Received: from fg-out-1718.google.com (fg-out-1718.google.com [72.14.220.156]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mASNTaQS005326 for ; Fri, 28 Nov 2008 15:29:37 -0800 Received: by fg-out-1718.google.com with SMTP id e21so1070162fga.6 for ; Fri, 28 Nov 2008 15:05:40 -0800 (PST) Received: by 10.187.247.15 with SMTP id z15mr1989300far.80.1227913539902; Fri, 28 Nov 2008 15:05:39 -0800 (PST) Received: by 10.187.188.10 with HTTP; Fri, 28 Nov 2008 15:05:39 -0800 (PST) Message-ID: <4bfc0da50811281505v61e29158w852d25449305a58d@mail.gmail.com> Date: Fri, 28 Nov 2008 23:05:39 +0000 From: "Ricardo Martins" To: "Andy Scott" In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: Cc: dtn-users@maillists.intel-research.net Subject: Re: [dtn-users] Should I Expect A Failure? X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Nov 2008 23:29:37 -0000 Hi, This and other issues are already fixed in the latest revisions of DTN's mercurial repository. Regards, Ricardo Martins 2008/11/28 Andy Scott : > Hi All > > Just downloaded the source (plus required other packages) for v2.6.0 > to Ubuntu and I am seeing a build error in SequenceID.cc in the > include of oasys/util/SafeRange.h > > size_t is not defined which is the root problem. > > Made the 1-line fix to get it past this issue and it builds no problems. > > I have now "downloaded" the bleeding edge stuff to try out - which I > guess has this issue well and truly resolved as it seems to be > building OK as I type this. > > My question is: is the error I encountered known or is it indicative > of a bust install of G++ &c? > > (I am also searching the posting archive - but my connection is slow > and my find kung fu weak) > > Thanks for any replies and look forward to getting into this stuff > some more it looks so cool. > > Andy > -- > Brain upgrade required: a working hypothalamus > _______________________________________________ > dtn-users mailing list > dtn-users@maillists.intel-research.net > http://maillists.intel-research.net/mailman/listinfo/dtn-users > Received: from yx-out-1718.google.com (yx-out-1718.google.com [74.125.44.152]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mASKtIOg030790 for ; Fri, 28 Nov 2008 12:55:18 -0800 Received: by yx-out-1718.google.com with SMTP id 4so603344yxp.34 for ; Fri, 28 Nov 2008 12:31:24 -0800 (PST) Received: by 10.65.54.9 with SMTP id g9mr9040432qbk.28.1227904284032; Fri, 28 Nov 2008 12:31:24 -0800 (PST) Received: by 10.64.178.14 with HTTP; Fri, 28 Nov 2008 12:31:23 -0800 (PST) Message-ID: Date: Fri, 28 Nov 2008 20:31:23 +0000 From: "Andy Scott" To: dtn-users@maillists.intel-research.net MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Subject: [dtn-users] Should I Expect A Failure? X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Nov 2008 20:55:18 -0000 Hi All Just downloaded the source (plus required other packages) for v2.6.0 to Ubuntu and I am seeing a build error in SequenceID.cc in the include of oasys/util/SafeRange.h size_t is not defined which is the root problem. Made the 1-line fix to get it past this issue and it builds no problems. I have now "downloaded" the bleeding edge stuff to try out - which I guess has this issue well and truly resolved as it seems to be building OK as I type this. My question is: is the error I encountered known or is it indicative of a bust install of G++ &c? (I am also searching the posting archive - but my connection is slow and my find kung fu weak) Thanks for any replies and look forward to getting into this stuff some more it looks so cool. Andy -- Brain upgrade required: a working hypothalamus Received: from asmtpout016.mac.com (asmtpout016.mac.com [17.148.16.91]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mASHmaLx022352 for ; Fri, 28 Nov 2008 09:48:36 -0800 MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Received: from [192.168.0.3] (78-33-82-89.static.enta.net [78.33.82.89]) by asmtp016.mac.com (Sun Java(tm) System Messaging Server 6.3-7.03 (built Aug 7 2008; 32bit)) with ESMTPSA id <0KB200J8T0D6NZ70@asmtp016.mac.com> for dtn-users@mailman.dtnrg.org; Fri, 28 Nov 2008 09:24:46 -0800 (PST) Message-id: <0F442FE1-D886-4B51-AAA0-3E5CA4EBB011@mac.com> From: Darren Long To: dtn-users mailing list In-reply-to: <59e8babf0811280823g45dcc56cy886d2acfef3b15d7@mail.gmail.com> Date: Fri, 28 Nov 2008 17:24:42 +0000 References: <200811271141.50178.peter@peter-b.co.uk> <59e8babf0811280823g45dcc56cy886d2acfef3b15d7@mail.gmail.com> X-Mailer: Apple Mail (2.929.2) Subject: Re: [dtn-users] Various questions about the DTN reference implementation X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Nov 2008 17:48:36 -0000 On 28 Nov 2008, at 16:23, Michael Demmer wrote: > 3) Is it possible to use multiple routing systems at once with DTN2? > For > instance, consider a large group of devices in an ad-hoc network, > but with > one or more of their number connected to fixed WANs. It would make > sense to > use something like LSR to route within the ad-hoc network, but the > 'boundary > nodes' would need to have LSR *and* possibly some static or name- > lookup-based > routing system for the WAN at the same time. Glancing through the > sources > gave the strong impression that only one routing approach could be > used at a > time. > > Unfortunately not at this time. This was something we've considered > for a while but never got around to implementing properly. I'm pretty sure it is possible (or at least was possible) to add static routes when using DTLSR. Darren Received: from an-out-0708.google.com (an-out-0708.google.com [209.85.132.248]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mASGlEbu019607 for ; Fri, 28 Nov 2008 08:47:14 -0800 Received: by an-out-0708.google.com with SMTP id d40so506334and.34 for ; Fri, 28 Nov 2008 08:23:25 -0800 (PST) Received: by 10.100.178.9 with SMTP id a9mr4226950anf.59.1227889405033; Fri, 28 Nov 2008 08:23:25 -0800 (PST) Received: by 10.100.95.9 with HTTP; Fri, 28 Nov 2008 08:23:25 -0800 (PST) Message-ID: <59e8babf0811280823g45dcc56cy886d2acfef3b15d7@mail.gmail.com> Date: Fri, 28 Nov 2008 08:23:25 -0800 From: "Michael Demmer" Sender: demmer@gmail.com To: "Peter TB Brett" In-Reply-To: <200811271141.50178.peter@peter-b.co.uk> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_64730_27869525.1227889405034" References: <200811271141.50178.peter@peter-b.co.uk> X-Google-Sender-Auth: 588af65bbb4330b8 Cc: dtn-users@maillists.intel-research.net Subject: Re: [dtn-users] Various questions about the DTN reference implementation X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Nov 2008 16:47:15 -0000 ------=_Part_64730_27869525.1227889405034 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline > > Hi all, > > I'm just getting started with the DTN2 impl., and have a few questions > arising > from reading the documentation. I'm more than happy to collate any answers > and wikify them for someone to add to dtnrg.org. > > 1) I note that the implementation supports a number of different endpoint > schema, but I couldn't find any documentation as to the purpose of each. Unfortunately that's because none exists. Most of my work has been with the standard 'dtn:' scheme though. > 2) It would be useful to be able to run multiple (3-10) agents on a single > system in order to test DTN applications in an automated fashion in a > controlled environment. I note that this is entirely possible (just by > setting different ports to listen on for each agent), but was wondering if > anyone was willing to share tips or even management scripts for setting up > and managing a set of dtnd instances. We've written a tcl based infrastructure that does a bunch of this. The basic model is that you write a tcl script that defines your test and then can interact with running daemons. It also has a framework for writing a topology that can either be implemented by running a bunch of daemons on one host (listening on diff't ports) or using a set of test machines. You can get the base infrastructure here: hg clone http://dtn.sourceforge.net/hg/test-utils DTN-specific test utilities are in DTN2/test-utils and example test scripts are in DTN2/test. You need to clone the test-utils repository in the same directory as DTN2. Then as an example, cd into DTN2 and run: test-utils/run-dtn.tcl test/send-one-bundle.tcl That should get you going with one functional example. 3) Is it possible to use multiple routing systems at once with DTN2? For > instance, consider a large group of devices in an ad-hoc network, but with > one or more of their number connected to fixed WANs. It would make sense to > use something like LSR to route within the ad-hoc network, but the > 'boundary > nodes' would need to have LSR *and* possibly some static or > name-lookup-based > routing system for the WAN at the same time. Glancing through the sources > gave the strong impression that only one routing approach could be used at > a > time. Unfortunately not at this time. This was something we've considered for a while but never got around to implementing properly. > 4) Does anyone have any example code for a simple external convergence > layer > that they would be willing to share, please? I don't. 5) Would the maintainers be willing to apply a patch for libdtnapi which > adds > API for DTN applications to specify the dtnd address/port other than by > environment variable? Since I'm no longer a maintainer, I can't say specifically, but my guess would be that's no problem as long as it remains backwards-compatible. -m ------=_Part_64730_27869525.1227889405034 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline
Hi all,

I'm just getting started with the DTN2 impl., and have a few questions arising
from reading the documentation. I'm more than happy to collate any answers
and wikify them for someone to add to dtnrg.org.

1) I note that the implementation supports a number of different endpoint
schema, but I couldn't find any documentation as to the purpose of each.

Unfortunately that's because none exists. Most of my work has been with the standard 'dtn:' scheme though.
 
2) It would be useful to be able to run multiple (3-10) agents on a single
system in order to test DTN applications in an automated fashion in a
controlled environment. I note that this is entirely possible (just by
setting different ports to listen on for each agent), but was wondering if
anyone was willing to share tips or even management scripts for setting up
and managing a set of dtnd instances.

We've written a tcl based infrastructure that does a bunch of this. The basic model is that you write a tcl script that defines your test and then can interact with running daemons. It also has a framework for writing a topology that can either be implemented by running a bunch of daemons on one host (listening on diff't ports) or using a set of test machines.

You can get the base infrastructure here:
hg clone http://dtn.sourceforge.net/hg/test-utils

DTN-specific test utilities are in DTN2/test-utils and example test scripts are in DTN2/test.

You need to clone the test-utils repository in the same directory as DTN2.

Then as an example, cd into DTN2 and run:

test-utils/run-dtn.tcl test/send-one-bundle.tcl

That should get you going with one functional example.

3) Is it possible to use multiple routing systems at once with DTN2? For
instance, consider a large group of devices in an ad-hoc network, but with
one or more of their number connected to fixed WANs. It would make sense to
use something like LSR to route within the ad-hoc network, but the 'boundary
nodes' would need to have LSR *and* possibly some static or name-lookup-based
routing system for the WAN at the same time. Glancing through the sources
gave the strong impression that only one routing approach could be used at a
time.

Unfortunately not at this time. This was something we've considered for a while but never got around to implementing properly.
 
4) Does anyone have any example code for a simple external convergence layer
that they would be willing to share, please?

I don't.

5) Would the maintainers be willing to apply a patch for libdtnapi which adds
API for DTN applications to specify the dtnd address/port other than by
environment variable?

Since I'm no longer a maintainer, I can't say specifically, but my guess would be that's no problem as long as it remains backwards-compatible.

-m

------=_Part_64730_27869525.1227889405034-- Received: from ppsw-5.csi.cam.ac.uk (ppsw-5.csi.cam.ac.uk [131.111.8.135]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mARC58Ol008186 for ; Thu, 27 Nov 2008 04:05:08 -0800 X-Cam-AntiVirus: no malware found X-Cam-SpamDetails: not scanned X-Cam-ScannerInfo: http://www.cam.ac.uk/cs/email/scanner/ Received: from ptbb2b.girton.cam.ac.uk ([128.232.240.168]:60223) by ppsw-5.csi.cam.ac.uk (ppsw.cam.ac.uk [131.111.8.135]:25) with esmtp id 1L5fFm-0004Cv-Hr (Exim 4.70) for dtn-users@maillists.intel-research.net (return-path ); Thu, 27 Nov 2008 11:41:50 +0000 Received: by ptbb2b.girton.cam.ac.uk (Postfix, from userid 500) id 7093078062; Thu, 27 Nov 2008 11:41:50 +0000 (GMT) From: Peter TB Brett Organization: Integral Informatics Ltd To: dtn-users@maillists.intel-research.net Date: Thu, 27 Nov 2008 11:41:47 +0000 User-Agent: KMail/1.9.9 X-Face: $N}{5?\D>xr}zi]BChHp)X[G,\Z"lc+zdwps1*EiY,=?utf-8?q?=7B!=26t+x49=5DlIKpn=60*U4cV=7E=5E-4=7ClNw=0A=092T9+3y?=@-cV}7TSB[F8'Cglfe#b6BrI.^AI-7|$; O@d6Ons!`h0\{-I/U6D]8v=P MIME-Version: 1.0 Content-Type: multipart/signed; boundary="nextPart1417671.9uZFBVeFHs"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit Message-Id: <200811271141.50178.peter@peter-b.co.uk> Subject: [dtn-users] Various questions about the DTN reference implementation X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 27 Nov 2008 12:05:09 -0000 --nextPart1417671.9uZFBVeFHs Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Hi all, I'm just getting started with the DTN2 impl., and have a few questions aris= ing=20 from reading the documentation. I'm more than happy to collate any answers= =20 and wikify them for someone to add to dtnrg.org. 1) I note that the implementation supports a number of different endpoint=20 schema, but I couldn't find any documentation as to the purpose of each. 2) It would be useful to be able to run multiple (3-10) agents on a single= =20 system in order to test DTN applications in an automated fashion in a=20 controlled environment. I note that this is entirely possible (just by=20 setting different ports to listen on for each agent), but was wondering if= =20 anyone was willing to share tips or even management scripts for setting up= =20 and managing a set of dtnd instances. 3) Is it possible to use multiple routing systems at once with DTN2? For=20 instance, consider a large group of devices in an ad-hoc network, but with= =20 one or more of their number connected to fixed WANs. It would make sense to= =20 use something like LSR to route within the ad-hoc network, but the 'boundar= y=20 nodes' would need to have LSR *and* possibly some static or name-lookup-bas= ed=20 routing system for the WAN at the same time. Glancing through the sources=20 gave the strong impression that only one routing approach could be used at = a=20 time. 4) Does anyone have any example code for a simple external convergence laye= r=20 that they would be willing to share, please? 5) Would the maintainers be willing to apply a patch for libdtnapi which ad= ds=20 API for DTN applications to specify the dtnd address/port other than by=20 environment variable? I'm sure there will be other questions later. I'm perfectly willing to=20 take, "You dummy! This is all clearly documented at !" as an answer. ;) Peter =2D-=20 Peter Brett Cambridge University Engineering Department --nextPart1417671.9uZFBVeFHs Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) iD8DBQBJLod7Z7Gbq7g7vpoRAkyxAJkB7EPSTUsogXPU7VkRNbRqvaH2MwCfRub6 BURnrWu5O940XEGgtJA8Up4= =W++l -----END PGP SIGNATURE----- --nextPart1417671.9uZFBVeFHs-- Received: from mail.jpl.nasa.gov (sentrion2.jpl.nasa.gov [128.149.139.106]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mAMKSeXg032341 for ; Sat, 22 Nov 2008 12:28:41 -0800 Received: from mail.jpl.nasa.gov (altvirehtstap02.jpl.nasa.gov [128.149.137.73]) by mail.jpl.nasa.gov (Switch-3.3.2mp/Switch-3.3.2mp) with ESMTP id mAMK7PIQ006573 (using TLSv1/SSLv3 with cipher RC4-MD5 (128 bits) verified FAIL) for ; Sat, 22 Nov 2008 20:07:25 GMT Received: from ALTPHYEMBEVSP10.RES.AD.JPL ([128.149.137.81]) by ALTVIREHTSTAP02.RES.AD.JPL ([128.149.137.73]) with mapi; Sat, 22 Nov 2008 12:08:29 -0800 From: "Burleigh, Scott C" To: dtn-users mailing list Date: Sat, 22 Nov 2008 12:07:30 -0800 Thread-Topic: [dtn-users] ION and DTN2 Thread-Index: AclM2a9MDfOZtO0MT9e4/2rfhsWSTwAA6F/w Message-ID: References: <5B4DCF3B-2C22-431B-9EDA-E612C24E37AD@mac.com> <6BB13DB1-79D5-4269-A088-07B79B37B30D@mac.com> In-Reply-To: <6BB13DB1-79D5-4269-A088-07B79B37B30D@mac.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 X-Source-IP: altvirehtstap02.jpl.nasa.gov [128.149.137.73] X-Source-Sender: scott.c.burleigh@jpl.nasa.gov X-AUTH: Authorized Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by maillists.intel-research.net id mAMKSeXg032341 Subject: Re: [dtn-users] ION and DTN2 X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 22 Nov 2008 20:28:41 -0000 Right, you'll need an "a scheme" line in your bprc file for "dtn", and you'll want to run dtnadmin at startup to configure your routes -- for now, ION only supports static routing for "dtn" endpoints. (Actually, you can omit the "ipn" scheme altogether if you're only going to be talking to "dtn" endpoints.) The guys at Ohio U. can give you a lot better help on your make problem than I could. Scott > -----Original Message----- > From: Darren Long [mailto:darren.long@mac.com] > Sent: Saturday, November 22, 2008 11:35 AM > To: Burleigh, Scott C > Cc: dtn-users mailing list > Subject: Re: [dtn-users] ION and DTN2 > > Thanks Scott, I'd pretty much drawn the same conclusion myself just > now after watching them trying to talk TCP. > > I guess I just need to add a an induct and outducts for udpcli/udpclo, > and a section in the for dtnadmin for the schema and some routes. I'm > a bit unsure about adding contacts to the ipnadmin section, and wonder > if using dtn schema addresses will work as well as ipn ones. > > I guess I'll find out though. Oh, I'm also having a problem getting > it to run on my OSX 10.5 system, make test fails with shared memory > problems. it seems to think its not a MACOSX machine. > > Darren > > On 22 Nov 2008, at 19:02, Burleigh, Scott C wrote: > > > Hi, Darren. ION and DTN2 are indeed interoperable (or at least they > > were as of the November 2007 IETF), but not via the TCP-CL. ION > > does have a TCP convergence-layer adapter implemented, but it's much > > simpler than the DTN2 TCP-CL because it doesn't try to support > > reactive fragmentation; support for the standard TCP-CL has been > > implemented, but it's not part of the ION distribution yet. To get > > ION and DTN2 interoperating you currently have to use the UDP > > convergence layer support built into both implementations. > > > > In the ION distribution there should be a "configs" directory > > containing some sample ION configuration files. They'd need to be > > tweaked for what you want to do, though, as none of them use the raw > > UDP convergence layer. > > > > I don't know much about the DTN2 apps, but if there's something that > > just prints text messages that arrive in bundles destined for > > whatever endpoint it opens then it should be able to receive and > > print what bpsource sends. Alternatively, maybe you could open > > bpsink on a given endpoint and use a DTN2 bundle source app to send > > text to it? > > > > Scott > > > >>> -----Original Message----- > >>> From: dtn-users-bounces@maillists.intel-research.net [mailto:dtn- > >> users- > >>> bounces@maillists.intel-research.net] On Behalf Of Darren Long > >>> Sent: Saturday, November 22, 2008 10:22 AM > >>> To: dtn-users mailing list > >>> Subject: [dtn-users] ION and DTN2 > >>> > >>> Hi all, > >>> > >>> I've just started dabbling with ION and have been going through the > >>> tutorials. > >>> > >>> I note that it is (or at least was) shown to be interoperable with > >>> DTN2 with the TCP-CL, so I wondered if anyone has any tips or some > >> rc > >>> file extracts for setting up an ION to DTN2 gateway using TCP-CL > >>> assuming an on-demand or always on link into a dtnd and handling > the > >>> mix of ipn and dtn schemas. > >>> > >>> I also suspect that the apps in the two implementations won't talk, > >>> having just tried to use bpsource to send to a dtn:// EID, so I > >>> suspect my best bet is to route between two DTN2 based sub-nets > over > >>> an IPN sub-net, with a pair of gateways. > >>> > >>> Cheers, > >>> > >>> Darren > >>> _______________________________________________ > >>> dtn-users mailing list > >>> dtn-users@maillists.intel-research.net > >>> http://maillists.intel-research.net/mailman/listinfo/dtn-users > > > > _______________________________________________ > > dtn-users mailing list > > dtn-users@maillists.intel-research.net > > http://maillists.intel-research.net/mailman/listinfo/dtn-users Received: from asmtpout013.mac.com (asmtpout013.mac.com [17.148.16.88]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mAMJv47H030900 for ; Sat, 22 Nov 2008 11:57:05 -0800 MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Received: from [192.168.0.3] (78-33-82-89.static.enta.net [78.33.82.89]) by asmtp013.mac.com (Sun Java(tm) System Messaging Server 6.3-7.03 (built Aug 7 2008; 32bit)) with ESMTPSA id <0KAR00G162F4VW20@asmtp013.mac.com> for dtn-users@mailman.dtnrg.org; Sat, 22 Nov 2008 11:35:31 -0800 (PST) Message-id: <6BB13DB1-79D5-4269-A088-07B79B37B30D@mac.com> From: Darren Long To: "Burleigh, Scott C" In-reply-to: Date: Sat, 22 Nov 2008 19:35:28 +0000 References: <5B4DCF3B-2C22-431B-9EDA-E612C24E37AD@mac.com> X-Mailer: Apple Mail (2.929.2) Cc: dtn-users mailing list Subject: Re: [dtn-users] ION and DTN2 X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 22 Nov 2008 19:57:05 -0000 Thanks Scott, I'd pretty much drawn the same conclusion myself just now after watching them trying to talk TCP. I guess I just need to add a an induct and outducts for udpcli/udpclo, and a section in the for dtnadmin for the schema and some routes. I'm a bit unsure about adding contacts to the ipnadmin section, and wonder if using dtn schema addresses will work as well as ipn ones. I guess I'll find out though. Oh, I'm also having a problem getting it to run on my OSX 10.5 system, make test fails with shared memory problems. it seems to think its not a MACOSX machine. Darren On 22 Nov 2008, at 19:02, Burleigh, Scott C wrote: > Hi, Darren. ION and DTN2 are indeed interoperable (or at least they > were as of the November 2007 IETF), but not via the TCP-CL. ION > does have a TCP convergence-layer adapter implemented, but it's much > simpler than the DTN2 TCP-CL because it doesn't try to support > reactive fragmentation; support for the standard TCP-CL has been > implemented, but it's not part of the ION distribution yet. To get > ION and DTN2 interoperating you currently have to use the UDP > convergence layer support built into both implementations. > > In the ION distribution there should be a "configs" directory > containing some sample ION configuration files. They'd need to be > tweaked for what you want to do, though, as none of them use the raw > UDP convergence layer. > > I don't know much about the DTN2 apps, but if there's something that > just prints text messages that arrive in bundles destined for > whatever endpoint it opens then it should be able to receive and > print what bpsource sends. Alternatively, maybe you could open > bpsink on a given endpoint and use a DTN2 bundle source app to send > text to it? > > Scott > >>> -----Original Message----- >>> From: dtn-users-bounces@maillists.intel-research.net [mailto:dtn- >> users- >>> bounces@maillists.intel-research.net] On Behalf Of Darren Long >>> Sent: Saturday, November 22, 2008 10:22 AM >>> To: dtn-users mailing list >>> Subject: [dtn-users] ION and DTN2 >>> >>> Hi all, >>> >>> I've just started dabbling with ION and have been going through the >>> tutorials. >>> >>> I note that it is (or at least was) shown to be interoperable with >>> DTN2 with the TCP-CL, so I wondered if anyone has any tips or some >> rc >>> file extracts for setting up an ION to DTN2 gateway using TCP-CL >>> assuming an on-demand or always on link into a dtnd and handling the >>> mix of ipn and dtn schemas. >>> >>> I also suspect that the apps in the two implementations won't talk, >>> having just tried to use bpsource to send to a dtn:// EID, so I >>> suspect my best bet is to route between two DTN2 based sub-nets over >>> an IPN sub-net, with a pair of gateways. >>> >>> Cheers, >>> >>> Darren >>> _______________________________________________ >>> dtn-users mailing list >>> dtn-users@maillists.intel-research.net >>> http://maillists.intel-research.net/mailman/listinfo/dtn-users > > _______________________________________________ > dtn-users mailing list > dtn-users@maillists.intel-research.net > http://maillists.intel-research.net/mailman/listinfo/dtn-users Received: from mail.jpl.nasa.gov (sentrion1.jpl.nasa.gov [128.149.139.105]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mAMJO63u029360 for ; Sat, 22 Nov 2008 11:24:06 -0800 Received: from mail.jpl.nasa.gov (ums-smtp.jpl.nasa.gov [128.149.137.72]) by mail.jpl.nasa.gov (Switch-3.3.2mp/Switch-3.3.2mp) with ESMTP id mAMJ2qW0017806 (using TLSv1/SSLv3 with cipher RC4-MD5 (128 bits) verified FAIL) for ; Sat, 22 Nov 2008 19:02:52 GMT Received: from ALTPHYEMBEVSP10.RES.AD.JPL ([128.149.137.81]) by ALTVIREHTSTAP01.RES.AD.JPL ([128.149.137.72]) with mapi; Sat, 22 Nov 2008 11:03:56 -0800 From: "Burleigh, Scott C" To: dtn-users mailing list Date: Sat, 22 Nov 2008 11:02:50 -0800 Thread-Topic: [dtn-users] ION and DTN2 Thread-Index: AclM0B5+Kxlk5WGgRUGL0/ta4q0ePwAAPulgAADkzpA= Message-ID: References: <5B4DCF3B-2C22-431B-9EDA-E612C24E37AD@mac.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 X-Source-IP: ums-smtp.jpl.nasa.gov [128.149.137.72] X-Source-Sender: scott.c.burleigh@jpl.nasa.gov X-AUTH: Authorized Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by maillists.intel-research.net id mAMJO63u029360 Subject: Re: [dtn-users] ION and DTN2 X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 22 Nov 2008 19:24:06 -0000 Hi, Darren. ION and DTN2 are indeed interoperable (or at least they were as of the November 2007 IETF), but not via the TCP-CL. ION does have a TCP convergence-layer adapter implemented, but it's much simpler than the DTN2 TCP-CL because it doesn't try to support reactive fragmentation; support for the standard TCP-CL has been implemented, but it's not part of the ION distribution yet. To get ION and DTN2 interoperating you currently have to use the UDP convergence layer support built into both implementations. In the ION distribution there should be a "configs" directory containing some sample ION configuration files. They'd need to be tweaked for what you want to do, though, as none of them use the raw UDP convergence layer. I don't know much about the DTN2 apps, but if there's something that just prints text messages that arrive in bundles destined for whatever endpoint it opens then it should be able to receive and print what bpsource sends. Alternatively, maybe you could open bpsink on a given endpoint and use a DTN2 bundle source app to send text to it? Scott > > -----Original Message----- > > From: dtn-users-bounces@maillists.intel-research.net [mailto:dtn- > users- > > bounces@maillists.intel-research.net] On Behalf Of Darren Long > > Sent: Saturday, November 22, 2008 10:22 AM > > To: dtn-users mailing list > > Subject: [dtn-users] ION and DTN2 > > > > Hi all, > > > > I've just started dabbling with ION and have been going through the > > tutorials. > > > > I note that it is (or at least was) shown to be interoperable with > > DTN2 with the TCP-CL, so I wondered if anyone has any tips or some > rc > > file extracts for setting up an ION to DTN2 gateway using TCP-CL > > assuming an on-demand or always on link into a dtnd and handling the > > mix of ipn and dtn schemas. > > > > I also suspect that the apps in the two implementations won't talk, > > having just tried to use bpsource to send to a dtn:// EID, so I > > suspect my best bet is to route between two DTN2 based sub-nets over > > an IPN sub-net, with a pair of gateways. > > > > Cheers, > > > > Darren > > _______________________________________________ > > dtn-users mailing list > > dtn-users@maillists.intel-research.net > > http://maillists.intel-research.net/mailman/listinfo/dtn-users Received: from asmtpout012.mac.com (asmtpout012.mac.com [17.148.16.87]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mAMIhvej027538 for ; Sat, 22 Nov 2008 10:43:58 -0800 MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Received: from [192.168.0.3] (78-33-82-89.static.enta.net [78.33.82.89]) by asmtp012.mac.com (Sun Java(tm) System Messaging Server 6.3-7.03 (built Aug 7 2008; 32bit)) with ESMTPSA id <0KAQ004XZZ1H0U00@asmtp012.mac.com> for dtn-users@mailman.dtnrg.org; Sat, 22 Nov 2008 10:22:31 -0800 (PST) Message-id: <5B4DCF3B-2C22-431B-9EDA-E612C24E37AD@mac.com> From: Darren Long To: dtn-users mailing list Date: Sat, 22 Nov 2008 18:22:28 +0000 X-Mailer: Apple Mail (2.929.2) Subject: [dtn-users] ION and DTN2 X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 22 Nov 2008 18:43:58 -0000 Hi all, I've just started dabbling with ION and have been going through the tutorials. I note that it is (or at least was) shown to be interoperable with DTN2 with the TCP-CL, so I wondered if anyone has any tips or some rc file extracts for setting up an ION to DTN2 gateway using TCP-CL assuming an on-demand or always on link into a dtnd and handling the mix of ipn and dtn schemas. I also suspect that the apps in the two implementations won't talk, having just tried to use bpsource to send to a dtn:// EID, so I suspect my best bet is to route between two DTN2 based sub-nets over an IPN sub-net, with a pair of gateways. Cheers, Darren Received: from yx-out-2324.google.com (yx-out-2324.google.com [74.125.44.30]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mAKLQiko032344 for ; Thu, 20 Nov 2008 13:26:44 -0800 Received: by yx-out-2324.google.com with SMTP id 3so287760yxj.79 for ; Thu, 20 Nov 2008 13:06:22 -0800 (PST) Received: by 10.142.199.10 with SMTP id w10mr1341650wff.84.1227215182057; Thu, 20 Nov 2008 13:06:22 -0800 (PST) Received: by 10.142.113.15 with HTTP; Thu, 20 Nov 2008 13:06:22 -0800 (PST) Message-ID: <4bfc0da50811201306u68de9a3cp42f61c3cc5be5e2@mail.gmail.com> Date: Thu, 20 Nov 2008 21:06:22 +0000 From: "Ricardo Martins" To: "Christopher Small" In-Reply-To: <49233E1E.9070709@bbn.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <49233E1E.9070709@bbn.com> Cc: dtn-users mailing list Subject: Re: [dtn-users] getting dtn2 to build on Ubuntu 8.10 (really gcc 4.3.2) X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Nov 2008 21:26:45 -0000 Hi, Usually something like "gcc -dumpversion | cut -f1-2 -d" does the trick to get the relevant version of gcc (or g++). I've sent you the dump files of 9 gcc versions as requested. Regards, Ricardo Martins 2008/11/18 Christopher Small : > DTN2 does not build out-of-the-box on a stock Ubuntu 8.10 machine. I've > got some candidate fixes, and to see if I've done the right thing it > would be helpful to get a small amount of data from a large variety of > machines. > > If you would be so kind as to email me (and not the entire list) the > output of "gcc -v" and "g++ -v" on your machine (Mac, Linux, BSD, > Cygwin, ...) I would be very grateful. > > running the following should do the job, if you have mail configured on > your machine. > > $ (uname -a ; gcc -v 2>&1 ; g++ -v 2>&1) | mail csmall@bbn.com > > Otherwise running the following, followed by a cut-and-paste into a mail > composition window, should work. > > $ (uname -a ; gcc -v 2>&1 ; g++ -v 2>&1) > > Thanks. > > - Chris > > -- > Dr. Christopher Small 617.873.6261 (vox) > Networking Research 617.873.6091 (fax) > BBN Technologies | MS 6/5C | 10 Moulton St | Cambridge MA, 02138 > > _______________________________________________ > dtn-users mailing list > dtn-users@maillists.intel-research.net > http://maillists.intel-research.net/mailman/listinfo/dtn-users > Received: from mx11.bbn.com (mx11.bbn.com [128.33.0.80]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mAIMXLLk000500 for ; Tue, 18 Nov 2008 14:33:21 -0800 Received: from dhcp89-081-089.bbn.com ([128.89.81.89]) by mx11.bbn.com with esmtp (Exim 4.60) (envelope-from ) id 1L2YpS-00070F-Em for dtn-users@mailman.dtnrg.org; Tue, 18 Nov 2008 17:13:50 -0500 Message-ID: <49233E1E.9070709@bbn.com> Date: Tue, 18 Nov 2008 17:13:50 -0500 From: Christopher Small User-Agent: Thunderbird 2.0.0.17 (Macintosh/20080914) MIME-Version: 1.0 To: dtn-users mailing list X-Enigmail-Version: 0.95.7 OpenPGP: url=http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x6F6F97BB Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: [dtn-users] getting dtn2 to build on Ubuntu 8.10 (really gcc 4.3.2) X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2008 22:33:21 -0000 DTN2 does not build out-of-the-box on a stock Ubuntu 8.10 machine. I've got some candidate fixes, and to see if I've done the right thing it would be helpful to get a small amount of data from a large variety of machines. If you would be so kind as to email me (and not the entire list) the output of "gcc -v" and "g++ -v" on your machine (Mac, Linux, BSD, Cygwin, ...) I would be very grateful. running the following should do the job, if you have mail configured on your machine. $ (uname -a ; gcc -v 2>&1 ; g++ -v 2>&1) | mail csmall@bbn.com Otherwise running the following, followed by a cut-and-paste into a mail composition window, should work. $ (uname -a ; gcc -v 2>&1 ; g++ -v 2>&1) Thanks. - Chris -- Dr. Christopher Small 617.873.6261 (vox) Networking Research 617.873.6091 (fax) BBN Technologies | MS 6/5C | 10 Moulton St | Cambridge MA, 02138 Received: from an-out-0708.google.com (an-out-0708.google.com [209.85.132.241]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mAHJL8va023158 for ; Mon, 17 Nov 2008 11:21:09 -0800 Received: by an-out-0708.google.com with SMTP id d40so1002697and.34 for ; Mon, 17 Nov 2008 11:02:09 -0800 (PST) Received: by 10.100.134.10 with SMTP id h10mr1712055and.150.1226948528792; Mon, 17 Nov 2008 11:02:08 -0800 (PST) Received: by 10.100.45.13 with HTTP; Mon, 17 Nov 2008 11:02:08 -0800 (PST) Message-ID: <59e8babf0811171102m2a3c3ec3s2a80de13785a542c@mail.gmail.com> Date: Mon, 17 Nov 2008 11:02:08 -0800 From: "Michael Demmer" Sender: demmer@gmail.com To: "Kurtis Heimerl" In-Reply-To: <9a4753250811162329h5faa284aqb48115778b448edd@mail.gmail.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_Part_46660_4477137.1226948528789" References: <9a4753250811162329h5faa284aqb48115778b448edd@mail.gmail.com> X-Google-Sender-Auth: 1610e2e6a5fb83f3 Cc: dtn-users@maillists.intel-research.net Subject: Re: [dtn-users] [dtn-interest] Bundle Protocol question X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 17 Nov 2008 19:21:09 -0000 ------=_Part_46660_4477137.1226948528789 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Are you sure you're correctly decoding the flags part of the block? Briefly, each block should be a 1 byte typecode, then a flags SDNV, then if there are any EID references in the block, the vector of EIDs (but there shouldn't be in the paylod) then the block length SDNV, then the block data. In the DTN2 reference code, this is constructed in BlockProcessor::generate_preamble -- is there something there that you're not understanding? -m On Sun, Nov 16, 2008 at 11:29 PM, Kurtis Heimerl wrote: > Hello again DTN-interest! I've made great strides on the python DTN > implementation, and can send and receive smaller bundles over the > TCPCL. > > I've begun work on allowing for larger files. For small files, the > implementation was allowed to strip off the type, flags and length > fields from the Bundle Payload block and assume the rest is payload. > However, as the payload gets larger, this fails. > > I need a field that indicates the length of the payload, but it turns > out that the block length field of the Bundle payload block isn't > anything obviously correct. It seems to return 8 if the bundle is > small and 4096 if large. It doesn't depend on the length of the > payload, that much I'm certain about. This means either I'm > incorrectly decoding this variable, which I find unlikely due to the > correctness of the rest of the fields, or it's not what I thought it > was. Where am I supposed to get this length? > > Also, are any of the DTN implementations more readable than the > reference of symbian implementations? It's hard to parse those, so a > simpler version might lead me to unblocking myself more often. > > Thanks! > _______________________________________________ > dtn-interest mailing list > dtn-interest@maillists.intel-research.net > http://maillists.intel-research.net/mailman/listinfo/dtn-interest > ------=_Part_46660_4477137.1226948528789 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Are you sure you're correctly decoding the flags part of the block?

Briefly, each block should be a 1 byte typecode, then a flags SDNV, then if there are any EID references in the block, the vector of EIDs (but there shouldn't be in the paylod) then the block length SDNV, then the block data.

In the DTN2 reference code, this is constructed in BlockProcessor::generate_preamble -- is there something there that you're not understanding?
-m

On Sun, Nov 16, 2008 at 11:29 PM, Kurtis Heimerl <kheimerl@eecs.berkeley.edu> wrote:
Hello again DTN-interest! I've made great strides on the python DTN
implementation, and can send and receive smaller bundles over the
TCPCL.

I've begun work on allowing for larger files. For small files, the
implementation was allowed to strip off the type, flags and length
fields from the Bundle Payload block and assume the rest is payload.
However, as the payload gets larger, this fails.

I need a field that indicates the length of the payload, but it turns
out that the block length field of the Bundle payload block isn't
anything obviously correct. It seems to return 8 if the bundle is
small and 4096 if large. It doesn't depend on the length of the
payload, that much I'm certain about. This means either I'm
incorrectly decoding this variable, which I find unlikely due to the
correctness of the rest of the fields, or it's not what I thought it
was. Where am I supposed to get this length?

Also, are any of the DTN implementations more readable than the
reference of symbian implementations? It's hard to parse those, so a
simpler version might lead me to unblocking myself more often.

Thanks!
_______________________________________________
dtn-interest mailing list
dtn-interest@maillists.intel-research.net
http://maillists.intel-research.net/mailman/listinfo/dtn-interest

------=_Part_46660_4477137.1226948528789-- Received: from fg-out-1718.google.com (fg-out-1718.google.com [72.14.220.154]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mADDNK2Y008075 for ; Thu, 13 Nov 2008 05:23:20 -0800 Received: by fg-out-1718.google.com with SMTP id e21so720275fga.6 for ; Thu, 13 Nov 2008 05:06:12 -0800 (PST) Received: by 10.181.207.9 with SMTP id j9mr3164140bkq.149.1226581572628; Thu, 13 Nov 2008 05:06:12 -0800 (PST) Received: by 10.180.214.5 with HTTP; Thu, 13 Nov 2008 05:06:12 -0800 (PST) Message-ID: Date: Thu, 13 Nov 2008 15:06:12 +0200 From: "Shengye Lu" To: "DTN Users" MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Subject: [dtn-users] about encoding mechanism of DTN Reference implementation X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 13 Nov 2008 13:23:21 -0000 Hi, I am doing packet eavesdropping test for DTN application which is based on DTN 2.4. As we know, DTN application API is on IPC client side, while DTNd sits on IPC server side. I use Wireshark to capture packets between application API and DTNd. Obviously, if I use Wireshark to capture packets among DTNds, Wireshark can't decode the packets, since Wireshark does not recognize BP. But my testing indicates: Wireshark can partly decode packets between application API and DTNd. I want to know: does Application API use the same encoding mechanism as DTNd? What encoding mechanism does DTN application API use? Cheers, -- Shengye Received: from c.painless.aaisp.net.uk (c.painless.aaisp.net.uk [81.187.30.53]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mADBpR30003392 for ; Thu, 13 Nov 2008 03:51:27 -0800 Received: from 250.254.187.81.in-addr.arpa ([81.187.254.250]) by c.painless.aaisp.net.uk with esmtp (Exim 4.69) (envelope-from ) id 1L0aSq-0001AJ-LE; Thu, 13 Nov 2008 11:34:20 +0000 From: Elwyn Davies To: Lloyd Wood In-Reply-To: <15765EEC-3C2A-4308-8CED-CA1CEC1F1DD1@surrey.ac.uk> References: <491B46E3.6080104@folly.org.uk> <15765EEC-3C2A-4308-8CED-CA1CEC1F1DD1@surrey.ac.uk> Content-Type: text/plain Organization: Folly Consulting Date: Thu, 13 Nov 2008 11:34:16 +0000 Message-Id: <1226576056.30272.189.camel@mightyatom.folly.org.uk> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 Content-Transfer-Encoding: 7bit X-Virus-Scanned: Clear (Version: ClamAV 0.94/8571/Wed Nov 5 11:25:50 2008, by smtp.aaisp.net.uk) Cc: DTN interest , dtn-users@maillists.intel-research.net Subject: Re: [dtn-users] [Fwd: Re: Oasys 1.3.0 does not compile on Cygwin] X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 13 Nov 2008 11:51:27 -0000 Hi, Lloyd. On Wed, 2008-11-12 at 21:45 +0000, Lloyd Wood wrote: > Elwyn, > > Previous list discussion of Cygwin stalled at the d_type problem. > Appears you've found a simple workaround and found that d_type use is > not essential to the dtn code. Right! Cheap hack. > > My take is that you're already over the hard part for getting the dtn > code to run, at least - but though cygwin can be convenient, it's > rarely preferable for development, and not good for networking work > (no multicast, no IPv6) or anything that stresses the filesystem > (slow, weirdness in mapping unix behaviour to windows, odd piping > behaviour too.) The original intention was just to get it up and running to provide me with a couple of extra testing nodes 'for free'. Well, TANSTAAFL!!! Anyway after a deal of head scratching I now have runnable executables. Whether they do what I need is the next question. I have sent a message to the users list documenting what I had to do. And it has provided me with a chance to run around the code a little. I'll discuss getting some patches in with Michael (possibly in Minneapolis). > > If you're familiar with Cygwin's weirdnesses, fine -- if not, you'll > be wondering if the behaviour you see is due to the program, or to > running the program on cygwin. Thanks for the warning! At the moment I only want very simple behaviour, so I'll see if it can provide. I don't intend to do any development on Cygwin anyway. I have Linux for that! Cheers, Elwyn > > cheers, > > L. > > On 12 Nov 2008, at 21:13, Folly Consulting wrote: > > > Stephen Farrell suggested the attached message might be usefully > > seen on > > the users list. > > > > Just to continue. Having got oasys to compile I tried DTN 2.6.0. It > > appears it has a similar number of minor problems, plus a problem with > > getting the linkage right between the xdr routines and the C++. the > > applib is currently refusing to link. > > > > However in the light of Michael's comments below, it is not clear > > whether there is any point in continuing. Maybe some sort of caveat > > ought to be added to the buiklding instructions or in the wiki - the > > manual currently implies that it can be built for Cygwin and there are > > clearly significant problems to sort out.. which may be intractable. > > > > Regards, > > Elwyn Davies > > > > -------- Original Message -------- > > Subject: Re: Oasys 1.3.0 does not compile on Cygwin > > Date: Wed, 12 Nov 2008 11:38:01 -0800 > > From: Michael Demmer > > To: Elwyn Davies > > CC: Delay Tolerant Networking Interest List > > > > References: <491AE958.3070001@dial.pipex.com> > > > > > > > > Yes. There have been cygwin incompatibilities for some time. From what > > I remember, getting things to compile in cygwin is a bit of a pain but > > not fundamentally a challenge, however there was something more > > fundamental in the threading and IO model that caused things to break > > in cygwin. That may have been fixed with a newer version of cygwin, > > but I haven't tried for some time. > > > > Also, I began an effort to port the code to use the Apache Portable > > Runtime (APR) which would mean that the system could run natively on > > either UNIX or Windows platforms. That work is currently checked in > > under the apr-port branch. > > > > -mike > > > > On Wed, Nov 12, 2008 at 6:34 AM, Elwyn Davies > > wrote: > >> Hi. > >> > >> I have been trying to compile DTN2 v2.6.0 and Oasys v1.3.0 on > >> Windows XP > >> SP2/Cygwin (version 1.5.25-15) > >> > >> There are 4 problems (so far): > >> io/FileUtils.cc: rm_all_from_dir function does not compile because > >> the > >> 'dirent structure does not contain 'd_type' on Cygwin (line 133). > >> This > >> field is only used when the 'recursive' flag is set - this isn't > >> used in > >> dtn, so commenting out the test for recursive delete provides a > >> workaround. > >> > >> io/TTY.h: Lines 36-39: cfgetispeed and cfgetospeed are defined as > >> preprocessor macros in sys/termios.h rather than global functions. > >> Consequently both the declarations and the functions that use them in > >> FileUtils.cc do not compile. AFAIK these functions are not used. I > >> commented out the functions and the declarations to allow it to > >> compile. > >> > >> io/TTY.h (Line 40) and ioTTY.cc(lines 96-101): There is no function > >> cfmakeraw in Cygwin. hence the wrapper will not compile. In order > >> to > >> support > >> the standard DTN code, we need to define TTY::cfmakeraw() as a No- > >> op in > >> oasys for cygwin. > >> > >> Makefile: 'make install' fails.The shell script executed for the > >> target > >> 'installlibs' (lines 305-419) is not good 'bash' code. The > >> '[' (test) > >> commends fail because the equality operator is '==' rather than '=' > >> and > >> the $DESTDIR argument needs to be enclosed in double quotes to > >> cater for > >> the empty string case (as is the default). Also the SHLIB_EXT > >> variable > >> doesn't seem to be defined. the following fragment works (if > >> SHLIB_EXT > >> := a is added): > >> > >> installlib: installdirs > >> [ "$(DESTDIR)" == . ] || \ > >> [ x$(SHLIBS) == x ] || \ > >> for lib in lib/liboasys-$(OASYS_VERSION).$(SHLIB_EXT) \ > >> lib/liboasyscompat-$(OASYS_VERSION).$(SHLIB_EXT) ; do \ > >> ($(INSTALL_PROGRAM) $$lib $(DESTDIR)$(libdir)) ; \ > >> done > >> > >> [ "$(DESTDIR)" == . ] || \ > >> for lib in liboasys liboasyscompat ; do \ > >> (cd $(DESTDIR)$(libdir) && rm -f $$lib.$(SHLIB_EXT) && \ > >> ln -s $$lib-$(OASYS_VERSION).$(SHLIB_EXT) $$lib.$ > >> (SHLIB_EXT)) \ > >> done > >> > >> > >> There are problems with DTN as well.. I will report these separately. > >> > >> Regards, > >> Elwyn > >> > > > > > > > > -- > > -------------------------------------------------------------------------- > > Folly Consulting Ltd, 60 The Butts, Soham, Ely, Cambridgeshire, CB7 > > 5AW > > Company number 6238664: Registered in the UK > > _______________________________________________ > > dtn-users mailing list > > dtn-users@maillists.intel-research.net > > http://maillists.intel-research.net/mailman/listinfo/dtn-users > > DTN work: http://info.ee.surrey.ac.uk/Personal/L.Wood/saratoga/ > > > > > > > Received: from c.painless.aaisp.net.uk (c.painless.aaisp.net.uk [81.187.30.53]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mADBg0mM002933 for ; Thu, 13 Nov 2008 03:42:01 -0800 Received: from 250.254.187.81.in-addr.arpa ([81.187.254.250]) by c.painless.aaisp.net.uk with esmtp (Exim 4.69) (envelope-from ) id 1L0aJi-00083g-EE for dtn-users@maillists.intel-research.net; Thu, 13 Nov 2008 11:24:54 +0000 From: Elwyn Davies To: DTN Users Content-Type: text/plain Organization: Folly Consulting Date: Thu, 13 Nov 2008 11:24:50 +0000 Message-Id: <1226575490.30272.179.camel@mightyatom.folly.org.uk> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 Content-Transfer-Encoding: 7bit X-Virus-Scanned: Clear (Version: ClamAV 0.94/8571/Wed Nov 5 11:25:50 2008, by smtp.aaisp.net.uk) Subject: [dtn-users] Problems compiling DTN 2.6.0 on Cygwin (and fixes) X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 13 Nov 2008 11:42:02 -0000 In addition to the problems reported with oasys v1.3.0, there are also problems with compiling DTN v2.6.0 on Windows XP SP2/Cygwin 1.5.25-15. util/SafeRange.h: This header uses size_t. This should probably pull in the stddef.h file supplied with the compiler for the environment automatically. For example this does happen on Gentoo Linux 2.6.25/gcc 4.1.2. However it does not seem to happen on Cygwin which uses a 'special' version of gcc 3.4.4. Adding two lines to the file before the start of the template fixes the problem: #define __need_size_t #include There is no need to specify additional include directories. stddef.h is buried in the compiler directories - /usr/lib/gcc/i686-pc-linux-gnu/4.1.2/include/stddef.h on my Linux build and /usr/libgcc/i686-pc-cygwin/3.4.4/include/stddef.h on Cygwin. servlib/routing/ExternalRouter.cc: This file does not compile because the Cygwin header netinet/in.h doesn't contain a definition of INADDR_ALLRTRS_GROUP. Adding the requisite #define after including netinet/in.h solves the problem: #ifdef __CYGWIN__ #define INADDR_ALLRTRS_GROUP ((inaddr_t) 0xE0000002) /* 224.0.0.2 */ #endif conv_layers/SerialConvergence.cc: There aren't any actual problems in this file - they are masked by fixing up the Oasys TTY.h/TTY.cc to provide a no-op cfmakeraw function - Windows doesn't have the concept of raw/cooked serial lines. (TTY.* are also fixed up to deal with some other issues with Cygwin termios definitions but they aren't relevant to DTN.) Applib is the source of most of the grief in sorting out Cygwin. The basic cause of the problems is that the XDR library routines used to marshal/unmarshal the basic types in the code that is created by rpcgen have archaic Kernigan and Ritchie (K&R) style definitions in /usr/include/rpc/xdr.h (definitions of xdr_ routines). Mostly this is not a problem. Almost all the calls to these routines are made in C code in dtn_types_xdr.c, where they do not cause any problems. However applib/APIServer.cc contains one call to xdr_u_int (line 1604). This call has two parameters (as it should) but is seen by g++ as conflicting with the K&R definition which implies no parameters. The fix is to define a C function xdr_u_int_ with a proper prototype and a macro which maps xdr_u_int -> xdr_u_int_ in APIServer.cc. This involves providing the prototype and macro definitions in oasys/compat/rpc.h and the declaration of the function should go in oasys/compat/rpc_compat.c - this is just a wrapper that calls the standard xdr_u_int. I also believe that applib/sdnv-c.c needs to include oasys/oasys-config.h to keep one of the other header files in bundling/SDNV.cc happy. Finally there are a couple of Makefile problems. 1. The calls to the xdr_ functions are all in functions that are parcelled up into libdtnapi libraries which appear in the (static) linker commands *after* the -lrpc argument that pulls in the implementations from librpc.a. This causes the link commands to fail because librpc.a is not rescanned (the dynamic linking case on Linux doesn't seem to suffer from this problem - not sure how it differs). This can be fixed by modifying the library lists in the Makefile to include a final -lrpc. 2. The linker gets upset about the xerces-c library. It complains that it needs to have the --enable-auto-import flag supplied in order to correctly pull in all the routines it needs (if I read the error messages rightly). Fixed this by specifying EXTRA_LDFLAGS = -Xlinker --enable-auto-import 3. Finally one of the static libraries was not being made with ar but with g++. 4. Installation doesn't work (don't think it remembers .exe suffices but it was getting a little late...). Proper fixing of the Makefile will require a little more investigation. Having worked all this through, I now have a set of usable executables. I haven't yet tested them much (dtnd.exe fires up and complains it can't find the configuration script - not surprising as I haven't made it yet - but that is as far as I have got). And I only wanted it to do a couple of simple tests... I think most of fixes could go into the main code line without problem. I may need some assistance with fixing up the Makefile parts. Regards, Elwyn Received: from mail71.messagelabs.com (mail71.messagelabs.com [193.109.255.131]) by maillists.intel-research.net (8.13.8/8.13.8) with SMTP id mACM2Z2A030446 for ; Wed, 12 Nov 2008 14:02:36 -0800 X-VirusChecked: Checked X-Env-Sender: L.Wood@surrey.ac.uk X-Msg-Ref: server-4.tower-71.messagelabs.com!1226526344!43744169!1 X-StarScan-Version: 5.5.12.14.2; banners=-,-,- X-Originating-IP: [131.227.102.140] Received: (qmail 10865 invoked from network); 12 Nov 2008 21:45:45 -0000 Received: from ads40.surrey.ac.uk (HELO ads40.surrey.ac.uk) (131.227.102.140) by server-4.tower-71.messagelabs.com with SMTP; 12 Nov 2008 21:45:45 -0000 Received: from ads31.surrey.ac.uk ([131.227.120.131]) by ads40.surrey.ac.uk with Microsoft SMTPSVC(6.0.3790.3959); Wed, 12 Nov 2008 21:45:44 +0000 Received: from [192.168.1.208] ([86.3.93.24]) by ads31.surrey.ac.uk over TLS secured channel with Microsoft SMTPSVC(6.0.3790.3959); Wed, 12 Nov 2008 21:45:42 +0000 Message-Id: <15765EEC-3C2A-4308-8CED-CA1CEC1F1DD1@surrey.ac.uk> From: Lloyd Wood To: Folly Consulting In-Reply-To: <491B46E3.6080104@folly.org.uk> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v929.2) Date: Wed, 12 Nov 2008 21:45:42 +0000 References: <491B46E3.6080104@folly.org.uk> X-Mailer: Apple Mail (2.929.2) X-OriginalArrivalTime: 12 Nov 2008 21:45:43.0765 (UTC) FILETIME=[03859850:01C94510] Cc: DTN interest , dtn-users@maillists.intel-research.net, Lloyd Wood Subject: Re: [dtn-users] [Fwd: Re: Oasys 1.3.0 does not compile on Cygwin] X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 12 Nov 2008 22:02:37 -0000 Elwyn, Previous list discussion of Cygwin stalled at the d_type problem. Appears you've found a simple workaround and found that d_type use is not essential to the dtn code. My take is that you're already over the hard part for getting the dtn code to run, at least - but though cygwin can be convenient, it's rarely preferable for development, and not good for networking work (no multicast, no IPv6) or anything that stresses the filesystem (slow, weirdness in mapping unix behaviour to windows, odd piping behaviour too.) If you're familiar with Cygwin's weirdnesses, fine -- if not, you'll be wondering if the behaviour you see is due to the program, or to running the program on cygwin. cheers, L. On 12 Nov 2008, at 21:13, Folly Consulting wrote: > Stephen Farrell suggested the attached message might be usefully > seen on > the users list. > > Just to continue. Having got oasys to compile I tried DTN 2.6.0. It > appears it has a similar number of minor problems, plus a problem with > getting the linkage right between the xdr routines and the C++. the > applib is currently refusing to link. > > However in the light of Michael's comments below, it is not clear > whether there is any point in continuing. Maybe some sort of caveat > ought to be added to the buiklding instructions or in the wiki - the > manual currently implies that it can be built for Cygwin and there are > clearly significant problems to sort out.. which may be intractable. > > Regards, > Elwyn Davies > > -------- Original Message -------- > Subject: Re: Oasys 1.3.0 does not compile on Cygwin > Date: Wed, 12 Nov 2008 11:38:01 -0800 > From: Michael Demmer > To: Elwyn Davies > CC: Delay Tolerant Networking Interest List > > References: <491AE958.3070001@dial.pipex.com> > > > > Yes. There have been cygwin incompatibilities for some time. From what > I remember, getting things to compile in cygwin is a bit of a pain but > not fundamentally a challenge, however there was something more > fundamental in the threading and IO model that caused things to break > in cygwin. That may have been fixed with a newer version of cygwin, > but I haven't tried for some time. > > Also, I began an effort to port the code to use the Apache Portable > Runtime (APR) which would mean that the system could run natively on > either UNIX or Windows platforms. That work is currently checked in > under the apr-port branch. > > -mike > > On Wed, Nov 12, 2008 at 6:34 AM, Elwyn Davies > wrote: >> Hi. >> >> I have been trying to compile DTN2 v2.6.0 and Oasys v1.3.0 on >> Windows XP >> SP2/Cygwin (version 1.5.25-15) >> >> There are 4 problems (so far): >> io/FileUtils.cc: rm_all_from_dir function does not compile because >> the >> 'dirent structure does not contain 'd_type' on Cygwin (line 133). >> This >> field is only used when the 'recursive' flag is set - this isn't >> used in >> dtn, so commenting out the test for recursive delete provides a >> workaround. >> >> io/TTY.h: Lines 36-39: cfgetispeed and cfgetospeed are defined as >> preprocessor macros in sys/termios.h rather than global functions. >> Consequently both the declarations and the functions that use them in >> FileUtils.cc do not compile. AFAIK these functions are not used. I >> commented out the functions and the declarations to allow it to >> compile. >> >> io/TTY.h (Line 40) and ioTTY.cc(lines 96-101): There is no function >> cfmakeraw in Cygwin. hence the wrapper will not compile. In order >> to >> support >> the standard DTN code, we need to define TTY::cfmakeraw() as a No- >> op in >> oasys for cygwin. >> >> Makefile: 'make install' fails.The shell script executed for the >> target >> 'installlibs' (lines 305-419) is not good 'bash' code. The >> '[' (test) >> commends fail because the equality operator is '==' rather than '=' >> and >> the $DESTDIR argument needs to be enclosed in double quotes to >> cater for >> the empty string case (as is the default). Also the SHLIB_EXT >> variable >> doesn't seem to be defined. the following fragment works (if >> SHLIB_EXT >> := a is added): >> >> installlib: installdirs >> [ "$(DESTDIR)" == . ] || \ >> [ x$(SHLIBS) == x ] || \ >> for lib in lib/liboasys-$(OASYS_VERSION).$(SHLIB_EXT) \ >> lib/liboasyscompat-$(OASYS_VERSION).$(SHLIB_EXT) ; do \ >> ($(INSTALL_PROGRAM) $$lib $(DESTDIR)$(libdir)) ; \ >> done >> >> [ "$(DESTDIR)" == . ] || \ >> for lib in liboasys liboasyscompat ; do \ >> (cd $(DESTDIR)$(libdir) && rm -f $$lib.$(SHLIB_EXT) && \ >> ln -s $$lib-$(OASYS_VERSION).$(SHLIB_EXT) $$lib.$ >> (SHLIB_EXT)) \ >> done >> >> >> There are problems with DTN as well.. I will report these separately. >> >> Regards, >> Elwyn >> > > > > -- > -------------------------------------------------------------------------- > Folly Consulting Ltd, 60 The Butts, Soham, Ely, Cambridgeshire, CB7 > 5AW > Company number 6238664: Registered in the UK > _______________________________________________ > dtn-users mailing list > dtn-users@maillists.intel-research.net > http://maillists.intel-research.net/mailman/listinfo/dtn-users DTN work: http://info.ee.surrey.ac.uk/Personal/L.Wood/saratoga/ Received: from a.painless.aaisp.net.uk (a.painless.aaisp.net.uk [81.187.30.51]) by maillists.intel-research.net (8.13.8/8.13.8) with ESMTP id mACLPM7J028737 for ; Wed, 12 Nov 2008 13:25:23 -0800 Received: from 247.254.187.81.in-addr.arpa ([81.187.254.247]) by a.painless.aaisp.net.uk with esmtps (TLSv1:AES256-SHA:256) (Exim 4.69) (envelope-from ) id 1L0Mwz-0003ez-3T for dtn-users@maillists.intel-research.net; Wed, 12 Nov 2008 21:08:33 +0000 Message-ID: <491B46E3.6080104@folly.org.uk> Date: Wed, 12 Nov 2008 21:13:07 +0000 From: Folly Consulting Organization: Folly Consulting Ltd User-Agent: Thunderbird 2.0.0.17 (Windows/20080914) MIME-Version: 1.0 To: dtn-users@maillists.intel-research.net Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: [dtn-users] [Fwd: Re: Oasys 1.3.0 does not compile on Cygwin] X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 12 Nov 2008 21:25:23 -0000 Stephen Farrell suggested the attached message might be usefully seen on the users list. Just to continue. Having got oasys to compile I tried DTN 2.6.0. It appears it has a similar number of minor problems, plus a problem with getting the linkage right between the xdr routines and the C++. the applib is currently refusing to link. However in the light of Michael's comments below, it is not clear whether there is any point in continuing. Maybe some sort of caveat ought to be added to the buiklding instructions or in the wiki - the manual currently implies that it can be built for Cygwin and there are clearly significant problems to sort out.. which may be intractable. Regards, Elwyn Davies -------- Original Message -------- Subject: Re: Oasys 1.3.0 does not compile on Cygwin Date: Wed, 12 Nov 2008 11:38:01 -0800 From: Michael Demmer To: Elwyn Davies CC: Delay Tolerant Networking Interest List References: <491AE958.3070001@dial.pipex.com> Yes. There have been cygwin incompatibilities for some time. From what I remember, getting things to compile in cygwin is a bit of a pain but not fundamentally a challenge, however there was something more fundamental in the threading and IO model that caused things to break in cygwin. That may have been fixed with a newer version of cygwin, but I haven't tried for some time. Also, I began an effort to port the code to use the Apache Portable Runtime (APR) which would mean that the system could run natively on either UNIX or Windows platforms. That work is currently checked in under the apr-port branch. -mike On Wed, Nov 12, 2008 at 6:34 AM, Elwyn Davies wrote: > Hi. > > I have been trying to compile DTN2 v2.6.0 and Oasys v1.3.0 on Windows XP > SP2/Cygwin (version 1.5.25-15) > > There are 4 problems (so far): > io/FileUtils.cc: rm_all_from_dir function does not compile because the > 'dirent structure does not contain 'd_type' on Cygwin (line 133). This > field is only used when the 'recursive' flag is set - this isn't used in > dtn, so commenting out the test for recursive delete provides a workaround. > > io/TTY.h: Lines 36-39: cfgetispeed and cfgetospeed are defined as > preprocessor macros in sys/termios.h rather than global functions. > Consequently both the declarations and the functions that use them in > FileUtils.cc do not compile. AFAIK these functions are not used. I > commented out the functions and the declarations to allow it to compile. > > io/TTY.h (Line 40) and ioTTY.cc(lines 96-101): There is no function > cfmakeraw in Cygwin. hence the wrapper will not compile. In order to > support > the standard DTN code, we need to define TTY::cfmakeraw() as a No-op in > oasys for cygwin. > > Makefile: 'make install' fails.The shell script executed for the target > 'installlibs' (lines 305-419) is not good 'bash' code. The '[' (test) > commends fail because the equality operator is '==' rather than '=' and > the $DESTDIR argument needs to be enclosed in double quotes to cater for > the empty string case (as is the default). Also the SHLIB_EXT variable > doesn't seem to be defined. the following fragment works (if SHLIB_EXT > := a is added): > > installlib: installdirs > [ "$(DESTDIR)" == . ] || \ > [ x$(SHLIBS) == x ] || \ > for lib in lib/liboasys-$(OASYS_VERSION).$(SHLIB_EXT) \ > lib/liboasyscompat-$(OASYS_VERSION).$(SHLIB_EXT) ; do \ > ($(INSTALL_PROGRAM) $$lib $(DESTDIR)$(libdir)) ; \ > done > > [ "$(DESTDIR)" == . ] || \ > for lib in liboasys liboasyscompat ; do \ > (cd $(DESTDIR)$(libdir) && rm -f $$lib.$(SHLIB_EXT) && \ > ln -s $$lib-$(OASYS_VERSION).$(SHLIB_EXT) $$lib.$(SHLIB_EXT)) \ > done > > > There are problems with DTN as well.. I will report these separately. > > Regards, > Elwyn > -- -------------------------------------------------------------------------- Folly Consulting Ltd, 60 The Butts, Soham, Ely, Cambridgeshire, CB7 5AW Company number 6238664: Registered in the UK Received: from web26504.mail.ukl.yahoo.com (web26504.mail.ukl.yahoo.com [217.146.176.41]) by maillists.intel-research.net (8.13.8/8.13.8) with SMTP id mAC2IPPP008569 for ; Tue, 11 Nov 2008 18:18:26 -0800 Received: (qmail 82288 invoked by uid 60001); 12 Nov 2008 02:01:57 -0000 X-YMail-OSG: sJv27FAVM1m7LB7qS8ll.FjZqTO79vHIeC.QN2Gdw34dEKbxVV3KszJH_f0_whxtC.3YOxoByrbuNU0SZ7WNNT0OvCEU0V3apkO8my_uDHNy3Cm6RM6jeuLYgOjhwhTHs8nrRhziFLj0kzcGFJEy8Pq8D3L9KaGzauuyrERkz5tA3_SLJa5dht9XP6nR Received: from [210.245.53.34] by web26504.mail.ukl.yahoo.com via HTTP; Wed, 12 Nov 2008 02:01:56 GMT X-Mailer: YahooMailRC/1155.20 YahooMailWebService/0.7.260.1 Date: Wed, 12 Nov 2008 02:01:56 +0000 (GMT) From: tsafack landry To: dtn-users@maillists.intel-research.net MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-1773422795-1226455316=:82248" Message-ID: <78162.82248.qm@web26504.mail.ukl.yahoo.com> Subject: [dtn-users] ok X-BeenThere: dtn-users@maillists.intel-research.net X-Mailman-Version: 2.1.9 Precedence: list List-Id: dtn-users mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 12 Nov 2008 02:18:26 -0000 --0-1773422795-1226455316=:82248 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable tsaf231@yahoo.fr=0Atland@ifi.edu.vn=0A=0A=0A=0A --0-1773422795-1226455316=:82248 Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable
tsaf231@yahoo.fr
tland@ifi.edu.vn

= =0A=0A=0A=0A --0-1773422795-1226455316=:82248--