Question about IBGP, OSPF relationship

[west33637]

This is a question of how IBGP operates with OSPF.

Firslty, synchronization has been disabled.

My understanding is that IBGP will check its BGP table for best next-hop to its intended destination. Then after that best next-hop is determined, BGP will then tell OSPF here is my best next hop. Get me there! Thats when the recursive routing now takes place. Is this correct?

Please no guess work. Answer, only if you are sure. Thanks,

[Giuseppe Larosa]

Hello west33637,

so it is in provider networks for successful forwarding one of the following conditions has to be met:

a) or iBGP is enabled everywhere using  route reflector servers and/or BGP confederations to avoid internal black holes

b) or MPLS is enabled, an LSP exists with destination = BGP next-hop, in this case a BGP free core can exist and internal MPLS routers don't need to know how to route the packet just how to switch the MPLS frame that carries the packet inside itself.

in several providers both conditions are true if a BGP free core is not  present.

with MPLS two levels of recursion can happen:

BGP choices the BGP next-hop

the BGP next-hop is used to check IGP routing table and MPLS forwarding table

traffic is sent over an LSP with destination = BGP next hop

the IP packet is encapsulated in an MPLS frame  with label value = the value expected by next router in the path to destination

Hope to help

Giuseppe

[west33637]

Hello giuslar. Thx for the detailed response. Taking MPLS and route reflectors/confeds out of the equation, can I assume from your response that my initial assumption of the workings between IBGP and OSPF are correct? Thanks,

[Giuseppe Larosa · Accepted solution]

Hello,

yes, your understanding or recursion is correct

Hope to help

Giuseppe