NG-MVPN Inter-AS Model B Using Non-Segmented mLDP Tunnels

The first step is to configure the router interfaces and IS-IS as the Interior Gateway Protocol (IGP) in AS 64496.

The router interfaces for PE-1 are configured as follows:

# on PE-1:
configure {
    router "Base" {
        interface "int-PE-1-P-2" {
            port 1/1/c1/1
            ipv4 {
                primary {
                    address 192.168.12.1
                    prefix-length 30
                }
            }
        }
        interface "system" {
            ipv4 {
                primary {
                    address 192.0.2.1
                    prefix-length 32
                }
            }
        }
    }
}

Each interface is configured to run IS-IS as the IGP. Each router is configured as a level 2 router.

# on PE-1:
configure {
    router "Base" {
        isis 0 {
            admin-state enable
            level-capability 2
            traffic-engineering true
            area-address [49.0001]
            interface "int-PE-1-P-2" {
                interface-type point-to-point
                level-capability 2
            }
            interface "system" {
                level-capability 2
            }
            level 2 {
                wide-metrics-only true
            }
        }
    }
}

The configuration for all other nodes in the AS is the same, apart from the IP addresses. The IP addresses can be derived from NG-MVPN Inter-AS Model B.

Label Distribution Protocol (LDP) is used as the MPLS protocol and must be enabled on each router interface, as follows:

# on PE-1:
configure {
    router "Base" {
        ldp {
            import-pmsi-routes {
                mvpn true    # needed for bindings
            }
            interface-parameters {
                interface "int-PE-1-P-2" {
                    ipv4 {
                        fec-type-capability {
                            p2mp-ipv4 true        # default
                        }
                    }
                }
            }
        }
    }
}

This configuration must be repeated on each of the other routers in the AS. As LDP is used as the provider tunnel interface for multicast traffic, each interface must also support P2MP LDP tunnels. Therefore, the FEC type capability for IPv4 P2MP tunnels must be enabled. The default value is enable, but is included in the preceding configuration for clarity.

AS 64497 Protocols shows the protocol requirements for AS 64497.

The first step is to configure router interfaces and OSPF on each router shown in AS 64497 Protocols. All router interfaces are members of a single backbone area: area 0.0.0.0.

The following router interfaces are configured on PE-6:

# on PE-6:
configure {
    router "Base" {
        interface "int-PE-6-ASBR-5" {
            port 1/1/c2/1
            ipv4 {
                primary {
                    address 192.168.56.2
                    prefix-length 30
                }
            }
        }
        interface "system" {
            ipv4 {
                primary {
                    address 192.0.2.6
                    prefix-length 32
                }
            }
        }
    }
}

The configuration for PE-6 to enable OSPF is:

# on PE-6:
configure {
    router "Base" {
        ospf 0 {
            area 0.0.0.0 {
                interface "int-PE-6-ASBR-5" {
                    interface-type point-to-point
                }
                interface "system" {
                    interface-type point-to-point
                }
            }
            admin-state enable
        }
    }
}

LDP is used as the MPLS protocol and must be enabled on each router interface, as follows.

# on PE-6:
configure {
    router "Base" {
        ldp {
            interface-parameters {
                interface "int-PE-6-ASBR-5" {
                    ipv4 {
                        fec-type-capability {
                            p2mp-ipv4 true        # default
                        }
                    }
                }
            }
        }
    }
}

This configuration must be repeated on each of the other routers in the AS. Again, the default value of FEC type capability for P2MP is enable, but is included for clarity.

Within AS 64497, internal BGP peering is required between ASBR-5 and PE-6 for the VPN-IPv4 and MVPN-IPv4 address families.

The following shows the BGP configuration for such a peering.

# on ASBR-5:
configure {
    router "Base" {
        autonomous-system 64497
        bgp {
            family {
                ipv4 false
                vpn-ipv4 true
                mvpn-ipv4 true
            }
            ebgp-default-reject-policy {
                import false
                export false
            }
            group "internal" {
                type internal
            }
            neighbor "192.0.2.6" {
                group "internal"
            }
        }
    }
}

# on PE-6:
configure {
    router "Base" {
        autonomous-system 64497
        bgp {
            rapid-withdrawal true
            ebgp-default-reject-policy {
                import false
                export false
            }
            rapid-update {
                vpn-ipv4 true
                mvpn-ipv4 true
            }
            group "internal" {
                type internal
                family {
                    vpn-ipv4 true
                    mvpn-ipv4 true
                }
            }
            neighbor "192.0.2.5" {
                group "internal"
            }
        }
    }
}

The following shows the eBGP peering configuration for ASBR-4. The peer address is the interface address of ASBR-5.

# on ASBR-4:
configure {
    router "Base" {
        autonomous-system 64496
        bgp {
            inter-as-vpn true    # required for inter-AS VPN option B
            split-horizon true
            family {
                ipv4 false
                vpn-ipv4 true
                mvpn-ipv4 true
            }
            ebgp-default-reject-policy {
                import false
                export false
            }
            group "external" {
                peer-as 64497
            }
            neighbor "192.168.45.2" {
                group "external"
            }
        }
    }
}

Similarly, the BGP configuration for ASBR-5 peering toward ASBR-4 is as follows:

# on ASBR-5:
configure {
    router "Base" {
        autonomous-system 64497
        bgp {
            inter-as-vpn true    # required for inter-AS VPN option B
            split-horizon true
            family {
                ipv4 false
                vpn-ipv4 true
                mvpn-ipv4 true
            }
            ebgp-default-reject-policy {
                import false
                export false
            }
            group "external" {
                peer-as 64496
            }
            neighbor "192.168.45.1" {
                group "external"
            }
        }
    }
}

Verification of the BGP peering session between ASBR-4 and ASBR-5 is shown in the following output:

# on ASBR-4:
[/]
A:admin@ASBR-4# show router bgp summary group "external"
===============================================================================
 BGP Router ID:192.0.2.4        AS:64496       Local AS:64496      
===============================================================================
BGP Admin State         : Up          BGP Oper State              : Up
Total Peers             : 1         
 
Total IPv4 Remote Rts   : 0           Total IPv4 Rem. Active Rts  : 0         
Total IPv6 Remote Rts   : 0           Total IPv6 Rem. Active Rts  : 0         
Total IPv4 Backup Rts   : 0           Total IPv6 Backup Rts       : 0         
Total LblIpv4 Rem Rts   : 0           Total LblIpv4 Rem. Act Rts  : 0         
Total LblIpv6 Rem Rts   : 0           Total LblIpv6 Rem. Act Rts  : 0         
Total LblIpv4 Bkp Rts   : 0           Total LblIpv6 Bkp Rts       : 0          
 
Total VPN-IPv4 Rem. Rts : 0           Total VPN-IPv4 Rem. Act. Rts: 0         
Total VPN-IPv6 Rem. Rts : 0           Total VPN-IPv6 Rem. Act. Rts: 0         
Total VPN-IPv4 Bkup Rts : 0           Total VPN-IPv6 Bkup Rts     : 0         
 
Total MVPN-IPv4 Rem Rts : 0           Total MVPN-IPv4 Rem Act Rts : 0         
Total MVPN-IPv6 Rem Rts : 0           Total MVPN-IPv6 Rem Act Rts : 0         
Total MDT-SAFI Rem Rts  : 0           Total MDT-SAFI Rem Act Rts  : 0         
Total McIPv4 Remote Rts : 0           Total McIPv4 Rem. Active Rts: 0         
Total McIPv6 Remote Rts : 0           Total McIPv6 Rem. Active Rts: 0         
Total McVpnIPv4 Rem Rts : 0           Total McVpnIPv4 Rem Act Rts : 0         
Total McVpnIPv6 Rem Rts : 0           Total McVpnIPv6 Rem Act Rts : 0         
 
Total EVPN Rem Rts      : 0           Total EVPN Rem Act Rts      : 0         
Total L2-VPN Rem. Rts   : 0           Total L2VPN Rem. Act. Rts   : 0         
Total MSPW Rem Rts      : 0           Total MSPW Rem Act Rts      : 0         
Total RouteTgt Rem Rts  : 0           Total RouteTgt Rem Act Rts  : 0         
Total FlowIpv4 Rem Rts  : 0           Total FlowIpv4 Rem Act Rts  : 0         
Total FlowIpv6 Rem Rts  : 0           Total FlowIpv6 Rem Act Rts  : 0         
Total FlowVpnv4 Rem Rts : 0           Total FlowVpnv4 Rem Act Rts : 0         
Total FlowVpnv6 Rem Rts : 0           Total FlowVpnv6 Rem Act Rts : 0         
Total Link State Rem Rts: 0           Total Link State Rem Act Rts: 0         
Total SrPlcyIpv4 Rem Rts: 0           Total SrPlcyIpv4 Rem Act Rts: 0         
Total SrPlcyIpv6 Rem Rts: 0           Total SrPlcyIpv6 Rem Act Rts: 0         

===============================================================================
BGP Summary
===============================================================================
Legend : D - Dynamic Neighbor
===============================================================================
Neighbor
Description
                   AS PktRcvd InQ  Up/Down   State|Rcv/Act/Sent (Addr Family)
                      PktSent OutQ
-------------------------------------------------------------------------------
192.168.45.2
                64497      12    0 00h04m00s 0/0/0 (VpnIPv4)
                           12    0           0/0/0 (MvpnIPv4)
-------------------------------------------------------------------------------

LDP is configured as the MPLS protocol between ASBR-4 and ASBR-5. On ASBR-4, the interface toward ASBR-5 has LDP enabled, as follows:

# on ASBR-4:
configure {
    router "Base" {
        ldp {
            import-pmsi-routes {
                mvpn true    # needed for bindings
            }
            interface-parameters {
                interface "int-ASBR-4-ASBR-5" {
                    ipv4 {
                        fec-type-capability {
                            p2mp-ipv4 true
                        }
                        transport-address interface
                    }
                }
            }
        }
    }
}

The P2MP FEC type capability for P2MP LDP is shown. This is the default value.

For completeness, the LDP configuration on ASBR-5 for the interface toward ASBR-4 is as follows:

# on ASBR-5:
configure {
    router "Base" {
        ldp {
            interface-parameters {
                interface "int-ASBR-5-ASBR-4" {
                    ipv4 {
                        fec-type-capability {
                            p2mp-ipv4 true        # default
                        }
                        transport-address interface
                    }
                }
            }
        }
    }
}

Verification that the LDP session is successfully established at ASBR-4 is shown in the following output:

# on ASBR-4:
[/]
A:admin@ASBR-4# show router ldp session 192.0.2.5

==============================================================================
LDP IPv4 Sessions
==============================================================================
Peer LDP Id         Adj Type  State         Msg Sent  Msg Recv  Up Time
------------------------------------------------------------------------------
192.0.2.5:0         Link      Established   54        54        0d 00:01:54
------------------------------------------------------------------------------
No. of IPv4 Sessions: 1
==============================================================================
---snip---

For completeness, the LDP session from ASBR-5 toward ASBR-4 is shown in the following output:

# on ASBR-5:
[/]
A:admin@ASBR-5# show router ldp session 192.0.2.4

==============================================================================
LDP IPv4 Sessions
==============================================================================
Peer LDP Id         Adj Type  State         Msg Sent  Msg Recv  Up Time
------------------------------------------------------------------------------
192.0.2.4:0         Link      Established   52        54        0d 00:01:53
------------------------------------------------------------------------------
No. of IPv4 Sessions: 1
==============================================================================
---snip---

When a label mapping message is received for an LDP FEC prefix, the next-hop for a FEC prefix is resolved in the routing table. The FEC is installed in the Label Information Base (LIB) if the next-hop matches a /32 route table entry.

The local interface configuration results in a route being installed with a subnet mask matching the interface configuration. In this case, the ASBR-to-ASBR route is 192.168.45.0/30.

For LDP to resolve the LDP FEC egress next-hop on ASBR-4, a /32 route matching the egress next-hop address is required in the FIB.

On ASBR-4, a static route is configured for the /32 address on ASBR-5, as follows.

# on ASBR-4:
configure {
    router "Base" {
        static-routes {
            route 192.168.45.2/32 route-type unicast {
                next-hop "192.168.45.2" {
                    admin-state enable
                }
            }
        }
    }
}

Similarly, a static route on ASBR-5 is configured for the /32 address on ASBR-4, as follows.

# on ASBR-5:
configure {
    router "Base" {
        static-routes {
            route 192.168.45.1/32 route-type unicast {
                next-hop "192.168.45.1" {
                    admin-state enable
                }
            }
        }
    }
}

The following output shows that the static route is installed in the ASBR-4 RIB.

# on ASBR-4:
[/]
A:admin@ASBR-4# show router route-table protocol static

===============================================================================
Route Table (Router: Base)
===============================================================================
Dest Prefix[Flags]                            Type    Proto     Age        Pref
      Next Hop[Interface Name]                                    Metric   
-------------------------------------------------------------------------------
192.168.45.2/32                               Remote  Static    00h00m00s  5
       192.168.45.2                                                 1
-------------------------------------------------------------------------------
No. of Routes: 1
Flags: n = Number of times nexthop is repeated
       B = BGP backup route available
       L = LFA nexthop available
       S = Sticky ECMP requested
===============================================================================

# on PE-1:
configure {
    service {
        vprn "VPRN 1" {
            admin-state enable
            service-id 1
            customer "1"
            igmp {
                interface "int-PE-1-VPRN-1-H-1" { }
            }
            pim { }
            mvpn {
                c-mcast-signaling bgp
                mdt-type receiver-only
                auto-discovery {
                    type bgp
                }
                vrf-target {
                    unicast true
                }
                provider-tunnel {
                    inclusive {
                        mldp {
                            admin-state enable
                        }
                    }
                }
            }
            bgp-ipvpn {
                mpls {
                    admin-state enable
                    route-distinguisher "192.0.2.1:1"
                    vrf-target {
                        community "target:64496:1"
                    }
                    auto-bind-tunnel {
                        resolution filter
                        resolution-filter {
                            ldp true
                        }
                    }
                }
            }
            interface "int-PE-1-VPRN-1-H-1" {
                ipv4 {
                    primary {
                        address 172.16.1.1
                        prefix-length 30
                    }
                }
                sap 1/1/c3/1:1 { }
            }
        }
    }
}

# on PE-6:
configure {
    service {
        vprn "VPRN 1" {
            admin-state enable
            service-id 1
            customer "1"
            pim {
                interface "int-PE-6-VPRN-1-source" { }
            }
            mvpn {
                c-mcast-signaling bgp
                mdt-type sender-only
                auto-discovery {
                    type bgp
                }
                vrf-target {
                    unicast true
                }
                provider-tunnel {
                    inclusive {
                        mldp {
                            admin-state enable
                        }
                    }
                }
            }
            bgp-ipvpn {
                mpls {
                    admin-state enable
                    route-distinguisher "192.0.2.6:1"
                    vrf-target {
                        community "target:64496:1"
                    }
                    auto-bind-tunnel {
                        resolution filter
                        resolution-filter {
                            ldp true
                        }
                    }
                }
            }
            interface "int-PE-6-VPRN-1-source" {
                ipv4 {
                    primary {
                        address 172.16.67.1
                        prefix-length 30
                    }
                }
                sap 1/1/c1/1 { }
            }
        }
    }
}

The use of non-segmented LDP provider tunnels requires that Intra-AD Auto Discovery routes must be advertised across the AS boundary between PEs. Each intra-AD route generated by a PE that is a member of an MVPN contains the well-known community "No-Export", which prevents a BGP speaker from advertising the route across an AS boundary to another external BGP speaker.

In inter-AS mode B, the ASBR router must support the MVPN address family. If it receives an intra-AD route containing the No-Export community, it is not advertised to any external peer. A route policy is required to remove the No-Export community before it can be advertised across the AS boundary to a BGP speaker that has negotiated the MVPN address family capability.

In the following example configuration, the policy removes the No-Export community on PE-6, the source router from all advertised routes, by configuring the community remove action as a default action.

# on PE-6:
configure {
    policy-options {
        community "NoExport" {
            member "no-export" { }
        }
        policy-statement "RemNoExport" {
            default-action {
                action-type accept
                community {
                    remove ["NoExport"]
                }
            }
        }
    }
}

This is applied as an export policy, so that the No-Export community is removed from all intra-AD routes advertised as updates to internal peers. The vpn-apply-export true command must be included to ensure that the export policy is applied to routes belonging to VPN address families; in this case, MVPN-IPv4 routes.

# on PE-6:
configure {
    router "Base" {
        bgp {
            vpn-apply-export true
            export {
                policy ["RemNoExport"]
            }
        }
    }
}

This policy should also be configured and applied on PE-1, so that intra-AD routes can be exported from MVPN PEs in AS 64496 to AS 64497.

BGP MVPN Intra-AD Route Advertisement shows the propagation of the BGP MVPN intra-AD route from PE-6 to PE-1 across the AS boundary. The original route has the No-Export community removed on PE-6 because the export route policy is applied. ASBR-5 receives the route and forwards it to ASBR-4. The BGP next-hop attribute is changed at the AS boundary to the interface address of ASBR-5: 192.168.45.2. ASBR-4 forwards the intra-AD route to the RR on P-2, and changes the BGP next-hop attribute to its system address: 192.0.2.4. P-2 reflects the route to PE-1.

PE-1 receives the route, and imports the route into VPRN 1 as the route target extended community matches the community configured in the MVPN context of the VPRN. PE-1 now uses the PTA contained within the intra-AD route to instantiate the provider tunnel.

The following output shows details of the MVPN intra-AD route received by PE-1, generated by PE-6.

# on PE-1:
[/]
A:admin@PE-1# show router bgp routes mvpn-ipv4 type intra-ad originator-ip 192.0.2.6 hunt
===============================================================================
 BGP Router ID:192.0.2.1        AS:64496       Local AS:64496      
===============================================================================
 Legend -
 Status codes  : u - used, s - suppressed, h - history, d - decayed, * - valid
                 l - leaked, x - stale, > - best, b - backup, p - purge
 Origin codes  : i - IGP, e - EGP, ? - incomplete

===============================================================================
BGP MVPN-IPv4 Routes
===============================================================================
-------------------------------------------------------------------------------
RIB In Entries
-------------------------------------------------------------------------------
Route Type     : Intra-Ad               
Route Dist.    : 192.0.2.6:1
Originator IP  : 192.0.2.6
Nexthop        : 192.0.2.4
Path Id        : None                   
From           : 192.0.2.2
Res. Nexthop   : 0.0.0.0
Local Pref.    : 100                    Interface Name : NotAvailable
Aggregator AS  : None                   Aggregator     : None
Atomic Aggr.   : Not Atomic             MED            : None
AIGP Metric    : None                   IGP Cost       : n/a
Connector      : None
Community      : target:64496:1
Cluster        : 192.0.2.2
Originator Id  : 192.0.2.4              Peer Router Id : 192.0.2.2
Origin         : IGP                    
Flags          : Used Valid Best 
Route Source   : Internal
AS-Path        : 64497 
Route Tag      : 0                      
Neighbor-AS    : 64497
DB Orig Val    : N/A                    Final Orig Val : N/A
Source Class   : 0                      Dest Class     : 0
Add Paths Send : Default                
Last Modified  : 00h00m08s              
VPRN Imported  : 1
-------------------------------------------------------------------------------
PMSI Tunnel Attributes : 
Tunnel-type    : LDP P2MP LSP           
Flags          : Type: RNVE(0) BM: 0 U: 0 Leaf: not required
MPLS Label     : 0                      
Root-Node      : 192.0.2.6              LSP-ID         : 8193
-------------------------------------------------------------------------------
 
-------------------------------------------------------------------------------
RIB Out Entries
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
Routes : 1
===============================================================================

The PTA lists the tunnel type as a P2MP LDP LSP. A P2MP label mapping message is originated on PE-1, with LSP-ID 8193, and the root of the mLDP tree is PE-6: 192.0.2.6. However, PE-1 does not have a route to PE-6, because inter-AS model B VPNs do not require the system addresses of the PEs to be advertised into the neighboring AS.

The intra-AD MVPN route is used to determine the path of the label mapping message from PE-1 toward PE-6. This is comparable to the unicast routing case, where a VPN-IPv4 labeled route is used to determine the path to the source.

The BGP next-hop of the intra-AD route is the system address of ASBR-4, so this can be used as the root address of the mLDP LSP, and the actual root can be contained inside the label mapping message as an inner root. The inner root becomes an opaque value that is known to the originator and receiver of the label mapping message.

P2MP LDP Label Mapping shows the path taken by the label mapping message from PE-1 to PE-6.

P-2 and P-3 do not have either a unicast or multicast (intra-AD) route toward PE-6, but have a route to the outer root ASBR-4. The label mapping message is forwarded from PE-1 to ASBR-4 via P-2 and P-3. At each hop, a label is allocated and a label binding entry is created. The labels can be different on PE-1, P-2, P-3, ASBR-4, ASBR-5, and PE-6. In this case, the labels are: {lbl-A}=524283, {lbl-B}=524283, {lbl-C}=524283, {lbl-D}=524279, and {lbl-E}=524279. In the following sections, the debug outputs are achieved using the following debug command:

debug
    router "Base"
        ldp
            peer <peer-ip-address>
                packet
                    label detail
                exit
            exit
        exit
    exit
exit

where <peer-ip-address> is the system address of the LDP peer.

Traffic is forwarded into multicast group 239.255.0.1 from the source using address 172.16.67.2.

An IGMPv3 group membership report is generated by the receiver H-1 and is shown on PE-1.

# on PE-1:
[/]
A:admin@PE-1# show router "1" igmp group
===============================================================================
IGMP Interface Groups
===============================================================================
 
(172.16.67.2,239.255.0.1)                                 UpTime: 0d 00:02:38
    Fwd List  : int-PE-1-VPRN-1-H-1 
-------------------------------------------------------------------------------
Entries : 1
===============================================================================
---snip---

The status of the PIM group for VPRN 1 for group 239.255.0.1 is shown in the following output.

# on PE-1:
[/]
A:admin@PE-1# show router "1" pim group

===============================================================================
Legend:  A = Active   S = Standby
===============================================================================
PIM Groups ipv4
===============================================================================
Group Address               Type                Spt Bit  Inc Intf       No.Oifs
   Source Address              RP                  State    Inc Intf(S) 
-------------------------------------------------------------------------------
239.255.0.1                 (S,G)                        mpls-if-73728  1
   172.16.67.2                                                             
-------------------------------------------------------------------------------
Groups : 1
===============================================================================

The incoming interface is an MPLS interface: mpls-if-73728. This is a PIM tunnel interface, as shown in the following output:

# on PE-1:
[/]
A:admin@PE-1# show router "1" pim tunnel-interface

===============================================================================
PIM Interfaces ipv4
===============================================================================
Interface                        Originator Address   Adm  Opr  Transport Type
-------------------------------------------------------------------------------
mpls-if-73728                    192.0.2.6            Up   Up   Rx-IPMSI
mpls-virt-if-1005857             192.0.2.1            Up   Up   Tx-IPMSI
-------------------------------------------------------------------------------
Interfaces : 2
===============================================================================

The originator address is 192.0.2.6, which is the root address of the mLDP tunnel on PE-6-the non-segmented mLDP tunnel.

For completeness, the PIM status of the group 239.255.0.1 on PE-6 is as follows:

# on PE-6:
[/]
A:admin@PE-6# show router "1" pim group

===============================================================================
Legend:  A = Active   S = Standby
===============================================================================
PIM Groups ipv4
===============================================================================
Group Address               Type                Spt Bit  Inc Intf       No.Oifs
   Source Address              RP                  State    Inc Intf(S) 
-------------------------------------------------------------------------------
239.255.0.1                 (S,G)                        int-PE-6-VPRN* 1
   172.16.67.2                                                             
-------------------------------------------------------------------------------
Groups : 1
===============================================================================
* indicates that the corresponding row element may have been truncated.