cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
753
Views
0
Helpful
3
Replies

nssa external routes

bruno.hodge
Level 1
Level 1

Hey everybody!

I have a question that pertains OSPF NSSA external routes.

The topology goes as follows:

R1 area 0 <--------->area0 R2 area29 <-------->area29 R3

Area 29 is configured as nssa.

In R3 i have static routes that are being redistributed into ospf and are being seen as external routes in R1. I happen to configure a new interface in R3, mistakinly, in area 28 (suposed to be 29). The moment i did that, the external routes disapeared from R1.

why is that??

thanks

Bruno

3 Replies 3

Giuseppe Larosa
Hall of Fame
Hall of Fame

Hello Bruno,

was the IP subnet of the new interface related to the IP addresses of next-hops of static routes defined on R3?

this can have caused issues with the forwarding address field in the external LSA data structures

Also to be noted that R1 can see these external routes because R2 the ABR between area 0 and area 29 was translating the O N1 route (LSA type 7) into LSA type 5.

So the same doubts apply to R2 first.

Hope to help

Giuseppe

Jon Marshall
Hall of Fame
Hall of Fame

Bruno

This is all to do with whether or not the P-Bit is set which tells the NSSA ABR whether to translate the type 7 LSA within the NSSA area into a type 5 LSA to advertise beyond the NSSA area.

From Cisco doc on OSPF NSSA -

If bit P = 0, then the NSSA ABR must not translate this LSA into Type 5. This happens when NSSA ASBR is also an NSSA ABR.
If bit P = 1, then the NSSA ABR must translate this type 7 LSA into a type 5 LSA. If there are multiple NSSA ABRs, the one with highest router ID.

full link - http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a0080094a88.shtml

Note P = 1 means P-Bit is set.

So in your original config

R1 area 0 <--------->area0 R2 area29 <-------->area29 R3

R3 was an NSSA ASBR because it was redistributing the statics. R2 was the NSSA ABR.

But as soon as you accidentally configured another area on R3 then R3 became both an NSSA ASBR and an NSSA ABR because it was a border router between area 28 and area 29. If you look at the definitions above R3 must now set the P-Bit to 0 because it is both ASBR and ABR. R2 now receives the LSAs with the P-Bit set to 0 which means it cannot now advertise these as external type 5 LSAs to R1.

If you look on R2 using the "sh ip ospf database" command you can see this for yourself. Before you add the area 28 interface then R2 will show the static routes as both Type 7 LSAs and the same routes as Type 5 LSAs. Because the P-Bit was set to 1 R2 actually translates the type 7 LSAs to type 5 LSAs.

Once you add the area command the P-Bit is set to 0. Now on R2 you can only see the type 7 LSAs and type 7s cannot be advertised beyond the area.

Jon

Jon,

perfect! Your description is exactly the behavior i observed and i did notice the P bit in R2, but i still have a couple of questions:

1 - The Cisco definition of ABR is that of a router connected to 2 or more areas, being  area 0 forcefully one of them. When i configurated mistakinly the area 28 in R3, the output for "sh ip ospf" acused still as an ASBR only. And R2, in "sh ip ospf", is ABR and ASBR at all times.

2- Why this behavior? OSPF loop protection? And also, the situation in R3 (2 areas configured and none of them area 0), what is the OSPF behavior for this configuration that goes against Cisco OSPF definitions?


Thank you both for the answers!

Bruno

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Innovations in Cisco Full Stack Observability - A new webinar from Cisco