Someone can hopefully help as I know this is predominantly Cisco.
I have a Juniper quetion (also valid for Cisco I expect). I have two customer using the same topology as below:
CustA: CE → PE → P → P → PE → CE: CustA:SiteB
CustB: CE → PE → P → P → PE → CE: CustB:SiteB
Customer A utilises eBGP between the CE and the PE and the VONv4 is created with no issue and it works between the end points.
The porblem is with Customer B who uses OSPF between the CE and the PE and the config (juniper) is as follows for the CE and PE (Basic):
Customer B - CE
set interfaces ge-0/0/0 unit 0 family inet address 10.10.10.2/30
set interfaces lo0 unit 0 family inet address 3.3.3.3/32
set routing-options router-id 3.3.3.3
set protocols ospf area 0.0.0.0 interface ge-0/0/0.0
set protocols ospf area 0.0.0.0 interface lo0.0 passive
PE Router:
set routing-instances london-custA instance-type vrf
set routing-instances london-custA interface ge-0/0/0.0
set routing-instances london-custA route-distinguisher 65533:10
set routing-instances london-custA vrf-target target:65533:10
set routing-instances london-custA protocols ospf area 0.0.0.0 interface ge-0/0/0.0
So, I think what is happening is that the routes received from the CE in the PE vrf are actually IPv4 routes and they need distributing into MP-BGP. The problem I have is I cannot find any docs to suggest how that is completed within an actual vrf with the iBGP being configured external to the vrf in global config.
Can anyone point me in the right direction please?