idnits 2.17.1 draft-ietf-bess-dci-evpn-overlay-08.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 : ---------------------------------------------------------------------------- ** There are 16 instances of lines with control characters in the document. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (February 8, 2018) is 2267 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) ** Downref: Normative reference to an Informational RFC: RFC 7041 == Outdated reference: A later version (-22) exists of draft-ietf-idr-tunnel-encaps-08 == Outdated reference: A later version (-12) exists of draft-ietf-bess-evpn-overlay-11 Summary: 2 errors (**), 0 flaws (~~), 3 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 BESS Workgroup J. Rabadan (Ed.) 3 Internet Draft S. Sathappan 4 Intended status: Standards Track W. Henderickx 5 Nokia 7 A. Sajassi 8 Cisco 10 J. Drake 11 Juniper 13 Expires: August 12, 2018 February 8, 2018 15 Interconnect Solution for EVPN Overlay networks 16 draft-ietf-bess-dci-evpn-overlay-08 18 Abstract 20 This document describes how Network Virtualization Overlays (NVO) can 21 be connected to a Wide Area Network (WAN) in order to extend the 22 layer-2 connectivity required for some tenants. The solution analyzes 23 the interaction between NVO networks running Ethernet Virtual Private 24 Networks (EVPN) and other L2VPN technologies used in the WAN, such as 25 Virtual Private LAN Services (VPLS), VPLS extensions for Provider 26 Backbone Bridging (PBB-VPLS), EVPN or PBB-EVPN. It also describes how 27 the existing Technical Specifications apply to the Interconnection 28 and extends the EVPN procedures needed in some cases. In particular, 29 this document describes how EVPN routes are processed on Gateways 30 (GWs) that interconnect EVPN-Overlay and EVPN-MPLS networks, as well 31 as the Interconnect Ethernet Segment (I-ES) to provide multi-homing, 32 and the use of the Unknown MAC route to avoid MAC scale issues on 33 Data Center Network Virtualization Edge (NVE) devices. 35 Status of this Memo 37 This Internet-Draft is submitted in full conformance with the 38 provisions of BCP 78 and BCP 79. 40 Internet-Drafts are working documents of the Internet Engineering 41 Task Force (IETF), its areas, and its working groups. Note that 42 other groups may also distribute working documents as Internet- 43 Drafts. 45 Internet-Drafts are draft documents valid for a maximum of six months 46 and may be updated, replaced, or obsoleted by other documents at any 47 time. It is inappropriate to use Internet-Drafts as reference 48 material or to cite them other than as "work in progress." 50 The list of current Internet-Drafts can be accessed at 51 http://www.ietf.org/ietf/1id-abstracts.txt 53 The list of Internet-Draft Shadow Directories can be accessed at 54 http://www.ietf.org/shadow.html 56 This Internet-Draft will expire on August 12, 2018. 58 Copyright Notice 60 Copyright (c) 2018 IETF Trust and the persons identified as the 61 document authors. All rights reserved. 63 This document is subject to BCP 78 and the IETF Trust's Legal 64 Provisions Relating to IETF Documents 65 (http://trustee.ietf.org/license-info) in effect on the date of 66 publication of this document. Please review these documents 67 carefully, as they describe your rights and restrictions with respect 68 to this document. Code Components extracted from this document must 69 include Simplified BSD License text as described in Section 4.e of 70 the Trust Legal Provisions and are provided without warranty as 71 described in the Simplified BSD License. 73 Table of Contents 75 1. Conventions and Terminology . . . . . . . . . . . . . . . . . . 3 76 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 77 3. Decoupled Interconnect solution for EVPN overlay networks . . . 6 78 3.1. Interconnect requirements . . . . . . . . . . . . . . . . . 6 79 3.2. VLAN-based hand-off . . . . . . . . . . . . . . . . . . . . 7 80 3.3. PW-based (Pseudowire-based) hand-off . . . . . . . . . . . 8 81 3.4. Multi-homing solution on the GWs . . . . . . . . . . . . . 8 82 3.5. Gateway Optimizations . . . . . . . . . . . . . . . . . . . 9 83 3.5.1. MAC Address Advertisement Control . . . . . . . . . . . 9 84 3.5.2. ARP/ND flooding control . . . . . . . . . . . . . . . . 9 85 3.5.3. Handling failures between GW and WAN Edge routers . . . 10 86 4. Integrated Interconnect solution for EVPN overlay networks . . 10 87 4.1. Interconnect requirements . . . . . . . . . . . . . . . . . 11 88 4.2. VPLS Interconnect for EVPN-Overlay networks . . . . . . . . 12 89 4.2.1. Control/Data Plane setup procedures on the GWs . . . . 12 90 4.2.2. Multi-homing procedures on the GWs . . . . . . . . . . 13 91 4.3. PBB-VPLS Interconnect for EVPN-Overlay networks . . . . . . 13 92 4.3.1. Control/Data Plane setup procedures on the GWs . . . . 13 93 4.3.2. Multi-homing procedures on the GWs . . . . . . . . . . 14 94 4.4. EVPN-MPLS Interconnect for EVPN-Overlay networks . . . . . 14 95 4.4.1. Control Plane setup procedures on the GWs . . . . . . . 14 96 4.4.2. Data Plane setup procedures on the GWs . . . . . . . . 16 97 4.4.3. Multi-homing procedure extensions on the GWs . . . . . 17 98 4.4.4. Impact on MAC Mobility procedures . . . . . . . . . . . 19 99 4.4.5. Gateway optimizations . . . . . . . . . . . . . . . . . 19 100 4.4.6. Benefits of the EVPN-MPLS Interconnect solution . . . . 20 101 4.5. PBB-EVPN Interconnect for EVPN-Overlay networks . . . . . . 21 102 4.5.1. Control/Data Plane setup procedures on the GWs . . . . 21 103 4.5.2. Multi-homing procedures on the GWs . . . . . . . . . . 21 104 4.5.3. Impact on MAC Mobility procedures . . . . . . . . . . . 22 105 4.5.4. Gateway optimizations . . . . . . . . . . . . . . . . . 22 106 4.6. EVPN-VXLAN Interconnect for EVPN-Overlay networks . . . . . 22 107 4.6.1. Globally unique VNIs in the Interconnect network . . . 23 108 4.6.2. Downstream assigned VNIs in the Interconnect network . 23 109 5. Security Considerations . . . . . . . . . . . . . . . . . . . . 24 110 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 25 111 7. References . . . . . . . . . . . . . . . . . . . . . . . . . . 25 112 7.1. Normative References . . . . . . . . . . . . . . . . . . . 25 113 7.2. Informative References . . . . . . . . . . . . . . . . . . 26 114 8. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . 27 115 9. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 27 116 10. Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . 27 118 1. Conventions and Terminology 120 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 121 "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and 122 "OPTIONAL" in this document are to be interpreted as described in BCP 123 14 [RFC2119] [RFC8174] when, and only when, they appear in all 124 capitals, as shown here. 126 AC: Attachment Circuit. 128 ARP: Address Resolution Protocol. 130 BUM: it refers to the Broadcast, Unknown unicast and Multicast 131 traffic. 133 CFM: Connectivity Fault Management. 135 DC and DCI: Data Center and Data Center Interconnect. 137 DC RR(s) and WAN RR(s): it refers to the Data Center and Wide Area 138 Network Route Reflectors, respectively. 140 DF and NDF: Designated Forwarder and Non-Designated Forwarder. 142 EVPN: Ethernet Virtual Private Network, as in [RFC7432]. 144 EVI: EVPN Instance. 146 EVPN Tunnel binding: it refers to a tunnel to a remote PE/NVE for a 147 given EVI. Ethernet packets in these bindings are encapsulated with 148 the Overlay or MPLS encapsulation and the EVPN label at the bottom of 149 the stack. 151 ES: Ethernet Segment. 153 ESI: Ethernet Segment Identifier. 155 GW: Gateway or Data Center Gateway. 157 I-ES and I-ESI: Interconnect Ethernet Segment and Interconnect 158 Ethernet Segment Identifier. An I-ES is defined on the GWs for multi- 159 homing to/from the WAN. 161 MAC-VRF: it refers to an EVI instance in a particular node. 163 MP2P and LSM tunnels: it refers to Multi-Point to Point and Label 164 Switched Multicast tunnels. 166 ND: Neighbor Discovery protocol. 168 NVE: Network Virtualization Edge. 170 NVGRE: Network Virtualization using Generic Routing Encapsulation. 172 NVO: refers to Network Virtualization Overlays. 174 OAM: Operations and Maintenance. 176 PBB: Provider Backbone Bridging. 178 PW: Pseudowire. 180 RD: Route-Distinguisher. 182 RT: Route-Target. 184 S/C-TAG: It refers to a combination of Service Tag and Customer Tag 185 in a 802.1Q frame. 187 TOR: Top-Of-Rack switch. 189 VNI/VSID: refers to VXLAN/NVGRE virtual identifiers. 191 VPLS: Virtual Private LAN Service. 193 VSI: Virtual Switch Instance or VPLS instance in a particular PE. 195 VXLAN: Virtual eXtensible LAN. 197 2. Introduction 199 [EVPN-Overlays] discusses the use of Ethernet Virtual Private 200 Networks (EVPN) [RFC7432] as the control plane for Network 201 Virtualization Overlays (NVO), where VXLAN [RFC7348], NVGRE [RFC7637] 202 or MPLS over GRE [RFC4023] can be used as possible data plane 203 encapsulation options. 205 While this model provides a scalable and efficient multi-tenant 206 solution within the Data Center, it might not be easily extended to 207 the Wide Area Network (WAN) in some cases due to the requirements and 208 existing deployed technologies. For instance, a Service Provider 209 might have an already deployed Virtual Private LAN Service (VPLS) 210 [RFC4761][RFC4762], VPLS extensions for Provider Backbone Bridging 211 (PBB-VPLS) [RFC7041], EVPN [RFC7432] or PBB-EVPN [RFC7623] network 212 that has to be used to interconnect Data Centers and WAN VPN users. A 213 Gateway (GW) function is required in these cases. [EVPN-Overlays] 214 refers to the architectures described in this document as "DCI using 215 GWs". 217 This document describes a Interconnect solution for EVPN overlay 218 networks, assuming that the NVO Gateway (GW) and the WAN Edge 219 functions can be decoupled in two separate systems or integrated into 220 the same system. The former option will be referred as "Decoupled 221 Interconnect solution" throughout the document, whereas the latter 222 one will be referred as "Integrated Interconnect solution". 224 The specified procedures are local to the redundant GWs connecting a 225 DC to the WAN. The document does not preclude any combination across 226 different DCs for the same tenant. For instance, a "Decoupled" 227 solution can be used in GW1 and GW2 (for DC1) and an "Integrated" 228 solution can be used in GW3 and GW4 (for DC2). 230 While the Gateways and WAN PEs use existing Technical Specifications 231 in some cases, the document also defines extensions to these 232 Technical Specifications so that the requirements of the 233 Interconnection can be met. In particular, the following EVPN 234 extensions are described: 236 o The Interconnect Ethernet Segment (I-ES). 238 o The use of the Unknown MAC route in a DCI scenario. 240 o The processing of EVPN routes on Gateways with MAC-VRFs connecting 241 EVPN-Overlay to EVPN-MPLS networks. 243 3. Decoupled Interconnect solution for EVPN overlay networks 245 This section describes the interconnect solution when the GW and WAN 246 Edge functions are implemented in different systems. Figure 1 depicts 247 the reference model described in this section. 249 +--+ 250 |CE| 251 +--+ 252 | 253 +----+ 254 +----| PE |----+ 255 +---------+ | +----+ | +---------+ 256 +----+ | +---+ +----+ +----+ +---+ | +----+ 257 |NVE1|--| | | |WAN | |WAN | | | |--|NVE3| 258 +----+ | |GW1|--|Edge| |Edge|--|GW3| | +----+ 259 | +---+ +----+ +----+ +---+ | 260 | NVO-1 | | WAN | | NVO-2 | 261 | +---+ +----+ +----+ +---+ | 262 | | | |WAN | |WAN | | | | 263 +----+ | |GW2|--|Edge| |Edge|--|GW4| | +----+ 264 |NVE2|--| +---+ +----+ +----+ +---+ |--|NVE4| 265 +----+ +---------+ | | +---------+ +----+ 266 +--------------+ 268 |<-EVPN-Overlay-->|<-VLAN->|<-WAN L2VPN->|<--PW-->|<--EVPN-Overlay->| 269 hand-off hand-off 271 Figure 1 Decoupled Interconnect model 273 The following section describes the interconnect requirements for 274 this model. 276 3.1. Interconnect requirements 278 The Decoupled Interconnect architecture is intended to be deployed in 279 networks where the EVPN-Overlay and WAN providers are different 280 entities and a clear demarcation is needed. This solution solves the 281 following requirements: 283 o A simple connectivity hand-off between the EVPN-Overlay network 284 provider and the WAN provider so that QoS and security enforcement 285 is easily accomplished. 287 o Independence of the Layer Two VPN (L2VPN) technology deployed in 288 the WAN. 290 o Multi-homing between GW and WAN Edge routers, including per-service 291 load balancing. Per-flow load balancing is not a strong requirement 292 since a deterministic path per service is usually required for an 293 easy QoS and security enforcement. 295 o Support of Ethernet OAM and Connectivity Fault Management (CFM) 296 [802.1AG][Y.1731] functions between the GW and the WAN Edge router 297 to detect individual AC failures. 299 o Support for the following optimizations at the GW: 300 + Flooding reduction of unknown unicast traffic sourced from the DC 301 Network Virtualization Edge devices (NVEs). 302 + Control of the WAN MAC addresses advertised to the DC. 303 + Address Resolution Protocol (ARP) and Neighbor Discovery (ND) 304 flooding control for the requests coming from the WAN. 306 3.2. VLAN-based hand-off 308 In this option, the hand-off between the GWs and the WAN Edge routers 309 is based on VLANs [802.1Q-2014]. This is illustrated in Figure 1 310 (between the GWs in NVO-1 and the WAN Edge routers). Each MAC-VRF in 311 the GW is connected to a different VSI/MAC-VRF instance in the WAN 312 Edge router by using a different C-TAG VLAN ID or a different 313 combination of S/C-TAG VLAN IDs that matches at both sides. 315 This option provides the best possible demarcation between the DC and 316 WAN providers and it does not require control plane interaction 317 between both providers. The disadvantage of this model is the 318 provisioning overhead since the service has to be mapped to a C-TAG 319 or S/C-TAG VLAN ID combination at both GW and WAN Edge routers. 321 In this model, the GW acts as a regular Network Virtualization Edge 322 (NVE) towards the DC. Its control plane, data plane procedures and 323 interactions are described in [EVPN-Overlays]. 325 The WAN Edge router acts as a (PBB-)VPLS or (PBB-)EVPN PE with 326 attachment circuits (ACs) to the GWs. Its functions are described in 327 [RFC4761], [RFC4762], [RFC6074] or [RFC7432], [RFC7623]. 329 3.3. PW-based (Pseudowire-based) hand-off 331 If MPLS between the GW and the WAN Edge router is an option, a PW- 332 based Interconnect solution can be deployed. In this option the 333 hand-off between both routers is based on FEC128-based PWs [RFC4762] 334 or FEC129-based PWs (for a greater level of network automation) 335 [RFC6074]. Note that this model still provides a clear demarcation 336 boundary between DC and WAN (since there is a single PW between each 337 MAC-VRF and peer VSI), and security/QoS policies may be applied on a 338 per PW basis. This model provides better scalability than a C-TAG 339 based hand-off and less provisioning overhead than a combined C/S-TAG 340 hand-off. The PW-based hand-off interconnect is illustrated in Figure 341 1 (between the NVO-2 GWs and the WAN Edge routers). 343 In this model, besides the usual MPLS procedures between GW and WAN 344 Edge router [RFC3031], the GW MUST support an interworking function 345 in each MAC-VRF that requires extension to the WAN: 347 o If a FEC128-based PW is used between the MAC-VRF (GW) and the VSI 348 (WAN Edge), the corresponding VCID MUST be provisioned on the MAC- 349 VRF and match the VCID used in the peer VSI at the WAN Edge router. 351 o If BGP Auto-discovery [RFC6074] and FEC129-based PWs are used 352 between the GW MAC-VRF and the WAN Edge VSI, the provisioning of 353 the VPLS-ID MUST be supported on the MAC-VRF and MUST match the 354 VPLS-ID used in the WAN Edge VSI. 356 3.4. Multi-homing solution on the GWs 358 EVPN single-active multi-homing, i.e. per-service load-balancing 359 multi-homing is required in this type of interconnect. 361 The GWs will be provisioned with a unique ES per WAN interconnect, 362 and the hand-off attachment circuits or PWs between the GW and the 363 WAN Edge router will be assigned an ESI for such ES. The ESI will be 364 administratively configured on the GWs according to the procedures in 365 [RFC7432]. This Interconnect ES will be referred as "I-ES" hereafter, 366 and its identifier will be referred as "I-ESI". [RFC7432] describes 367 different ESI Types. The use of Type 0 for the I-ESI is RECOMMENDED 368 in this document. 370 The solution (on the GWs) MUST follow the single-active multi-homing 371 procedures as described in [EVPN-Overlays] for the provisioned I-ESI, 372 i.e. Ethernet A-D routes per ES and per EVI will be advertised to the 373 DC NVEs for the multi-homing functions, ES routes will be advertised 374 so that ES discovery and Designated Forwarder (DF) procedures can be 375 followed. The MAC addresses learned (in the data plane) on the hand- 376 off links will be advertised with the I-ESI encoded in the ESI field. 378 3.5. Gateway Optimizations 380 The following GW features are optional and optimize the control plane 381 and data plane in the DC. 383 3.5.1. MAC Address Advertisement Control 385 The use of EVPN in NVO networks brings a significant number of 386 benefits as described in [EVPN-Overlays]. However, if multiple DCs 387 are interconnected into a single EVI, each DC will have to import all 388 of the MAC addresses from each of the other DCs. 390 Even if optimized BGP techniques like RT-constraint [RFC4684] are 391 used, the number of MAC addresses to advertise or withdraw (in case 392 of failure) by the GWs of a given DC could overwhelm the NVEs within 393 that DC, particularly when the NVEs reside in the hypervisors. 395 The solution specified in this document uses the 'Unknown MAC' route 396 which is advertised into a given DC by each of the DC's GWs. This 397 route is a regular EVPN MAC/IP Advertisement route in which the MAC 398 Address Length is set to 48, the MAC address is set to 399 00:00:00:00:00:00, the IP length is set to 0, and the ESI field is 400 set to the DC GW's I-ESI. 402 An NVE within that DC that understands and process the Unknown MAC 403 route will send unknown unicast frames to one of the DCs GWs, which 404 will then forward that packet to the correct egress PE. Note that, 405 because the ESI is set to the DC GW's I-ESI, all-active multi-homing 406 can be applied to unknown unicast MAC addresses. An NVE that does not 407 understand the Unknown MAC route will handle unknown unicast as 408 described in [RFC7432]. 410 This document proposes that local policy determines whether MAC 411 addresses and/or the Unknown MAC route are advertised into a given 412 DC. As an example, when all the DC MAC addresses are learned in the 413 control/management plane, it may be appropriate to advertise only the 414 Unknown MAC route. Advertising all the DC MAC addresses in the 415 control/management plane is usually the case when the NVEs reside in 416 hypervisors. Refer to [EVPN-Overlays] section 7. 418 3.5.2. ARP/ND flooding control 420 Another optimization mechanism, naturally provided by EVPN in the 421 GWs, is the Proxy ARP/ND function. The GWs should build a Proxy 422 ARP/ND cache table as per [RFC7432]. When the active GW receives an 423 ARP/ND request/solicitation coming from the WAN, the GW does a Proxy 424 ARP/ND table lookup and replies as long as the information is 425 available in its table. 427 This mechanism is especially recommended on the GWs, since it 428 protects the DC network from external ARP/ND-flooding storms. 430 3.5.3. Handling failures between GW and WAN Edge routers 432 Link/PE failures are handled on the GWs as specified in [RFC7432]. 433 The GW detecting the failure will withdraw the EVPN routes as per 434 [RFC7432]. 436 Individual AC/PW failures may be detected by OAM mechanisms. For 437 instance: 439 o If the Interconnect solution is based on a VLAN hand-off, Ethernet- 440 CFM [802.1AG][Y.1731] may be used to detect individual AC failures 441 on both, the GW and WAN Edge router. An individual AC failure will 442 trigger the withdrawal of the corresponding A-D per EVI route as 443 well as the MACs learned on that AC. 445 o If the Interconnect solution is based on a PW hand-off, the Label 446 Distribution Protocol (LDP) PW Status bits TLV [RFC6870] may be 447 used to detect individual PW failures on both, the GW and WAN Edge 448 router. 450 4. Integrated Interconnect solution for EVPN overlay networks 452 When the DC and the WAN are operated by the same administrative 453 entity, the Service Provider can decide to integrate the GW and WAN 454 Edge PE functions in the same router for obvious CAPEX and OPEX 455 saving reasons. This is illustrated in Figure 2. Note that this model 456 does not provide an explicit demarcation link between DC and WAN 457 anymore. 459 +--+ 460 |CE| 461 +--+ 462 | 463 +----+ 464 +----| PE |----+ 465 +---------+ | +----+ | +---------+ 466 +----+ | +---+ +---+ | +----+ 467 |NVE1|--| | | | | |--|NVE3| 468 +----+ | |GW1| |GW3| | +----+ 469 | +---+ +---+ | 470 | NVO-1 | WAN | NVO-2 | 471 | +---+ +---+ | 472 | | | | | | 473 +----+ | |GW2| |GW4| | +----+ 474 |NVE2|--| +---+ +---+ |--|NVE4| 475 +----+ +---------+ | | +---------+ +----+ 476 +--------------+ 478 |<--EVPN-Overlay--->|<-----VPLS--->|<---EVPN-Overlay-->| 479 |<--PBB-VPLS-->| 480 Interconnect -> |<-EVPN-MPLS-->| 481 options |<--EVPN-Ovl-->|* 482 |<--PBB-EVPN-->| 484 Figure 2 Integrated Interconnect model 486 * EVPN-Ovl stands for EVPN-Overlay (and it's an Interconnect option). 488 4.1. Interconnect requirements 490 The Integrated Interconnect solution meets the following 491 requirements: 493 o Control plane and data plane interworking between the EVPN-overlay 494 network and the L2VPN technology supported in the WAN, irrespective 495 of the technology choice, i.e. (PBB-)VPLS or (PBB-)EVPN, as 496 depicted in Figure 2. 498 o Multi-homing, including single-active multi-homing with per-service 499 load balancing or all-active multi-homing, i.e. per-flow load- 500 balancing, as long as the technology deployed in the WAN supports 501 it. 503 o Support for end-to-end MAC Mobility, Static MAC protection and 504 other procedures (e.g. proxy-arp) described in [RFC7432] as long as 505 EVPN-MPLS is the technology of choice in the WAN. 507 o Independent inclusive multicast trees in the WAN and in the DC. 508 That is, the inclusive multicast tree type defined in the WAN does 509 not need to be the same as in the DC. 511 4.2. VPLS Interconnect for EVPN-Overlay networks 513 4.2.1. Control/Data Plane setup procedures on the GWs 515 Regular MPLS tunnels and TLDP/BGP sessions will be setup to the WAN 516 PEs and RRs as per [RFC4761], [RFC4762], [RFC6074] and overlay 517 tunnels and EVPN will be setup as per [EVPN-Overlays]. Note that 518 different route-targets for the DC and for the WAN are normally 519 required. A single type-1 RD per service may be used. 521 In order to support multi-homing, the GWs will be provisioned with an 522 I-ESI (see section 3.4), that will be unique per interconnection. All 523 the [RFC7432] procedures are still followed for the I-ES, e.g. any 524 MAC address learned from the WAN will be advertised to the DC with 525 the I-ESI in the ESI field. 527 A MAC-VRF per EVI will be created in each GW. The MAC-VRF will have 528 two different types of tunnel bindings instantiated in two different 529 split-horizon-groups: 531 o VPLS PWs will be instantiated in the "WAN split-horizon-group". 533 o Overlay tunnel bindings (e.g. VXLAN, NVGRE) will be instantiated 534 in the "DC split-horizon-group". 536 Attachment circuits are also supported on the same MAC-VRF, but they 537 will not be part of any of the above split-horizon-groups. 539 Traffic received in a given split-horizon-group will never be 540 forwarded to a member of the same split-horizon-group. 542 As far as BUM flooding is concerned, a flooding list will be composed 543 of the sub-list created by the inclusive multicast routes and the 544 sub-list created for VPLS in the WAN. BUM frames received from a 545 local Attachment Circuit (AC) will be forwarded to the flooding list. 546 BUM frames received from the DC or the WAN will be forwarded to the 547 flooding list observing the split-horizon-group rule described above. 549 Note that the GWs are not allowed to have an EVPN binding and a PW to 550 the same far-end within the same MAC-VRF, so that loops and packet 551 duplication are avoided. In case a GW can successfully establish 552 both, an EVPN binding and a PW to the same far-end PE, the EVPN 553 binding will prevail and the PW will be brought operationally down. 555 The optimizations procedures described in section 3.5 can also be 556 applied to this model. 558 4.2.2. Multi-homing procedures on the GWs 560 This model supports single-active multi-homing on the GWs. All-active 561 multi-homing is not supported by VPLS, therefore it cannot be used on 562 the GWs. 564 In this case, for a given EVI, all the PWs in the WAN split-horizon- 565 group are assigned to I-ES. All the single-active multi-homing 566 procedures as described by [EVPN-Overlays] will be followed for the 567 I-ES. 569 The non-DF GW for the I-ES will block the transmission and reception 570 of all the PWs in the "WAN split-horizon-group" for BUM and unicast 571 traffic. 573 4.3. PBB-VPLS Interconnect for EVPN-Overlay networks 575 4.3.1. Control/Data Plane setup procedures on the GWs 577 In this case, there is no impact on the procedures described in 578 [RFC7041] for the B-component. However the I-component instances 579 become EVI instances with EVPN-Overlay bindings and potentially local 580 attachment circuits. A number of MAC-VRF instances can be multiplexed 581 into the same B-component instance. This option provides significant 582 savings in terms of PWs to be maintained in the WAN. 584 The I-ESI concept described in section 4.2.1 will also be used for 585 the PBB-VPLS-based Interconnect. 587 B-component PWs and I-component EVPN-overlay bindings established to 588 the same far-end will be compared. The following rules will be 589 observed: 591 o Attempts to setup a PW between the two GWs within the B- 592 component context will never be blocked. 594 o If a PW exists between two GWs for the B-component and an 595 attempt is made to setup an EVPN binding on an I-component linked 596 to that B-component, the EVPN binding will be kept operationally 597 down. Note that the BGP EVPN routes will still be valid but not 598 used. 600 o The EVPN binding will only be up and used as long as there is no 601 PW to the same far-end in the corresponding B-component. The EVPN 602 bindings in the I-components will be brought down before the PW in 603 the B-component is brought up. 605 The optimizations procedures described in section 3.5 can also be 606 applied to this Interconnect option. 608 4.3.2. Multi-homing procedures on the GWs 610 This model supports single-active multi-homing on the GWs. All-active 611 multi-homing is not supported by this scenario. 613 The single-active multi-homing procedures as described by [EVPN- 614 Overlays] will be followed for the I-ES for each EVI instance 615 connected to the B-component. Note that in this case, for a given 616 EVI, all the EVPN bindings in the I-component are assigned to the I- 617 ES. The non-DF GW for the I-ES will block the transmission and 618 reception of all the I-component EVPN bindings for BUM and unicast 619 traffic. When learning MACs from the WAN, the non-DF MUST NOT 620 advertise EVPN MAC/IP routes for those MACs. 622 4.4. EVPN-MPLS Interconnect for EVPN-Overlay networks 624 If EVPN for MPLS tunnels, EVPN-MPLS hereafter, is supported in the 625 WAN, an end-to-end EVPN solution can be deployed. The following 626 sections describe the proposed solution as well as the impact 627 required on the [RFC7432] procedures. 629 4.4.1. Control Plane setup procedures on the GWs 631 The GWs MUST establish separate BGP sessions for sending/receiving 632 EVPN routes to/from the DC and to/from the WAN. Normally each GW will 633 setup one BGP EVPN session to the DC RR (or two BGP EVPN sessions if 634 there are redundant DC RRs) and one session to the WAN RR (or two 635 sessions if there are redundant WAN RRs). 637 In order to facilitate separate BGP processes for DC and WAN, EVPN 638 routes sent to the WAN SHOULD carry a different route-distinguisher 639 (RD) than the EVPN routes sent to the DC. In addition, although 640 reusing the same value is possible, different route-targets are 641 expected to be handled for the same EVI in the WAN and the DC. Note 642 that the EVPN service routes sent to the DC RRs will normally include 643 a [TUNNEL-ENCAP] BGP encapsulation extended community with a 644 different tunnel type than the one sent to the WAN RRs. 646 As in the other discussed options, an I-ES and its assigned I-ESI 647 will be configured on the GWs for multi-homing. This I-ES represents 648 the WAN EVPN-MPLS PEs to the DC but also the DC EVPN-Overlay NVEs to 649 the WAN. Optionally, different I-ESI values are configured for 650 representing the WAN and the DC. If different EVPN-Overlay networks 651 are connected to the same group of GWs, each EVPN-Overlay network 652 MUST get assigned a different I-ESI. 654 Received EVPN routes will never be reflected on the GWs but consumed 655 and re-advertised (if needed): 657 o Ethernet A-D routes, ES routes and Inclusive Multicast routes 658 are consumed by the GWs and processed locally for the 659 corresponding [RFC7432] procedures. 661 o MAC/IP advertisement routes will be received, imported and if 662 they become active in the MAC-VRF, the information will be re- 663 advertised as new routes with the following fields: 665 + The RD will be the GW's RD for the MAC-VRF. 667 + The ESI will be set to the I-ESI. 669 + The Ethernet-tag value will be kept from the received NLRI. 671 + The MAC length, MAC address, IP Length and IP address values 672 will be kept from the received NLRI. 674 + The MPLS label will be a local 20-bit value (when sent to the 675 WAN) or a DC-global 24-bit value (when sent to the DC for 676 encapsulations using a VNI). 678 + The appropriate Route-Targets (RTs) and [TUNNEL-ENCAP] BGP 679 Encapsulation extended community will be used according to 680 [EVPN-Overlays]. 682 The GWs will also generate the following local EVPN routes that will 683 be sent to the DC and WAN, with their corresponding RTs and [TUNNEL- 684 ENCAP] BGP Encapsulation extended community values: 686 o ES route(s) for the I-ESI(s). 688 o Ethernet A-D routes per ES and EVI for the I-ESI(s). The A-D 689 per-EVI routes sent to the WAN and the DC will have consistent 690 Ethernet-Tag values. 692 o Inclusive Multicast routes with independent tunnel type value 693 for the WAN and DC. E.g. a P2MP LSP may be used in the WAN 694 whereas ingress replication may be used in the DC. The routes 695 sent to the WAN and the DC will have a consistent Ethernet-Tag. 697 o MAC/IP advertisement routes for MAC addresses learned in local 698 attachment circuits. Note that these routes will not include the 699 I-ESI, but ESI=0 or different from 0 for local multi-homed 700 Ethernet Segments (ES). The routes sent to the WAN and the DC 701 will have a consistent Ethernet-Tag. 703 Assuming GW1 and GW2 are peer GWs of the same DC, each GW will 704 generate two sets of the above local service routes: Set-DC will be 705 sent to the DC RRs and will include A-D per EVI, Inclusive Multicast 706 and MAC/IP routes for the DC encapsulation and RT. Set-WAN will be 707 sent to the WAN RRs and will include the same routes but using the 708 WAN RT and encapsulation. GW1 and GW2 will receive each other's set- 709 DC and set-WAN. This is the expected behavior on GW1 and GW2 for 710 locally generated routes: 712 o Inclusive multicast routes: when setting up the flooding lists 713 for a given MAC-VRF, each GW will include its DC peer GW only in 714 the EVPN-MPLS flooding list (by default) and not the EVPN- 715 Overlay flooding list. That is, GW2 will import two Inclusive 716 Multicast routes from GW1 (from set-DC and set-WAN) but will 717 only consider one of the two, having the set-WAN route higher 718 priority. An administrative option MAY change this preference so 719 that the set-DC route is selected first. 721 o MAC/IP advertisement routes for local attachment circuits: as 722 above, the GW will select only one, having the route from the 723 set-WAN a higher priority. As with the Inclusive multicast 724 routes, an administrative option MAY change this priority. 726 4.4.2. Data Plane setup procedures on the GWs 728 The procedure explained at the end of the previous section will make 729 sure there are no loops or packet duplication between the GWs of the 730 same EVPN-Overlay network (for frames generated from local ACs) since 731 only one EVPN binding per EVI (or per Ethernet Tag in case of VLAN- 732 aware bundle services) will be setup in the data plane between the 733 two nodes. That binding will by default be added to the EVPN-MPLS 734 flooding list. 736 As for the rest of the EVPN tunnel bindings, they will be added to 737 one of the two flooding lists that each GW sets up for the same MAC- 738 VRF: 740 o EVPN-overlay flooding list (composed of bindings to the remote 741 NVEs or multicast tunnel to the NVEs). 743 o EVPN-MPLS flooding list (composed of MP2P or LSM tunnel to the 744 remote PEs) 746 Each flooding list will be part of a separate split-horizon-group: 747 the WAN split-horizon-group or the DC split-horizon-group. Traffic 748 generated from a local AC can be flooded to both 749 split-horizon-groups. Traffic from a binding of a split-horizon-group 750 can be flooded to the other split-horizon-group and local ACs, but 751 never to a member of its own split-horizon-group. 753 When either GW1 or GW2 receive a BUM frame on an MPLS tunnel 754 including an ESI label at the bottom of the stack, they will perform 755 an ESI label lookup and split-horizon filtering as per [RFC7432] in 756 case the ESI label identifies a local ESI (I-ESI or any other non- 757 zero ESI). 759 4.4.3. Multi-homing procedure extensions on the GWs 761 This model supports single-active as well as all-active multi-homing. 763 All the [RFC7432] multi-homing procedures for the DF election on I- 764 ES(s) as well as the backup-path (single-active) and aliasing (all- 765 active) procedures will be followed on the GWs. Remote PEs in the 766 EVPN-MPLS network will follow regular [RFC7432] aliasing or backup- 767 path procedures for MAC/IP routes received from the GWs for the same 768 I-ESI. So will NVEs in the EVPN-Overlay network for MAC/IP routes 769 received with the same I-ESI. 771 As far as the forwarding plane is concerned, by default, the EVPN- 772 Overlay network will have an analogous behavior to the access ACs in 773 [RFC7432] multi-homed Ethernet Segments. 775 The forwarding behavior on the GWs is described below: 777 o Single-active multi-homing; assuming a WAN split-horizon-group 778 (comprised of EVPN-MPLS bindings), a DC split-horizon-group 779 (comprised of EVPN-Overlay bindings) and local ACs on the GWs: 781 + Forwarding behavior on the non-DF: the non-DF MUST block 782 ingress and egress forwarding on the EVPN-Overlay bindings 783 associated to the I-ES. The EVPN-MPLS network is considered to 784 be the core network and the EVPN-MPLS bindings to the remote 785 PEs and GWs will be active. 787 + Forwarding behavior on the DF: the DF MUST NOT forward BUM or 788 unicast traffic received from a given split-horizon-group to a 789 member of his own split-horizon group. Forwarding to other 790 split-horizon-groups and local ACs is allowed (as long as the 791 ACs are not part of an ES for which the node is non-DF). As 792 per [RFC7432] and for split-horizon purposes, when receiving 793 BUM traffic on the EVPN-Overlay bindings associated to an I- 794 ES, the DF GW SHOULD add the I-ESI label when forwarding to 795 the peer GW over EVPN-MPLS. 797 + When receiving EVPN MAC/IP routes from the WAN, the non-DF 798 MUST NOT re-originate the EVPN routes and advertise them to 799 the DC peers. In the same way, EVPN MAC/IP routes received 800 from the DC MUST NOT be advertised to the WAN peers. This is 801 consistent with [RFC7432] and allows the remote PE/NVEs know 802 who the primary GW is, based on the reception of the MAC/IP 803 routes. 805 o All-active multi-homing; assuming a WAN split-horizon-group 806 (comprised of EVPN-MPLS bindings), a DC split-horizon-group 807 (comprised of EVPN-Overlay bindings) and local ACs on the GWs: 809 + Forwarding behavior on the non-DF: the non-DF follows the same 810 behavior as the non-DF in the single-active case but only for 811 BUM traffic. Unicast traffic received from a split-horizon- 812 group MUST NOT be forwarded to a member of its own split- 813 horizon-group but can be forwarded normally to the other 814 split-horizon-groups and local ACs. If a known unicast packet 815 is identified as a "flooded" packet, the procedures for BUM 816 traffic MUST be followed. 818 + Forwarding behavior on the DF: the DF follows the same 819 behavior as the DF in the single-active case but only for BUM 820 traffic. Unicast traffic received from a split-horizon-group 821 MUST NOT be forwarded to a member of its own split-horizon- 822 group but can be forwarded normally to the other split- 823 horizon-group and local ACs. If a known unicast packet is 824 identified as a "flooded" packet, the procedures for BUM 825 traffic MUST be followed. As per [RFC7432] and for split- 826 horizon purposes, when receiving BUM traffic on the EVPN- 827 Overlay bindings associated to an I-ES, the DF GW MUST add the 828 I-ESI label when forwarding to the peer GW over EVPN-MPLS. 830 + Contrary to the single-active multi-homing case, both DF and 831 non-DF re-originate and advertise MAC/IP routes received from 832 the WAN/DC peers, adding the corresponding I-ESI so that the 833 remote PE/NVEs can perform regular aliasing as per [RFC7432]. 835 The example in Figure 3 illustrates the forwarding of BUM traffic 836 originated from an NVE on a pair of all-active multi-homing GWs. 838 |<--EVPN-Overlay--->|<--EVPN-MPLS-->| 840 +---------+ +--------------+ 841 +----+ BUM +---+ | 842 |NVE1+----+----> | +-+-----+ | 843 +----+ | | DF |GW1| | | | 844 | | +-+-+ | | ++--+ 845 | | | | +--> |PE1| 846 | +--->X +-+-+ | ++--+ 847 | NDF| | | | 848 +----+ | |GW2<-+ | 849 |NVE2+--+ +-+-+ | 850 +----+ +--------+ | +------------+ 851 v 852 +--+ 853 |CE| 854 +--+ 856 Figure 3 Multi-homing BUM forwarding 858 GW2 is the non-DF for the I-ES and blocks the BUM forwarding. GW1 is 859 the DF and forwards the traffic to PE1 and GW2. Packets sent to GW2 860 will include the ESI-label for the I-ES. Based on the ESI-label, GW2 861 identifies the packets as I-ES-generated packets and will only 862 forward them to local ACs (CE in the example) and not back to the 863 EVPN-Overlay network. 865 4.4.4. Impact on MAC Mobility procedures 867 MAC Mobility procedures described in [RFC7432] are not modified by 868 this document. 870 Note that an intra-DC MAC move still leaves the MAC attached to the 871 same I-ES, so under the rules of [RFC7432] this is not considered a 872 MAC mobility event. Only when the MAC moves from the WAN domain to 873 the DC domain (or from one DC to another) the MAC will be learned 874 from a different ES and the MAC Mobility procedures will kick in. 876 The sticky bit indication in the MAC Mobility extended community MUST 877 be propagated between domains. 879 4.4.5. Gateway optimizations 881 All the Gateway optimizations described in section 3.5 MAY be applied 882 to the GWs when the Interconnect is based on EVPN-MPLS. 884 In particular, the use of the Unknown MAC route, as described in 885 section 3.5.1, solves some transient packet duplication issues in 886 cases of all-active multi-homing, as explained below. 888 Consider the diagram in Figure 2 for EVPN-MPLS Interconnect and all- 889 active multi-homing, and the following sequence: 891 a) MAC Address M1 is advertised from NVE3 in EVI-1. 893 b) GW3 and GW4 learn M1 for EVI-1 and re-advertise M1 to the WAN 894 with I-ESI-2 in the ESI field. 896 c) GW1 and GW2 learn M1 and install GW3/GW4 as next-hops following 897 the EVPN aliasing procedures. 899 d) Before NVE1 learns M1, a packet arrives at NVE1 with 900 destination M1. If the Unknown MAC route had not been 901 advertised into the DC, NVE1 would have flooded the packet 902 throughout the DC, in particular to both GW1 and GW2. If the 903 same VNI/VSID is used for both known unicast and BUM traffic, 904 as is typically the case, there is no indication in the packet 905 that it is a BUM packet and both GW1 and GW2 would have 906 forwarded it, creating packet duplication. However, because the 907 Unknown MAC route had been advertised into the DC, NVE1 will 908 unicast the packet to either GW1 or GW2. 910 e) Since both GW1 and GW2 know M1, the GW receiving the packet 911 will forward it to either GW3 or GW4. 913 4.4.6. Benefits of the EVPN-MPLS Interconnect solution 915 Besides retaining the EVPN attributes between Data Centers and 916 throughout the WAN, the EVPN-MPLS Interconnect solution on the GWs 917 has some benefits compared to pure BGP EVPN RR or Inter-AS model B 918 solutions without a gateway: 920 o The solution supports the connectivity of local attachment 921 circuits on the GWs. 923 o Different data plane encapsulations can be supported in the DC 924 and the WAN. 926 o Optimized multicast solution, with independent inclusive 927 multicast trees in DC and WAN. 929 o MPLS Label aggregation: for the case where MPLS labels are 930 signaled from the NVEs for MAC/IP Advertisement routes, this 931 solution provides label aggregation. A remote PE MAY receive a 932 single label per GW MAC-VRF as opposed to a label per NVE/MAC- 933 VRF connected to the GW MAC-VRF. For instance, in Figure 2, PE 934 would receive only one label for all the routes advertised for a 935 given MAC-VRF from GW1, as opposed to a label per NVE/MAC-VRF. 937 o The GW will not propagate MAC mobility for the MACs moving 938 within a DC. Mobility intra-DC is solved by all the NVEs in the 939 DC. The MAC Mobility procedures on the GWs are only required in 940 case of mobility across DCs. 942 o Proxy-ARP/ND function on the DC GWs can be leveraged to reduce 943 ARP/ND flooding in the DC or/and in the WAN. 945 4.5. PBB-EVPN Interconnect for EVPN-Overlay networks 947 PBB-EVPN [RFC7623] is yet another Interconnect option. It requires 948 the use of GWs where I-components and associated B-components are 949 part of EVI instances. 951 4.5.1. Control/Data Plane setup procedures on the GWs 953 EVPN will run independently in both components, the I-component MAC- 954 VRF and B-component MAC-VRF. Compared to [RFC7623], the DC C-MACs are 955 no longer learned in the data plane on the GW but in the control 956 plane through EVPN running on the I-component. Remote C-MACs coming 957 from remote PEs are still learned in the data plane. B-MACs in the B- 958 component will be assigned and advertised following the procedures 959 described in [RFC7623]. 961 An I-ES will be configured on the GWs for multi-homing, but its I-ESI 962 will only be used in the EVPN control plane for the I-component EVI. 963 No non-reserved ESIs will be used in the control plane of the B- 964 component EVI as per [RFC7623], that is, the I-ES will be represented 965 to the WAN PBB-EVPN PEs using shared or dedicated B-MACs. 967 The rest of the control plane procedures will follow [RFC7432] for 968 the I-component EVI and [RFC7623] for the B-component EVI. 970 From the data plane perspective, the I-component and B-component EVPN 971 bindings established to the same far-end will be compared and the I- 972 component EVPN-overlay binding will be kept down following the rules 973 described in section 4.3.1. 975 4.5.2. Multi-homing procedures on the GWs 977 This model supports single-active as well as all-active multi-homing. 979 The forwarding behavior of the DF and non-DF will be changed based on 980 the description outlined in section 4.4.3, only replacing the "WAN 981 split-horizon-group" for the B-component, and using [RFC7623] 982 procedures for the traffic sent or received on the B-component. 984 4.5.3. Impact on MAC Mobility procedures 986 C-MACs learned from the B-component will be advertised in EVPN within 987 the I-component EVI scope. If the C-MAC was previously known in the 988 I-component database, EVPN would advertise the C-MAC with a higher 989 sequence number, as per [RFC7432]. From a Mobility perspective and 990 the related procedures described in [RFC7432], the C-MACs learned 991 from the B-component are considered local. 993 4.5.4. Gateway optimizations 995 All the considerations explained in section 4.4.5 are applicable to 996 the PBB-EVPN Interconnect option. 998 4.6. EVPN-VXLAN Interconnect for EVPN-Overlay networks 1000 If EVPN for Overlay tunnels is supported in the WAN and a GW function 1001 is required, an end-to-end EVPN solution can be deployed. This 1002 section focuses on the specific case of EVPN for VXLAN (EVPN-VXLAN 1003 hereafter) and the impact on the [RFC7432] procedures. 1005 The procedures described in section 4.4 apply to this section too, 1006 only replacing EVPN-MPLS for EVPN-VXLAN control plane specifics and 1007 using [EVPN-Overlays] "Local Bias" procedures instead of section 1008 4.4.3. Since there are no ESI-labels in VXLAN, GWs need to rely on 1009 "Local Bias" to apply split-horizon on packets generated from the I- 1010 ES and sent to the peer GW. 1012 This use-case assumes that NVEs need to use the VNIs or VSIDs as a 1013 globally unique identifiers within a data center, and a Gateway needs 1014 to be employed at the edge of the data center network to translate 1015 the VNI or VSID when crossing the network boundaries. This GW 1016 function provides VNI and tunnel IP address translation. The use-case 1017 in which local downstream assigned VNIs or VSIDs can be used (like 1018 MPLS labels) is described by [EVPN-Overlays]. 1020 While VNIs are globally significant within each DC, there are two 1021 possibilities in the Interconnect network: 1023 a) Globally unique VNIs in the Interconnect network: 1024 In this case, the GWs and PEs in the Interconnect network will 1025 agree on a common VNI for a given EVI. The RT to be used in the 1026 Interconnect network can be auto-derived from the agreed 1027 Interconnect VNI. The VNI used inside each DC MAY be the same 1028 as the Interconnect VNI. 1030 b) Downstream assigned VNIs in the Interconnect network. 1031 In this case, the GWs and PEs MUST use the proper RTs to 1032 import/export the EVPN routes. Note that even if the VNI is 1033 downstream assigned in the Interconnect network, and unlike 1034 option B, it only identifies the pair and 1035 not the pair. The VNI used inside 1036 each DC MAY be the same as the Interconnect VNI. GWs SHOULD 1037 support multiple VNI spaces per EVI (one per Interconnect 1038 network they are connected to). 1040 In both options, NVEs inside a DC only have to be aware of a single 1041 VNI space, and only GWs will handle the complexity of managing 1042 multiple VNI spaces. In addition to VNI translation above, the GWs 1043 will provide translation of the tunnel source IP for the packets 1044 generated from the NVEs, using their own IP address. GWs will use 1045 that IP address as the BGP next-hop in all the EVPN updates to the 1046 Interconnect network. 1048 The following sections provide more details about these two options. 1050 4.6.1. Globally unique VNIs in the Interconnect network 1052 Considering Figure 2, if a host H1 in NVO-1 needs to communicate with 1053 a host H2 in NVO-2, and assuming that different VNIs are used in each 1054 DC for the same EVI, e.g. VNI-10 in NVO-1 and VNI-20 in NVO-2, then 1055 the VNIs MUST be translated to a common Interconnect VNI (e.g. VNI- 1056 100) on the GWs. Each GW is provisioned with a VNI translation 1057 mapping so that it can translate the VNI in the control plane when 1058 sending BGP EVPN route updates to the Interconnect network. In other 1059 words, GW1 and GW2 MUST be configured to map VNI-10 to VNI-100 in the 1060 BGP update messages for H1's MAC route. This mapping is also used to 1061 translate the VNI in the data plane in both directions, that is, VNI- 1062 10 to VNI-100 when the packet is received from NVO-1 and the reverse 1063 mapping from VNI-100 to VNI-10 when the packet is received from the 1064 remote NVO-2 network and needs to be forwarded to NVO-1. 1066 The procedures described in section 4.4 will be followed, considering 1067 that the VNIs advertised/received by the GWs will be translated 1068 accordingly. 1070 4.6.2. Downstream assigned VNIs in the Interconnect network 1072 In this case, if a host H1 in NVO-1 needs to communicate with a host 1073 H2 in NVO-2, and assuming that different VNIs are used in each DC for 1074 the same EVI, e.g. VNI-10 in NVO-1 and VNI-20 in NVO-2, then the VNIs 1075 MUST be translated as in section 4.6.1. However, in this case, there 1076 is no need to translate to a common Interconnect VNI on the GWs. Each 1077 GW can translate the VNI received in an EVPN update to a locally 1078 assigned VNI advertised to the Interconnect network. Each GW can use 1079 a different Interconnect VNI, hence this VNI does not need to be 1080 agreed on all the GWs and PEs of the Interconnect network. 1082 The procedures described in section 4.4 will be followed, taking the 1083 considerations above for the VNI translation. 1085 5. Security Considerations 1087 This document applies existing Technical Specifications to a number 1088 of Interconnect models. The Security Considerations included in those 1089 documents, such as [RFC7432], [EVPN-Overlays], [RFC7623], [RFC4761] 1090 and [RFC4762] apply to this document whenever those technologies are 1091 used. 1093 [EVPN-Overlays] discusses two main DCI solution groups: "DCI using 1094 GWs" and "DCI using ASBRs". This document specifies the solutions 1095 that correspond to the "DCI using GWs" group. It is important to note 1096 that use of GWs provide a superior level of security on a per tenant 1097 basis, compared to the use of ASBRs. This is due to the fact that GWs 1098 need to perform a MAC lookup on the frames being received from the 1099 WAN, and they apply security procedures, such as filtering of 1100 undesired frames, filtering of frames with a source MAC that matches 1101 a protected MAC in the DC or application of MAC duplication 1102 procedures defined in [RFC7432]. On ASBRs though, traffic is 1103 forwarded based on a label or VNI swap and there is usually no 1104 visibility of the encapsulated frames, which can carry malicious 1105 traffic. 1107 In addition, the GW optimizations specified in this document, provide 1108 additional protection of the DC Tenant Systems. For instance, the MAC 1109 address advertisement control and Unknown MAC route defined in 1110 section 3.5.1 protect the DC NVEs from being overwhelmed with an 1111 excessive number MAC/IP routes being learned on the GWs from the WAN. 1112 The ARP/ND flooding control described in 3.5.2 can reduce/suppress 1113 broadcast storms being injected from the WAN. 1115 Finally, the reader should be aware of the potential security 1116 implications of designing a DCI with the Decoupled Interconnect 1117 solution (section 2) or the Integrated Interconnect solution (section 1118 3). In the Decoupled Interconnect solution the DC is typically easier 1119 to protect from the WAN, since each GW has a single logical link to 1120 one WAN PE, whereas in the Integrated solution, the GW has logical 1121 links to all the WAN PEs that are attached to the tenant. In either 1122 model, proper control plane and data plane policies should be put in 1123 place in the GWs in order to protect the DC from potential attacks 1124 coming from the WAN. 1126 6. IANA Considerations 1128 This document has no IANA actions. 1130 7. References 1132 7.1. Normative References 1134 [RFC4761] Kompella, K., Ed., and Y. Rekhter, Ed., "Virtual Private 1135 LAN Service (VPLS) Using BGP for Auto-Discovery and Signaling", 1136 RFC 4761, DOI 10.17487/RFC4761, January 2007, . 1139 [RFC4762] Lasserre, M., Ed., and V. Kompella, Ed., "Virtual Private 1140 LAN Service (VPLS) Using Label Distribution Protocol (LDP) 1141 Signaling", RFC 4762, DOI 10.17487/RFC4762, January 2007, 1142 . 1144 [RFC6074] Rosen, E., Davie, B., Radoaca, V., and W. Luo, 1145 "Provisioning, Auto-Discovery, and Signaling in Layer 2 Virtual 1146 Private Networks (L2VPNs)", RFC 6074, DOI 10.17487/RFC6074, January 1147 2011, . 1149 [RFC7041] Balus, F., Ed., Sajassi, A., Ed., and N. Bitar, Ed., 1150 "Extensions to the Virtual Private LAN Service (VPLS) Provider Edge 1151 (PE) Model for Provider Backbone Bridging", RFC 7041, DOI 1152 10.17487/RFC7041, November 2013, . 1155 [RFC7432] Sajassi, A., Ed., Aggarwal, R., Bitar, N., Isaac, A., 1156 Uttaro, J., Drake, J., and W. Henderickx, "BGP MPLS-Based Ethernet 1157 VPN", RFC 7432, DOI 10.17487/RFC7432, February 2015, . 1160 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1161 Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1162 1997, . 1164 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 1165 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May 2017, 1166 . 1168 [TUNNEL-ENCAP] Rosen et al., "The BGP Tunnel Encapsulation 1169 Attribute", draft-ietf-idr-tunnel-encaps-08, work in progress, 1170 January 11, 2018. 1172 [RFC7623] Sajassi et al., "Provider Backbone Bridging Combined with 1173 Ethernet VPN (PBB-EVPN)", RFC 7623, September, 2015, . 1176 [EVPN-Overlays] Sajassi-Drake et al., "A Network Virtualization 1177 Overlay Solution using EVPN", draft-ietf-bess-evpn-overlay-11.txt, 1178 work in progress, January, 2018 1180 7.2. Informative References 1182 [RFC4684] Marques, P., Bonica, R., Fang, L., Martini, L., Raszuk, 1183 R., Patel, K., and J. Guichard, "Constrained Route Distribution for 1184 Border Gateway Protocol/MultiProtocol Label Switching (BGP/MPLS) 1185 Internet Protocol (IP) Virtual Private Networks (VPNs)", RFC 4684, 1186 DOI 10.17487/RFC4684, November 2006, . 1189 [RFC7348] Mahalingam, M., Dutt, D., Duda, K., Agarwal, P., Kreeger, 1190 L., Sridhar, T., Bursell, M., and C. Wright, "Virtual eXtensible 1191 Local Area Network (VXLAN): A Framework for Overlaying Virtualized 1192 Layer 2 Networks over Layer 3 Networks", RFC 7348, DOI 1193 10.17487/RFC7348, August 2014, . 1196 [RFC7637] Garg, P., et al., "NVGRE: Network Virtualization using 1197 Generic Routing Encapsulation", RFC 7637, September, 2015 1199 [RFC4023] Worster, T., Rekhter, Y., and E. Rosen, Ed., 1200 "Encapsulating MPLS in IP or Generic Routing Encapsulation (GRE)", 1201 RFC 4023, DOI 10.17487/RFC4023, March 2005, . 1204 [Y.1731] ITU-T Recommendation Y.1731, "OAM functions and mechanisms 1205 for Ethernet based networks", July 2011. 1207 [802.1AG] IEEE 802.1AG_2007, "IEEE Standard for Local and 1208 Metropolitan Area Networks - Virtual Bridged Local Area Networks 1209 Amendment 5: Connectivity Fault Management", January 2008. 1211 [802.1Q-2014] IEEE 802.1Q-2014, "IEEE Standard for Local and 1212 metropolitan area networks--Bridges and Bridged Networks", December 1213 2014. 1215 [RFC6870] Muley, P., Ed., and M. Aissaoui, Ed., "Pseudowire 1216 Preferential Forwarding Status Bit", RFC 6870, DOI 10.17487/RFC6870, 1217 February 2013, . 1219 [RFC3031] Rosen, E., Viswanathan, A., and R. Callon, "Multiprotocol 1220 Label Switching Architecture", RFC 3031, DOI 10.17487/RFC3031, 1221 January 2001, . 1223 8. Acknowledgments 1225 The authors would like to thank Neil Hart, Vinod Prabhu and Kiran 1226 Nagaraj for their valuable comments and feedback. We would also like 1227 to thank Martin Vigoureux and Alvaro Retana for his detailed review 1228 and comments. 1230 9. Contributors 1232 In addition to the authors listed on the front page, the following 1233 co-authors have also contributed to this document: 1235 Ravi Shekhar 1236 Anil Lohiya 1237 Wen Lin 1238 Juniper Networks 1240 Florin Balus 1241 Patrice Brissette 1242 Cisco 1244 Senad Palislamovic 1245 Nokia 1247 Dennis Cai 1248 Alibaba 1250 10. Authors' Addresses 1252 Jorge Rabadan 1253 Nokia 1254 777 E. Middlefield Road 1255 Mountain View, CA 94043 USA 1256 Email: jorge.rabadan@nokia.com 1258 Senthil Sathappan 1259 Nokia 1260 Email: senthil.sathappan@nokia.com 1262 Wim Henderickx 1263 Nokia 1264 Email: wim.henderickx@nokia.com 1266 Ali Sajassi 1267 Cisco 1268 Email: sajassi@cisco.com 1270 John Drake 1271 Juniper 1272 Email: jdrake@juniper.net