IOSXR 6.1.3 InterAS option-C

[aghufran]

Hello,

I am testing InterAS option-C setup in eveNG lab using Cisco IOSXRv running code 6.1.3.

As part of the required configuration I am placing static routes on both the IOSXRv devices for each other, pointing out the connected interface. On one of the devices, if I use a next-hop IP address in the static route, I get bgp next-hop as inaccessible in the bgp ipv4 labeled-unicast table. 

So if place something like this:

router static
address-family ipv4 unicast
1.2.3.4/32 GigabitEthernet0/0/0/2.998 1.2.3.4

I get something like this:

RP/0/0/CPU0:lab-asbr-1#sh bgp ipv4 labeled-unicast a.b.c.d/32

...

BGP routing table entry for a.b.c.d/32

...

1.2.3.4 (inaccessible) from 1.2.3.4 (RTR-ID)
....

If, however I change the static route as follows:

router static
address-family ipv4 unicast
1.2.3.4/32 GigabitEthernet0/0/0/2.998

i.e. removing the next-hop IP address of 1.2.3.4, it does not show the next-hop as inaccessible in the bgp ipv4 labeled-unicast table.

Interesting, I am not getting this sort of behaviour on the other remote asbr device 1.2.3.4. 

Please advise.

Thanks,

Abid Ghufran

[Harold Ritter]

Hi @aghufran ,

The proper way to configure the static route is the second you mentioned.

router static
address-family ipv4 unicast
1.2.3.4/32 GigabitEthernet0/0/0/2.998

Note that this route is a static route and doesn't need to be in BGP.

Please refer to the following document:

https://www.cisco.com/c/en/us/support/docs/multiprotocol-label-switching-mpls/mpls/200523-Configuration-and-Verification-of-Layer.html

Regards,

[aghufran]

Hi Harold,

As you suggested it works fine if I do not specify the next-hop IP address. However, what I find confusing is that there is another IOS-XR running the same code and there the format of static route, with a next-hop IP address works fine. So not sure why is there this difference in behaviour. But I guess it would require more of your time to delve into it and as its working, I will leave it for now. Plus this is a lab environment so will investigate it further later.

Thanks and regards,

Abid Ghufran