Junos OS

Hi Everyone,

I have a MPLS network which is using AS6500 for iBGP.

I am using an SRX300.
After further checking, it seems like I am not receiving routes from other iBGP neighbor.

These routes are hidden.

For example...

10.15.0.0/16 [BGP/170] 00:52:49, MED 0, localpref 100, from 10.60.0.10
AS path: 64620 I, validation-state: unverified
Unusable

When I perform the show route hidden extensive, the output for the above route is...

10.15.0.0/16 (1 entry, 0 announced)
BGP Preference: 170/-101
Route Distinguisher: 64512:100
Next hop type: Unusable, Next hop index: 0
Address: 0x16a1658
Next-hop reference count: 101
State: <Secondary Hidden Int Ext ProtectionCand>
Local AS: 65000 Peer AS: 65000
Age: 53:49 Metric: 0
Validation State: unverified
Task: BGP_65000.10.60.0.10
AS path: 64620 I (Originator)
Cluster list: 10.60.0.10
Originator ID: 10.255.30.34
Communities: target:64512:100
Import Accepted
VPN Label: 18
Localpref: 100
Router ID: 10.255.30.1
Primary Routing Table bgp.l3vpn.0
Indirect next hops: 1
Protocol next hop: 10.255.30.34
Label operation: Push 18
Label TTL action: prop-ttl
Load balance label: Label 18: None;
Indirect next hop: 0x0 - INH Session ID: 0x0

Can someone please guide me on how to overcome this issue?

Do note that this PE is a route-reflector-client.

Thanks.

Hello,

For 10.15.0.0/16 route to be unhidden, 10.255.30.34 route must be present in inet.3 table.

And inet.3 table is populated by either LDP, RSVP or BGP-LU.

If You don't have at least 1 of these protocols enabled, then it's high time to do it, otherwise Your MPLS services won't work.

HTH

Thx

Alex

Hi Alex,

Thanks for the response.
I am running MPLS TE via RSVP in my network which also consist of Cisco network devices.

I have attached the simplified version of the network diagram in this post.

Each SRX has only the LSP and BGP to the Main_PE device (which is a cisco 6500).

The Cisco 6500 is acting as a route-reflector.

You are right in the sense that there is no 10.255.30.34 in my inet.3 table.

I only have the Main_PE device in my inet.3 table.

The reason why I have route-reflectors is because the number of PEs that I will have will be expanding and for scalability purpose, route-reflector was implemented.

This would be an issue with the route-reflector right?

Regards,

Han

Hello,

---

@cHrome08 wrote:

 

You are right in the sense that there is no 10.255.30.34 in my inet.3 table.

I only have the Main_PE device in my inet.3 table.

The reason why I have route-reflectors is because the number of PEs that I will have will be expanding and for scalability purpose, route-reflector was implemented.

This would be an issue with the route-reflector right?

 

 

 

---

I would say this is an issue with Your network architecture.

The RR is clearly doing its job and reflecting the L3VPN route WITHOUT changing the nexthop.

Which is the default and proper behaviour for iBGP L3VPN RR. And then You would need the end-to-end, or "full-span", or PE-to-PE LSP for the L3VPN to work properly.

You may get away with curent "half-span" LSP but then You need to :

1/ make RR to change nexthop to self, and

2/ configure all VRFs on the RR

HTH

Thx

Alex

Hi Alex,

Thanks for pointing me to the right direction.

It seems like the route-reflector command was not reflecting the routes from iBGP neighbor.

Its only reflecting the routes from other eBGP neighbor.

I just had to modify the config from...

address-family vpnv4

neighbor PeerGroup2 send-community extended

 neighbor PeerGroup2 route-reflector-client
neighbor PeerGroup2 next-hop-self

exit address-family

to

address-family vpnv4

neighbor PeerGroup2 send-community extended

 neighbor PeerGroup2 route-reflector-client
neighbor PeerGroup2 next-hop-self all

exit address-family

The "all" command on the cisco 6500 will route-reflect every routes learned from eBGP and iBGP neighbor and that solves the issue.

Thanks for your help.

Regards,

Han