Dear Authors,
please find my comments below:
- page 9, first para s/we loose out/we lose out/ - two times
- page 9, second to last para "Partition of the network into incompatible
and unconnected islands is neither desirable nor acceptable." While I agree
with the former, the latter is highly subjective and absolutely
unenforceable. As vendors we give operators loaded gun and a warning. What
they do with them might surprise and amaze us all.
- Section 4.4, first para: "There are three MPLS signaling control
protocols used for distributing labels to set up LSPs and PWs in MPLS
networks: LDP, RSVP-TE, and GMPLS." Perhaps GMPLS in this list should be
replaced by BGP/MPLS. RSVP-TE is equally used as signaling protocol to
distribute MPLS and GMPLS label information. There are three paradigms that
operate with distinct sets of constructs:
- LDP MPLS, a.k.a. IP/MPLS;
- TE-MPLS;
- TE-GMPLS.
- Section 4.4, second para. Would note that relationship between LDP and
TE-(G)MPLS networks might be not only as peering but as client-server, e.g.
LDP over RSVP-TE tunnels that could be MSPL-TP.
Regards,
Greg
_______________________________________________
Ietf mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ietf
