idnits 2.17.1 draft-ietf-bess-evpn-etree-06.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (June 10, 2016) is 2877 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) No issues found here. Summary: 0 errors (**), 0 flaws (~~), 1 warning (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 L2VPN Workgroup A. Sajassi, Ed. 3 INTERNET-DRAFT S. Salam 4 Intended Status: Standards Track Cisco 5 J. Drake 6 Juniper 7 J. Uttaro 8 ATT 9 S. Boutros 10 VMware 11 J. Rabadan 12 Nokia 14 Expires: December 10, 2016 June 10, 2016 16 E-TREE Support in EVPN & PBB-EVPN 17 draft-ietf-bess-evpn-etree-06 19 Abstract 21 The Metro Ethernet Forum (MEF) has defined a rooted-multipoint 22 Ethernet service known as Ethernet Tree (E-Tree). A solution 23 framework for supporting this service in MPLS networks is proposed in 24 and RFC called "A Framework for E-Tree Service over MPLS Network". 25 This document discusses how those functional requirements can be 26 easily met with (PBB-)EVPN and how (PBB-)EVPN offers a more efficient 27 implementation of these functions. 29 Status of this Memo 31 This Internet-Draft is submitted to IETF in full conformance with the 32 provisions of BCP 78 and BCP 79. 34 Internet-Drafts are working documents of the Internet Engineering 35 Task Force (IETF), its areas, and its working groups. Note that 36 other groups may also distribute working documents as 37 Internet-Drafts. 39 Internet-Drafts are draft documents valid for a maximum of six months 40 and may be updated, replaced, or obsoleted by other documents at any 41 time. It is inappropriate to use Internet-Drafts as reference 42 material or to cite them other than as "work in progress." 44 The list of current Internet-Drafts can be accessed at 45 http://www.ietf.org/1id-abstracts.html 46 The list of Internet-Draft Shadow Directories can be accessed at 47 http://www.ietf.org/shadow.html 49 Copyright and License Notice 51 Copyright (c) 2016 IETF Trust and the persons identified as the 52 document authors. All rights reserved. 54 This document is subject to BCP 78 and the IETF Trust's Legal 55 Provisions Relating to IETF Documents 56 (http://trustee.ietf.org/license-info) in effect on the date of 57 publication of this document. Please review these documents 58 carefully, as they describe your rights and restrictions with respect 59 to this document. Code Components extracted from this document must 60 include Simplified BSD License text as described in Section 4.e of 61 the Trust Legal Provisions and are provided without warranty as 62 described in the Simplified BSD License. 64 Table of Contents 66 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 67 1.1 Terminology . . . . . . . . . . . . . . . . . . . . . . . . 4 68 2 E-Tree Scenarios and EVPN / PBB-EVPN Support . . . . . . . . . 4 69 2.1 Scenario 1: Leaf OR Root site(s) per PE . . . . . . . . . . 4 70 2.2 Scenario 2: Leaf OR Root site(s) per AC . . . . . . . . . . 5 71 2.3 Scenario 3: Leaf OR Root site(s) per MAC . . . . . . . . . . 6 72 3 Operation for EVPN . . . . . . . . . . . . . . . . . . . . . . . 7 73 3.1 Known Unicast Traffic . . . . . . . . . . . . . . . . . . . 7 74 3.2 BUM Traffic . . . . . . . . . . . . . . . . . . . . . . . . 8 75 3.2.1 BUM traffic originated from a single-homed site on a 76 leaf AC . . . . . . . . . . . . . . . . . . . . . . . . 9 77 3.2.2 BUM traffic originated from a single-homed site on a 78 root AC . . . . . . . . . . . . . . . . . . . . . . . . 9 79 3.2.3 BUM traffic originated from a multi-homed site on a 80 leaf AC . . . . . . . . . . . . . . . . . . . . . . . . 9 81 3.2.4 BUM traffic originated from a multi-homed site on a 82 root AC . . . . . . . . . . . . . . . . . . . . . . . . 9 83 3.3 E-TREE Traffic Flows for EVPN . . . . . . . . . . . . . . . 10 84 3.3.1 E-Tree with MAC Learning . . . . . . . . . . . . . . . . 10 85 3.3.2 E-Tree without MAC Learning . . . . . . . . . . . . . . 11 86 4 Operation for PBB-EVPN . . . . . . . . . . . . . . . . . . . . . 11 87 4.1 Known Unicast Traffic . . . . . . . . . . . . . . . . . . . 12 88 4.2 BUM Traffic . . . . . . . . . . . . . . . . . . . . . . . . 12 89 4.3 E-Tree without MAC Learning . . . . . . . . . . . . . . . . 13 90 5 BGP Encoding . . . . . . . . . . . . . . . . . . . . . . . . . . 13 91 5.1 E-TREE Extended Community . . . . . . . . . . . . . . . . . 13 92 5.2 PMSI Tunnel Attribute . . . . . . . . . . . . . . . . . . . 14 93 6 Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . 15 94 7 Security Considerations . . . . . . . . . . . . . . . . . . . . 15 95 8 IANA Considerations . . . . . . . . . . . . . . . . . . . . . . 15 96 9 References . . . . . . . . . . . . . . . . . . . . . . . . . . 15 97 9.1 Normative References . . . . . . . . . . . . . . . . . . . 15 98 9.2 Informative References . . . . . . . . . . . . . . . . . . 15 99 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 100 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 16 102 1 Introduction 104 The Metro Ethernet Forum (MEF) has defined a rooted-multipoint 105 Ethernet service known as Ethernet Tree (E-Tree). In an E-Tree 106 service, endpoints are labeled as either Root or Leaf sites. Root 107 sites can communicate with all other sites. Leaf sites can 108 communicate with Root sites but not with other Leaf sites. 110 [RFC7387] proposes the solution framework for supporting E-Tree 111 service in MPLS networks. The document identifies the functional 112 components of the overall solution to emulate E-Tree services in 113 addition to Ethernet LAN (E-LAN) services on an existing MPLS 114 network. 116 [RFC7432] is a solution for multipoint L2VPN services, with advanced 117 multi-homing capabilities, using BGP for distributing customer/client 118 MAC address reach-ability information over the MPLS/IP network. 119 [RFC7623] combines the functionality of EVPN with [802.1ah] Provider 120 Backbone Bridging for MAC address scalability. 122 This document discusses how the functional requirements for E-Tree 123 service can be easily met with (PBB-)EVPN and how (PBB-)EVPN offers a 124 more efficient implementation of these functions. 126 1.1 Terminology 128 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 129 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 130 document are to be interpreted as described in RFC 2119 [KEYWORDS]. 132 2 E-Tree Scenarios and EVPN / PBB-EVPN Support 134 In this section, we will categorize support for E-Tree into three 135 different scenarios, depending on the nature of the site association 136 (Root/Leaf) per PE or per Ethernet Segment: 138 - Leaf OR Root site(s) per PE 140 - Leaf OR Root site(s) per AC 142 - Leaf OR Root site(s) per MAC 144 2.1 Scenario 1: Leaf OR Root site(s) per PE 146 In this scenario, a PE may receive traffic from either Root sites OR 147 Leaf sites for a given MAC-VRF/bridge table, but not both 148 concurrently. In other words, a given EVI on a PE is either 149 associated with a root or leaf. The PE may have both Root and Leaf 150 sites albeit for different EVIs. 152 +---------+ +---------+ 153 | PE1 | | PE2 | 154 +---+ | +---+ | +------+ | +---+ | +---+ 155 |CE1+---ES1----+--+ | | | MPLS | | | +--+----ES2-----+CE2| 156 +---+ (Root) | |MAC| | | /IP | | |MAC| | (Leaf) +---+ 157 | |VRF| | | | | |VRF| | 158 | | | | | | | | | | +---+ 159 | | | | | | | | +--+----ES3-----+CE3| 160 | +---+ | +------+ | +---+ | (Leaf) +---+ 161 +---------+ +---------+ 163 Figure 1: Scenario 1 165 In such scenario, an EVPN PE implementation MAY provide E-TREE 166 service using topology constraint among the PEs belonging to the same 167 EVI. The purpose of this topology constraint is to avoid having PEs 168 with only Leaf sites importing and processing BGP MAC routes from 169 each other. To support such topology constrain in EVPN, two BGP 170 Route-Targets (RTs) are used for every EVPN Instance (EVI): one RT is 171 associated with the Root sites and the other is associated with the 172 Leaf sites. On a per EVI basis, every PE exports the single RT 173 associated with its type of site(s). Furthermore, a PE with Root 174 site(s) imports both Root and Leaf RTs, whereas a PE with Leaf 175 site(s) only imports the Root RT. If the number of EVIs is very large 176 (e.g., more than 64K), then RT type 0 as defined in [RFC4360] SHOULD 177 be used; otherwise, RT type 2 is sufficient [RFC7153]. 179 2.2 Scenario 2: Leaf OR Root site(s) per AC 181 In this scenario, a PE receives traffic from either Root OR Leaf 182 sites (but not both) on a given Attachment Circuit (AC) of an EVI. In 183 other words, an AC (ES or ES/VLAN) is either associated with a Root 184 or Leaf (but not both). 186 +---------+ +---------+ 187 | PE1 | | PE2 | 188 +---+ | +---+ | +------+ | +---+ | +---+ 189 |CE1+-----ES1----+--+ | | | | | | +--+---ES2/AC1--+CE2| 190 +---+ (Leaf) | |MAC| | | MPLS | | |MAC| | (Leaf) +---+ 191 | |VRF| | | /IP | | |VRF| | 192 | | | | | | | | | | +---+ 193 | | | | | | | | +--+---ES2/AC2--+CE3| 194 | +---+ | +------+ | +---+ | (Root) +---+ 195 +---------+ +---------+ 197 Figure 2: Scenario 2 199 In this scenario, if there are PEs with only root (or leaf) sites per 200 EVI, then the RT constrain procedures described in section 2.1 can 201 also be used here. However, when a Root site is added to a Leaf PE, 202 then that PE needs to process MAC routes from all other Leaf PEs and 203 add them to its forwarding table. For this scenario, if for a given 204 EVI, the majority of PEs will eventually have both Leaf and Root 205 sites attached, even though they may start as Root-only or Leaf-only 206 PEs, then it is recommended to use a single RT per EVI and avoid 207 additional configuration and operational overhead. 209 2.3 Scenario 3: Leaf OR Root site(s) per MAC 211 In this scenario, a PE may receive traffic from both Root AND Leaf 212 sites on a given Attachment Circuit (AC) of an EVI. Since an 213 Attachment Circuit (ES or ES/VLAN) carries traffic from both Root and 214 Leaf sites, the granularity at which Root or Leaf sites are 215 identifies is on a per MAC address. This scenario is considered in 216 this draft for EVPN service with only known unicast traffic - i.e., 217 there is no BUM traffic. 219 +---------+ +---------+ 220 | PE1 | | PE2 | 221 +---+ | +---+ | +------+ | +---+ | +---+ 222 |CE1+-----ES1----+--+ | | | | | | +--+---ES2/AC1--+CE2| 223 +---+ (Root) | | E | | | MPLS | | | E | | (Leaf/Root)+---+ 224 | | V | | | /IP | | | V | | 225 | | I | | | | | | I | | +---+ 226 | | | | | | | | +--+---ES2/AC2--+CE3| 227 | +---+ | +------+ | +---+ | (Leaf) +---+ 228 +---------+ +---------+ 230 Figure 3: Scenario 3 232 3 Operation for EVPN 234 [RFC7432] defines the notion of ESI MPLS label used for split-horizon 235 filtering of BUM traffic at the egress PE. Such egress filtering 236 capabilities can be leveraged in provision of E-TREE services as seen 237 shortly. In other words, [RFC7432] has inherent capability to support 238 E-TREE services without defining any new BGP routes but by just 239 defining a new BGP Extended Community for leaf indication as shown 240 later in this document. 242 3.1 Known Unicast Traffic 244 Since in EVPN, MAC learning is performed in control plane via 245 advertisement of BGP routes, the filtering needed by E-TREE service 246 for known unicast traffic can be performed at the ingress PE, thus 247 providing very efficient filtering and avoiding sending known unicast 248 traffic over MPLS/IP core to be filtered at the egress PE as done in 249 traditional E-TREE solutions (e.g., E-TREE for VPLS). 251 To provide such ingress filtering for known unicast traffic, a PE 252 MUST indicate to other PEs what kind of sites (root or leaf) its MAC 253 addresses are associated with by advertising a leaf indication flag 254 (via an Extended Community) along with each of its MAC/IP 255 Advertisement route. The lack of such flag indicates that the MAC 256 address is associated with a root site. This scheme applies to all 257 scenarios described in section 2. 259 Furthermore, for multi-homing scenario of section 2.2, where an AC is 260 either root or leaf (but not both), the PE MAY advertise leaf 261 indication along with the Ethernet A-D per EVI route. This 262 advertisement is used for sanity checking in control-plane to ensure 263 that there is no discrepancy in configuration among different PEs of 264 the same redundancy group. For example, if a leaf site is multi-homed 265 to PE1 an PE2, and PE1 advertises the Ethernet A-D per EVI 266 corresponding to this leaf site with the leaf-indication flag but PE2 267 does not, then the receiving PE notifies the operator of such 268 discrepancy and ignore the leaf-indication flag on PE1. In other 269 words, in case of discrepancy, the multi-homing for that pair of PEs 270 is assumed to be in default "root" mode for that or . The leaf indication flag on Ethernet A-D per EVI route 272 tells the receiving PEs that all MAC addresses associated with this 273 or are from a leaf site. Therefore, if a 274 PE receives a leaf indication for an AC via the Ethernet A-D per EVI 275 route but doesn't receive a leaf indication in the corresponding MAC 276 route, then it notify the operator and ignore the leaf indication on 277 the Ethernet A-D per EVI route. 279 Tagging MAC addresses with a leaf indication enables remote PEs to 280 perform ingress filtering for known unicast traffic - i.e., on the 281 ingress PE, the MAC destination address lookup yields, in addition to 282 the forwarding adjacency, a flag which indicates whether the target 283 MAC is associated with a Leaf site or not. The ingress PE cross- 284 checks this flag with the status of the originating AC, and if both 285 are Leafs, then the packet is not forwarded. 287 To support the above ingress filtering functionality, a new E-TREE 288 Extended Community with a Leaf indication flag is introduced [section 289 5.2]. This new Extended Community MUST be advertised with MAC/IP 290 Advertisement route and MAY be advertised with an Ethernet A-D per 291 EVI route as described above. 293 3.2 BUM Traffic 295 For BUM traffic, it is not possible to perform filtering on the 296 ingress PE, as is the case with known unicast, because of the multi- 297 destination nature of the traffic. As such, the solution relies on 298 egress filtering. In order to apply the proper egress filtering, 299 which varies based on whether a packet is sent from a Leaf AC or a 300 root AC, the MPLS-encapsulated frames MUST be tagged with an 301 indication when they originated from a Leaf AC. In other words, leaf 302 indication for BUM traffic is done at the granularity of AC. This can 303 be achieved in EVPN through the use of a MPLS label where it can be 304 used to either identify the Ethernet segment of origin per [RFC7432] 305 (i.e., ESI label) or it can be used to indicate that the packet is 306 originated from a leaf site (Leaf label). 308 BUM traffic sent over a P2MP LSP or ingress replication, may need to 309 carry an upstream assigned or downstream assigned MPLS label 310 (respectively) for the purpose of egress filtering to indicate to the 311 egress PEs whether this packet is originated from a leaf AC. 313 The main difference between downstream and upstream assigned MPLS 314 label is that in case of downstream assigned not all egress PE 315 devices need to receive the label just like ingress replication 316 procedures defined in [RFC7432]. 318 There are four scenarios to consider as follow. In all these 319 scenarios, the imposition PE imposes the right MPLS label associated 320 with the originated Ethernet Segment (ES) depending on whether the 321 Ethernet frame originated from a Root or a Leaf site on that Ethernet 322 Segment (ESI or Leaf label). The mechanism by which the PE identifies 323 whether a given frame originated from a Root or a Leaf site on the 324 segment is based on the Ethernet Tag associated with the frame (e.g., 325 whether the frame received on a leaf or a root AC). Other mechanisms 326 for identifying whether an ingress AC is a root or leaf is beyond the 327 scope of this document. 329 3.2.1 BUM traffic originated from a single-homed site on a leaf AC 331 In this scenario, the ingress PE adds a special MPLS label indicating 332 a Leaf site. This special Leaf MPLS label, used for single-homing 333 scenarios, is not on a per ES basis but rather on a per PE basis - 334 i.e., a single Leaf MPLS label is used for all single-homed ES's on 335 that PE. This Leaf label is advertised to other PE devices, using a 336 new EVPN Extended Community called E-TREE Extended Community (section 337 5.1) along with an Ethernet A-D per ES route with ESI of zero and a 338 set of Route Targets (RTs) corresponding to all EVIs on the PE with 339 at least one leaf site per EVI. The set of Ethernet A-D per ES routes 340 may be needed if the number of Route Targets (RTs) that need to be 341 sent exceed the limit on a single route per [RFC7432]. The ESI for 342 the Ethernet A-D per ES route is set to zero to indicate single-homed 343 sites. 345 When a PE receives this special Leaf label in the data path, it 346 blocks the packet if the destination AC is of type Leaf; otherwise, 347 it forwards the packet. 349 3.2.2 BUM traffic originated from a single-homed site on a root AC 351 In this scenario, the ingress PE does not add any ESI or Leaf label 352 and it operates per [RFC7432] procedures. 354 3.2.3 BUM traffic originated from a multi-homed site on a leaf AC 356 In this scenario, it is assumed that While different ACs (VLANs) on 357 the same ES could have different root/leaf designation (some being 358 roots and some being leaves), the same VLAN does have the same 359 root/leaf designation on all PEs on the same ES. Furthermore, it is 360 assumed that there is no forwarding among subnets - ie, the service 361 is EVPN L2 and not EVPN IRB. IRB use case is outside the scope of 362 this document. 364 In such scenarios, If a multicast packet is originated from a leaf 365 AC, then it only needs to carry Leaf label described in section 366 3.2.1. This label is sufficient in providing the necessary egress 367 filtering of BUM traffic from getting sent to leaf ACs including the 368 leaf AC on the same Ethernet Segment. 370 3.2.4 BUM traffic originated from a multi-homed site on a root AC 372 In this scenario, both the ingress and egress PE devices follows the 373 procedure defined in [RFC7432] for adding and/or processing an ESI 374 MPLS label. 376 3.3 E-TREE Traffic Flows for EVPN 378 Per [RFC7387], a generic E-Tree service supports all of the following 379 traffic flows: 381 - Ethernet Unicast from Root to Roots & Leaf 382 - Ethernet Unicast from Leaf to Root 383 - Ethernet Broadcast/Multicast from Root to Roots & Leafs 384 - Ethernet Broadcast/Multicast from Leaf to Roots 386 A particular E-Tree service may need to support all of the above 387 types of flows or only a select subset, depending on the target 388 application. In the case where unicast flows need not be supported, 389 the L2VPN PEs can avoid performing any MAC learning function. 391 In the subsections that follow, we will describe the operation of 392 EVPN to support E-Tree service with and without MAC learning. 394 3.3.1 E-Tree with MAC Learning 396 The PEs implementing an E-Tree service must perform MAC learning when 397 unicast traffic flows must be supported among Root and Leaf sites. In 398 this case, the PE with Root sites performs MAC learning in the data- 399 path over the Ethernet Segments, and advertises reachability in EVPN 400 MAC Advertisement routes. These routes will be imported by all PEs 401 for that EVI (i.e., PEs that have Leaf sites as well as PEs that have 402 Root sites). Similarly, the PEs with Leaf sites perform MAC learning 403 in the data-path over their Ethernet Segments, and advertise 404 reachability in EVPN MAC Advertisement routes. For the scenario 405 described in section 2.1 (or possibly section 2.2), these routes are 406 imported only by PEs with at least one Root site in the EVI - i.e., a 407 PE with only Leaf sites will not import these routes. PEs with Root 408 and/or Leaf sites may use the Ethernet A-D routes for aliasing (in 409 the case of multi-homed segments) and for mass MAC withdrawal per 410 [RFC7432]. 412 To support multicast/broadcast from Root to Leaf sites, either a P2MP 413 tree rooted at the PE(s) with the Root site(s) or ingress replication 414 can be used. The multicast tunnels are set up through the exchange of 415 the EVPN Inclusive Multicast route, as defined in [RFC7432]. 417 To support multicast/broadcast from Leaf to Root sites, ingress 418 replication should be sufficient for most scenarios where there are 419 only a few Roots (typically two). Therefore, in a typical scenario, a 420 root PE needs to support both a P2MP tunnel in transmit direction 421 from itself to leaf PEs and at the same time it needs to support 422 ingress-replication tunnels in receive direction from leaf PEs to 423 itself. In order to signal this efficiently from the root PE, a new 424 composite tunnel type is defined per section 5.3. This new composite 425 tunnel type is advertised by the root PE to simultaneously indicate a 426 P2MP tunnel in transmit direction and an ingress-replication tunnel 427 in the receive direction for the BUM traffic. 429 If the number of Roots is large, P2MP tunnels originated at the PEs 430 with Leaf sites may be used and thus there will be no need to use the 431 modified PMSI tunnel attribute in section 5.2 for composite tunnel 432 type. 434 3.3.2 E-Tree without MAC Learning 436 The PEs implementing an E-Tree service need not perform MAC learning 437 when the traffic flows between Root and Leaf sites are only multicast 438 or broadcast. In this case, the PEs do not exchange EVPN MAC 439 Advertisement routes. Instead, the Inclusive Multicast Ethernet Tag 440 (IMET) routes are used to support BUM traffic. 442 The fields of the IMET route are populated per the procedures defined 443 in [RFC7432], and the multicast tunnel setup criteria are as 444 described in the previous section. 446 Just as in the previous section, if the number of PEs with root sites 447 are only a few and thus ingress replication is desired from leaf PEs 448 to these root PEs, then the modified PMSI attribute as defined in 449 section 5.3 should be used. 451 4 Operation for PBB-EVPN 453 In PBB-EVPN, the PE advertises a Root/Leaf indication along with each 454 B-MAC Advertisement route, to indicate whether the associated B-MAC 455 address corresponds to a Root or a Leaf site. Just like the EVPN 456 case, the new E-TREE Extended Community defined in section [5.1] is 457 advertised with each MAC Advertisement route. 459 In the case where a multi-homed Ethernet Segment has both Root and 460 Leaf sites attached, two B-MAC addresses are advertised: one B-MAC 461 address is per ES as specified in [RFC7623] and implicitly denoting 462 Root, and the other B-MAC address is per PE and explicitly denoting 463 Leaf. The former B-MAC address is not advertised with the E-TREE 464 extended community but the latter B-MAC denoting Leaf is advertised 465 with the new E-TREE extended community where "Leaf-indication" flag 466 is set. In such multi-homing scenarios where and Ethernet Segment has 467 both Root and Leaf ACs, it is assumed that While different ACs 468 (VLANs) on the same ES could have different root/leaf designation 469 (some being roots and some being leaves), the same VLAN does have the 470 same root/leaf designation on all PEs on the same ES. Furthermore, it 471 is assumed that there is no forwarding among subnets - ie, the 472 service is L2 and not IRB. IRB use case is outside the scope of this 473 document. 475 The ingress PE uses the right B-MAC source address depending on 476 whether the Ethernet frame originated from the Root or Leaf AC on 477 that Ethernet Segment. The mechanism by which the PE identifies 478 whether a given frame originated from a Root or Leaf site on the 479 segment is based on the Ethernet Tag associated with the frame. Other 480 mechanisms of identification, beyond the Ethernet Tag, are outside 481 the scope of this document. 483 Furthermore, a PE advertises two special global B-MAC addresses: one 484 for Root and another for Leaf, and tags the Leaf one as such in the 485 MAC Advertisement route. These B-MAC addresses are used as source 486 addresses for traffic originating from single-homed segments. The B- 487 MAC address used for indicating Leaf sites can be the same for both 488 single-homed and multi-homed segments. 490 4.1 Known Unicast Traffic 492 For known unicast traffic, the PEs perform ingress filtering: On the 493 ingress PE, the C-MAC destination address lookup yields, in addition 494 to the target B-MAC address and forwarding adjacency, a flag which 495 indicates whether the target B-MAC is associated with a Root or a 496 Leaf site. The ingress PE cross-checks this flag with the status of 497 the originating site, and if both are a Leaf, then the packet is not 498 forwarded. 500 4.2 BUM Traffic 502 For BUM traffic, the PEs must perform egress filtering. When a PE 503 receives a MAC advertisement route (which will be used as a source B- 504 MAC), it updates its Ethernet Segment egress filtering function 505 (based on the source B-MAC address), as follows: 507 - If the MAC Advertisement route indicates that the advertised B-MAC 508 is a Leaf, and the local Ethernet Segment is a Leaf as well, then the 509 source B-MAC address is added to the B-MAC filtering list. 511 - Otherwise, the B-MAC filtering list is not updated. 513 When the egress PE receives the packet, it examines the B-MAC source 514 address to check whether it should filter or forward the frame. Note 515 that this uses the same filtering logic as baseline [RFC7623] and 516 does not require any additional flags in the data-plane. 518 The PE places all Leaf Ethernet Segments of a given bridge domain in 519 a single split-horizon group in order to prevent intra-PE forwarding 520 among Leaf segments. This split-horizon function applies to BUM 521 traffic. 523 4.3 E-Tree without MAC Learning 525 In scenarios where the traffic of interest is only Multicast and/or 526 broadcast, the PEs implementing an E-Tree service do not need to do 527 any MAC learning. In such scenarios the filtering must be performed 528 on egress PEs. For PBB-EVPN, the handling of such traffic is per 529 section 4.2 without C-MAC learning part of it at both ingress and 530 egress PEs. 532 5 BGP Encoding 534 This document defines two new BGP Extended Community for EVPN. 536 5.1 E-TREE Extended Community 538 This Extended Community is a new transitive Extended Community having 539 a Type field value of 0x06 (EVPN) and the Sub-Type 0x05. It is used 540 for leaf indication of known unicast and BUM traffic. For BUM 541 traffic, the Leaf Label field is set to a valid MPLS label and this 542 EC is advertised along with Ethernet A-D per ES route with an ESI of 543 zero to enable egress filtering on disposition PEs per section 3.2.1 544 and 3.2.3. There is no need to send ESI Label Extended Community when 545 sending Ethernet A-D per ES route with an ESI of zero. For known 546 unicast traffic, the Leaf flag bit is set to one and this EC is 547 advertised along with MAC/IP Advertisement route per section 3.1. 549 The E-TREE Extended Community is encoded as an 8-octet value as 550 follows: 552 0 1 2 3 553 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 554 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 555 | Type=0x06 | Sub-Type=0x05 | Flags(1 Octet)| Reserved=0 | 556 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 557 | Reserved=0 | Leaf Label | 558 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 560 The low-order bit of the Flags octet is defined as the "Leaf- 561 Indication" bit. A value of one indicates a Leaf AC/Site. 563 When this EC is advertised along with MAC/IP Advertisement route (for 564 known unicast traffic), the Leaf-Indication flag MUST be set to one 565 and Leaf Label is set to zero. The received PE should ignore Leaf 566 Label and only processes Leaf-Indication flag. A value of zero for 567 Leaf-Indication flag is invalid when sent along with MAC/IP 568 advertisement route and an error should be logged. 570 When this EC is advertised along with Ethernet A-D per ES route (with 571 ESI of zero) for BUM traffic, the Leaf Label MUST be set to a valid 572 MPLS label and the Leaf-Indication flag should be set to zero. The 573 received PE should ignore the Leaf-Indication flag. A non-valid MPLS 574 label when sent along with the Ethernet A-D per ES route, should be 575 logged as an error. 577 5.2 PMSI Tunnel Attribute 579 [RFC6514] defines PMSI Tunnel attribute which is an optional 580 transitive attribute with the following format: 582 +---------------------------------+ 583 | Flags (1 octet) | 584 +---------------------------------+ 585 | Tunnel Type (1 octets) | 586 +---------------------------------+ 587 | MPLS Label (3 octets) | 588 +---------------------------------+ 589 | Tunnel Identifier (variable) | 590 +---------------------------------+ 592 This draft uses all the fields per existing definition except for the 593 following modifications to the Tunnel Type and Tunnel Identifier: 595 When receiver ingress-replication label is needed, the high-order bit 596 of the tunnel type field (C bit - Composite tunnel bit) is set while 597 the remaining low-order seven bits indicate the tunnel type as 598 before. When this C bit is set, the "tunnel identifier" field would 599 begin with a three-octet label, followed by the actual tunnel 600 identifier for the transmit tunnel. PEs that don't understand the 601 new meaning of the high-order bit would treat the tunnel type as an 602 invalid tunnel type. For the PEs that do understand the new meaning 603 of the high-order, if ingress replication is desired when sending BUM 604 traffic, the PE will use the the label in the Tunnel Identifier field 605 when sending its BUM traffic. 607 6 Acknowledgement 609 We would like to thank Dennis Cai, Antoni Przygienda, and Jeffrey 610 Zhang for their valuable comments. 612 7 Security Considerations 614 Since this draft uses the EVPN constructs of [RFC7432] and [RFC7623], 615 the same security considerations in these drafts are also applicable 616 here. Furthermore, this draft provides additional security check by 617 allowing sites (or ACs) of an EVPN instance to be designated as 618 "Root" or "Leaf" and preventing any traffic exchange among "Leaf" 619 sites of that VPN through ingress filtering for known unicast traffic 620 and egress filtering for BUM traffic. 622 8 IANA Considerations 624 This document requests the allocation of value 5 in the "EVPN 625 Extended Community Sub-Types" registry defined in [RFC7153] and 626 modification of the registry as follow: 628 SUB-TYPE VALUE NAME Reference 630 0x05 E-TREE Extended Community This document 631 6-255 Unassigned 633 9 References 635 9.1 Normative References 637 [KEYWORDS] Bradner, S., "Key words for use in RFCs to Indicate 638 Requirement Levels", BCP 14, RFC 2119, March 1997. 640 [RFC7432] Sajassi et al., "BGP MPLS Based Ethernet VPN", February, 641 2015. 643 [RFC7623] Sajassi et al., "Provider Backbone Bridging Combined with 644 Ethernet VPN (PBB-EVPN)", September, 2015. 646 9.2 Informative References 648 [RFC7387] Key et al., "A Framework for E-Tree Service over MPLS 649 Network", October 2014. 651 [RFC4360] S. Sangli et al, "BGP Extended Communities Attribute", 652 February, 2006. 654 [RFC7153] Rosen et al., "IANA Registries for BGP Extended 655 Communities", March, 2014. 657 [RFC6514] Aggarwal et al., "BGP Encodings and Procedures for 658 Multicast in MPLS/BGP IP VPNs", February, 2012. 660 Contributors 662 In addition to the authors listed on the front page, the following 663 co-authors have also contributed to this document: 665 Wim Henderickx 666 Nokia 668 Aldrin Isaac 669 Wen Lin 670 Juniper 672 Authors' Addresses 674 Ali Sajassi 675 Cisco 676 Email: sajassi@cisco.com 678 Samer Salam 679 Cisco 680 Email: ssalam@cisco.com 682 John Drake 683 Juniper 684 Email: jdrake@juniper.net 686 Jim Uttaro 687 AT&T 688 Email: ju1738@att.com 690 Sami Boutros 691 VMware 692 Email: sboutros@vmware.com 694 Jorge Rabadan 695 Nokia 696 Email: jorge.rabadan@nokia.com