March 2013 - Ibr-dtn - lists.ibr.cs.tu-bs.de

Re: [ibr-dtn] logfile
by Johannes Morgenroth 30 Mar '13

30 Mar '13

Hello. There is no logging of the bundle size in the logs. The only information you can get is the calculated transmission rate using the debug option with level 15 or greater. That would look like this... Sat Mar 30 10:31:23 2013 DEBUG.5 TCPConnection: transfer finished after 0.25 ms with 8301.88 kb/s Since the project is open source, feel free to add any logging you need. Kind regards Johannes Morgenroth Am 29.03.2013 14:13, schrieb Bruno Tavares: > Thank you very much for your reply. > > In this logfile is there any information about the size of the bundles > sent? the size of the fragments? Or is there some way to get this > information ? > > Thank you > > Bruno Tavares

1 0

Multicast Broadcast
by justin＠gardettoengineering.com 29 Mar '13

29 Mar '13

1 0

logfile
by Bruno Tavares 28 Mar '13

28 Mar '13

Hello all I have some doubts about the information displayed in the log file. Can anyone tell me the meaning of the numbers between the brackets in this example? Wed Mar 27 10:07:51 2013 NOTICE: BundleReceivedEvent: Bundle received [417714860.14.16920458] dtn://drivein253/ee -> dtn://drivein250/ee from dtn://drivein101 Wed Mar 27 10:07:52 2013 NOTICE: QueueBundleEvent: New bundle queued [417714860.14.16920458] dtn://drivein253/ee Wed Mar 27 10:07:52 2013 NOTICE: BundleEvent: bundle [417714860.14.16920458] dtn://drivein253/ee received What does this mean? [417714860.14.16920458] Thank you all Best regards, Bruno Tavares

2 1

Re: [ibr-dtn] Ibr-dtn Digest, Vol 32, Issue 9
by Bruno Tavares 26 Mar '13

26 Mar '13

I have installed from source last week so I suppose it is the latest version. But I made confusion. After several tests I think it is making the reactive fragmentation. I sent on file that took something like 90 seconds to be delivered. Then I sent it again but stopped the transmission at 70 seconds. Then I connected again the two nodes and the message was transmitted for 20 seconds more or less. So I suppose it is making the reactive fragmentation right? One more question that I have is about the log file. Is there any way to get the information about how many quantity of information was transmitted from one node to other when the bundle is not sent completely ? Is there any way to have access to this kind of information ? Thank you very much for you answer and for your time. Best regards, Bruno Tavares ________________________________________ De: ibr-dtn-bounces(a)ibr.cs.tu-bs.de [ibr-dtn-bounces(a)ibr.cs.tu-bs.de] em nome de ibr-dtn-request(a)ibr.cs.tu-bs.de [ibr-dtn-request(a)ibr.cs.tu-bs.de] Enviado: terça-feira, 26 de Março de 2013 11:00 Para: ibr-dtn(a)ibr.cs.tu-bs.de Assunto: Ibr-dtn Digest, Vol 32, Issue 9 Send Ibr-dtn mailing list submissions to ibr-dtn(a)ibr.cs.tu-bs.de To subscribe or unsubscribe via the World Wide Web, visit https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn or, via email, send a message with subject or body 'help' to ibr-dtn-request(a)ibr.cs.tu-bs.de You can reach the person managing the list at ibr-dtn-owner(a)ibr.cs.tu-bs.de When replying, please edit your Subject line so it is more specific than "Re: Contents of Ibr-dtn digest..." Today's Topics: 1. interruption of transmission (Bruno Tavares) 2. Re: interruption of transmission (Johannes Morgenroth) ---------------------------------------------------------------------- Message: 1 Date: Mon, 25 Mar 2013 11:58:22 +0000 From: Bruno Tavares <bmpt(a)ua.pt> Subject: [ibr-dtn] interruption of transmission To: "ibr-dtn(a)ibr.cs.tu-bs.de" <ibr-dtn(a)ibr.cs.tu-bs.de> Message-ID: <90494CFA1222DC45A1855F79747890D82078FF06(a)DBXPRD0210MB358.eurprd02.prod.outlook.com> Content-Type: text/plain; charset="us-ascii" Hello I'm making some tests with ibr-dtn. I noticed that when I'm transferring a file from one node to another one, if during the transmission I manage to interrupt it, the receiving node does not receive any bundle. It should receive some fragmented bundle right? I putted in both configuration files : fragmentation=yes Best regards Bruno Tavares ------------------------------ Message: 2 Date: Mon, 25 Mar 2013 14:23:25 +0100 From: Johannes Morgenroth <morgenroth(a)ibr.cs.tu-bs.de> Subject: Re: [ibr-dtn] interruption of transmission To: ibr-dtn(a)ibr.cs.tu-bs.de Message-ID: <1364217805.764.8.camel@syrah> Content-Type: text/plain; charset="ISO-8859-15" Well, that depends on the version you use. Reactive fragmentation support is a very new feature and if I do remember right, not supported by the latest release. Moreover, a fragment bundle never gets delivered to an application. Only if the bundle is reassembled the bundle will forwarded to the application. Which version are you testing? Kind regards, Johannes Am Montag, den 25.03.2013, 11:58 +0000 schrieb Bruno Tavares: > Hello > > I'm making some tests with ibr-dtn. I noticed that when I'm transferring a file from one node to another one, if during the transmission I manage to interrupt it, the receiving node does not receive any bundle. It should receive some fragmented bundle right? > > I putted in both configuration files : fragmentation=yes > > Best regards > > Bruno Tavares > -- > !! This message is brought to you via the `ibr-dtn' mailing list. > !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust > !! your settings, send a mail message to <ibr-dtn-request(a)ibr.cs.tu-bs.de> > !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn. ------------------------------ -- !! This message is brought to you via the `ibr-dtn' mailing list. !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust !! your settings, send a mail message to <ibr-dtn-request(a)ibr.cs.tu-bs.de> !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn. End of Ibr-dtn Digest, Vol 32, Issue 9 **************************************

1 0

interruption of transmission
by Bruno Tavares 25 Mar '13

25 Mar '13

Hello I'm making some tests with ibr-dtn. I noticed that when I'm transferring a file from one node to another one, if during the transmission I manage to interrupt it, the receiving node does not receive any bundle. It should receive some fragmented bundle right? I putted in both configuration files : fragmentation=yes Best regards Bruno Tavares

2 1

[patch] abort if something fails during configuration
by Philipp Hagemeister 14 Mar '13

14 Mar '13

Hi list, I haven't found an official contribution guideline, so please find attached a trivial patch that aborts autogen.sh if a step fails. By the way, why are all the configuration files starting with #!/bin/bash instead of #!/bin/sh ? I haven't found a single bashism so far, just outdated `command` instead of $(command). Cheers, Philipp

1 0

Doubts about custody
by Alex López 14 Mar '13

14 Mar '13

Hello guys, I hope someone can help me. I've implemented some logging to the ibr-dtn source code to know when the priority is assigned and what priority the bundles have in the queue. After some research I found that: 1.- when you send a file with dtnsend -p 1, you save the priority in the bundle with b.setpriority(p) which assigns the correct bits to the PRIORITY_BIT1 and 2. 2.- this bundle is saved in a priority queue which checks the PRIORITY_BITS and order the bundles according to this. (here I have a logger that tells me with which priority the bundle was saved. 3.- after this, the bundle is waiting until the destination node is detected. The problem I've found happens when I have an intermediate node. All the routers are set with epidemic routing and the R1 can only see R2 R1 -------- R2 --------- R3 (destination) Now, happens this: 1.- I use dtnsend -p 1 in R1 to send the file to R3. 2.- this bundle is saved in a priority queue which checks the PRIORITY_BITS and order the bundles according to this. (here I have a logger that tells me that the priority of the saved bundle was 1, perfect) 3.- Now i think that tries to process the epidemic routing Thu Jan 31 11:48:32 2013 DEBUG.5: processing static routing task >> ProcessBundleTask: [412948112.1] dtn://node1.dtn/XdkTGFPAfiJdlFB > > Thu Jan 31 11:48:32 2013 DEBUG.50: search static route for [412948112.1] >> dtn://node1.dtn/XdkTGFPAfiJdlFB > > Thu Jan 31 11:48:32 2013 DEBUG.5: processing neighbor routing task >> ProcessBundleTask: [412948112.1] dtn://node1.dtn/XdkTGFPAfiJdlFB > > Thu Jan 31 11:48:32 2013 DEBUG.5: processing neighbor routing task >> SearchNextBundleTask: dtn://node2.dtn > > Thu Jan 31 11:48:32 2013 DEBUG.50: processing epidemic task >> SearchNextBundleTask: dtn://node2.dtn > > Thu Jan 31 11:48:32 2013 DEBUG.5: got 0 items to transfer to >> dtn://node2.dtn > > Thu Jan 31 11:48:32 2013 DEBUG.40: search some bundles not known by >> dtn://node2.dtn > > NOTICE: BundleGeneratedEvent: Bundle generated [412948112.4] > dtn://node1.dtn/routing -> dtn://node2.dtn/routing The problem is that when it generates this new bundle to send it to R2, my priority logger says that the priority of the bundle saved is different from the one that it saved previously, so my questions are: Why do you need to generate a new bundle to send to the intermediate nodes instead of copying the one that is already stored? Why the priority is lost? I want to implement that the intermediate nodes save the custody bundles keeping the priority of the original bundle, so the QoS can be keeped too. It makes no sense that the priority of a bundle only affects the local queue from where it is sent. So where in the source code is the class that generates this bundle from the original one to send it as a custory for the R2? Thanks in advance!

2 2

Re: [ibr-dtn] Dht ibrdtn package
by Sebastian Schildt 12 Mar '13

12 Mar '13

Hello, no IBR-DTN can not parse the UCI configuration syntax, so if you use it with the -c option possibly the Parser will bail out early, and default values will be used. On open WRT (and only there), the /etc/init.d/ibrdtn script translates the luci syntax to a valid IBR-DTN configuration file, puts it into /tmp and then starts the daemon. So when using the init script, modify the config in /et/config, when you want to start ibrdtn manually use a valid IBR-DTN config (for example from the debian version or the example in the sourcecode archive , or you can grab the one in /tmp copy it to another place and modify it) MfG Sebastian Am 11.03.2013 um 20:50 schrieb Dimitrios Giotas <dimigiotas(a)gmail.com>: > Ok so I stumbled across a rather weird fact. > > First of all, is it right to call the daemon on the openwrt with the dtnd -c /etc/config/ibrdtn command, > or is it better to just use the /etc/init.d/ibrdtn <...> ? > > Anyway, I use the 1st method along with the script to resolve the issue of which conf file is loaded. > So, when I call the daemon with the -c parameter I get that the local node name is dtn://OpenWrt . > However when I see the /tmp/ibrdtn.config the local node name, as well as some other parameters > I'm currently trying out, are different. > > I've tried restarting the init.d script but to no avail. > Also in the LuCi I see many /usr/sbin/dtnd -c /tmp/ibrdtn.config processes. Is that normal? > > Ideas are welcome. > > Regards > > On Mon, Mar 11, 2013 at 3:12 PM, Sebastian Schildt <schildt(a)ibr.cs.tu-bs.de> wrote: > > On our debian packages there is also a > > /etc/init.d/ibrdtn > > that can be used to start, stop or restart. However thee is no "tmeporary" confug file that is generated. Instead it diriectly uses the config in > /etc/ibrdtn.conf or /etc/ibrdtn/ibrdtnd.conf (I am not usre, depending on version, we changd that osmetime. When IBR-DTN is running you can use ps -ax to check, with which config is has been loaded) > > > Sebastian > > > Am 10.03.2013 um 23:03 schrieb Dimitrios Giotas <dimigiotas(a)gmail.com>: > >> One last question regarding the matter.Just out of curiosity, >> is there something similar for a daemon running on a PC? >> I mean the restart and the tmp config file. >> >> Thank you, >> >> Regards >> >> On Thu, Mar 7, 2013 at 2:45 PM, Dimitrios Giotas <dimigiotas(a)gmail.com> wrote: >> Many thanks Sebastian. The restart magic solved the problems altogether. Awesome. >> >> >> On Thu, Mar 7, 2013 at 1:46 PM, Sebastian Schildt <schildt(a)ibr.cs.tu-bs.de> wrote: >> >> Just a blind guess: For the OpenWRT uci magic to work you need to start ibrdtn with the supplied start script,i.e. >> >> /etc/init.d/ibrdtn restart >> >> this will convert the uci configuration to a ibrdtn configfile in /tmp and then start the daemon >> >> MfG >> >> Sebastian >> >> >> Am 07.03.2013 um 11:53 schrieb Dimitrios Giotas <dimigiotas(a)gmail.com>: >> >>> Hello, >>> >>> >>> I am having some troubles with my router which is too running openwrt. >>> When I run the daemon with a configuration file I don't get anything in the /tmp/ibrdtn.conf >>> It doesn't exist. What might be the cause of that? >>> I have a few more issues as well, which I will post on future mails. >>> Thank you. >>> >>> On Wed, Mar 6, 2013 at 4:45 PM, Sebastian Schildt <schildt(a)ibr.cs.tu-bs.de> wrote: >>> >>> >>> Regarding OpenWRT the "uci" systemis used for configuration, You can find that configuration file in /etc/config . This will be converted to a config file in ibrdtn syntax when the dameon is started. So if you are unsure, you can check /tmp/ibrdtn.config while the daemon is running to see which configuration is really used by an OpenWRT node. >>> >>> The DHT module allows to look up EIDs of other DTN nodes, if they are also suing DHT and have a direct internet connection. See this paper for details: >>> >>> Schildt, Sebastian, Lorentzen, Till, Morgenroth, Johannes, Pöttner, Wolf-Bastian and Wolf, Lars: Free-riding the BitTorrent DHT to improve DTN connectivity, in Proceedings of the seventh ACM international workshop on Challenged networks, CHANTS '12, New York, NY, USA, Seite 9-16, ACM, 2012 >>> >>> http://dx.doi.org/10.1145%2F2348616.2348619 >>> >>> >>> Sebastian >>> >>> Am 06.03.2013 um 13:26 schrieb Francisco Cuesta <ndarkness(a)gmail.com>: >>> >>>> Hello, >>>> >>>> I'm struggling with a weird fact, I can not ping from one dtn node(C) to another dtn node (A), and I cannot understand completely why. I have set the right clocks and the configuration but first I have to ping (dtnping) from A to C in order that C can ping A, if I start directly pinging from C to A the nodes cannot see each other. Indeed, looking at the log of C there aren't other dtn nodes present... The discovery_announce are set on C but I don't know how to check whether it's enabled on A (this node uses openwrt). Any hint of what's going on? >>>> >>>> Furthermore, I would like to know the functionality of DHT package, what is it used for? >>>> >>>> Thanks in advance, >>>> >>>> Ragards! >>>> -- >>>> !! This message is brought to you via the `ibr-dtn' mailing list. >>>> !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust >>>> !! your settings, send a mail message to <ibr-dtn-request(a)ibr.cs.tu-bs.de> >>>> !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn. >>> >>> >>> -- >>> !! This message is brought to you via the `ibr-dtn' mailing list. >>> !! Please do not reply to this message to unsubscribe. To unsubscribe or adjust >>> !! your settings, send a mail message to <ibr-dtn-request(a)ibr.cs.tu-bs.de> >>> !! or look at https://www.ibr.cs.tu-bs.de/mailman/listinfo/ibr-dtn. >>> >> >> >> > >

1 0

Re: [ibr-dtn] Android Apps Neighbor Discovery
by justin＠gardettoengineering.com 11 Mar '13

11 Mar '13

1 0

Priority of DTN
by Alex López 11 Mar '13

11 Mar '13

Hello all, I've been studying how the priority parameter works with IBRDTN in a MANET that I have with 4 openwrt routers. My lab setup is the following: Source 1 --\ \ \ Node 3 -------> End 4 / / Source 2---/ I've observed that if I try to send a file with a high priority from source 1 to End 4, and a file with low priority from source 2 to End 4, the file with low priority is received in End 4 before the file with high priority. Reading the RFC 4838, it says the following: Applications specify their requested priority class and data lifetime (see > below) for each ADU they send. This information, coupled with policy > applied at DTN nodes that select how messages are forwarded and which > routing algorithms are in use, affects the overall likelihood and > timeliness of ADU delivery. > > The priority class of a bundle is only required to relate to other bundles > from the same source. This means that a high priority bundle from one > source may not be delivered faster (or with some other superior quality of > service) than a medium priority bundle from a different source. It does > mean that a high priority bundle from one source will be handled > preferentially to a lower priority bundle sent from the same source. > > Depending on a particular DTN node’s forwarding/scheduling policy, > priority may or may not be enforced across different sources. That is, in > some DTN nodes, expedited bundles might always be sent prior to any bulk > bundles, irrespective of source. Many variations are possible. If I understand well, the priority marking that is done in the Sources, it's lost in the intermediate Node 3, right? I cannot imagine why DTN works like this? It is really an stupid QoS no? Whats the reason behind this implementation? There is any way that the Node 3 recognizes the priority of the bundles that it has in custody? And from what source they are from? Maybe it is because DTN is aiming for a only bidirectional communication between only 1 source and 1 end and multiple nodes? So a MANET could not be the best option to set a DTN implementation? Thanks in advance.

2 1