BGP Route Not Redistributing Into OSPF

[Robert Juric]

I have a BGP route, being learned from a Route-Reflector-Client, that should be advertised into OSPF but it is not. It shows redistributing via ospf, but I'm unable to see it in the OSPF database and when I go the next-hop router it is not being learned there either. Normal redistribution is working fine, but this route-reflector-client peer was just turned up and this isn't working.

router#sh ip ro 192.168.227.192
Routing entry for 192.168.227.192/26
  Known via "bgp 65006", distance 200, metric 0
  Tag 65506, type internal
  Redistributing via ospf 1001
  Last update from 192.168.224.82 2d10h ago
  Routing Descriptor Blocks:
  * 192.168.224.82, from x.x.5.113, 2d10h ago
      Route metric is 0, traffic share count is 1
      AS Hops 1
      Route tag 65506

router#sh ip bgp 192.168.227.192/26
BGP routing table entry for 192.168.227.192/26, version 7041
Paths: (2 available, best #2, table Default-IP-Routing-Table)
  Advertised to update-groups:
     1          2
  65506
    192.168.224.82 from 10.255.2.7 (10.255.2.7)
      Origin IGP, localpref 100, valid, internal
      Originator: x.x.5.150, Cluster list: 10.255.2.7
  65506, (Received from a RR-client), (received & used)
    192.168.224.82 from x.x.5.113 (x.x.5.150)
      Origin IGP, localpref 100, valid, internal, best
      Community: 1:1 1:3
      Extended Community: RT:4093:4093

router#sh ip ospf database | i 192.168.227.192
router#

The OSPF config:

router ospf 1001
 router-id x.x.0.246
 log-adjacency-changes
 area x.x.254.0 authentication
 redistribute connected subnets
 redistribute static subnets route-map Static-to-OSPF
 redistribute bgp 65006 subnets
 network x.x.0.243 0.0.0.0 area x.x.254.0
 network x.x.5.24 0.0.0.3 area x.x.254.0
 network x.x.59.0 0.0.0.255 area x.x.254.0
 network 172.29.29.0 0.0.0.255 area x.x.254.0
 network 172.29.59.0 0.0.0.255 area x.x.254.0
!
 
route-map Static-to-OSPF permit 10
 match tag 10
!

[Jon Marshall · Accepted solution]

Robert

BGP does not by default redistribute IBGP routes into an IGP.

For this to happen you need to add this under your BGP configuration, assuming you haven't already done it -

router bgp <AS no>
bgp redistribute-internal

however there is a good reason why it does not redistribute by default and that is because it can cause routing loops if you are not careful.

Cisco recommend to always use route maps to control exactly which IBGP routes are redistributed and even then only you can say whether it will cause a routing issue.

If you already have that command apologies for wasting your time :)

Jon

[Robert Juric]

Yep, that was it!

Thanks!