05-27-2009 07:30 AM - edited 03-04-2019 04:54 AM
Hi All, I have inherited a network. One location has two routers, one is in OSPF Area 0 and the second is configured as a stub area. The second router has an external link to BGP also and is trying to redistribute ospf into BGP. This is not working. Would changing the stub area to a Not-So-Stubby-Area then allow me to redistribute into BGP?
Cheers
Brian
Solved! Go to Solution.
05-28-2009 08:03 AM
Brian
Thank you for the additional information. This does confirm that there are OSPF routes in the routing table which could be redistributed into BGP. If they are not being redistributed then we need to find the cause why not. In a previous post you gave this as the config of BGP:
router bgp 65500
no synchronization
bgp log-neighbor-changes
redistribute ospf 6 metric 100 route-map ospf2bgp
neighbor x.x.x.x remote-as xxxx
no auto-summary
is this still accurate?
If so my guess would be an issue with the route map, since the rest of the config seems pretty straight forward and without problem. Can you post the route map and any associated access lists?
HTH
Rick
05-27-2009 07:39 AM
Brian
If you are redistributing from OSPF into BGP then the stub area should not make any difference. It would matter if you were trying to redistribute from BGP into OSPF. If the redistribution is not working then there must be some other explanation. Perhaps if you post the config of the router we might be able to identify what the problem is.
HTH
Rick
05-27-2009 07:58 AM
Hi Rick, I am also going to be trying to sort out BGP to OSPF redistribution. It is supposed to be working both ways but figured I would tackle one direction first. Config looks as follows:
router ospf 6
router-id x.x.x.x
auto-cost reference-bandwidth 10000
area 38 stub no-summary
timers throttle spf 10 100 5000
timers throttle lsa all 10 100 5000
redistribute bgp 65500 metric 500 subnets
network x.x.x.x 0.0.0.255 area 38
router bgp 65500
no synchronization
bgp log-neighbor-changes
redistribute ospf 6 metric 100 route-map ospf2bgp
neighbor x.x.x.x remote-as xxxx
no auto-summary
I have verified that the route-maps are allowing the correct subnets.
Cheers
Brian
05-27-2009 07:59 AM
Hi Rick, I am also going to be trying to sort out BGP to OSPF redistribution. It is supposed to be working both ways but figured I would tackle one direction first. Config looks as follows:
router ospf 6
router-id x.x.x.x
auto-cost reference-bandwidth 10000
area 38 stub no-summary
timers throttle spf 10 100 5000
timers throttle lsa all 10 100 5000
redistribute bgp 65500 metric 500 subnets
network x.x.x.x 0.0.0.255 area 38
router bgp 65500
no synchronization
bgp log-neighbor-changes
redistribute ospf 6 metric 100 route-map ospf2bgp
neighbor x.x.x.x remote-as xxxx
no auto-summary
I have verified that the route-maps are allowing the correct subnets.
Cheers
Brian
05-27-2009 09:30 AM
Hello Brian,
as Rick has noted you cannot redistribute from BGP into OSPF in a router internal to a totally stub area.
area 38 stub no-summary
Second point about red of OSPF into BGP:
it is possible to redistribute OSPF into BGP but be aware that OSPF DB of this router is almost empty:
it contains only routes internal to area 38 and a default route.
So in comparison to a router in area 0 the DB is for sure incomplete you should move this router to a standard area to have results comparable to the other node in area 0
Hope to help
Giuseppe
05-27-2009 10:52 AM
Brian
I certainly agree with the strategy of dividing the problem and solving one part (OSPF to BGP) before you tackle the other part (BGP to OSPF). Based on what we know so far, that this appears to be the only router in OSPF area 38 and that the area is totally stubby I believe that the suggestion from Giuseppe is the most likely explanation - that there is very little, or nothing, in the OSPF routing table to redistribute.
If you are not satisfied that this is the explanation then it might help if you would post the output of show ip route from this router and indicate which routes you expect to be redistributed into BGP.
HTH
Rick
05-27-2009 02:13 PM
Hi Brian,
BGP-> OSPF
Yes. You need to use area 38 nssa no-summary to be able to redistribute BGP into OSPF. This is primarily what NSSA is used for.
OSPF->BGP
As mentioned by Rick and Giuseppe , if you dont have internal routes in area 38 you will end up redistributing only OSPF enabled directly connected inteffaces on the ASBR.
HTH,
-Rakesh
05-28-2009 01:44 AM
Hi Guys, appreciate all the responses. Ok, the topology is as follows:
Two switches connected via Layer 3 links to Gateway 1. Gateway 1 is the ABR and is advertising itself as the default gateway into the stub area. Gateway 2 which is the one I am concerned with is directly connected to Gateway 1 and to a BGP cloud. Gateway 2 knows about the two switches as they are all in area 38 but doesn't redistribute these routes into BGP. So what I am hoping to acheive is to get the route to the two switches via Gateway 2 redistributed into BGP. Hope I am explaining myself ok.
Cheers
Brian
05-28-2009 04:30 AM
Brian
In your explanation you have not mentioned area 0. Is there an area 0 or is everything in area 38? (if gateway 2 knows about the switches connected to gateway 1 it would suggest that they are all in area 38)
It would be very helpful if you would post the output of show ip route from gateway 2 and to indicate which routes are the ones that you want to be redistributed.
HTH
Rick
05-28-2009 04:44 AM
HI Rick, Here is the output of show ip route ospf on Gateway 2, I would like all these routes redistributed into BGP:
10.0.0.0/8 is variably subnetted, 84 subnets, 4 masks
O 10.99.202.1/32 [110/106] via 192.168.236.197, 2d09h,
FastEthernet0/0
O 10.99.203.1/32 [110/111] via 192.168.236.197, 2d09h,
FastEthernet0/0
O 10.104.178.0/24 [110/115] via 192.168.236.197, 2w4d,
FastEthernet0/0
O 10.104.177.0/24 [110/120] via 192.168.236.197, 2w4d,
FastEthernet0/0
O 10.102.177.0/24
[110/120] via 192.168.236.197, 20:52:50, FastEthernet0/0
O 10.102.178.0/24 [110/115] via 192.168.236.197, 2w4d,
FastEthernet0/0
192.168.236.0/24 is variably subnetted, 6 subnets, 2 masks
O 192.168.236.180/30
[110/105] via 192.168.236.197, 2w4d, FastEthernet0/0
O 192.168.236.184/30
[110/110] via 192.168.236.197, 2w4d, FastEthernet0/0
Gateway 1 is in Area 0 and Area 38, all other devices are in area 38.
Cheers
Brian
05-28-2009 08:03 AM
Brian
Thank you for the additional information. This does confirm that there are OSPF routes in the routing table which could be redistributed into BGP. If they are not being redistributed then we need to find the cause why not. In a previous post you gave this as the config of BGP:
router bgp 65500
no synchronization
bgp log-neighbor-changes
redistribute ospf 6 metric 100 route-map ospf2bgp
neighbor x.x.x.x remote-as xxxx
no auto-summary
is this still accurate?
If so my guess would be an issue with the route map, since the rest of the config seems pretty straight forward and without problem. Can you post the route map and any associated access lists?
HTH
Rick
05-28-2009 08:12 AM
Ah Jasus you are not going to believe this. The access-list is specified but no route-map!!! Staring at a problem too long and miss the obvious! Thanks for your help Rick, wouldn't have figured it out otherwise!
05-28-2009 08:26 AM
Brian
Sometimes it is the simple and obvious things that are the most difficult to find. I am glad that you got it resolved and that my suggestion pointed you in the right direction. Thank you for using the rating system to indicate that you have resolved the issue (and thanks for the rating). It makes the forum more useful when people can read about a problem and can know that they will read the solution to the problem.
The forum is an excellent place to get help and to learn about Cisco networking. I encourage you to continue your participation in the forum.
HTH
Rick
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide