I want to apply Stub in one of my sites
My Design is Hub and Spokes
Each spoke in site have many routers behind it, Each Site has internet from it’s Site itself
Internet—Area 0–Area 1–Internet
If I applied Stub, My sites will not reach O E1, O E2 from Other sites because no default route my Spoke will get as long as this spoke already have default route toward ISP to reach internet
What is the best solution here?
Hello Ali
Typically, at each spoke, the default route should lead you out to the Internet. That means that you need to have something more specific than a default route that routes your traffic from one spoke to another. In the lesson, the NSSA uses a default route. That’s the beauty of the NSSA…
However, you can’t have that in your topology, because your default route leads out to the Internet. In your specific scenario, each spoke router must know the specific subnets behind all the other spoke routers in order to route traffic correctly. Stubs will not allow you to do that. I suggest that you do not configure any kind of stub network, but perform route summarization in order to reduce the size of the routing table at each spoke. It seems that each site is in a different area, so summarization between areas is possible. Take a look at the following lesson for more info…
I hope this has been helpful!
Laz
Which is best ospf network type for my design?
All connected through isp with same vlan.
@lagapides
Hello Ali
Take a look at this response here:
I hope this has been helpful!
Laz
Hi
Why I have in a routing table a summary route without configure"area X nssa default-information-originate" ?
After configuring the command on ABR the output didnt change.
I’ve built a symple lab topology with 3 areas…A1(stub)<—(backbone)–>A2(nssa)
Hello Giovanni
The area X nssa default-information-originate
command does not introduce a summary route into the routing table. It should introduce a default route. Now having said that, there are various commands that may have caused a summary route to appear in your routing table. This includes commands such as area range
or summary-address
to name a couple.
The area X nssa default-information-originate
command when applied to an ABR should create a default route that should show up as an NSSA External Type 2 (O*N2) in the routing table of the neighbor that exists within the NSSA (as seen in the lesson). If that hasn’t happened, you should examine the rest of your configuration.
Let us know a little more about your topology and your configuration so that we can help you further.
I hope this has been helpful!
Laz
Hi ,
I will ask about O*IA notation.
Does that mean → This route is the candidate default route which comes from A different OSPF Area?
Question 2 → If I see a route entry in routing table with O*IA Which scenario is valid ?
Scenario 1 → There is only one candidate default route for this specific destination so there is one candidate for that destination therefore Router selects this route into RIB as default route
Scenario 2-> There are more than one candidate default route to the destination bur for some reasons (what is these reasons also a question) this specific candidate default route get into RIB.
Question 3 → for example let’s say there are 2 routes OIA 0.0.0.0/0 [110/2] via 192.168.23.2, 00:02:27, FastEthernet0/0 and OIA 0.0.0.0/0 [110/2] via 192.168.13.2, 00:02:27, FastEthernet0/1 as candidate default routes
which one will be selected into RIB and according to what? Or it will be equal cost load balancing?
Hello Görgen
Yes, the O IA
notation means “Inter-Area.” This means that this is a route learned from another OSPF area. The “*” indicates a candidate default route. That means that somewhere, OSPF was configured with the default-information originate
command which injects a default route into OSPF and is advertised.
It depends on how many candidate default routes exist. If there is only one, you will see that route chosen. If there are multiple candidate routes, then the administrative distance (AD) of each route will be used to determine which one will be the active default route.
So the default route that is chosen and actually installed is the one with the lowest AD.
Specifically, you will see Gateway of last resort is...
and the next hop IP of that default route.
The same rules apply here as to all routes. If the metric is the same in your example, then both will be installed with load balancing taking place.
For more information, take a look at these NetworkLessons notes:
- Candidate default route
- How the routing table is populated
- Process by which entries are matched on the routing table
I hope this has been helpful!
Laz
Hello,
Please, would you explain one case where generating the default route in the ASBR will be useful?
I have thought about it, but is a little contradictory, because if we generate the Default Route on the ASBR, then is redundant, as Type 7- External LSA are allowed on NSSA areas.
Thanks,
Leo Grazzi
Hello Leonardo
If you notice in the configuration of the NSSA, the default-information-originate
command is issued on the ABR R2. This causes R2 to advertise itself as the default route for the NSSA area which is Area 1. So in the scenario in the lesson, R2 (the ABR) is acting as the default route for all of the stub Area 1, and thus for our ASBR R3. But using the NSSA, R3 is able to advertise it’s route to 3.3.3.3 which is outside of the OSPF domain (thus the ASBR designation) to the rest of the OSFP domain via the ABR R2, because of the NSSA configuration.
So your specific question is somewhat different from the scenario of the lesson. You’re asking about use cases to cause the ASBR to be a default route to the whole OSPF domain, correct?
In the lesson, the ASBR is advertising the 3.3.3.0/24 network. This could easily be changed to the advertising of a default route. R2 (the ABR) could propagate that route into the backbone area, and thus provide a default route for the whole OSPF domain. This is somewhat unusual, but it is not unheard of. By keeping the stub area an NSSA, you can achieve this. Does that make sense?
I hope this has been helpful!
Laz
Hello Lazarus,
Thanks for the response.
Yes, I don’t understand why the default route is not originated by default on the ABR as in a stubby area, because I cannot think of a case when the ASBR is the one generating the default route, because it would be redundant to the Type 7 LSAs generated by the ASBR, is that correct?
However, I am assuming there must be a reason why Cisco doesn’t automatically inject the default route on NSSA areas on the ABR. I believe is because, since ASBRs are allowed on NSSA, then on an NSSA area initially both the ABR and the ASBR could receive Type 5 LSA (in a scenario where different external routes are being advertised to the ABR and the ASBR (by redistribution to the ASBR) ) so then OSPF “waits” for us to decide which router would be the one generating the default route. Is this correct?
Thanks,
Leo
Hello Leonardo
The logic of a stub area is that you have only one possible route out of the area. Therefore, why not replace all routing information with a single default route to the ABR? That gives us more efficient routing tables within the stub. If you have an ASBR however, there are now two possible exit points from the area, via the ABR to the rest of the OSPF domain, or out of the ASBR to a routing domain external to OSPF. Regardless of the LSA types allowed or not, the automatic default route generation is left up to the network engineer.
Yes, your understanding makes sense. Although not explicitly stated in the RFC 3101 which describes NSSAs, the RFC does indicate that ABRs do not automatically generate default routes (Type-3 LSAs for 0.0.0.0/0) into NSSA areas unless explicitly configured to do so. This design choice ensures that NSSA areas maintain control over routing information and avoid unintended routing behavior.
I hope this has been helpful!
Laz