DIME 75 minutes Victor Fajardo chairing. Status (charts) Victor Agenda accepted. Diameter API status (Victor) ============================ Charts. Rejected by IESG. Question to group: any interest in it now? No one indicated it would be useful. Glen suggested it might be helpful for beginners. Will decide fate on the list. Base protocol status (Victor) ============================= Waiting for Hannes's review. Glen sent E-mails re section 9 (July 14) that need action. QoS attributes status (Mark Jones) ================================== Charts. Review of changes -> versions 12, 13 Additional DISCUSSES to resolve. Dan Romascanu gave further information. Diameter ERP (Stephane Decugis) =============================== Charts. Number of recent E-mails. Question: new application for ERP or reuse EAP? Presentation of information flows: Glen questioned term "EAP Proxy". Intended to be a Diameter routing entity. Simon Mizokovsky questioned how "Local EAP server" is reached. Qin Wu talked about home EAP server, but Stephane avoided that terminology because it is confusing. Routing issue: Lionel: additional complications in routing. Clarified intention here. Simon: could achieve separation by providing different user identifiers for EAP and ERP Other discussion Hannes question on Jabber. Answer "Yes" Pros and cons of separate application ID Qin Wu -- clarifications "Hum" re separate app ID: no disagreement, positive support. Last slide -- lots of open questions for further consideration NAT Control (Dec Wojciech on behalf of Frank) ============================================= Charts. More complete analysis of existing protocols vs. required capabilities. WG comment requested. Dan R reporting Hannes on Jabber: protocol analysis cannot become WG item. Protocol can, assuming it is Diameter. Diameter Capabilities Update (Glen) =================================== Need a couple of more reviews. PCN Data Collection (Tom) ========================= Hannes pointed out that some of the structures defined in the QoS attribute work might be applicable -- particularly the classifiers. Given that the details of the PCN parameters are still under discussion, Dan suggests waiting for that to settle and consequent update of this draft before deciding whether to take it on as a work item. Diameter WebAuth (Xiaoming Fu) ============================== Comment: RFC 5090 never implemented -- too much latency. Talk to the authors to learn more. Hannes commented: It is a small sub-set of the Diameter SIP application. The diameter SIP application turned out to be quite SIP specific. This document re-uses the relevant AVPs from the Diameter SIP application document. Realm-based redirection (Tom) ============================= Suggestion that we could think about the equivalent to Redirect-Host-Usage is required. General feeling that this is a "FOREVER" type of redirection. Could check with list. One or two more reviews forthcoming. RADIUS-Diameter Gateway Application (Glen) ========================================== Mark: problem statement different from what Glen had suggested. Comment: what do people think the RADIUS-Diameter transition problem is, and why? Simon: WiMAX specifies use of both RADIUS and Diameter. Need separate sets of auth data depending whether you come in through Diameter or RADIUS-Diameter because Diameter RFC 4005 provides data too large for RADIUS. Hannes comment: My investigations on the transition concluded that there isn't really a need to translate RADIUS to Diameter (and vice versa). Nobody seems to be doing that. Regarding Wimax: Currently most Wimax RADIUS and Diameter work is still on paper. We will have to wait and see how their deployments will work out. Further discussion. Bernard summation: it would be really nice to capture this discussion in a draft. He agreed to take it up as a work item. Question whether the architecture implied by this application is useful. Victor conclusion: need more information to decide on this document. Bernard requested people to post to list their experiences in using RFC 4005. Document Set (Glen) ============ 3588bis document set was horribly out of date. Glen proposes BCP to be updated whenever content changes. Question on scope. Would it be worth the effort? Ask on the list. Locating Mobility Anchor Resolution for PMIP6 (Marco) ============================================= Update on problem statement. Had proposed a query mechanism addressed to the policy store. New development: new work in other WGs could use this capability. Simon questioned scope of the proposal. AVPs for Cryptographic Key Transport (Qin Wu) ==================================== Charts. Backward-compatible update to RFC 4072 to accommodate ERP as well as EAP. Concern over reuse of EAP-Key-Names because it changes the meaning of the contents. Will break RADIUS. (Glen points out draft keeps original meaning.) May be missing AVP within EAP-Key as shown on chart. HannesTschofenig asked: Isn't the transport of ERP keys defined already in the Diameter ERP document? Can someone ask this question on the microphone? Why do we want to make the key transport more generic? It isn't that we have so many document that send keys around. EAP and ERP and that's it. He continued by questioning whether 3GPP would dump existing stuff in favour of the new transport. Take it to the list for comment. Idea for Future Hums (Victor presenting) ==================== Charts. Reviewed suggestions, asked for comments on the list. Lionel now: looking for more info in draft or by other means? Victor: in draft. Generally a check list. Glen: So happy that DIME has so much work being presented to it that we need flow control! HannesTschofenig: The problem statement in the draft is sometimes related. But it does not provide the same amount of information. This is not flow control. We would like to get some additional background on the documents. If you, for example, plan to implement something then please state it. If you did investigations regarding deployment then the group needs to know about it. It is valuable information. Dan: not that we need flow control, but there is a fair amount of work. Sees a problem because two documents reached IESG in very poor shape and out of scope. Glen questions how this could have happened. Skeptical. Lionel: need clear problem statement in draft itself. Later, no need for tutorial in draft once it is mature. Glen OK with executive summary. Appears some of IESG have major need for education. Further discussion to list. Dan: two different issues. First is when we start work on something. The second is when we hand off to IESG.