Hello all, we are trying use BGP-EVPN + VRF on Linux in L3 mode. For each VPN instance, we add two snippets into the FRR configuration: Snippet 1: ==== vrf <vrf-iface-name> vni <vni> prefix-routes-only exit-vrf ==== Snippet 2: ==== router bgp 64512 vrf <vrf-iface-name> bgp router-id 169.254.0.1 ! address-family ipv4 unicast redistribute kernel exit-address-family ! address-family l2vpn evpn advertise ipv4 unicast exit-address-family ==== Our use case requires dynamic creation and removal of VRF instances on demand. I understand that FRRouting can autodetect VRF+VxLAN interface groups that are already present in the system and start serving them, but only as "classic" L2VPNs. My question is: should it be made possible (or maybe it is already possible?) to set default attributes for VRF/EVPNs that FRR autodetects? So that one could add something like this just once: ==== vrf DEFAULT vni AUTODETECT prefix-routes-only exit-vrf ==== ==== router bgp 64512 vrf DEFAULT bgp router-id 169.254.0.1 ! address-family ipv4 unicast redistribute kernel exit-address-family ! address-family l2vpn evpn advertise ipv4 unicast exit-address-family ==== and then all newly created VRF+VxLAN groups would be automatically picked up by FRR and configured as L3 EVPNs, without the need to rewrite configuration file and restart the daemon. Does this make sense or am I missing something? On a related note, I understand that currently FRR can fetch FDB only from 'bridge' interfaces, but not directly from VxLAN interfaces. So VxLAN interface cannot be directly enslaved into the VRF interface, but instead we have to add an intermediate bridge: "vrfXX---brXX---vxlanXX". (while client interfaces are directly enslaved into the VRF interface). In such setup, 'br' interface seems unnecessary. Should it be made possible to avoid this redundant 'br' interface? Thank you, and don't get too angry if I am messing things up: I am pretty new to this. Eugene