01-28-2025 05:53 AM
I currently have a routing issue, that I'm not sure if it is a bug or configuration issue, but here IT goes.
I have service provider network that have two ebgp peers to two service providers both send full routes plus default route 0.0.0.0/0
With a route policy I only tag default route with an ext community xxxx:xxxx export it and import it to all access nodes. In this case the access nodes are NCS540.
The setup and configuration mentioned above works great no issue here. I can access everything fine.
Here is my problem
I have a third NCS55A2 which connects to a content provider which has several Ebgp peers using a bvi bridge domain. With a route policy, I tag these routes coming from these ebgp peers with ext community 5555:5555 and import this same ext community in all access nodes.
Once I do this I can see all the routes from the ebgp content provider because obviously it is a shorter path, but can no longer reach them. When I do traceroutes it clearly trying to go thru the ebgp peer with direct connections, but it dies at the NCS55a2 box. If I remove the imported routes with ext yyyy:yyyy it can reach those routes because now it's using the default route from the other two service providers. See picture attached.
I after going some extensive research I concluded that there might be a limitation on the NCS55A2 using bridge domains using many ebgp peers per this documentation when using the label allocation mode per vf per ce. See below documentation from cisco
Do you guys give it the same conclusion or am I interpreting this wrong. I can't find anything wrong with the configuration. is there a work around this??
######################################################################################
Restrictions for MPLS VPN per CE Label
• Enabling the MPLS VPN per CE Label feature causes Border Gateway Protocol (BGP) reconvergence,
which can result in data loss for traffic coming from the Multiprotocol Label Switching (MPLS) VPN
core.
You can minimize network disruption by enabling this feature during a scheduled MPLS
maintenance window. Also, if possible, avoid enabling this feature on a live device.
Note
• IPv6 Provider Edge devices (6PE) are not supported.
• Prefix-Independent Convergence (PIC) is not supported.Per CE Label with only multipath is supported.
You cannot use this feature with:
• Internal Border Gateway Protocol (IBGP) multipath feature
• Carrier Supporting Carrier (CSC) feature
• When per CE label is configured, MPLS Forwarding Infrastructure (MFI) has to back up key and label
information to a standby device. This will impact software downgrades.
• The BGP Best External feature provides the network with a backup external route to avoid loss of
connectivity of the primary external route. This feature is not supported.
• Importing routes from protocols other than BGP on a PE device is not supported.
• Any network with a zero next hop is assigned one label per network, because the next hop cannot be
reliably determined.
• Do not use per CE labels if there are multiple neighbors with the same address in a VRF domain.
• Only single hop EBGP is supported. Multihop EBGP is not supported.
• In high availability configurations, labels will be preserved after switchover from standby only if BGP
Graceful Restart is configured before establishing BGP sessions
########################################################################
01-28-2025 07:02 AM
Hi friend
I write note after check many doc. About label type
Per-vrf have two limitations
1- always use ip lookup
2- no support backup path <<- I think you hit this' that why it not work properly
MHM
01-28-2025 07:26 AM
yes and that's I was thinking, but what's the work around. Each ebgp peer routes are getting a single mpls label.
Like >>>
ebgp peer 1 routes gets for example 24050
ebgp peer 2 routes gets 24151
how can we do that additional lookup or what could be the work around for it
01-28-2025 08:25 AM
Hi @DanielGutierrez615 ,
This is a well know limitation, as stated in the following document:
"Label allocation mode per-CE with BVI is not supported in an access network along with PE-CE protocols enabled"
The workaround is to change the "per-ce" label allocation mode to "per-vrf" or "per-prefix" (if limited number of prefixes).
Regards,
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