Re: [Idr] draft-litkowski-idr-flowspec-interfaceset => NLRI vs EXTCT (part 2)

[Jeffrey Haas <[e-mail]>]

Acee,

On Fri, Jul 28, 2017 at 12:02:41PM +0000, Acee Lindem (acee) wrote:
> What was the reason for not making the interface-group part a new component type in the NLRI? It seems that making it just another Flow Spec match condition would  handle all these situations naturally.
> 
>  I remember that this was discussed but don’t recall whether it was backward compatibility or that the semantics of interface group-id was a better fit for an extended community.

A few motivations:
- Unknown components in the filter cannot match.
- The match elements in the filter (NLRI) might be as-transitive, but policy
  engines don't interact with such filters transitively very well in
  implementations we're familiar with.  However, pretty much everything can
  manipulate extended communities.

The first consideration is partially about incremental deployment.  You
either have the choice of shipping a filter across your network that may be
selectively restrictive per the intent of the group-id or completely
ignored.

The best answer, of course, is consistent deployment of the updated feature.
However, this is the course we'd chosen as part of the second consideration.

-- Jeff