Need Help Redistributing OSPF in to BGP

I am having a bit of trouble in one of my labs trying to pass OSPF traffic in to BGP. In this scenario I have a simple OSPF Instance above and a few loopback interfaces with different subnets on the Hosts router. At the bottom I have a simple instance of BGP set up, all routers belong to the same AS, the clouds, which represent ISPs, each have a different AS. As far as I can tell both OSPF and BGP are configured correctly, I am distributing connected routes in BGP and can ping any router or ISP.

On the Edina router, I am redistributing connected in to BGP, I am also redistributing OSPF 1, which from what I understand should distribute inter and intra area routes from OSPF. I am also redistributing BGP in to OSPF so the Hosts router knows about the other subnets.

Looking at the top ISP (it's called L3, not sure why I edited that out) I can see that the OSPF routes are being passed to it, and are recognized as being BGP routes at that point, however, they do not make it any further than that, none of the routers or ISP below know about the routes that are being pushed in to BGP from OSPF. What am I missing here? Below are the configs from the routers and ISP involved.

Edina

Edina#sho run | b router
router ospf 1
log-adjacency-changes
redistribute bgp 100 metric-type 1 subnets
network 100.10.1.4 0.0.0.3 area 0
network 172.16.254.0 0.0.0.3 area 0
!
router bgp 100
no synchronization
bgp log-neighbor-changes
redistribute connected
redistribute ospf 1
neighbor 100.10.1.5 remote-as 200
no auto-summary

L3

L3#sho run | b router
router bgp 200
no synchronization
bgp log-neighbor-changes
redistribute connected
neighbor 100.10.1.6 remote-as 100
neighbor 100.10.1.18 remote-as 100
neighbor 100.10.1.42 remote-as 100
no auto-summary

Chennai

Chennai#sho run | b router
router bgp 100
no synchronization
bgp log-neighbor-changes
redistribute connected
neighbor 100.0.0.1 remote-as 300
neighbor 100.10.1.41 remote-as 200
no auto-summary

Bangalore

Bangalore#sho run | b router
router bgp 100
no synchronization
bgp log-neighbor-changes
redistribute connected
neighbor 100.0.0.5 remote-as 300
neighbor 100.10.1.17 remote-as 200
no auto-summary

Capture.jpg

Find more posts tagged with

Comments

CodeBlox

I'm gonna take a stab at it based on the info given and say it is due to the AS's you're using. Edina is using AS 100 and sends to AS 200 (L3). From there, it would be expected that Chennai and Bangalor would learn these routes. The issue I believe you are experiences are loop prevention mechanisms in BGP since Chennai and Bangalor are also using AS 100. BGP will not add routes to the BGP table if the AS-Path from eBGP contains it's own AS. I would say you'd need to pick another AS for either Edina or both Chennai and Bangalor. Only 1 of them should be using AS 100.

Danielh22185

CodeBlox wrote: »

I'm gonna take a stab at it based on the info given and say it is due to the AS's you're using. Edina is using AS 100 and sends to AS 200 (L3). From there, it would be expected that Chennai and Bangalor would learn these routes. The issue I believe you are experiences are loop prevention mechanisms in BGP since Chennai and Bangalor are also using AS 100. BGP will not add routes to the BGP table if the AS-Path from eBGP contains it's own AS. I would say you'd need to pick another AS for either Edina or both Chennai and Bangalor. Only 1 of them should be using AS 100.

Correct the Lvl3 Router has no idea what to do with sending the traffic because it learned the routes from AS 100 and you currently have Chennai and Bangalore in the same AS as which the routes were initially learned from.

I went ahead and annotated it so you can see what we are talking about.

Capture.JPG

bugzy3188

That was it, I have been scratching my head over that for a few days, thanks!

Danielh22185

bugzy3188 wrote: »

That was it, I have been scratching my head over that for a few days, thanks!

Easy mistake. Happens to us all when we are grinding out the labs! Always best to take a step back and get a second opinion if you are stuck.