On Mar 27, 2009, at 8:28 PM, UTTARO, JAMES, ATTLABS wrote:

How and which routes send back for importation is
really the discussion.

Indeed.  And to my point again, I hope vendors don't
blindly reflect all routes back to a client in order
to implement this particular capability, as it's effects
are far wider and systemic, and these extraneous updates
are simply not benign.

I think I've stated my point several times now, and even
illustrated some of the concerns with this particular
issue.  I trust operators and implementers that are
concerned with implementation and systemic scale and the
effects on convergence and stability in their environments
will have considered it.  Those that don't believe it's an
issue, I have no intentions to further concern you as such.

-danny
_______________________________________________
GROW mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/grow

Reply via email to