cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
644
Views
0
Helpful
2
Replies

Unable to ping external n/w under ebgp from ibgp

babaim2010
Level 1
Level 1

Dear Team,

 

As per attached diag, I a unable to ping 5.5.5.5 from R1 where I am getting *> routes of 5.5.5.5

I have used next-hop-self command at R4

I don't want to redistribute external routes into OSPF which is running among R1-R2-R3-R4 for reachablity only.

If I do so, I am able to ping 5.5.5.5 because of getting O E2 routes at R1 but then, external routes will also be flowed to R2 & R3 as well that I don't want.

Please help me giving resolution, suggestion & guidance.

Note : I have just used 9.9.9.9 at R1 , assuming a internal n/w configuring Lo

 

Rgds

Kaushik Mitra

1 Accepted Solution

Accepted Solutions

trfinkenstadt
Level 1
Level 1

Kaushik,


If I understood everything completely, you are not running BGP on R2 and R3, which are the "corners" that separate R1 and R4 on your drawing.  Without redistribution of 5.5.5.5 at R4 into OSPF, then R2 and R3 do not know how to reach 5.5.5.5 and follow their default routes if they have one.  Otherwise you probably are receiving a !U (i think) in a traceroute.  

 

We have run into a similar problem when you have L3 switches that have default only with /30s to two routers with different connectivity.

 

Ways to fix this:

1.  Establish a GRE tunnel between R1 and R4 to be able to have a direct L3 link between R1 and R4.  Be careful when doing this because you can get into a recursive loop if you aren't careful.

 

2.  Get R1 and R4 to be on the same L3 subnet to be able to route 5.5.5.5 directly.

 

HTH,

 

 

Tim

 

View solution in original post

2 Replies 2

trfinkenstadt
Level 1
Level 1

Kaushik,


If I understood everything completely, you are not running BGP on R2 and R3, which are the "corners" that separate R1 and R4 on your drawing.  Without redistribution of 5.5.5.5 at R4 into OSPF, then R2 and R3 do not know how to reach 5.5.5.5 and follow their default routes if they have one.  Otherwise you probably are receiving a !U (i think) in a traceroute.  

 

We have run into a similar problem when you have L3 switches that have default only with /30s to two routers with different connectivity.

 

Ways to fix this:

1.  Establish a GRE tunnel between R1 and R4 to be able to have a direct L3 link between R1 and R4.  Be careful when doing this because you can get into a recursive loop if you aren't careful.

 

2.  Get R1 and R4 to be on the same L3 subnet to be able to route 5.5.5.5 directly.

 

HTH,

 

 

Tim

 

Thanks a lot dear for your support.

 

Kaushik