Configuration is completed. Check distribution of routing information and network availability of nodes:
PE1# sh bgp vpnv4 unicast all
Status codes: * - valid, > - best, i - internal, S - stale
Origin codes: i - IGP, e - EGP, ? - incomplete
Codes Route Distinguisher IP Prefix Next hop Metric Label LocPrf
Weight Path
----- --------------------- ------------------ --------------- ---------- ------- ----------
------ ----------------
*> 65500:1 10.110.0.1/32 -- -- 37 100 --
65501 i
*> 65500:1 10.111.0.1/32 -- -- 35 100 --
65501 i
*>i 65500:1 10.113.0.1/32 10.10.1.2 -- 43 100 0
?
*>i 65500:1 10.114.0.1/32 10.10.1.2 -- 48 100 0
?
CE1# ping 10.113.0.1 source ip 10.110.0.1 detailed
PING 10.113.0.1 (10.113.0.1) from 10.110.0.1 : 56 bytes of data.
64 bytes from 10.113.0.1: icmp_seq=1 ttl=0 time=1.31 ms
64 bytes from 10.113.0.1: icmp_seq=2 ttl=0 time=1.14 ms
64 bytes from 10.113.0.1: icmp_seq=3 ttl=0 time=1.08 ms
64 bytes from 10.113.0.1: icmp_seq=4 ttl=0 time=1.06 ms
64 bytes from 10.113.0.1: icmp_seq=5 ttl=0 time=1.16 ms
Unlike Option A, there is no need to use VRF between ASBRs: when passing traffic between ASBRs, an MPLS
label will be assigned. This scheme has the best scalability.
In the current implementation, Option B is only supported for VPN-IPv4 routes (AFI = 1, SAFI = 128).