From mayutan.arumaithurai at gmail.com Thu Dec 1 11:51:06 2016 From: mayutan.arumaithurai at gmail.com (Mayutan A.) Date: Thu, 1 Dec 2016 20:51:06 +0100 Subject: [Ndn-interest] [CFP] ACM SIGCOMM 2017 - Standard and Experience Track Message-ID: ==================================================== CALL FOR PAPERS: Standard Track, Experience Track ACM SIGCOMM 2017 UCLA, Los Angeles, CA, USA on August 21-25, 2017 http://conferences.sigcomm.org/sigcomm/2017/cfp.html ==================================================== The ACM SIGCOMM 2017 conference seeks papers describing significant research contributions to the field of computer and data communication networks. We invite submissions on a wide range of networking research, including, but not limited to: - Network architectures and algorithms - Distributed systems design, implementation & application - Enterprise, datacenter, and storage area networks - SDN, NFV, and network programming - Experimental results from operational networks or network applications - Measurement of and statistics methods/machine learning/dat mining applied to networks - Network monitoring and diagnosis - Formal methods and network verification - Economic aspects of the Internet - Energy-aware communication - Network management and operations - Network security and privacy, censorship, transparency - Network, transport, and application-layer protocols - Fault-tolerance, reliability, and troubleshooting - Operating system and host support for networking - Novel physical layer technologies - Networking hardware (Router/switch/NIC) design - P2P and content (including video) distribution networks - Cloud and wide-area networking systems & infrastructure - Resource management, QoS, and signaling - Routing, traffic engineering, switching, and addressing - Wireless, mobile, and sensor networks - Edge, access, and home networks - Innovative uses of network data beyond communication In addition to the main conference, SIGCOMM 2017 will have a series of co-located workshops, tutorials, poster and demo sessions, a travel grant program, and conference best paper(s) and SIGCOMM awards. For the criteria for the selection of the conference best paper award(s), see Best Paper Award Guidelines. SUBMISSIONS SIGCOMM is a highly selective conference where full papers typically report novel results firmly substantiated by experimentation, deployment, simulation, or analysis. Submissions should be in two-column, 10-point format, and can be up to 12 pages in length with as many additional pages as necessary for references. Detailed submission instructions can be found on the conference web site. EXPERIENCE TRACK In addition to submissions considered for the standard track, SIGCOMM 2017 will also include a submission category for experience papers on the design, analysis, and evaluation of techniques in commercial (or otherwise widespread) deployment. Authors must explicitly indicate whether a submission is to be considered for this track or the standard track. (Further information in the detailed submission instructions.) ETHICAL CONCERNS Authors must as part of the submission process attest that their work complies with all applicable ethical standards of their home institution(s), including, but not limited to privacy policies and policies on experiments involving humans. Note that submitting research for approval by one's institution's ethics review body is necessary, but not sufficient ? in cases where the PC has concerns about the ethics of the work in a submission, the PC will have its own discussion of the ethics of that work. The PC takes a broad view of what constitutes an ethical concern, and authors agree to be available at any time during the review process to rapidly respond to queries from the PC chairs regarding ethical standards. ==================================================== IMPORTANT DATES Paper Registration Deadline: January 20, 2017 (5:00pm PST) Paper Submission Deadline: January 27, 2017 (5:00pm PST) Acceptance Notification: May 5, 2017 Conference dates: August 21-25, 2017 ==================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: From NourElHouda.BenYoussef at wevioo.com Thu Dec 1 13:01:40 2016 From: NourElHouda.BenYoussef at wevioo.com (Nour El Houda Ben Youssef) Date: Thu, 1 Dec 2016 21:01:40 +0000 Subject: [Ndn-interest] Collision detection In-Reply-To: References: Message-ID: <712FE4E7257849499414892DD92704BC7A9E4417@INFRAEX02.oxia.corp> Thank you for your reply If I don't want NDN to be working as an overlay network and to replace the whole protocol stack, I surly will be needing a solution for collision detection -----Message d'origine----- De?: Ndn-interest [mailto:ndn-interest-bounces at lists.cs.ucla.edu] De la part de ndn-interest-request at lists.cs.ucla.edu Envoy??: jeudi 1 d?cembre 2016 21:02 ??: ndn-interest at lists.cs.ucla.edu Objet?: Ndn-interest Digest, Vol 33, Issue 1 Send Ndn-interest mailing list submissions to ndn-interest at lists.cs.ucla.edu To subscribe or unsubscribe via the World Wide Web, visit http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest or, via email, send a message with subject or body 'help' to ndn-interest-request at lists.cs.ucla.edu You can reach the person managing the list at ndn-interest-owner at lists.cs.ucla.edu When replying, please edit your Subject line so it is more specific than "Re: Contents of Ndn-interest digest..." Today's Topics: 1. Re: Collision detection (Christos Papadopoulos) 2. [CFP] ACM SIGCOMM 2017 - Standard and Experience Track (Mayutan A.) ---------------------------------------------------------------------- Message: 1 Date: Wed, 30 Nov 2016 23:58:20 -0700 From: Christos Papadopoulos To: Subject: Re: [Ndn-interest] Collision detection Message-ID: Content-Type: text/plain; charset="windows-1252"; Format="flowed" The MAC layer is typically outside the scope of NDN, which operates at higher layers. But let me ask, why are you looking for an NDN solution at that layer? Christos. On 11/30/2016 11:49 PM, Nour El Houda Ben Youssef wrote: > > Dear ndn users > > I would like to ask about collision detection in NDN > > In case of using NDN as an overlay network, the mac layer will do the > job but for a pure ndn solution? > > Best regards > > *logoslogan.png* > > > > *Nour El Houda BEN YOUSSEF KOUB?A*|Doctorante > > Technopark El Ghazela 2088 Tunis- Tunisia > > Phone: +216 31 34 00 14 > > Mobile: +216 40 01 73 56 > > www.wevioo.com > > Paris ? Tunis - Dubai ? Alger > > > > > > _______________________________________________ > Ndn-interest mailing list > Ndn-interest at lists.cs.ucla.edu > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: image/png Size: 4997 bytes Desc: not available URL: ------------------------------ Message: 2 Date: Thu, 1 Dec 2016 20:51:06 +0100 From: "Mayutan A." To: "mayutan.a at gmail.com" Subject: [Ndn-interest] [CFP] ACM SIGCOMM 2017 - Standard and Experience Track Message-ID: Content-Type: text/plain; charset="utf-8" ==================================================== CALL FOR PAPERS: Standard Track, Experience Track ACM SIGCOMM 2017 UCLA, Los Angeles, CA, USA on August 21-25, 2017 http://conferences.sigcomm.org/sigcomm/2017/cfp.html ==================================================== The ACM SIGCOMM 2017 conference seeks papers describing significant research contributions to the field of computer and data communication networks. We invite submissions on a wide range of networking research, including, but not limited to: - Network architectures and algorithms - Distributed systems design, implementation & application - Enterprise, datacenter, and storage area networks - SDN, NFV, and network programming - Experimental results from operational networks or network applications - Measurement of and statistics methods/machine learning/dat mining applied to networks - Network monitoring and diagnosis - Formal methods and network verification - Economic aspects of the Internet - Energy-aware communication - Network management and operations - Network security and privacy, censorship, transparency - Network, transport, and application-layer protocols - Fault-tolerance, reliability, and troubleshooting - Operating system and host support for networking - Novel physical layer technologies - Networking hardware (Router/switch/NIC) design - P2P and content (including video) distribution networks - Cloud and wide-area networking systems & infrastructure - Resource management, QoS, and signaling - Routing, traffic engineering, switching, and addressing - Wireless, mobile, and sensor networks - Edge, access, and home networks - Innovative uses of network data beyond communication In addition to the main conference, SIGCOMM 2017 will have a series of co-located workshops, tutorials, poster and demo sessions, a travel grant program, and conference best paper(s) and SIGCOMM awards. For the criteria for the selection of the conference best paper award(s), see Best Paper Award Guidelines. SUBMISSIONS SIGCOMM is a highly selective conference where full papers typically report novel results firmly substantiated by experimentation, deployment, simulation, or analysis. Submissions should be in two-column, 10-point format, and can be up to 12 pages in length with as many additional pages as necessary for references. Detailed submission instructions can be found on the conference web site. EXPERIENCE TRACK In addition to submissions considered for the standard track, SIGCOMM 2017 will also include a submission category for experience papers on the design, analysis, and evaluation of techniques in commercial (or otherwise widespread) deployment. Authors must explicitly indicate whether a submission is to be considered for this track or the standard track. (Further information in the detailed submission instructions.) ETHICAL CONCERNS Authors must as part of the submission process attest that their work complies with all applicable ethical standards of their home institution(s), including, but not limited to privacy policies and policies on experiments involving humans. Note that submitting research for approval by one's institution's ethics review body is necessary, but not sufficient ? in cases where the PC has concerns about the ethics of the work in a submission, the PC will have its own discussion of the ethics of that work. The PC takes a broad view of what constitutes an ethical concern, and authors agree to be available at any time during the review process to rapidly respond to queries from the PC chairs regarding ethical standards. ==================================================== IMPORTANT DATES Paper Registration Deadline: January 20, 2017 (5:00pm PST) Paper Submission Deadline: January 27, 2017 (5:00pm PST) Acceptance Notification: May 5, 2017 Conference dates: August 21-25, 2017 ==================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ Subject: Digest Footer _______________________________________________ Ndn-interest mailing list Ndn-interest at lists.cs.ucla.edu http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest ------------------------------ End of Ndn-interest Digest, Vol 33, Issue 1 ******************************************* From lixia at CS.UCLA.EDU Thu Dec 1 13:19:36 2016 From: lixia at CS.UCLA.EDU (Lixia Zhang) Date: Thu, 1 Dec 2016 13:19:36 -0800 Subject: [Ndn-interest] Collision detection In-Reply-To: <712FE4E7257849499414892DD92704BC7A9E4417@INFRAEX02.oxia.corp> References: <712FE4E7257849499414892DD92704BC7A9E4417@INFRAEX02.oxia.corp> Message-ID: > On Dec 1, 2016, at 1:01 PM, Nour El Houda Ben Youssef wrote: > > Thank you for your reply > > If I don't want NDN to be working as an overlay network and to replace the whole protocol stack, I surly will be needing a solution for collision detection Wonder what would be your definition of "the whole protocol stack"? e.g. The TCP/IP protocol stack starts from IP, not link layer. This has allowed the protocol stack to out-live all the link layer protocols it started with back in early 80's. Lixia From NourElHouda.BenYoussef at wevioo.com Thu Dec 1 13:24:07 2016 From: NourElHouda.BenYoussef at wevioo.com (Nour El Houda Ben Youssef) Date: Thu, 1 Dec 2016 21:24:07 +0000 Subject: [Ndn-interest] Collision detection In-Reply-To: References: <712FE4E7257849499414892DD92704BC7A9E4417@INFRAEX02.oxia.corp> Message-ID: <712FE4E7257849499414892DD92704BC7A9E5438@INFRAEX02.oxia.corp> Thank you -----Message d'origine----- De?: Lixia Zhang [mailto:lixia at cs.ucla.edu] Envoy??: jeudi 1 d?cembre 2016 22:20 ??: Nour El Houda Ben Youssef Cc?: ndn-interest at lists.cs.ucla.edu Objet?: Re: [Ndn-interest] Collision detection > On Dec 1, 2016, at 1:01 PM, Nour El Houda Ben Youssef wrote: > > Thank you for your reply > > If I don't want NDN to be working as an overlay network and to replace the whole protocol stack, I surly will be needing a solution for collision detection Wonder what would be your definition of "the whole protocol stack"? e.g. The TCP/IP protocol stack starts from IP, not link layer. This has allowed the protocol stack to out-live all the link layer protocols it started with back in early 80's. Lixia From Matteo.Bertolino at eurecom.fr Fri Dec 2 06:38:16 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Fri, 02 Dec 2016 15:38:16 +0100 Subject: [Ndn-interest] Launching several instances of nlsr Message-ID: <20161202153816.rd41nmdrc48oosgs@webmail.eurecom.fr> Good morning, I need to perform a complicate question. I would like to launch several NDN nodes in different processes (i.e. 3 processes) on the same machine. Exactly the same behaviour of mini-ndn, but without. Then, about the NFD I had no problem: 1- Open three terminals. 2- On each terminal, change the HOME environment variable. 3- Put on the new home a new nfd.conf, changing the socket name (nfd.sock, nfd1.sock, nfd2.sock), the port numbers, and repeat in the client.config.sample. 4- On each terminal, launch nfd --config ~/nfd.conf & It works well. The problem arrives with NLSR. In each HOME, I put a different version of nlsr.conf, where I changed the names and the ip of the neighborhoods. Then I run on each node nlsr -d -f ~/nlsr.conf And here I noticed a strange thing: all the output, nlsr related, converges in one terminal. So I started to think that actually there are not 3 separated nlsr instances despite the three different processes. If I try to do nlsrc advertise /prefix , it tells me "success in advertisement", but no FIBS are actually updated. This demonstrate that the three instances are actually one. But, however, mini obtains a perfect separation. I tried to replicate exactly the same steps in nlsr.py, but I have no success. In your opinion, what do I forgot ? Thanks ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From mesarpe at gmail.com Fri Dec 2 07:34:59 2016 From: mesarpe at gmail.com (=?UTF-8?Q?C=C3=A9sar_A=2E_Bernardini?=) Date: Fri, 2 Dec 2016 16:34:59 +0100 Subject: [Ndn-interest] Launching several instances of nlsr In-Reply-To: <20161202153816.rd41nmdrc48oosgs@webmail.eurecom.fr> References: <20161202153816.rd41nmdrc48oosgs@webmail.eurecom.fr> Message-ID: I don't remember the parameters of nslr . Maybe the thing you could try is to create tap devices in Linux and attach the app to this fake interfaces. I remember I did this three years ago. Bests, On Dec 2, 2016 15:40, "Matteo Bertolino" wrote: > Good morning, I need to perform a complicate question. > I would like to launch several NDN nodes in different processes (i.e. 3 > processes) on the same machine. Exactly the same behaviour of mini-ndn, but > without. > Then, about the NFD I had no problem: > 1- Open three terminals. > 2- On each terminal, change the HOME environment variable. > 3- Put on the new home a new nfd.conf, changing the socket name (nfd.sock, > nfd1.sock, nfd2.sock), the port numbers, and repeat in the > client.config.sample. > 4- On each terminal, launch nfd --config ~/nfd.conf & > > It works well. The problem arrives with NLSR. > In each HOME, I put a different version of nlsr.conf, where I changed the > names and the ip of the neighborhoods. > Then I run on each node nlsr -d -f ~/nlsr.conf > > And here I noticed a strange thing: all the output, nlsr related, > converges in one terminal. So I started to think that actually there are > not 3 separated nlsr instances despite the three different processes. > If I try to do nlsrc advertise /prefix , it tells me "success in > advertisement", but no FIBS are actually updated. This demonstrate that the > three instances are actually one. > > But, however, mini obtains a perfect separation. I tried to replicate > exactly the same steps in nlsr.py, but I have no success. In your opinion, > what do I forgot ? Thanks > > ------------------------------------------------------------ > ------------------- > This message was sent using EURECOM Webmail: http://webmail.eurecom.fr > > > _______________________________________________ > Ndn-interest mailing list > Ndn-interest at lists.cs.ucla.edu > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Fri Dec 2 07:41:02 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Fri, 02 Dec 2016 16:41:02 +0100 Subject: [Ndn-interest] Launching several instances of nlsr In-Reply-To: References: <20161202153816.rd41nmdrc48oosgs@webmail.eurecom.fr> Message-ID: <20161202164102.1fzq7cbmswo0kw8g@webmail.eurecom.fr> Dear Cesar, thanks for the answer. Actually I did it, with linux namespaces. I am convinced that nlsr is the problem, because I compared two FIB in this topology: node1, node2 both attached to nodeGW. FIB of node1 in MINI: (just first line) /ndn/NLSR/sync nexthops={faceid=257 (cost=0), faceid=260 (cost=10)} FIB of node1 without MINI: (just first line) /ndn/NLSR/sync nexthops={faceid=263 (cost=0), faceid=265 (cost=0), faceid=266 (cost=0), faceid=264 (cost=10), faceid=267 (cost=10), faceid=268 (cost=10)} 6 entries vs 2. Then the NLSR instances are actually "together"! Quoting "C?sar A. Bernardini" : > I don't remember the parameters of nslr . Maybe the thing you could try is > to create tap devices in Linux and attach the app to this fake interfaces. > I remember I did this three years ago. > > Bests, > > On Dec 2, 2016 15:40, "Matteo Bertolino" > wrote: > >> Good morning, I need to perform a complicate question. >> I would like to launch several NDN nodes in different processes (i.e. 3 >> processes) on the same machine. Exactly the same behaviour of mini-ndn, but >> without. >> Then, about the NFD I had no problem: >> 1- Open three terminals. >> 2- On each terminal, change the HOME environment variable. >> 3- Put on the new home a new nfd.conf, changing the socket name (nfd.sock, >> nfd1.sock, nfd2.sock), the port numbers, and repeat in the >> client.config.sample. >> 4- On each terminal, launch nfd --config ~/nfd.conf & >> >> It works well. The problem arrives with NLSR. >> In each HOME, I put a different version of nlsr.conf, where I changed the >> names and the ip of the neighborhoods. >> Then I run on each node nlsr -d -f ~/nlsr.conf >> >> And here I noticed a strange thing: all the output, nlsr related, >> converges in one terminal. So I started to think that actually there are >> not 3 separated nlsr instances despite the three different processes. >> If I try to do nlsrc advertise /prefix , it tells me "success in >> advertisement", but no FIBS are actually updated. This demonstrate that the >> three instances are actually one. >> >> But, however, mini obtains a perfect separation. I tried to replicate >> exactly the same steps in nlsr.py, but I have no success. In your opinion, >> what do I forgot ? Thanks >> >> ------------------------------------------------------------ >> ------------------- >> This message was sent using EURECOM Webmail: http://webmail.eurecom.fr >> >> >> _______________________________________________ >> Ndn-interest mailing list >> Ndn-interest at lists.cs.ucla.edu >> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest >> > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From lanwang at memphis.edu Fri Dec 2 07:48:03 2016 From: lanwang at memphis.edu (Lan Wang (lanwang)) Date: Fri, 2 Dec 2016 15:48:03 +0000 Subject: [Ndn-interest] Launching several instances of nlsr In-Reply-To: <20161202164102.1fzq7cbmswo0kw8g@webmail.eurecom.fr> References: <20161202153816.rd41nmdrc48oosgs@webmail.eurecom.fr> <20161202164102.1fzq7cbmswo0kw8g@webmail.eurecom.fr> Message-ID: Why do you want to run multiple instances of NLSR on one node without mini-NDN? Lan > On Dec 2, 2016, at 9:41 AM, Matteo Bertolino wrote: > > Dear Cesar, > thanks for the answer. Actually I did it, with linux namespaces. > I am convinced that nlsr is the problem, because I compared two FIB in this topology: node1, node2 both attached to nodeGW. > > FIB of node1 in MINI: (just first line) > /ndn/NLSR/sync nexthops={faceid=257 (cost=0), faceid=260 (cost=10)} > > FIB of node1 without MINI: (just first line) > /ndn/NLSR/sync nexthops={faceid=263 (cost=0), faceid=265 (cost=0), faceid=266 (cost=0), faceid=264 (cost=10), faceid=267 (cost=10), faceid=268 (cost=10)} > > 6 entries vs 2. Then the NLSR instances are actually "together"! > > > Quoting "C?sar A. Bernardini" : > >> I don't remember the parameters of nslr . Maybe the thing you could try is >> to create tap devices in Linux and attach the app to this fake interfaces. >> I remember I did this three years ago. >> >> Bests, >> >> On Dec 2, 2016 15:40, "Matteo Bertolino" >> wrote: >> >>> Good morning, I need to perform a complicate question. >>> I would like to launch several NDN nodes in different processes (i.e. 3 >>> processes) on the same machine. Exactly the same behaviour of mini-ndn, but >>> without. >>> Then, about the NFD I had no problem: >>> 1- Open three terminals. >>> 2- On each terminal, change the HOME environment variable. >>> 3- Put on the new home a new nfd.conf, changing the socket name (nfd.sock, >>> nfd1.sock, nfd2.sock), the port numbers, and repeat in the >>> client.config.sample. >>> 4- On each terminal, launch nfd --config ~/nfd.conf & >>> >>> It works well. The problem arrives with NLSR. >>> In each HOME, I put a different version of nlsr.conf, where I changed the >>> names and the ip of the neighborhoods. >>> Then I run on each node nlsr -d -f ~/nlsr.conf >>> >>> And here I noticed a strange thing: all the output, nlsr related, >>> converges in one terminal. So I started to think that actually there are >>> not 3 separated nlsr instances despite the three different processes. >>> If I try to do nlsrc advertise /prefix , it tells me "success in >>> advertisement", but no FIBS are actually updated. This demonstrate that the >>> three instances are actually one. >>> >>> But, however, mini obtains a perfect separation. I tried to replicate >>> exactly the same steps in nlsr.py, but I have no success. In your opinion, >>> what do I forgot ? Thanks >>> >>> ------------------------------------------------------------ >>> ------------------- >>> This message was sent using EURECOM Webmail: http://webmail.eurecom.fr >>> >>> >>> _______________________________________________ >>> Ndn-interest mailing list >>> Ndn-interest at lists.cs.ucla.edu >>> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest >>> >> > > > > ------------------------------------------------------------------------------- > This message was sent using EURECOM Webmail: http://webmail.eurecom.fr > > > _______________________________________________ > Ndn-interest mailing list > Ndn-interest at lists.cs.ucla.edu > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest From jdd at wustl.edu Fri Dec 2 08:42:44 2016 From: jdd at wustl.edu (Dehart, John) Date: Fri, 2 Dec 2016 16:42:44 +0000 Subject: [Ndn-interest] Launching several instances of nlsr In-Reply-To: References: <20161202153816.rd41nmdrc48oosgs@webmail.eurecom.fr> <20161202164102.1fzq7cbmswo0kw8g@webmail.eurecom.fr> Message-ID: Lan, Just out of curiosity, how does NLSR communicate with NFD? Does it use ?transport? from client.conf (via ndn-cxx library) to find the nfd socket like other NFD clients? John > On Dec 2, 2016, at 9:48 AM, Lan Wang (lanwang) wrote: > > Why do you want to run multiple instances of NLSR on one node without mini-NDN? > > Lan > >> On Dec 2, 2016, at 9:41 AM, Matteo Bertolino wrote: >> >> Dear Cesar, >> thanks for the answer. Actually I did it, with linux namespaces. >> I am convinced that nlsr is the problem, because I compared two FIB in this topology: node1, node2 both attached to nodeGW. >> >> FIB of node1 in MINI: (just first line) >> /ndn/NLSR/sync nexthops={faceid=257 (cost=0), faceid=260 (cost=10)} >> >> FIB of node1 without MINI: (just first line) >> /ndn/NLSR/sync nexthops={faceid=263 (cost=0), faceid=265 (cost=0), faceid=266 (cost=0), faceid=264 (cost=10), faceid=267 (cost=10), faceid=268 (cost=10)} >> >> 6 entries vs 2. Then the NLSR instances are actually "together"! >> >> >> Quoting "C?sar A. Bernardini" : >> >>> I don't remember the parameters of nslr . Maybe the thing you could try is >>> to create tap devices in Linux and attach the app to this fake interfaces. >>> I remember I did this three years ago. >>> >>> Bests, >>> >>> On Dec 2, 2016 15:40, "Matteo Bertolino" >>> wrote: >>> >>>> Good morning, I need to perform a complicate question. >>>> I would like to launch several NDN nodes in different processes (i.e. 3 >>>> processes) on the same machine. Exactly the same behaviour of mini-ndn, but >>>> without. >>>> Then, about the NFD I had no problem: >>>> 1- Open three terminals. >>>> 2- On each terminal, change the HOME environment variable. >>>> 3- Put on the new home a new nfd.conf, changing the socket name (nfd.sock, >>>> nfd1.sock, nfd2.sock), the port numbers, and repeat in the >>>> client.config.sample. >>>> 4- On each terminal, launch nfd --config ~/nfd.conf & >>>> >>>> It works well. The problem arrives with NLSR. >>>> In each HOME, I put a different version of nlsr.conf, where I changed the >>>> names and the ip of the neighborhoods. >>>> Then I run on each node nlsr -d -f ~/nlsr.conf >>>> >>>> And here I noticed a strange thing: all the output, nlsr related, >>>> converges in one terminal. So I started to think that actually there are >>>> not 3 separated nlsr instances despite the three different processes. >>>> If I try to do nlsrc advertise /prefix , it tells me "success in >>>> advertisement", but no FIBS are actually updated. This demonstrate that the >>>> three instances are actually one. >>>> >>>> But, however, mini obtains a perfect separation. I tried to replicate >>>> exactly the same steps in nlsr.py, but I have no success. In your opinion, >>>> what do I forgot ? Thanks >>>> >>>> ------------------------------------------------------------ >>>> ------------------- >>>> This message was sent using EURECOM Webmail: http://webmail.eurecom.fr >>>> >>>> >>>> _______________________________________________ >>>> Ndn-interest mailing list >>>> Ndn-interest at lists.cs.ucla.edu >>>> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest >>>> >>> >> >> >> >> ------------------------------------------------------------------------------- >> This message was sent using EURECOM Webmail: http://webmail.eurecom.fr >> >> >> _______________________________________________ >> Ndn-interest mailing list >> Ndn-interest at lists.cs.ucla.edu >> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest > > > _______________________________________________ > Ndn-interest mailing list > Ndn-interest at lists.cs.ucla.edu > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest From Matteo.Bertolino at eurecom.fr Fri Dec 2 08:56:30 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Fri, 02 Dec 2016 17:56:30 +0100 Subject: [Ndn-interest] Launching several instances of nlsr (quite solved) In-Reply-To: References: <20161202153816.rd41nmdrc48oosgs@webmail.eurecom.fr> <20161202164102.1fzq7cbmswo0kw8g@webmail.eurecom.fr> Message-ID: <20161202175630.cw3scs1yos48ows8@webmail.eurecom.fr> Dear all, my problem was that client.conf must be under $HOME/.ndn and not only under $HOME. I was tricky from the example in the /usr/local/etc that put the client.conf.sample under the same level of the other nfd.conf and nlsr.conf. I will also answer to the other messages, you were very kind however. At the end I re-write the question of John since I am not sure that I could answer well to it. > Ashlesh > [...] You can create your own network namespaces, but why not use Mininet? > If you want to anyway - I would suggest to see how exactly Mininet does it. > Do these NLSR have the same log file, what does the neighbor section > look like? I followed this advice making a little of reverse engineering, thanks a lot. > Lan Why do you want to run multiple instances of NLSR on one node > without mini-NDN? Actually because I would like to obtain the same effects of mini without using mini: they adviced me that in order to understand very well how the things works, and in my opinion it is a good suggestion. before this work, I never take care about nlsr.conf, nfd.conf etc. The question of John is still open: > Lan, > Just out of curiosity, how does NLSR communicate with NFD? > Does it use ?transport? from client.conf (via ndn-cxx library) to > find the nfd > socket like other NFD clients? > John In my opinion yes, but absolutely I am not sure. It has sense ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From agawande at memphis.edu Fri Dec 2 09:04:56 2016 From: agawande at memphis.edu (Ashlesh Gawande (agawande)) Date: Fri, 2 Dec 2016 17:04:56 +0000 Subject: [Ndn-interest] Launching several instances of nlsr In-Reply-To: References: <20161202153816.rd41nmdrc48oosgs@webmail.eurecom.fr> <20161202164102.1fzq7cbmswo0kw8g@webmail.eurecom.fr> , Message-ID: Yes, NLSR uses transport from client.conf to find NFD socket. Ashlesh ________________________________ From: Ndn-interest on behalf of Dehart, John Sent: Friday, December 2, 2016 10:42:44 AM To: Lan Wang (lanwang) Cc: ndn-interest at lists.cs.ucla.edu Subject: Re: [Ndn-interest] Launching several instances of nlsr Lan, Just out of curiosity, how does NLSR communicate with NFD? Does it use ?transport? from client.conf (via ndn-cxx library) to find the nfd socket like other NFD clients? John > On Dec 2, 2016, at 9:48 AM, Lan Wang (lanwang) wrote: > > Why do you want to run multiple instances of NLSR on one node without mini-NDN? > > Lan > >> On Dec 2, 2016, at 9:41 AM, Matteo Bertolino wrote: >> >> Dear Cesar, >> thanks for the answer. Actually I did it, with linux namespaces. >> I am convinced that nlsr is the problem, because I compared two FIB in this topology: node1, node2 both attached to nodeGW. >> >> FIB of node1 in MINI: (just first line) >> /ndn/NLSR/sync nexthops={faceid=257 (cost=0), faceid=260 (cost=10)} >> >> FIB of node1 without MINI: (just first line) >> /ndn/NLSR/sync nexthops={faceid=263 (cost=0), faceid=265 (cost=0), faceid=266 (cost=0), faceid=264 (cost=10), faceid=267 (cost=10), faceid=268 (cost=10)} >> >> 6 entries vs 2. Then the NLSR instances are actually "together"! >> >> >> Quoting "C?sar A. Bernardini" : >> >>> I don't remember the parameters of nslr . Maybe the thing you could try is >>> to create tap devices in Linux and attach the app to this fake interfaces. >>> I remember I did this three years ago. >>> >>> Bests, >>> >>> On Dec 2, 2016 15:40, "Matteo Bertolino" >>> wrote: >>> >>>> Good morning, I need to perform a complicate question. >>>> I would like to launch several NDN nodes in different processes (i.e. 3 >>>> processes) on the same machine. Exactly the same behaviour of mini-ndn, but >>>> without. >>>> Then, about the NFD I had no problem: >>>> 1- Open three terminals. >>>> 2- On each terminal, change the HOME environment variable. >>>> 3- Put on the new home a new nfd.conf, changing the socket name (nfd.sock, >>>> nfd1.sock, nfd2.sock), the port numbers, and repeat in the >>>> client.config.sample. >>>> 4- On each terminal, launch nfd --config ~/nfd.conf & >>>> >>>> It works well. The problem arrives with NLSR. >>>> In each HOME, I put a different version of nlsr.conf, where I changed the >>>> names and the ip of the neighborhoods. >>>> Then I run on each node nlsr -d -f ~/nlsr.conf >>>> >>>> And here I noticed a strange thing: all the output, nlsr related, >>>> converges in one terminal. So I started to think that actually there are >>>> not 3 separated nlsr instances despite the three different processes. >>>> If I try to do nlsrc advertise /prefix , it tells me "success in >>>> advertisement", but no FIBS are actually updated. This demonstrate that the >>>> three instances are actually one. >>>> >>>> But, however, mini obtains a perfect separation. I tried to replicate >>>> exactly the same steps in nlsr.py, but I have no success. In your opinion, >>>> what do I forgot ? Thanks >>>> >>>> ------------------------------------------------------------ >>>> ------------------- >>>> This message was sent using EURECOM Webmail: http://webmail.eurecom.fr >>>> >>>> >>>> _______________________________________________ >>>> Ndn-interest mailing list >>>> Ndn-interest at lists.cs.ucla.edu >>>> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest >>>> >>> >> >> >> >> ------------------------------------------------------------------------------- >> This message was sent using EURECOM Webmail: http://webmail.eurecom.fr >> >> >> _______________________________________________ >> Ndn-interest mailing list >> Ndn-interest at lists.cs.ucla.edu >> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest > > > _______________________________________________ > Ndn-interest mailing list > Ndn-interest at lists.cs.ucla.edu > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest _______________________________________________ Ndn-interest mailing list Ndn-interest at lists.cs.ucla.edu http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From josh at caida.org Fri Dec 2 16:00:48 2016 From: josh at caida.org (Josh Polterock) Date: Fri, 2 Dec 2016 16:00:48 -0800 Subject: [Ndn-interest] Named Data Networking (NDN) Project Newsletter for October/November 2016 Message-ID: <20161203000048.GB85943@caida.org> Named Data Networking (NDN) Project Newsletter for October/November 2016 The NDN project team compiles and publishes this newsletter periodically to inform the community about recent activities, technical news, meetings, publications, presentations, code releases, and upcoming events. You can find these newsletters posted on the Named Data Networking Project blog. COMMUNITY OUTREACH * We will host our next NDNcomm at the University of Memphis March 23-24, 2017 immediately preceding IETF 98 in Chicago. We will hold the 4th NDN Hackathon immediately following NDNcomm on 25-26 March, 2017. You can find details regarding the meeting, travel grants, and accomodations at http://www.caida.org/workshops/ndn/1703/ * Coming up in December 2016, adjacent to IEEE Globecom 2016 at the Workshop on Information Centric Networking Solutions for Real World Applications (ICNSRA 2016) in Washington D.C., the NDN team will present a publication (listed below) as well as participate on a panel discussing "Application of ICN in Infrastructure-Free Environments: Rural Areas, Disaster Recovery, and Military Tactical Environments. See more details at http://icnsra.nz.comm.waseda.ac.jp/?page_id=61 * On the heels of our recent NDN Retreat, we held the 3rd Named Data Networking (NDN) Hackathon November 4-5th, Colorado State University, Fort Collins, CO. See which projects took home the coveted Quadcopter Mini Drones at http://3rd-ndn-hackathon.named-data.net/. TECHNICAL NEWS * We released the NDN Platform version 0.5.0 with updates to NFD, ndn-cxx version 0.5.0, the latest Common Client Library releases, and updates to NLSR and Mini-NDN. For details and links to the packages, please see: https://named-data.net/codebase/platform/#Version_050 * We released version 0.5.0 of Named Data Networking Forwarding Daemon (NFD) and ndn-cxx library. Please find detailed release notes: - NFD: http://named-data.net/doc/NFD/0.5.0-rc1/RELEASE_NOTES.html - ndn-cxx library: http://named-data.net/doc/ndn-cxx/0.5.0-rc1/RELEASE_NOTES.html More details about NFD, source code, install instructions, tutorials, HOWTOs, a FAQ and other useful resources are available on the official webpages of NFD and ndn-cxx: - http://named-data.net/doc/NFD/0.5.0/ - http://named-data.net/doc/ndn-cxx/0.5.0/ The latest version of the NFD developer's guide: - https://named-data.net/publications/techreports/ndn-0021-7-nfd-developer-guide/ * We released the NDN Python library PyNDN version v2.4beta1. with support for the ImplicitSha256Digest Name component. https://github.com/named-data/PyNDN2/releases/tag/v2.4beta1 https://pypi.python.org/pypi/PyNDN * We released the NDN Java library jNDN version v0.14 with support for the ImplicitSha256Digest Name component. https://github.com/named-data/jndn/releases/tag/v0.14 http://search.maven.org/#search|ga|1|g%3A%22net.named-data%22 * We released the NDN JavaScript library NDN-JS version v0.12.0. with support for the ImplicitSha256Digest Name component. https://github.com/named-data/ndn-js/releases/tag/v0.12.0 https://www.npmjs.org/package/ndn-js It also includes experimental updates to the Micro Forwarder to support multicast on the local LAN. https://github.com/named-data/ndn-js/blob/master/tools/micro-forwarder/INSTALL.md * As part of the Enterprise Building Automation & Management activities UCLA began publishing live building monitoring system data to a repository accessible on the testbed. Researchers interested in using the data for experiments should contact Zhehao @ REMAP for more information on the data and namespace. https://github.com/zhehaowang/bms-node NDN PUBLICATIONS, PRESENTATIONS, and TECHNICAL REPORTS * Wentao Shang, Alexander Afanasyev, and Lixia Zhang, "The Design and Implementation of the NDN Protocol Stack for RIOT-OS" to be presented at GLOBECOM Workshop on Information Centric Networking Solutions for Real World Applications, December 2016. https://named-data.net/publications/design_implementation_ndn_protocol/ * NDN TR-28 Revision 2: Spyridon Mastorakis, Alexander Afanasyev, Ilya Moiseenko, and Lixia Zhang, "ndnSIM 2: An updated NDN simulator for NS-3" to match the simulation platform to the latest advancements of NDN research. It uses the ndn-cxx library (NDN C++ library with eXperimental eXtensions) and the NDN Forwarding Daemon (NFD) to enable experiments with real code in a simulation environment. https://named-data.net/publications/techreports/ndn-0028-2-ndnsim-v2/ SEMINARS * The NDN seminars are internally focused. If you would like to participate in the NDN Seminars, please contact the PoC, UCLA PhD. candidate Spyridon Mastorakis for the most up-to-date information regarding upcoming seminars. - 5 October 2016 2pm Wentao Shang (UCLA), "The Design and Implementation of the NDN Protocol Stack for RIOT-OS" - 30 November 2016 2pm Susmit Shannigrahi, Chengyu Fan, Christos Papadopoulos (CSU), "Applying NDN to large scientific data" - 14 December 2016 2pm Wentao Shang, "VectorSync design overview" - 11 January 2016 2pm Junxiao Shi, "Technical Discussion of Self-Learning Strategy For more information about the Named Data Networking (NDN) Project please visit http://www.named-data.net/. From klaus at cs.arizona.edu Sun Dec 4 15:53:50 2016 From: klaus at cs.arizona.edu (Klaus Schneider) Date: Sun, 4 Dec 2016 16:53:50 -0700 Subject: [Ndn-interest] PCON source code release Message-ID: <8242f73f-93ad-9b77-0d1a-58f42279b19d@cs.arizona.edu> Dear all, since some have requested it, I just released the source code that I used in the experiments of our paper "A Practical Congestion Control Scheme for Named Data Networking" (http://conferences2.sigcomm.org/acm-icn/2016/proceedings/p21-schneider.pdf). You can find the code and documentation on how to run the experiments here: https://github.com/schneiderklaus/ndnSIM Please let me know if you have any comments or questions. Many thanks to the NFD and ndnSIM developers for making it possible. Best regards, Klaus From Matteo.Bertolino at eurecom.fr Thu Dec 8 02:51:50 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Thu, 08 Dec 2016 11:51:50 +0100 Subject: [Ndn-interest] Repo-ng does not work well? Message-ID: <20161208115150.6bu4atnsow4wg8sw@webmail.eurecom.fr> Good morning everyone. I've launched repo-ng on the host 192.168.0.100 and not "localhost" as usual, and I have some problems. Indeed, when I try to do: "base64 -d site.ndncert | nc 192.168.0.100 7376" it works, it works, but finally it does nothing, like a sleep. I launched the program with the following configuration file: [...] tcp_bulk_insert { host "192.168.0.100" port 7376 } Is it enough or do I miss something? Best regards, Matteo ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Thu Dec 8 03:31:01 2016 From: shijunxiao at email.arizona.edu (shijunxiao at email.arizona.edu) Date: Thu, 8 Dec 2016 04:31:01 -0700 Subject: [Ndn-interest] Repo-ng does not work well? In-Reply-To: <20161208115150.6bu4atnsow4wg8sw@webmail.eurecom.fr> References: <20161208115150.6bu4atnsow4wg8sw@webmail.eurecom.fr> Message-ID: <58494471.829d630a.51de8.d43e@mx.google.com> Hi Matteo Can you execute this command successfully on 192.168.0.110 machine itself? If so, check whether the TCP connection between the original machine and 192.168.0.110 is established by capturing a tcpdump trace and analyze it with Wireshark. If no, look at `netstat -an` output on 192.168.0.110 and see whether repo-ng is indeed listening on this endpoint. Yours, Junxiao From: Matteo Bertolino Sent: Thursday, December 8, 2016 03:52 To: ndn-interest at lists.cs.ucla.edu Subject: [Ndn-interest] Repo-ng does not work well? Good morning everyone. I've launched repo-ng on the host 192.168.0.100 and not "localhost" as usual, and I have some problems. Indeed, when I try to do: "base64 -d site.ndncert | nc 192.168.0.100 7376" it works, it works, but finally it does nothing, like a sleep. I launched the program with the following configuration file: [...] tcp_bulk_insert { host "192.168.0.100" port 7376 } Is it enough or do I miss something? Best regards, Matteo ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr _______________________________________________ Ndn-interest mailing list Ndn-interest at lists.cs.ucla.edu http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Thu Dec 8 08:04:43 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Thu, 08 Dec 2016 17:04:43 +0100 Subject: [Ndn-interest] Repo-ng does not work well? In-Reply-To: <58494471.829d630a.51de8.d43e@mx.google.com> References: <20161208115150.6bu4atnsow4wg8sw@webmail.eurecom.fr> <58494471.829d630a.51de8.d43e@mx.google.com> Message-ID: <20161208170443.7m6xa4x6skkwk4go@webmail.eurecom.fr> Dear Junxiao, thanks for your answer. It was a problem not NDN-related. My machine had not the interface lo up, so also the basic network operations like "ping myself" work well. Have a nice day, matteo Quoting shijunxiao at email.arizona.edu: > Hi Matteo > > Can you execute this command successfully on 192.168.0.110 machine itself? > If so, check whether the TCP connection between the original machine > and 192.168.0.110 is established by capturing a tcpdump trace and > analyze it with Wireshark. > If no, look at `netstat -an` output on 192.168.0.110 and see whether > repo-ng is indeed listening on this endpoint. > > Yours, Junxiao > > From: Matteo Bertolino > Sent: Thursday, December 8, 2016 03:52 > To: ndn-interest at lists.cs.ucla.edu > Subject: [Ndn-interest] Repo-ng does not work well? > > Good morning everyone. > I've launched repo-ng on the host 192.168.0.100 and not "localhost" as > usual, and I have some problems. Indeed, when I try to do: > "base64 -d site.ndncert | nc 192.168.0.100 7376" it works, it works, > but finally it does nothing, like a sleep. > > I launched the program with the following configuration file: > [...] > tcp_bulk_insert { > host "192.168.0.100" > port 7376 > } > > Is it enough or do I miss something? > Best regards, > Matteo > > ------------------------------------------------------------------------------- > This message was sent using EURECOM Webmail: http://webmail.eurecom.fr > > _______________________________________________ > Ndn-interest mailing list > Ndn-interest at lists.cs.ucla.edu > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest > > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From Matteo.Bertolino at eurecom.fr Mon Dec 12 07:39:24 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Mon, 12 Dec 2016 16:39:24 +0100 Subject: [Ndn-interest] Repo-ng freshness period Message-ID: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> Good morning, I would like to modify the repo-ng source code in order to put a Freshness period on the data that it returns. I tried to insert the modification immediately before each "face.put(*data)" making a research in the souce code with "grep" tool. Unfortunately, the data published continues to stay indefinitely inside the content store of the intermediate nodes. Where can I inject my modification? Bests, Matteo ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Mon Dec 12 10:47:47 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Mon, 12 Dec 2016 11:47:47 -0700 Subject: [Ndn-interest] Repo-ng freshness period In-Reply-To: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> References: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> Message-ID: Hi Matteo repo-ng cannot modify FreshnessPeriod on Data packets because it does not possess the signing key. Instead, the producer who creates and signs the Data should set FreshnessPeriod prior to inserting the Data into FreshnessPeriod. NFD does not erase Data whose FreshnessPeriod has expires from the ContentStore, because stale Data is still valid Data, and can be retrieved by an Interest without MustBeFresh selector. Stale Data cannot satisfy an Interest whose MustBeFresh selector is set. Staleness can be considered in a CS policy. For example, the priority_fifo policy evicts stale Data before fresh Data. This eviction occurs only if CS is full; stale Data won't be erased just because it becomes stale, and would stay indefinitely if the CS is never full. Yours, Junxiao On Mon, Dec 12, 2016 at 8:39 AM, Matteo Bertolino < Matteo.Bertolino at eurecom.fr> wrote: > I would like to modify the repo-ng source code in order to put a Freshness > period on the data that it returns. > I tried to insert the modification immediately before each > "face.put(*data)" making a research in the souce code with "grep" tool. > Unfortunately, the data published continues to stay indefinitely inside > the content store of the intermediate nodes. > Where can I inject my modification? > -------------- next part -------------- An HTML attachment was scrubbed... URL: From aa at CS.UCLA.EDU Mon Dec 12 13:18:29 2016 From: aa at CS.UCLA.EDU (Alex Afanasyev) Date: Mon, 12 Dec 2016 13:18:29 -0800 Subject: [Ndn-interest] ndnSIM 2.3 released Message-ID: <388F5737-0DA9-477E-98EF-9B3C92C098C3@cs.ucla.edu> Dear all, We are pleased to announce the release of ndnSIM version 2.3, a new release of the NDN simulator. The new version includes the updated submodules of NFD and ndn-cxx (baseed on 0.5.0 with updates from the master branch) and features full support of NDNLPv2, including support for network NACKs. For more details about the release, refer to https://ndnsim.net/2.3/RELEASE_NOTES.html Source code, instruction how to install how to compile ndnSIM, tutorials, FAQs, and other useful resources are available on official ndnSIM website: https://ndnsim.net/2.3/. * * * The new revision of the ndnSIM technical report will be available shortly at https://named-data.net/publications/techreports/ * * * A special welcome to our new contributor Mohammad Sabet. * * * The ndnSIM Team: Spyridon (Spyros) Mastorakis, Alexander Afanasyev, and others (https://ndnsim.net/2.3/meta/authors.html) From Matteo.Bertolino at eurecom.fr Mon Dec 12 14:27:16 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Mon, 12 Dec 2016 23:27:16 +0100 Subject: [Ndn-interest] Repo-ng freshness period In-Reply-To: References: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> Message-ID: <20161212232716.2tadyxne0w4ksoos@webmail.eurecom.fr> Dear Junxiao, first of all thanks. Even if you explanation seems very clear, for my poor knowledge I have still some black holes. Supposing that I would like to use repo-ng to store and to distribute the certificates. I can publish a certificate, for example, writing: base64 -d cert.ndncert | nc localhost 7376 in the very basic case. Now, client express an interest in order to retrieve cert.ndncert. I imagine, seeing the code, that repo-ng take the data and put the data in the face in order to satisfy the interest. Exactly like any other application. I would like to avoid that another node , that desire FRESH content, retrieve that certificate after 10 seconds on an intermediate cache. The request should arrive to repo-ng, instead of intermediate node. The solution you proposed is not enough for this purpose. Are you sure that I cannot obtain this behaviour? It seems quite strange, probably I don't explained very well what I wish. Bests, Matteo Quoting Junxiao Shi : > Hi Matteo > > repo-ng cannot modify FreshnessPeriod on Data packets because it does not > possess the signing key. Instead, the producer who creates and signs the > Data should set FreshnessPeriod prior to inserting the Data into > FreshnessPeriod. > > NFD does not erase Data whose FreshnessPeriod has expires from the > ContentStore, because stale Data is still valid Data, and can be retrieved > by an Interest without MustBeFresh selector. Stale Data cannot satisfy an > Interest whose MustBeFresh selector is set. > Staleness can be considered in a CS policy. For example, the priority_fifo > policy evicts stale Data before fresh Data. This eviction occurs only if CS > is full; stale Data won't be erased just because it becomes stale, and > would stay indefinitely if the CS is never full. > > Yours, Junxiao > > On Mon, Dec 12, 2016 at 8:39 AM, Matteo Bertolino < > Matteo.Bertolino at eurecom.fr> wrote: > >> I would like to modify the repo-ng source code in order to put a Freshness >> period on the data that it returns. >> I tried to insert the modification immediately before each >> "face.put(*data)" making a research in the souce code with "grep" tool. >> Unfortunately, the data published continues to stay indefinitely inside >> the content store of the intermediate nodes. >> Where can I inject my modification? >> > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Mon Dec 12 14:44:44 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Mon, 12 Dec 2016 15:44:44 -0700 Subject: [Ndn-interest] Repo-ng freshness period In-Reply-To: <20161212232716.2tadyxne0w4ksoos@webmail.eurecom.fr> References: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> <20161212232716.2tadyxne0w4ksoos@webmail.eurecom.fr> Message-ID: Hi Matteo You should change the program that generates and signs the certificate to add FreshnessPeriod, not repo-ng. If you patch repo-ng to update the certificate, it would not be valid anymore because the FreshnessPeriod is protected by the signature. Yours, Junxiao On Mon, Dec 12, 2016 at 3:27 PM, Matteo Bertolino < Matteo.Bertolino at eurecom.fr> wrote: > Supposing that I would like to use repo-ng to store and to distribute the > certificates. I can publish a certificate, for example, writing: > base64 -d cert.ndncert | nc localhost 7376 in the very basic case. > > Now, client express an interest in order to retrieve cert.ndncert. > I imagine, seeing the code, that repo-ng take the data and put the data in > the face in order to satisfy the interest. Exactly like any other > application. > I would like to avoid that another node , that desire FRESH content, > retrieve that certificate after 10 seconds on an intermediate cache. The > request should arrive to repo-ng, instead of intermediate node. > > The solution you proposed is not enough for this purpose. Are you sure > that I cannot obtain this behaviour? It seems quite strange, probably I > don't explained very well what I wish. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Mon Dec 12 14:53:11 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Mon, 12 Dec 2016 23:53:11 +0100 Subject: [Ndn-interest] Repo-ng freshness period In-Reply-To: References: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> <20161212232716.2tadyxne0w4ksoos@webmail.eurecom.fr> Message-ID: <20161212235311.xply6qtcpw8wcw88@webmail.eurecom.fr> Dear Junxiao, I understand (finally) what you mean. Since I generate the certificates with ndnsec, is there a method to obtain that? Thanks a lot, Bests, Matteo Quoting Junxiao Shi : > Hi Matteo > > You should change the program that generates and signs the certificate to > add FreshnessPeriod, not repo-ng. > If you patch repo-ng to update the certificate, it would not be valid > anymore because the FreshnessPeriod is protected by the signature. > > Yours, Junxiao > > On Mon, Dec 12, 2016 at 3:27 PM, Matteo Bertolino < > Matteo.Bertolino at eurecom.fr> wrote: > >> Supposing that I would like to use repo-ng to store and to distribute the >> certificates. I can publish a certificate, for example, writing: >> base64 -d cert.ndncert | nc localhost 7376 in the very basic case. >> >> Now, client express an interest in order to retrieve cert.ndncert. >> I imagine, seeing the code, that repo-ng take the data and put the data in >> the face in order to satisfy the interest. Exactly like any other >> application. >> I would like to avoid that another node , that desire FRESH content, >> retrieve that certificate after 10 seconds on an intermediate cache. The >> request should arrive to repo-ng, instead of intermediate node. >> >> The solution you proposed is not enough for this purpose. Are you sure >> that I cannot obtain this behaviour? It seems quite strange, probably I >> don't explained very well what I wish. >> > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From Marc.Mosko at parc.com Mon Dec 12 16:42:06 2016 From: Marc.Mosko at parc.com (Marc.Mosko at parc.com) Date: Tue, 13 Dec 2016 00:42:06 +0000 Subject: [Ndn-interest] Repo-ng freshness period In-Reply-To: <20161212235311.xply6qtcpw8wcw88@webmail.eurecom.fr> References: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> <20161212232716.2tadyxne0w4ksoos@webmail.eurecom.fr> <20161212235311.xply6qtcpw8wcw88@webmail.eurecom.fr> Message-ID: <26419491-FCB5-4D36-BD19-C80A6E5E05E4@parc.com> Matteo, If you expect to put a FreshnessSeconds in a certificate Data object that corresponds to the certificate?s validity period, I do not think you will get the behavior you want from NDN. The FreshnessPeriod is the relative time (in seconds) from when a ContentStore receives a Data Object to when it will consider it stale. This process is repeated at each hop. If you are putting a 10 second FreshnessSeconds in a certificate Data object to force nodes to ?phone home? to the repo every 10 seconds, you will not always get that either, because the 10 seconds is reset at each hop. Each hop along A -> B -> C -> ? fetches the certificate after 9 seconds, it will always be considered fresh for every hop and could thus live as fresh in the network for a long time. Usually getting the most recent data object is controlled by using a version (timestamp or sequence number) in the Data object?s name, then using selectors to get the right-most child (which due to caching may need to be repeated). Here?s an example: As an example, let?s assume it?s January 1 at 00:00:00 and your certificate expires at January 11 00:00:00 (864,000 seconds). So, node A publishes the certificate with a FreshnessSeconds of 864,000. It is then retrieved as follows: A -> 1 hour -> B -> 5 days -> C Node B will receive the certificate at January 1 01:00:00 (3600 seconds), so node B will consider it fresh until 864,000 + 3,600 = 867,600 seconds (January 11 01:00:00). Node C will receive the certificate after a delay of 3,600 + 432,000 = 435,600 seconds, so node C will consider it fresh until 1,299,600 seconds after node A published it (January 15 01:00:00). You could take a look at what NDN did in one of the recent NDNfit security papers, where they encoded the validity periods as part of the key?s name (I don?t remember the paper title right now), but it was something like NDN for constrained environments. Marc On Dec 12, 2016, at 2:53 PM, Matteo Bertolino > wrote: Dear Junxiao, I understand (finally) what you mean. Since I generate the certificates with ndnsec, is there a method to obtain that? Thanks a lot, Bests, Matteo Quoting Junxiao Shi >: Hi Matteo You should change the program that generates and signs the certificate to add FreshnessPeriod, not repo-ng. If you patch repo-ng to update the certificate, it would not be valid anymore because the FreshnessPeriod is protected by the signature. Yours, Junxiao On Mon, Dec 12, 2016 at 3:27 PM, Matteo Bertolino < Matteo.Bertolino at eurecom.fr> wrote: Supposing that I would like to use repo-ng to store and to distribute the certificates. I can publish a certificate, for example, writing: base64 -d cert.ndncert | nc localhost 7376 in the very basic case. Now, client express an interest in order to retrieve cert.ndncert. I imagine, seeing the code, that repo-ng take the data and put the data in the face in order to satisfy the interest. Exactly like any other application. I would like to avoid that another node , that desire FRESH content, retrieve that certificate after 10 seconds on an intermediate cache. The request should arrive to repo-ng, instead of intermediate node. The solution you proposed is not enough for this purpose. Are you sure that I cannot obtain this behaviour? It seems quite strange, probably I don't explained very well what I wish. ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr _______________________________________________ Ndn-interest mailing list Ndn-interest at lists.cs.ucla.edu http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Tue Dec 13 01:16:19 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Tue, 13 Dec 2016 10:16:19 +0100 Subject: [Ndn-interest] Repo-ng freshness period In-Reply-To: <26419491-FCB5-4D36-BD19-C80A6E5E05E4@parc.com> References: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> <20161212232716.2tadyxne0w4ksoos@webmail.eurecom.fr> <20161212235311.xply6qtcpw8wcw88@webmail.eurecom.fr> <26419491-FCB5-4D36-BD19-C80A6E5E05E4@parc.com> Message-ID: <20161213101619.yxhvyljy844gg8kg@webmail.eurecom.fr> Dear Marc, thank you your explanation, it is really clear and you convinced me that I cannot get the desired behavior just from NDN. About this case I really convinced, however I need to renew the question, because repo-ng could be used for the storage of different file. The one of the certificate was only an example, but at the end how can I avoid that intermediate content store consider the data published by repo-ng fresh indefinitely? Quoting Marc.Mosko at parc.com: > Matteo, > > If you expect to put a FreshnessSeconds in a certificate Data object > that corresponds to the certificate?s validity period, I do not > think you will get the behavior you want from NDN. The > FreshnessPeriod is the relative time (in seconds) from when a > ContentStore receives a Data Object to when it will consider it > stale. This process is repeated at each hop. > > If you are putting a 10 second FreshnessSeconds in a certificate > Data object to force nodes to ?phone home? to the repo every 10 > seconds, you will not always get that either, because the 10 seconds > is reset at each hop. Each hop along A -> B -> C -> ? fetches the > certificate after 9 seconds, it will always be considered fresh for > every hop and could thus live as fresh in the network for a long time. > > Usually getting the most recent data object is controlled by using a > version (timestamp or sequence number) in the Data object?s name, > then using selectors to get the right-most child (which due to > caching may need to be repeated). > > Here?s an example: > > As an example, let?s assume it?s January 1 at 00:00:00 and your > certificate expires at January 11 00:00:00 (864,000 seconds). So, > node A publishes the certificate with a FreshnessSeconds of 864,000. > It is then retrieved as follows: > > A -> 1 hour -> B -> 5 days -> C > > Node B will receive the certificate at January 1 01:00:00 (3600 > seconds), so node B will consider it fresh until 864,000 + 3,600 = > 867,600 seconds (January 11 01:00:00). > > Node C will receive the certificate after a delay of 3,600 + 432,000 > = 435,600 seconds, so node C will consider it fresh until 1,299,600 > seconds after node A published it (January 15 01:00:00). > > You could take a look at what NDN did in one of the recent NDNfit > security papers, where they encoded the validity periods as part of > the key?s name (I don?t remember the paper title right now), but it > was something like NDN for constrained environments. > > Marc > > > > On Dec 12, 2016, at 2:53 PM, Matteo Bertolino > > > wrote: > > Dear Junxiao, > I understand (finally) what you mean. > Since I generate the certificates with ndnsec, is there a method to > obtain that? > > Thanks a lot, > Bests, Matteo > > Quoting Junxiao Shi > >: > > Hi Matteo > > You should change the program that generates and signs the certificate to > add FreshnessPeriod, not repo-ng. > If you patch repo-ng to update the certificate, it would not be valid > anymore because the FreshnessPeriod is protected by the signature. > > Yours, Junxiao > > On Mon, Dec 12, 2016 at 3:27 PM, Matteo Bertolino < > Matteo.Bertolino at eurecom.fr> wrote: > > Supposing that I would like to use repo-ng to store and to distribute the > certificates. I can publish a certificate, for example, writing: > base64 -d cert.ndncert | nc localhost 7376 in the very basic case. > > Now, client express an interest in order to retrieve cert.ndncert. > I imagine, seeing the code, that repo-ng take the data and put the data in > the face in order to satisfy the interest. Exactly like any other > application. > I would like to avoid that another node , that desire FRESH content, > retrieve that certificate after 10 seconds on an intermediate cache. The > request should arrive to repo-ng, instead of intermediate node. > > The solution you proposed is not enough for this purpose. Are you sure > that I cannot obtain this behaviour? It seems quite strange, probably I > don't explained very well what I wish. > > > > > > ------------------------------------------------------------------------------- > This message was sent using EURECOM Webmail: > http://webmail.eurecom.fr > > > _______________________________________________ > Ndn-interest mailing list > Ndn-interest at lists.cs.ucla.edu > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest > > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Tue Dec 13 04:48:44 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Tue, 13 Dec 2016 05:48:44 -0700 Subject: [Ndn-interest] Repo-ng freshness period In-Reply-To: <20161212235311.xply6qtcpw8wcw88@webmail.eurecom.fr> References: <20161212163924.vw1n8f2twswo8ow4@webmail.eurecom.fr> <20161212232716.2tadyxne0w4ksoos@webmail.eurecom.fr> <20161212235311.xply6qtcpw8wcw88@webmail.eurecom.fr> Message-ID: Hi Matteo Since I generate the certificates with ndnsec, is there a method to obtain > that? > You'll have to patch ndnsec cert-gen to add FreshnessPeriod before the certificate is signed. https://github.com/named-data/ndn-cxx/blob/2fa5939375d39eb63239148aca4af6 a92f45637d/tools/ndnsec/cert-gen.hpp#L217 The one of the certificate was only an example, but at the end how can I > avoid that intermediate content store consider the data published by > repo-ng fresh indefinitely? > Current protocol does not have a mechanism for that. Consumers are expected to use a name discovery mechanism to determine whether the Data is latest. But I wonder what's the motivation for this? Yours, Junxiao -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Wed Dec 14 01:36:56 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Wed, 14 Dec 2016 10:36:56 +0100 Subject: [Ndn-interest] Again about repo: memory Message-ID: <20161214103656.ljynyl4lc04skc8s@webmail.eurecom.fr> Good morning community, I am wondering what about end of memory in repo. Sometimes it could be useful remove data from repo, but I don't see in repo-ng tools folder a program able to do that. Sometimes it could be more comfortable removing automatically data inside repo-ng after a certain amount of time. Any idea? Thanks to all, matteo ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Wed Dec 14 01:57:18 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Wed, 14 Dec 2016 09:57:18 +0000 Subject: [Ndn-interest] Again about repo: memory In-Reply-To: <20161214103656.ljynyl4lc04skc8s@webmail.eurecom.fr> References: <20161214103656.ljynyl4lc04skc8s@webmail.eurecom.fr> Message-ID: Hi Matteo repo-ng has a command for deleting a packet: https://redmine.named-data.net/projects/repo-ng/wiki/Repo_Deletion_Protocol The tool for executing this command is incomplete, and has been blocked in code review for two years: https://gerrit.named-data.net/1089 https://redmine.named-data.net/issues/1801 If your repo-ng is running out of memory, check out repo-sql which is designed to support unlimited amount of Data without being bound by server memory https://github.com/3rd-ndn-hackathon/repo-sql http://3rd-ndn-hackathon.named-data.net/hacks/junxiao-repo-sql.pdf Yours, Junxiao On Wed, Dec 14, 2016 at 02:38 Matteo Bertolino wrote: > Good morning community, > > I am wondering what about end of memory in repo. > > Sometimes it could be useful remove data from repo, but I don't see in > repo-ng tools folder a program able to do that. > > Sometimes it could be more comfortable removing automatically data > inside repo-ng after a certain amount of time. > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lfan1 at memphis.edu Wed Dec 14 22:34:49 2016 From: lfan1 at memphis.edu (Laqin Fan (lfan1)) Date: Thu, 15 Dec 2016 06:34:49 +0000 Subject: [Ndn-interest] Questions about broadcasting an interest Message-ID: Hello all, I have questions about broadcasting an interest, if a consumer asks for content by broadcasting its interest over all available connectivity, any node hearing the interest and having the satisfied data will respond with a data packet, the consumer will receive the first data packet? and how? what about the others? And if there are two data packets arriving at the consumer at the same time, what happens? Could anyone please make me clear? I really appreciate it. Thanks, Laqin Fan -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.h.ahmed at ieee.org Wed Dec 14 22:46:29 2016 From: s.h.ahmed at ieee.org (Syed Hassan Ahmed) Date: Thu, 15 Dec 2016 06:46:29 +0000 Subject: [Ndn-interest] Questions about broadcasting an interest In-Reply-To: References: Message-ID: Dear Laqin, While looking at the question technically, I would give two aspects: 1. Either we make an assumption that without any milisecond difference, if both packets are trying to reach same consumer: packets might collide/drop. In that case, Consumer's PIT entry is still having an entry and it will wait for a certain amount of time and will rebroadcast Interest as it is unsatisfied. 2. Or, if we avoid this narrow assumption, when the first copy of Data packet arrives, consumer matches the Data name with entries in PIT. Once it is determined that the received packet is not unsolicited, the Data is stored at CS (let's say) and corresponding PIT entry is deleted. On the arrival of second copy, it is treated as unsolicited packet and thus discarded. Meanwhile, it depends on the application as well. If the consumer prefers freshness as priority one, it may use Pub/Sub and wait for the fresh copy. However, that will lead the discussion towards another debate. I hope point 1 and 2 clears the question. Regards, Hassan https://sites.google.com/site/shahmedknu/ On Thu, Dec 15, 2016 at 3:35 PM Laqin Fan (lfan1) wrote: > > > > > > > > > > > > > > > > > Hello all, > > > > > > I have questions about broadcasting an interest, if a consumer asks for > content by broadcasting its interest over all available connectivity, any > node hearing the interest and having the satisfied data will respond with a > data packet, the > > consumer will receive the first data packet? and how? what about the > others? And if there are two data packets arriving at the consumer at the > same time, what happens? > > > Could anyone please make me clear? I really appreciate it. > > > > > > Thanks, > > > Laqin Fan > > > > > > > > > _______________________________________________ > > Ndn-interest mailing list > > Ndn-interest at lists.cs.ucla.edu > > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest > > -- Best Regards, Syed Hassan Ahmed. (??) Kyungpook National University, Daegu, Republic of Korea. https://sites.google.com/site/shahmedknu/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From Marc.Mosko at parc.com Wed Dec 14 23:56:22 2016 From: Marc.Mosko at parc.com (Marc.Mosko at parc.com) Date: Thu, 15 Dec 2016 07:56:22 +0000 Subject: [Ndn-interest] Questions about broadcasting an interest In-Reply-To: References: Message-ID: Unless some extra mechanism is used, one could have a Data object implosion problem when many other systems have a Data to satisfy the interest. The sender?s PIT will filter out multiple responses, but it is expensive network-wise. I am not sure if NFD will unicast or broadcast the response Data. The old CCNx 0.8 forwarder would apply a randomized delay before answering a broadcast and if it saw another response during that period, it would suppress its transmission. This is like a gossip style suppression sometimes used in MANETs. It does rely on the response Data being broadcast too so everyone else sees it. In a high-speed LAN, using this kind of broadcast-broadcast communications is pretty inefficient as a large switch often runs much slower with broadcast traffic. Over other styles of networks, like wifi, it might make sense, though one would lose RTS-CTS and ACKs, which introduces other problems. Personally, I think broadcasts should be use sparingly, such as for initial discovery or to elect a cluster head, etc., and then use unicast. Marc On Dec 14, 2016, at 10:34 PM, Laqin Fan (lfan1) > wrote: Hello all, I have questions about broadcasting an interest, if a consumer asks for content by broadcasting its interest over all available connectivity, any node hearing the interest and having the satisfied data will respond with a data packet, the consumer will receive the first data packet? and how? what about the others? And if there are two data packets arriving at the consumer at the same time, what happens? Could anyone please make me clear? I really appreciate it. Thanks, Laqin Fan _______________________________________________ Ndn-interest mailing list Ndn-interest at lists.cs.ucla.edu http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From christos at colostate.edu Thu Dec 15 07:10:32 2016 From: christos at colostate.edu (Christos Papadopoulos) Date: Thu, 15 Dec 2016 08:10:32 -0700 Subject: [Ndn-interest] Questions about broadcasting an interest In-Reply-To: References: Message-ID: <4b74468b-2f5d-b1a7-8475-9c3e4acb4de6@colostate.edu> Laqin I think we need more information before we can give you an answer. Why is the broadcast happening? Is only on the first hop from the user? Are you talking about a MAC layer broadcast or a message sent to all faces at the NDN layer? The following may or may not relate to your question. The way I understand the question is that, the NDN stack at the end user decides to broadcast an interest over all available faces, perhaps because it has no route or as part of an experiment looking for a better route to the content. I would not expect implosion at the end point, since typically there are not that many next hop potential responders (implosion would happen if there were many next hop NDN routers that receive the broadcast *and* they had the requested content). While I can construct scenarios where that may potentially happen, I can't imagine they will be very common. I expect the broadcast (at the MAC or NDN layers) to be handled by the strategy layer. Once responses start to arrive, the strategy layer will select one or more faces (possibly those that responded faster) and send future interests there. Christos. On 12/15/2016 12:56 AM, Marc.Mosko at parc.com wrote: > Unless some extra mechanism is used, one could have a Data object > implosion problem when many other systems have a Data to satisfy the > interest. The sender?s PIT will filter out multiple responses, but it > is expensive network-wise. I am not sure if NFD will unicast or > broadcast the response Data. > > The old CCNx 0.8 forwarder would apply a randomized delay before > answering a broadcast and if it saw another response during that > period, it would suppress its transmission. This is like a gossip > style suppression sometimes used in MANETs. It does rely on the > response Data being broadcast too so everyone else sees it. In a > high-speed LAN, using this kind of broadcast-broadcast communications > is pretty inefficient as a large switch often runs much slower with > broadcast traffic. Over other styles of networks, like wifi, it might > make sense, though one would lose RTS-CTS and ACKs, which introduces > other problems. > > Personally, I think broadcasts should be use sparingly, such as for > initial discovery or to elect a cluster head, etc., and then use unicast. > > Marc > >> On Dec 14, 2016, at 10:34 PM, Laqin Fan (lfan1) > > wrote: >> >> Hello all, >> I have questions about broadcasting an interest, if a consumer asks >> for content by broadcasting its interest over all available >> connectivity, any node hearing the interest and having the satisfied >> data will respond with a data packet, the consumer will receive the >> first data packet? and how? what about the others? And if there are >> two data packets arriving at the consumer at the same time, what happens? >> Could anyone please make me clear? I really appreciate it. >> Thanks, >> Laqin Fan >> _______________________________________________ >> Ndn-interest mailing list >> Ndn-interest at lists.cs.ucla.edu >> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest > > > > _______________________________________________ > Ndn-interest mailing list > Ndn-interest at lists.cs.ucla.edu > http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From lfan1 at memphis.edu Thu Dec 15 10:58:49 2016 From: lfan1 at memphis.edu (Laqin Fan (lfan1)) Date: Thu, 15 Dec 2016 18:58:49 +0000 Subject: [Ndn-interest] Questions about broadcasting an interest In-Reply-To: References: Message-ID: Hi Marc, Thank you for your response. My understanding is there will be a Data object implosion problem when many other nodes have a Data to satisfy the interest. PIT entry can filter the multiple response, but it will be expensive. And data broadcast traffic will make router (in the LAN) run much slower. And I also wanna know if NFD uses unicast or broadcast the response Data. ? Yours, Laqin From: Marc.Mosko at parc.com [mailto:Marc.Mosko at parc.com] Sent: Thursday, December 15, 2016 1:56 AM To: Laqin Fan (lfan1) Cc: ndn-interest at lists.cs.ucla.edu Subject: Re: [Ndn-interest] Questions about broadcasting an interest Unless some extra mechanism is used, one could have a Data object implosion problem when many other systems have a Data to satisfy the interest. The sender?s PIT will filter out multiple responses, but it is expensive network-wise. I am not sure if NFD will unicast or broadcast the response Data. The old CCNx 0.8 forwarder would apply a randomized delay before answering a broadcast and if it saw another response during that period, it would suppress its transmission. This is like a gossip style suppression sometimes used in MANETs. It does rely on the response Data being broadcast too so everyone else sees it. In a high-speed LAN, using this kind of broadcast-broadcast communications is pretty inefficient as a large switch often runs much slower with broadcast traffic. Over other styles of networks, like wifi, it might make sense, though one would lose RTS-CTS and ACKs, which introduces other problems. Personally, I think broadcasts should be use sparingly, such as for initial discovery or to elect a cluster head, etc., and then use unicast. Marc On Dec 14, 2016, at 10:34 PM, Laqin Fan (lfan1) > wrote: Hello all, I have questions about broadcasting an interest, if a consumer asks for content by broadcasting its interest over all available connectivity, any node hearing the interest and having the satisfied data will respond with a data packet, the consumer will receive the first data packet? and how? what about the others? And if there are two data packets arriving at the consumer at the same time, what happens? Could anyone please make me clear? I really appreciate it. Thanks, Laqin Fan _______________________________________________ Ndn-interest mailing list Ndn-interest at lists.cs.ucla.edu http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From lfan1 at memphis.edu Thu Dec 15 11:04:50 2016 From: lfan1 at memphis.edu (Laqin Fan (lfan1)) Date: Thu, 15 Dec 2016 19:04:50 +0000 Subject: [Ndn-interest] Questions about broadcasting an interest In-Reply-To: References: Message-ID: Hi Hassan, Thank you for your response, now I understand the question about two data packets arrive at the same time, what the consumer does and after the first data is received, the PIT entry will be deleted, other satisfied data packets will be discarded. Thanks. Yours, Laqin From: Syed Hassan Ahmed [mailto:s.h.ahmed at ieee.org] Sent: Thursday, December 15, 2016 12:46 AM To: Laqin Fan (lfan1) ; ndn-interest at lists.cs.ucla.edu Subject: Re: [Ndn-interest] Questions about broadcasting an interest Dear Laqin, While looking at the question technically, I would give two aspects: 1. Either we make an assumption that without any milisecond difference, if both packets are trying to reach same consumer: packets might collide/drop. In that case, Consumer's PIT entry is still having an entry and it will wait for a certain amount of time and will rebroadcast Interest as it is unsatisfied. 2. Or, if we avoid this narrow assumption, when the first copy of Data packet arrives, consumer matches the Data name with entries in PIT. Once it is determined that the received packet is not unsolicited, the Data is stored at CS (let's say) and corresponding PIT entry is deleted. On the arrival of second copy, it is treated as unsolicited packet and thus discarded. Meanwhile, it depends on the application as well. If the consumer prefers freshness as priority one, it may use Pub/Sub and wait for the fresh copy. However, that will lead the discussion towards another debate. I hope point 1 and 2 clears the question. Regards, Hassan https://sites.google.com/site/shahmedknu/ On Thu, Dec 15, 2016 at 3:35 PM Laqin Fan (lfan1) > wrote: Hello all, I have questions about broadcasting an interest, if a consumer asks for content by broadcasting its interest over all available connectivity, any node hearing the interest and having the satisfied data will respond with a data packet, the consumer will receive the first data packet? and how? what about the others? And if there are two data packets arriving at the consumer at the same time, what happens? Could anyone please make me clear? I really appreciate it. Thanks, Laqin Fan _______________________________________________ Ndn-interest mailing list Ndn-interest at lists.cs.ucla.edu http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -- Best Regards, Syed Hassan Ahmed. (??) Kyungpook National University, Daegu, Republic of Korea. https://sites.google.com/site/shahmedknu/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Mon Dec 19 08:32:07 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Mon, 19 Dec 2016 17:32:07 +0100 Subject: [Ndn-interest] Interest Fields Message-ID: <20161219173207.7z143ogoysw0cs4c@webmail.eurecom.fr> Good morning guys, I have some questions about interest fields and the content store. Specifically: 1- In a paper I've seen that an Interest could take the following optional fields: a) PublisherPublicKeyDigest (I just see PublisherPublicKeyLocator) b) AnswerOriginKind. But, grepping on ndn-cxx repo, I did not see that names. Does ndn-cxx provide the following fields? 2- Is it possible just caching some prefixes in the content store? Thanks to everyone! matteo ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Mon Dec 19 08:50:02 2016 From: shijunxiao at email.arizona.edu (shijunxiao at email.arizona.edu) Date: Mon, 19 Dec 2016 09:50:02 -0700 Subject: [Ndn-interest] Interest Fields In-Reply-To: <20161219173207.7z143ogoysw0cs4c@webmail.eurecom.fr> References: <20161219173207.7z143ogoysw0cs4c@webmail.eurecom.fr> Message-ID: <58580fb6.012f630a.79037.47d4@mx.google.com> Hi Matteo PublisherPublicKeyDigest has been replaced by PublisherPublicKeyLocator. The benefit is that routers do not need to compute the digest; the drawback is that the field is longer. AnswerOriginKind belongs to CCNx 0.7.2 but never made to NDN packet format. Its primary purpose is to restrict only producer or only cache can answer an Interest, but we believe it can be used for DoS attacks. Its secondary purpose is to delete a Data packet from local ccnd CS; we plan to implement a management command for this, but there has never been a reasonable use case. Yes, you can implement a CS policy to cache only certain prefixes. In afterInsert trigger of this policy, it should immediately evict any entries you do not want to cache by emitting beforeEvict signal. Yours, Junxiao From: Matteo Bertolino Sent: Monday, December 19, 2016 09:34 To: ndn-interest at lists.cs.ucla.edu Subject: [Ndn-interest] Interest Fields Good morning guys, I have some questions about interest fields and the content store. Specifically: 1- In a paper I've seen that an Interest could take the following optional fields: a) PublisherPublicKeyDigest (I just see PublisherPublicKeyLocator) b) AnswerOriginKind. But, grepping on ndn-cxx repo, I did not see that names. Does ndn-cxx provide the following fields? 2- Is it possible just caching some prefixes in the content store? Thanks to everyone! matteo ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr _______________________________________________ Ndn-interest mailing list Ndn-interest at lists.cs.ucla.edu http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest -------------- next part -------------- An HTML attachment was scrubbed... URL: From aa at CS.UCLA.EDU Mon Dec 19 09:40:12 2016 From: aa at CS.UCLA.EDU (Alex Afanasyev) Date: Mon, 19 Dec 2016 09:40:12 -0800 Subject: [Ndn-interest] Interest Fields In-Reply-To: <58580fb6.012f630a.79037.47d4@mx.google.com> References: <20161219173207.7z143ogoysw0cs4c@webmail.eurecom.fr> <58580fb6.012f630a.79037.47d4@mx.google.com> Message-ID: <862E1423-E651-4482-87D1-FDADB7C7553E@cs.ucla.edu> Hi Matteo, Just to add to what Junxiao said. The paper you're looking described the packet format at the time the paper was written. The format is updated from time to time to reflect the latest understanding: there were a few changes since the paper and we are planning to make a few more soon. The current format is described in the NDN packet specification: http://named-data.net/doc/ndn-tlv/ --- Alex > On Dec 19, 2016, at 8:50 AM, shijunxiao at email.arizona.edu wrote: > > Hi Matteo > > PublisherPublicKeyDigest has been replaced by PublisherPublicKeyLocator. The benefit is that routers do not need to compute the digest; the drawback is that the field is longer. > > AnswerOriginKind belongs to CCNx 0.7.2 but never made to NDN packet format. Its primary purpose is to restrict only producer or only cache can answer an Interest, but we believe it can be used for DoS attacks. Its secondary purpose is to delete a Data packet from local ccnd CS; we plan to implement a management command for this, but there has never been a reasonable use case. > > Yes, you can implement a CS policy to cache only certain prefixes. In afterInsert trigger of this policy, it should immediately evict any entries you do not want to cache by emitting beforeEvict signal. > > Yours, Junxiao > > From: Matteo Bertolino > Sent: Monday, December 19, 2016 09:34 > To: ndn-interest at lists.cs.ucla.edu > Subject: [Ndn-interest] Interest Fields > > Good morning guys, > I have some questions about interest fields and the content store. > Specifically: > > 1- In a paper I've seen that an Interest could take the following > optional fields: > a) PublisherPublicKeyDigest (I just see PublisherPublicKeyLocator) > b) AnswerOriginKind. > But, grepping on ndn-cxx repo, I did not see that names. Does ndn-cxx > provide the following fields? > > > 2- Is it possible just caching some prefixes in the content store? > > Thanks to everyone! > matteo > From Matteo.Bertolino at eurecom.fr Tue Dec 20 01:20:22 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Tue, 20 Dec 2016 10:20:22 +0100 Subject: [Ndn-interest] Interest Fields In-Reply-To: <862E1423-E651-4482-87D1-FDADB7C7553E@cs.ucla.edu> References: <20161219173207.7z143ogoysw0cs4c@webmail.eurecom.fr> <58580fb6.012f630a.79037.47d4@mx.google.com> <862E1423-E651-4482-87D1-FDADB7C7553E@cs.ucla.edu> Message-ID: <20161220102022.xe9kqalfs0gkcswg@webmail.eurecom.fr> Dears Junxiao and Alex, thank you you the answers and the link, I understand all now! Just to precise, the paper I'm referring is this one: http://www.ics.uci.edu/~gts/paps/ndn-dos-ICCCN13.pdf (page 2, just before section III) bests and happy xmas, matteo Quoting Alex Afanasyev : > Hi Matteo, > > Just to add to what Junxiao said. The paper you're looking > described the packet format at the time the paper was written. The > format is updated from time to time to reflect the latest > understanding: there were a few changes since the paper and we are > planning to make a few more soon. The current format is described > in the NDN packet specification: http://named-data.net/doc/ndn-tlv/ > > --- > Alex > >> On Dec 19, 2016, at 8:50 AM, shijunxiao at email.arizona.edu wrote: >> >> Hi Matteo >> >> PublisherPublicKeyDigest has been replaced by >> PublisherPublicKeyLocator. The benefit is that routers do not need >> to compute the digest; the drawback is that the field is longer. >> >> AnswerOriginKind belongs to CCNx 0.7.2 but never made to NDN packet >> format. Its primary purpose is to restrict only producer or only >> cache can answer an Interest, but we believe it can be used for DoS >> attacks. Its secondary purpose is to delete a Data packet from >> local ccnd CS; we plan to implement a management command for this, >> but there has never been a reasonable use case. >> >> Yes, you can implement a CS policy to cache only certain prefixes. >> In afterInsert trigger of this policy, it should immediately evict >> any entries you do not want to cache by emitting beforeEvict signal. >> >> Yours, Junxiao >> >> From: Matteo Bertolino >> Sent: Monday, December 19, 2016 09:34 >> To: ndn-interest at lists.cs.ucla.edu >> Subject: [Ndn-interest] Interest Fields >> >> Good morning guys, >> I have some questions about interest fields and the content store. >> Specifically: >> >> 1- In a paper I've seen that an Interest could take the following >> optional fields: >> a) PublisherPublicKeyDigest (I just see PublisherPublicKeyLocator) >> b) AnswerOriginKind. >> But, grepping on ndn-cxx repo, I did not see that names. Does ndn-cxx >> provide the following fields? >> >> >> 2- Is it possible just caching some prefixes in the content store? >> >> Thanks to everyone! >> matteo >> > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From lanwang at memphis.edu Tue Dec 20 12:36:23 2016 From: lanwang at memphis.edu (Lan Wang (lanwang)) Date: Tue, 20 Dec 2016 20:36:23 +0000 Subject: [Ndn-interest] registration open for NDNcomm 2017 and 4th NDN Hackathon Message-ID: <386C22A0-DEEE-4A3E-AB59-A9D76C783074@memphis.edu> Hi all, NDNcomm 2017 will be held at University of Memphis on March 23-24, 2017, followed by the 4th NDN hackathon on March 25-26, 2017. You are invited to register for NDNcomm 2017 at http://www.caida.org/workshops/ndn/1703/register.xml and 4th NDN Hackathon at http://4th-ndn-hackathon.named-data.net. We?ll send a Call for Contribution (presentations, posters, demos, and panels) in a few days. Lan, Christos and Lotfi NDNcomm 2017 Co-chairs From Matteo.Bertolino at eurecom.fr Wed Dec 21 08:22:10 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Wed, 21 Dec 2016 17:22:10 +0100 Subject: [Ndn-interest] Manually finalize an interest Message-ID: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> Dear all, I am wondering if my application can directly finalize an Interest calling the method onInterestFinalized (forwarder.cpp) on it. I would like to locally remove a PIT entry neither satisfying it with a Data or a NACK. Is it possible? Best regards, Matteo ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Wed Dec 21 09:27:13 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Wed, 21 Dec 2016 17:27:13 +0000 Subject: [Ndn-interest] Manually finalize an interest In-Reply-To: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> References: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> Message-ID: Hi Matteo No, you cannot. Forwarding is in NFD process, not in application process. You cannot directly call a function in another process. Yours, Junxiao On Wed, Dec 21, 2016 at 09:22 Matteo Bertolino wrote: > Dear all, > > I am wondering if my application can directly finalize an Interest > > calling the method onInterestFinalized (forwarder.cpp) on it. I would > > like to locally remove a PIT entry neither satisfying it with a Data > > or a NACK. Is it possible? > > Best regards, > > Matteo > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Wed Dec 21 09:53:57 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Wed, 21 Dec 2016 18:53:57 +0100 Subject: [Ndn-interest] Manually finalize an interest In-Reply-To: References: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> Message-ID: <20161221185357.if99q47qv44kssoc@webmail.eurecom.fr> Thanks! I suppose that I should act directly at NFD level. Stupid example: Malicious (stupid) guy: ehi man, I would like to fill-up the network with my malicious content! The prefix is /very/malicious! Then, all the NFDs of the world could filter the /very/malicious prefix calling immediately onInterestFinalized on it! This should be possible, i think (sorry for the not smart example but it is difficult imagine one) :) Matteo Quoting Junxiao Shi : > Hi Matteo > > No, you cannot. > Forwarding is in NFD process, not in application process. You cannot > directly call a function in another process. > > Yours, Junxiao > > On Wed, Dec 21, 2016 at 09:22 Matteo Bertolino > wrote: > >> Dear all, >> >> I am wondering if my application can directly finalize an Interest >> >> calling the method onInterestFinalized (forwarder.cpp) on it. I would >> >> like to locally remove a PIT entry neither satisfying it with a Data >> >> or a NACK. Is it possible? >> >> Best regards, >> >> Matteo >> >> > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Wed Dec 21 10:04:38 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Wed, 21 Dec 2016 18:04:38 +0000 Subject: [Ndn-interest] Manually finalize an interest In-Reply-To: <20161221185357.if99q47qv44kssoc@webmail.eurecom.fr> References: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> <20161221185357.if99q47qv44kssoc@webmail.eurecom.fr> Message-ID: Hi Matteo In this scenario, 1. Are you talking about a cache poisoning attack or an Interest flooding attack? 2. How does the network know /very/malicious prefix is malicious? 3. How does "every NFD in the world" (excluding attacker's NFD) get this alert? Yours, Junxiao On Wed, Dec 21, 2016 at 10:55 Matteo Bertolino wrote: > Thanks! I suppose that I should act directly at NFD level. > > Stupid example: > > Malicious (stupid) guy: ehi man, I would like to fill-up the network > > with my malicious content! The prefix is /very/malicious! > > Then, all the NFDs of the world could filter the /very/malicious > > prefix calling immediately onInterestFinalized on it! > > This should be possible, i think (sorry for the not smart example but > > it is difficult imagine one) :) > > Matteo > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Wed Dec 21 10:23:10 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Wed, 21 Dec 2016 19:23:10 +0100 Subject: [Ndn-interest] Manually finalize an interest In-Reply-To: References: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> <20161221185357.if99q47qv44kssoc@webmail.eurecom.fr> Message-ID: <20161221192310.nrf36622as0ckks4@webmail.eurecom.fr> Hi Junxiao, then: > 1. Are you talking about a cache poisoning attack or an Interest flooding > attack? No attacks in particular, it was just a stupid example in order to explain if the operation of calling onInterestFinalize was possible locally in an NFD :) > 2. How does the network know /very/malicious prefix is malicious? Assumption, like above :) The example was very stupid just to explain better the feasibility of "remove an interest from the PIT without satisfy it with nack or data" > 3. How does "every NFD in the world" (excluding attacker's NFD) get this > alert? If we focus on this example, just modifying the code of all the NFD inside an institution :) Probably a better example helps better to express my question. I am Jon and I know that Peter is my enemy. I don't want to receive anything from him. Jon offers services public in the world, but for his reason he doesn t want to perform this service to Peter. Jon desire to remove the Peter Interests as soon as possible without satisfy them neither with a data or nack. The PIT table has just space for 3 interests pending per time, then It is suitable do not occupy it. Jon could modify its NFD in order to do: if the interest received is from Peter then remove it by my PIT ? Thanks a lot, Matteo :) ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Wed Dec 21 11:09:19 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Wed, 21 Dec 2016 19:09:19 +0000 Subject: [Ndn-interest] Manually finalize an interest In-Reply-To: <20161221192310.nrf36622as0ckks4@webmail.eurecom.fr> References: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> <20161221185357.if99q47qv44kssoc@webmail.eurecom.fr> <20161221192310.nrf36622as0ckks4@webmail.eurecom.fr> Message-ID: Hi Matteo In this example, how does Jon detect the Interest is from Peter instead of someone else? If Peter is directly connected, just shutdown its face. Otherwise, you cannot distinguish who expressed the Interest. Yours, Junxiao On Wed, Dec 21, 2016 at 11:23 Matteo Bertolino wrote: > > > Probably a better example helps better to express my question. > > I am Jon and I know that Peter is my enemy. I don't want to receive > > anything from him. Jon offers services public in the world, but for > > his reason he doesn t want to perform this service to Peter. Jon > > desire to remove the Peter Interests as soon as possible without > > satisfy them neither with a data or nack. The PIT table has just space > > for 3 interests pending per time, then It is suitable do not occupy > > it. Jon could modify its NFD in order to do: > > if the interest received is from Peter then > > remove it by my PIT > > ? > > Thanks a lot, > > Matteo :) > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Wed Dec 21 11:16:46 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Wed, 21 Dec 2016 20:16:46 +0100 Subject: [Ndn-interest] Manually finalize an interest In-Reply-To: References: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> <20161221185357.if99q47qv44kssoc@webmail.eurecom.fr> <20161221192310.nrf36622as0ckks4@webmail.eurecom.fr> Message-ID: <20161221201646.bu8sux62gwgw8ogs@webmail.eurecom.fr> Dear Junxiao, let's suppose by assumption that Jon could surely identify all Peter interest because the interest must by signed to be taken into consideration :) Then Jon could have a black list of the enemy's keys. Since this assumption, is it possible deleting the PIT entry with a customized NFD code? If yes, is it enough calling onInterestFInalize method if the extracted key match with the blacklist ? :) Thank you again, Matteo Quoting Junxiao Shi : > Hi Matteo > > In this example, how does Jon detect the Interest is from Peter instead of > someone else? If Peter is directly connected, just shutdown its face. > Otherwise, you cannot distinguish who expressed the Interest. > > Yours, Junxiao > > On Wed, Dec 21, 2016 at 11:23 Matteo Bertolino > wrote: > >> >> >> Probably a better example helps better to express my question. >> >> I am Jon and I know that Peter is my enemy. I don't want to receive >> >> anything from him. Jon offers services public in the world, but for >> >> his reason he doesn t want to perform this service to Peter. Jon >> >> desire to remove the Peter Interests as soon as possible without >> >> satisfy them neither with a data or nack. The PIT table has just space >> >> for 3 interests pending per time, then It is suitable do not occupy >> >> it. Jon could modify its NFD in order to do: >> >> if the interest received is from Peter then >> >> remove it by my PIT >> >> ? >> >> Thanks a lot, >> >> Matteo :) >> >> > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From shijunxiao at email.arizona.edu Wed Dec 21 11:21:39 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Wed, 21 Dec 2016 19:21:39 +0000 Subject: [Ndn-interest] Manually finalize an interest In-Reply-To: <20161221201646.bu8sux62gwgw8ogs@webmail.eurecom.fr> References: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> <20161221185357.if99q47qv44kssoc@webmail.eurecom.fr> <20161221192310.nrf36622as0ckks4@webmail.eurecom.fr> <20161221201646.bu8sux62gwgw8ogs@webmail.eurecom.fr> Message-ID: Hi Matteo In this case, the closest you can get is to reject Peter's Interests in the strategy. The Interests will stay in the Interest table for a very short time (straggler timer), but won't be forwarded. Adjusting straggler timer to be very small will cause the Interest table entry to be ejected almost immediately, achieving the effect you wanted, but this would affect all Interests. Yours, Junxiao On Wed, Dec 21, 2016 at 12:16 Matteo Bertolino wrote: > Dear Junxiao, let's suppose by assumption that Jon could surely > > identify all Peter interest because the interest must by signed to be > > taken into consideration :) Then Jon could have a black list of the > > enemy's keys. Since this assumption, is it possible deleting the PIT > > entry with a customized NFD code? If yes, is it enough calling > > onInterestFInalize method if the extracted key match with the > > blacklist ? :) > > Thank you again, > > Matteo > > > > Quoting Junxiao Shi : > > > > > Hi Matteo > > > > > > In this example, how does Jon detect the Interest is from Peter instead > of > > > someone else? If Peter is directly connected, just shutdown its face. > > > Otherwise, you cannot distinguish who expressed the Interest. > > > > > > Yours, Junxiao > > > > > > On Wed, Dec 21, 2016 at 11:23 Matteo Bertolino < > Matteo.Bertolino at eurecom.fr> > > > wrote: > > > > > >> > > >> > > >> Probably a better example helps better to express my question. > > >> > > >> I am Jon and I know that Peter is my enemy. I don't want to receive > > >> > > >> anything from him. Jon offers services public in the world, but for > > >> > > >> his reason he doesn t want to perform this service to Peter. Jon > > >> > > >> desire to remove the Peter Interests as soon as possible without > > >> > > >> satisfy them neither with a data or nack. The PIT table has just space > > >> > > >> for 3 interests pending per time, then It is suitable do not occupy > > >> > > >> it. Jon could modify its NFD in order to do: > > >> > > >> if the interest received is from Peter then > > >> > > >> remove it by my PIT > > >> > > >> ? > > >> > > >> Thanks a lot, > > >> > > >> Matteo :) > > >> > > >> > > > > > > > > > > > > ------------------------------------------------------------------------------- > > This message was sent using EURECOM Webmail: http://webmail.eurecom.fr > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matteo.Bertolino at eurecom.fr Thu Dec 22 02:50:01 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Thu, 22 Dec 2016 11:50:01 +0100 Subject: [Ndn-interest] Manually finalize an interest In-Reply-To: References: <20161221172210.8wdyioupwkoo4ooo@webmail.eurecom.fr> <20161221185357.if99q47qv44kssoc@webmail.eurecom.fr> <20161221192310.nrf36622as0ckks4@webmail.eurecom.fr> <20161221201646.bu8sux62gwgw8ogs@webmail.eurecom.fr> Message-ID: <20161222115001.va7u097fggko8wgc@webmail.eurecom.fr> Dear Junxiao, that's clear! Thanks a lot, Bests Matteo Quoting Junxiao Shi : > Hi Matteo > > In this case, the closest you can get is to reject Peter's Interests in the > strategy. The Interests will stay in the Interest table for a very short > time (straggler timer), but won't be forwarded. > Adjusting straggler timer to be very small will cause the Interest table > entry to be ejected almost immediately, achieving the effect you wanted, > but this would affect all Interests. > > Yours, Junxiao > > On Wed, Dec 21, 2016 at 12:16 Matteo Bertolino > wrote: > >> Dear Junxiao, let's suppose by assumption that Jon could surely >> >> identify all Peter interest because the interest must by signed to be >> >> taken into consideration :) Then Jon could have a black list of the >> >> enemy's keys. Since this assumption, is it possible deleting the PIT >> >> entry with a customized NFD code? If yes, is it enough calling >> >> onInterestFInalize method if the extracted key match with the >> >> blacklist ? :) >> >> Thank you again, >> >> Matteo >> >> >> >> Quoting Junxiao Shi : >> >> >> >> > Hi Matteo >> >> > >> >> > In this example, how does Jon detect the Interest is from Peter instead >> of >> >> > someone else? If Peter is directly connected, just shutdown its face. >> >> > Otherwise, you cannot distinguish who expressed the Interest. >> >> > >> >> > Yours, Junxiao >> >> > >> >> > On Wed, Dec 21, 2016 at 11:23 Matteo Bertolino < >> Matteo.Bertolino at eurecom.fr> >> >> > wrote: >> >> > >> >> >> >> >> >> >> >> >> Probably a better example helps better to express my question. >> >> >> >> >> >> I am Jon and I know that Peter is my enemy. I don't want to receive >> >> >> >> >> >> anything from him. Jon offers services public in the world, but for >> >> >> >> >> >> his reason he doesn t want to perform this service to Peter. Jon >> >> >> >> >> >> desire to remove the Peter Interests as soon as possible without >> >> >> >> >> >> satisfy them neither with a data or nack. The PIT table has just space >> >> >> >> >> >> for 3 interests pending per time, then It is suitable do not occupy >> >> >> >> >> >> it. Jon could modify its NFD in order to do: >> >> >> >> >> >> if the interest received is from Peter then >> >> >> >> >> >> remove it by my PIT >> >> >> >> >> >> ? >> >> >> >> >> >> Thanks a lot, >> >> >> >> >> >> Matteo :) >> >> >> >> >> >> >> >> > >> >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------- >> >> This message was sent using EURECOM Webmail: http://webmail.eurecom.fr >> >> >> >> > ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From jasper at endlessm.com Thu Dec 22 12:14:22 2016 From: jasper at endlessm.com (Jasper St. Pierre) Date: Thu, 22 Dec 2016 12:14:22 -0800 Subject: [Ndn-interest] Quick question on names and name structure Message-ID: Hi all, I'm part of a team looking at deploying NDN as a data distribution technology for our other technologies. We're using the NDN naming conventions established in NDN-0022: http://named-data.net/doc/tech-memos/naming-conventions.pdf This is a nice piece of initial standardization that allows us to talk about segment/block numbers. However, its unfortunate that all parts of our NDN-based stack have to "know" about these conventions. e.g. if we want to receive segment 5 of version 3 of name /A/B/C, we could either have the resulting name /A/B/C/%FD%03/%00%05 or /A/B/C/%00%05/%FD%03 While this is a simple case that seems to have an obvious "correct" answer, in more complex scenarios, it is difficult to build generic code and structures that can handle all of these cases, especially as we add more and more components to these names. While we could support both orders, this loses our ability to cache routes in NFD. It would be nicer if these more "parameters" to the name could be specified in an unordered set, or if there was at least a canonical correct ordering established in the tech report. -- Jasper -------------- next part -------------- An HTML attachment was scrubbed... URL: From lanwang at memphis.edu Fri Dec 23 04:55:51 2016 From: lanwang at memphis.edu (Lan Wang (lanwang)) Date: Fri, 23 Dec 2016 12:55:51 +0000 Subject: [Ndn-interest] Named Data Networking Community Meeting (NDNcomm) 2017 Message-ID: Call for Contributions Named Data Networking Community Meeting (NDNcomm) 2017 March 23-24, 2017 Memphis, TN, USA http://www.caida.org/workshops/ndn/1703/ Submission deadline: 23:59 EST, Monday, Jan. 23, 2017 The organizing committee invites you to contribute to NDNComm 2017 to be held at the University of Memphis, Memphis, TN, USA on March 23-24, 2017. NDNComm is an annual event that brings together a large community of researchers, users and other parties interested in Named Data Networking (NDN). NDN is an instance of Information Centric Networking (ICN), which allows named content to be accessed directly rather than through network hosts resulting in a network service model that is much better aligned with application requirements. Named content offers many advantages, such as reliability, security, scalability and robustness in communication. In addition, NDN is also highly suitable for newly emerging IoT applications. NDNcomm provides a community forum to discuss the state of the NDN software platform and testbed, and guide the evolution of the NDN architecture. As the NSF NDN-NP project concludes its final project year, NDNcomm 2017 will highlight the research results from this project team, in particular the IoT and multimedia applications developed by the project. Moreover, we invite submissions from the broader NDN community for presentations, posters, demos and panels on topics including but not limited to: ? NDN applications, such as scientific data, education, entertainment, tactical edge, smart cities and transportation; ? NDN support for IoT, mobile, and ad hoc environments; ? Security and privacy at different layers of the architecture; ? Management of NDN networks; ? Experience with NDN experimentation; ? Strategies to stimulate NDN development, both from research and commercial perspectives. ------------------ Submission Instructions Submissions should be one page in PDF format containing a title, authors and an abstract. The title should start with ?Presentation:?, ?Poster:?, ?Demo:? or ?Panel:?. Please submit your abstract using http://ndncomm2017.named-data.net/. ------------------- Deadlines ? Abstracts for presentations, posters, demos and panels: 23:59 EST, Monday, Jan. 23, 2017 ? Notifications: 23:59 EST, Friday, Feb. 10, 2017 Abstract submission is not required for participation in NDNcomm 2017. Interested parties can register for NDNcomm 2017 at http://www.caida.org/workshops/ndn/1703/register.xml and the 4th NDN hackathon on March 25-26, 2017 at http://4th-ndn-hackathon.named-data.net. The community meeting will be webcast. Please visit the NDN website (www.named-data.net) closer to the meeting for instructions. Lan ************************************************ Lan Wang Professor & Chair Department of Computer Science University of Memphis Memphis, TN 38152 Phone: 901-678-1643 URL: http://www.cs.memphis.edu/~lanwang *********************************************** From Matteo.Bertolino at eurecom.fr Tue Dec 27 10:15:28 2016 From: Matteo.Bertolino at eurecom.fr (Matteo Bertolino) Date: Tue, 27 Dec 2016 19:15:28 +0100 Subject: [Ndn-interest] Retrieve the identity name from signed interest Message-ID: <20161227191528.alt9esj9gkskkcwo@webmail.eurecom.fr> Good morning community, I would like to know if, by C++ code, is possible retrieving the identity name when I receive a signed interest. If yes, how? A signed interest has the following fields: prefix, timestamp, randomValue, SignatureInfo, SignatureValue. The I think that I could retrieve the resource pointed by the KeyLocator, but not the identity name of who signed the interest. Am I wrong? Please tell me yes :) Bests, Matteo ------------------------------------------------------------------------------- This message was sent using EURECOM Webmail: http://webmail.eurecom.fr From NourElHouda.BenYoussef at wevioo.com Wed Dec 28 00:23:18 2016 From: NourElHouda.BenYoussef at wevioo.com (Nour El Houda Ben Youssef) Date: Wed, 28 Dec 2016 08:23:18 +0000 Subject: [Ndn-interest] Nack reason Message-ID: <712FE4E7257849499414892DD92704BC7A9F6CB9@INFRAEX02.oxia.corp> Dear community Each time I emit an interest, I receive a Nack where nack.hetReason equals 150 What does exactly this nack means Best regards [logoslogan.png] Nour El Houda BEN YOUSSEF KOUB?A|Doctorante Technopark El Ghazela 2088 Tunis- Tunisia Phone: +216 31 34 00 14 Mobile: +216 40 01 73 56 www.wevioo.com Paris - Tunis - Dubai - Alger -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 4997 bytes Desc: image001.png URL: From shijunxiao at email.arizona.edu Wed Dec 28 19:10:46 2016 From: shijunxiao at email.arizona.edu (Junxiao Shi) Date: Thu, 29 Dec 2016 03:10:46 +0000 Subject: [Ndn-interest] Retrieve the identity name from signed interest In-Reply-To: <20161227191528.alt9esj9gkskkcwo@webmail.eurecom.fr> References: <20161227191528.alt9esj9gkskkcwo@webmail.eurecom.fr> Message-ID: Hi Matteo No, there is no reliable and universal way to infer identity name from KeyLocator. Data packets and signed Interests are signed by a *key*. The private key is held by the signer, while the public key is published in a *certificate*. KeyLocator contains the name of that certificate. *Identity* is a way used by ndn-cxx to organize one's own keys and certificates. It has no meaning for a relying party. The trust schema should be defined in terms of keys and certificates, and should not use identity names. Yours, Junxiao On Tue, Dec 27, 2016 at 11:16 Matteo Bertolino wrote: > Good morning community, > > I would like to know if, by C++ code, is possible retrieving the > > identity name when I receive a signed interest. > > If yes, how? > > A signed interest has the following fields: > > prefix, timestamp, randomValue, SignatureInfo, SignatureValue. > > The I think that I could retrieve the resource pointed by the > > KeyLocator, but not the identity name of who signed the interest. Am I > > wrong? Please tell me yes :) > > Bests, > > Matteo > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From s.h.ahmed at ieee.org Fri Dec 30 20:49:05 2016 From: s.h.ahmed at ieee.org (Syed Hassan Ahmed) Date: Sat, 31 Dec 2016 13:49:05 +0900 Subject: [Ndn-interest] New Year Greetings and CFPs Message-ID: Dear Colleagues, Seniors, and students, Seasonal Greetings! First of all, I would like to wish you a great and successful New Year ahead. I hope that the Year 2017 will be bringing a lot of innovations and further developments in the domain wireless communications and computer networks. Through this email, I am sharing the opportunities of publishing your recent findings regarding the Future Networking domains and paradigms: 1. 2nd RAFNET Workshop in conjunction with IEEE VTC Spring 2017, Sydney, Australia. URL: http://rafnetworkshop.wixsite.com/2017 Submission Due: January 15, 2017 [Please note that the authors of selected papers will be invited to submit the extended versions to IF journals. Details can be found on RAFNET URL provided.] 2. IEEE INFOCOM; SCAN Workshop 2017, Atlanta, USA. URL: http://infocom2017.ieee-infocom.org/workshop/scan-advances-software-defined-and-context%E2%80%90aware-cognitive-networks Submission Due: January 10, 2017 3. IWCMC 2017, Vehicular Communications Symposium, Valencia, Spain. URL: http://iwcmc.org/2017/symposia/ Submission Due: January 10, 2017 4. IEEE Access Journal Special Section on Future Networks. URL: http://ieeeaccess.ieee.org/special-sections/future-networks-architectures-protocols-applications/ Submission Due: January 31, 2017 5. WCMC Journal, SI on Wireless Networking Technologies for Smart Cities URL: https://www.hindawi.com/journals/wcmc/si/972967/cfp/ Submission Due: March 31, 2017 6. Elsevier FGCS Journal: Special Issue on Future Networking Research Plethora for Smart Cities URL: http://www.journals.elsevier.com/future-generation-computer-systems/call-for-papers/special-issue-on-future-networking-research-plethora-for-sma Submission Due: June 01, 2017 7. IEEE Internet Initiative e-News Letter URL: http://internetinitiative.ieee.org/newsletter Submission Type: 800-1200 words related to the subject. How Future Internet can help in making policies!! For details, check the URL. Looking forward for interesting contributions and in case of any query, feel free to contact me. ?Regards? *Syed Hassan Ahmed* PhD Scholar, Kyungpook National University [image: photo] Mobile: +82-10-9883-0786 Skype: hassan.shah2728 <#> Website: https://sites.google.com/site/shahmedknu/ Member: IEEE ComSoc, IEEE Young Professionals, ACM Get a signature like this: Click here! -------------- next part -------------- An HTML attachment was scrubbed... URL: