RE: [L2tpext] draft-galtzur-l2tpext-gr-01.txt and draft-ietf-l2tpext-failover-02.txt

Certainly failover-02 needed to benefit from feedback and refinement, but this is not unusual.  Vipin and Paul have made great progress offline.  I don't think the "newness" of the technology should be viewed as an obstacle.  

That said, I don't believe it is possible to have a single mechanism that will satisfy the perspectives of both the "static" and "dynamic" models.  If the group believes with deep conviction that these two models should be defined as important separate applications of l2tp, important enough that each perspective warrants such satisfaction, then I think the only way to proceed is to allow two solutions.  Each solution would have to come with a clear explanation of the tradeoffs between them.  At a minimum, the failover-02 draft will have to be enhanced to take care of v3.

I think it is most unfortunate for the (nay, any) protocol to have more than one mechanism for failure recovery.  It sets a bad precedent, as someone in the future may think of another l2tp application that benefits from yet another recovery mechanism.

Tom

_______________________________________________
L2tpext mailing list
L2tpext@ietf.org
https://www1.ietf.org/mailman/listinfo/l2tpext