The way I see it it is not the number of vrf:s that is the limitation, it will be the rib and fib size. Then how you distribute that between vrf:s wont matter.
I havent seen any official tests of number of vrf:s, but assume that 500 vrf:s should work. But I figure 4000 vrf:s might work aswell. This is based on that Juniper creates a label per vrf and PE, not per route and PE that Cisco does, so L3VPN is not consuming so much fib. But all routes will be in rib of course. Patrik _______________________________________________ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp
