MPLS Redistribution between BGP and OSPF

[Evlaa2021]

In my lab configuration i am not able to redistribute between BGP and OSPF in an MPLS type setup, i've attached a topology diagram and full configs for reference. On my PE routers i can see that they are learning routes from the CE/CPE routers via OSPF, example below. Routes are being redistributed into BGP but then at the remote side of the MPLS VPN they do not get put back into OSPF.

PE1#show ip route vrf MFT_CORP 192.168.200.0
Routing entry for 192.168.200.0/32, 1 known subnets
Redistributing via bgp 50

O 192.168.200.1
[110/2] via 192.168.1.2, 00:06:12, GigabitEthernet3/0.100

This is a loopback interface on CPE 1

The routes are being learned across the BGP neighborship and passed to PE2 but not redistributed back into OSPF.

PE1#show run | sec router
router ospf 100 vrf MFT_CORP
router-id 55.55.55.55
log-adjacency-changes
redistribute bgp 50 subnets
router ospf 20 vrf HSCN
router-id 5.5.5.5
log-adjacency-changes
redistribute bgp 50 subnets
router ospf 1
mpls ldp autoconfig
log-adjacency-changes
router bgp 50
no synchronization
bgp log-neighbor-changes
bgp redistribute-internal
neighbor 10.0.0.6 remote-as 50
neighbor 10.0.0.6 update-source Loopback0
no auto-summary
!
address-family vpnv4
neighbor 10.0.0.6 activate
neighbor 10.0.0.6 send-community extended
exit-address-family
!
address-family ipv4 vrf MFT_CORP
redistribute ospf 100 vrf MFT_CORP
no synchronization
exit-address-family
!
address-family ipv4 vrf HSCN
redistribute ospf 20 vrf HSCN
no synchronization
exit-address-family
mpls ldp router-id Loopback0

PE2(config-router)#do show run | sec router
router ospf 100 vrf MFT_CORP
router-id 66.66.66.66
log-adjacency-changes
redistribute bgp 50 subnets
router ospf 20 vrf HSCN
router-id 6.6.6.6
log-adjacency-changes
redistribute bgp 50 subnets
router ospf 1
mpls ldp autoconfig
log-adjacency-changes
router bgp 50
no synchronization
bgp log-neighbor-changes
bgp redistribute-internal
neighbor 10.0.0.5 remote-as 50
neighbor 10.0.0.5 update-source Loopback0
no auto-summary
!
address-family vpnv4
neighbor 10.0.0.5 activate
neighbor 10.0.0.5 send-community extended
exit-address-family
!
address-family ipv4 vrf MFT_CORP
redistribute ospf 100 vrf MFT_CORP
no synchronization
exit-address-family
!
address-family ipv4 vrf HSCN
redistribute ospf 20 vrf HSCN
no synchronization
exit-address-family
mpls ldp router-id Loopback0

Recently added command 'bgp redistribute-internal' as i read iBGP learned routes will not redistribute into an IGP, i've never had to add this command before though?

[MHM Cisco World]

Mr.Harold and Mr.Giuseppe

After I make deep study to MPLS multi-VRF I found 

1- Down Bit effect ALL OPSF aware VRF  router 

2- route target import/export need only when we use BGP as protocol between PE-CE

the point is I don't get why we need MPLS label push toward CE, why we don't depend on L2 tag to forward different traffic for different VRF?

[Harold Ritter]

Hi @MHM Cisco World ,

Happy New Year!

> 2- route target import/export need only when we use BGP as protocol between PE-CE

You do not need to configure the import/export route target even if you use BGP between the PE and the CE. VRF lite provides isolation by creating a separate VRF table for each VRF no matter what is the routing protocol used between the PE and the CE.

> the point is I don't get why we need MPLS label push toward CE, why we don't depend on L2 tag to forward different traffic > for different VRF?

In a VRF lite scenario, you do not propagate labels between the PE CE. You do depend on the L2 tag to provide isolation between the different VRFs.

Regards,

[MHM Cisco World]

Happy New Year

Thanks a lot for your answer

wish you all success this Year.