0

Topology

Hi, I am working with 4 routers: A B, P and C

A and B have a P router between them that has an ospf neighborship with both. B is directly connected to router C. There's no routing protocol between B and C. Network 172.16.55. /31 is utilized in between router B and C. with .50 configured on B and .51 on C.

Router C utilises a default route towards B.

What I am trying to achieve is to be able to ping loopback 3.3.3.3 configured on router C from router A.

On router B I configured a static route for 3.3.3.3 that has 172.16.55.51 as its next-hop. I can ping from B to C just fine.

I then advertised the interface of router B that connects to C in OSPF, both router P and A learn of the addresses and from both routers I can successfully ping to 172.16.55.50 and 51.

I then tried to configure on A. a static route to 3.3.3.3 with 172.16.55.51 as next-hop, the ping fails. Same happens if I try to configure 172.16.55.50 as next-hop.

However, utilizing router B ospf-router id (10.0.0.2) as next-hop results in a successful ping from A to C.

I am not sure why it's working like this but not when I utilise 172.16.55.51 as next hop.

-Router A and P can reach both next-hops (172.16.55.51 and 10.0.0.2) so does router P.

-Router B and P can reach the source interface utilized by router A for the ping.

So it seems to me that utilizing 172.16.55.51 as next-hop should be every bit as valid as utilizing 10.0.0.2 however, only the latter works. Can anyone help figure out why?

Thanks

EDIT:

Adding the routing tables for routers A, B and P. Note that A B and P are also running MPLS between them with P acting purely as a transport router.

I omitted the non relevant routes from the tables for brevity, Loopback0 is being used as ping source:

Router A
S 3.3.3.3/32 [1/0] via 172.16.55.51, 00:00:07
O IA 10.0.0.2/32 [110/21] via 172.16.6.24, 1d09h, HundredGigE0/0/0/1
O IA 172.16.55.50/31 [110/30] via 172.16.6.24, 1d09h, HundredGigE0/0/0/1
Router B
O 10.0.0.1/32 [110/11] via 172.16.6.25, 12w0d, HundredGigE0/0/0/8
O 10.0.0.2/32 [110/11] via 172.16.55.49, 1d09h, Bundle-Ether1
O 172.16.55.50/31 [110/20] via 172.16.55.49, 1d09h, Bundle-Ether1
Router C
S 3.3.3.3/32 [1/0] via 172.16.55.50, 07:51:48
C 172.16.55.50/31 is directly connected, 2w3d, HundredGigE0/0/0/37
L 172.16.55.51/32 is directly connected, 2w3d, HundredGigE0/0/0/37
O IA 10.0.0.1/32 [110/21] via 172.16.55.48, 1d09h, Bundle-Ether1
3
  • 1
    Please edit your post to include the routing tables from each router. I suspect that when you do, the answer will become apparent.
    – Ron Trunk
    Commented Feb 16 at 18:40
  • Thanks for pointing this out Ron, I've added the routing tables
    – V Dp
    Commented Feb 16 at 19:48
  • On which router do you have L 3.3.3.3/32 route? Router C has a route to reach 3.3.3.3 via router B interface address, while router A tries to find 3.3.3.3 via router C interface address. But nobody actually has it...
    – manish ma
    Commented Feb 17 at 5:44

1 Answer 1

0

3.3.3.3/32 IS an interface on Router C, so no static route is necessary. The route needs to be on Router B, and it needs to announce the route into OSPF. Furthermore, "interfaces" under "router ospf" only indicates where OSPF should run, not what to announce. "redistribute static" would be the method to get Router B's static route shared via OSPF.

1
  • Thanks for catching that, I made a typo. For the OSPF, on XR router to get the network of an interface advertised, we usually add the interface under the OSPF process in passive. The red static works but I was wondering why if I configure a static on router A it will work when using Router B router id loopback interface but not when using the network between B and C as both are known to A in OSPF and reachable
    – V Dp
    Commented Feb 16 at 20:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.