This topic is to discuss the following lesson::
Thanks for great lesson.
I’m quite new to Vpn l3. Could you explain about label stack of vpn packet when it travels from CE1 to PE1, then through isp , then out on PE2, then to CE2. Tks !!
For sure, it’s explained here:
Thank you for the this great site.
Can we redistribute into OSPF instead of redistributing into BGP.
Redistributing RIP into OSPF is no problem but you can’t use it for MPLS VPN. You need BGP since it has address families for VPN routes.
Very nice explanation. I have a general question: We are using IGP between PE-CE routers, Aren’t we only allowed to use e-BGP between different AS routers?
In an MPLS PE-CE scenario, you can use IGPs or BGP. Even static routes are possible.
I have the lab for MPLS L3 VPN and RIP is my IGP.
I could see all the routes but not able to R5 loopback 0 from R1 with source address of loopback 0 and vice versa.
I could see the 18.104.22.168 routes on R1 and 22.214.171.124.0 routes on R5.
Please help me if I am missing something.
In order to be able to help you we’d need a little more information. Have you attempted connectivity between the Fa0/0 interfaces on CE1 and CE2? (I assume when you say R1 and R5 you mean CE1 and CE2). If that is not successful, you should check out the configurations on all your routers. If it is successful, check to make sure the Lo interfaces are participating in BGP.
I hope this has been helpful!
Will you please advise me how do I advertise loopback I/F of PE1 to CE1?
I have added
PE1(config)#router rip PE1(config-router)#address-family ipv4 vrf CUSTOMER PE1(config-router-af)#version 2 PE1(config-router-af)#no auto-summary PE1(config-router-af)#network 192.168.12.0 **PE1(config-router-af)#network 126.96.36.199** and even **PE1(config-router-af)#redistribute connnected**
It seems the PE1 are not advertise its loopback to CE1.
This behaviour is normal. Not only that, it is actually desired. The whole purpose of the MPLS Layer 3 VPN is to be able to create a transparent MPLS VPN between the two CE routers as if they are directly connected to each other. You’ve configured the Loopback to participate in the RIP routing, but you’ve also configured the
ip vrf forwarding CUSTOMER VRF on the customer facing Fa0/0 interface and the
mpls ldp autoconfig on the network facing interface. This will cause all additional network interfaces to be “inside the VPN” and not visible to the RIP process.
I hope this has been helpful!
Thank you for your kindly explaination.
I also find some hint in the next few lessons.
I understand that bet PE and CE we can use RIP/BGP/OSPF as well.
But can we use BGP in the Core Service provider network (bet PE1, P and PE2 routers), or we only have to use OSPF as our IGP in the core MPLS network?
Can any other protocol (except OSPF as shown here) can be used as IGP in MPLS cloud network?
You’ll need an IGP so that your IBGP routers know how to reach each other. Here is an example:
Another IGP is no problem, you could use EIGRP or IS-IS.
So we cant use eBGP bet PE1,PE and PE2 routers right? I tried using eBGP, tht did not worked.
It worked fine with OSPF.
Also, 1 more thing, can we use eBGP between PE 1 and PE2 routers (edge routers), or only iBGP works?
I tried with eBGP between PE1 and PE2, it did not work. It worked fine with iBGP.
Technically, you should be able to make it work. In a scenario like this, the PE and P routers belong to a single SP so it’s more likely that they use a single AS internally.