERO [RFC6706]. This document 3733 obsoletes [RFC6706], and together with [I-D.templin-6man-omni] 3734 reclaims the UDP port number "8060" for 'aero' as the service port 3735 for UDP/IP encapsulation. (Note that, although [RFC6706] was not 3736 widely implemented or deployed, any messages coded to that 3737 specification can be easily distinguished and ignored since they use 3738 the invalid ICMPv6 message type number '0'.) This document makes no 3739 request of IANA, since [I-D.templin-6man-omni] already provides 3740 instructions. 3742 No further IANA actions are required. 3744 6. Security Considerations 3746 AERO Bridges configure secured tunnels with AERO Proxy/Servers and 3747 Relays within their local OMNI link segments. Applicable secured 3748 tunnel alternatives include IPsec [RFC4301], TLS/SSL [RFC8446], DTLS 3749 [RFC6347], WireGuard [WG], etc. The AERO Bridges of all OMNI link 3750 segments in turn configure secured tunnels for their neighboring AERO 3751 Bridges in a secured spanning tree topology. Therefore, control 3752 messages exchanged between any pair of OMNI link neighbors over the 3753 secured spanning tree are already protected. 3755 To prevent spoofing vectors, Proxy/Servers MUST discard without 3756 responding to any unsecured NS(AR) messages. Also, Proxy/Servers 3757 MUST discard without forwarding any original IP packets received from 3758 one of their own Clients (whether directly or following OAL 3759 reassembly) with a source address that does not match the Client's 3760 MNP and/or a destination address that does match the Client's MNP. 3761 Finally, Proxy/Servers MUST discard without forwarding any carrier 3762 packets with an OAL source and destination that both match the same 3763 MNP (i.e., after consulting the ORH if present). 3765 For INET partitions that require strong security in the data plane, 3766 two options for securing communications include 1) disable route 3767 optimization so that all traffic is conveyed over secured tunnels, or 3768 2) enable on-demand secure tunnel creation between Client neighbors. 3769 Option 1) would result in longer routes than necessary and impose 3770 traffic concentration on critical infrastructure elements. Option 2) 3771 could be coordinated between Clients using NS/NA messages with OMNI 3772 Host Identity Protocol (HIP) "Initiator/Responder" message sub- 3773 options [RFC7401][I-D.templin-6man-omni] to create a secured tunnel 3774 on-demand. 3776 AERO Clients that connect to secured ANETs need not apply security to 3777 their ND messages, since the messages will be authenticated and 3778 forwarded by a perimeter Proxy/Server that applies security on its 3779 INET-facing interface as part of the spanning tree (see above). AERO 3780 Clients connected to the open INET can use network and/or transport 3781 layer security services such as VPNs or can by some other means 3782 establish a direct link to a Proxy/Server. When a VPN or direct link 3783 may be impractical, however, INET Clients and Proxy/Servers SHOULD 3784 include and verify authentication signatures for their IPv6 ND 3785 messages as specified in [I-D.templin-6man-omni]. 3787 Application endpoints SHOULD use transport-layer (or higher-layer) 3788 security services such as TLS/SSL, DTLS or SSH [RFC4251] to assure 3789 the same level of protection as for critical secured Internet 3790 services. AERO Clients that require host-based VPN services SHOULD 3791 use network and/or transport layer security services such as IPsec, 3792 TLS/SSL, DTLS, etc. AERO Proxys and Proxy/Servers can also provide a 3793 network-based VPN service on behalf of the Client, e.g., if the 3794 Client is located within a secured enclave and cannot establish a VPN 3795 on its own behalf. 3797 AERO Proxy/Servers and Bridges present targets for traffic 3798 amplification Denial of Service (DoS) attacks. This concern is no 3799 different than for widely-deployed VPN security gateways in the 3800 Internet, where attackers could send spoofed packets to the gateways 3801 at high data rates. This can be mitigated through the AERO/OMNI data 3802 origin authentication procedures, as well as connecting Proxy/Servers 3803 and Bridges over dedicated links with no connections to the Internet 3804 and/or when connections to the Internet are only permitted through 3805 well-managed firewalls. Traffic amplification DoS attacks can also 3806 target an AERO Client's low data rate links. This is a concern not 3807 only for Clients located on the open Internet but also for Clients in 3808 secured enclaves. AERO Proxy/Servers and Proxys can institute rate 3809 limits that protect Clients from receiving packet floods that could 3810 DoS low data rate links. 3812 AERO Relays must implement ingress filtering to avoid a spoofing 3813 attack in which spurious messages with ULA addresses are injected 3814 into an OMNI link from an outside attacker. AERO Clients MUST ensure 3815 that their connectivity is not used by unauthorized nodes on their 3816 EUNs to gain access to a protected network, i.e., AERO Clients that 3817 act as routers MUST NOT provide routing services for unauthorized 3818 nodes. (This concern is no different than for ordinary hosts that 3819 receive an IP address delegation but then "share" the address with 3820 other nodes via some form of Internet connection sharing such as 3821 tethering.) 3823 The MAP list MUST be well-managed and secured from unauthorized 3824 tampering, even though the list contains only public information. 3825 The MAP list can be conveyed to the Client in a similar fashion as in 3826 [RFC5214] (e.g., through layer 2 data link login messaging, secure 3827 upload of a static file, DNS lookups, etc.). 3829 The AERO service for open INET Clients depends on a public key 3830 distribution service in which Client public keys and identities are 3831 maintained in a shared database accessible to all open INET Proxy/ 3832 Servers. Similarly, each Client must be able to determine the public 3833 key of each Proxy/Server, e.g. by consulting an online database. 3834 When AERO nodes register their public keys indexed by a unique Host 3835 Identity Tag (HIT) [RFC7401] in a distributed database such as the 3836 DNS, and use the HIT as an identity for applying IPv6 ND message 3837 authentication signatures, a means for determining public key 3838 attestation is available. 3840 Security considerations for IPv6 fragmentation and reassembly are 3841 discussed in [I-D.templin-6man-omni]. In environments where spoofing 3842 is considered a threat, OMNI nodes SHOULD employ Identification 3843 window synchronization and OAL destinations SHOULD configure an (end- 3844 system-based) firewall. 3846 SRH authentication facilities are specified in [RFC8754]. Security 3847 considerations for accepting link-layer ICMP messages and reflected 3848 packets are discussed throughout the document. 3850 7. Acknowledgements 3852 Discussions in the IETF, aviation standards communities and private 3853 exchanges helped shape some of the concepts in this work. 3854 Individuals who contributed insights include Mikael Abrahamsson, Mark 3855 Andrews, Fred Baker, Bob Braden, Stewart Bryant, Brian Carpenter, 3856 Wojciech Dec, Pavel Drasil, Ralph Droms, Adrian Farrel, Nick Green, 3857 Sri Gundavelli, Brian Haberman, Bernhard Haindl, Joel Halpern, Tom 3858 Herbert, Sascha Hlusiak, Lee Howard, Zdenek Jaron, Andre Kostur, 3859 Hubert Kuenig, Ted Lemon, Andy Malis, Satoru Matsushima, Tomek 3860 Mrugalski, Madhu Niraula, Alexandru Petrescu, Behcet Saikaya, Michal 3861 Skorepa, Joe Touch, Bernie Volz, Ryuji Wakikawa, Tony Whyman, Lloyd 3862 Wood and James Woodyatt. Members of the IESG also provided valuable 3863 input during their review process that greatly improved the document. 3864 Special thanks go to Stewart Bryant, Joel Halpern and Brian Haberman 3865 for their shepherding guidance during the publication of the AERO 3866 first edition. 3868 This work has further been encouraged and supported by Boeing 3869 colleagues including Kyle Bae, M. Wayne Benson, Dave Bernhardt, Cam 3870 Brodie, John Bush, Balaguruna Chidambaram, Irene Chin, Bruce Cornish, 3871 Claudiu Danilov, Don Dillenburg, Joe Dudkowski, Wen Fang, Samad 3872 Farooqui, Anthony Gregory, Jeff Holland, Seth Jahne, Brian Jaury, 3873 Greg Kimberly, Ed King, Madhuri Madhava Badgandi, Laurel Matthew, 3874 Gene MacLean III, Kyle Mikos, Rob Muszkiewicz, Sean O'Sullivan, Vijay 3875 Rajagopalan, Greg Saccone, Rod Santiago, Kent Shuey, Brian Skeen, 3876 Mike Slane, Carrie Spiker, Katie Tran, Brendan Williams, Amelia 3877 Wilson, Julie Wulff, Yueli Yang, Eric Yeh and other members of the 3878 Boeing mobility, networking and autonomy teams. Kyle Bae, Wayne 3879 Benson, Madhuri Madhava Badgandi, Vijayasarathy Rajagopalan, Katie 3880 Tran and Eric Yeh are especially acknowledged for implementing the 3881 AERO functions as extensions to the public domain OpenVPN 3882 distribution. Chuck Klabunde is honored and remembered for his early 3883 leadership, and we mourn his untimely loss. 3885 Earlier works on NBMA tunneling approaches are found in 3886 [RFC2529][RFC5214][RFC5569]. 3888 Many of the constructs presented in this second edition of AERO are 3889 based on the author's earlier works, including: 3891 o The Internet Routing Overlay Network (IRON) 3892 [RFC6179][I-D.templin-ironbis] 3894 o Virtual Enterprise Traversal (VET) 3895 [RFC5558][I-D.templin-intarea-vet] 3897 o The Subnetwork Encapsulation and Adaptation Layer (SEAL) 3898 [RFC5320][I-D.templin-intarea-seal] 3900 o AERO, First Edition [RFC6706] 3902 Note that these works cite numerous earlier efforts that are not also 3903 cited here due to space limitations. The authors of those earlier 3904 works are acknowledged for their insights. 3906 This work is aligned with the NASA Safe Autonomous Systems Operation 3907 (SASO) program under NASA contract number NNA16BD84C. 3909 This work is aligned with the FAA as per the SE2025 contract number 3910 DTFAWA-15-D-00030. 3912 This work is aligned with the Boeing Commercial Airplanes (BCA) 3913 Internet of Things (IoT) and autonomy programs. 3915 This work is aligned with the Boeing Information Technology (BIT) 3916 MobileNet program. 3918 8. References 3920 8.1. Normative References 3922 [I-D.templin-6man-omni] 3923 Templin, F. L. and T. Whyman, "Transmission of IP Packets 3924 over Overlay Multilink Network (OMNI) Interfaces", draft- 3925 templin-6man-omni-03 (work in progress), April 2021. 3927 [RFC0791] Postel, J., "Internet Protocol", STD 5, RFC 791, 3928 DOI 10.17487/RFC0791, September 1981, 3929 . 3931 [RFC0792] Postel, J., "Internet Control Message Protocol", STD 5, 3932 RFC 792, DOI 10.17487/RFC0792, September 1981, 3933 . 3935 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 3936 Requirement Levels", BCP 14, RFC 2119, 3937 DOI 10.17487/RFC2119, March 1997, 3938 . 3940 [RFC2473] Conta, A. and S. Deering, "Generic Packet Tunneling in 3941 IPv6 Specification", RFC 2473, DOI 10.17487/RFC2473, 3942 December 1998, . 3944 [RFC3971] Arkko, J., Ed., Kempf, J., Zill, B., and P. Nikander, 3945 "SEcure Neighbor Discovery (SEND)", RFC 3971, 3946 DOI 10.17487/RFC3971, March 2005, 3947 . 3949 [RFC3972] Aura, T., "Cryptographically Generated Addresses (CGA)", 3950 RFC 3972, DOI 10.17487/RFC3972, March 2005, 3951 . 3953 [RFC4191] Draves, R. and D. Thaler, "Default Router Preferences and 3954 More-Specific Routes", RFC 4191, DOI 10.17487/RFC4191, 3955 November 2005, . 3957 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast 3958 Addresses", RFC 4193, DOI 10.17487/RFC4193, October 2005, 3959 . 3961 [RFC4380] Huitema, C., "Teredo: Tunneling IPv6 over UDP through 3962 Network Address Translations (NATs)", RFC 4380, 3963 DOI 10.17487/RFC4380, February 2006, 3964 . 3966 [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, 3967 "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, 3968 DOI 10.17487/RFC4861, September 2007, 3969 . 3971 [RFC4862] Thomson, S., Narten, T., and T. Jinmei, "IPv6 Stateless 3972 Address Autoconfiguration", RFC 4862, 3973 DOI 10.17487/RFC4862, September 2007, 3974 . 3976 [RFC6081] Thaler, D., "Teredo Extensions", RFC 6081, 3977 DOI 10.17487/RFC6081, January 2011, 3978 . 3980 [RFC7401] Moskowitz, R., Ed., Heer, T., Jokela, P., and T. 3981 Henderson, "Host Identity Protocol Version 2 (HIPv2)", 3982 RFC 7401, DOI 10.17487/RFC7401, April 2015, 3983 . 3985 [RFC7739] Gont, F., "Security Implications of Predictable Fragment 3986 Identification Values", RFC 7739, DOI 10.17487/RFC7739, 3987 February 2016, . 3989 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 3990 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 3991 May 2017, . 3993 [RFC8200] Deering, S. and R. Hinden, "Internet Protocol, Version 6 3994 (IPv6) Specification", STD 86, RFC 8200, 3995 DOI 10.17487/RFC8200, July 2017, 3996 . 3998 [RFC8415] Mrugalski, T., Siodelski, M., Volz, B., Yourtchenko, A., 3999 Richardson, M., Jiang, S., Lemon, T., and T. Winters, 4000 "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)", 4001 RFC 8415, DOI 10.17487/RFC8415, November 2018, 4002 . 4004 8.2. Informative References 4006 [BGP] Huston, G., "BGP in 2015, http://potaroo.net", January 4007 2016. 4009 [I-D.bonica-6man-comp-rtg-hdr] 4010 Bonica, R., Kamite, Y., Alston, A., Henriques, D., and L. 4011 Jalil, "The IPv6 Compact Routing Header (CRH)", draft- 4012 bonica-6man-comp-rtg-hdr-24 (work in progress), January 4013 2021. 4015 [I-D.bonica-6man-crh-helper-opt] 4016 Li, X., Bao, C., Ruan, E., and R. Bonica, "Compressed 4017 Routing Header (CRH) Helper Option", draft-bonica-6man- 4018 crh-helper-opt-03 (work in progress), April 2021. 4020 [I-D.ietf-intarea-frag-fragile] 4021 Bonica, R., Baker, F., Huston, G., Hinden, R. M., Troan, 4022 O., and F. Gont, "IP Fragmentation Considered Fragile", 4023 draft-ietf-intarea-frag-fragile-17 (work in progress), 4024 September 2019. 4026 [I-D.ietf-intarea-tunnels] 4027 Touch, J. and M. Townsley, "IP Tunnels in the Internet 4028 Architecture", draft-ietf-intarea-tunnels-10 (work in 4029 progress), September 2019. 4031 [I-D.ietf-ipwave-vehicular-networking] 4032 (editor), J. (. J., "IPv6 Wireless Access in Vehicular 4033 Environments (IPWAVE): Problem Statement and Use Cases", 4034 draft-ietf-ipwave-vehicular-networking-20 (work in 4035 progress), March 2021. 4037 [I-D.ietf-rtgwg-atn-bgp] 4038 Templin, F. L., Saccone, G., Dawra, G., Lindem, A., and V. 4039 Moreno, "A Simple BGP-based Mobile Routing System for the 4040 Aeronautical Telecommunications Network", draft-ietf- 4041 rtgwg-atn-bgp-10 (work in progress), January 2021. 4043 [I-D.templin-6man-dhcpv6-ndopt] 4044 Templin, F. L., "A Unified Stateful/Stateless 4045 Configuration Service for IPv6", draft-templin-6man- 4046 dhcpv6-ndopt-11 (work in progress), January 2021. 4048 [I-D.templin-intarea-seal] 4049 Templin, F. L., "The Subnetwork Encapsulation and 4050 Adaptation Layer (SEAL)", draft-templin-intarea-seal-68 4051 (work in progress), January 2014. 4053 [I-D.templin-intarea-vet] 4054 Templin, F. L., "Virtual Enterprise Traversal (VET)", 4055 draft-templin-intarea-vet-40 (work in progress), May 2013. 4057 [I-D.templin-ipwave-uam-its] 4058 Templin, F. L., "Urban Air Mobility Implications for 4059 Intelligent Transportation Systems", draft-templin-ipwave- 4060 uam-its-04 (work in progress), January 2021. 4062 [I-D.templin-ironbis] 4063 Templin, F. L., "The Interior Routing Overlay Network 4064 (IRON)", draft-templin-ironbis-16 (work in progress), 4065 March 2014. 4067 [I-D.templin-v6ops-pdhost] 4068 Templin, F. L., "IPv6 Prefix Delegation and Multi- 4069 Addressing Models", draft-templin-v6ops-pdhost-27 (work in 4070 progress), January 2021. 4072 [OVPN] OpenVPN, O., "http://openvpn.net", October 2016. 4074 [RFC1035] Mockapetris, P., "Domain names - implementation and 4075 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 4076 November 1987, . 4078 [RFC1812] Baker, F., Ed., "Requirements for IP Version 4 Routers", 4079 RFC 1812, DOI 10.17487/RFC1812, June 1995, 4080 . 4082 [RFC2003] Perkins, C., "IP Encapsulation within IP", RFC 2003, 4083 DOI 10.17487/RFC2003, October 1996, 4084 . 4086 [RFC2004] Perkins, C., "Minimal Encapsulation within IP", RFC 2004, 4087 DOI 10.17487/RFC2004, October 1996, 4088 . 4090 [RFC2236] Fenner, W., "Internet Group Management Protocol, Version 4091 2", RFC 2236, DOI 10.17487/RFC2236, November 1997, 4092 . 4094 [RFC2464] Crawford, M., "Transmission of IPv6 Packets over Ethernet 4095 Networks", RFC 2464, DOI 10.17487/RFC2464, December 1998, 4096 . 4098 [RFC2529] Carpenter, B. and C. Jung, "Transmission of IPv6 over IPv4 4099 Domains without Explicit Tunnels", RFC 2529, 4100 DOI 10.17487/RFC2529, March 1999, 4101 . 4103 [RFC2983] Black, D., "Differentiated Services and Tunnels", 4104 RFC 2983, DOI 10.17487/RFC2983, October 2000, 4105 . 4107 [RFC3168] Ramakrishnan, K., Floyd, S., and D. Black, "The Addition 4108 of Explicit Congestion Notification (ECN) to IP", 4109 RFC 3168, DOI 10.17487/RFC3168, September 2001, 4110 . 4112 [RFC3330] IANA, "Special-Use IPv4 Addresses", RFC 3330, 4113 DOI 10.17487/RFC3330, September 2002, 4114 . 4116 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 4117 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 4118 DOI 10.17487/RFC3810, June 2004, 4119 . 4121 [RFC4122] Leach, P., Mealling, M., and R. Salz, "A Universally 4122 Unique IDentifier (UUID) URN Namespace", RFC 4122, 4123 DOI 10.17487/RFC4122, July 2005, 4124 . 4126 [RFC4251] Ylonen, T. and C. Lonvick, Ed., "The Secure Shell (SSH) 4127 Protocol Architecture", RFC 4251, DOI 10.17487/RFC4251, 4128 January 2006, . 4130 [RFC4271] Rekhter, Y., Ed., Li, T., Ed., and S. Hares, Ed., "A 4131 Border Gateway Protocol 4 (BGP-4)", RFC 4271, 4132 DOI 10.17487/RFC4271, January 2006, 4133 . 4135 [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing 4136 Architecture", RFC 4291, DOI 10.17487/RFC4291, February 4137 2006, . 4139 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 4140 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 4141 December 2005, . 4143 [RFC4389] Thaler, D., Talwar, M., and C. Patel, "Neighbor Discovery 4144 Proxies (ND Proxy)", RFC 4389, DOI 10.17487/RFC4389, April 4145 2006, . 4147 [RFC4443] Conta, A., Deering, S., and M. Gupta, Ed., "Internet 4148 Control Message Protocol (ICMPv6) for the Internet 4149 Protocol Version 6 (IPv6) Specification", STD 89, 4150 RFC 4443, DOI 10.17487/RFC4443, March 2006, 4151 . 4153 [RFC4511] Sermersheim, J., Ed., "Lightweight Directory Access 4154 Protocol (LDAP): The Protocol", RFC 4511, 4155 DOI 10.17487/RFC4511, June 2006, 4156 . 4158 [RFC4541] Christensen, M., Kimball, K., and F. Solensky, 4159 "Considerations for Internet Group Management Protocol 4160 (IGMP) and Multicast Listener Discovery (MLD) Snooping 4161 Switches", RFC 4541, DOI 10.17487/RFC4541, May 2006, 4162 . 4164 [RFC4605] Fenner, B., He, H., Haberman, B., and H. Sandick, 4165 "Internet Group Management Protocol (IGMP) / Multicast 4166 Listener Discovery (MLD)-Based Multicast Forwarding 4167 ("IGMP/MLD Proxying")", RFC 4605, DOI 10.17487/RFC4605, 4168 August 2006, . 4170 [RFC4982] Bagnulo, M. and J. Arkko, "Support for Multiple Hash 4171 Algorithms in Cryptographically Generated Addresses 4172 (CGAs)", RFC 4982, DOI 10.17487/RFC4982, July 2007, 4173 . 4175 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 4176 "Bidirectional Protocol Independent Multicast (BIDIR- 4177 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 4178 . 4180 [RFC5214] Templin, F., Gleeson, T., and D. Thaler, "Intra-Site 4181 Automatic Tunnel Addressing Protocol (ISATAP)", RFC 5214, 4182 DOI 10.17487/RFC5214, March 2008, 4183 . 4185 [RFC5320] Templin, F., Ed., "The Subnetwork Encapsulation and 4186 Adaptation Layer (SEAL)", RFC 5320, DOI 10.17487/RFC5320, 4187 February 2010, . 4189 [RFC5522] Eddy, W., Ivancic, W., and T. Davis, "Network Mobility 4190 Route Optimization Requirements for Operational Use in 4191 Aeronautics and Space Exploration Mobile Networks", 4192 RFC 5522, DOI 10.17487/RFC5522, October 2009, 4193 . 4195 [RFC5558] Templin, F., Ed., "Virtual Enterprise Traversal (VET)", 4196 RFC 5558, DOI 10.17487/RFC5558, February 2010, 4197 . 4199 [RFC5569] Despres, R., "IPv6 Rapid Deployment on IPv4 4200 Infrastructures (6rd)", RFC 5569, DOI 10.17487/RFC5569, 4201 January 2010, . 4203 [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection 4204 (BFD)", RFC 5880, DOI 10.17487/RFC5880, June 2010, 4205 . 4207 [RFC6106] Jeong, J., Park, S., Beloeil, L., and S. Madanapalli, 4208 "IPv6 Router Advertisement Options for DNS Configuration", 4209 RFC 6106, DOI 10.17487/RFC6106, November 2010, 4210 . 4212 [RFC6139] Russert, S., Ed., Fleischman, E., Ed., and F. Templin, 4213 Ed., "Routing and Addressing in Networks with Global 4214 Enterprise Recursion (RANGER) Scenarios", RFC 6139, 4215 DOI 10.17487/RFC6139, February 2011, 4216 . 4218 [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful 4219 NAT64: Network Address and Protocol Translation from IPv6 4220 Clients to IPv4 Servers", RFC 6146, DOI 10.17487/RFC6146, 4221 April 2011, . 4223 [RFC6179] Templin, F., Ed., "The Internet Routing Overlay Network 4224 (IRON)", RFC 6179, DOI 10.17487/RFC6179, March 2011, 4225 . 4227 [RFC6221] Miles, D., Ed., Ooghe, S., Dec, W., Krishnan, S., and A. 4228 Kavanagh, "Lightweight DHCPv6 Relay Agent", RFC 6221, 4229 DOI 10.17487/RFC6221, May 2011, 4230 . 4232 [RFC6273] Kukec, A., Krishnan, S., and S. Jiang, "The Secure 4233 Neighbor Discovery (SEND) Hash Threat Analysis", RFC 6273, 4234 DOI 10.17487/RFC6273, June 2011, 4235 . 4237 [RFC6347] Rescorla, E. and N. Modadugu, "Datagram Transport Layer 4238 Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347, 4239 January 2012, . 4241 [RFC6355] Narten, T. and J. Johnson, "Definition of the UUID-Based 4242 DHCPv6 Unique Identifier (DUID-UUID)", RFC 6355, 4243 DOI 10.17487/RFC6355, August 2011, 4244 . 4246 [RFC6438] Carpenter, B. and S. Amante, "Using the IPv6 Flow Label 4247 for Equal Cost Multipath Routing and Link Aggregation in 4248 Tunnels", RFC 6438, DOI 10.17487/RFC6438, November 2011, 4249 . 4251 [RFC6706] Templin, F., Ed., "Asymmetric Extended Route Optimization 4252 (AERO)", RFC 6706, DOI 10.17487/RFC6706, August 2012, 4253 . 4255 [RFC6935] Eubanks, M., Chimento, P., and M. Westerlund, "IPv6 and 4256 UDP Checksums for Tunneled Packets", RFC 6935, 4257 DOI 10.17487/RFC6935, April 2013, 4258 . 4260 [RFC6936] Fairhurst, G. and M. Westerlund, "Applicability Statement 4261 for the Use of IPv6 UDP Datagrams with Zero Checksums", 4262 RFC 6936, DOI 10.17487/RFC6936, April 2013, 4263 . 4265 [RFC7333] Chan, H., Ed., Liu, D., Seite, P., Yokota, H., and J. 4266 Korhonen, "Requirements for Distributed Mobility 4267 Management", RFC 7333, DOI 10.17487/RFC7333, August 2014, 4268 . 4270 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 4271 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 4272 Multicast - Sparse Mode (PIM-SM): Protocol Specification 4273 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 4274 2016, . 4276 [RFC8402] Filsfils, C., Ed., Previdi, S., Ed., Ginsberg, L., 4277 Decraene, B., Litkowski, S., and R. Shakir, "Segment 4278 Routing Architecture", RFC 8402, DOI 10.17487/RFC8402, 4279 July 2018, . 4281 [RFC8446] Rescorla, E., "The Transport Layer Security (TLS) Protocol 4282 Version 1.3", RFC 8446, DOI 10.17487/RFC8446, August 2018, 4283 . 4285 [RFC8754] Filsfils, C., Ed., Dukes, D., Ed., Previdi, S., Leddy, J., 4286 Matsushima, S., and D. Voyer, "IPv6 Segment Routing Header 4287 (SRH)", RFC 8754, DOI 10.17487/RFC8754, March 2020, 4288 . 4290 [WG] Wireguard, "WireGuard, https://www.wireguard.com", August 4291 2020. 4293 Appendix A. Non-Normative Considerations 4295 AERO can be applied to a multitude of Internetworking scenarios, with 4296 each having its own adaptations. The following considerations are 4297 provided as non-normative guidance: 4299 A.1. Implementation Strategies for Route Optimization 4301 Route optimization as discussed in Section 3.14 results in the route 4302 optimization source (ROS) creating a NCE for the target neighbor. 4303 The NCE state is set to REACHABLE for at most ReachableTime seconds. 4304 In order to refresh the NCE lifetime before the ReachableTime timer 4305 expires, the specification requires implementations to issue a new 4306 NS/NA exchange to reset ReachableTime while data packets are still 4307 flowing. However, the decision of when to initiate a new NS/NA 4308 exchange and to perpetuate the process is left as an implementation 4309 detail. 4311 One possible strategy may be to monitor the NCE watching for data 4312 packets for (ReachableTime - 5) seconds. If any data packets have 4313 been sent to the neighbor within this timeframe, then send an NS to 4314 receive a new NA. If no data packets have been sent, wait for 5 4315 additional seconds and send an immediate NS if any data packets are 4316 sent within this "expiration pending" 5 second window. If no 4317 additional data packets are sent within the 5 second window, reset 4318 the NCE state to STALE. 4320 The monitoring of the neighbor data packet traffic therefore becomes 4321 an ongoing process during the NCE lifetime. If the NCE expires, 4322 future data packets will trigger a new NS/NA exchange while the 4323 packets themselves are delivered over a longer path until route 4324 optimization state is re-established. 4326 A.2. Implicit Mobility Management 4328 OMNI interface neighbors MAY provide a configuration option that 4329 allows them to perform implicit mobility management in which no ND 4330 messaging is used. In that case, the Client only transmits packets 4331 over a single interface at a time, and the neighbor always observes 4332 packets arriving from the Client from the same link-layer source 4333 address. 4335 If the Client's underlying interface address changes (either due to a 4336 readdressing of the original interface or switching to a new 4337 interface) the neighbor immediately updates the NCE for the Client 4338 and begins accepting and sending packets according to the Client's 4339 new address. This implicit mobility method applies to use cases such 4340 as cellphones with both WiFi and Cellular interfaces where only one 4341 of the interfaces is active at a given time, and the Client 4342 automatically switches over to the backup interface if the primary 4343 interface fails. 4345 A.3. Direct Underlying Interfaces 4347 When a Client's OMNI interface is configured over a Direct interface, 4348 the neighbor at the other end of the Direct link can receive packets 4349 without any encapsulation. In that case, the Client sends packets 4350 over the Direct link according to traffic selectors. If the Direct 4351 interface is selected, then the Client's IP packets are transmitted 4352 directly to the peer without going through an ANET/INET. If other 4353 interfaces are selected, then the Client's IP packets are transmitted 4354 via a different interface, which may result in the inclusion of 4355 Proxy/Servers and Bridges in the communications path. Direct 4356 interfaces must be tested periodically for reachability, e.g., via 4357 NUD. 4359 A.4. AERO Critical Infrastructure Considerations 4361 AERO Bridges can be either Commercial off-the Shelf (COTS) standard 4362 IP routers or virtual machines in the cloud. Bridges must be 4363 provisioned, supported and managed by the INET administrative 4364 authority, and connected to the Bridges of other INETs via inter- 4365 domain peerings. Cost for purchasing, configuring and managing 4366 Bridges is nominal even for very large OMNI links. 4368 AERO cloud Proxy/Servers can be standard dedicated server platforms, 4369 but most often will be deployed as virtual machines in the cloud. 4370 The only requirements for cloud Proxy/Servers are that they can run 4371 the AERO user-level code and have at least one network interface 4372 connection to the INET. Cloud Proxy/Servers must be provisioned, 4373 supported and managed by the INET administrative authority. Cost for 4374 purchasing, configuring and managing cloud Proxy/Servers is nominal 4375 especially for virtual machines. 4377 AERO ANET Proxy/Servers are most often standard dedicated server 4378 platforms with one underlying interface connected to the ANET and a 4379 second interface connected to an INET. As with cloud Proxy/Servers, 4380 the only requirements are that they can run the AERO user-level code 4381 and have at least one interface connection to the INET. ANET Proxy/ 4382 Servers must be provisioned, supported and managed by the ANET 4383 administrative authority. Cost for purchasing, configuring and 4384 managing Proxys is nominal, and borne by the ANET administrative 4385 authority. 4387 AERO Relays are simply Proxy/Servers connected to INETs and/or EUNs 4388 that provide forwarding services for non-MNP destinations. The Relay 4389 connects to the OMNI link and engages in eBGP peering with one or 4390 more Bridges as a stub AS. The Relay then injects its MNPs and/or 4391 non-MNP prefixes into the BGP routing system, and provisions the 4392 prefixes to its downstream-attached networks. The Relay can perform 4393 ROS/ROR services the same as for any Proxy/Server, and can route 4394 between the MNP and non-MNP address spaces. 4396 A.5. AERO Server Failure Implications 4398 AERO Proxy/Servers may appear as a single point of failure in the 4399 architecture, but such is not the case since all Proxy/Servers on the 4400 link provide identical services and loss of a Proxy/Server does not 4401 imply immediate and/or comprehensive communication failures. Proxy/ 4402 Server failure is quickly detected and conveyed by Bidirectional 4403 Forward Detection (BFD) and/or proactive NUD allowing Clients to 4404 migrate to new Proxy/Servers. 4406 If a Proxy/Server fails, ongoing packet forwarding to Clients will 4407 continue by virtue of the neighbor cache entries that have already 4408 been established in route optimization sources (ROSs). If a Client 4409 also experiences mobility events at roughly the same time the Proxy/ 4410 Server fails, unsolicited NA messages may be lost but neighbor cache 4411 entries in the DEPARTED state will ensure that packet forwarding to 4412 the Client's new locations will continue for up to DepartTime 4413 seconds. 4415 If a Client is left without a Proxy/Server for a considerable length 4416 of time (e.g., greater than ReachableTime seconds) then existing 4417 neighbor cache entries will eventually expire and both ongoing and 4418 new communications will fail. The original source will continue to 4419 retransmit until the Client has established a new Proxy/Server 4420 relationship, after which time continuous communications will resume. 4422 Therefore, providing many Proxy/Servers on the link with high 4423 availability profiles provides resilience against loss of individual 4424 Proxy/Servers and assurance that Clients can establish new Proxy/ 4425 Server relationships quickly in event of a Proxy/Server failure. 4427 A.6. AERO Client / Server Architecture 4429 The AERO architectural model is client / server in the control plane, 4430 with route optimization in the data plane. The same as for common 4431 Internet services, the AERO Client discovers the addresses of AERO 4432 Proxy/Servers and connects to one or more of them. The AERO service 4433 is analogous to common Internet services such as google.com, 4434 yahoo.com, cnn.com, etc. However, there is only one AERO service for 4435 the link and all Proxy/Servers provide identical services. 4437 Common Internet services provide differing strategies for advertising 4438 server addresses to clients. The strategy is conveyed through the 4439 DNS resource records returned in response to name resolution queries. 4440 As of January 2020 Internet-based 'nslookup' services were used to 4441 determine the following: 4443 o When a client resolves the domainname "google.com", the DNS always 4444 returns one A record (i.e., an IPv4 address) and one AAAA record 4445 (i.e., an IPv6 address). The client receives the same addresses 4446 each time it resolves the domainname via the same DNS resolver, 4447 but may receive different addresses when it resolves the 4448 domainname via different DNS resolvers. But, in each case, 4449 exactly one A and one AAAA record are returned. 4451 o When a client resolves the domainname "ietf.org", the DNS always 4452 returns one A record and one AAAA record with the same addresses 4453 regardless of which DNS resolver is used. 4455 o When a client resolves the domainname "yahoo.com", the DNS always 4456 returns a list of 4 A records and 4 AAAA records. Each time the 4457 client resolves the domainname via the same DNS resolver, the same 4458 list of addresses are returned but in randomized order (i.e., 4459 consistent with a DNS round-robin strategy). But, interestingly, 4460 the same addresses are returned (albeit in randomized order) when 4461 the domainname is resolved via different DNS resolvers. 4463 o When a client resolves the domainname "amazon.com", the DNS always 4464 returns a list of 3 A records and no AAAA records. As with 4465 "yahoo.com", the same three A records are returned from any 4466 worldwide Internet connection point in randomized order. 4468 The above example strategies show differing approaches to Internet 4469 resilience and service distribution offered by major Internet 4470 services. The Google approach exposes only a single IPv4 and a 4471 single IPv6 address to clients. Clients can then select whichever IP 4472 protocol version offers the best response, but will always use the 4473 same IP address according to the current Internet connection point. 4474 This means that the IP address offered by the network must lead to a 4475 highly-available server and/or service distribution point. In other 4476 words, resilience is predicated on high availability within the 4477 network and with no client-initiated failovers expected (i.e., it is 4478 all-or-nothing from the client's perspective). However, Google does 4479 provide for worldwide distributed service distribution by virtue of 4480 the fact that each Internet connection point responds with a 4481 different IPv6 and IPv4 address. The IETF approach is like google 4482 (all-or-nothing from the client's perspective), but provides only a 4483 single IPv4 or IPv6 address on a worldwide basis. This means that 4484 the addresses must be made highly-available at the network level with 4485 no client failover possibility, and if there is any worldwide service 4486 distribution it would need to be conducted by a network element that 4487 is reached via the IP address acting as a service distribution point. 4489 In contrast to the Google and IETF philosophies, Yahoo and Amazon 4490 both provide clients with a (short) list of IP addresses with Yahoo 4491 providing both IP protocol versions and Amazon as IPv4-only. The 4492 order of the list is randomized with each name service query 4493 response, with the effect of round-robin load balancing for service 4494 distribution. With a short list of addresses, there is still 4495 expectation that the network will implement high availability for 4496 each address but in case any single address fails the client can 4497 switch over to using a different address. The balance then becomes 4498 one of function in the network vs function in the end system. 4500 The same implications observed for common highly-available services 4501 in the Internet apply also to the AERO client/server architecture. 4502 When an AERO Client connects to one or more ANETs, it discovers one 4503 or more AERO Proxy/Server addresses through the mechanisms discussed 4504 in earlier sections. Each Proxy/Server address presumably leads to a 4505 fault-tolerant clustering arrangement such as supported by Linux-HA, 4506 Extended Virtual Synchrony or Paxos. Such an arrangement has 4507 precedence in common Internet service deployments in lightweight 4508 virtual machines without requiring expensive hardware deployment. 4509 Similarly, common Internet service deployments set service IP 4510 addresses on service distribution points that may relay requests to 4511 many different servers. 4513 For AERO, the expectation is that a combination of the Google/IETF 4514 and Yahoo/Amazon philosophies would be employed. The AERO Client 4515 connects to different ANET access points and can receive 1-2 Proxy/ 4516 Server ADM-LLAs at each point. It then selects one AERO Proxy/Server 4517 address, and engages in RS/RA exchanges with the same Proxy/Server 4518 from all ANET connections. The Client remains with this Proxy/Server 4519 unless or until the Proxy/Server fails, in which case it can switch 4520 over to an alternate Proxy/Server. The Client can likewise switch 4521 over to a different Proxy/Server at any time if there is some reason 4522 for it to do so. So, the AERO expectation is for a balance of 4523 function in the network and end system, with fault tolerance and 4524 resilience at both levels. 4526 Appendix B. Change Log 4528 << RFC Editor - remove prior to publication >> 4530 Changes from draft-templin-6man-aero-12 to draft-templin-6man-aero- 4531 13: 4533 o Final editorial review pass resulting in multiple changes. 4534 Document now submit for final approval (with reference to rfcdiff 4535 from previous version). 4537 Changes from draft-templin-6man-aero-11 to draft-templin-6man-aero- 4538 12: 4540 o Final editorial review pass resulting in multiple changes. 4541 Document now submit for final approval (with reference to rfcdiff 4542 from previous version). 4544 Changes from draft-templin-6man-aero-10 to draft-templin-6man-aero- 4545 11: 4547 o Final editorial review pass resulting in multiple changes. 4548 Document now submit for final approval (with reference to rfcdiff 4549 from previous version). 4551 Changes from draft-templin-6man-aero-09 to draft-templin-6man-aero- 4552 10: 4554 o Final editorial review pass resulting in multiple changes. 4555 Document now submit for final approval (with reference to rfcdiff 4556 from previous version). 4558 Changes from draft-templin-6man-aero-08 to draft-templin-6man-aero- 4559 09: 4561 o Final editorial review pass resulting in multiple changes. 4562 Document now submit for final approval (with reference to rfcdiff 4563 from previous version). 4565 Changes from draft-templin-6man-aero-07 to draft-templin-6man-aero- 4566 08: 4568 o Final editorial review pass resulting in multiple changes. 4569 Document now submit for final approval (with reference to rfcdiff 4570 from previous version). 4572 Changes from draft-templin-6man-aero-06 to draft-templin-6man-aero- 4573 07: 4575 o Final editorial review pass resulting in multiple changes. 4576 Document now submit for final approval (with reference to rfcdiff 4577 from previous version). 4579 Changes from draft-templin-6man-aero-05 to draft-templin-6man-aero- 4580 06: 4582 o Final editorial review pass resulting in multiple changes. 4583 Document now submit for final approval. 4585 Changes from draft-templin-6man-aero-04 to draft-templin-6man-aero- 4586 05: 4588 o Changed to use traffic selectors instead of the former multilink 4589 selection strategy. 4591 Changes from draft-templin-6man-aero-03 to draft-templin-6man-aero- 4592 04: 4594 o Removed documents from "Obsoletes" list. 4596 o Introduced the concept of "secured" and "unsecured" spanning tree. 4598 o Additional security considerations. 4600 o Additional route optimization considerations. 4602 Changes from draft-templin-6man-aero-02 to draft-templin-6man-aero- 4603 03: 4605 o Support for extended route optimization from ROR to target over 4606 target's underlying interfaces. 4608 Changes from draft-templin-6man-aero-01 to draft-templin-6man-aero- 4609 02: 4611 o Changed reference citations to "draft-templin-6man-omni". 4613 o Several important updates to IPv6 ND cache states and route 4614 optimization message addressing. 4616 o Included introductory description of the "6M's". 4618 o Updated Multicast specification. 4620 Changes from draft-templin-6man-aero-00 to draft-templin-6man-aero- 4621 01: 4623 o Changed category to "Informational". 4625 o Updated implementation status. 4627 Changes from earlier versions to draft-templin-6man-aero-00: 4629 o Established working baseline reference. 4631 Author's Address 4633 Fred L. Templin (editor) 4634 Boeing Research & Technology 4635 P.O. Box 3707 4636 Seattle, WA 98124 4637 USA 4639 Email: fltemplin@acm.org