Dear WG, Based on the discussion in IETF 109, here are the detailed response to the questions in the meeting:
1. How to differentiate the condition the route is node down vs. link down? There are 3 stages in the mechanism: in the 1st stage, link failure and node failure are both treated as node failure, just as FRR mechanism; in the 2nd stage, node failure will cause FIB miss and link failure will converge; in the 3rd stage, it not relevant whether it is node failure and link failure; 2. What about the function supposed to be executed at node E? E as a middle point, the possible functions defined in E include: End, End.X, End.T, End.B6 ( https://tools.ietf.org/html/draft-ietf-spring-srv6-network-programming-25) ; Except for End.T , other functions are all used for directing forwarding (End.B6 is also used to bind a path which is composed of End, End.X, directing forwarding). The proxy behavior is for path repair which guarantees the reachability and other functions can't be agented. So End.T protection is not supported. 3. Could TE path be changed when doing protection? 1) Middle point protection is for temporary reachability repair when failure happens in the TE path. In the stage 3 of the mechanism, the ingress/controller/PCE will reroute and the attributes of TE will recover. Although unexpected path will be caused by this mechanism temporarily, we believe reachability repair is still necessary. 2) In MPLS TE Protection, the similar behavior is allowed which is defined in RFC4090. 3) (why reroute triggered by ingress node is not used here) middle point protection is usually used when there is no E2E BFD. And in this case, middle point protection could be the supplement to the E2E protection. Hope this will help to understand the mechanism. The document will be updated based on these comments. And if there are any other questions, please also let us know. Best Xuesong
_______________________________________________ rtgwg mailing list [email protected] https://www.ietf.org/mailman/listinfo/rtgwg
