In message <[EMAIL PROTECTED]> JP Vasseur writes: > > Hi Curtis, > > On Nov 30, 2006, at 8:48 PM, Curtis Villamizar wrote: > > > > > In message <[EMAIL PROTECTED]> > > JP Vasseur writes: > >> > >> Other attributes such as affinity should be used to not allows 0-bw > >> TE LSP to traverse a specific link. This TLV is only used to report > >> the number of such TE LSPs traversing the link. > >> > >> Thanks. > >> > >> JP. > > > > > > The provider already has the necessary tools that can be used to > > accomplish this. If a general purpose tool (attributes and > > affinities) is available which accomplishes something a special > > purpose tool to accomplish the same thing is not needed. > > > > Such a tool would only be useful if the administration of the MPLS > > midpoint (where the attribute is set) had no control over the > > administration of the MPLS ingress or a border that is doing route > > computation (where the affinity is set). I don't see any anticipated > > real world deployment that would benefit from this. If you do, then > > please explain the deployment scenario. > > > > not sure to see your point here ... I was mentioning that the aim of > this TLV was not to avoid some links. > Looks like you're saying the same thing. > > JP. > > > Curtis
My point was regarding a mention in that conversation that the value of zero might mean no zero-BW LSPs have been set up or none are allowed. Looks like I cut the wrong paragraph out of the conversation. The existing link attributes and affinity are sufficient to indicate the condition that no zero-BW LSPs are allowed if the administration of that network decides to use an attribute for that purpose. I'm agreeing with you. Curtis _______________________________________________ OSPF mailing list [email protected] https://www1.ietf.org/mailman/listinfo/ospf
