Minutes for LISP at IETF-92
minutes-92-lisp-1

Administration
---------------------------------
	
	Chairs remind the decision to move LISP to routing area. 
	Deborah Brungard will be the new AD following the group activities.
	
	Chairs list of active documents
		draft-ietf-lisp-ddt is waiting for the shepherd writeup
		draft-ietf-lisp-impact is on last call
	
	Chairs asking about the plans for draft-ietf-lisp-lcaf
		the document has some very stable part about the 
		procedure and the general format. It is suggested to 
		focus the document on that part and propose clear 
		procedure to decided how to add now types, via 
		the creation of a dedicated IANA registry
		 Luigi: will provide information about how this 
			can be done with IANA

	Chairs asking about the plans for draft-ietf-lisp-sec
		Vina: indicates the document is stable since 
		a long time and can be put in last call.
		Luigi: will review the document to decide 
		if it can go for last call.


WG Documents Update
---------------------------------
	
	LISP Threats (Damien Saucez)
		Ron: asks for clarification about the risk 
		of threats caused by the mitigations themselves
		Damien: the document explains that the 
		mitigation can be used for attacks and the 
		threat model is such that it covers this case.
		Luigi: reminds that the draft is not about 
		implementations details.
		Chairs asking if document is ready for last 
		call: there was consensus on accepting last call.
	
	LISP Impact (Damien Saucez)
		Damien: asks people to share their experience 
		with LISP on the mailing list.
		Chairs asking if document is ready for 
		last call: there was consensus on accepting 
		last call.
		
	LISP crypto (Dino Farinacci)
		Dino: summaries the comments from security 
		experts.
		Dino: asks the attendees for deep review 
		during the week.
		Dino: asks for inter-operatbility tests.
		Ron: asks for clarification on authentication 
		and privacy in the exchanges.


Non WG Documents
---------------------------------

	Signal-Free LISP Multicast (Dino Farinacci)
		Joel: asks for clarification on the number 
		of layers.
		Dino: it can be up to 128 layers and explains 
		how it works.
		Lucy: what type of packet travel from 
		ITR to RTR?
		Dino: multicast packet encaspulated in 
		unicast packet.
		<Missed Name>: asks if the failure of the 
		RTR can cause problems
		Dino: no single point of failure, you can 
		have as many RTR as you want
		<Missed Name>: asks how levelling is done 
		and decided.
		Dino: that depends, this is the RLE that 
		defines it. 
		<Missed Name>: Up to how many replication 
		can an ITR replicate to?
		Dino: we tried replication of 4 and it worked.
		Luigi: what if you have multiple servers 
		for the tree?
		Dino: in practice you should be register 
		in the same place in the DDT so it should not 
		cause problem.
		Luigi: ask to add the DDT discussion in the 
		document
		Dino/Joel: explain that bi-directional flows 
		are implemented using two different trees.
		Joel: the assumption is that the number of 
		sources is small vs the number of destinations. 
		This document does strictly solve that, other 
		solutions have to be used if the number of 
		sources is large.
		

	LISP YANG Model (A. Rodriguez-Natal)
		Joel: asks if authors have asked the YANG 
		doctors to see if the choices are pertinents
		Alberto: not discussed with them yet.
		Dino: asks if DDT is covered.
		Alberto: says it is not covered it yet.
		Dino: asks how it can deal with different 
		types of mapping system, one model per 
		mapping system or a general one?
		Vina: the current model has the option to say 
		which mapping system you use.

	Overflow Time/Discussion
		Chairs: what do we want to be put in the new charter?
		Darrel: we could start with some use cases.
		Dino: agrees with Darrel. We should give a chance to 
		each of these use cases. Probably most is already 
		there and few new machinery would be necessary.
		Larry: says that the group should look more on how 
		to integrate with NVO3.
		Dino: if LISP solves the VM mobility use case, is 
		that ok NVO3?
		Larry: it is one case among among others.
		Dino: asks if there is any use case in NVO3 that 
		LISP can’t do.
		Dino: asks for new about draft-ietf-lisp-intro
		chairs: indicate that the last few fixes are 
		being done.