Some times we may find problems during the advertisements of the clients routes in BGP by adding network in the vrf (case of MPLS) & with the help of network command (In case of IP VPN). When the routes advertisements are forwarded they always carry the next hop ip address as the interface ip. A problem which persists in this scenario is that the interface ip address is not in the IGP table as consequence of packet drops. To overcome this problem we need to use the {next-hop-self} during the IBGP peering with the RR or with PE.