Internet Working Group Ali Sajassi Internet Draft Samer Salam Category: Standards Track Cisco Expires: September 7, 2011 March 7, 2011 PBB E-VPN draft-sajassi-l2vpn-pbb-evpn-00.txt Status of this Memo This Internet-Draft is submitted to IETF in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet- Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html This Internet-Draft will expire on August 11, 2011. Copyright Notice Copyright (c) 2011 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Sajassi, et. al. [Page 1] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 Abstract This document discusses how Ethernet Provider Backbone Bridging [802.1ah] can be combined with E-VPN in order to reduce the number of BGP MAC advertisement routes, provide Customer MAC address mobility with MAC sub-netting, provide Customer MAC address scoping, offer per site policies and avoid Customer MAC address flushing on topology changes. The combined solution is referred to as PBB-EVPN. Conventions The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 Table of Contents 1. Introduction.................................................... 3 2. Terminology..................................................... 3 3. Requirements.................................................... 3 3.1. MAC Advertisement Route Scalability........................... 3 3.2. C-MAC Mobility with MAC Sub-netting........................... 4 3.3. C-MAC Address Scoping......................................... 4 3.4. Per Site Policy Support....................................... 4 3.5. Avoiding C-MAC Address Flushing............................... 5 4. Solution Overview............................................... 5 5. BGP Encoding.................................................... 6 5.1. BGP MAC Advertisement Route................................... 6 5.2. Ethernet Auto-Discovery Route................................. 6 5.3. Per VPN Route Targets......................................... 7 6. Operation....................................................... 7 6.1. MAC Address Distribution over Core............................ 7 6.2. Device Multi-homing........................................... 7 6.2.1. MES MAC Layer Addressing & Multi-homing..................... 7 6.2.2. Split Horizon and Designated Forwarder Election............ 10 6.3. Frame Forwarding............................................. 10 6.3.1. Unicast.................................................... 10 6.3.2. Multicast/Broadcast........................................ 11 7. Acknowledgements............................................... 11 8. Security Considerations........................................ 11 9. IANA Considerations............................................ 11 10. Intellectual Property Considerations.......................... 11 Sajassi, et al. [Page 2] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 11. Normative References.......................................... 11 12. Informative References........................................ 11 13. Authors' Addresses............................................ 12 1. Introduction [E-VPN] introduces a solution for multipoint L2VPN services with advanced multi-homing capabilities using BGP for distributing customer MAC address reach-ability information over the core MPLS/IP network. [802.1ah] defines an architecture for Ethernet Provider Backbone Bridging (PBB), where MAC tunneling is employed to improve service instance and MAC address scalability in Ethernet networks. In this document, we discuss how PBB can be combined with E-VPN in order to reduce the number of BGP MAC advertisement routes, provide Customer MAC address mobility with MAC sub-netting, provide Customer MAC address scoping, offer per site policies and avoid Customer MAC address flushing on topology changes. The combined solution is referred to as PBB-EVPN. 2. Terminology BEB: Backbone Edge Bridge B-MAC: Backbone MAC Address CE: Customer Edge C-MAC: Customer MAC Address DHD: Dual-homed Device DHN: Dual-homed Network LACP: Link Aggregation Control Protocol LSM: Label Switched Multicast MDT: Multicast Delivery Tree MES: MPLS Edge Switch MP2MP: Multipoint to Multipoint P2MP: Point to Multipoint P2P: Point to Point PoA: Point of Attachment PW: Pseudowire E-VPN: Ethernet VPN 3. Requirements The requirements for PBB-EVPN include all the requirements for E-VPN that were described in [EVPN-REQ], in addition to the following: 3.1. MAC Advertisement Route Scalability Sajassi, et al. [Page 3] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 In typical operation, an [E-VPN] MES sends a BGP MAC Advertisement Route per customer MAC (C-MAC) address. In certain applications, this poses scalability challenges, as is the case in virtualized data center environments where the number of virtual machines (VMs), and hence the number of C-MAC addresses, can be in the millions. In such scenarios, it is required to reduce the number of BGP MAC Advertisement routes by relying on a MAC 'summarization' scheme, as is provided by PBB. Note that the MAC sub-netting capability already built into E-VPN is not sufficient in those environments, as will be discussed next. 3.2. C-MAC Mobility with MAC Sub-netting Certain applications, such as virtual machine mobility, require support for fast C-MAC address mobility. For these applications, it is not possible to use MAC address sub-netting in E-VPN, i.e. advertise reach-ability to a MAC address prefix. Rather, the exact virtual machine MAC address needs to be transmitted in BGP MAC Advertisement route. Otherwise, traffic would be forwarded to the wrong segment when a virtual machine moves from one Ethernet segment to another. This hinders the scalability benefits of sub-netting. It is required to support C-MAC address mobility, while retaining the scalability benefits of MAC sub-netting. This can be achieved by leveraging PBB technology, which defines a Backbone MAC (B-MAC) address space that is independent of the C-MAC address space. 3.3. C-MAC Address Scoping In E-VPN, all the MES nodes participating in the same EVI are exposed to all the C-MAC addresses learnt by any one of these MES nodes. This is the case even if the MES in question is not involved in forwarding traffic to, or from, these C-MAC addresses. Even if an implementation doesn't install hardware forwarding entries for C-MAC addresses that are not part of active traffic flows on that MES, the device memory is still consumed by keeping record of the C-MAC addresses in the control-plane. In network applications with millions of C-MAC addresses, this introduces a non-trivial waste of MES resources. As such, it is required to confine the scope of visibility of C-MAC addresses only to those MES nodes that are actively involved in forwarding traffic to, or from, these addresses. 3.4. Interworking with TRILL and 802.1aq Access Networks with C-MAC Address Transparency [TRILL] and [802.1aq] define next generation Ethernet bridging technologies that offer optimal forwarding using IS-IS control plane, and C-MAC address transparency via Ethernet tunneling technologies. When access networks based on TRILL or 802.1aq are interconnected over an MPLS/IP network, it is required to guarantee Sajassi, et al. [Page 4] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 C-MAC address transparency on the hand-off point and the edge (i.e. MES) of the MPLS network. As such, solutions that require termination of the access data-plane encapsulation (i.e. TRILL or 802.1ah) at the hand-off to the MPLS network do not meet this transparency requirement, and expose the MPLS edge devices to the MAC address scalability problem. PBB-EVPN supports seamless interconnect with these next generation Ethernet solutions while guaranteeing C-MAC address transparency on the MES nodes. 3.5. Per Site Policy Support In many applications, it is required to be able to enforce connectivity policy rules at the granularity of a site (or segment). This includes the ability to control which MES nodes in the network can forward traffic to, or from, a given site. PBB-EVPN is capable of providing this granularity of policy control. In the case where per C-MAC address granularity is required, the EVI can always continue to operate in E-VPN mode. 3.6. Avoiding C-MAC Address Flushing It is required to avoid C-MAC address flushing upon link, port or node failure for multi-homed devices and networks. This is in order to speed up re-convergence upon failure. 4. Solution Overview The solution involves incorporating IEEE 802.1ah Backbone Edge Bridge (BEB) functionality on the E-VPN MES nodes. The MES devices would then receive 802.1Q Ethernet frames from their attachment circuits, encapsulate them in the PBB header and forward the frames over the IP/MPLS core. On the egress E-VPN MES, the PBB header is removed following the MPLS disposition, and the original 802.1Q Ethernet frame is delivered to the customer equipment. BEB +--------------+ BEB || | | || \/ | | \/ +----+ AC1 +----+ | | +----+ +----+ | CE1|-----| | | | | |---| CE2| +----+\ |MES1| | IP/MPLS | |MES3| +----+ \ +----+ | Network | +----+ \ | | AC2\ +----+ | | \| | | | |MES2| | | +----+ | | /\ +--------------+ Sajassi, et al. [Page 5] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 || BEB <-802.1Q-> <------PBB over MPLS------> <-802.1Q-> Figure 1: PBB-EVPN Network The MES nodes perform the following functions: - Learn customer MAC addresses (C-MACs) over the attachment circuits in the data-plane, per normal bridge operation. - Learn remote C-MAC to B-MAC bindings in the data-plane from traffic ingress from the core. - Advertise local B-MAC address reach-ability information in BGP to all other MES nodes in the same set of service instances. Note that every MES has a set of local B-MAC addresses that uniquely identify the device. More on the MES addressing in section 5. - Build a forwarding table from remote BGP advertisements received associating remote B-MAC addresses with remote MES IP addresses. 5. BGP Encoding PBB-EVPN leverages the same BGP Routes and Attributes defined in [E- VPN], adapted as follows: 5.1. BGP MAC Advertisement Route The E-VPN MAC Advertisement Route is used to distribute B-MAC addresses of the MES nodes instead of the C-MAC addresses of end- stations/hosts. This is because the C-MAC addresses are learnt in the data-plane for traffic arriving from the core. The MAC Advertisement Route is encoded as follows: -The RD is set to a Type 1 RD RD [RFC4364]. The value field encodes the IP address of the MES (typically, the loopback address) followed by 0. The reason for such encoding is that the RD cannot be that of a single EVI since the same B-MAC address can span across multiple EVIs. -The MAC address field contains the B-MAC address. -The Ethernet Tag field is set to 0. The route is tagged with the set of RTs corresponding to all EVIs associated with the B-MAC address. All other fields are set as defined in [E-VPN]. 5.2. Ethernet Auto-Discovery Route Sajassi, et al. [Page 6] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 The Ethernet A-D Route is used in PBB-EVPN to advertise reach- ability of Ethernet segments. The route is encoded as follows: -The RD is set to a Type 1 RD RD [RFC4364]. The value field encodes the IP address of the MES (typically, the loopback address) followed by 0. The reason for such encoding is that the RD cannot be that of a single EVI since the same Ethernet segment can be associated with multiple EVIs. -The Ethernet Tag field is always set to 0. -The MPLS label is downstream assigned and identifies the Ethernet segment (i.e. B-MAC address) on the advertising MES. The Ethernet A-D Route is tagged with the set of RTs corresponding to all EVIs associated with the B-MAC address. All other fields are set as defined in [E-VPN]. 5.3. Per VPN Route Targets PBB-EVPN uses the same set of route targets defined in [E-VPN]. More specifically, the RT associated with a VPN is set to the value of the I-SID associated with the service instance. This eliminates the need for manually configuring the VPN-RT. Note that all other BGP messages and/or attributes are used as defined in [E-VPN]. 6. Operation This section discusses the operation of PBB-EVPN, specifically in areas where it differs from [E-VPN]. 6.1. MAC Address Distribution over Core In PBB-EVPN, host MAC addresses (i.e. C-MAC addresses) need not be distributed in BGP. Rather, every MES independently learns the C-MAC addresses in the data-plane via normal bridging operation. Every MES has a set of one or more unicast B-MAC addresses associated with it, and those are the addresses distributed over the core in MAC Advertisement routes. Given that these B-MAC addresses are global within the provider's network, there is no need to advertise them on a per service instance basis. 6.2. Device Multi-homing 6.2.1. MES MAC Layer Addressing & Multi-homing In [802.1ah] every BEB is uniquely identified by one or more B-MAC addresses. These addresses are usually locally administered by the Service Provider. For PBB-EVPN, the choice of B-MAC address(es) for Sajassi, et al. [Page 7] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 the MES nodes must be examined carefully as it has implications on the proper operation of multi-homing. In particular, for the scenario where a CE is multi-homed to a number of MES nodes with all-active redundancy and flow-based load-balancing, a given C-MAC address would be reachable via multiple MES nodes concurrently. Given that any given remote MES will bind the C-MAC address to a single B-MAC address, then the various MES nodes connected to the same CE must share the same B-MAC address. Otherwise, the MAC address table of the remote MES nodes will keep flip-flopping between the B-MAC addresses of the various MES devices. For example, consider the network of Figure 1, and assume that MES1 has B-MAC BM1 and MES2 has B-MAC BM2. Also, assume that both links from CE1 to the MES nodes are part of an all-active multi-chassis Ethernet link aggregation group. If BM1 is not equal to BM2, the consequence is that the MAC address table on MES3 will keep oscillating such that the C-MAC address CM of CE1 would flip-flop between BM1 or BM2, depending on the load-balancing decision on CE1 for traffic destined to the core. Considering that there could be multiple sites (e.g. CEs) that are multi-homed to the same set of MES nodes, then it is required for all the MES devices in a Redundancy Group to have a unique B-MAC address per site. This way, it is possible to achieve fast convergence in the case where a link or port failure impacts the attachment circuit connecting a single site to a given MES. +---------+ +-------+ MES1 | IP/MPLS | / | | CE1 | Network | MESr M1 \ | | +-------+ MES2 | | +-------+ | | / | | CE2 | | M2 \ | | +-------+ MES3 | | +---------+ Figure 2: B-MAC Address Assignment In the example network shown in Figure 2 above, two sites corresponding to CE1 and CE2 are dual-homed to MES1/MES2 and MES2/MES3, respectively. Assume that BM1 is the B-MAC used for the site corresponding to CE1. Similarly, BM2 is the B-MAC used for the site corresponding to CE2. On MES1, a single B-MAC address (BM1) is required for the site corresponding to CE1. On MES2, two B-MAC addresses (BM1 and BM2) are required, one per site. Whereas on MES3, a single B-MAC address (BM2) is required for the site corresponding Sajassi, et al. [Page 8] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 to CE2. All three MES nodes would advertise their respective B-MAC addresses in BGP using the MAC Advertisement routes defined in [E- VPN]. The remote MES, MESr, would learn via BGP that BM1 is reachable via MES1 and MES2, whereas BM2 is reachable via both MES2 and MES3. Furthermore, MESr establishes via the normal bridge learning that C-MAC M1 is reachable via BM1, and C-MAC M2 is reachable via BM2. As a result, MESr can load-balance traffic destined to M1 between MES1 and MES2, as well as traffic destined to M2 between both MES2 and MES3. In the case of a failure that causes, for example, CE1 to be isolated from MES1, the latter can withdraw the route it has advertised for BM1. This way, MESr would update its path list for BM1, and will send all traffic destined to M1 over to MES2 only. For single-homed sites, it is possible to assign a unique B-MAC address per site, or have all the single-homed sites connected to a given MES share a single B-MAC address. The advantage of the first model is the ability to avoid C-MAC destination address lookup on the disposition PE (even though source C-MAC learning is still required in the data-plane). Also, by assigning the B-MAC addresses from a contiguous range, it is possible to advertise a single B-MAC subnet for all single-homed sites, thereby rendering the number of MAC advertisement routes required at par with the second model. In summary, every MES may use a unicast B-MAC address shared by all single-homed CEs or a unicast B-MAC address per single-homed CE, and in addition a unicast B-MAC address per dual-homed CE. In the latter case, the B-MAC address MUST be the same for all MES nodes in a Redundancy Group connected to the same CE. 6.2.1.1. Automating B-MAC Address Assignment The MES B-MAC address used for single-homed sites can be automatically derived from the hardware (using for e.g. the backplane's address). However, the B-MAC address used for multi- homed sites must be coordinated among the RG members. To automate the assignment of this latter address, the MES can derive this B-MAC address from the MAC Address portion of the CE's LACP System Identifier by flipping the 'Locally Administered' bit of the CE's address. This guarantees the uniqueness of the B-MAC address within the network, and ensures that all MES nodes connected to the same multi-homed CE use the same value for the B-MAC address. Note that with this automatic provisioning of the B-MAC address associated with mult-homed CEs, it is not possible to support the uncommon scenario where a CE has multiple bundles towards the MES nodes, and the service involves hair-pinning traffic from one bundle to another. This is because the split-horizon filtering relies on B- MAC addresses rather than Site-ID Labels (as will be described in the next section). The operator must explicitly configure the B-MAC address for this fairly uncommon service scenario. Sajassi, et al. [Page 9] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 Whenever a B-MAC address is provisioned on the MES, either manually or automatically (as an outcome of CE auto-discovery), the MES MUST transmit an MAC Advertisement Route for the B-MAC address with a downstream assigned MPLS label that uniquely identifies the address on the advertising MES. The route is tagged with the RTs of the associated EVIs as described above. 6.2.2. Split Horizon and Designated Forwarder Election [E-VPN] relies on access split horizon, where the Ethernet Segment Label is used for egress filtering on the attachment circuit in order to prevent forwarding loops. In PBB-EVPN, the B-MAC source address can be used for the same purpose, as it uniquely identifies the originating site of a given frame. As such, Segment Labels are not used in PBB-EVPN, and the egress filtering is done based on the B-MAC source address. It is worth noting here that [802.1ah] defines this B-MAC address based filtering function as part of the I- Component options, hence no new functions are required to support split-horizon beyond what is already defined in [802.1ah]. Given that the Segment label is not used in PBB-EVPN, the MES sets the Label field in the Ethernet Segment Route to 0. The Designated Forwarder election procedures remain unchanged from [E-VPN]. 6.3. Frame Forwarding The frame forwarding functions are divided in between the Bridge Module, which hosts the [802.1ah] Backbone Edge Bridge (BEB) functionality, and the MPLS Forwarder which handles the MPLS imposition/disposition. The details of frame forwarding for unicast and multi-destination frames are discussed next. 6.3.1. Unicast Known unicast traffic received from the AC will be PBB-encapsulated by the MES using the B-MAC source address corresponding to the originating site. The unicast B-MAC destination address is determined based on a lookup of the C-MAC destination address (the binding of the two is done via transparent learning of reverse traffic). The resulting frame is then encapsulated with an LSP tunnel label and the MPLS label which uniquely identifies the B-MAC destination address on the egress MES. If per flow load-balancing over ECMPs in the MPLS core is required, then a flow label is added as the end of stack label. For unknown unicast traffic, the MES can optionally forward these frames over MPLS core; however, the default is not to forward. If these frames are to be forwarded, then the same set of options used Sajassi, et al. [Page 10] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 for forwarding multicast/broadcast frames (as described in next section) are used. 6.3.2. Multicast/Broadcast Multi-destination frames received from the AC will be PBB- encapsulated by the MES using the B-MAC source address corresponding to the originating site. The multicast B-MAC destination address is selected based on the value of the I-SID as defined in [802.1ah]. The resulting frame is then forwarded over the MPLS core using one out of the following two options: Option 1: the MPLS Forwarder can perform ingress replication over a set of MP2P tunnel LSPs. The frame is encapsulated with an LSP tunnel label and the E-VPN ingress replication label advertised in the Inclusive Multicast Route. Option 2: the MPLS Forwarder can use P2MP LSM tunnel per the procedures defined in [E-VPN]. This includes either the use of Inclusive or Aggregate Inclusive trees. Note that the same procedures for advertising and handling the Inclusive Multicast Route defined in [E-VPN] apply here. 7. Acknowledgements TBD. 8. Security Considerations There are no additional security aspects beyond those of VPLS/H-VPLS that need to be discussed here. 9. IANA Considerations This document requires IANA to assign a new SAFI value for L2VPN_MAC SAFI. 10. Intellectual Property Considerations This document is being submitted for use in IETF standards discussions. 11. Normative References [802.1ah] "Virtual Bridged Local Area Networks Amendment 7: Provider Backbone Bridges", IEEE Std. 802.1ah-2008, August 2008. 12. Informative References Sajassi, et al. [Page 11] draft-sajassi-l2vpn-pbb-evpn-00.txt February 2011 [EVPN-REQ] Sajassi et al., "Requirements for Ethernet VPN (E-VPN)", draft-sajassi-raggarwa-l2vpn-evpn-req-00.txt, work in progress, October, 2010. [E-VPN] Aggarwal et al., "BGP MPLS Based Ethernet VPN", draft- raggarwa-sajassi-l2vpn-evpn-01.txt, November, 2010. , work in progress, June, 2010. 13. Authors' Addresses Ali Sajassi Cisco 170 West Tasman Drive San Jose, CA 95134, US Email: sajassi@cisco.com Samer Salam Cisco 595 Burrard Street, Suite 2123 Vancouver, BC V7X 1J1, Canada Email: ssalam@cisco.com Sajassi, et al. [Page 12]