08-04-2005 07:32 AM
Hi all,
I need help. For days, I spent my time to check my config many times. So far, I still did not get the clue what happen to my config. I am setting up Inter-AS MPLS VPN, with VPNv4 eBGP session between ASBR. Both CE have their complete route between each other. Both PE also has complete routing table on its VRF. But, I still can not ping end-to-end. What's the problem. FYI, I am running MPLS on 2801 box with IOS c2801-adventerprisek9-mz.123-14.YT.bin.
my topology is simple, as follow:
CE1--PE1--ASPE1--ASPE2--PE2--CE2
Can you give comment on my config?
Thank for your help,
Rivalino
08-04-2005 08:37 AM
Your configs look good. Could you please post the output of the following commands:
from PE1:
sh ip cef vrf TES 6.6.6.6
sh ip cef vrf TES 56.56.56.0
show mpls forw det
from PE2:
sh ip cef vrf TES 1.1.1.1
sh ip cef vrf TES 12.12.12.0
show mpls forw det
From ASPE1 and ASPE2:
sh mpls for detail
Thanks,
08-08-2005 05:11 AM
you are not advertising the dmz link in your IGP. Add network 34.34.34.0 0.0.0.255 under your ospf statements on each asbr pe and you will then be fine. Oh and you will need to passive interface that baby as well or else you shall have an ospf neighbour. OR just do a redist connected.
08-08-2005 05:36 AM
It has to be something else since next-hop-self is configured on both ASBRs and looking at the routes on the PEs we can see the next hop is the ASBR loopback address.
Regards,
08-08-2005 06:21 AM
08-08-2005 07:32 AM
Looking at the output you just provided, it looks like there is something wrong with LDP session between PE1 and AS-PE1 and between PE2 and AS-PE2. The reason I'm saying that is on PE1 the LFIB entry for 3.3.3.3 shows as untagged when it should be "Pop tag"
Can you join a "sh mpls ldp nei" and "sh mpls ldp dis" on PE1, PE2, AS-PE1 and AS-PE2.
Also "show mpls ldp bind 3.3.3.3 32" from PE1 and "show mpls ldp bind 4.4.4.4 32" from PE2.
Thanks,
08-09-2005 06:04 AM
Hi,
I surprise to know that ldp neighbor is not up. You can see at attachment. What you said, that LFIB entry for 3.3.3.3 on PE1 should be "Pop tag", I think it should be untagged ,since from PE1 to reach 3.3.3.3 is only one hop away. And PE1 is th first LSR in this case, it receive untagged packet from CE1, then he should not do 'pop tagging' since it receive untagged label. Am I right?
So, what about my config? It seems ios problem..!?
Thanks,
Rivalino
08-09-2005 07:24 AM
The issue is really with the LDP session not being setup properly. The "no route" on both PE1 and PE2 is what is causing the issue.
PE1#sh mpls ldp dis
Local LDP Identifier:
23.23.23.2:0
Discovery Sources:
Interfaces:
FastEthernet0/1 (ldp): xmit/recv
LDP Id: 34.34.34.3:0; no route
PE2#sh mpl ld dis
Local LDP Identifier:
5.5.5.5:0
Discovery Sources:
Interfaces:
FastEthernet0/1 (ldp): xmit/recv
LDP Id: 34.34.34.4:0; no route
This is because AS-PE1 and AS-PE2 use the serial interface address as ther LDP router-id instead of the loopback interface IP address.
Make sure that all routers use the loopback interface IP address instead by configuring the following command:
mpls ldp router-id loopback 0 force
Hope this helps,
08-10-2005 05:07 AM
Hi,
It solves my problem. Now CE1 can reach CE2.
Thanks for your help :))))))
Rivalino
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide