MPLS Layer 3 VPN PE-CE EIGRP

Hello everyone,

I joined this site awhile ago and wanted to say hello. I noticed that the MPLS Layer 3 VPN PE-CE EIGRP lesson is using an OSPF configuration. I think this is a mistake, and should be EIGRP?

Thanks,

Richard Aaron

Hello Richard

The lesson describes a situation where a customer wants to run EIGRP on their internal networks, and thus also between their CE routers and the ISP’s PE routers. In the EIGRP between PE and CE routers section of the lesson, you can see this configuration using EIGRP.

In this MPLS topology, as in most MPLS implementations, OSPF is used as the IGP within the infrastructure of the MPLS network. Most MPLS networks are deployed using the MPLS/OSPF/BGP combination of technologies, even though MPLS/EIGRP/BGP is also possible.

The main reason OSPF is used is because it is a link state routing protocol. This is especially useful when using what is known as MPLS traffic engineering (MPLS-TE). You can find out more about MPLS-TE at the following link, although it is outside the scope of any current Cisco CCNP or CCIE Enterprise certifications.

IS-IS can also be used as the IGP within an MPLS backbone as it too is a link state routing protocol.

I hope this has been helpful!

Laz

Hi guys.
One think it’s driving me crazy.
For what reason there’s an EIGRP AS number 5 on the CE3 router?
In the configuration I haven’t found anything.

I guess it was just for illustrating.
In the scenario 2, using different AS (AS 2), I understood the concept about it.

Hello Tiago

In this specific lesson, it seems that there is no CE3 router. Assuming that was a typo, I was unable to find a configured EIGRP AS of 5. Can you clarify where you have seen this so we can answer your question?

Thanks!

Laz

Hi Laz

Over here, highlighted on yellow

It wasn’t in the configuration , just on the picture instead.

Hello Tiago

The AS numbers shown in this diagram are those used by BGP and not EIGRP. Because this series of lessons on MPLS VPN uses various implementations, one of those is to use eBGP between CE1 and PE1 and between CE2 and PE2, which is what has been done in the MPLS L3 VPN Configuration lesson as well as many subsequent lessons. The same diagram is being used.

However, in this case, we are not configuring any BGP parameters in CE1 and CE5 so you are right, these AS numbers should not appear in the diagram.

I will let Rene take a look and make any necessary changes.

Thanks for pointing that out!

I hope this has been helpful!

Laz

Hello, everyone!

So if I understand this right, although the BGP VPNV4 routes are being redistributed into EIGRP
obrázok

These routes will appear as internal routes?

So is it safe to assume that the MPLS network is technically transparent to the two customers? It’s as if they were directly connected and just exchanged this prefix.

However, how is it possible that despite being redistributed, the routes are still internal? Is this an MPLS-related behaviour? And why exactly is it like this?

Thank you.

Hello David

Yes, you’re correct in your understanding that the MPLS network is technically transparent to the two customers. It does indeed appear as if they were directly connected and just exchanged this prefix. This is the “magic” of MPLS.

As for your question about the routes appearing as internal despite being redistributed, this is because of the nature of EIGRP and how it classifies routes. You are correct that when a route is redistributed into EIGRP, it’s classified as an external route. However, in this case, the BGP VPNv4 routes are not being redistributed directly into EIGRP. Instead, they’re being imported into the VRF and then redistributed into EIGRP. Notice the redistribute command is being applied under the IPv4 address family VRF Customer configuration mode and not directly under the router eigrp 1 configuration mode. This causes the routes to appear as internal because, as far as EIGRP is concerned, they’re originating from within the VRF.

This behavior is not specific to MPLS but it is specific to VRFs. It has to do with how EIGRP handles redistributed routes from VRFs. The key factor here is that the routes are being imported into the VRF before they’re redistributed into EIGRP, which is why they’re appearing as internal routes.

A similar issue appears with OSPF. When redistribution is applied, you would initially expect O E1 or O E2 routers (external), but you don’t, and this is due to the same reason that I explained above for EIGRP. However, what is “strange” is that you actually get an O AI route which is an inter-area route, even though both CE1 and CE2 are in the same OSPF area. The reasoning behind this, which is actually quite interesting, is explained in the Verification section of this lesson:

I hope this has been helpful!

Laz

Hi all
Cannot understand the scope of using another routing protocol between CE and PE instead of using EBGP.
Maintain an EBGP connection between them Is in my opinion straight easier than redistributing something in IBGP and viceversa, isn’t It?
Anyway thanks for the great lesson, it’s explained well
Aronne

Hello Aronne

You’re correct that maintaining an eBGP connection between CE and PE routers is typically simpler, and it’s indeed a common practice. However, we must remember that the CE and PE devices typically belong to two different entities: the customer and the ISP. The requirements of each are not always driven by technological factors, but also by policies and rules that keep conformity throughout each entity’s network.

For example, it may be the policy of the enterprise (customer) to use OSPF in all of their networks. The ISP must be flexible enough to accommodate such requirements, and this is why EIGRP, OSPF, and even RIP (!) can be used as the edge routing protocol.

I hope this has been helpful!

Laz

Hi Laz,
Understood. Thanks for your explanation.
Always on the top !
Have a nice day.
Aronne

1 Like