Re: mpls vpn filtering on interface

On a second thought, being this a QoS issue, it might be a better idea
to differentiate the flows with different FECs based on exp bits.
TE policy can do different things on different exp bits too.

-Carlos

Carlos G Mendioroz @ 30/09/2011 06:57 -0300 dixit:
> TE is your friend, combined with an export map to differentiate NH to
> different VRFs ? I'll check if this works, but sounds doable :)
> -Carlos
>
> Sergey Matashuk @ 30/09/2011 05:44 -0300 dixit:
>> Dear experts
>>
>> I have an interesting question.
>> Lets say there are two VRFs VRF_A and VRF_B on PE router Router1. Router1
>> has BGP connection from loopback interface with two BGP RRs, located
>> at MPLS
>> could and there is OSPF used as IGP. Both VRFs exits at abother PE
>> device -
>> router2, located on another side of our MPLS cloud, so routes on that
>> VRFs
>> have same BGP NH.
>>
>> Router1 has two links to MPLS cloud, link1 is 10Gbps and link2 is
>> 100Mbps.
>>
>> VRF_A is heavy loaded by traffic, say it has 5Gbps throughput, assume,
>> VRF_B
>> with throughput about 80 Mbps used for critical data.
>>
>> Under normal condition both VRF uses link1. But when link1 fails ospf
>> gets
>> them rerouted to link2 (as BGP NH accessible on both links). Now, as
>> VRF_A
>> data is not critical service the task is to filter VRF_A on link2.
>>
>> What are your thoughts, guys?
>>
>> --
>> Regards,
>>
>> Siarhei Matashuk
>> CCIE #27340
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
>>
>>
>>
>>
>>
>>
>>
>

-- 
Carlos G Mendioroz  <tron_at_huapi.ba.ar>  LW7 EQI  Argentina
Blogs and organic groups at http://www.ccie.net