I have reviewed this document as part of the security directorate's ongoing effort to review all IETF documents being processed by the IESG. These comments were written primarily for the benefit of the security area directors. Document editors and WG chairs should treat these comments just like any other last call comments. This document alters how BGP Route Reflector computes the optimal routes on behalf of clients. Instead using its own IGP cost to the AS Exit points, the document describes the steps for RR to compute the optimal route by using Clients' position to the AS Exit points. The described method is useful when RR is centralized. For deployment with distributed RR closer to the clients, the described method doesn't have any benefits. Security Concern: If RR's information of its clients topology is compromised, then the optimal paths selected by the RR might not be accurate anymore. Minor nits: Page 7: Section 3.2. "If the routing routing optimization requires ..." Is it a typo? duplicated word "routing"? Last sentence: "This needed for use cases ..." Do you mean "This is needed for use cases ..." Cheers, Linda Dunbar