Unable to ping each other over MPLS VPN network

[ltgcisco]

Hi,

I am facing a strange problem, as shown below, CPE-HQ WAN IP cannot ping CPE-Branch's WAN IP,

From 25.25.1.2 cannot ping 25.25.2.2, the routes (25.25.1.0 255.255.255.252 and 25.25.2.0 255.255.255.252) present in routing table in respective VRF; Both CPE are configured with default route.

However,

ping vrf hq 25.25.2.2 source 25.25.1.1, able to reach.

ping vrf branch 25.25.1.2 source 25.25.2.1, able to reach.

from CPE WAN to CPE WAN, unreachable.

I appreciate it if anyone can help.

Best Regards,

TG

[balaji.bandi]

how is your routing between PE to CE , static route ? or IGP ?

below view help you :

https://wannabelab.blogspot.com/2019/03/configuring-mpls-vpn-between-pe-routers.html?m=1

https://www.youtube.com/watch?v=Mb44vevUu4M

still issue post the config of the PE and CE to verify along with show route outputs.

[ltgcisco]

Hi Balaji,

Thanks for helping me!

Exactly I am facing this issue in production environment, I simulated in simulator and I get no issue. Unfortunately, I can't share the configuration here as all the IP addresses are public IP which are confidential.

the routing-table in each VRF is perfect, and the next hop are correct.

I'd like to know if anyone experience such weird issue.

each CPE has one static default route point to PE, and static and connected routes are redistributed to BGP

thanks

[MHM Cisco World]

in PE can you see the WAN subnet for both HQ and Branch?

[ltgcisco]

Hi

Each VRF can see the wan subnet from the other one.

[MHM Cisco World]

..

[MHM Cisco World]

do traceroute see where the packet is drop.

[ltgcisco]

as shown above, it stops at 25.25.2.1 for both directions

[MHM Cisco World]

Both directions same router begin,

What I mean 
the traceroute 
1 <-PE local 
2 <- P depend on TTL config 
3 <- PE remote 
4 <- CE remote 

here both trace route use same PE local how this happened?

some step to troubleshooting 

R#sh mpls interface

check that all interface is run MPLS 

R#sh mpls ldp disco
check that this PE is send/receive LDP label from neighbor "please notice if any line have no route then there is problem in IGP connect two router"

[Harold Ritter]

Hi @ltgcisco ,

Who owns the MPLS core? You or the service provider? If you own the core, did you configure LDP between PE1 and PE2?

Regards,

[ltgcisco]

yes, configured, route import and export working as normal, can see mpls interface: Te0/0/20 Yes (ldp)

[Georg Pauwen]

Hello,

post the full running configurations of all four routers...

[emre]

merhaba

There seems to be a problem in your mpls config, there is no more than local ip !

You need to be able to ping the mpls ends ip !

if you can't ping there is a problem with QinQ or vlan .

If you post the PE and CE end configs, I'll check.
you can delete your ip addresses

[ltgcisco]

Hi mere,

ping 25.25.2.1 source 25.25.2.2 has no issue, but traceroute 25.25.1.2 source 25.25.2.2, it stop at 25.25.2.1

How could it be VLAN or QnQ issue?

Thanks