Users Guide
Enabling MBGP Congurations
Multiprotocol BGP (MBGP) is an enhanced BGP that carries IP multicast routes. BGP carries two sets of routes: one set for unicast routing
and one set for multicast routing. The routes associated with multicast routing are used by the protocol independent multicast (PIM) to
build data distribution trees.
The Dell Networking OS MBGP is implemented per RFC 1858. You can enable the MBGP feature per router and/or per peer/peer-group.
The default is IPv4 Unicast routes.
When you congure a peer to support IPv4 multicast, the system takes the following actions:
• Send a capability advertisement to the peer in the BGP Open message specifying IPv4 multicast as a supported AFI/SAFI (Subsequent
Address Family Identier).
• If the corresponding capability is received in the peer’s Open message, BGP marks the peer as supporting the AFI/SAFI.
• When exchanging updates with the peer, BGP sends and receives IPv4 multicast routes if the peer is marked as supporting that AFI/
SAFI.
• Exchange of IPv4 multicast route information occurs through the use of two new attributes called MP_REACH_NLRI and
MP_UNREACH_NLRI, for feasible and withdrawn routes, respectively.
• If the peer has not been activated in any AFI/SAFI, the peer remains in Idle state.
Most Dell Networking OS BGP IPv4 unicast commands are extended to support the IPv4 multicast RIB using extra options to the
command. For a detailed description of the MBGP commands, refer to the Dell Networking OS Command Line Interface Reference Guide.
• Enables support for the IPv4 multicast family on the BGP node.
CONFIG-ROUTER-BGP mode
address family ipv4 multicast
• Enable IPv4 multicast support on a BGP neighbor/peer group.
CONFIG-ROUTER-BGP-AF (Address Family) mode
neighbor [ip-address | peer-group-name] activate
Congure IPv6 NH Automatically for IPv6 Prex Advertised over IPv4
Neighbor
You can congure the system to pick the next hop IPv6 address dynamically for IPv6 prex advertised over an IPv4 neighbor congured
under IPv6 address family. If there is no IPv6 address congured on the local interface, the system uses the IPv4 mapped IPv6 address. If
there are multiple IPv6 addresses congured on the interface, the system uses the lowest IPv6 address congured on that interface. If the
conguration is already present and a new IPv6 address, which is lower than the lowest existing address, is assigned to one of the peer
interfaces, that address is not used as the NH until you ap the interface manually.
Figure 24. Congure IPv6 NH to advertise IPv6 prex over IPv4 neighbor
Border Gateway Protocol IPv4 (BGPv4)
221