Re: [Idr] Revised text on how to construct the AS path info

Enke Chen <enkechen@cisco.com> wrote:
> 
> What types of "inconsistencies" between AS_PATH and NEW_AS_PATH you are 
> concerned about, beyond the one described in the Transition Section 

   To tell truth, I'm not concerned about any particular case: I'm
concerned about the general case. I expect whatever algorithms we
specify to be still in use five years from now, in weird situations
we _cannot_ imagine.

> If your concern is about bogus AS path info somehow got propagated 
> around, that seems to be a general issue, independent of the 4byte AS.

   Indeed, I am largely concerned about things which are a general
problem now. My concern is that these problems will lead to different
behaviors and that the trail will be harder to follow.

   I believe there are simple consistency checks that could be run,
and that if the consistency checks (between 2-byte AS_PATH and 4-byte
NEW_AS_PATH) fail, we could either discard the NLRI or tag the 4-byte
AS_PATH we will pass on so that other BGP speakers can know the
inconsistency was found.

   (Paul and I, BTW, haven't reached agreement on which of these options
is better. My concern is to establish the general principle, not to
argue the details.)

--
John Leslie <john@jlc.net>

_______________________________________________
Idr mailing list
Idr@ietf.org
https://www1.ietf.org/mailman/listinfo/idr