[j-nsp] Limit on interfaces in bundle

[Saku Ytti]

On 2 November 2015 at 15:30, Jesper Skriver <jesper at skriver.dk> wrote:

Hey,

> 1) Keep all the logic as today, but for different received
>    ethertypes point to different L3 tables, that result in
>    different rewrite objects that ensure the outgoing packets
>    have the different ethertypes requested.  This require that you
>    replicate the MPLS table for each of your ethertypes.
>
> 2) Change the logic such that the incoming ethertype is
>    remembered, and on egress do not use a static ethertype, but
>    instead do something like
>       if ( input ethertype is one of the MPLS ones &&
>            output ethertype is 'mpls' &&
>            no operation performed changed the nature of the payload ) {
>          output_ethertype = input_ethertype
>       }

What I'm thinking is quite like 2), but not as explicit/tight
coupling, and employing existing classification infra.

We already classify packet in many ways on ingress, QoS type, L2, L3,
L4 protocol etc. This would just be another classification we assign
on the packet on ingress. And on egress, based on this internal
classification different ethertype is written.

-- 
  ++ytti