Prerequisites for MPLS VPN
None.
Restrictions for MPLS VPN
- When configuring MPLS VPN, note that VPNs are recirculated when the number of VPNs is over 511.
- MPLS VPN supports these commands:
–
address-family
–
exit-address-family
–
import map
–
ip route vrf
–
ip route forwarding
–
ip vrf
–
neighbor activate
–
rd
–
route-target
For information about these commands, see these publications:
http://www.cisco.com/en/US/products/ps11846/prod_command_reference_list.html
Cisco IOS Release 15.4SY supports only Ethernet interfaces. Cisco IOS Release 15.4SY does not support any WAN features or commands.
Information About MPLS VPN Support
The IP VPN feature for MPLS allows a Cisco IOS network to deploy scalable IP Layer 3 VPN backbone services to multiple sites deployed on a shared infrastructure while also providing the same access or security policies as a private network. VPN based on MPLS technology provides the benefits of routing isolation and security, as well as simplified routing and better scalability. See this publication for more information about MPLS VPNs:
http://www.cisco.com/en/US/docs/ios-xml/ios/mpls/config_library/15-sy/mp-15-sy-library.html
Figure 38-1 VPNs with MPLS Service Provider Backbone
At the ingress PE, the PFC makes a forwarding decision based on the packet headers. The PFC contains a table that maps VLANs to VPNs. In the switch architecture, all physical ingress interfaces in the system are associated with a specific VPN. The PFC looks up the IP destination address in the CEF table but only against prefixes that are in the specific VPN. (The table entry points to a specific set of adjacencies and one is chosen as part of the load-balancing decision if multiple parallel paths exist.)
The table entry contains the information on the Layer 2 header that the packet needs, as well as the specific MPLS labels to be pushed onto the frame. The information to rewrite the packet goes back to the ingress module where it is rewritten and forwarded to the egress line interface.
VPN traffic is handled at the egress from the PE based upon the per-prefix labels or aggregate labels. If per-prefix labels are used, then each VPN prefix has a unique label association; this allows the PE to forward the packet to the final destination based upon a label lookup in the FIB.
Note
The PFC allocates only one aggregate label per VRF.
If aggregate labels are used for disposition in an egress PE, many prefixes on the multiple interfaces may be associated with the label. In this case, the PFC must perform an IP lookup to determine the final destination. The IP lookup may require recirculation.
Configuration Example for MPLS VPNs
This sample configuration shows LAN CE-facing interfaces. MPLS switching configuration in Cisco IOS Release 15.4SY is identical to configuration in other releases.
route-target export 100:1
route-target import 100:1
mpls ldp logging neighbor-changes
ip address 10.4.4.4 255.255.255.255
interface GigabitEthernet4/2
description Catalyst link to P2
interface GigabitEthernet4/2.42
ip address 10.0.3.2 255.255.255.0
interface GigabitEthernet7/3
description Catalyst link to CE2
interface GigabitEthernet7/3.73
ip address 10.19.7.1 255.255.255.0
network 10.4.4.4 0.0.0.0 area 0
network 10.0.0.0 0.0.255.255 area 0
router ospf 65000 vrf blues
redistribute bgp 100 subnets
network 10.19.0.0 0.0.255.255 area 0
neighbor 10.3.3.3 remote-as 100
neighbor 10.3.3.3 description MP-BGP to PE1
neighbor 10.3.3.3 update-source Loopback0
neighbor 10.3.3.3 activate
neighbor 10.3.3.3 send-community extended
address-family ipv4 vrf blues
redistribute ospf 65000 match internal external 1 external 2