idnits 2.17.1 draft-pfister-bier-mld-03.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 seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). -- The document date (March 9, 2017) is 2605 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(See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-08) exists of draft-ietf-bier-architecture-05 -- Obsolete informational reference (is this intentional?): RFC 2460 (Obsoleted by RFC 8200) Summary: 0 errors (**), 0 flaws (~~), 3 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group P. Pfister 3 Internet-Draft IJ. Wijnands 4 Intended status: Standards Track S. Venaas 5 Expires: September 10, 2017 Cisco Systems 6 C. Wang 7 Z. Zhang 8 ZTE Corporation 9 M. Stenberg 10 March 9, 2017 12 BIER Ingress Multicast Flow Overlay using Multicast Listener Discovery 13 Protocols 14 draft-pfister-bier-mld-03 16 Abstract 18 This document specifies the ingress part of a multicast flow overlay 19 for BIER networks. Using existing multicast listener discovery 20 protocols, it enables multicast membership information sharing from 21 egress routers, acting as listeners, toward ingress routers, acting 22 as queriers. Ingress routers keep per-egress-router state, used to 23 construct the BIER bit mask associated with IP multicast packets 24 entering the BIER domain. 26 Status of This Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at http://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on September 10, 2017. 43 Copyright Notice 45 Copyright (c) 2017 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (http://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 61 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 62 3. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 3 63 4. Applicability Statement . . . . . . . . . . . . . . . . . . . 4 64 5. Querier and Listener Specifications . . . . . . . . . . . . . 4 65 5.1. Configuration Parameters . . . . . . . . . . . . . . . . 5 66 5.2. MLDv2 instances. . . . . . . . . . . . . . . . . . . . . 5 67 5.2.1. Sending Queries . . . . . . . . . . . . . . . . . . . 6 68 5.2.2. Sending Reports . . . . . . . . . . . . . . . . . . . 6 69 5.2.3. Receiving Queries . . . . . . . . . . . . . . . . . . 6 70 5.2.4. Receiving Reports . . . . . . . . . . . . . . . . . . 7 71 5.3. Packet Forwarding . . . . . . . . . . . . . . . . . . . . 7 72 6. Security Considerations . . . . . . . . . . . . . . . . . . . 7 73 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8 74 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 8 75 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 8 76 9.1. Normative References . . . . . . . . . . . . . . . . . . 8 77 9.2. Informative References . . . . . . . . . . . . . . . . . 9 78 Appendix A. BIER Use Case in Data Centers . . . . . . . . . . . 9 79 A.1. Convention and Terminology . . . . . . . . . . . . . . . 11 80 A.2. BIER in data centers . . . . . . . . . . . . . . . . . . 11 81 A.3. A BIER MLD solution for Virtual Network information . . . 12 82 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 13 84 1. Introduction 86 The Bit Index Explicit Replication (BIER - 87 [I-D.ietf-bier-architecture]) forwarding technique enables IP 88 multicast transport across a BIER domain. When receiving or 89 originating a packet, ingress routers have to construct a bit mask 90 indicating which BIER egress routers located within the same BIER 91 domain will receive the packet. A stateless approach would consist 92 in forwarding all incoming packets toward all egress routers, which 93 would in turn make a forwarding decision based on local information. 94 But any more efficient approach would require ingress routers to keep 95 some state about egress routers multicast membership information, 96 hence requiring state sharing from egress routers toward ingress 97 routers. 99 This document specifies how to use the Multicast Listener Discovery 100 protocol version 2 [RFC3810] (resp. the Internet Group Management 101 protocol version 3 [RFC3376]) as the ingress part of a BIER multicast 102 flow overlay (BIER layering is described in 103 [I-D.ietf-bier-architecture]) for IPv6 (resp. IPv4). It enables 104 multicast membership information sharing from egress routers, acting 105 as listeners, toward ingress routers, acting as queriers. Ingress 106 routers keep per-egress-router state, used to construct the BIER bit 107 mask associated with IP multicast packets entering the BIER domain. 109 This specification is applicable to both IP version 4 and version 6. 110 It therefore specifies two separate mechanisms operating 111 independently. For the sake of simplicity, the rest of this document 112 uses IPv6 terminology. It can be applied to IPv4 by replacing 113 'MLDv2' with 'IGMPv3', and following specific requirements when 114 explicitly stated. 116 2. Terminology 118 In this document, the key words "MAY", "MUST", "MUST NOT", 119 "RECOMMENDED", and "SHOULD", are to be interpreted as described in 120 [RFC2119]. 122 The terms "Bit-Forwarding Router" (BFR), "Bit-Forwarding Egress 123 Router" (BFER), "Bit-Forwarding Ingress Router" (BFIR), "BFR-id" and 124 "BFR-Prefix" are to be interpreted as described in 125 [I-D.ietf-bier-architecture]. 127 Additionally, the following definitions are used: 129 BIER Multicast Listener Discovery (BMLD): The modified version of 130 MLD specified in this document. 132 BMLD Querier: A BFR implementing the Querier part of this 133 specification. A BMLD Node MAY be both a Querier and a Listener. 135 BMLD Listener: A BFR implementing the Listener part of this 136 specification. A BMLD Node MAY be both a Querier and a Listener. 138 3. Overview 140 This document proposes to use the mechanisms described in MLDv2 in 141 order to enable multicast membership information sharing from BFERs 142 toward BFIRs within a given BIER domain. BMLD queries (resp. 143 reports) are sent over BIER toward all BMLD Nodes (resp. BMLD 144 Queriers) using modified MLDv2 messages which IP destination is set 145 to a configured 'all BMLD Nodes' (resp. 'all BMLD Queriers') IP 146 multicast address. 148 By running MLDv2 instances with per-listener explicit tracking, BMLD 149 Queriers are able to map BMLD Listeners with MLDv2 membership states. 150 This state is then used to construct the set of BFERs associated with 151 each incoming IP multicast data packet. 153 4. Applicability Statement 155 BMLD runs on top of a BIER Layer and provides the ingress part of a 156 BIER multicast flow overlay, i.e, it specifies how BFIRs construct 157 the set of BFERs for each ingress IP multicast data packet. The BFER 158 part of the Multicast Flow Overlay is out of scope of this document. 160 The BIER Layer MUST be able to transport BMLD messages toward all 161 BMLD Queriers and Listeners. Such packets are IP multicast packets 162 with a BFR-Prefix as source address, a multicast destination address, 163 and containing a MLDv2 message. 165 BMLD only requires state to be kept by Queriers, and is therefore 166 more scalable than PIMv2 [RFC7761] in terms of overall state, but is 167 also likely to be less scalable than PIMv2 in terms of the amount of 168 control traffic and the size of the state that is kept by individual 169 routers. 171 This specification is applicable to both IP version 4 and version 6. 172 It therefore specifies two separate mechanisms operating 173 independently. For the sake of simplicity, this document uses IPv6 174 terminology. It can be applied to IPv4 by replacing 'MLDv2' with 175 'IGMPv3', and following specific requirements when explicitly stated. 177 5. Querier and Listener Specifications 179 Routers desiring to receive IP multicast traffic (e.g., for their own 180 use, or for forwarding) MUST behave as BMLD Listeners. Routers 181 receiving IP multicast traffic from outside the BIER domain, or 182 originating multicast traffic, MUST behave as BMLD Queriers. 184 BMLD Queriers (resp. BMLD Listeners) MUST act as MLDv2 Queriers 185 (resp. MLDv2 Listeners) as specified in [RFC3810] unless stated 186 otherwise in this section. 188 5.1. Configuration Parameters 190 Both Queriers and Listeners MUST operate as BFIRs and BFERs within 191 the BIER domain in order to send and receive BMLD messages. They 192 MUST therefore be configured accordingly, as specified in 193 [I-D.ietf-bier-architecture]. 195 All Listeners MUST be configured with a 'all BMLD Queriers' multicast 196 address and the BFR-ids of all the BMLD Queriers. This is used by 197 Listeners to send BMLD reports over BIER toward all Queriers. All 198 Queriers MUST be configured to accept BMLD reports sent to this 199 address. 201 All Queriers MUST be configured with a 'all BMLD Nodes' multicast 202 address and the BFR-ids of all the Queriers and Listeners. This 203 information is used by Queriers to send BMLD queries over BIER toward 204 all BMLD Nodes. All BMLD Nodes MUST be configured to accept BMLD 205 queries sent to this address. 207 Note that BMLD (unlike MLDv2) makes use of per-instance configured 208 multicast group addresses rather than well-known addresses so that 209 multiple instances of BMLD (using different group addresses) can be 210 run simultaneously within the same BIER domain. Configured group 211 addresses MAY be obtained from allocated IP prefixes using [RFC3306]. 212 One MAY choose to use the well-known MLDv2 addresses in one instance, 213 but different instances MUST use different addresses. 215 IP packets coming from outside of the BIER domain and having a 216 destination address set to the configured 'all BMLD Queriers' or the 217 'all BMLD Nodes' group address MUST be dropped. It is RECOMMENDED 218 that these configured addresses have a limited scope, enforcing this 219 behavior by scope-based filtering on BIER domain's egress interfaces. 221 5.2. MLDv2 instances. 223 BMLD Queriers MUST run a MLDv2 Querier instance with per-host 224 tracking, which means they keep track of the MLDv2 state associated 225 with each BMLD Listener. For that purpose, Listeners are identified 226 by their respective BFR-Prefix, used as IP source address in all BMLD 227 reports. 229 BMLD Listeners MUST run a MLDv2 Listener instance expressing their 230 interest in the multicast traffic they are supposed to receive for 231 local use or forwarding. 233 BMLD Listeners and Queriers MUST NOT run the MLDv1 (IGMPv2 and IGMPv1 234 for IPv4) backward compatibility procedures. 236 5.2.1. Sending Queries 238 BMLD Queries are IP packets sent over BIER by BMLD Queriers: 240 o Toward all BMLD Nodes (i.e., providing to the BIER Layer the BFR- 241 ids of all BMLD Nodes). 243 o Without the IPv6 router alert option [RFC2711] in the hop-by-hop 244 extension header [RFC2460] (or the IPv4 router alert option 245 [RFC2113] for IPv4). 247 o With the IP destination address set to the 'all BMLD Nodes' group 248 address. 250 o With the IP source address set to the BFR-Prefix of the sender. 252 o With a TTL value great enough such that the packet can be received 253 by all BMLD Nodes, depending on the underlying BIER layer (whether 254 it decrements the IP TTL or not) and the size of the network. The 255 default value is 64. 257 5.2.2. Sending Reports 259 BMLD Reports are IP packets sent over BIER by BMLD Listeners: 261 o Toward all BMLD Queriers (i.e., providing to the BIER layer the 262 BFR-ids of all BMLD Queriers). 264 o Without the IPv6 router alert option [RFC2711] in the hop-by-hop 265 extension header [RFC2460] (or the IPv4 router alert option 266 [RFC2113] for IPv4). 268 o With the IP destination address set to the 'all BMLD Queriers' 269 group address. 271 o With the IP source address set to the BFR-Prefix of the sender. 273 o With a TTL value great enough such that the packet can be received 274 by all BMLD Queriers, depending on the underlying BIER layer 275 (whether it decrements the IP TTL or not) and the size of the 276 network. The default value is 64. 278 5.2.3. Receiving Queries 280 BMLD Queriers and Listeners MUST check the destination address of all 281 the IP packets that are received or forwarded over BIER whenever 282 their own BIER bit is set in the packet. If the destination address 283 is equal to the 'all BMLD Nodes' group address the packet is 284 processed as specified in this section. 286 If the IPv6 (resp. IPv4) packet contains an ICMPv6 (resp. IGMP) 287 message of type 'Multicast Listener Query' (resp. of type 'Membership 288 Query'), it is processed by the MLDv2 (resp. IGMPv3) instance run by 289 the BMLD Querier. It MUST be dropped otherwise. 291 During the MLDv2 processing, the packet MUST NOT be checked against 292 the MLDv2 consistency conditions (i.e., the presence of the router 293 alert option, the TTL equaling 1 and, for IPv6 only, the source 294 address being link-local). 296 5.2.4. Receiving Reports 298 BMLD Queriers MUST check the destination address of all the IP 299 packets that are received or forwarded over BIER whenever their own 300 BIER bit is set. If the destination address is equal to the 'all 301 BMLD Queriers' the packet is processed as specified in this section. 303 If the IPv6 (resp. IPv4) packet contains an ICMPv6 (resp. IGMP) 304 message of type 'Multicast Listener Report Message v2' (resp. 305 'Version 3 Membership Report'), it is processed by the MLDv2 (resp. 306 IGMPv3) instance run by the BMLD Querier. It MUST be dropped 307 otherwise. 309 During the MLDv2 processing, the packet MUST NOT be checked against 310 the MLDv2 consistency conditions (i.e., the presence of the router 311 alert option, the TTL equaling 1 and, for IPv6 only, the source 312 address being link-local). 314 5.3. Packet Forwarding 316 BMLD Queriers configure the BIER Layer using the information obtained 317 using BMLD, which associates BMLD Listeners (identified by their BFR- 318 Prefixes) with their respective MLDv2 membership state. 320 More specifically, the MLDv2 state associated with each BMLD Listener 321 is provided to the BIER layer such that whenever a multicast packet 322 enters the BIER domain, if that packet matches the membership 323 information from a BMLD Listener, its BFR-id is added to the set of 324 BFR-ids the packet should be forwarded to by the BIER-Layer. 326 6. Security Considerations 328 BMLD makes use of IP MLDv2 messages transported over BIER in order to 329 configure the BIER Layer of BFIRs. BMLD messages MUST be secured, 330 either by relying on physical or link-layer security, by securing the 331 IP packets (e.g., using IPSec [RFC4301]), or by relying on security 332 features provided by the BIER Layer. 334 Whenever an attacker would be able to spoof the identity of a router, 335 it could: 337 o Redirect undesired traffic toward the spoofed router by 338 subscribing to undesired multicast traffic. 340 o Prevent desired multicast traffic from reaching the spoofed router 341 by unsubscribing to some desired multicast traffic. 343 7. IANA Considerations 345 This specification does not require any action from IANA. 347 8. Acknowledgements 349 Comments concerning this document are very welcome. 351 9. References 353 9.1. Normative References 355 [RFC2113] Katz, D., "IP Router Alert Option", RFC 2113, 356 DOI 10.17487/RFC2113, February 1997, 357 . 359 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 360 Requirement Levels", BCP 14, RFC 2119, 361 DOI 10.17487/RFC2119, March 1997, 362 . 364 [RFC3376] Cain, B., Deering, S., Kouvelas, I., Fenner, B., and A. 365 Thyagarajan, "Internet Group Management Protocol, Version 366 3", RFC 3376, DOI 10.17487/RFC3376, October 2002, 367 . 369 [RFC3810] Vida, R., Ed. and L. Costa, Ed., "Multicast Listener 370 Discovery Version 2 (MLDv2) for IPv6", RFC 3810, 371 DOI 10.17487/RFC3810, June 2004, 372 . 374 [I-D.ietf-bier-architecture] 375 Wijnands, I., Rosen, E., Dolganow, A., Przygienda, T., and 376 S. Aldrin, "Multicast using Bit Index Explicit 377 Replication", draft-ietf-bier-architecture-05 (work in 378 progress), October 2016. 380 9.2. Informative References 382 [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 383 (IPv6) Specification", RFC 2460, DOI 10.17487/RFC2460, 384 December 1998, . 386 [RFC2711] Partridge, C. and A. Jackson, "IPv6 Router Alert Option", 387 RFC 2711, DOI 10.17487/RFC2711, October 1999, 388 . 390 [RFC3306] Haberman, B. and D. Thaler, "Unicast-Prefix-based IPv6 391 Multicast Addresses", RFC 3306, DOI 10.17487/RFC3306, 392 August 2002, . 394 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 395 Internet Protocol", RFC 4301, DOI 10.17487/RFC4301, 396 December 2005, . 398 [RFC5015] Handley, M., Kouvelas, I., Speakman, T., and L. Vicisano, 399 "Bidirectional Protocol Independent Multicast (BIDIR- 400 PIM)", RFC 5015, DOI 10.17487/RFC5015, October 2007, 401 . 403 [RFC7348] Mahalingam, M., Dutt, D., Duda, K., Agarwal, P., Kreeger, 404 L., Sridhar, T., Bursell, M., and C. Wright, "Virtual 405 eXtensible Local Area Network (VXLAN): A Framework for 406 Overlaying Virtualized Layer 2 Networks over Layer 3 407 Networks", RFC 7348, DOI 10.17487/RFC7348, August 2014, 408 . 410 [RFC7365] Lasserre, M., Balus, F., Morin, T., Bitar, N., and Y. 411 Rekhter, "Framework for Data Center (DC) Network 412 Virtualization", RFC 7365, DOI 10.17487/RFC7365, October 413 2014, . 415 [RFC7761] Fenner, B., Handley, M., Holbrook, H., Kouvelas, I., 416 Parekh, R., Zhang, Z., and L. Zheng, "Protocol Independent 417 Multicast - Sparse Mode (PIM-SM): Protocol Specification 418 (Revised)", STD 83, RFC 7761, DOI 10.17487/RFC7761, March 419 2016, . 421 Appendix A. BIER Use Case in Data Centers 423 In current data center virtualization, virtual eXtensible Local Area 424 Network (VXLAN) [RFC7348] is a kind of network virtualization overlay 425 technology which is overlaid between NVEs and is intended for multi- 426 tenancy data center networks, whose reference architecture is 427 illustrated as per Figure 1. 429 +--------+ +--------+ 430 | Tenant +--+ +----| Tenant | 431 | System | | (') | System | 432 +--------+ | ................ ( ) +--------+ 433 | +-+--+ . . +--+-+ (_) 434 | | NVE|--. .--| NVE| | 435 +--| | . . | |---+ 436 +-+--+ . . +--+-+ 437 / . . 438 / . L3 Overlay . +--+-++--------+ 439 +--------+ / . Network . | NVE|| Tenant | 440 | Tenant +--+ . .--| || System | 441 | System | . . +--+-++--------+ 442 +--------+ ................ 444 Figure 1: NVO3 Architecture 446 And there are two kinds of most common methods about how to forward 447 BUM packets in this virtualization overlay network. One is using PIM 448 as underlay multicast routing protocol to build explicit multicast 449 distribution tree, such as PIM-SM [RFC7761] or PIM-BIDIR [RFC5015] 450 multicast routing protocol. Then, when BUM packets arrive at NVE, it 451 requires NVE to have a mapping between the VXLAN Network Identifier 452 and the IP multicast group. According to the mapping, NVE can 453 encapsulate BUM packets in a multicast packet which group address is 454 the mapping IP multicast group address and steer them through 455 explicit multicast distribution tree to the destination NVEs. This 456 method has two serious drawbacks. It need the underlay network 457 supports complicated multicast routing protocol and maintains 458 multicast related per-flow state in every transit nodes. What is 459 more, how to configure the ratio of the mapping between VNI and IP 460 multicast group is also an issue. If the ratio is 1:1, there should 461 be 16M multicast groups in the underlay network at maximum to map to 462 the 16M VNIs, which is really a significant challenge for the data 463 center devices. If the ratio is n:1, it would result in inefficiency 464 bandwidth utilization which is not optimal in data center networks. 466 The other method is using ingress replication to require each NVE to 467 create a mapping between the VXLAN Network Identifier and the remote 468 addresses of NVEs which belong to the same virtual network. When NVE 469 receives BUM traffic from the attached tenant, NVE can encapsulate 470 these BUM packets in unicast packets and replicate them and tunnel 471 them to different remote NVEs respectively. Although this method can 472 eliminate the burden of running multicast protocol in the underlay 473 network, it has a significant disadvantage: large waste of bandwidth, 474 especially in big-sized data center where there are many receivers. 476 BIER [I-D.ietf-bier-architecture] is an architecture that provides 477 optimal multicast forwarding through a "BIER domain" without 478 requiring intermediate routers to maintain any multicast related per- 479 flow state. BIER also does not require any explicit tree-building 480 protocol for its operation. A multicast data packet enters a BIER 481 domain at a "Bit-Forwarding Ingress Router" (BFIR), and leaves the 482 BIER domain at one or more "Bit-Forwarding Egress Routers" (BFERs). 483 The BFIR router adds a BIER header to the packet. The BIER header 484 contains a bit-string in which each bit represents exactly one BFER 485 to forward the packet to. The set of BFERs to which the multicast 486 packet needs to be forwarded is expressed by setting the bits that 487 correspond to those routers in the BIER header. Specifically, for 488 BIER-TE, the BIER header may also contain a bit-string in which each 489 bit indicates the link the flow passes through. 491 The following sub-sections try to propose how to take full advantage 492 of overlay multicast protocol to carry virtual network information, 493 and create a mapping between the virtual network information and the 494 bit-string to implement BUM services in data centers. 496 A.1. Convention and Terminology 498 The terms about NVO3 are defined in [RFC7365]. The most common 499 terminology used in this appendix is listed below. 501 NVE: Network Virtualization Edge, which is the entity that 502 implements the overlay functionality. An NVE resides at the 503 boundary between a Tenant System and the overlay network. 505 VXLAN: Virtual eXtensible Local Area Network 507 VNI: VXLAN Network Identifier 509 Virtal Network Context Identifier: Field in an overlay encapsulation 510 header that identifies the specific VN the packet belongs to. 512 A.2. BIER in data centers 514 This section tries to describe how to use BIER as an optimal scheme 515 to forward the broadcast, unknown and multicast (BUM) packets when 516 they arrive at the ingress NVE in data centers. 518 The principle of using BIER to forward BUM traffic is that: firstly, 519 it requires each ingress NVE to have a mapping between the Virtual 520 Network Context Identifier and the bit-string in which each bit 521 represents exactly one egress NVE to forward the packet to. And 522 then, when receiving the BUM traffic, the BFIR/Ingree NVE maps the 523 receiving BUM traffic to the mapping bit-string, encapsulates the 524 BIER header, and forwards the encapsulated BUM traffic into the BIER 525 domain to the other BFERs/Egress NVEs indicated by the bit-string. 527 Furthermore, as for how each ingress NVE knows the other egress NVEs 528 that belong to the same virtual network and creates the mapping is 529 the main issue discussed below. Basically, BIER Multicast Listener 530 Discovery is an overlay solution to support ingress routers to keep 531 per-egress-router state to construct the BIER bit-string associated 532 with IP multicast packets entering the BIER domain. The following 533 section tries to extend BIER MLD to carry virtual network 534 information(such as Virtual Network Context identifier), and 535 advertise them between NVEs. When each NVE receive these 536 information, they create the mapping between the virtual network 537 information and the bit-string representing the other NVEs belonged 538 to the same virtual network. 540 A.3. A BIER MLD solution for Virtual Network information 542 The BIER MLD solution allows having multiple MLD instances by having 543 unique pairs of BMLD Nodes and BMLD Querier addresses for each 544 instance. Assume for now that we have a unique instance per VNI and 545 that all BMLD routers are using the same mapping between VNIs and 546 BMLD address pairs. Also for each VNI there is a multicast group 547 used for encapsulation of BUM traffic over BIER. This group may 548 potentially be shared by some or all of the VNIs. 550 Each NVE acquires the Virtual Network information, and advertises 551 this Virtual Network information to other NVEs through the MLD 552 messages. For a given VNI it sends BMLD reports to the BMLD nodes 553 address used for that VNI, for the group used for delivering BUM 554 traffic for that VNI. This allows all NVE routers to know which 555 other NVE routers have interest in BUM traffic for a particular VNI. 556 If one attached virtual network is migrated, the NVE will withdraw 557 the Virtual Network information by sending an unsolicited BMLD 558 report. Note that NVEs also respond to periodic queries to BMLD 559 Nodes addresses corresponding to VNIs for which they have interest. 561 When ingress NVE receives the Virtual Network information 562 advertisement message, it builds a mapping between the receiving 563 Virtual Network Context Identifier in this message and the bit-string 564 in which each bit represents one egress NVE who sends the same 565 Virtual Network information. Subsequently, once this ingress NVE 566 receives some other MLD advertisements which include the same Virtual 567 Network information from some other NVEs , it updates the bit-string 568 in the mapping and adds the corresponding sending NVE to the updated 569 bit-string. Once the ingress NVE removes one virtual network, it 570 will delete the mapping corresponding to this virtual network as well 571 as send withdraw message to other NVEs. 573 After finishing the above interaction of MLD messages, each ingress 574 NVE knows where the other egress NVEs are in the same virtual 575 network. When receiving BUM traffic from the attached virtual 576 network, each ingress NVE knows exactly how to encapsulate this 577 traffic and where to forward them to. 579 This can be used in both IPv4 network and IPv6 network. In IPv4, 580 IGMP protocol does the similar extension for carrying Virtual Network 581 information TLV in Version 2 membership report message. 583 Note that it is possible to have multiple VNIs map to the same pair 584 of BMLD addresses. Provided VNIs that map to the same BMLD address 585 uses different multicast groups for encapsulation, this is not a 586 problem, because each instance is tracking interest for each 587 multicast group separately. If multiple VNIs map to the same pair 588 and the multicast group used is not unique, some NVEs may receive BUM 589 traffic for which they are not interested. An NVE would drop packets 590 for an unknown VNI, but it means wasting some bandwidth and 591 processing. This is similar to the non-BIER case where there is not 592 a unique multicast group for encapsulation. The improvement offered 593 by using BMLD is by using multiple instance, hence reducing the 594 problems caused by using the same transport group for multiple VNIs. 596 Authors' Addresses 598 Pierre Pfister 599 Cisco Systems 600 Paris 601 France 603 Email: pierre.pfister@darou.fr 605 IJsbrand Wijnands 606 Cisco Systems 607 De Kleetlaan 6a 608 Diegem 1831 609 Belgium 611 Email: ice@cisco.com 612 Stig Venaas 613 Cisco Systems 614 Tasman Drive 615 San Jose, CA 95134 616 USA 618 Email: stig@cisco.com 620 Cui(Linda) Wang 621 ZTE Corporation 622 No.50 Software Avenue, Yuhuatai District 623 Nanjing, CA 624 China 626 Email: wang.cui1@zte.com.cn 628 Zheng(Sandy) Zhang 629 ZTE Corporation 630 No.50 Software Avenue, Yuhuatai District 631 Nanjing, CA 632 China 634 Email: zhang.zheng@zte.com.cn 636 Markus Stenberg 637 Helsinki 00930 638 Finland 640 Email: markus.stenberg@iki.fi