********************************************************************** IETF 88 PWE3 - Tuesday, November 5, 2013 - 13:00 - 14:00 Regency C (60**/60 min allocated; ** Please note the time adjustments.) ********************************************************************** Chairs: Matthew Bocci and Andy Malis Secretary: David Sinicrope (x = slide sets NOT received as of November 4, 2013 20:00 Pacific Standard Time) 15 min - Agenda bash, WG Agenda and Status - Andy MALIS and Matthew BOCCI Matthew opened the meeting. The slide date is incorrect. Should be November. Matthew went through the status slides. Many expired drafts that need action. Chairs will determine with group which should be progressed. Address IESG and auth 48 requests quickly. 2. 10 min - Update on PWE3 congestion draft - Yaakov STEIN http://tools.ietf.org/id/draft-ietf-pwe3-congcons-01 Objectives: The authors have performed additional calculations and significantly changed the graphics in the draft. Yaakov went through the slides. Expired before IETF88. He will have a new version soon. If TDM PW is under standard top rate equation it is ok or if a bit over. Slides detail the performance for TDM PW New version when graphs sorted out. 3. 10 min - MAC Address Withdrawal over Static Pseudowire - Himanshu SHAH http://tools.ietf.org/html/draft-boutros-l2vpn-mpls-tp-mac-wd-02 Objectives: To introduce new draft. We are increasingly finding deployments of static MPLS TP PWs, especially in MTU-s that are dual-homed in HVPLS configurations. Sending MAC withdraw over static PW has become a key requirement. Himanshu presented slides. Presented by Sami in 2011. Back burnered. Demand for static pw increasing. Work resurrected, Himanshu added as author. Sam Aldrin - going to use oam pkt, Vccv? Himanshu: GAL,GACH,pwtype,... Responder send ack response inband Sam: If mspw, and assuming fault in pw, does spe respond or process the message? HS: Mac learning is done at ingress & egress, so nothing to do with mac flushing at intermediate points. Sam: if packet received at intermediate node then mac flushed on wrong node HS: ms path broken and spe received, but spe not part of instance. George: Mac withdrawal or flushing? HS: can flush individuals or by using mac TLV can flush whole mac table George: the implicit ack looks like it is broken HS: yes, if you missed first 9 and ackd 10 then can't really assume first 9 were processed. The information was lost. Luca: isn't this l2vpn? HS: no, this only discusses PW signaling not what you do with mac list TLV Giles: what happens if you miss acks. This is optimization because bidirectional and will switch over, so corner case. Mathew: # of those that read? 6 or so. Send comments to the list. HS: ach type can get prereserved only for wg drafts so some urgency. 4. 10 min - STP Application of ICCP - Mingui ZHANG http://tools.ietf.org/html/draft-zhang-pwe3-iccp-stp-00 Objectives: Ask for WG adoption so that it becomes a new application of ICCP. Mingui went through slides. Mingui called for wg adoption Claims that similar to DNRI are not true Matthew: How many read the draft? 6 or so. Need comments on the list 5. 5 min - ICCP Application TLVs for VPN Route Label Sharing - Mingui ZHANG http://tools.ietf.org/html/draft-zhang-pwe3-iccp-label-sharing-00 Objectives: Solicit comments and suggestions from the WG. Mingui went through slides. No comments. Matthew: Application of ICCP in L3VPN? Mingui: yes Matthew: Will be presented there. Need feedback from them as to how it fits in their environment. 6. 5 min - Multi-chassis PON Protection in MPLS - Yuanlong JIANG http://tools.ietf.org/html/draft-jiang-pwe3-mc-pon-00 Objectives: Discuss multi-homing of PON attachment circuit (AC) issues in MPLS, get feedbacks from the WG whether ICCP extension for PON is a good direction to pursue. Yuanlong presented slides. Yaakov: in slide 3 how can an optical splitter take 2 inputs? Yuanlong: it is an optical device Yaakov: so what is distributed? ???: need to clarify relationship to PON Yaakov: GPON or EPON? GPON phase locks so splitter can throw whole thing out of phase Andy Malis (via Jabber): It's GPON Type B protection, allows you to dual home splitters to multiple OLTs. Defined in G.983.1 Appendix IV. ???: GPON ranging Yaakov: but ONUs have to phase lock, so if any disruption all onus drop. ???: still not clear Matthew: let author finish Matthew: How many read the draft? 2 people. Comments to the list. **** Time permitting 7. 5 min - MPLS-TP PWE3 dual-homed protection (MPDP) - Hui DENG http://tools.ietf.org/html/draft-cheng-mpls-tp-pwe3-dual-homed-protection-00 Objectives: We would like to get the direction on PWE3 based dual homed protection in MPLS-TP networks from the WG. Hui presented the slides. They have implementation and would like feed back. ???- ALU: can't see why we need this when we have RFC 6718. Sharam: don't need to change the pw, just the PE or AC ???: with AC fail you just set preferences Sharam: no, you need to change Matthew: need to be clear about differences between this and RFC 6718. It has DNI PW which was removed because not in alignment with PW architecture. Sharam: MPLS-tp psc, Amy (HUA): ITU-T has an applicable solution Meeting adjourned at approx. 14:00. ----------------------- ********************************************************************** Overflow (Will be presented if time permits.) ********************************************************************** xx. - None currently ********************************************************************** REMOTE INFORMATION FOR THE PWE3 SESSION(S) ********************************************************************** Remote Participation Info: http://www.ietf.org/meeting/88/remote-participation.html - No WebEx - Audio will be on: Regency C Channel 7 http://ietf88streaming.dnsalias.net/ietf/ietf887.m3u - Jabber room is "pwe3" @jabber.ietf.org