Cisco Support Community
Showing results for 
Search instead for 
Did you mean: 

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

OSPF default route originator problem


I have two edge routers and two core routers:

+--------+ +--------+

| edge A | | edge B |

+--------+ +--------+

| |

+--------+ +--------+

| core A |-----------------| core B |

+--------+ +--------+

Core A has a static default route to edge A and core B has a static default route to edge B. The two core routers exchange routing information via OSPF.

I want core A to change its default route to core B in the case edge A become unreachable, so I added this to my OSPF configuration:

default-information originate metric 100

It works very well, but I want core B to change its default route too if edge B becomes down.

Unfortunaly, a default route originator doesn't accept default route announcements.

How can I force the core routers to accept them ?



Re: OSPF default route originator problem

Both routers can accept both default routes and stores into OSPF database("sh ip os da"). If the existing static default route is not there, then the learned OSPF default route comes into route table. I can recommend other configuration use "default-information originate route-map test". I am giving one example from Core A aspect....

route-map test permit 10

match ip address 1

match ip next-hop 2

access-list 1 permit

access-list 2 permit

So that, Core A advertises default route to Core B, only if it is having connectivity to edge A.

Or in other way, assuming that you are advertising core-B and edge-B connected network to Core-A. Now, if core-B and edge-B network fails, then this network dropped from Core-B to Core-A. The following configuration at Core-A checks for that one and Core-A advertises the default route if Core-B and edge-B network is down.

route-map test1 permit 10

match ip address

match ip next-hop

I strongly believe that the above configuration might work. Because when Core-B withdraws the route, then Core-A matches that network with it's existing static default route pointed to edge-A.

CreatePlease login to create content