Hi Adrian, I tend disagree with this statement: "My contention is that it is quite likely that someone would try to use a legacy NMS to manage a new flexigrid network since "it is all just GMPLS". And it is not an unreasonable assumption to be able to do this if some fields (for example, label) can be displayed as opaque quantities." I'm not sure an operator would be happy to see 96 channels called lambda1, lambda2,..., lambda 96 from his NMS while the network is flexi grid. We'll have to face these issues when we'll be speaking about YANG models for flexi grid but for the time being let's be on the safe side and address the compatibility issue as you're suggesting. Cheers, Daniele > -----Original Message----- > From: Adrian Farrel [mailto:adrian@xxxxxxxxxxxx] > Sent: martedì 8 settembre 2015 18:02 > To: Daniele Ceccarelli; 'Adrian Farrel'; 'Robert Sparks'; draft-ietf-ccamp- > flexigrid-lambda-label@xxxxxxxx; ccamp@xxxxxxxx; 'General Area Review > Team'; ietf@xxxxxxxx > Subject: RE: Gen-art LC review: draft-ietf-ccamp-flexigrid-lambda-label-04 > > Hi Daniele, > > > Thanks for the careful review and your comments. > > I pretty much agree with Adrian's reply but I think explicitly having > > some backward compatibility text in the draft could be helpful. > > > > Adrian, authors, I'd suggest changing section 5 from "Manageability > > Considerations" to "Backward Compatibility and Manageability > Considerations" > > adding to the existing text backward compatibility considerations > > against legacy GMPLS and legacy NMS (mostly what you've already written > below). > > I can work on this. > > > WRT the legacy NMS I don't think it is a reasonable scenario, since > > before operating the nodes with a GMPLS implementing this draft, the > > node needs to be configured and the NMS must be flexi-grid compatible. > > But wait! This is exactly the point. > Suppose there is an NMS that is inspecting an LSP at a transit node. > That NMS does not need to be flexigrid compatible to read the details of the > LSP at that transit node. > However, it *does* need to have some capabilities in order to not barf when > it gets the response from the LSR. > The first thing is to handle a 64 bit label as an opaque string. > The advanced thing is to be able to parse out the 64 bits into the relevant > fields. > > Indeed, to request the setup of a flexigrid LSP does not require a flexigrid > compatible NMS since it is possible to simply request a path and bandwidth > (or not even request a path). > > My contention is that it is quite likely that someone would try to use a legacy > NMS to manage a new flexigrid network since "it is all just GMPLS". And it is > not an unreasonable assumption to be able to do this if some fields (for > example, label) can be displayed as opaque quantities. > > Now, if you are talking about an EMS, I completely agree. > > Cheers, > Adrian >