<?xmlversion='1.0' encoding='utf-8'?>version="1.0" encoding="UTF-8"?> <!DOCTYPE rfc SYSTEM"rfc2629.dtd" [ <!ENTITY RFC4761 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4761.xml"> <!ENTITY RFC4762 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4762.xml"> <!ENTITY RFC6074 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6074.xml"> <!ENTITY RFC7041 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7041.xml"> <!ENTITY RFC7432 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7432.xml"> <!ENTITY RFC2119 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"> <!ENTITY RFC8174 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"> <!ENTITY I-D.ietf-idr-tunnel-encaps SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I-D.ietf-idr-tunnel-encaps.xml"> <!ENTITY RFC7623 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7623.xml"> <!ENTITY I-D.ietf-bess-evpn-overlay SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I-D.ietf-bess-evpn-overlay.xml"> <!ENTITY RFC7543 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7543.xml"> <!ENTITY RFC4684 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4684.xml"> <!ENTITY RFC7348 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7348.xml"> <!ENTITY RFC7637 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7637.xml"> <!ENTITY RFC4023 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4023.xml"> <!ENTITY RFC6870 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6870.xml"> <!ENTITY RFC3031 SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3031.xml"> <!ENTITY I-D.sajassi-bess-evpn-virtual-eth-segment SYSTEM "https://xml2rfc.ietf.org/public/rfc/bibxml3/reference.I-D.sajassi-bess-evpn-virtual-eth-segment.xml"> ]>"rfc2629-xhtml.ent"> <rfc xmlns:xi="http://www.w3.org/2001/XInclude" submissionType="IETF"docName="draft-ietf-bess-dci-evpn-overlay-10"category="std"ipr="trust200902">consensus="true" docName="draft-ietf-bess-dci-evpn-overlay-10" number="9014" ipr="trust200902" obsoletes="" updates="" xml:lang="en" symRefs="true" sortRefs="true" tocInclude="true" version="3"> <!-- xml2rfc v2v3 conversion 2.45.2 --> <!-- Generated by id2xml 1.5.0 on 2020-05-27T19:59:24Z --><?rfc strict="yes"?> <?rfc compact="yes"?> <?rfc subcompact="no"?> <?rfc symrefs="yes"?> <?rfc sortrefs="no"?> <?rfc text-list-symbols="o++*-"?> <!-- draft-ietf-bess-dci-evpn-overlay-10-manual.txt(13): Warning: Expected an expires indication top left, found none --><?rfc toc="yes"?><front> <title abbrev="InterconnectSolutionfor EVPN-Overlays">Interconnect Solution forEVPNEthernet VPN (EVPN) Overlaynetworks</title>Networks</title> <seriesInfo name="RFC" value="9014"/> <author initials="J." surname="Rabadan" fullname="Jorge Rabadan" role="editor"> <organization>Nokia</organization> <address> <postal> <street>777 E. Middlefield Road</street> <city>Mountain View</city> <region>CA</region> <code>94043</code> <country>USA</country> </postal> <email>jorge.rabadan@nokia.com</email> </address> </author> <author initials="S." surname="Sathappan" fullname="Senthil Sathappan"> <organization>Nokia</organization> <address> <email>senthil.sathappan@nokia.com</email> </address> </author> <author initials="W." surname="Henderickx" fullname="Wim Henderickx"> <organization>Nokia</organization> <address> <email>wim.henderickx@nokia.com</email> </address> </author> <author initials="A." surname="Sajassi" fullname="Ali Sajassi"> <organization>Cisco</organization> <address> <email>sajassi@cisco.com</email> </address> </author> <author initials="J." surname="Drake" fullname="John Drake"> <organization>Juniper</organization> <address> <email>jdrake@juniper.net</email> </address> </author> <dateyear="2020"year="2021" month="May"/> <workgroup>BESS Workgroup</workgroup><abstract><t><abstract> <t> This document describes how Network Virtualization Overlays(NVO)(NVOs) can be connected to a Wide Area Network (WAN) in order to extend thelayer-2Layer 2 connectivity required for some tenants. The solution analyzes the interaction between NVO networks running Ethernet Virtual Private Networks(EVPN)(EVPNs) and otherL2VPNLayer 2 VPN (L2VPN) technologies used in the WAN, such as Virtual Private LAN Services(VPLS),(VPLSs), VPLS extensions for Provider Backbone Bridging (PBB-VPLS),EVPNEVPN, or PBB-EVPN. It also describes how the existing technical specifications apply to theInterconnectioninterconnection and extends the EVPN procedures needed in some cases. In particular, this document describes how EVPN routes are processed on Gateways (GWs) that interconnect EVPN-Overlay and EVPN-MPLS networks, as well as the Interconnect Ethernet Segment(I-ES)(I-ES), to providemulti-homing, andmultihoming. This document also describes the use of the Unknown MACrouteRoute (UMR) to avoidMAC scaleissues of a Media Access Control (MAC) scale on Data Center Network Virtualization Edge (NVE) devices.</t> </abstract> </front> <middle> <sectiontitle="Conventions and Terminology" anchor="sect-1"><t> The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they appear in all capitals, as shown here.</t>anchor="sect-2" numbered="true" toc="default"> <name>Introduction</name> <t>AC: Attachment Circuit.</t> <t> ARP: Address Resolution Protocol.</t> <t> BUM: refers to Broadcast, Unknown unicast and Multicast traffic.</t> <t> CE: Customer Equipment.</t> <t> CFM: Connectivity Fault Management.</t> <t> DC and DCI: Data Center and Data Center Interconnect.</t> <t> DC RR(s) and WAN RR(s): refers to the Data Center and Wide Area Network Route Reflectors, respectively.</t> <t> DF and NDF: Designated Forwarder and Non-Designated Forwarder.</t> <t> EVPN: Ethernet Virtual Private Network, as in<xreftarget="RFC7432"/>.</t> <t> EVI: EVPN Instance.</t> <t> EVPN Tunnel binding: refers to a tunnel to a remote PE/NVE for a given EVI. Ethernet packets in these bindings are encapsulated with the Overlay or MPLS encapsulation and the EVPN label at the bottom of the stack.</t> <t> ES and vES: Ethernet Segment and virtual Ethernet Segment.</t> <t> ESI: Ethernet Segment Identifier.</t> <t> GW: Gateway or Data Center Gateway.</t> <t> I-ES and I-ESI: Interconnect Ethernet Segment and Interconnect Ethernet Segment Identifier. An I-ES is defined on the GWs for multi-homing to/from the WAN.</t> <t> MAC-VRF: refers to an EVI instance in a particular node.</t> <t> MP2P and LSM tunnels: refer to Multi-Point to Point and Label Switched Multicast tunnels.</t> <t> ND: Neighbor Discovery protocol.</t> <t> NVE: Network Virtualization Edge.</t> <t> NVGRE: Network Virtualization using Generic Routing Encapsulation.</t> <t> NVO: refers to Network Virtualization Overlays.</t> <t> OAM: Operations and Maintenance.</t> <t> PBB: Provider Backbone Bridging.</t> <t> PE: Provider Edge.</t> <t> PW: Pseudowire.</t> <t> RD: Route-Distinguisher.</t> <t> RT: Route-Target.</t> <t> S/C-TAG: refers to a combination of Service Tag and Customer Tag in a 802.1Q frame.</t> <t> TOR: Top-Of-Rack switch.</t> <t> UMR: Unknown MAC Route.</t> <t> VNI/VSID: refers to VXLAN/NVGRE virtual identifiers.</t> <t> VPLS: Virtual Private LAN Service.</t> <t> VSI: Virtual Switch Instance or VPLS instance in a particular PE.</t> <t> VXLAN: Virtual eXtensible LAN.</t> </section> <section title="Introduction" anchor="sect-2"><t> <xref target="I-D.ietf-bess-evpn-overlay"/>target="RFC8365" format="default"/> discusses the use of Ethernet Virtual Private Networks(EVPN)(EVPNs) <xreftarget="RFC7432"/>target="RFC7432" format="default"/> as the control plane for Network Virtualization Overlays(NVO),(NVOs), where VXLAN <xreftarget="RFC7348"/>,target="RFC7348" format="default"/>, NVGRE <xreftarget="RFC7637"/>target="RFC7637" format="default"/>, or MPLS over GRE <xreftarget="RFC4023"/>target="RFC4023" format="default"/> can be used as possible data plane encapsulation options.</t> <t> While this model provides a scalable and efficientmulti-tenantmultitenant solution within the Data Center, it might not be easily extended to the Wide Area Network (WAN) in somecasescases, due to the requirements and existing deployed technologies. For instance, a Service Provider might have an already deployed Virtual Private LAN Service (VPLS) <xreftarget="RFC4761"/><xref target="RFC4762"/>,target="RFC4761" format="default"/> <xref target="RFC4762" format="default"/>, VPLS extensions for Provider Backbone Bridging (PBB-VPLS) <xreftarget="RFC7041"/>,target="RFC7041" format="default"/>, EVPN <xreftarget="RFC7432"/>target="RFC7432" format="default"/>, or PBB-EVPN <xreftarget="RFC7623"/>target="RFC7623" format="default"/> network that has to be used to interconnect Data Centers and WAN VPN users. A Gateway (GW) function is required in these cases. In fact, <xreftarget="I-D.ietf-bess-evpn-overlay"/>target="RFC8365" format="default"/> discusses two main Data Center Interconnect (DCI) solution groups: "DCI using GWs" and "DCI using ASBRs". This document specifies the solutions that correspond to the "DCI using GWs" group.</t> <t> It is assumed that the NVOGateway (GW)GW and the WAN Edge functions can be decoupledininto two separate systems or integrated into the same system. The former option will be referred to as"Decoupled Interconnect"decoupled interconnect solution" throughout the document, whereas the latter one will be referred to as"Integrated Interconnect"integrated interconnect solution".</t> <t> The specified procedures are local to the redundant GWs connecting a DC to the WAN. The document does not preclude any combination across different DCs for the same tenant. For instance, a "Decoupled" solution can be used in GW1 and GW2 (forDC1)DC1), and an "Integrated" solution can be used in GW3 and GW4 (for DC2).</t> <t> While the Gateways and WANPEsProvider Edges (PEs) use existing specifications in some cases, the document also defines extensions that are specific to DCI. In particular, those extensions are:</t><t><list style="symbols"><t>The<ul spacing="normal"> <li>The Interconnect Ethernet Segment (I-ES), an Ethernet Segment that can be associated to a set ofPWspseudowires (PWs) or other tunnels. The I-ES defined in this document is not associated with a set of Ethernet links, as per <xref target="RFC7432" format="default"/>, but rather with a set of virtual tunnels (e.g., a set of PWs). This set of virtual tunnels is referred to as vES <xref target="I-D.ietf-bess-evpn-virtual-eth-segment" format="default"/>.</li> <li>The use of the Unknown MAC Route (UMR) in a DCI scenario.</li> <li>The processing of EVPN routes on Gateways with MAC-VRFs connecting EVPN-Overlay and EVPN-MPLS networks, or EVPN-Overlay and EVPN-Overlay networks.</li> </ul> </section> <section anchor="sect-1" numbered="true" toc="default"> <name>Conventions and Terminology</name> <t> The key words "<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>", "<bcp14>MAY</bcp14>", and "<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as described in BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they appear in all capitals, as shown here. </t> <dl> <dt>AC:</dt><dd> Attachment Circuit</dd> <dt>ARP:</dt><dd>Address Resolution Protocol</dd> <dt>BUM:</dt><dd>Broadcast, Unknown Unicast and Multicast (traffic)</dd> <dt>CE:</dt><dd>Customer Equipment</dd> <dt>CFM:</dt><dd>Connectivity Fault Management</dd> <dt>DC:</dt><dd>Data Center</dd> <dt>DCI:</dt><dd>Data Center Interconnect</dd> <dt>DF:</dt><dd>Designated Forwarder</dd> <dt>EVI:</dt><dd>EVPN Instance</dd> <dt>EVPN:</dt><dd>Ethernet Virtual Private Network, as in <xref target="RFC7432" format="default"/></dd> <dt>EVPN Tunnel binding:</dt><dd>refers to a tunnel to a remote PE/NVE for a given EVI. Ethernet packets in these bindings are encapsulated with the Overlay or MPLS encapsulation and the EVPN label at the bottom of the stack.</dd> <dt>ES:</dt><dd>Ethernet Segment</dd> <dt>ESI:</dt><dd>Ethernet Segment Identifier</dd> <dt>GW:</dt><dd>Gateway or Data Center Gateway</dd> <dt>I-ES and I-ESI:</dt><dd>Interconnect Ethernet Segment and Interconnect Ethernet Segment Identifier. An I-ES is defined on the GWs for multihoming to/from the WAN.</dd> <dt>MAC</dt><dd>Media Access Control</dd> <dt>MAC-VRF:</dt><dd>refers to an EVI instance inthis document is not associated withaset of Ethernet links, as per <xref target="RFC7432"/>, but rather withparticular node</dd> <dt>MP2P and LSM tunnels:</dt><dd>refer to multipoint-to-point and label switched multicast tunnels</dd> <dt>ND:</dt><dd>Neighbor Discovery</dd> <dt>NDF:</dt><dd>Non-Designated Forwarder</dd> <dt>NVE:</dt><dd>Network Virtualization Edge</dd> <dt>NVGRE:</dt><dd>Network Virtualization using Generic Routing Encapsulation</dd> <dt>NVO:</dt><dd>Network Virtualization Overlay</dd> <dt>OAM:</dt><dd>Operations, Administration, and Maintenance</dd> <dt>PBB:</dt><dd>Provider Backbone Bridging</dd> <dt>PE:</dt><dd>Provider Edge</dd> <dt>PW:</dt><dd>Pseudowire</dd> <dt>RD:</dt><dd>Route Distinguisher</dd> <dt>RR:</dt><dd>Route Reflector</dd> <dt>RT:</dt><dd>Route Target</dd> <dt>S/C-TAG:</dt><dd>refers to asetcombination ofvirtual tunnels (e.g.,Service Tag and Customer Tag in aset of PWs). This set of virtual tunnels is referred to as vES <xref target="I-D.sajassi-bess-evpn-virtual-eth-segment"/>.</t> <t>The use of the Unknown802.1Q frame</dd> <dt>TOR:</dt><dd>Top-Of-Rack</dd> <dt>UMR:</dt><dd>Unknown MACrouteRoute</dd> <dt>vES:</dt><dd>virtual Ethernet Segment</dd> <dt>VNI/VSID:</dt><dd>refers to VXLAN/NVGRE virtual identifiers</dd> <dt>VPLS:</dt><dd>Virtual Private LAN Service</dd> <dt>VSI:</dt><dd>Virtual Switch Instance or VPLS instance in aDCI scenario.</t> <t>The processing of EVPN routes on Gateways with MAC-VRFs connecting EVPN-Overlay and EVPN-MPLS networks, or EVPN-Overlay and EVPN-Overlay networks.</t> </list> </t>particular PE</dd> <dt>VXLAN:</dt><dd>Virtual eXtensible LAN</dd> </dl> </section> <sectiontitle="Decoupledanchor="sect-3" numbered="true" toc="default"> <name>Decoupled InterconnectsolutionSolution forEVPN overlay networks" anchor="sect-3"><t>EVPN-Overlay Networks</name> <t> This section describes the interconnect solution when the GW and WAN Edge functions are implemented in different systems.Figure 1<xref target="fig-1"/> depicts the reference model described in this section. Note that, although not shown inFigure 1,<xref target="fig-1"/>, GWs may have localACs (Attachment Circuits).</t>Attachment Circuits (ACs).</t> <figuretitle="Decoupledanchor="fig-1"> <name>Decoupled Interconnectmodel" anchor="ure-decoupled-interconnect-model"><artwork><![CDATA[Model</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +--+ |CE| +--+ | +----+ +----| PE |----+ +---------+ | +----+ | +---------+ +----+ | +---+ +----+ +----+ +---+ | +----+ |NVE1|--| | | |WAN | |WAN | | | |--|NVE3| +----+ | |GW1|--|Edge| |Edge|--|GW3| | +----+ | +---+ +----+ +----+ +---+ | | NVO-1 | | WAN | | NVO-2 | | +---+ +----+ +----+ +---+ | | | | |WAN | |WAN | | | | +----+ | |GW2|--|Edge| |Edge|--|GW4| | +----+ |NVE2|--| +---+ +----+ +----+ +---+ |--|NVE4| +----+ +---------+ | | +---------+ +----+ +--------------+ |<-EVPN-Overlay-->|<-VLAN->|<-WAN L2VPN->|<--PW-->|<--EVPN-Overlay->|hand-off hand-offhandoff handoff ]]></artwork> </figure> <t> The following section describes the interconnect requirements for this model.</t> <sectiontitle="Interconnect requirements" anchor="sect-3.1"><t>anchor="sect-3.1" numbered="true" toc="default"> <name>Interconnect Requirements</name> <t> TheDecoupled Interconnectdecoupled interconnect architecture is intended to be deployed in networks where the EVPN-Overlay and WAN providers are different entities and a clear demarcation is needed. This solution solves the following requirements:</t><t><list style="symbols"><t>A<ul spacing="normal"> <li>A simple connectivityhand-offhandoff between the EVPN-Overlay network provider and the WAN provider so that QoS and security enforcementisare easilyaccomplished.</t> <t>Independenceaccomplished.</li> <li>Independence of theLayer Two VPN (L2VPN)L2VPN technology deployed in theWAN.</t> <t>Multi-homingWAN.</li> <li>Multihoming between GW and WAN Edge routers, including per-service load balancing. Per-flow load balancing is not a strongrequirementrequirement, since a deterministic path per service is usually required for an easy QoS and securityenforcement.</t> <t>Supportenforcement.</li> <li>Support of Ethernet OAM and Connectivity Fault Management (CFM) <xreftarget="IEEE.802.1AG"/><xref target="Y.1731"/>target="IEEE.802.1AG" format="default"/> <xref target="Y.1731" format="default"/> functions between the GW and the WAN Edge router to detect individual ACfailures.</t>failures.</li> <li> <t>Support for the following optimizations at theGW:<list style="symbols"><t>FloodingGW:</t> <ul spacing="normal"> <li>Flooding reduction of unknown unicast traffic sourced from the DC Network Virtualization Edgedevices (NVEs).</t> <t>Control(NVE) devices.</li> <li>Control of the WAN MAC addresses advertised to theDC.</t> <t>AddressDC.</li> <li>Address Resolution Protocol (ARP) and Neighbor Discovery (ND) flooding control for the requests coming from theWAN.</t> </list> </t> </list> </t>WAN.</li> </ul> </li> </ul> </section> <sectiontitle="VLAN-based hand-off" anchor="sect-3.2"><t>anchor="sect-3.2" numbered="true" toc="default"> <name>VLAN-Based Handoff</name> <t> In this option, thehand-offhandoff between the GWs and the WAN Edge routers is based on VLANs <xreftarget="IEEE.802.1Q-2014"/>.target="IEEE.802.1Q" format="default"/>. This is illustrated inFigure 1<xref target="fig-1"/> (between the GWs in NVO-1 and the WAN Edge routers). Each MAC-VRF in the GW is connected to a different VSI/MAC-VRF instance in the WAN Edge router by using a different C-TAG VLAN ID or a different combination of S/C-TAG VLAN IDs that matches at both sides.</t> <t> This option provides the best possible demarcation between the DC and WANprovidersproviders, and it does not require control plane interaction between both providers. The disadvantage of this model is the provisioningoverheadoverhead, since the service has to be mapped to a C-TAG or S/C-TAG VLAN ID combination at both GW and WAN Edge routers.</t> <t> In this model, the GW acts as a regular Network Virtualization Edge (NVE) towards the DC. Its control plane, data planeproceduresprocedures, and interactions are described in <xreftarget="I-D.ietf-bess-evpn-overlay"/>.</t>target="RFC8365" format="default"/>.</t> <t> The WAN Edge router acts as a (PBB-)VPLS or (PBB-)EVPN PE withattachment circuitsAttachment Circuits (ACs) to the GWs. Its functions are described in <xreftarget="RFC4761"/>,target="RFC4761" format="default"/>, <xreftarget="RFC4762"/>,target="RFC4762" format="default"/>, <xreftarget="RFC6074"/> ortarget="RFC6074" format="default"/>, <xreftarget="RFC7432"/>,target="RFC7432" format="default"/>, and <xreftarget="RFC7623"/>.</t>target="RFC7623" format="default"/>.</t> </section> <sectiontitle="PW-based (Pseudowire-based) hand-off" anchor="sect-3.3"><t>anchor="sect-3.3" numbered="true" toc="default"> <name>PW-Based Handoff</name> <t> If MPLS between the GW and the WAN Edge router is an option, a PW-basedInterconnectinterconnect solution can be deployed. In thisoptionoption, thehand-offhandoff between both routers is based on FEC128-based PWs <xreftarget="RFC4762"/>target="RFC4762" format="default"/> or FEC129-based PWs (for a greater level of network automation) <xreftarget="RFC6074"/>.target="RFC6074" format="default"/>. Note that this model still provides a clear demarcationboundarybetween DC and WAN (since there is a single PW between each MAC-VRF and peer VSI), and security/QoS policies may be applied on aper PWper-PW basis. This model provides better scalability than aC-TAG based hand-offC-TAG-based handoff and less provisioning overhead than a combined C/S-TAGhand-off.handoff. The PW-basedhand-offhandoff interconnect is illustrated inFigure 1<xref target="fig-1"/> (between the NVO-2 GWs and the WAN Edge routers).</t> <t> In this model, besides the usual MPLS procedures between GW and WAN Edge router <xreftarget="RFC3031"/>,target="RFC3031" format="default"/>, the GWMUST<bcp14>MUST</bcp14> support an interworking function in each MAC-VRF that requires extension to the WAN:</t><t><list style="symbols"><t>If<ul spacing="normal"> <li>If a FEC128-based PW is used between the MAC-VRF (GW) and the VSI (WAN Edge), the correspondingVCID MUSTVirtual Connection Identifier (VCID) <bcp14>MUST</bcp14> be provisioned on the MAC-VRF and match the VCID used in the peer VSI at the WAN Edgerouter.</t> <t>Ifrouter.</li> <li>If BGP Auto-discovery <xreftarget="RFC6074"/>target="RFC6074" format="default"/> and FEC129-based PWs are used between the GW MAC-VRF and the WAN Edge VSI, the provisioning of the VPLS-IDMUST<bcp14>MUST</bcp14> be supported on the MAC-VRF andMUST<bcp14>MUST</bcp14> match the VPLS-ID used in the WAN EdgeVSI.</t> </list> </t>VSI.</li> </ul> <t> If a PW-based handoff is used, the GW's AC (or point of attachment to the EVPNInstance)instance) uses a combination of a PW label and VLAN IDs. PWs are treated as serviceinterfacesinterfaces, defined in <xreftarget="RFC7432"/>.</t>target="RFC7432" format="default"/>.</t> </section> <sectiontitle="Multi-homing solutionanchor="sect-3.4" numbered="true" toc="default"> <name>Multihoming Solution on theGWs" anchor="sect-3.4"><t>GWs</name> <t> EVPN single-activemulti-homing, i.e.multihoming -- i.e., per-service load-balancingmulti-homingmultihoming -- is required in this type of interconnect.</t> <t> The GWs will be provisioned with a unique ESperfor each WAN interconnect, and thehand-offhandoff attachment circuits or PWs between the GW and the WAN Edge router will be assigned an ESI for each such ES. The ESI will be administratively configured on the GWs according to the procedures in <xreftarget="RFC7432"/>.target="RFC7432" format="default"/>. ThisInterconnect ESI-ES will be referred to as "I-ES" hereafter, and its identifier will be referred to as "I-ESI".<xref target="RFC7432"/> describes differentDifferent ESITypes.types are described in <xref target="RFC7432" format="default"/>. The use of Type 0 for the I-ESI isRECOMMENDED<bcp14>RECOMMENDED</bcp14> in this document.</t> <t> The solution (on the GWs)MUST<bcp14>MUST</bcp14> follow the single-activemulti-homingmultihoming procedures as described in <xreftarget="I-D.ietf-bess-evpn-overlay"/>target="RFC8365" format="default"/> for the provisionedI-ESI, i.e.I-ESI -- i.e., Ethernet A-D routes per ES and per EVI will be advertised to the DC NVEs for themulti-homingmultihoming functions, and ES routes will be advertised so that ES discovery and Designated Forwarder (DF) procedures can be followed. The MAC addresses learned (in the data plane) on thehand-offhandoff links will be advertised with the I-ESI encoded in the ESI field.</t> </section> <sectiontitle="Gateway Optimizations" anchor="sect-3.5"><t>anchor="sect-3.5" numbered="true" toc="default"> <name>Gateway Optimizations</name> <t> The following GW features are optional and optimize the control plane and data plane in the DC.</t> <sectiontitle="MACanchor="sect-3.5.1" numbered="true" toc="default"> <name>MAC Address AdvertisementControl" anchor="sect-3.5.1"><t>Control</name> <t> The use of EVPN in NVO networks brings a significant number ofbenefitsbenefits, as described in <xreftarget="I-D.ietf-bess-evpn-overlay"/>.target="RFC8365" format="default"/>. However, if multiple DCs are interconnected into a single EVI, each DC will have to import all of the MAC addresses from each of the other DCs.</t> <t> Even if optimized BGP techniques likeRT-constraintRT constraint <xreftarget="RFC4684"/>target="RFC4684" format="default"/> are used, the number of MAC addresses to advertise or withdraw (in case of failure) by the GWs of a given DC could overwhelm the NVEs within that DC, particularly when the NVEs reside in the hypervisors.</t> <t> The solution specified in this document uses the'UnknownUnknown MACRoute'Route (UMR)whichthat is advertised into a given DC by each of the DC's GWs. This route is defined in <xreftarget="RFC7543"/>target="RFC7543" format="default"/> and is a regular EVPN MAC/IP Advertisement route in which the MAC Address Length is set to 48, the MAC address is set to 0, and the ESI field is set to the DC GW's I-ESI.</t> <t> An NVE within that DC that understands andprocessprocesses the UMR will send unknown unicast frames to one of theDCsDC's GWs, which will then forward that packet to the correct egress PE. Note that, because the ESI is set to the DC GW's I-ESI, all-activemulti-homingmultihoming can be applied to unknown unicast MAC addresses. An NVE that does not understand the Unknown MACrouteRoute will handle unknown unicast as described in <xreftarget="RFC7432"/>.</t>target="RFC7432" format="default"/>.</t> <t> This document proposes that local policydeterminesdetermine whether MAC addresses and/or the UMR are advertised into a given DC. As an example, when all the DC MAC addresses are learned in the control/management plane, it may be appropriate to advertise only the UMR. Advertising all the DC MAC addresses in the control/management plane is usually the case when the NVEs reside in hypervisors. Refer to <xreftarget="I-D.ietf-bess-evpn-overlay"/> section 7.</t>target="RFC8365" sectionFormat="comma" section="7"/>.</t> <t> It is worth noting that the UMR usage in <xreftarget="RFC7543"/>target="RFC7543" format="default"/> and the UMR usage in this document are different. In the former, a Virtual Spoke (V-spoke) does not necessarily learn all the MAC addresses pertaining to hosts in other V-spokes of the same network. The communication between two V-spokes is done through theDMG,Default MAC Gateway (DMG) until the V-spokes learn each other's MAC addresses. In this document, two leaf switches in the same DC are recommended for V-spokes to learn each other's MAC addresses for the same EVI. Theleaf to leafleaf-to-leaf communication is always direct and does not go through the GW.</t> </section> <sectiontitle="ARP/ND flooding control" anchor="sect-3.5.2"><t>anchor="sect-3.5.2" numbered="true" toc="default"> <name>ARP/ND Flooding Control</name> <t> Another optimization mechanism, naturally provided by EVPN in the GWs, is the Proxy ARP/ND function. The GWs should build a Proxy ARP/ND cachetabletable, as per <xreftarget="RFC7432"/>.target="RFC7432" format="default"/>. When the active GW receives an ARP/ND request/solicitation coming from the WAN, the GW does a Proxy ARP/ND table lookup and replies as long as the information is available in its table.</t> <t> This mechanism is especially recommended on the GWs, since it protects the DC network from external ARP/ND-flooding storms.</t> </section> <sectiontitle="Handling failuresanchor="sect-3.5.3" numbered="true" toc="default"> <name>Handling Failures between GW and WAN Edgerouters" anchor="sect-3.5.3"><t>Routers</name> <t> Link/PE failures are handled on the GWs as specified in <xreftarget="RFC7432"/>.target="RFC7432" format="default"/>. The GW detecting the failure will withdraw the EVPNroutesroutes, as per <xreftarget="RFC7432"/>.</t>target="RFC7432" format="default"/>.</t> <t> Individual AC/PW failures may be detected by OAM mechanisms. For instance:</t><t><list style="symbols"><t>If<ul spacing="normal"> <li>If theInterconnectinterconnect solution is based on a VLANhand-off,handoff, Ethernet-CFM <xreftarget="IEEE.802.1AG"/><xref target="Y.1731"/>target="IEEE.802.1AG" format="default"/> <xref target="Y.1731" format="default"/> may be used to detect individual AC failures onboth,both the GW and WAN Edge router. An individual AC failure will trigger the withdrawal of the corresponding A-D per EVI route as well as the MACs learned on thatAC.</t> <t>IfAC.</li> <li>If theInterconnectinterconnect solution is based on a PWhand-off,handoff, the Label Distribution Protocol (LDP) PW Status bits TLV <xreftarget="RFC6870"/>target="RFC6870" format="default"/> may be used to detect individual PW failures onboth,both the GW and WAN Edgerouter.</t> </list> </t>router.</li> </ul> </section> </section> </section> <sectiontitle="Integratedanchor="sect-4" numbered="true" toc="default"> <name>Integrated InterconnectsolutionSolution forEVPN overlay networks" anchor="sect-4"><t>EVPN-Overlay Networks</name> <t> When the DC and the WAN are operated by the same administrative entity, the Service Provider can decide to integrate the GW and WAN Edge PE functions in the same router for obviousCAPEXreasons to save as relates to Capital Expenditure (CAPEX) andOPEX saving reasons.Operating Expenses (OPEX). This is illustrated inFigure 2.<xref target="fig-2"/>. Note that this model does not provide an explicit demarcation link between DC and WAN anymore. Although not shown inFigure 2,<xref target="fig-2"/>, note that the GWs may have local ACs.</t> <figuretitle="Integratedanchor="fig-2"> <name>Integrated Interconnectmodel" anchor="ure-integrated-interconnect-model"><artwork><![CDATA[Model</name> <artwork name="" type="" align="left" alt=""><![CDATA[ +--+ |CE| +--+ | +----+ +----| PE |----+ +---------+ | +----+ | +---------+ +----+ | +---+ +---+ | +----+ |NVE1|--| | | | | |--|NVE3| +----+ | |GW1| |GW3| | +----+ | +---+ +---+ | | NVO-1 | WAN | NVO-2 | | +---+ +---+ | | | | | | | +----+ | |GW2| |GW4| | +----+ |NVE2|--| +---+ +---+ |--|NVE4| +----+ +---------+ | | +---------+ +----+ +--------------+ |<--EVPN-Overlay--->|<-----VPLS--->|<---EVPN-Overlay-->| |<--PBB-VPLS-->| Interconnect -> |<-EVPN-MPLS-->| options |<--EVPN-Ovl-->|* |<--PBB-EVPN-->|]]></artwork> </figure> <t>** EVPN-Ovl stands for EVPN-Overlay (and it's anInterconnect option).</t>interconnect option). ]]></artwork> </figure> <sectiontitle="Interconnect requirements" anchor="sect-4.1"><t>anchor="sect-4.1" numbered="true" toc="default"> <name>Interconnect Requirements</name> <t> TheIntegrated Interconnectintegrated interconnect solution meets the following requirements:</t><t><list style="symbols"><t>Control<ul spacing="normal"> <li>Control plane and data plane interworking between theEVPN-overlayEVPN-Overlay network and the L2VPN technology supported in the WAN, irrespective of the technologychoice, i.e.choice -- i.e., (PBB-)VPLS or (PBB-)EVPN, as depicted inFigure 2.</t> <t>Multi-homing,<xref target="fig-2"/>.</li> <li>Multihoming, including single-activemulti-homingmultihoming with per-service load balancing or all-activemulti-homing, i.e.multihoming -- i.e., per-flowload-balancing,load-balancing -- as long as the technology deployed in the WAN supportsit.</t> <t>Supportit.</li> <li>Support for end-to-end MAC Mobility, Static MAC protection and other procedures(e.g.(e.g., proxy-arp) described in <xreftarget="RFC7432"/>target="RFC7432" format="default"/> as long as EVPN-MPLS is the technology of choice in theWAN.</t> <t>IndependentWAN.</li> <li>Independent inclusive multicast trees in the WAN and in the DC. That is, the inclusive multicast tree type defined in the WAN does not need to be the same as in theDC.</t> </list> </t>DC.</li> </ul> </section> <sectiontitle="VPLSanchor="sect-4.2" numbered="true" toc="default"> <name>VPLS Interconnect for EVPN-Overlaynetworks" anchor="sect-4.2"><section title="Control/DataNetworks</name> <section anchor="sect-4.2.1" numbered="true" toc="default"> <name>Control/Data Planesetup proceduresSetup Procedures on theGWs" anchor="sect-4.2.1"><t>GWs</name> <t> Regular MPLS tunnels andTLDP/BGPTargeted LDP (tLDP) / BGP sessions will besetupset up to the WAN PEs and RRs as per <xreftarget="RFC4761"/>,target="RFC4761" format="default"/>, <xreftarget="RFC4762"/>,target="RFC4762" format="default"/>, and <xreftarget="RFC6074"/>target="RFC6074" format="default"/>, and overlay tunnels and EVPN will besetupset up as per <xreftarget="I-D.ietf-bess-evpn-overlay"/>.target="RFC8365" format="default"/>. Note that differentroute-targetsroute targets for the DC andforthe WAN are normally required (unless <xreftarget="RFC4762"/>target="RFC4762" format="default"/> is used in the WAN, in which case no WANroute-targetroute target is needed). A single type-1 RD per service may be used.</t> <t> In order to supportmulti-homing,multihoming, the GWs will be provisioned with an I-ESI (seesection 3.4), that<xref target="sect-3.4"/>), which will be uniqueperfor each interconnection.The I-ES inIn thiscasecase, the I-ES will represent the group of PWs to the WAN PEs and GWs. All the <xreftarget="RFC7432"/>target="RFC7432" format="default"/> procedures are still followed for theI-ES, e.g.I-ES -- e.g., any MAC address learned from the WAN will be advertised to the DC with the I-ESI in the ESI field.</t> <t> A MAC-VRF per EVI will be created in each GW. The MAC-VRF will have two different types of tunnel bindings instantiated in two differentsplit-horizon-groups:</t> <t><list style="symbols"> <t>split-horizon groups:</t> <ul spacing="normal"> <li> VPLS PWs will be instantiated in the"WAN split-horizon-group".</t> <t>WAN split-horizon group.</li> <li> Overlay tunnel bindings(e.g.(e.g., VXLAN, NVGRE) will be instantiated in the"DC split-horizon-group".</t> </list> </t>DC split-horizon group.</li> </ul> <t> Attachment circuits are also supported on the same MAC-VRF (although not shown inFigure 2),<xref target="fig-2"/>), but they will not be part of any of the abovesplit-horizon-groups.</t>split-horizon groups.</t> <t> Traffic received in a givensplit-horizon-groupsplit-horizon group will never be forwarded to a member of the samesplit-horizon-group.</t>split-horizon group.</t> <t> As far as BUM flooding is concerned, a flooding list will be composed of thesub-listsublist created by the inclusive multicast routes and thesub-listsublist created for VPLS in the WAN. BUM frames received from a local Attachment Circuit (AC) will be forwarded to the flooding list. BUM frames received from the DC or the WAN will be forwarded to the floodinglistlist, observing thesplit-horizon-groupsplit-horizon group rule described above.</t> <t> Note that the GWs are not allowed to have an EVPN binding and a PW to the samefar-endfar end within the same MAC-VRF, so that loops and packet duplication are avoided. In case a GW can successfully establishboth,both an EVPN binding and a PW to the same far-end PE, the EVPN binding willprevailprevail, and the PW will be broughtoperationally down.</t>down operationally.</t> <t> Theoptimizationsoptimization procedures described insection 3.5<xref target="sect-3.5"/> can also be applied to this model.</t> </section> <sectiontitle="Multi-homing proceduresanchor="sect-4.2.2" numbered="true" toc="default"> <name>Multihoming Procedures on theGWs" anchor="sect-4.2.2"><t>GWs</name> <t> This model supports single-activemulti-homingmultihoming on the GWs. All-activemulti-homingmultihoming is not supported byVPLS, thereforeVPLS; therefore, it cannot be used on the GWs.</t> <t> In this case, for a given EVI, all the PWs in the WANsplit-horizon-groupsplit-horizon group are assigned to I-ES. All the single-activemulti-homingmultihoming procedures as described by <xreftarget="I-D.ietf-bess-evpn-overlay"/>target="RFC8365" format="default"/> will be followed for the I-ES.</t> <t> The non-DF GW for the I-ES will block the transmission and reception of all the PWs in the"WAN split-horizon-group"WAN split-horizon group for BUM and unicast traffic.</t> </section> </section> <sectiontitle="PBB-VPLSanchor="sect-4.3" numbered="true" toc="default"> <name>PBB-VPLS Interconnect for EVPN-Overlaynetworks" anchor="sect-4.3"><section title="Control/DataNetworks</name> <section anchor="sect-4.3.1" numbered="true" toc="default"> <name>Control/Data Planesetup proceduresSetup Procedures on theGWs" anchor="sect-4.3.1"><t>GWs</name> <t> In this case, there is no impact on the procedures described in <xreftarget="RFC7041"/>target="RFC7041" format="default"/> for the B-component.HoweverHowever, the I-component instances become EVI instances with EVPN-Overlay bindings and potentially local attachment circuits. A number of MAC-VRF instances can be multiplexed into the same B-component instance. This option provides significant savings in terms of PWs to be maintained in the WAN.</t> <t> The I-ESI concept described insection 4.2.1<xref target="sect-4.2.1"/> will also be used for the PBB-VPLS-basedInterconnect.</t>interconnect.</t> <t> B-component PWs and I-componentEVPN-overlayEVPN-Overlay bindings established to the samefar-endfar end will be compared. The following rules will be observed:</t><t><list style="symbols"> <t><ul spacing="normal"> <li> Attempts tosetupset up a PW between the two GWs within the B-component context will never beblocked.</t> <t>blocked.</li> <li> If a PW exists between two GWs for the B-component and an attempt is made tosetupset up an EVPN binding on an I-component linked to that B-component, the EVPN binding will be keptoperationally down.down operationally. Note that the BGP EVPN routes will still be valid but notused.</t> <t>used.</li> <li> The EVPN binding will only be up and used as long as there is no PW to the samefar-endfar end in the corresponding B-component. The EVPN bindings in the I-components will be brought down before the PW in the B-component is broughtup.</t> </list> </t>up.</li> </ul> <t> Theoptimizationsoptimization procedures described insection 3.5<xref target="sect-3.5"/> can also be applied to thisInterconnectinterconnect option.</t> </section> <sectiontitle="Multi-homing proceduresanchor="sect-4.3.2" numbered="true" toc="default"> <name>Multihoming Procedures on theGWs" anchor="sect-4.3.2"><t>GWs</name> <t> This model supports single-activemulti-homingmultihoming on the GWs. All-activemulti-homingmultihoming is not supported by this scenario.</t> <t> The single-activemulti-homingmultihoming procedures as described by <xreftarget="I-D.ietf-bess-evpn-overlay"/>target="RFC8365" format="default"/> will be followed for the I-ES for each EVI instance connected to the B-component. Note that in this case, for a given EVI, all the EVPN bindings in the I-component are assigned to the I-ES. The non-DF GW for the I-ES will block the transmission and reception of all the I-component EVPN bindings for BUM and unicast traffic. When learning MACs from the WAN, the non-DFMUST NOT<bcp14>MUST NOT</bcp14> advertise EVPN MAC/IP routes for those MACs.</t> </section> </section> <sectiontitle="EVPN-MPLSanchor="sect-4.4" numbered="true" toc="default"> <name>EVPN-MPLS Interconnect for EVPN-Overlaynetworks" anchor="sect-4.4"><t>Networks</name> <t> If EVPN for MPLStunnels, EVPN-MPLS hereafter, istunnels (referred to as "EVPN-MPLS" hereafter) are supported in the WAN, an end-to-end EVPN solution can be deployed. The following sections describe the proposed solution as well astheits impactrequiredon the procedures from <xreftarget="RFC7432"/> procedures.</t>target="RFC7432" format="default"/>.</t> <sectiontitle="Control Plane setup proceduresanchor="sect-4.4.1" numbered="true" toc="default"> <name>Control plane Setup Procedures on theGWs" anchor="sect-4.4.1"><t>GWs</name> <t> The GWsMUST<bcp14>MUST</bcp14> establish separate BGP sessions for sending/receiving EVPN routes to/from the DC and to/from the WAN.NormallyNormally, each GW willsetupset up one BGP EVPN session to the DC RR (or two BGP EVPN sessions if there are redundant DC RRs) and one session to the WAN RR (or two sessions if there are redundant WAN RRs).</t> <t> In order to facilitate separate BGP processes for DC and WAN, EVPN routes sent to the WANSHOULD<bcp14>SHOULD</bcp14> carry a differentroute-distinguisherRoute Distinguisher (RD) than the EVPN routes sent to the DC. In addition, although reusing the same value is possible, differentroute-targetsroute targets are expected to be handled for the same EVI in the WAN and the DC. Note that the EVPN service routes sent to the DC RRs will normally include a <xreftarget="I-D.ietf-idr-tunnel-encaps"/>target="RFC9012" format="default"/> BGP encapsulation extended community with a different tunnel type than the one sent to the WAN RRs.</t> <t> As in the other discussed options, an I-ES and its assigned I-ESI will be configured on the GWs formulti-homing.multihoming. This I-ES represents the WAN EVPN-MPLS PEs to the DC but also the DC EVPN-Overlay NVEs to the WAN. Optionally, different I-ESI values are configured for representing the WAN and the DC. If different EVPN-Overlay networks are connected to the same group of GWs, each EVPN-Overlay networkMUST<bcp14>MUST</bcp14> get assigned a different I-ESI.</t> <t> Received EVPN routes will never be reflected on the GWs but instead will be consumed and re-advertised (if needed):</t><t><list style="symbols"> <t>Ethernet<ul spacing="normal"> <li>Ethernet A-D routes, ESroutesroutes, and Inclusive Multicast routes are consumed by the GWs and processed locally for the corresponding <xreftarget="RFC7432"/> procedures.</t>target="RFC7432" format="default"/> procedures.</li> <li> <t>MAC/IP advertisement routes will bereceived, importedreceived and imported, and if they become active in the MAC-VRF, the information will be re-advertised as new routes with the following fields:<list style="symbols"> <t>The</t> <ul spacing="normal"> <li>The RD will be the GW's RD for theMAC-VRF.</t> <t>TheMAC-VRF.</li> <li>The ESI will be set to theI-ESI.</t> <t>TheI-ESI.</li> <li>The Ethernet-tag value will be kept from the received NLRI the receivedNLRI.</t> <t>TheNLRI.</li> <li>The MAC length, MAC address, IPLengthLength, and IP address values will be kept from the receivedNLRI.</t> <t>TheNLRI.</li> <li>The MPLS label will be a local 20-bit value (when sent to the WAN) or a DC-global 24-bit value (when sent to the DC for encapsulations using aVNI).</t> <t>TheVNI).</li> <li>The appropriateRoute-TargetsRoute Targets (RTs) and <xreftarget="I-D.ietf-idr-tunnel-encaps"/>target="RFC9012" format="default"/> BGPEncapsulationencapsulation extended community will be used according to <xreftarget="I-D.ietf-bess-evpn-overlay"/>.</t> </list> </t> </list> </t>target="RFC8365" format="default"/>.</li> </ul> </li> </ul> <t> The GWs will also generate the following local EVPN routes that will be sent to the DC and WAN, with their corresponding RTs and <xreftarget="I-D.ietf-idr-tunnel-encaps"/>target="RFC9012" format="default"/> BGPEncapsulationencapsulation extended community values:</t><t><list style="symbols"> <t>ES<ul spacing="normal"> <li>ES route(s) for theI-ESI(s).</t> <t>EthernetI-ESI(s).</li> <li>Ethernet A-D routes per ES and EVI for the I-ESI(s). The A-D per-EVI routes sent to the WAN and the DC will have consistent Ethernet-Tagvalues.</t> <t>Inclusivevalues.</li> <li>Inclusive Multicast routes with independenttunnel typetunnel-type value for the WAN and DC.E.g.For example, a P2MPLSPLabel Switched Path (LSP) may be used in theWANWAN, whereas ingress replication may be used in the DC. The routes sent to the WAN and the DC will have a consistentEthernet-Tag.</t> <t>MAC/IPEthernet-Tag.</li> <li>MAC/IP advertisement routes for MAC addresses learned in local attachment circuits. Note that these routes will not include theI-ESI, but ESI=0I-ESI value in the ESI field. These routes will include a zero ESI ordifferent from 0a non-zero ESI for localmulti-homedmultihomed Ethernet Segments (ES). The routes sent to the WAN and the DC will have a consistentEthernet-Tag.</t> </list> </t>Ethernet-Tag.</li> </ul> <t> Assuming GW1 and GW2 are peer GWs of the same DC, each GW will generate two sets of the above local service routes:Set-DCset-DC will be sent to the DC RRs and will include an A-D per EVI, InclusiveMulticastMulticast, and MAC/IP routes for the DC encapsulation and RT. Set-WAN will be sent to the WAN RRs and will include the same routes but using the WAN RT and encapsulation. GW1 and GW2 will receive each other's set-DC and set-WAN. This is the expected behavior on GW1 and GW2 for locally generated routes:</t><t><list style="symbols"> <t>Inclusive<ul spacing="normal"> <li>Inclusive multicast routes:whenWhen setting up the flooding lists for a given MAC-VRF, each GW will include its DC peer GW only in the EVPN-MPLS flooding list (by default) and not the EVPN-Overlay flooding list. That is, GW2 will import two Inclusive Multicast routes from GW1 (from set-DC and set-WAN) but will only consider one of the two,havinggiving the set-WAN route higher priority. An administrative optionMAY<bcp14>MAY</bcp14> change this preference so that the set-DC route is selectedfirst.</t> <t>MAC/IPfirst.</li> <li>MAC/IP advertisement routes for local attachment circuits:asAs above, the GW will select only one,havinggiving the route from the set-WAN a higher priority. As with the Inclusive multicast routes, an administrative optionMAY<bcp14>MAY</bcp14> change thispriority.</t> </list> </t>priority.</li> </ul> </section> <sectiontitle="Dataanchor="sect-4.4.2" numbered="true" toc="default"> <name>Data Planesetup proceduresSetup Procedures on theGWs" anchor="sect-4.4.2"><t>GWs</name> <t> The procedure explained at the end of the previous section will make sure there are no loops or packet duplication between the GWs of the same EVPN-Overlay network (for frames generated from localACs)ACs), since only one EVPN binding per EVI (or per Ethernet Tag in the case of VLAN-aware bundle services) will besetupset up in the data plane between the two nodes. That binding will by default be added to the EVPN-MPLS flooding list.</t> <t> As for the rest of the EVPN tunnel bindings, they will be added to one of the two flooding lists that each GW sets up for the same MAC-VRF:</t><t><list style="symbols"> <t>EVPN-overlay<ul spacing="normal"> <li>EVPN-Overlay flooding list (composed of bindings to the remote NVEs or multicast tunnel to theNVEs).</t> <t>EVPN-MPLSNVEs).</li> <li>EVPN-MPLS flooding list (composed of MP2P or LSM tunnel to the remotePEs)</t> </list> </t>PEs).</li> </ul> <t> Each flooding list will be part of a separatesplit-horizon-group:split-horizon group: the WANsplit-horizon-groupsplit-horizon group or the DCsplit-horizon-group.split-horizon group. Traffic generated from a local AC can be flooded to bothsplit-horizon-groups.split-horizon groups. Traffic from a binding of asplit-horizon-groupsplit-horizon group can be flooded to the othersplit-horizon-groupsplit-horizon group and local ACs, but never to a member of its ownsplit-horizon-group.</t>split-horizon group.</t> <t> When either GW1 or GW2receivereceives a BUM frame on an MPLStunneltunnel, including an ESI label at the bottom of the stack, they will perform an ESI label lookup and split-horizon filtering as per <xreftarget="RFC7432"/>target="RFC7432" format="default"/>, in case the ESI label identifies a local ESI (I-ESI or any othernon-zerononzero ESI).</t> </section> <sectiontitle="Multi-homing procedure extensionsanchor="sect-4.4.3" numbered="true" toc="default"> <name>Multihoming Procedure Extensions on theGWs" anchor="sect-4.4.3"><t>GWs</name> <t> This model supports single-active as well as all-activemulti-homing.</t>multihoming.</t> <t> All the <xreftarget="RFC7432"/> multi-homingtarget="RFC7432" format="default"/> multihoming procedures for the DF election onI-ES(s)I-ES(s), as well as the backup-path (single-active) and aliasing (all-active)proceduresprocedures, will be followed on the GWs. Remote PEs in the EVPN-MPLS network will follow regular <xreftarget="RFC7432"/>target="RFC7432" format="default"/> aliasing or backup-path procedures for MAC/IP routes received from the GWs for the same I-ESI. So will NVEs in the EVPN-Overlay network for MAC/IP routes received with the same I-ESI.</t> <t> As far as the forwarding plane is concerned, by default, the EVPN-Overlay network will have an analogous behavior to the access ACs in <xreftarget="RFC7432"/> multi-homedtarget="RFC7432" format="default"/> multihomed Ethernet Segments.</t><t><list style="symbols"><t>The forwarding behavior on the GWs is described below:</t> <ul spacing="normal"> <li> <t>Single-activemulti-homing;multihoming; assuming a WANsplit-horizon-groupsplit-horizon group (comprised of EVPN-MPLS bindings), a DCsplit-horizon-groupsplit-horizon group (comprised of EVPN-Overlaybindings)bindings), and local ACs on the GWs:<list style="symbols"> <t>Forwarding</t> <ul spacing="normal"> <li>Forwarding behavior on the non-DF:theThe non-DFMUST<bcp14>MUST</bcp14> block ingress and egress forwarding on the EVPN-Overlay bindings associated to the I-ES. The EVPN-MPLS network is considered to be the corenetworknetwork, and the EVPN-MPLS bindings to the remote PEs and GWs will beactive.</t> <t>Forwardingactive.</li> <li>Forwarding behavior on the DF:theThe DFMUST NOT<bcp14>MUST NOT</bcp14> forward BUM or unicast traffic received from a givensplit-horizon-groupsplit-horizon group to a member ofhisits own split-horizon group. Forwarding to othersplit-horizon-groupssplit-horizon groups and local ACs is allowed (as long as the ACs are not part of an ES for which the node is non-DF). As per <xreftarget="RFC7432"/>target="RFC7432" format="default"/> and for split-horizon purposes, when receiving BUM traffic on the EVPN-Overlay bindings associated to an I-ES, the DF GWSHOULD<bcp14>SHOULD</bcp14> add the I-ESI label when forwarding to the peer GW overEVPN-MPLS.</t> <t>WhenEVPN-MPLS.</li> <li>When receiving EVPN MAC/IP routes from the WAN, the non-DFMUST NOT re-originate<bcp14>MUST NOT</bcp14> reoriginate the EVPN routes and advertise them to the DC peers. In the same way, EVPN MAC/IP routes received from the DCMUST NOT<bcp14>MUST NOT</bcp14> be advertised to the WAN peers. This is consistent with <xreftarget="RFC7432"/>target="RFC7432" format="default"/> and allows the remote PE/NVEs to know who the primary GW is, based on the reception of the MAC/IProutes.</t> </list> </t> </list> </t> <t><list style="symbols">routes.</li> </ul> </li> </ul> <ul spacing="normal"> <li> <t>All-activemulti-homing;multihoming; assuming a WANsplit-horizon-groupsplit-horizon group (comprised of EVPN-MPLS bindings), a DCsplit-horizon-groupsplit-horizon group (comprised of EVPN-Overlaybindings)bindings), and local ACs on the GWs:<list style="symbols"> <t>Forwarding</t> <ul spacing="normal"> <li>Forwarding behavior on the non-DF:theThe non-DF follows the same behavior as the non-DF in the single-activecasecase, but only for BUM traffic. Unicast traffic received from asplit-horizon-group MUST NOTsplit-horizon group <bcp14>MUST NOT</bcp14> be forwarded to a member of its ownsplit-horizon-groupsplit-horizon group but can be forwarded normally to the othersplit-horizon-groupssplit-horizon groups and local ACs. If a known unicast packet is identified as a "flooded" packet, the procedures for BUM trafficMUST<bcp14>MUST</bcp14> befollowed.</t> <t>Forwardingfollowed.</li> <li>Forwarding behavior on the DF:theThe DF follows the same behavior as the DF in the single-activecasecase, but only for BUM traffic. Unicast traffic received from asplit-horizon-group MUST NOTsplit-horizon group <bcp14>MUST NOT</bcp14> be forwarded to a member of its ownsplit-horizon-groupsplit-horizon group but can be forwarded normally to the othersplit-horizon-groupsplit-horizon group and local ACs. If a known unicast packet is identified as a "flooded" packet, the procedures for BUM trafficMUST<bcp14>MUST</bcp14> be followed. As per <xreftarget="RFC7432"/>target="RFC7432" format="default"/> and for split-horizon purposes, when receiving BUM traffic on the EVPN-Overlay bindings associated to an I-ES, the DF GWMUST<bcp14>MUST</bcp14> add the I-ESI label when forwarding to the peer GW overEVPN-MPLS.</t> <t>ContraryEVPN-MPLS.</li> <li>Contrary to the single-activemulti-homingmultihoming case, both DF and non-DFre-originatereoriginate and advertise MAC/IP routes received from the WAN/DC peers, adding the corresponding I-ESI so that the remote PE/NVEs can perform regularaliasingaliasing, as per <xreftarget="RFC7432"/>.</t> </list> </t> </list> </t>target="RFC7432" format="default"/>.</li> </ul> </li> </ul> <t> The example inFigure 3<xref target="fig-3"/> illustrates the forwarding of BUM traffic originated from an NVE on a pair of all-activemulti-homingmultihoming GWs.</t> <figuretitle="Multi-hominganchor="fig-3"> <name>Multihoming BUMforwarding" anchor="ure-multi-homing-bum-forwarding"><artwork><![CDATA[Forwarding</name> <artwork name="" type="" align="left" alt=""><![CDATA[ |<--EVPN-Overlay--->|<--EVPN-MPLS-->| +---------+ +--------------+ +----+ BUM +---+ | |NVE1+----+----> | +-+-----+ | +----+ | | DF |GW1| | | | | | +-+-+ | | ++--+ | | | | +--> |PE1| | +--->X +-+-+ | ++--+ | NDF| | | | +----+ | |GW2<-+ | |NVE2+--+ +-+-+ | +----+ +--------+ | +------------+ v +--+ |CE| +--+ ]]></artwork> </figure> <t> GW2 is the non-DF for the I-ES and blocks the BUM forwarding. GW1 is the DF and forwards the traffic to PE1 and GW2. Packets sent to GW2 will include theESI-labelESI label for the I-ES. Based on theESI-label,ESI label, GW2 identifies the packets as I-ES-generated packets and will only forward them to local ACs (CE in the example) and not back to the EVPN-Overlay network.</t> </section> <sectiontitle="Impactanchor="sect-4.4.4" numbered="true" toc="default"> <name>Impact on MAC Mobilityprocedures" anchor="sect-4.4.4"><t>Procedures</name> <t> MAC Mobility procedures described in <xreftarget="RFC7432"/>target="RFC7432" format="default"/> are not modified by this document.</t> <t> Note that an intra-DC MAC move still leaves the MAC attached to the same I-ES, so under the rules of <xreftarget="RFC7432"/>target="RFC7432" format="default"/>, this is not considered a MACmobilityMobility event. Only when the MAC moves from the WAN domain to the DC domain (or from one DC to another) will the MACwillbe learned from a differentESES, and the MAC Mobility procedures will kick in.</t> <t> Thesticky bitsticky-bit indication in the MAC Mobility extended communityMUST<bcp14>MUST</bcp14> be propagated between domains.</t> </section> <sectiontitle="Gateway optimizations" anchor="sect-4.4.5"><t>anchor="sect-4.4.5" numbered="true" toc="default"> <name>Gateway Optimizations</name> <t> All the Gateway optimizations described insection 3.5 MAY<xref target="sect-3.5"/> <bcp14>MAY</bcp14> be applied to the GWs when theInterconnectinterconnect is based on EVPN-MPLS.</t> <t> In particular, the use of the Unknown MAC Route, as described insection 3.5.1,<xref target="sect-3.5.1"/>, solves some transientpacket duplicationpacket-duplication issues in cases of all-activemulti-homing,multihoming, as explained below.</t> <t> Consider the diagram inFigure 2<xref target="fig-2"/> for EVPN-MPLSInterconnectinterconnect and all-activemulti-homing,multihoming, and the following sequence:</t><t><list style="format (%c)"> <t>MAC<ol spacing="normal" type="(%c)"> <li>MAC Address M1 is advertised from NVE3 inEVI-1.</t> <t>GW3EVI-1.</li> <li>GW3 and GW4 learn M1 for EVI-1 and re-advertise M1 to the WAN with I-ESI-2 in the ESIfield.</t> <t>GW1field.</li> <li>GW1 and GW2 learn M1 and install GW3/GW4 asnext-hopsnext hops following the EVPN aliasingprocedures.</t> <t>Beforeprocedures.</li> <li>Before NVE1 learns M1, a packet arrives at NVE1 with destination M1. If the Unknown MAC Route had not been advertised into the DC, NVE1 would have flooded the packet throughout the DC, in particular to both GW1 and GW2. If the same VNI/VSID is used for both known unicast and BUM traffic, as is typically the case, there is no indication in the packet that it is a BUMpacketpacket, and both GW1 and GW2 would have forwarded it, creating packet duplication. However, because the Unknown MAC Route had been advertised into the DC, NVE1 will unicast the packet to either GW1 orGW2.</t> <t>SinceGW2.</li> <li>Since both GW1 and GW2 know M1, the GW receiving the packet will forward it to either GW3 orGW4.</t> </list> </t>GW4.</li> </ol> </section> <sectiontitle="Benefitsanchor="sect-4.4.6" numbered="true" toc="default"> <name>Benefits of the EVPN-MPLS Interconnectsolution" anchor="sect-4.4.6"><t>Solution</name> <t> The<xref target="I-D.ietf-bess-evpn-overlay"/>"DCI using ASBRs" solution described in <xref target="RFC8365" format="default"/> and the GW solution with EVPN-MPLSInterconnectinterconnect may be seensimilaras similar, since they both retain the EVPN attributes between Data Centers and throughout the WAN.HoweverHowever, the EVPN-MPLSInterconnectinterconnect solution on the GWs has significant benefits compared to the "DCI using ASBRs" solution:</t><t><list style="symbols"> <t>As<ul spacing="normal"> <li>As in any of the described GW models, this solution supports the connectivity of local attachment circuits on the GWs. This is not possible in a "DCI using ASBRs"solution.</t> <t>Differentsolution.</li> <li>Different data plane encapsulations can be supported in the DC and the WAN, while a uniform encapsulation is needed in the "DCI using ASBRs"solution.</t> <t>Optimizedsolution.</li> <li>Optimized multicast solution, with independent inclusive multicast trees in DC andWAN.</t> <t>MPLS LabelWAN.</li> <li>MPLS label aggregation:forFor the case where MPLS labels are signaled from the NVEs for MAC/IPAdvertisementadvertisement routes, this solution provides label aggregation. A remote PEMAY<bcp14>MAY</bcp14> receive a single label per GWMAC-VRFMAC-VRF, as opposed to a label per NVE/MAC-VRF connected to the GW MAC-VRF. For instance, inFigure 2,<xref target="fig-2"/>, PE would receive only one label for all the routes advertised for a given MAC-VRF from GW1, as opposed to a label perNVE/MAC-VRF.</t> <t>TheNVE/MAC-VRF.</li> <li>The GW will not propagate MACmobilityMobility for the MACs moving within a DC. Mobility intra-DC is solved by all the NVEs in the DC. The MAC Mobility procedures on the GWs are only required in case of mobility acrossDCs.</t> <t>Proxy-ARP/NDDCs.</li> <li>Proxy-ARP/ND function on the DC GWs can be leveraged to reduce ARP/ND flooding in the DC or/andintheWAN.</t> </list> </t>WAN.</li> </ul> </section> </section> <sectiontitle="PBB-EVPNanchor="sect-4.5" numbered="true" toc="default"> <name>PBB-EVPN Interconnect for EVPN-Overlaynetworks" anchor="sect-4.5"><t>Networks</name> <t> PBB-EVPN <xreftarget="RFC7623"/>target="RFC7623" format="default"/> is yet anotherInterconnectinterconnect option. It requires the use of GWs where I-components and associated B-components are part of EVI instances.</t> <sectiontitle="Control/Dataanchor="sect-4.5.1" numbered="true" toc="default"> <name>Control/Data Planesetup proceduresSetup Procedures on theGWs" anchor="sect-4.5.1"><t>GWs</name> <t> EVPN will run independently in both components, the I-component MAC-VRF and B-component MAC-VRF. Compared to <xreftarget="RFC7623"/>,target="RFC7623" format="default"/>, the DCC-MACscustomer MACs (C-MACs) are no longer learned in the data plane on the GW but in the control plane through EVPN running on the I-component. Remote C-MACs coming from remote PEs are still learned in the data plane. B-MACs in the B-component will be assigned and advertised following the procedures described in <xreftarget="RFC7623"/>.</t>target="RFC7623" format="default"/>.</t> <t> An I-ES will be configured on the GWs formulti-homing,multihoming, but its I-ESI will only be used in the EVPN control plane for the I-component EVI. Nonon-reservedunreserved ESIs will be used in the control plane of the B-componentEVIEVI, as per <xreftarget="RFC7623"/>, thattarget="RFC7623" format="default"/>. That is, the I-ES will be represented to the WAN PBB-EVPN PEs using shared or dedicated B-MACs.</t> <t> The rest of the control plane procedures will follow <xreftarget="RFC7432"/>target="RFC7432" format="default"/> for the I-component EVI and <xreftarget="RFC7623"/>target="RFC7623" format="default"/> for the B-component EVI.</t> <t> From the data plane perspective, the I-component and B-component EVPN bindings established to the samefar-endfar end will becomparedcompared, and the I-componentEVPN-overlayEVPN-Overlay binding will be kept down following the rules described insection 4.3.1.</t><xref target="sect-4.3.1"/>.</t> </section> <sectiontitle="Multi-homing proceduresanchor="sect-4.5.2" numbered="true" toc="default"> <name>Multihoming Procedures on theGWs" anchor="sect-4.5.2"><t>GWs</name> <t> This model supports single-active as well as all-activemulti-homing.</t>multihoming.</t> <t> The forwarding behavior of the DF and non-DF will be changed based on the description outlined insection 4.4.3, only replacing<xref target="sect-4.4.3"/>, substituting the"WAN split-horizon-group"WAN split-horizon group for the B-component, and using <xreftarget="RFC7623"/>target="RFC7623" format="default"/> procedures for the traffic sent or received on the B-component.</t> </section> <sectiontitle="Impactanchor="sect-4.5.3" numbered="true" toc="default"> <name>Impact on MAC Mobilityprocedures" anchor="sect-4.5.3"><t>Procedures</name> <t> C-MACs learned from the B-component will be advertised in EVPN within the I-component EVI scope. If the C-MAC was previously known in the I-component database, EVPN would advertise the C-MAC with a higher sequence number, as per <xreftarget="RFC7432"/>.target="RFC7432" format="default"/>. Froma Mobilitythe perspective of Mobility and the related procedures described in <xreftarget="RFC7432"/>,target="RFC7432" format="default"/>, the C-MACs learned from the B-component are considered local.</t> </section> <sectiontitle="Gateway optimizations" anchor="sect-4.5.4"><t>anchor="sect-4.5.4" numbered="true" toc="default"> <name>Gateway Optimizations</name> <t> All the considerations explained insection 4.4.5<xref target="sect-4.4.5"/> are applicable to the PBB-EVPNInterconnectinterconnect option.</t> </section> </section> <sectiontitle="EVPN-VXLANanchor="sect-4.6" numbered="true" toc="default"> <name>EVPN-VXLAN Interconnect for EVPN-Overlaynetworks" anchor="sect-4.6"><t>Networks</name> <t> If EVPN for Overlay tunnels is supported in theWANWAN, and a GW function is required, an end-to-end EVPN solution can be deployed. While multiple Overlay tunnel combinations at the WAN and the DC are possible (MPLSoGRE,nvGRE,NVGRE, etc.), VXLAN is described here, given its popularity in the industry. This section focuses on the specific case of EVPN for VXLAN (EVPN-VXLAN hereafter) and the impact on the <xreftarget="RFC7432"/>target="RFC7432" format="default"/> procedures.</t> <t> The procedures described insection 4.4<xref target="sect-4.4"/> apply to thissectionsection, too, onlyreplacingsubstituting EVPN-MPLS for EVPN-VXLAN control plane specifics and using <xreftarget="I-D.ietf-bess-evpn-overlay"/>target="RFC8365" format="default"/> "Local Bias" procedures instead ofsection 4.4.3.<xref target="sect-4.4.3"/>. Since there are noESI-labelsESI labels in VXLAN, GWs need to rely on "Local Bias" to applysplit-horizonsplit horizon on packets generated from the I-ES and sent to the peer GW.</t> <t> Thisuse-caseuse case assumes that NVEs need to use the VNIs or VSIDs asaglobally unique identifiers within adata center,Data Center, and a Gateway needs to be employed at the edge of thedata centerData-Center network to translate the VNI or VSID when crossing the network boundaries. This GW function provides VNI andtunnel IP addresstunnel-IP-address translation. Theuse-caseuse case in which localdownstream assigneddownstream-assigned VNIs or VSIDs can be used (like MPLS labels) is described by <xreftarget="I-D.ietf-bess-evpn-overlay"/>.</t>target="RFC8365" format="default"/>.</t> <t> While VNIs are globally significant within each DC, there are two possibilities in theInterconnectinterconnect network:</t><t><list style="format (%c)"> <t>Globally<ol spacing="normal"> <li>Globally unique VNIs in theInterconnect network:interconnect network. In this case, the GWs and PEs in theInterconnectinterconnect network will agree on a common VNI for a given EVI. The RT to be used in theInterconnectinterconnect network can beauto-derivedautoderived from theagreed Interconnectagreed-upon interconnect VNI. The VNI used inside each DCMAY<bcp14>MAY</bcp14> be the same as theInterconnect VNI.</t> <t>Downstream assignedinterconnect VNI.</li> <li>Downstream-assigned VNIs in theInterconnectinterconnect network. In this case, the GWs and PEsMUST<bcp14>MUST</bcp14> use the proper RTs to import/export the EVPN routes. Note that even if the VNI is downstream assigned in theInterconnectinterconnect network, and unlike option (a), it only identifies the <Ethernet Tag, GW> pair and not the <Ethernet Tag, egress PE> pair. The VNI used inside each DCMAY<bcp14>MAY</bcp14> be the same as theInterconnectinterconnect VNI. GWsSHOULD<bcp14>SHOULD</bcp14> support multiple VNI spaces per EVI (one perInterconnectinterconnect network they are connected to).</t> </list> </t></li> </ol> <t> In both options, NVEs inside a DC only have to be aware of a single VNI space, and only GWs will handle the complexity of managing multiple VNI spaces. In addition to VNI translation above, the GWs will provide translation of the tunnel source IP for the packets generated from the NVEs, using their own IP address. GWs will use that IP address as the BGPnext-hopnext hop in all the EVPN updates to theInterconnectinterconnect network.</t> <t> The following sections provide more details about these two options.</t> <sectiontitle="Globally uniqueanchor="sect-4.6.1" numbered="true" toc="default"> <name>Globally Unique VNIs in the Interconnectnetwork" anchor="sect-4.6.1"><t>Network</name> <t> ConsideringFigure 2,<xref target="fig-2"/>, if a host H1 in NVO-1 needs to communicate with a host H2 in NVO-2, and assuming that different VNIs are used in each DC for the sameEVI, e.g.EVI (e.g., VNI-10 in NVO-1 and VNI-20 inNVO-2,NVO-2), then the VNIsMUST<bcp14>MUST</bcp14> be translated to a commonInterconnectinterconnect VNI(e.g.(e.g., VNI-100) on the GWs. Each GW is provisioned with a VNI translation mapping so that it can translate the VNI in the control plane when sending BGP EVPN route updates to theInterconnectinterconnect network. In other words, GW1 and GW2MUST<bcp14>MUST</bcp14> be configured to map VNI-10 to VNI-100 in the BGP update messages for H1's MAC route. This mapping is also used to translate the VNI in the data plane in bothdirections,directions: that is,VNI- 10VNI-10 to VNI-100 when the packet is received from NVO-1 and the reverse mapping from VNI-100 to VNI-10 when the packet is received from the remote NVO-2 network and needs to be forwarded to NVO-1.</t> <t> The procedures described insection 4.4<xref target="sect-4.4"/> will be followed, considering that the VNIs advertised/received by the GWs will be translated accordingly.</t> </section> <sectiontitle="Downstream assignedanchor="sect-4.6.2" numbered="true" toc="default"> <name>Downstream-Assigned VNIs in the Interconnectnetwork" anchor="sect-4.6.2"><t>Network</name> <t> In this case, if a host H1 in NVO-1 needs to communicate with a host H2 in NVO-2, and assuming that different VNIs are used in each DC for the same EVI,e.g.e.g., VNI-10 in NVO-1 and VNI-20 in NVO-2, then the VNIsMUST<bcp14>MUST</bcp14> be translated as insection 4.6.1.<xref target="sect-4.6.1"/>. However, in this case, there is no need to translate to a commonInterconnectinterconnect VNI on the GWs. Each GW can translate the VNI received in an EVPN update to a locally assigned VNI advertised to theInterconnectinterconnect network. Each GW can use a differentInterconnect VNI, henceinterconnect VNI; hence, this VNI does not need to be agreed upon on all the GWs and PEs of theInterconnectinterconnect network.</t> <t> The procedures described insection 4.4<xref target="sect-4.4"/> will be followed, taking into account the considerations above for the VNI translation.</t> </section> </section> </section> <sectiontitle="Security Considerations" anchor="sect-5"><t>anchor="sect-5" numbered="true" toc="default"> <name>Security Considerations</name> <t> This document applies existing specifications to a number ofInterconnectinterconnect models. TheSecurity Considerationssecurity considerations included in those documents, such as <xreftarget="RFC7432"/>,target="RFC7432" format="default"/>, <xreftarget="I-D.ietf-bess-evpn-overlay"/>,target="RFC8365" format="default"/>, <xreftarget="RFC7623"/>,target="RFC7623" format="default"/>, <xreftarget="RFC4761"/>target="RFC4761" format="default"/>, and <xreftarget="RFC4762"/>target="RFC4762" format="default"/> apply to this document whenever those technologies are used.</t> <t> As discussed, <xreftarget="I-D.ietf-bess-evpn-overlay"/>target="RFC8365" format="default"/> discusses two main DCI solution groups: "DCI using GWs" and "DCI using ASBRs". This document specifies the solutions that correspond to the "DCI using GWs" group. It is important to note that the use of GWsprovideprovides a superior level of security on aper tenantper-tenant basis, compared to the use of ASBRs. This is due to the fact that GWs need to perform a MAC lookup on the frames being received from the WAN, and they apply security procedures, such as filtering of undesired frames, filtering of frames with a source MAC that matches a protected MAC in theDCDC, or application ofMAC duplicationMAC-duplication procedures defined in <xreftarget="RFC7432"/>.target="RFC7432" format="default"/>. OnASBRsASBRs, though, traffic is forwarded based on a label or VNIswapswap, and there is usually no visibility of the encapsulated frames, which can carry malicious traffic.</t> <t> In addition, the GW optimizations specified in thisdocument,document provide additional protection of the DCTenant Systems.tenant systems. For instance, theMAC addressMAC-address advertisement control and Unknown MAC Route defined insection 3.5.1<xref target="sect-3.5.1"/> protect the DC NVEs from being overwhelmed with an excessive number of MAC/IP routes being learned on the GWs from the WAN. The ARP/ND flooding control described in3.5.2<xref target="sect-3.5.2"/> can reduce/suppress broadcast storms being injected from the WAN.</t> <t> Finally, the reader should be aware of the potential security implications of designing a DCI with theDecoupled Interconnectdecoupled interconnect solution(section 3)(<xref target="sect-3"/>) or theIntegrated Interconnectintegrated interconnect solution(section 4).(<xref target="sect-4"/>). In theDecoupled Interconnect solutiondecoupled interconnect solution, the DC is typically easier to protect from the WAN, since each GW has a single logical link to one WAN PE, whereas in the Integrated solution, the GW has logical links to all the WAN PEs that are attached to the tenant. In either model, proper control plane and data plane policies should be put in place in the GWs in order to protect the DC from potential attacks coming from the WAN.</t> </section> <sectiontitle="IANA Considerations" anchor="sect-6"><t>anchor="sect-6" numbered="true" toc="default"> <name>IANA Considerations</name> <t> This document has no IANA actions.</t> </section> </middle> <back><references title="Normative References"> &RFC4761; &RFC4762; &RFC6074; &RFC7041; &RFC7432; &RFC2119; &RFC8174; &I-D.ietf-idr-tunnel-encaps; &RFC7623; &I-D.ietf-bess-evpn-overlay; &RFC7543;<displayreference target="I-D.ietf-bess-evpn-virtual-eth-segment" to="VIRTUAL-ES"/> <references> <name>References</name> <references> <name>Normative References</name> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4761.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4762.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6074.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7041.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7432.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/> <!-- draft-ietf-idr-tunnel-encaps-15 is now RFC 9012--> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.9012.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7623.xml"/> <!-- [I-D.ietf-bess-evpn-overlay] Published as RFC 8365 --> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8365.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7543.xml"/> </references><references title="Informative References"> &RFC4684; &RFC7348; &RFC7637; &RFC4023;<references> <name>Informative References</name> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4684.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7348.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7637.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4023.xml"/> <referenceanchor="Y.1731"><front>anchor="Y.1731"> <front> <title>OAM functions and mechanisms for Ethernet based networks</title> <author><organization>ITU-T Recommendation Y.1731</organization><organization>ITU-T</organization> </author> <datemonth="July" year="2011"/>month="August" year="2019"/> </front> <seriesInfo name="ITU-T Recommendation" value="Y.1731" /> </reference> <referenceanchor="IEEE.802.1AG"><front>anchor="IEEE.802.1AG"> <front> <title>IEEE Standard for Local and Metropolitan Area Networks-Virtual Bridged Local Area Networks Amendment 5: Connectivity Fault Management</title> <author><organization>IEEE 802.1AG_2007</organization><organization>IEEE</organization> </author> <date month="January" year="2008"/> </front> <seriesInfo name="IEEE standard" value="802.1ag-2007"/> </reference> <referenceanchor="IEEE.802.1Q-2014"><front>anchor="IEEE.802.1Q"> <front> <title>IEEE Standard for Local and metropolitan area networks--Bridges and Bridged Networks</title> <author><organization>IEEE 802.1Q-2014</organization><organization>IEEE</organization> </author> <date month="December" year="2014"/> </front> <seriesInfo name="IEEE standard" value="802.1Q-2014" /> <seriesInfo name="DOI" value="10.1109/IEEESTD.2014.6991462"/> </reference>&RFC6870; &RFC3031; &I-D.sajassi-bess-evpn-virtual-eth-segment;<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6870.xml"/> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3031.xml"/> <!-- [I-D.sajassi-bess-evpn-virtual-eth-segment] Replaced by draft-ietf-bess-evpn-virtual-eth-segment; IESG state I-D Exists --> <xi:include href="https://datatracker.ietf.org/doc/bibxml3/reference.I-D.ietf-bess-evpn-virtual-eth-segment.xml"/> </references> </references> <sectiontitle="Acknowledgments" anchor="sect-8"><t>anchor="sect-8" numbered="false" toc="default"> <name>Acknowledgments</name> <t> The authors would like to thankNeil Hart, Vinod Prabhu and Kiran Nagaraj<contact fullname="Neil Hart"/>, <contact fullname="Vinod Prabhu"/>, and <contact fullname="Kiran Nagaraj"/> for their valuable comments and feedback. We would also like to thankMartin Vigoureux and Alvaro Retana<contact fullname="Martin Vigoureux"/> and <contact fullname="Alvaro Retana"/> forhistheir detailedreviewreviews and comments.</t> </section> <sectiontitle="Contributors" anchor="sect-9"><t>anchor="sect-9" numbered="false" toc="default"> <name>Contributors</name> <t> In addition to the authors listed on the front page, the followingco-authorscoauthors have also contributed to this document:</t><figure><artwork><![CDATA[ Ravi Shekhar Anil Lohiya Wen Lin Juniper Networks Florin Balus Patrice Brissette Cisco Senad Palislamovic Nokia Dennis Cai Alibaba ]]></artwork> </figure><contact fullname="Ravi Shekhar"> <organization>Juniper Networks</organization> </contact> <contact fullname="Anil Lohiya"> <organization>Juniper Networks</organization> </contact> <contact fullname="Wen Lin"> <organization>Juniper Networks</organization> </contact> <contact fullname="Florin Balus"> <organization>Cisco</organization> </contact> <contact fullname="Patrice Brissette"> <organization>Cisco</organization> </contact> <contact fullname="Senad Palislamovic"> <organization>Nokia</organization> </contact> <contact fullname="Dennis Cai"> <organization>Alibaba</organization> </contact> </section> </back> </rfc>