Hi Chris, "for each interface intf of x" in Figure 8: Computing Low-Point value & "for each link (x, w)" in Figure 7: Basic Depth-First Search algorithm is based on section : "5.1. Interface Ordering" ? If yes can it be made more specific, since this very important part of logic, if not let me know how & where section : "5.1. Interface Ordering" is used in the algorithm.
Thanks & Regards Anil S N "Be liberal in what you accept, and conservative in what you send" - Jon Postel From: Chris Bowers [mailto:[email protected]] Sent: 10 June 2015 00:29 To: Anil Kumar S N (VRP Network BL); Gábor Sándor Enyedi; [email protected]; Alia Atlas; [email protected] Cc: [email protected]; [email protected] Subject: RE: [rtgwg] draft-ietf-rtgwg-mrt-frr-algorithm-03 Anil, Thanks for the suggestion to clarify the use of root to mean gadag_root or spf_root in the pseudo-code, as well as the typo. I made the changes on github. The diff can be found at: https://github.com/cbowers/draft-ietf-rtgwg-mrt-frr-algorithm/commit/ada619050ec9d773b7919a1c622f068d5a5a5e88 Tell me if you agree with these changes. With respect to comment#3, if ((D is F) or (D.order_proxy is F)), then there are several cases to consider: 1) If the link from S to F is a cut-link, a) if this is a single cut-link between S and F, then there is no alternate b) if there are parallel cut-links between S and F, then one can, for example, ECMP across the remaining links, noting that there is no link protection. 2) if the link from S to F is not a cut-link, then at least one of the MRT next-hops for D (red or blue) will not be the same as the primary next-hop for D. In which case, one should use the color that is not the primary next-hop as the alternate, noting that the alternate does not provide node protection. I agree that the existing pseudo-code is not very clear here. I am planning to update this part of the pseudo-code in the near future to make it clearer, but hopefully the explanation above suffices for the moment. Chris From: rtgwg [mailto:[email protected]] On Behalf Of Anil Kumar S N (VRP Network BL) Sent: Tuesday, June 09, 2015 1:27 AM To: Gábor Sándor Enyedi; [email protected]<mailto:[email protected]>; Alia Atlas; Chris Bowers; [email protected]<mailto:[email protected]> Cc: [email protected]<mailto:[email protected]>; [email protected]<mailto:[email protected]> Subject: [rtgwg] draft-ietf-rtgwg-mrt-frr-algorithm-03 Hi Authors, As discussed before, Please find my review comments : Comment 1: Can we rename parameter which is passed to these functions as real SPF root or GADAG root ? Run_DFS(node root) Run_Lowpoint(node root) Compute_Localroot(root, root) Construct_GADAG_via_Lowpoint(topology, root) Add_Undirected_Links(topo, root) Assign_Block_ID(root, max_block_id) Compute_MRT_NextHops(x, root) Comment 2: Here parenthesis are not matching, four '(' and five ')'. This must be typo mistake. In_Common_Block(x, y) if ( (x.block_id is y.block_id)) or (x is y.localroot) or (y is x.localroot) ) return true return false Comment 3: Is it possible to rephrase "if an MRT doesn't use primary_intf" What does the sentence "MRT doesn't use primary_intf" mean ? Dose it mean neither Red interface nor Blue interface is same as primary interface ? What does the sentence "return that MRT color" means ? Select_Alternates_Internal(S, D, F, primary_intf, D_lower, D_higher, D_topo_order) //When D==F, we can do only link protection if ((D is F) or (D.order_proxy is F)) if an MRT doesn't use primary_intf indicate alternate is not node-protecting return that MRT color Thanks & Regards Anil S N "Be liberal in what you accept, and conservative in what you send" - Jon Postel
_______________________________________________ rtgwg mailing list [email protected] https://www.ietf.org/mailman/listinfo/rtgwg
