BGP routing with TE Tunnel

Hi,

Loopback is advertised via OSPF and all all MPLS protocols are configured (LDP, RSVP).

I tried to read the wiki “http://wiki.mikrotik.com/wiki/Manual:TE_Tunnels” but the traffic is not forwarded on the tunnel for all L3 VPN.

I need to statically route the traffic to the tunnel for it to work but it is not possible with all 2500 routes received via bgp vpn.

Setup is a mix of Juniper and Mikrotik CCR1072 in an ISP environment (traffic is automatically forwarded on LSP tunnels on Juniper as soon as it is configured but not on Mikrotik).

Has anyone have the same issue? What to do insuch case?

Thanks,

Manantsoa R.

Traffic will only be automatically forwarded through the tunnels if some criteria is met, read http://wiki.mikrotik.com/wiki/Manual:TE_Tunnels#Forwarding_traffic_onto_TE_tunnels carefully.

Already read it but L3 VPN BGP traffic is not forwarded on the tunnel even if the endpoint of the tunnel is the BGP next-hop received via IGP.

It is only working if I make a static route for the remote destination to go through the tunnel.

Any ideas as it seems that someone else posted it in the forum in the past but never had a reply?

The same here.

Can someone show a real example of using RSVP LSPs to forward traffic towards BGP known destinations?

For now, the workaround is to set a route-filter on import with "set-in-nexthop-direct=“LSP_to_whatsoever_ROUTER” and to filter out OSPF prefixes leaving just loopbacks and interconnection links.

On Junos software is much easier and understandable:
“set protocols mpls traffic-engineering bgp-igp-both-ribs”.

This command copies routes from table inet.3 (RSVP) to inet.0 (global table).

That way, BGP and OSPF known routes are less specific than RSVP because of the administrative distance.

So, you need OSPF for initial convergence and BGP for other, external routes announcing as well as other protocol signaling. After that, RSVP comes in places and does all the magic.

I’ve tried in the lab all kind of different scenarios to meet the requirements stated on wiki:
(http://wiki.mikrotik.com/wiki/Manual:TE_Tunnels#Forwarding_traffic_onto_TE_tunnels)

  • traffic that is routed using route route learned from BGP, if BGP NextHop is tunnel endpoint (this default behaviour can be changed by setting route porperty “use-te-nexthop” to “no”), both - regular IP and VPNv4 (MP-BGP IP VPN) routes fit in this category;

^^^ DOES NOT WORK!

  • traffic for VPLS interfaces, if remote endpoint of VPLS pseudowire is the same as TE tunnel endpoint.

^^^ Confirm, it behaves how it’s written

If anybody can, please say that I’m wrong and show us a real working example! : )

PS: and stop publishing posts with examples of static routes over LSPs and even worse, - /24 subnets on 2 LSPs as they are an Ethernet link.
LSPs are unidirectional:

“An MPLS connection (LSP) is unidirectional—allowing data to flow in only one direction between two endpoints. Establishing two-way communications between endpoints requires a pair of LSPs to be established. Because 2 LSPs are required for connectivity, data flowing in the forward direction may use a different path from data flowing in the reverse direction.”

Guys, serious… For those who create those articles, …serious?! :

Same problem here as well. Even created an import filter for my BGP peer to specifically set “set use te-nexthop” to yes, and no go still. Any advice to avoid having to static route from the head-end in to TE tunnels would be great. Thank you.

So I found that while my routing table won’t update to show the TE tunnel as the forwarding next-hop, nor do I see the TX stats increment at all on the “MPLS->Traffic Engineering->Traffic Engineering Interface” window, I do see traffic when I monitor the TE tunnel on the terminal, so it does appear to be forwarding. I also see RSVP working to adjust bandwidth as I go. My question now would be does anyone have a best practice for the TE bandwidth tab? I have my tunnel set to a minimum bandwidth of 1Meg, an auto-range of 1M-100M, bandwidth reserve of 30%, avg. interval of 10s and an update interval of 1min. When I activate the tunnel, I in turn appear to be choking traffic to my subscriber base down to an unusable level before the first update interval hits.

@ntsoa79 did you find a workaround?

Is this bug solved already ? or V7 , 8 , 9 , 10?