MPLS VPN - Problem with communication between CustA-CustA1 and PE1-CA1

[TommyKay]

Hello everyone,

I hope you are doing well. I am having an issue with communication between Customer A router and CustomerA1 router. Additionally, what is strange is that when I try to ping from PEdge2 router to CustomerA router's loopback0, the ping is successful

(ping vrf 101:CustA 100.100.100.100)

However, when I try to ping from PEdge1 router to CustomerA1 router's loopback0, the ping fails

(ping vrf 101:CustA 111.111.111.111)

The topology and configuration have been created with the implementation of 6VPE (once I can resolve the communication issues) and tunneling in the lower part of the topology. Please find the attached images of the topology, interfaces, and their addresses and configurations of the routers(zip file) and the project files(zip and link to Google drive).

I have tried various commands to test the network, but apparently, I am missing something that is causing this problem. The network is set up in GNS3, and the routers are from the 7200 series

(c7200-adventerprisek9-mz.124-24.T5.image)

I would greatly appreciate any guidance that could help me solve this communication problem.

Best regards!

Link to project: https://drive.google.com/drive/folders/1NQ5XiPm_Ic-JKMFjXVsRmPIFr2klTRcV?usp=sharing

[David Ruess]

Hello,

At first glance it looks like you dont have mutual

redistribution

2 things assuming L3 MPLS VPN:

On PE2 and PE1 you need to

redistribute

the OSPF process into BGP vrf and vice versa. You can tell you have

redistributed 

into the BGP VPNv4 process when you do a

show ip bgp vpnv4 all

and see the customer routes in that table. 

address-family ipv4 vrf 101:CustA

redistribution here -> redistribute ospf 10 vrf CustA internal external
neighbor 11.0.0.1 remote-as 65532
neighbor 11.0.0.1 activate
neighbor 11.0.0.1 send-community
no synchronization
network 11.0.0.0
exit-address-family

2. Your OSPF process on PE1 (hadn't checked PE2 yet but assume its the same) is not in the VRF. Do you have an OSPF neighbor ship between your PE and CE devices?

router ospf 1
log-adjacency-changes

**no vrf or network statements. If you are using OSPF for the MPLS you need another OSPF process in the VRF like below

router ospf 10 vrf 101:CustA

network 11.0.0.2 0.0.0.0 area <#>

redistribute bgp 65535

-David

[Harold Ritter]

Hi @David Ruess ,

router ospf 1

is the core routing protocol. BGP is used as the PE-CE protocol. Therefore there is no need to

 redistribute

anything here.

> **no vrf or network statements. If you are using OSPF for the MPLS you need another OSPF process in the VRF like below

The network statements are not required under

"router ospf 1", as the " ip ospf 1 area 0"

command is used on all appropriate interfaces.

Regards,

[David Ruess]

Harold,

Good catch. After I saw your answer I meant to go back and check. I did skim over the configurations and jumped to a couple of conclusions. The config did seem a bit odd and I did not catch the BGP CE/PE peering, but I should have slowed down. Thank you for the correction as I always aim to increase my knowledge and troubleshooting.

-David

[MHM Cisco World]

sorry I can not open zip file can share you config as text ?

Thanks

[TommyKay]

Hi @MHM Cisco World , of course. These routers are essential so I didn't include P3/P4 and the bottom side of the topology.

[MHM Cisco World]

Your config have following misconfig

1- you must use

redistrubte connect

under pe bgp address vrf that connect to ce

2- you must never use

next-hop-self

under bgp address vpnv4 

That it' 

Do above 

Clear ip bgp xxx

And check again ping from CE to CE

[TommyKay]

I did change the settings and use 

clear ip bgp 

on both PE routers. Below I present the changes in both PE routers. Additionally I attach commands showing

CustA bgp neighbour 10.0.0.2(PEdge1 g1/0 int address)

routes.

Ping from CustA to CustA1 still doesn't work. Ping using

VRF 101:CustA from PEdge1 to loopback0 CustomerA1

still doesn't work and ping from PEdge2 to CustA does work, as at the beginning.

[MHM Cisco World]

sh bgp vpnv4 uni all summ

sh bgp vpnv4 uni all

Share this after change 

Note:- change the lo you use in P1 and P2' make it 

101.101.101.101

102.102.102.102

The

ospf lo/32

with mpls have issue to build label.

Thanks 

[TommyKay]

Thank you @MHM Cisco World  for trying to help me with this one :).

I did change the l0 on P1 and P2. I did also set the mask to /32, but you wrote that lo with /32 mask have problems building labels in mpls, so I wasn't sure whether it'll be correct.

I did issue the commands. Below are the results.

[MHM Cisco World]

Wait let me check something here 

Thanks 

[MHM Cisco World · Accepted solution]

PEdge2

The link between

PEdge2

and CE change it subnet from

10.x.x.x tp 20.x.x.x

Do

clear ip bgp 

in all router and try again.

Thanks 

[TommyKay]

@MHM Cisco World You mean from

11.x.x.x to f.e. 22.x.x.x 

? In the bottom part of the topology, the link between

PEdge3

and CustomerB has network

20.0.0.0 /8

so we have to allocate a different subnet. Can I allocate subnet

 22.x.x.x /8

to the link between

PEdge2 - CustomerA1

?

[MHM Cisco World]

Sorry if you use

20.x.x.x

in other link then use different like using between

PEdge2 - CustomerA1 subnet 25.0.0.0

Thnaks 

[TommyKay]

I did change the subnet to

25.0.0.0 /8 between PEdge2 and CustomerA1

and on every router

(CustA, CustA1, PEdge 1/2, Provider 1/2) cleared bgp using clear ip bgp 

Pings from

CustA to CustA1

and conversely still doesn't work. Ping from

PEdge2 to CustA

works as it did and from

 PEdg1 to CustA1

still doesn't work.

I include outputs from the two commands:

sh bgp vpnv4 uni all summ

sh bgp vpnv4 uni all

issued on PEdge1 and PEdge2. 

I'm open to speaking with you or anybody who wants to help me to solve this issue on MS Teams, Zoom, or other platforms you choose.