[c-nsp] Sending traffic over different paths

[A. Rahman Isnaini R Sutan]

Quite similar case :)
Since routers don't support MPLS, so we have semi automatic Routing 
Redundancy and Load Balancing with PBR on Router B.


^ Salam
^ A Rahman Isnaini R Sutan



----- Original Message ----- 
From: "Oliver Boehmer (oboehmer)" <oboehmer at cisco.com>
To: "Kristofer Sigurdsson" <kristosig at gmail.com>; 
<cisco-nsp at puck.nether.net>
Sent: Monday, January 23, 2006 11:27 PM
Subject: RE: [c-nsp] Sending traffic over different paths


> Kristofer Sigurdsson <> wrote on Monday, January 23, 2006 11:54 AM:
>
>
>> I was wondering if anyone on the list had any suggestions on how to
>> implement the following:
> [...]
>>
>> Traffic to and from peers via RTR_D (identified by source IP
>> addresses or BGP communities) use Link A to go to and from RTR_A.
>>
>> Traffic to and from peers via RTR_C (identified by a default route or
>> BGP communities) use Link B for traffic to and from RTR_A.
>>
>> Should either Link A or Link B go down, all traffic goes through the
>> other one.
>
> Even though I'm not sure I understood your exact topology, MPLS Traffic
> Engineering, possibly using static and/or policy-routing at the edge, is
> usually a good technology to achieve this. You build two tunnels from A
> to C and D, and auto-route C's next-hop to the tunnel (and likewise
> towards D). Same goes for the reverse direction. No need for
> policy-routing in the middle as this is done during tunnel setup. In
> case your links go down, tunnel can be re-optimized over the alternate
> path or can go down and you will just forward using your regular IGP..
>
> oli
>
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>