Network Working Group T. Clausen Internet-Draft A. Camacho Intended status: Informational J. Yi Expires: June 14, 2013 A. Colin de Verdiere LIX, Ecole Polytechnique Y. Igarashi H. Satoh Y. Morii Hitachi, Ltd., Yokohama Research Laboratory U. Herberg Fujitsu Laboratories of America C. Lavenu EDF R&D December 11, 2012 Interoperability Report for the Lightweight On-demand Ad hoc Distance- vector Routing Protocol - Next Generation (LOADng) draft-lavenu-lln-loadng-interoperability-report-04 Abstract This document reports experience with the LOADng routing protocol, as obtained by way of a number of interoperability tests during the protocol development. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at http://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on June 14, 2013. Copyright Notice Copyright (c) 2012 IETF Trust and the persons identified as the document authors. All rights reserved. Clausen, et al. Expires June 14, 2013 [Page 1] Internet-Draft LOADng Interop Report December 2012 This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. This document may contain material from IETF Documents or IETF Contributions published or made publicly available before November 10, 2008. The person(s) controlling the copyright in some of this material may not have granted the IETF Trust the right to allow modifications of such material outside the IETF Standards Process. Without obtaining an adequate license from the person(s) controlling the copyright in such materials, this document may not be modified outside the IETF Standards Process, and derivative works of it may not be created outside the IETF Standards Process, except to format it for publication as an RFC or to translate it into languages other than English. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 3. Interoperability Scenarios . . . . . . . . . . . . . . . . . . 5 3.1. Scenario 01: 1-hop Bidirectional Route Establishment - Forward Route and Reverse Route initial installation . . . 5 3.1.1. Scenario Topology . . . . . . . . . . . . . . . . . . 5 3.1.2. Expected Message Sequencing . . . . . . . . . . . . . 6 3.2. Scenario 02: 1-hop Bidirectional Route Establishment -Forward Route and Reverse Route updating . . . . . . . . 6 3.2.1. Scenario Topology . . . . . . . . . . . . . . . . . . 6 3.2.2. Expected Message Sequencing . . . . . . . . . . . . . 7 3.3. Scenario 03: 2-hop bidirectional route establishment - Forward Route and Reverse Route initial installation . . . 7 3.3.1. Scenario Topology . . . . . . . . . . . . . . . . . . 8 3.3.2. Expected Message Sequencing . . . . . . . . . . . . . 8 3.4. Scenario 04: 2-hop bidirectional route establishment - Forward Route and Reverse Route updating . . . . . . . . . 9 3.4.1. Scenario Topology . . . . . . . . . . . . . . . . . . 9 3.4.2. Expected Message Sequencing . . . . . . . . . . . . . 9 3.5. Scenario 05: 2-hop bidirectional route establishment - Link breakage handling . . . . . . . . . . . . . . . . . . 10 3.5.1. Scenario Topology . . . . . . . . . . . . . . . . . . 10 3.5.2. Expected Message Sequencing . . . . . . . . . . . . . 11 3.6. Scenario 06: 3-hop bidirectional route establishment - Clausen, et al. Expires June 14, 2013 [Page 2] Internet-Draft LOADng Interop Report December 2012 Forward Route and Reverse Route initial installation . . . 11 3.6.1. Scenario Topology . . . . . . . . . . . . . . . . . . 12 3.6.2. Expected Message Sequencing . . . . . . . . . . . . . 12 3.7. Scenario 07: 3-hop bidirectional route establishment - Forward Route and Reverse Route updating . . . . . . . . . 13 3.7.1. Scenario Topology . . . . . . . . . . . . . . . . . . 14 3.7.2. Expected Message Sequencing . . . . . . . . . . . . . 14 3.8. Scenario 08: 3-hop bidirectional route establishment - Link breakage handling . . . . . . . . . . . . . . . . . . 15 3.8.1. Scenario Topology . . . . . . . . . . . . . . . . . . 15 3.8.2. Expected Message Sequencing . . . . . . . . . . . . . 16 3.9. Scenario 09: 4-hop bidirectional route establishment - Forward Route and Reverse Route initial installation . . . 16 3.9.1. Scenario Topology . . . . . . . . . . . . . . . . . . 17 3.9.2. Expected Message Sequencing . . . . . . . . . . . . . 17 3.10. Scenario 10: 4-hop bidirectional route establishment - Link breakage handling . . . . . . . . . . . . . . . . . . 19 3.10.1. Scenario Topology . . . . . . . . . . . . . . . . . . 19 3.10.2. Expected Message Sequencing . . . . . . . . . . . . . 20 3.11. Scenario 11: Establishment of the best bidirectional route . . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.11.1. Scenario Topology . . . . . . . . . . . . . . . . . . 21 3.11.2. Expected Message Sequencing . . . . . . . . . . . . . 21 3.12. Scenario 12: Blacklisting . . . . . . . . . . . . . . . . 22 3.12.1. Scenario Topology . . . . . . . . . . . . . . . . . . 23 3.12.2. Expected Message Sequencing . . . . . . . . . . . . . 23 4. Interop 01: Yokohama, Japan, October 2011 . . . . . . . . . . 26 4.1. Version of LOADng Specification Tested . . . . . . . . . . 26 4.2. Place and Date of Interoperability Test . . . . . . . . . 27 4.3. Participating Implementations . . . . . . . . . . . . . . 27 4.4. Scenarios Tested . . . . . . . . . . . . . . . . . . . . . 27 4.5. Additional Interoperability Test Considerations . . . . . 27 4.6. Results For Scenario 01 . . . . . . . . . . . . . . . . . 28 4.7. Results For Scenario 02 . . . . . . . . . . . . . . . . . 28 4.8. Results For Scenario 03 . . . . . . . . . . . . . . . . . 28 4.9. Results For Scenario 04 . . . . . . . . . . . . . . . . . 29 4.10. Results For Scenario 05 . . . . . . . . . . . . . . . . . 29 4.11. Results For Scenario 06 . . . . . . . . . . . . . . . . . 30 4.12. Results For Scenario 07 . . . . . . . . . . . . . . . . . 30 4.13. Results For Scenario 08 . . . . . . . . . . . . . . . . . 30 4.14. Results For Scenario 09 . . . . . . . . . . . . . . . . . 31 4.15. Results For Scenario 10 . . . . . . . . . . . . . . . . . 31 4.16. Results For Scenario 11 . . . . . . . . . . . . . . . . . 31 4.17. Results For Scenario 12 . . . . . . . . . . . . . . . . . 32 4.18. Conclusions . . . . . . . . . . . . . . . . . . . . . . . 32 5. Interop 02: San Jose, USA March 2012 . . . . . . . . . . . . . 34 5.1. LOADng version tested . . . . . . . . . . . . . . . . . . 34 5.2. Place and Date of Interoperability Test . . . . . . . . . 34 Clausen, et al. Expires June 14, 2013 [Page 3] Internet-Draft LOADng Interop Report December 2012 5.3. Participating Implementations . . . . . . . . . . . . . . 34 5.4. Interoperability Test Considerations . . . . . . . . . . . 35 5.5. Results For Scenario 01 . . . . . . . . . . . . . . . . . 35 5.6. Results For Scenrio 03 . . . . . . . . . . . . . . . . . . 35 5.7. Results For Scenario 05 . . . . . . . . . . . . . . . . . 35 6. Interop 03: Los Angeles, USA, June 2012 . . . . . . . . . . . 36 6.1. Version of LOADng Specification Tested . . . . . . . . . . 36 6.2. Place and Date of Interoperability Test . . . . . . . . . 36 6.3. Participating Implementations . . . . . . . . . . . . . . 36 6.4. Scenarios Tested . . . . . . . . . . . . . . . . . . . . . 36 6.5. Additional Interoperability Test Considerations . . . . . 36 6.6. Results For Scenario 01-02 . . . . . . . . . . . . . . . . 37 6.7. Results For Scenario 03-04-05 . . . . . . . . . . . . . . 37 6.8. Results For Scenario 06-07-08 . . . . . . . . . . . . . . 38 6.9. Results For Scenario 09-10 . . . . . . . . . . . . . . . . 39 6.10. Results For Scenario 11 . . . . . . . . . . . . . . . . . 39 6.11. Conclusions . . . . . . . . . . . . . . . . . . . . . . . 39 7. Interop 04: Vancouver, Canada, August, 2011 . . . . . . . . . 39 7.1. Version of LOADng Specifiation Tested . . . . . . . . . . 39 7.2. Place and Date of Interoperability Test . . . . . . . . . 40 7.3. Participating Implementations . . . . . . . . . . . . . . 40 7.4. Scenarios Tested . . . . . . . . . . . . . . . . . . . . . 40 7.5. Additional Interoperability Test Considerations . . . . . 40 7.6. Results for Scenario 01-02 . . . . . . . . . . . . . . . . 40 7.7. Results for Scenario 03-04-05 . . . . . . . . . . . . . . 41 7.8. Conclusions . . . . . . . . . . . . . . . . . . . . . . . 43 8. Security Considerations . . . . . . . . . . . . . . . . . . . 43 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 43 10. Contributors . . . . . . . . . . . . . . . . . . . . . . . . . 43 11. Informative References . . . . . . . . . . . . . . . . . . . . 44 Clausen, et al. Expires June 14, 2013 [Page 4] Internet-Draft LOADng Interop Report December 2012 1. Introduction This document reports experience with the LOADng [LOADng] routing protocol, as obtained by way of a number of interoperability tests during the protocol development. Interoperability tests between LOADng Routers implemented on the basis of the different versions of the protocol have been undertaken mainly to: o Show evidence that interoperable LOADng implementations do exist. o Clarify and improve the overall quality of the LOADng specification. o Demonstrate that the final LOADng internet draft can be considered as a standalone specification allowing the development of interoperable implementations of LOADng. 2. Terminology This document uses the terminology of [LOADng]. 3. Interoperability Scenarios This section describes the various tests and scenarios carried out between the implementations involved in the various interoperability tests. The testbed required is composed of up to five LOADng Routers, connected according to the specific topology described for each test scenario below. The LOADng routing protocol was run over UDP and IPv4. Either Ethernet or 802.11 wireless network was used in the test. 3.1. Scenario 01: 1-hop Bidirectional Route Establishment - Forward Route and Reverse Route initial installation For each implementation, this test aims to verify the initial installation of a bidirectional route (Forward Route and Reverse Route from A to B) within the LOADng Router routing tables (Routing Sets) through the effective generation and processing of LOADng control messages (RREQ, RREP, RREP-ACK). 3.1.1. Scenario Topology The testbed is composed of two LOADng Routers: Clausen, et al. Expires June 14, 2013 [Page 5] Internet-Draft LOADng Interop Report December 2012 +-------+ +-------+ | A |________| B | | | | | +-------+ +-------+ This test suite consists in establishing a bidirectional route between LOADng Router A and LOADng Router B. 3.1.2. Expected Message Sequencing The expected message sequencing is as follows: o LOADng Router A generates an RREQ message intended for LOADng Router B. o Upon receiving the RREQ, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router B to LOADng Router A) and sends an unicast RREP message intended for LOADng Router A, soliciting an RREP-ACK message. o Upon receiving the RREP, LOADng Router A installs a new tuple in its Routing Set towards LOADng Router B (Forward Route from LOADng Router A to LOADng Router B) and sends an unicast RREP-ACK message to LOADng Router B. A B | RREQ | --------------------> | RREP | <-------------------- | RREP-ACK | --------------------> | | 3.2. Scenario 02: 1-hop Bidirectional Route Establishment -Forward Route and Reverse Route updating For each implementation, this test aims to verify the refreshment of a bidirectional route (Forward Route and Reverse Route from A to B) already installed within the LOADng Router routing tables (Routing Sets) through the effective generation and processing of LOADng control messages (RREQ, RREP, RREP-ACK). 3.2.1. Scenario Topology The testbed is composed of two LOADng Routers: Clausen, et al. Expires June 14, 2013 [Page 6] Internet-Draft LOADng Interop Report December 2012 +-------+ +-------+ | A |________| B | | | | | +-------+ +-------+ This test suite consists in updating a bidirectional route between LOADng Router A and LOADng Router B. 3.2.2. Expected Message Sequencing The expected message sequencing is as follows: o LOADng Router A generates an RREQ message intended for LOADng Router B. o Upon receiving the RREQ, LOADng Router B updates the corresponding route (Reverse Route from LOADng Router B to LOADng Router A) already installed within its Routing Set and sends an unicast RREP message intended for LOADng Router A, soliciting an RREP-ACK message. o Upon receiving the RREP, LOADng Router A updates the corresponding route (Forward Route from LOADng Router A to LOADng Router B) already installed within its Routing Set and sends an unicast RREP-ACK message to LOADng Router B. A B | RREQ | --------------------> | RREP | <-------------------- | RREP-ACK | --------------------> | | 3.3. Scenario 03: 2-hop bidirectional route establishment - Forward Route and Reverse Route initial installation This test aims to verify the initial installation of a bidirectional route (Forward Route and Reverse Route from A to C) within the LOADng Router routing tables (Routing Sets) through the effective forwarding of LOADng control traffic by LOADng Router B which is located between LOADng Router A and LOADng Router C. It is also verified that RREP- ACK messages are not forwarded by the LOADng Routers these messages are intended for. Clausen, et al. Expires June 14, 2013 [Page 7] Internet-Draft LOADng Interop Report December 2012 3.3.1. Scenario Topology The testbed is composed of three LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router C or LOADng Router C towards LOADng Router A has to be forwarded by LOADng Router B: +-------+ +-------+ +-------+ | A |________| B |________| C | | | | | | | +-------+ +-------+ +-------+ This test suite consists in establishing a bidirectional route between LOADng Router A and LOADng Router C. 3.3.2. Expected Message Sequencing The expected message sequencing is as follows: o LOADng Router A generates an RREQ message intended for LOADng Router C. o Upon receiving the RREQ, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router B to LOADng Router A) and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router C installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router C to LOADng Router A) and a new tuple towards LOADng Router B (Reverse route from LOADng Router C to LOADng Router B). The reception of the RREQ also triggers the generation of an unicast RREP message intended for LOADng Router A, soliciting an RREP-ACK message. o Upon receiving the RREP, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router C (Forward Route from LOADng Router B to LOADng Router C), sends an unicast RREP-ACK message to LOADng Router C and forwards the RREP received previously. o Upon receiving the RREP, LOADng Router A installs a new tuple in its Routing Set towards LOADng Router B (Forward Route from LOADng Router A to LOADng Router B) and a new tuple towards LOADng Router C (Forward Route from LOADng Router A to LOADng Router C). The reception of the RREP also triggers an unicast RREP-ACK message intended for LOADng Router B. Clausen, et al. Expires June 14, 2013 [Page 8] Internet-Draft LOADng Interop Report December 2012 A B C | RREQ | | --------------------> | | | RREQ | | --------------------> | | RREP | | <-------------------- | | RREP-ACK | | --------------------> | RREP | | <-------------------- | | RREP-ACK | | --------------------> | | | | 3.4. Scenario 04: 2-hop bidirectional route establishment - Forward Route and Reverse Route updating This test aims to verify the refreshment of a bidirectional route (Forward Route and Reverse Route from A to C) already installed within the LOADng Router routing tables (Routing Sets) through the effective forwarding of LOADng control traffic by LOADng Router B which is located between LOADng Router A and LOADng Router C. 3.4.1. Scenario Topology The testbed is composed of three LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router C or LOADng Router C towards LOADng Router A has to be forwarded by LOADng Router B: +-------+ +-------+ +-------+ | A |________| B |________| C | | | | | | | +-------+ +-------+ +-------+ This test suite consists in updating a bidirectional route between LOADng Router A and LOADng Router C. 3.4.2. Expected Message Sequencing The expected message sequencing is as follows: o LOADng Router A generates an RREQ message intended for LOADng Router C. o Upon receiving the RREQ, LOADng Router B updates the corresponding route (Reverse Route from LOADng Router B to LOADng Router A) Clausen, et al. Expires June 14, 2013 [Page 9] Internet-Draft LOADng Interop Report December 2012 already installed within its Routing Set and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router C updates the corresponding routes (Reverse Routes from LOADng Router C to LOADng Router A and from LOADng Router C to LOADng Router B). The reception of the RREQ also triggers the generation of an unicast RREP message intended for LOADng Router A, soliciting an RREP-ACK message. o Upon receiving the RREP, LOADng Router B updates the corresponding route (Forward route from LOADng Router B to LOADng Router C), sends an unicast RREP-ACK message to LOADng Router C and forwards the RREP received previously. o Upon receiving the RREP, LOADng Router A updates the corresponding routes (Forward routes from LOADng Router A to LOADng Router B and from LOADng Router A to LOADng Router C). The reception of the RREP also triggers an unicast RREP-ACK message intended for LOADng Router B. A B C | RREQ | | --------------------> | | | RREQ | | --------------------> | | RREP | | <-------------------- | | RREP-ACK | | --------------------> | RREP | | <-------------------- | | RREP-ACK | | --------------------> | | | | 3.5. Scenario 05: 2-hop bidirectional route establishment - Link breakage handling This test aims to verify the proper generation and processing of an RERR message after an artificially created link breakage on an previously established bidirectional route. 3.5.1. Scenario Topology The testbed is composed of three LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router C or LOADng Clausen, et al. Expires June 14, 2013 [Page 10] Internet-Draft LOADng Interop Report December 2012 Router C towards LOADng Router A has to be forwarded by LOADng Router B: +-------+ +-------+ +-------+ | A |________| B |________| C | | | | | | | +-------+ +-------+ +-------+ This test suite consists in handling link breakages between routers. 3.5.2. Expected Message Sequencing The expected message sequencing is as follows: o A bidirectional route is already established between LOADng Routers A and C. o At some time, link breakage is detected by LOADng Router B. Consequently, an unicast RERR message intended for LOADng Router A (here the assumption is made that the unsuccessful delivered data traffic would have been generated by LOADng Router A) is transmitted. Note: link breakage is provoked artificially and its detection by LOADng Router B is triggered manually (normally, this would be triggered by failure in sending data traffic intended for LOADng Router C). o Upon receiving the RERR, LOADng Router A updates its Routing Set by invalidating the existing Forward Route from LOADng Router A to LOADng Router C. A B C | | | | | B-C link breakage | | | X | RERR | X <-------------------- X | | X 3.6. Scenario 06: 3-hop bidirectional route establishment - Forward Route and Reverse Route initial installation This test aims to verify the initial installation of a bidirectional route (Forward Route and Reverse Route from A to D) within the LOADng Router routing tables (Routing Sets) through the effective forwarding Clausen, et al. Expires June 14, 2013 [Page 11] Internet-Draft LOADng Interop Report December 2012 of LOADng control traffic by LOADng Routers B and C, which are located between LOADng Router A and LOADng Router D. It is also verified that RREP-ACK messages are not forwarded by the LOADng Routers these messages are intended for. 3.6.1. Scenario Topology The testbed is composed of four LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router D or LOADng Router D towards LOADng Router A has to be forwarded by LOADng Routers B and C: +-------+ +-------+ +-------+ +-------+ | A |________| B |________| C |________| D | | | | | | | | | +-------+ +-------+ +-------+ +-------+ This test suite consists in establishing a bidirectional route between LOADng Router A and LOADng Router D. 3.6.2. Expected Message Sequencing The expected message sequencing is as follows: o LOADng Router A generates an RREQ message intended for LOADng Router D. o Upon receiving the RREQ, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router B to LOADng Router A) and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router C installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router C to LOADng Router A) and a new tuple towards LOADng Router B (Reverse route from LOADng Router C to LOADng Router B) and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router D installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router D to LOADng Router A) and a new tuple towards LOADng Router C (Reverse route from LOADng Router D to LOADng Router C). The reception of the RREQ also triggers the generation of an unicast RREP message intended for LOADng Router A, soliciting an RREP-ACK message. o Upon receiving the RREP, LOADng Router C installs a new tuple in its Routing Set towards LOADng Router D (Forward Route from LOADng Router C to LOADng Router D), sends an unicast RREP-ACK message to Clausen, et al. Expires June 14, 2013 [Page 12] Internet-Draft LOADng Interop Report December 2012 LOADng Router D and forwards the RREP received previously. o Upon receiving the RREP, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router D (Forward Route from LOADng Router B to LOADng Router D) and a new tuple towards LOADng Router C (Forward Route from LOADng Router B to LOADng Router C). An unicast RREP-ACK message is also sent to LOADng Router C and the RREP received previously is forwarded. o Upon receiving the RREP, LOADng Router A installs a new tuple in its Routing Set towards LOADng Router B (Forward Route from LOADng Router A to LOADng Router B) and a new tuple towards LOADng Router D (Forward Route from LOADng Router A to LOADng Router D). The reception of the RREP also triggers an unicast RREP-ACK message intended for LOADng Router B. A B C D | RREQ | | | --------------------> | | | | RREQ | | | --------------------> | | | | RREQ | | | --------------------> | | | RREP | | | <-------------------- | | | RREP-ACK | | | --------------------> | | RREP | | | <-------------------- | | | RREP-ACK | | | --------------------> | | RREP | | | <-------------------- | | | RREP-ACK | | | --------------------> | | | | | | 3.7. Scenario 07: 3-hop bidirectional route establishment - Forward Route and Reverse Route updating This test aims to verify the refreshment of a bidirectional route (Forward Route and Reverse Route from A to D) already installed within the LOADng Router routing tables (Routing Sets) through the effective forwarding of LOADng control traffic by LOADng Routers B and C which are located between LOADng Router A and LOADng Router D. Clausen, et al. Expires June 14, 2013 [Page 13] Internet-Draft LOADng Interop Report December 2012 3.7.1. Scenario Topology The testbed is composed of four LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router D or LOADng Router D towards LOADng Router A has to be forwarded by LOADng Routers B and C: +-------+ +-------+ +-------+ +-------+ | A |________| B |________| C |________| D | | | | | | | | | +-------+ +-------+ +-------+ +-------+ This test suite consists in updating a bidirectional route between LOADng Router A and LOADng Router D. 3.7.2. Expected Message Sequencing The expected message sequencing is as follows: o LOADng Router A generates an RREQ message intended for LOADng Router D. o Upon receiving the RREQ, LOADng Router B updates the corresponding route (Reverse Route from LOADng Router B to LOADng Router A) already installed within its Routing Set and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router C updates the corresponding routes (Reverse Routes from LOADng Router C to LOADng Router A and from LOADng Router C to LOADng Router B) already installed within its Routing Set and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router D updates the corresponding routes (Reverse Routes from LOADng Router D to LOADng Router A and from LOADng Router D to LOADng Router C) already installed within its Routing Set. The reception of the RREQ also triggers the generation of an unicast RREP message intended for LOADng Router A, soliciting an RREP-ACK message. o Upon receiving the RREP, LOADng Router C updates the corresponding route (Forward Route from LOADng Router C to LOADng Router D), sends an unicast RREP-ACK message to LOADng Router D and forwards the RREP received previously. o Upon receiving the RREP, LOADng Router B updates the corresponding routes (Forward Route from LOADng Router B to LOADng Router D and from LOADng Router B to LOADng Router C). An unicast RREP-ACK message is also sent to LOADng Router C and the RREP received Clausen, et al. Expires June 14, 2013 [Page 14] Internet-Draft LOADng Interop Report December 2012 previously is forwarded. o Upon receiving the RREP, LOADng Router A updates the corresponding routes (Forward Route from LOADng Router A to LOADng Router B and from LOADng Router A to LOADng Router D). The reception of the RREP also triggers an unicast RREP-ACK message intended for LOADng Router B. A B C D | RREQ | | | --------------------> | | | | RREQ | | | --------------------> | | | | RREQ | | | --------------------> | | | RREP | | | <-------------------- | | | RREP-ACK | | | --------------------> | | RREP | | | <-------------------- | | | RREP-ACK | | | --------------------> | | RREP | | | <-------------------- | | | RREP-ACK | | | --------------------> | | | | | | 3.8. Scenario 08: 3-hop bidirectional route establishment - Link breakage handling This test aims to verify the proper generation, processing and forwarding of a RERR message after an artificially created link breakage on an previously established bidirectional route. 3.8.1. Scenario Topology The testbed is composed of four LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router D or LOADng Router D towards LOADng Router A has to be forwarded by LOADng Routers B and C: +-------+ +-------+ +-------+ +-------+ | A |________| B |________| C |________| D | | | | | | | | | +-------+ +-------+ +-------+ +-------+ Clausen, et al. Expires June 14, 2013 [Page 15] Internet-Draft LOADng Interop Report December 2012 This test suite consists in handling link breakages between LOADng Routers. 3.8.2. Expected Message Sequencing The expected message sequencing is as follows: o A bidirectional route is already established between LOADng Routers A and D. o At some time, link breakage is detected by LOADng Router C. Consequently, an unicast RERR message intended for LOADng Router A (here the assumption is made that the unsuccessful delivered data traffic would have been generated by LOADng Router A) is transmitted to LOADng Router B according to the Reverse Route from LOADng Router C to LOADng Router A computed previously. Note: link breakage is provoked artificially and its detection by LOADng Router C is triggered manually (normally, this would be triggered by failure in sending data traffic intended for LOADng Router D). o Upon receiving the RERR, LOADng Router B updates its Routing Set by invalidating the existing Forward Route from LOADng Router B to LOADng Router D. Afterwards, the RERR message is forwarded according to the existing Reverse Route from LOADng Router B to LOADng Router A. o Upon receiving the RERR, LOADng Router A updates its Routing Set by invalidating the existing Forward Route from LOADng Router A to LOADng Router D. A B C D | | | | | | | C-D link breakage X | | | X | | RERR | X | <-------------------- X | RERR | | X <-------------------- | X | | | X 3.9. Scenario 09: 4-hop bidirectional route establishment - Forward Route and Reverse Route initial installation This test aims to verify the initial installation of a bidirectional route (Forward Route and Reverse Route from A to E) within the LOADng Clausen, et al. Expires June 14, 2013 [Page 16] Internet-Draft LOADng Interop Report December 2012 Router routing tables (Routing Sets) through the effective forwarding of LOADng control traffic by LOADng Routers B, C and D, which are located between LOADng Router A and LOADng Router E. It is also verified that RREP-ACK messages are not forwarded by the LOADng Routers these messages are intended for. 3.9.1. Scenario Topology The testbed is composed of five LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router E or LOADng Router E towards LOADng Router A has to be forwarded by LOADng Routers B, C and D: +-------+ +-------+ +-------+ +-------+ +-------+ | A |______| B |______| C |______| D |______| E | | | | | | | | | | | +-------+ +-------+ +-------+ +-------+ +-------+ This test suite consists in establishing a bidirectional route between LOADng Router A and LOADng Router E. 3.9.2. Expected Message Sequencing The expected message sequencing is as follows: o LOADng Router A generates an RREQ message intended for LOADng Router E. o Upon receiving the RREQ, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router B to LOADng Router A) and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router C installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router C to LOADng Router A) and a new tuple towards LOADng Router B (Reverse route from LOADng Router C to LOADng Router B) and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router D installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router D to LOADng Router A) and a new tuple towards LOADng Router C (Reverse route from LOADng Router D to LOADng Router C) and forwards the received RREQ. o Upon receiving the RREQ, LOADng Router E installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router E to LOADng Router A) and a new tuple towards LOADng Router D (Reverse route from LOADng Router E to LOADng Router D). The Clausen, et al. Expires June 14, 2013 [Page 17] Internet-Draft LOADng Interop Report December 2012 reception of the RREQ also triggers the generation of an unicast RREP message intended for LOADng Router A, soliciting an RREP-ACK message. o Upon receiving the RREP, LOADng Router D installs a new tuple in its Routing Set towards LOADng Router E (Forward Route from LOADng Router D to LOADng Router E), sends an unicast RREP-ACK message to LOADng Router E and forwards the RREP received previously. o Upon receiving the RREP, LOADng Router C installs a new tuple in its Routing Set towards LOADng Router E (Forward Route from LOADng Router C to LOADng Router E) and a new tuple towards LOADng Router D (Forward Route from LOADng Router C to LOADng Router D). An unicast RREP-ACK message is also sent to LOADng Router D and the RREP received previously is forwarded. o Upon receiving the RREP, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router E (Forward Route from LOADng Router B to LOADng Router E) and a new tuple towards LOADng Router C (Forward Route from LOADng Router B to LOADng Router C). An unicast RREP-ACK message is also sent to LOADng Router C and the RREP received previously is forwarded. o Upon receiving the RREP, LOADng Router A installs a new tuple in its Routing Set towards LOADng Router B (Forward Route from LOADng Router A to LOADng Router B) and a new tuple towards LOADng Router E (Forward Route from LOADng Router A to LOADng Router E). The reception of the RREP also triggers an unicast RREP-ACK message intended for LOADng Router B. Clausen, et al. Expires June 14, 2013 [Page 18] Internet-Draft LOADng Interop Report December 2012 A B C D E | RREQ | | | | ---------------> | | | | | RREQ | | | | ---------------> | | | | | RREQ | | | | ---------------> | | | | | RREQ | | | | ---------------> | | | | RREP | | | | <--------------- | | | | RREP-ACK | | | | ---------------> | | | RREP | | | | <--------------- | | | | RREP-ACK | | | | ---------------> | | | RREP | | | | <--------------- | | | | RREP-ACK | | | | ---------------> | | | RREP | | | | <--------------- | | | | RREP-ACK | | | | ---------------> | | | | | | | | 3.10. Scenario 10: 4-hop bidirectional route establishment - Link breakage handling This test aims to verify the proper generation, processing and forwarding of a RERR message after an artificially created link breakage on an previously established bidirectional route. 3.10.1. Scenario Topology The testbed is composed of five LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router E or LOADng Router E towards LOADng Router A has to be forwarded by LOADng Routers B, C and D: +-------+ +-------+ +-------+ +-------+ +-------+ | A |______| B |______| C |______| D |______| E | | | | | | | | | | | +-------+ +-------+ +-------+ +-------+ +-------+ This test suite consists in handling link breakages between routers. Clausen, et al. Expires June 14, 2013 [Page 19] Internet-Draft LOADng Interop Report December 2012 3.10.2. Expected Message Sequencing The expected message sequencing is as follows: o A bidirectional route is already established between LOADng Routers A and E. o At some time, a link breakage to E is detected by LOADng Router D. Consequently, an unicast RERR message intended for LOADng Router A (here the assumption is made that the unsuccessful delivered data traffic would have been generated by LOADng Router A) is transmitted to LOADng Router C according to the Reverse Route from LOADng Router C to LOADng Router A computed previously. Note: link breakage is provoked artificially and its detection by LOADng Router D is triggered manually (normally, this would be triggered by failure in sending data traffic intended for LOADng Router E). o Upon receiving the RERR, LOADng Router C updates its Routing Set by invalidating the existing Forward Route from LOADng Router C to LOADng Router E. Afterwards, the RERR message is forwarded according to the existing Reverse Route from LOADng Router C to LOADng Router A. o Upon receiving the RERR, LOADng Router B updates its Routing Set by invalidating the existing Forward Route from LOADng Router B to LOADng Router E. Afterwards, the RERR message is forwarded according to the existing Reverse Route from LOADng Router B to LOADng Router A. o Upon receiving the RERR, LOADng Router A updates its Routing Set by invalidating the existing Forward Route from LOADng Router A to LOADng Router E. A B C D E | | | | | | | | D-E link breakage | | | | X | | | RERR | X | | <--------------- X | | RERR | | X | <--------------- | X | RERR | | | X <--------------- | | X | | | | X Clausen, et al. Expires June 14, 2013 [Page 20] Internet-Draft LOADng Interop Report December 2012 3.11. Scenario 11: Establishment of the best bidirectional route This test aims to verify the processing of multiple RREQs when installing a bidirectional route (Forward Route and Reverse Route from A to C) within the LOADng Router routing tables (Routing Sets). 3.11.1. Scenario Topology The testbed is composed of three LOADng Routers. Control traffic generated by either LOADng Router A towards LOADng Router C or LOADng Router C towards LOADng Router A can be forwarded by LOADng Router B or transmitted via the direct link between LOADng Routers A and C: +-------+ +-------+ +-------+ | A |________| B |________| C | | | | | | | +-------+ +-------+ +-------+ |_________________________________| This test consists in establishing a bidirectional route between LOADng Router A and LOADng Router C. Hop count metric is used for measuring differet routes. 3.11.2. Expected Message Sequencing The expected message sequencing is as follows: o LOADng Router A generates an RREQ message intended for LOADng Router C. According to RREQ transmission rules, the generated RREQ message is transmitted to all neighbor LOADng Routers. o Upon receiving the RREQ, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router B to LOADng Router A) and forwards the received RREQ. At the same time, upon receiving the same RREQ via its direct link with LOADng Router A, LOADng Router C installs a new tuple in its Routing Set (Reverse Route from LOADng Router C to LOADng Router A). The reception of the RREQ also triggers the generation of an unicast RREP message intended for LOADng Router A, requiring RREP- ACK message. o Upon receiving the same RREQ via LOADng Router B, LOADng Router C compares the RREQ.route-metric information carried by the RREQ with the already existing tuple within its Routing Set (Reverse Route from LOADng Router C to LOADng Router A) according to the comparison operator specified by the metric used (the "hop count" metric was used). Thus, the best route is chosen considering only Clausen, et al. Expires June 14, 2013 [Page 21] Internet-Draft LOADng Interop Report December 2012 the hop count: Already existing tuple: = 1 Tuple corresponding to the newly received RREQ: = 2 According to the comparison operator specified by the metric used: 1 < 2 Consequently, the newly received RREQ message is discarded without affecting the Routing Set or triggering the generation of any RREP message. o Upon receiving the RREP via its direct link with LOADng Router C, LOADng Router A installs a new tuple in its Routing Set (Forward Route from LOADng Router A to LOADng Router C). The reception of the RREP also triggers an unicast RREP-ACK message intended for LOADng Router C. A B C | RREQ | | --------------------> RREQ | ----------------------------------------> | | RREQ | | --------------------> | | RREP | <---------------------------------------- | | RREP-ACK | ----------------------------------------> | | | Note: the RREQ forwarded by LOADng Router B towards C is not necessarily received before LOADng Router C generates the RREP message intended for LOADng Router A. Indeed, the order in which those messages are transmitted is dependent on the transmission delays of each single link between LOADng Routers A, B and C. 3.12. Scenario 12: Blacklisting This test aims to verify the effectiveness of avoiding unidirectional links using blacklisting. Clausen, et al. Expires June 14, 2013 [Page 22] Internet-Draft LOADng Interop Report December 2012 3.12.1. Scenario Topology The testbed is composed of four LOADng Routers with a unidirectional link between LOADng Routers A and D (direct communication from D towards A is impossible). +-------+ +-------+ | A |_________| B | | | | | +-------+ +-------+ | | V | +-------+ +-------+ | D |_________| C | | | | | +-------+ +-------+ This test consists in establishing a bidirectional route between LOADng Router A and LOADng Router D. 3.12.2. Expected Message Sequencing First attempt to establish a bidirectional route between LOADng Routers A and D: o LOADng Router A generates an RREQ message (RREQ.seq-num = 0, RREQ.originator = A) intended for LOADng Router D. o Upon receiving the RREQ, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router B to LOADng Router A) and forwards the received RREQ. At the same time, upon receiving the same RREQ via its direct (unidirectional) link with LOADng Router A, LOADng Router D installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router D to LOADng Router A). The reception of the RREQ also triggers the generation of an unicast RREP message intended for LOADng Router A. The RREP.ackrequired the sent RREP message is set ('1'). o Upon receiving the RREQ, LOADng Router C installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router C to LOADng Router A) and a new tuple towards LOADng Router B (Reverse route from LOADng Router C to LOADng Router B) and forwards the received RREQ. o Upon receiving the same RREQ (RREQ.seq-num = 0, RREQ.originator = A) again via LOADng Router C, LOADng Router D compares the Clausen, et al. Expires June 14, 2013 [Page 23] Internet-Draft LOADng Interop Report December 2012 RREQ.route-metric information carried by the RREQ with the already existing tuple within its Routing Set (Reverse Route from LOADng Router D to LOADng Router A) according to the comparison operator specified by the metric used (hop count): Already existing tuple: = 1 Tuple corresponding to the newly received RREQ: = 2 According to the comparison operator specified by the metric used: 1 < 2 Consequently, the newly received RREQ message is discarded without affecting the Routing Set or triggering the generation of any RREP message. o Due to the unidirectional nature of the existing link between LOADng Routers A and D, the RREP message previously sent by LOADng Router D intended for LOADng Router A did not reach its destination. After an elapsed time equaling RREP_ACK_TIMEOUT, LOADng Router D is not expecting an RREP-ACK message anymore. This results in recording LOADng Router A neighbor in LOADng Router D's Blacklist. Second attempt to establish a bidirectional route between LOADng Routers A and D: o LOADng Router A generates an RREQ message (RREQ.seq-num = 1, RREQ.originator = A) intended for LOADng Router D. o Upon receiving the RREQ, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router A (Reverse Route from LOADng Router B to LOADng Router A) and forwards the received RREQ. At the same time, upon receiving the same RREQ via its blacklisted neighbor LOADng Router A, LOADng Router D discards the message. o Upon receiving the RREQ, LOADng Router C updates the corresponding routes (Reverse Routes from LOADng Router C to LOADng Router A and from LOADng Router C to LOADng Router B) and forwards the received RREQ. Clausen, et al. Expires June 14, 2013 [Page 24] Internet-Draft LOADng Interop Report December 2012 o Upon receiving the RREQ, LOADng Router D updates the already installed route (Reverse Route from LOADng Router C to LOADng Router A) and installs a new tuple towards LOADng Router C (Reverse route from LOADng Router D to LOADng Router C). The reception of the RREQ also triggers the generation of an unicast RREP message intended for LOADng Router A. The RREP.ackrequired of the sent RREP message is set ('1'). o Upon receiving the RREP, LOADng Router C installs a new tuple in its Routing Set towards LOADng Router D (Forward Route from LOADng Router C to LOADng Router D), sends an unicast RREP-ACK message to LOADng Router D and forwards the RREP received previously. o Upon receiving the RREP, LOADng Router B installs a new tuple in its Routing Set towards LOADng Router D (Forward Route from LOADng Router B to LOADng Router D) and a new tuple towards LOADng Router C (Forward Route from LOADng Router B to LOADng Router C). An unicast RREP-ACK message is also sent to LOADng Router C and the RREP received previously is forwarded. o Upon receiving the RREP, LOADng Router A installs a new tuple in its Routing Set towards LOADng Router D (Forward Route from LOADng Router A to LOADng Router D) and a new tuple towards LOADng Router B (Forward Route from LOADng Router A to LOADng Router B). The reception of the RREP also triggers an unicast RREP-ACK message intended for LOADng Router B. Clausen, et al. Expires June 14, 2013 [Page 25] Internet-Draft LOADng Interop Report December 2012 A B C D | | | | First attempt ///////////////////////////////////////// | RREQ | | | ------------------> RREQ | | ------------------------------------------------------> | | RREP | | |XXXXX <----------------------------------------------- | | RREQ | | | ------------------> | | | | RREQ | | | ----------------->X RREQ | | | | Discarded Second attempt //////////////////////////////////////// | RREQ | | | ------------------> RREQ | | ----------------------------------------------------->X RREQ | | RREQ | | Discarded | ------------------> | | | | RREQ | | | ------------------> | | | RREP | | | <------------------ | | | RREP-ACK | | | ------------------> | | RREP | | | <------------------ | | | RREP-ACK | | | ------------------> | | RREP | | | <------------------ | | | RREP-ACK | | | ------------------> | | 4. Interop 01: Yokohama, Japan, October 2011 4.1. Version of LOADng Specification Tested The interoperability tests were conducted according to the specification in [LOADng-00]. NOTE: Due to the evolution of [LOADng] and this document, ome of the conventions used in Section 3, such as routing metric and some fields of messages, may be different from the description in [LOADng-00]. Clausen, et al. Expires June 14, 2013 [Page 26] Internet-Draft LOADng Interop Report December 2012 4.2. Place and Date of Interoperability Test This section reports experience with the LOADng routing protocol, resulting from interoperability testing performed at Hitachi YRL in Yokohama, Japan, from october 17th to october 19th 2011. 4.3. Participating Implementations The following implementations were used to perform the interoperability tests this section, listed alphabetically: Ecole Polytechnique: "LIX" - This implementation was jointly developed by Axel Colin de Verdiere, Jiazi Yi, Ulrich Herberg and Thomas Clausen of Ecole Ploytechnique's networking team. It consists of approximately 6000 lines of JAVA code running in a Mac OS environment. It supports RREQ, RREP, RREP-ACK and RERR generation, processing, forwarding and transmission. Hitachi YRL 1: "Hitachi 1" - This implementation was fully developed by Yuichi Igarashi of Hitachi YRL. It consists of 1589 lines of C code running in the Hitachi proprietary micro OS environment embedded in a 16MHz H8 micro processor. It supports RREQ, RREP, RREP-ACK and RERR generation, processing, forwarding and transmission. Hitachi YRL 2: "Hitachi 2" - This implementation was jointly developed by Nobukatsu Inomata of Hitachi ULSI Systems and Yoko Morii of Hitachi YRL. It consists of 1987 lines of C++ code running in a Mac OS environment. It supports RREQ, RREP, RREP-ACK generation, processing, forwarding and transmission, and RERR processing. 4.4. Scenarios Tested This interoperability test includes all scenarios 01-12 (inclusive). 4.5. Additional Interoperability Test Considerations Wireshark packet sniffers, modified to interpret LOADng control traffic, were used to monitor each link, so as to verify propper message sequencing. For each test, the initiation of the communication resulting in the generation of the first LOADng control traffic message is always triggered manually. In addition, RREP-ACK LOADng control messages were systematically expected from each LOADng Router upon reception of a RREP LOADng control message in order to allow the detection of unidirectional links. Clausen, et al. Expires June 14, 2013 [Page 27] Internet-Draft LOADng Interop Report December 2012 4.6. Results For Scenario 01 The following table is summarizing the results obtained for the different combinations for which a 1-hop Forward Route and Reverse Route initial installation test was performed: +-----------+------+-----------+-----------+ | | LIX | Hitachi 1 | Hitachi 2 | +-----------+------+-----------+-----------+ | LIX | N/R | Pass | Pass | | Hitachi 1 | Pass | N/R | Pass | | Hitachi 2 | Pass | Pass | N/R | +-----------+------+-----------+-----------+ Table 1 4.7. Results For Scenario 02 The following table is summarizing the results obtained for the different combinations for which a 1-hop Forward Route and Reverse Route updating test was performed: +-----------+------+-----------+-----------+ | | LIX | Hitachi 1 | Hitachi 2 | +-----------+------+-----------+-----------+ | LIX | N/R | Pass | Pass | | Hitachi 1 | Pass | N/R | Pass | | Hitachi 2 | Pass | Pass | N/R | +-----------+------+-----------+-----------+ Table 2 4.8. Results For Scenario 03 The following table is summarizing the results obtained for the different combinations for which a 2-hop Forward Route and Reverse Route initial installation test was performed: Clausen, et al. Expires June 14, 2013 [Page 28] Internet-Draft LOADng Interop Report December 2012 +-----------+-----------+-----------+--------+ | A | B | C | Result | +-----------+-----------+-----------+--------+ | Hitachi 1 | LIX | Hitachi 2 | Pass | | Hitachi 2 | LIX | Hitachi 1 | Pass | | LIX | Hitachi 1 | Hitachi 2 | Pass | | Hitachi 2 | Hitachi 1 | LIX | Pass | | LIX | Hitachi 2 | Hitachi 1 | Pass | | Hitachi 1 | Hitachi 2 | LIX | Pass | +-----------+-----------+-----------+--------+ Table 3 4.9. Results For Scenario 04 The following table is summarizing the results obtained for the different combinations for which a 2-hop Forward Route and Reverse Route updating test was performed: +-----------+-----------+-----------+--------+ | A | B | C | Result | +-----------+-----------+-----------+--------+ | Hitachi 1 | LIX | Hitachi 2 | Pass | | Hitachi 2 | LIX | Hitachi 1 | Pass | | LIX | Hitachi 1 | Hitachi 2 | Pass | | Hitachi 2 | Hitachi 1 | LIX | Pass | | LIX | Hitachi 2 | Hitachi 1 | Pass | | Hitachi 1 | Hitachi 2 | LIX | Pass | +-----------+-----------+-----------+--------+ Table 4 4.10. Results For Scenario 05 The following table is summarizing the results obtained for the different combinations for which a Link breakage handling test was performed: +-----------+-----------+-----+--------+ | A | B | C | Result | +-----------+-----------+-----+--------+ | Hitachi 1 | LIX | LIX | Pass | | LIX | Hitachi 1 | LIX | Pass | +-----------+-----------+-----+--------+ Table 5 Clausen, et al. Expires June 14, 2013 [Page 29] Internet-Draft LOADng Interop Report December 2012 4.11. Results For Scenario 06 The following table is summarizing the results obtained for the different combinations for which a 3-hop Forward Route and Reverse Route initial installation test was performed: +-----------+-----------+-----------+-----------+--------+ | A | B | C | D | Result | +-----------+-----------+-----------+-----------+--------+ | Hitachi 1 | LIX | LIX | Hitachi 2 | Pass | | Hitachi 1 | LIX | Hitachi 2 | LIX | Pass | | LIX | Hitachi 2 | Hitachi 1 | LIX | Pass | +-----------+-----------+-----------+-----------+--------+ Table 6 4.12. Results For Scenario 07 The following table is summarizing the results obtained for the different combinations for which a 3-hop Forward Route and Reverse Route updating test was performed: +-----------+-----------+-----------+-----------+--------+ | A | B | C | D | Result | +-----------+-----------+-----------+-----------+--------+ | Hitachi 1 | LIX | LIX | Hitachi 2 | Pass | | Hitachi 1 | LIX | Hitachi 2 | LIX | Pass | | LIX | Hitachi 2 | Hitachi 1 | LIX | Pass | +-----------+-----------+-----------+-----------+--------+ Table 7 4.13. Results For Scenario 08 The following table is summarizing the results obtained for the different combinations for which a Link breakage handling test was performed: +-----------+-----------+-----+-----------+--------+ | A | B | C | D | Result | +-----------+-----------+-----+-----------+--------+ | Hitachi 1 | LIX | LIX | Hitachi 2 | Pass | | LIX | Hitachi 1 | LIX | Hitachi 2 | Pass | +-----------+-----------+-----+-----------+--------+ Table 8 Clausen, et al. Expires June 14, 2013 [Page 30] Internet-Draft LOADng Interop Report December 2012 4.14. Results For Scenario 09 The following table is summarizing the results obtained for the different combinations for which a 4-hop Forward Route and Reverse Route initial installation test was performed: +-----------+-----------+-----+-----------+-----+--------+ | A | B | C | D | E | Result | +-----------+-----------+-----+-----------+-----+--------+ | Hitachi 2 | Hitachi 1 | LIX | Hitachi 1 | LIX | Pass | +-----------+-----------+-----+-----------+-----+--------+ Table 9 4.15. Results For Scenario 10 The following table is summarizing the results obtained for the different combinations for which a Link breakage handling test was performed: +-----------+-----------+-----+-----------+-----+--------+ | A | B | C | D | E | Result | +-----------+-----------+-----+-----------+-----+--------+ | Hitachi 2 | Hitachi 1 | LIX | Hitachi 1 | LIX | Pass | +-----------+-----------+-----+-----------+-----+--------+ Table 10 4.16. Results For Scenario 11 The following table is summarizing the results obtained for the different combinations for which a test consisting in the establishment of the best bidirectional route was performed: +-----------+-----------+-----------+--------+ | A | B | C | Result | +-----------+-----------+-----------+--------+ | LIX | Hitachi 1 | Hitachi 2 | Pass | | LIX | Hitachi 2 | Hitachi 1 | Pass | | Hitachi 2 | Hitachi 1 | LIX | Pass | | Hitachi 1 | LIX | Hitachi 2 | Pass | +-----------+-----------+-----------+--------+ Table 11 Clausen, et al. Expires June 14, 2013 [Page 31] Internet-Draft LOADng Interop Report December 2012 4.17. Results For Scenario 12 The following table is summarizing the results obtained for the different combinations for which a Blacklisting test was performed: +-----------+-----+-----------+-----------+--------+ | A | B | C | D | Result | +-----------+-----+-----------+-----------+--------+ | Hitachi 2 | LIX | Hitachi 1 | LIX | Pass | | LIX | LIX | Hitachi 1 | Hitachi 2 | Pass | | Hitachi 2 | LIX | LIX | Hitachi 1 | Pass | +-----------+-----+-----------+-----------+--------+ Table 12 4.18. Conclusions The different test scenarios carried that were carried out for different interoperable and independent implementations allowed to completely cover the [LOADng-00] specification by checking each technical feature one by one. In addition, the completion of this process permitted the improvement of the overall quality and accuracy of the [LOADng-00] specification. +------+----------------+-----------------------+-----------+ | | | |Test suites| |Chap. | Item | Technical feature +-----------+ | | | |A|B|C|D|E|F| +------+----------------+------------+----------+-+-+-+-+-+-+ |6.1 | | |Originator|X|X|X| |X|X| +------+ Information |Routing Set +----------+-+-+-+-+-+-+ |6.1 | Base | |Previous | |X|X|X| |X| +------+ +------------+----------+-+-+-+-+-+-+ |6.2 | |Blacklist Neighbor set | | | | | |X| +------+----------------+-----------------------+-+-+-+-+-+-+ |8.1 | |TLV |X|X|X|X|X|X| +------+ +-----------------------+-+-+-+-+-+-+ |8.2.1 | Packet |Route Request Message |X|X|X|X|X|X| +------+ Format +-----------------------+-+-+-+-+-+-+ |8.2.1 | |Route Reply Message |X|X|X|X|X|X| +------+ +-----------------------+-+-+-+-+-+-+ |8.2.2 | |Route Reply Ack Message|X|X|X|X|X|X| +------+ +-----------------------+-+-+-+-+-+-+ |8.2.3 | |Route Error Message | |X|X|X| | | +------+----------------+-----------------------+-+-+-+-+-+-+ |10.1 | Unidirectional |Blacklist | | | | | |X| | | link handling | | | | | | | | +------+----------------+-----------------------+-+-+-+-+-+-+ Clausen, et al. Expires June 14, 2013 [Page 32] Internet-Draft LOADng Interop Report December 2012 |11.1 | |Invalid RREQ, RREP |X|X|X|X|X|X| +------+ Common rules +-----------------------+-+-+-+-+-+-+ |11.2 | for RREQ, RREP |RREQ, RREP Processing |X|X|X|X|X|X| +------+ Message +-----------------------+-+-+-+-+-+-+ |11.3 | |Updating RREQ, RREP |X|X|X|X|X|X| +------+----------------+-----------------------+-+-+-+-+-+-+ |12.1 | |RREQ Generation |X|X|X|X|X|X| +------+ +-----------------------+-+-+-+-+-+-+ |12.2 | Route |RREQ Processing |X|X|X|X|X|X| +------+ Requests +-----------------------+-+-+-+-+-+-+ |12.3 | (RREQs) |RREQ Forwarding | |X|X|X|X|X| +------+ +-----------------------+-+-+-+-+-+-+ |12.4 | |RREQ Transmission |X|X|X|X|X|X| +------+----------------+-----------------------+-+-+-+-+-+-+ |13.1 | |RREP Generation |X|X|X|X|X|X| +------+ +-----------------------+-+-+-+-+-+-+ |13.2 | Route |RREP Processing |X|X|X|X|X|X| +------+ Replies +-----------------------+-+-+-+-+-+-+ |13.3 | (RREPs) |RREP Forwarding | |X|X|X|X|X| +------+ +-----------------------+-+-+-+-+-+-+ |13.4 | |RREP Transmission |X|X|X|X|X|X| +------+----------------+-----------------------+-+-+-+-+-+-+ |14.1 | |RERR Generation | |X|X|X| | | +------+ +-----------------------+-+-+-+-+-+-+ |14.2 | Route |RERR Processing | |X|X|X| | | +------+ Errors +-----------------------+-+-+-+-+-+-+ |14.3 | (RERRs) |RERR Forwarding | | |X|X| | | +------+ +-----------------------+-+-+-+-+-+-+ |14.4 | |RERR Transmission | |X|X|X| | | +------+----------------+-----------------------+-+-+-+-+-+-+ |15.1 | |RREP-ACK Generation |X|X|X|X|X|X| +------+ +-----------------------+-+-+-+-+-+-+ |15.2 | Route |RREQ-ACK Processing |X|X|X|X|X|X| +------+ Reply +-----------------------+-+-+-+-+-+-+ |15.3 | Acknowledgement|RREQ-ACK Forwarding |X|X|X|X|X|X| +------+ (RREP-ACKs) +-----------------------+-+-+-+-+-+-+ |15.4 | |RREQ-ACK Transmission |X|X|X|X|X|X| +------+----------------+-----------------------+-+-+-+-+-+-+ |16 | Metrics |Hop Count While |X|X|X|X|X|X| | | |Avoiding Weak Links | | | | | | | +------+----------------+-----------------------+-+-+-+-+-+-+ Test suite A : 1-hop bidirectional route establishment (scenarios 01, 02) Test suite B : 2-hop bidirectional route establishment (scenarios 03, 04, 05) Clausen, et al. Expires June 14, 2013 [Page 33] Internet-Draft LOADng Interop Report December 2012 Test suite C : 3-hop bidirectional route establishment (scenarios 06, 07, 08) Test suite D : 4-hop bidirectional route establishment (scenarios 09, 10) Test suite E : Establishment of the best bidirectional route (scenario 11) Test suite F : Blacklisting (scenario 12) 5. Interop 02: San Jose, USA March 2012 5.1. LOADng version tested The interoperability tests were conducted according to the specification in [LOADng-03]. NOTE: Due to the evolution of [LOADng] and this document, ome of the conventions used in Section 3, such as routing metric and some fields of messages, may be different from the description in [LOADng-03]. 5.2. Place and Date of Interoperability Test This section reports experience with the LOADng routing protocol, resulting from interoperability testing performed at Fujitsu Laboratories of America (FLA), San Jose, USA, on April 13, 2012. 5.3. Participating Implementations The following implementations were used to perform the interoperability tests this section, listed alphabetically: Ecole Polytechnique: "LIX" - This implementation was jointly developed by Axel Colin de Verdiere, Jiazi Yi, Ulrich Herberg and Thomas Clausen of Ecole Ploytechnique's networking team. It consists of approximately 6000 lines of JAVA code running in a Mac OS environment. It supports RREQ, RREP, RREP-ACK and RERR generation, processing, forwarding and transmission. Fujitsu Laboratories of America: "FLA" - This implementation was developed by Ulrich Herberg from Fujitsu Laboratories of America. It is a Java implementation, supporting basic features (RREQ, RREP, RREP-ACK generation, processing, forwarding and transmision). Clausen, et al. Expires June 14, 2013 [Page 34] Internet-Draft LOADng Interop Report December 2012 5.4. Interoperability Test Considerations As an intermediate test, only a subset of the scenarios described were tested (01, 03 and 05), for verifying interoperability bug- fixing the involved implementations. 5.5. Results For Scenario 01 The following table is summarizing the results obtained for the different combinations for which a 1-hop Forward Route and Reverse Route initial installation test was performed: +-----+------+------+ | | LIX | FLA | +-----+------+------+ | LIX | N/R | Pass | | FLA | Pass | N/R | +-----+------+------+ Table 13 5.6. Results For Scenrio 03 The following table is summarizing the results obtained for the different combinations for which a 2-hop Forward Route and Reverse Route initial installation test was performed: +-----+-----+-----+--------+ | A | B | C | Result | +-----+-----+-----+--------+ | LIX | FLA | LIX | Pass | | LIX | LIX | FLA | Pass | +-----+-----+-----+--------+ Table 14 5.7. Results For Scenario 05 The following table is summarizing the results obtained for the different combinations for which a Link breakage handling test was performed: +-----+-----+-----+--------+ | A | B | C | Result | +-----+-----+-----+--------+ | LIX | FLA | LIX | Pass | +-----+-----+-----+--------+ Clausen, et al. Expires June 14, 2013 [Page 35] Internet-Draft LOADng Interop Report December 2012 Table 15 6. Interop 03: Los Angeles, USA, June 2012 6.1. Version of LOADng Specification Tested The interoperability tests were conducted according to the specification in [LOADng-04]. NOTE: Due to the evolution of [LOADng] and this document, some of the conventions used in Section 3, such as routing metric and some fields of messages, may be different from the description in [LOADng-04]. 6.2. Place and Date of Interoperability Test This section reports experience with the LOADng routing protocol, resulting from interoperability testing performed at the Los Angeles Airport Hilton, USA, on June 6, 2012. 6.3. Participating Implementations The following implementations were used to perform the interoperability tests this section, listed alphabetically: Ecole Polytechnique: "LIX" - This implementation was jointly developed by Axel Colin de Verdiere, Jiazi Yi, Ulrich Herberg and Thomas Clausen of Ecole Ploytechnique's networking team. It consists of approximately 6000 lines of JAVA code running in a Mac OS environment. It supports RREQ, RREP, RREP-ACK and RERR generation, processing, forwarding and transmission. Fujitsu Laboratories of America: "FLA" - This implementation was developed by Ulrich Herberg from Fujitsu Laboratories of America. It is a Java implementation, supporting basic features (RREQ, RREP, RREP-ACK generation, processing, forwarding and transmision). 6.4. Scenarios Tested This interoperability test includes scenarios 01-12 (inclusive). 6.5. Additional Interoperability Test Considerations Wireshark packet sniffers, that have been modified to interpret LOADng control traffic, were used to monitor each single underlying link. For each test, the initiation of the communication resulting in the Clausen, et al. Expires June 14, 2013 [Page 36] Internet-Draft LOADng Interop Report December 2012 generation of the first LOADng control traffic message is always triggered manually. In addition, RREP-ACK LOADng control messages were systematically expected from each LOADng Router upon reception of a RREP LOADng control message in order to allow the detection of unidirectional links. 6.6. Results For Scenario 01-02 The following table is summarizing the results obtained for the different combinations for which test 1 (Forward Route and Reverse Route initial installation) was performed: +-----+------+------+ | | LIX | FLA | +-----+------+------+ | LIX | N/R | Pass | | FLA | Pass | N/R | +-----+------+------+ Table 16 The following table is summarizing the results obtained for the different combinations for which test 2 (Forward Route and Reverse Route updating) was performed: +-----+------+------+ | | LIX | FLA | +-----+------+------+ | LIX | N/R | Pass | | FLA | Pass | N/R | +-----+------+------+ Table 17 6.7. Results For Scenario 03-04-05 The following table is summarizing the results obtained for the different combinations for which these test 1 (Forward Route and Reverse Route initial installation) and test 2 (Forward Route and Reverse Route updating) were performed: +-----+-----+-----+--------+--------+ | A | B | C | Test 1 | Test 2 | +-----+-----+-----+--------+--------+ | LIX | FLA | LIX | Pass | Pass | | LIX | LIX | FLA | Pass | Pass | +-----+-----+-----+--------+--------+ Clausen, et al. Expires June 14, 2013 [Page 37] Internet-Draft LOADng Interop Report December 2012 Table 18 The following table is summarizing the results obtained for the different combinations for which these test 3 (Link breakage handling) was performed: +-----+-----+-----+--------+ | A | B | C | Test 3 | +-----+-----+-----+--------+ | FLA | LIX | LIX | Pass | | LIX | FLA | LIX | Pass | +-----+-----+-----+--------+ Table 19 6.8. Results For Scenario 06-07-08 The following table is summarizing the results obtained for the different combinations for which these test 1 (Forward Route and Reverse Route initial installation) and test 2 (Forward Route and Reverse Route updating) were performed: +-----+-----+-----+-----+--------+--------+ | A | B | C | D | Test 1 | Test 2 | +-----+-----+-----+-----+--------+--------+ | LIX | FLA | LIX | LIX | Pass | Pass | | LIX | LIX | FLA | LIX | Pass | Pass | | FLA | LIX | LIX | LIX | Pass | Pass | | LIX | LIX | LIX | FLA | Pass | Pass | +-----+-----+-----+-----+--------+--------+ Table 20 The following table is summarizing the results obtained for the different combinations for which these test 3 (Link breakage handling) was performed: +-----+-----+-----+-----+--------+ | A | B | C | D | Test 3 | +-----+-----+-----+-----+--------+ | FLA | LIX | LIX | LIX | Pass | | LIX | LIX | LIX | FLA | Pass | +-----+-----+-----+-----+--------+ Table 21 Clausen, et al. Expires June 14, 2013 [Page 38] Internet-Draft LOADng Interop Report December 2012 6.9. Results For Scenario 09-10 The following table is summarizing the results obtained for the different combinations for which test 1 (Forward Route and Reverse Route initial installation) and test 2 (Link breakage handling) were performed: +-----+-----+-----+-----+-----+--------+--------+ | A | B | C | D | E | Test 1 | Test 2 | +-----+-----+-----+-----+-----+--------+--------+ | FLA | FLA | LIX | LIX | LIX | Pass | Pass | | LIX | LIX | LIX | FLA | FLA | Pass | Pass | +-----+-----+-----+-----+-----+--------+--------+ Table 22 6.10. Results For Scenario 11 The following table is summarizing the results obtained for the different combinations for which this test was performed: +-----+-----+-----+--------+ | A | B | C | Result | +-----+-----+-----+--------+ | LIX | FLA | LIX | Pass | | LIX | LIX | FLA | Pass | | FLA | LIX | LIX | Pass | +-----+-----+-----+--------+ Table 23 6.11. Conclusions The different test scenarios that were carried out for different interoperable and independent implementations allowed to cover all major features of the LOADng specification by checking each technical feature one by one. In addition, the completion of this process permitted the improvement of the overall quality and accuracy of the [LOADng] specification. 7. Interop 04: Vancouver, Canada, August, 2011 7.1. Version of LOADng Specifiation Tested The interoperability tests were conducted according to the specification in [LOADng-05]. Clausen, et al. Expires June 14, 2013 [Page 39] Internet-Draft LOADng Interop Report December 2012 7.2. Place and Date of Interoperability Test This section reports experience with the LOADng routing protocol, resulting from interoperability testing performed at Hyatt Hotel, Vancouver, August 2nd, 2012. 7.3. Participating Implementations The following implementations were used to perform the interoperability tests this section, listed alphabetically: Ecole Polytechnique: "LIX" - This implementation was jointly developed by Axel Colin de Verdiere, Jiazi Yi, Ulrich Herberg and Thomas Clausen of Ecole Ploytechnique's networking team. It consists of approximately 6000 lines of JAVA code running in a Mac OS environment. It supports RREQ, RREP, RREP-ACK and RERR generation, processing, forwarding and transmission. Fujitsu Laboratories of America: "FLA" - This implementation was developed by Ulrich Herberg from Fujitsu Laboratories of America. It is a Java implementation, supporting all LOADng features (RREQ, RREP, RREP-ACK generation, processing, forwarding and transmission). 7.4. Scenarios Tested This interoperability test includes scenarios 01-05 (inclusive). 7.5. Additional Interoperability Test Considerations For each test, the initiation of the communication resulting in the generation of the first LOADng control traffic message is always triggered manually. In addition, RREP-ACK LOADng control messages were systematically expected from each LOADng Router upon reception of an RREP LOADng control message in order to allow the detection of unidirectional links. In this interop event, the use of different metrics types in the protocol were specifically considered. 7.6. Results for Scenario 01-02 The following table summarizes the results obtained for the different combinations for which test 1 (Forward Route and Reverse Route initial installation) was performed: Clausen, et al. Expires June 14, 2013 [Page 40] Internet-Draft LOADng Interop Report December 2012 +-----+------+------+ | | LIX | FLA | +-----+------+------+ | LIX | N/R | Pass | | FLA | Pass | N/R | +-----+------+------+ Table 24 The following table summarizes the results obtained for the different combinations for which test 2 (Forward Route and Reverse Route updating) was performed: +-----+------+------+ | | LIX | FLA | +-----+------+------+ | LIX | N/R | Pass | | FLA | Pass | N/R | +-----+------+------+ Table 25 7.7. Results for Scenario 03-04-05 The following table summarizes the results obtained for the different combinations for which these test 1 (Forward Route and Reverse Route initial installation) and test 2 (Forward Route and Reverse Route updating) were performed: +-----+-----+-----+--------+--------+ | A | B | C | Test 1 | Test 2 | +-----+-----+-----+--------+--------+ | LIX | FLA | LIX | Pass | Pass | | LIX | LIX | FLA | Pass | Pass | +-----+-----+-----+--------+--------+ Table 26 The following table is summarizing the results obtained for the different combinations for which these test 3 (Link breakage handling) was performed: Clausen, et al. Expires June 14, 2013 [Page 41] Internet-Draft LOADng Interop Report December 2012 +-----+-----+-----+--------+ | A | B | C | Test 3 | +-----+-----+-----+--------+ | FLA | LIX | LIX | Pass | | LIX | FLA | LIX | Pass | +-----+-----+-----+--------+ Table 27 In addition to conventional scenarios as described in Scenario 03 and Scenario 04 with the same metric type (HOP_COUNT, type 0), different metric types are tested in the same network. In the test, the originator of the RREQ initiates a route discovery using a metric type that the intermediate router does not understand (type 1). The following table summarizes the results obtained for the different combinations for which these test 1 (Forward Route and Reverse Route initial installation) and test 2 (Forward Route and Reverse Route updating), with different metric types: +-----+-----+-----+--------+--------+ | A | B | C | Test 1 | Test 2 | +-----+-----+-----+--------+--------+ | LIX | FLA | LIX | Pass | Pass | | LIX | LIX | FLA | Pass | Fail | +-----+-----+-----+--------+--------+ Table 28 One of the tests failed because handling unknown metric types was not defined properly in [LOADng-05] (but corrected in [LOADng-06], as a direct result of this interop test). Some changes from [LOADng-05] to [LOADng-06] that were proposed (and integrated in [LOADng-06]): 1. In section 13.1 ("RREP Generation"): o RREP.metric-type set to the same value as the RREQ.metric-type in the corresponding RREQ; is changed to o RREP.metric-type set to the same value as the RREQ.metric-type in the corresponding RREQ if the metric type is known to the router. Otherwise, RREP.metric-type is set to HOP_COUNT. Rationale: If a router that generates an RREP for an incoming RREQ does not know the metric from the RREQ, it will use the HOP_COUNT metric as fall-back. Per definition, all routers Clausen, et al. Expires June 14, 2013 [Page 42] Internet-Draft LOADng Interop Report December 2012 running LOADng support the HOP_COUNT metric. 2. In section 12.3 ("RREQ forwarding"): 3. RREQ.route-metric := received-route-metric + link-metric is changed to 3. If used-metric-type is not HOP_COUNT, then RREQ.route-metric := route-metric + link-metric Rationale: When the HOP_COUNT metric is used, the metric TLV value should remain unchanged, and instead the hop count from the message header is used to calculate the distance. 7.8. Conclusions As an intermediate test, and because of the limited time, only a subset of the scenarios (01, 02, 03, 04, 05) have been tested. In the performed tests, in addition to the conventional behaviors (regular message exchanges), different metric types, especially unknown metric types have been used in the network. The results show that for scenarios with only one metric type, the two implementations are able to interoperate with each other. However, when different metrics exist in the same network, the test failed in some scenarios. The problems are identified, and corresponding resolutions are proposed. The updates have been integrated in [LOADng-06]. 8. Security Considerations This document does currently not specify any security considerations. 9. IANA Considerations This document has no actions for IANA. 10. Contributors This specification is the result of the joint efforts of the following contributors -- listed alphabetically. o Alberto Camacho, LIX, France, o Thomas Heide Clausen, LIX, France, Clausen, et al. Expires June 14, 2013 [Page 43] Internet-Draft LOADng Interop Report December 2012 o Axel Colin de Verdiere, LIX, France, o Ulrich Herberg, Fujitsu Laboratories of America, USA, o Yuichi Igarashi, HITACHI YRL, Japan, o Nobukatsu Inomata, HITACHI ULSI Systems, Japan, o Yoko Morii, HITACHI YRL, Japan, o Hiroki Satoh, HITACHI YRL, Japan, o Jiazi Yi, LIX, France, 11. Informative References [LOADng] Clausen, T., Colin de Verdiere, A., Yi, J., Niktash, A., Igarashi, Y., Satoh, H., Herberg, U., Lavenu, C., Lys, T., Perkins, C., and J. Dean, "The Lightweight On-demand Ad hoc Distance-vector Routing Protocol - Next Generation (LOADng)", draft-clausen-lln-loadng (work in progress), October 2012. [LOADng-00] Clausen, T., Colin de Verdiere, A., Yi, J., Lavenu, C., Lys, T., Niktash, A., Igarashi, Y., and H. Satoh, "The LLN On-demand Ad hoc Distance-vector Routing Protocol - Next Generation (LOADng)", draft-clausen-lln-loadng-00 (work in progress), October 2011. [LOADng-03] Clausen, T., Colin de Verdiere, A., Yi, J., Niktash, A., Igarashi, Y., Satoh, H., Herberg, U., Lavenu, C., and T. Lys, "The LLN On-demand Ad hoc Distance-vector Routing Protocol - Next Generation (LOADng)", draft-clausen-lln-loadng-03 (work in progress), March 2012. [LOADng-04] Clausen, T., Colin de Verdiere, A., Yi, J., Niktash, A., Igarashi, Y., Satoh, H., Herberg, U., Lavenu, C., and T. Lys, "The LLN On-demand Ad hoc Distance-vector Routing Protocol - Next Generation (LOADng)", draft-clausen-lln-loadng-04 (work in progress), April 2012. [LOADng-05] Clausen, T., Colin de Verdiere, A., Yi, J., Niktash, A., Igarashi, Y., Satoh, H., Herberg, U., Lavenu, C., Lys, Clausen, et al. Expires June 14, 2013 [Page 44] Internet-Draft LOADng Interop Report December 2012 T., and C. Perkins, "The LLN On-demand Ad hoc Distance- vector Routing Protocol - Next Generation (LOADng)", draft-clausen-lln-loadng-05 (work in progress), July 2012. [LOADng-06] Clausen, T., Colin de Verdiere, A., Yi, J., Niktash, A., Igarashi, Y., Satoh, H., Herberg, U., Lavenu, C., Lys, T., Perkins, C., and J. Dean, "The Lightweight On-demand Ad hoc Distance-vector Routing Protocol - Next Generation (LOADng)", draft-clausen-lln-loadng-06 (work in progress), October 2012. Authors' Addresses Thomas Heide Clausen LIX, Ecole Polytechnique Phone: +33 6 6058 9349 EMail: T.Clausen@computer.org URI: http://www.ThomasClausen.org/ Alberto Camacho LIX, Ecole Polytechnique Phone: +34 636 309 835 EMail: alberto@albertocamacho.com URI: http://www.albertocamacho.com/ Jiazi Yi LIX, Ecole Polytechnique Phone: +33 1 6933 4031 EMail: jiazi@jiaziyi.com URI: http://www.jiaziyi.com/ Axel Colin de Verdiere LIX, Ecole Polytechnique Phone: +33 6 1264 7119 EMail: axel@axelcdv.com URI: http://www.axelcdv.com/ Clausen, et al. Expires June 14, 2013 [Page 45] Internet-Draft LOADng Interop Report December 2012 Yuichi Igarashi Hitachi, Ltd., Yokohama Research Laboratory Phone: +81 45 860 3083 EMail: yuichi.igarashi.hb@hitachi.com URI: http://www.hitachi.com/rd/yrl/index.html Hiroki Satoh Hitachi, Ltd., Yokohama Research Laboratory Phone: +81 44 959 0205 EMail: hiroki.satoh.yj@hitachi.com URI: http://www.hitachi.com/rd/yrl/index.html Yoko Morii Hitachi, Ltd., Yokohama Research Laboratory Phone: +81 45 860 3083 EMail: yoko.morii.cs@hitachi.com URI: http://www.hitachi.com/rd/yrl/index.html Ulrich Herberg Fujitsu Laboratories of America Phone: +1 408 530 4528 EMail: ulrich@herberg.name URI: http://www.herberg.name/ Cedric Lavenu EDF R&D Phone: +33 1 4765 2729 EMail: cedric-2.lavenu@edf.fr URI: http://www.edf.fr/ Clausen, et al. Expires June 14, 2013 [Page 46]