Hi, I have the following questions/comments.
The use case seem to be: > to have multiple subnets using a single broadcast domain - only one broadcast > domain and one IRB for "N" subnets compare to the "N" broadcast domain and > "N" IRB interface to manage. > to have multiple subnets from same ES on multihoming All-Active mode What exactly is a broadcast domain? From the following RFC7432 text: > VLAN-Based Service Interface: With this service interface, an EVPN instance > consists of only a single broadcast domain (e.g., a single VLAN). > VLAN Bundle Service Interface: With this service interface, an EVPN instance > corresponds to multiple broadcast domains (e.g., multiple VLANs) > VLAN-Aware Bundle Service Interface: With this service interface, an EVPN > instance consists of multiple broadcast domains (e.g., multiple VLANs) with > each VLAN having its own bridge table It seems that a broadcast domain is equivalent to a vlan. If that is the case, I have two confusions: - How are multiple subnets in a single broadcast domain (i.e. a single vlan) differentiated? Why does the draft say the following where multiple VLANs are used? > For example, lets take the case from Figure 1 where PE1 learns MAC of H1 on > VLAN 1 (subnet S1). > In Figure 1, BD-1 has multiple subnets where each subnet is distinguished by > VLAN 1, 2,3 and 4. - Since a VLAN bundle in RFC7432 has *multiple* broadcast domains, why does the draft say the following? > From the definition, it seems like VLAN Bundle Service Interface does provide > flexibility to support multiple subnets within *a single* broadcast domain. Or is it that the requirement is not about "a single broadcast domain", but "a single IRB"? If so, we should remove all text about broadcast domain. https://datatracker.ietf.org/doc/html/draft-ietf-bess-evpn-vpws-fxc-08 says: [RFC8214] describes a solution to deliver P2P services using BGP constructs defined in [RFC7432]. It delivers this P2P service between a pair of Attachment Circuits (ACs), where an AC can designate on a PE, a port, a VLAN on a port, or a group of VLANs on a port. It seems that: - Each pair of AC in vpws-fxc is given a normalized VID as the AC ID. - In this draft, each VLAN across the MHES is given an ACID and signaled in the attachment circuit ID Extended Community (ACID EC) I am confused as to why this draft applies only to MHES. Let's say there is no MH and the CE is only homed to PE1, but both PE1 and PE2 host subnet S1. When PE2 gets a packet destined to H1 and it is to be routed down PE2's IRB (and then switched to PE1 - asymmetric mode), what VLAN ID should PE2 use? I suppose what is needed is a mapping between vlan and subnet - something to be configured locally. But once that is done, the MHES problem is also solved, right? And the vlan/subnet mapping is comparable to the ACID configuration. In the case of syncing IGMP/MLD state across MHES PEs, we just need to attach the VLAN info to the routes. Finally, even if we want to go with the solution in the draft, I don't think we should call this the 4th service interface. It is just about how to do single IRB on the vlan bundle service interface. Adding the 4th service interface just adds confusion. Thanks. Jeffrey Juniper Business Use Only _______________________________________________ BESS mailing list BESS@ietf.org https://www.ietf.org/mailman/listinfo/bess
