From kunchanl@pollux.usc.edu Fri Mar 1 05:15:10 2002 From: kunchanl@pollux.usc.edu (kunchanl) Date: Fri Mar 1 05:15:10 2002 Subject: [ns] Problem with DSR sample script In-Reply-To: <200202252250200660.00099BEC@smtp.idola.net.id> Message-ID: You might want to try ns-2.1b7 which seems to work fine for this script. Kun-chan Lan On Mon, 25 Feb 2002, Lukman Hakim wrote: > > Hi.. > > I'm running ns-2.1b8b on RedHat Linux 7.1 the installation process just fine and working well with other sample script. But when I try to execute the sample script (wireless-mip-test.tcl) and got some result like this : > > [lukman@northpole ex]$ ns wireless-mip-test.tcl > no value given for parameter "errproc" to "_o33" > (Node/MobileNode add-interface line 1) > invoked from within > "$node add-interface $chan $prop $opt(ll) $opt(mac) $opt(ifq) > $opt(ifqlen) $opt(netif) $opt(ant)" > (procedure "create-dsr-bs-node" line 5) > invoked from within > "create-$opt(rp)-bs-node $node $id" > (procedure "create-base-station-node" line 21) > invoked from within > "create-base-station-node 1.0.0" > invoked from within > "set HA [create-base-station-node 1.0.0]" > (file "wireless-mip-test.tcl" line 137) > > and also for script wireless3.tcl that I already modified for adhoc routing protocol from DSDV to DSR, and also got an error message like this : > > [lukman@northpole examples]$ ns wireless3.tcl > num_nodes is set 3 > warning: Please use -channel as shown in tcl/ex/wireless-mitf.tcl > can't read "ragent_": no such variable > while executing > "subst $[subst $var" > (procedure "_o36" line 5) > (SplitObject set line 5) > invoked from within > "$self set ragent_" > (procedure "_o36" line 3) > (Node/MobileNode agenttrace line 3) > invoked from within > "$node agenttrace $tracefd" > (procedure "_o3" line 75) > (Simulator create-wireless-node line 75) > invoked from within > "_o3 create-wireless-node 1.0.0" > ("eval" body line 1) > invoked from within > "eval $self create-wireless-node $args" > (procedure "_o3" line 14) > (Simulator node line 14) > invoked from within > "$ns_ node 1.0.0" > invoked from within > "set HA [$ns_ node 1.0.0]" > (file "wireless3b.tcl" line 124) > > > Could anyone help me with these problem ?? Does NS work well for DSR adhoc routing ?? > > Thank's in advance guys.. > > Regards, > > -lukman- > From kunchanl@pollux.usc.edu Fri Mar 1 05:15:20 2002 From: kunchanl@pollux.usc.edu (kunchanl) Date: Fri Mar 1 05:15:20 2002 Subject: [ns] Counting packets sents by nodes In-Reply-To: Message-ID: You might want to see if you can use QueueMonitor, see chap 22 of ns manual. Kun-chan Lan On Wed, 27 Feb 2002, Asaf Shabtai wrote: > > Hi all, > I would like to get some help in counting packets sent by each node int the > simulation, > Where should I place the counter (In the node object ??) and where does the > node execute the send method to the next node (so I can change the counter). > Thank U > Asaf > > > _________________________________________________________________ > Chat with friends online, try MSN Messenger: http://messenger.msn.com > From kunchanl@pollux.usc.edu Fri Mar 1 05:15:29 2002 From: kunchanl@pollux.usc.edu (kunchanl) Date: Fri Mar 1 05:15:29 2002 Subject: [ns] Calling C++ procedure in TCL In-Reply-To: Message-ID: You will need to use "command", please see chapter 1 of ns manual. Kun-chan Lan On Wed, 27 Feb 2002, Steve Miles wrote: > > Hi, > > I made some extensions to a queue (e.g. i added a function void test(); ) > and i want to call that procedure directly from my tcl script. > > I already tried with creating a Queue instance, queue > and then : > $ns at 0 "$queue test" > But it gives an error when running the TCL script > > How can I do that ??? > > thanks in advance > > > _________________________________________________________________ > Send and receive Hotmail on your mobile device: http://mobile.msn.com > From kunchanl@pollux.usc.edu Fri Mar 1 05:15:38 2002 From: kunchanl@pollux.usc.edu (kunchanl) Date: Fri Mar 1 05:15:38 2002 Subject: [ns] Urgently need Tcl debgger for ns-2.1b6!!! In-Reply-To: <20020301003732.91703.qmail@web14804.mail.yahoo.com> Message-ID: Try http://www.cs.ucla.edu/classes/cs217/ns-allinone-2.1b4/ Kun-chan Lan On Thu, 28 Feb 2002, kaiduan xie wrote: > > hi, > > Anyone who are using ns-2.1b6,please give me a hand. > Since 2.1b6 used Tcl 8.04, the current tcl-debugger > 2.0 cannot be used in ns-2.1b6. I have been searching > for tcl-debugger for ns-2.1b6 for several days without > any results. The author in NIST didnot remmeber where > he put the code. If you have one, please give me a > copy. I needed it urgently. Thanks for help. > > kaiduan xie > > ______________________________________________________________________ > Find, Connect, Date! http://personals.yahoo.ca > From kunchanl@pollux.usc.edu Fri Mar 1 05:15:46 2002 From: kunchanl@pollux.usc.edu (kunchanl) Date: Fri Mar 1 05:15:46 2002 Subject: [ns] "advance" and "advanceby" command in TcpAgent In-Reply-To: Message-ID: the code in tcp.cc will kinda self-explain int TcpAgent::command(int argc, const char*const* argv) { if (argc == 3) { if (strcmp(argv[1], "advance") == 0) { int newseq = atoi(argv[2]); if (newseq > maxseq_) advanceby(newseq - curseq_); else advanceby(maxseq_ - curseq_); return (TCL_OK); } if (strcmp(argv[1], "advanceby") == 0) { advanceby(atoi(argv[2])); return (TCL_OK); } Kun-chan Lan On Tue, 26 Feb 2002, Kai Xu wrote: > > Hi, > > What are the meaning of "advance" and "advanceby" commands for Agent/TCP? > eg. > > set tcp [new Agent/TCP] > $tcp advance 10 > > Thanks in advance. > > -- > Kai > > From johnh@ISI.EDU Fri Mar 1 05:15:58 2002 From: johnh@ISI.EDU (John Heidemann) Date: Fri Mar 1 05:15:58 2002 Subject: [ns] ns-2 FAQ Message-ID: <200203010904.g2194Xt12820@dash.isi.edu> The Network Simulator ns-2: Frequently Asked Questions (This FAQ is also on the web at http://www.isi.edu/nsnam/ns/ns-faq.html.) * _Where do I get ns?_ From the ns web site at http://www.isi.edu/nsnam/ns/ns.html and the download page http://www.isi.edu/nsnam/ns/ns-tests.html. * _What platforms does ns run on and what kind of hardware do I need?_ Please see "where to start" on the building ns web page: http://www.isi.edu/nsnam/ns/ns-build.html#start. * _What should I do if I have trouble downloading/extracting ns?_ This question is answered in detail at http://www.isi.edu/nsnam/ns/ns-problems.html#downloading. * _What should I do if I encounter problems building ns?_ Check: 1. the README that comes in the distribution (very brief), 2. the "installation problems, bug fixes and help" web page http://www.isi.edu/nsnam/ns/ns-problems.html, 3. the archives of the ns-users mailing list http://www.isi.edu/nsnam/ns/ns-lists.html, 4. post a bug report (see below) http://www.isi.edu/cgi-bin/nsnam/reportbug.cgi. * _What do I do after I successfully build ns?_ + Put the path to your ns executable into your PATH environment + Put the path to your otcl into your LD_LIBRARY_PATH environment + Put the path to your tcl library into your TCL_LIBRARY environment * _Where can I find documentation for ns?_ All documentation is linked from the main ns web page http://www.isi.edu/nsnam/ns/. Documentation includes a tutorial (originally from Marc Greis) and a reference manual (ns notes and documentation). * _Words, words, words... that documentation is nice, but where are some sample scripts I can start from?_ Many sample scripts can be found in the ns distribution in ~ns-2/tcl/ex and ~ns-2/tcl/test. * _What protocols does ns support?_ A lot! Almost all variants of TCP, several forms of multicast, wired networking, several ad hoc routing protocols and propagation models (but not cellular phones), data diffusion, satellite, and other stuff. See the documentation (described above) for details, or download ns and look. * _How do I know that ns correctly implements these protocols?_ Ns has validation tests that cover many protocols, see http://www.isi.edu/nsnam/ns/ns-tests.html. However, ultimately users are responsible for verifying that ns is accurate for their purposes---since we cannot foresee all the ways ns may be used, we cannot test all cases with all inputs. * _Are there any contributed/additional protocols not in the main distribution?_ Yes, please see the contributed code web page http://www.isi.edu/nsnam/ns/ns-contributed.html. The mailing list archives can also be helpful (see below). * _How should I get started doing something (like implementing a new protocol or trying an experiment)?_ We recommend that you look through the tutorial (see documentation, above), then start with an example program that is most similar to yours (in the tutorial, or in tcl/ex or tcl/test in the distribution), and then start changing things. * _What should I do to compile ns to reflect my changes if I've modified some .cc or .h files?_ go to ns directory and run "make" or "make depend; make" * _How do I subscribe to the ns-users mailing list? How do I search old list archives? I can't take any more---how do I get off this list?_ To subscribe or unsubscribe, see http://www.isi.edu/nsnam/ns/ns-lists.html. The list archive is at http://www.isi.edu/nsnam/ns/ns-lists.html. * _What if I have a question that's not answered here?_ If you've checked the installation problems and bug fixes web page (http://www.isi.edu/nsnam/ns/ns-problems.html) and there's no answer to your question, you may want to file a bug report or post a question to the ns-user's mailing list. First, you should check the archive of the list at http://www.isi.edu/nsnam/ns/ns-lists.html. Your question may already be answered there. If not, you can post a bug report using the web form at http://www.isi.edu/cgi-bin/nsnam/reportbug.cgi. If your question is NOT about ns implementation bugs, you may wish to post to the list. First you should subscribe. Subscription instructions are at http://www.isi.edu/nsnam/ns/ns-lists.html. When posting bug reports, please _always_ include information including at least (the bug report form includes spaces for these): + what version of ns you're using, + what operating system you're running on (not just Linux or Solaris, but RedHat version 7.0 or Solaris 2.4---send us the output of "uname -a"), + what specific behavior you see (if ns doesn't compile, what's the specific error; if TCP does something strange, what exactly did it do [send a pointer to a packet trace]), + what behavior you expected to see (if ns doesn't compile this is obvious, but if TCP does something strange, why is it strange, where is the TCP spec violated?), + pointers to your script detailed output files, + a statement that "yes, I've read the FAQ, ns-problems page, and manual and I couldn't find the answer there" (or a statement about why you didn't do that yet :-) A reminder about mailing list etiquette: + Please check the web pages and list archives before posting your question. + Please keep the body of your post to simple ASCII, not HTML. + Please do _not_ send large attachments (if what you have is bigger than a few kilobytes, put it on a web page and send a URL) + Before posting a question like "did people see my post" or "the list seems down", please check the archives (you can answer this question more accurately by checking yourself rather than asking). + Please don't post subscribe/unsubscribe requests directly to the list, send them to majordomo@isi.edu (see the web page mentioned above for details). _________________________________________________________________ From davosnitti@katamail.com Fri Mar 1 05:16:11 2002 From: davosnitti@katamail.com (davosnitti@katamail.com) Date: Fri Mar 1 05:16:11 2002 Subject: [ns] Nam is compatible with Windows ME (Millennium Edition) Message-ID: <20020301095449.QGPB309.mta1@[194.153.172.176]> Hi all, I have a lot of problems with nam building process under Windows ME. I only wish to know if there is someone that had compiled nam with success, or if nobody succeeded in this "enterprise". Please help me! Let me Know somewhat on this subject. Every suggestion will be appreciated. By Davide Oscar Nitti __________________________________________ Fai i tuoi acquisti su www.kwshopping.it From dbelen@netmode.ece.ntua.gr Fri Mar 1 05:16:20 2002 From: dbelen@netmode.ece.ntua.gr (Dimitris Velenis) Date: Fri Mar 1 05:16:20 2002 Subject: [ns] Problems with the Queue Class and its "resume" function In-Reply-To: <20020228010739.D31020@jungla>; from jafernan@dit.upm.es on Thu, Feb 28, 2002 at 01:07:39AM +0100 References: <20020228010739.D31020@jungla> Message-ID: <20020301133654.B89075@netmode.ece.ntua.gr> Hi, In order to call Queue::resume() from TCL you need an appropriate entry in the Queue::command() as you said. Since you have created this entry, I think you should check the number of arguments you use. Calling resume with no arguments will invoke Queue::command() with argc = 2. Dimitris. On Thu, Feb 28, 2002 at 01:07:39AM +0100, Juan Antonio Fernandez Alonso wrote: > > Hello, > > I'm trying to unblock a queue using the "resume" procedure but I get a run-time error from ns: > > "catch "$self cmd $args" ret" > (procedure "_o1348" line 2) > (SplitObject unknown line 2) > invoked from within > "$q resume" > > I think this error is due to there isn't a command procedure defined into the Queue Class, so I cannot use the "resume" procedure implemented in queue.cc. > > I have tried to implement a command procedure for Queue Class but I still get the same error. > > Can someone help me resolve this problem? Maybe the QueueHandler has something to say here, but I don't understand how I must use it (I read in the NS manual that the "handle" function of this Class invokes the "resume" procedure, but How can I call this function from TCL Interpreter?). > > Vielen Danke, > > Juan Antonio Fernandez. -- Dimitris Velenis Network Management & Optimal Design Lab National Technical University of Athens From malvarez@dit.upm.es Fri Mar 1 05:16:29 2002 From: malvarez@dit.upm.es (Miguel Angel Alvarez) Date: Fri Mar 1 05:16:29 2002 Subject: [ns] Video traffic generator available (MPEG4) References: <3C7C3DD0.6EA4B913@sce.carleton.ca> <016401c1bf7f$f17fc940$e101048a@dit.upm.es> <3C7E8247.7DA0F2BD@sce.carleton.ca> Message-ID: <006501c1c116$89bb9da0$e101048a@dit.upm.es> This is a multi-part message in MIME format. ------=_NextPart_000_0062_01C1C11E.EB138830 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Ashraf, Thank you very much for your reply. I am not still using your code, but, = would like to use it.=20 My current work is developing a simulation of MPEG4 videoconference = service, now using Trace Traffic Source based on real MPEG4 traces of = videoconferences. I wonder which kind of videos did you use to obtain the probabilistic = distrubition of MPEG4 packets. I suppose you used movies, but, do you = have probalistic distributions of videoconference? M.A. =C1lvarez ----- Original Message -----=20 From: Ashraf Matrawy=20 To: Miguel Angel Alvarez=20 Sent: Thursday, February 28, 2002 8:17 PM Subject: Re: [ns] Video traffic generator available (MPEG4) =20 Hi Miguel,=20 Thanks for your interest in my work.=20 rateFactor_ only scales the output (up or down). It will afect both = the Max and Avg values but you can not use that to set a restriction on = the Max bitrate because this need buffering. You can not simply set a = max and make an "if" statement (if rate > max then rate=3Dmax) because = this will change the properties of the generated video.=20 Hope this helps=20 Have you installed it?=20 Please let me know if you need any help.=20 Ashraf=20 =20 Miguel Angel Alvarez wrote:=20 Hi,=20 I am very interested in using your MPEG4 traffic source. But, first, = i have=20 been reading your "mpeg4_traffic_README" file, and i wonder if it is = possible to limit the rate generated by your MPEG-4 traffic source, = with the=20 rateFactor. If possible, which relation is there between this factor = and=20 coder output bitrate, both average and peak bit rates???=20 Thank you in advance,=20 Congratulations, your work seems to be very interesting.=20 M.A. =C1lvarez=20 ----- Original Message -----=20 From: Ashraf Matrawy=20 To: ns-users@ISI.EDU=20 Sent: Wednesday, February 27, 2002 3:00 AM=20 Subject: [ns] Video traffic generator available (MPEG4)=20 Hi everyone,=20 I have developed a traffic generator based on TES (Transform Expand = Sample)=20 model of MPEG4 trace files.=20 It generates traffic that has the same first and second order = statistics as=20 an original MPEG4 trace.=20 If anyone is interested, I made it available on=20 http://www.sce.carleton.ca/~amatrawy/mpeg4=20 Please use it and give me some feedback=20 thanks=20 Ashraf --=20 ---------------------------------------------- Ashraf Matrawy, Ph.D. Candidate=20 Broadband Networks Laboratory =20 Department of Systems and Computer Engineering =20 Carleton University,=20 1125 Colonel By Dr., Ottawa, Canada K1S 5B6 Phone : 613-520-2600 ext. 2913=20 http://www.sce.carleton.ca/~amatrawy =20 ------=_NextPart_000_0062_01C1C11E.EB138830 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Hi Ashraf,
 
Thank you very much for your reply. I = am not still=20 using your code, but, would like to use it.
 
My current work is developing a=20 simulation of MPEG4 videoconference service, now using Trace = Traffic=20 Source based on real MPEG4 traces of videoconferences.
 
I wonder which kind of videos did = you use=20 to obtain the probabilistic distrubition of MPEG4 packets. I suppose=20 you used movies, but, do you have probalistic distributions of=20 videoconference?
 
M.A. =C1lvarez
----- Original Message -----
From:=20 Ashraf Matrawy
Sent: Thursday, February 28, = 2002 8:17=20 PM
Subject: Re: [ns] Video traffic = generator=20 available (MPEG4)

 
Hi = Miguel,=20

Thanks for your interest in my work.=20

rateFactor_ only scales the output (up or down). It will afect both = the Max=20 and Avg values but you can not use that to set a restriction on the = Max=20 bitrate because this need buffering. You can not simply set a max and = make an=20 "if" statement (if rate > max then rate=3Dmax) because this will = change the=20 properties of the generated video.=20

Hope this helps=20

Have you installed it?=20

Please let me know if you need any help.=20

Ashraf
 =20

Miguel Angel Alvarez wrote:=20

Hi,=20

I am very interested in using your MPEG4 traffic source. But, = first, i=20 have
been reading your "mpeg4_traffic_README" file, and i wonder = if it=20 is
possible to limit the rate generated by your MPEG-4 traffic = source,=20 with the
rateFactor. If possible, which relation is there = between this=20 factor and
coder output bitrate, both  average and peak bit = rates???=20

Thank you in advance,
Congratulations, your work seems to be = very=20 interesting.=20

M.A. =C1lvarez=20

----- Original Message -----
From: Ashraf Matrawy
To:=20 ns-users@ISI.EDU
Sent: Wednesday, February 27, 2002 3:00 AM =
Subject:=20 [ns] Video traffic generator available (MPEG4)=20

Hi everyone,=20

I have developed a traffic generator based on TES (Transform = Expand=20 Sample)
model of MPEG4 trace files.
It generates traffic = that has=20 the same first and second order statistics as
an original MPEG4 = trace.=20
If anyone is interested, I made it available on
http://www.sce.carlet= on.ca/~amatrawy/mpeg4=20
Please use it and give me some feedback=20

thanks
Ashraf

-- 
----------------------------------------------
Ashraf Matrawy, Ph.D. Candidate 
Broadband Networks =
Laboratory          &nb=
sp;
Department of Systems and Computer Engineering  
Carleton University, 
1125 Colonel By Dr., Ottawa, Canada K1S 5B6
Phone : 613-520-2600 ext. 2913 
http://www.sce.carleton.ca/=
~amatrawy
 =20 ------=_NextPart_000_0062_01C1C11E.EB138830-- From rskrn@hotmail.com Fri Mar 1 06:00:04 2002 From: rskrn@hotmail.com (Sampath Kumaran) Date: Fri Mar 1 06:00:04 2002 Subject: [ns] help needed for xgraph Message-ID: Hi, Iam doing a project in ns.i tried to build multimedia application over udp.I executed the code given in "add new application/agent " in "NS by example".I dont know how to get the output in xgraph.could anyone can pleasehelp me in telling me the modifications i have to do in calling a xgraph file instead of nam file. I could not save the output(nam)and therefore iam unable to take print out. could anyone please help me in this regard. It would be very helpful for me if anyone could reply soon . Thankx in advance sampath _________________________________________________________________ Join the world’s largest e-mail service with MSN Hotmail. http://www.hotmail.com From kkco@cin.ufpe.br Fri Mar 1 06:05:01 2002 From: kkco@cin.ufpe.br (Karina Karla C. de Oliveira) Date: Fri Mar 1 06:05:01 2002 Subject: [ns] More Diffserv Examples. Message-ID: Hi, I would like to know if someone has some example of a network diffserv with more than one core router. In the ns examples, in ns/tcl/ex/diffserv, only a configuration is used with only one core router. My network configuration is: edge_router1 ------ core_router1 -------- core_router2 ------- edge_ router2 How will be the configuration for this case? Best Regards, Karina Oliveira. ********************************************************* Karina Karla Cavalcante de Oliveira Computer Science Universidade Federal de Pernambuco Centro de Informática - CIn Caixa Postal 7851 CEP: 50732-970 Recife-PE - Brasil Phone: +55 (081) 3271-8430 Email: kkco@cin.ufpe.br WWW: www.cin.ufpe.br/~kkco Phones: (HOME) +55 (081) 3338-2054 ********************************************************* From kkco@cin.ufpe.br Fri Mar 1 09:15:02 2002 From: kkco@cin.ufpe.br (Karina Karla C. de Oliveira) Date: Fri Mar 1 09:15:02 2002 Subject: [ns] Help to DiffServ Configuration... Message-ID: Dear All, I'm trying to set a diffserv configuration to the following topology: n1 ----- - n2 ----- - - n6 n3 ----- ----- edge 1 ---- core 1 ----- core 2 ----- edge 2 ----- - n7 n4 ----- - - n8 n5 ----- - "n?" are nodes that have source traffic associated, such as FTP, HTTP, Telnet, CBR and On/Off (on/off traffic is representing an voice traffic). For example, n5 has Telnet, HTTPClient and On/Off. All nodes have more than one type of source associated. I want configure the voice (on/off) traffic as EF Traffic and the others traffics as BE Traffic. How will be then this configuration? Note that I have two core routers and more than one source per node. Please, some ideas or contributions are welcome. P.S. I saw the examples in ns/tcl/ex/diffserv and in the chapter 9 in ns manual, but they don't represent my study case. Thanks in advance, Karina Oliveira. ********************************************************* Karina Karla Cavalcante de Oliveira Computer Science Universidade Federal de Pernambuco Centro de Informática - CIn Caixa Postal 7851 CEP: 50732-970 Recife-PE - Brasil Phone: +55 (081) 3271-8430 Email: kkco@cin.ufpe.br WWW: www.cin.ufpe.br/~kkco Phones: (HOME) +55 (081) 3338-2054 ********************************************************* From xuanc@ISI.EDU Fri Mar 1 09:25:02 2002 From: xuanc@ISI.EDU (Xuan Chen) Date: Fri Mar 1 09:25:02 2002 Subject: [ns] Re: webcache - multi-clients and multi-caches (fwd) In-Reply-To: Message-ID: I am sorry that I am not familiar with this part in ns. You may want to search for the mailing list archives or ask the question to ns-users@isi.edu (NOT to ns-users-admin directly!), which is the reason I forward your email to ns-users@isi.edu. Hope somebody there can give you more help. Cheers, -chen On Fri, 1 Mar 2002, cs152 Student wrote: > > Hi Xuan, > > You replied to my email but there were no comments. I didn't understand > what you meant to do so I am replying back to you. > > I have noticed that you give very useful and helpful replies to questions. > I appreciate it if you can direct me in the right direction. Like I said, > I got confused reading the archieved emails (some of yours in fact.) > > Stephanie > > > > On Thu, 28 Feb 2002, Xuan Chen wrote: > > > ---------- Forwarded message ---------- > > Date: Thu, 28 Feb 2002 16:32:33 -0800 (PST) > > From: cs152 Student > > Subject: webcache - multi-clients and multi-caches > > > > > > I am trying to perform an experiment to find the performance of web-cache > > using NS v2.1b81. I need to set up an experiment with multiple > > client-cache configuration connected to a router that is connected to > > multiple servers. > > > > I studied the example tcl but couldn't modify it to work for my configuration. > > I read other emails related to my problem but it confused me more because > > some emails said my configuration is not supported in NS and others said > > it is supported. > > > > Would you tell me if I can set-up my configuration and if there is any > > example TCL I can look at? Learning NS is taking longer than I expected > > and I have only 2 more weeks to complete my project. > > > > I appreciate your prompt response. Thank you. > > Stephanie > > > -- Xuan Chen USC/ISI From xuanc@ISI.EDU Fri Mar 1 09:35:02 2002 From: xuanc@ISI.EDU (Xuan Chen) Date: Fri Mar 1 09:35:02 2002 Subject: [ns] Re: Problem with DSR sample script In-Reply-To: <200202281440220060.000FCF79@smtp.idola.net.id> Message-ID: These example scripts might be out of date, you may want to look at ns/tcl/test/test-suite-wireless-lan-newnode.tcl for the latest updates. It contains both DSR and MIP. To better assist you to understand a test-suite, you may want to read chapter 23 (test suite support) in ns manual. Thanks. Hope it helps, -chen On Thu, 28 Feb 2002, Lukman Hakim wrote: > Hi.. I hope you can help me for some tutorial script that doesnt work at my NS... > > I'm running ns-2.1b8b on RedHat Linux 7.1 the installation process just fine and working well with other sample script. But when I try to execute the sample script (wireless-mip-test.tcl) and got some result like this : > > [lukman@northpole ex]$ ns wireless-mip-test.tcl > no value given for parameter "errproc" to "_o33" > (Node/MobileNode add-interface line 1) > invoked from within > "$node add-interface $chan $prop $opt(ll) $opt(mac) $opt(ifq) > $opt(ifqlen) $opt(netif) $opt(ant)" > (procedure "create-dsr-bs-node" line 5) > invoked from within > "create-$opt(rp)-bs-node $node $id" > (procedure "create-base-station-node" line 21) > invoked from within > "create-base-station-node 1.0.0" > invoked from within > "set HA [create-base-station-node 1.0.0]" > (file "wireless-mip-test.tcl" line 137) > > and also for script wireless3.tcl that I already modified for adhoc routing protocol from DSDV to DSR, and also got an error message like this : > > [lukman@northpole examples]$ ns wireless3.tcl > num_nodes is set 3 > warning: Please use -channel as shown in tcl/ex/wireless-mitf.tcl > can't read "ragent_": no such variable > while executing > "subst $[subst $var" > (procedure "_o36" line 5) > (SplitObject set line 5) > invoked from within > "$self set ragent_" > (procedure "_o36" line 3) > (Node/MobileNode agenttrace line 3) > invoked from within > "$node agenttrace $tracefd" > (procedure "_o3" line 75) > (Simulator create-wireless-node line 75) > invoked from within > "_o3 create-wireless-node 1.0.0" > ("eval" body line 1) > invoked from within > "eval $self create-wireless-node $args" > (procedure "_o3" line 14) > (Simulator node line 14) > invoked from within > "$ns_ node 1.0.0" > invoked from within > "set HA [$ns_ node 1.0.0]" > (file "wireless3b.tcl" line 124) > > > Could anyone help me with these problem ?? Does NS work well for DSR adhoc routing ?? > > Thank's in advance guys.. > > Regards, > > -lukman- > -- Xuan Chen USC/ISI From xuanc@ISI.EDU Fri Mar 1 09:35:18 2002 From: xuanc@ISI.EDU (Xuan Chen) Date: Fri Mar 1 09:35:18 2002 Subject: [ns] More Diffserv Examples. In-Reply-To: Message-ID: You can basically config the queue between two core routers as Core/Edge Queue. Other configs should be similar to the sample scripts. Hope it helps, -chen On Fri, 1 Mar 2002, Karina Karla C. de Oliveira wrote: > > > Hi, > > I would like to know if someone has some example of a network diffserv > with more than one core router. In the ns examples, in > ns/tcl/ex/diffserv, only a configuration is used with only one core > router. > > My network configuration is: > > edge_router1 ------ core_router1 -------- core_router2 ------- edge_ > router2 > > How will be the configuration for this case? > > Best Regards, > Karina Oliveira. > ********************************************************* > Karina Karla Cavalcante de Oliveira > Computer Science > Universidade Federal de Pernambuco > Centro de Informática - CIn > Caixa Postal 7851 > CEP: 50732-970 > Recife-PE - Brasil > Phone: +55 (081) 3271-8430 > > Email: kkco@cin.ufpe.br > WWW: www.cin.ufpe.br/~kkco > Phones: (HOME) +55 (081) 3338-2054 > ********************************************************* > -- Xuan Chen USC/ISI From xuanc@ISI.EDU Fri Mar 1 09:40:01 2002 From: xuanc@ISI.EDU (Xuan Chen) Date: Fri Mar 1 09:40:01 2002 Subject: [ns] Help to DiffServ Configuration... In-Reply-To: Message-ID: Although I have not tried it by myself, but the author of the diffserv model told me that you can config the routers to support both EF and BF. I think you may need to insert two policies in the policy table and play with different policers a little bit... btw, when you figure out how this works, I would be happy to add your simulation script as a new diffserv example (If you don't mind...:)). Cheers, -chen On Fri, 1 Mar 2002, Karina Karla C. de Oliveira wrote: > > > Dear All, > > I'm trying to set a diffserv configuration to the following topology: > > n1 ----- - > n2 ----- - - n6 > n3 ----- ----- edge 1 ---- core 1 ----- core 2 ----- edge 2 ----- - n7 > n4 ----- - - n8 > n5 ----- - > > > "n?" are nodes that have source traffic associated, such as FTP, HTTP, > Telnet, CBR and On/Off (on/off traffic is representing an voice traffic). > > For example, n5 has Telnet, HTTPClient and On/Off. All nodes have more > than one type of source associated. > > I want configure the voice (on/off) traffic as EF Traffic and the others > traffics as BE Traffic. How will be then this configuration? > > Note that I have two core routers and more than one source per node. > > Please, some ideas or contributions are welcome. > > P.S. I saw the examples in ns/tcl/ex/diffserv and in the chapter 9 in > ns manual, but they don't represent my study case. > > Thanks in advance, > Karina Oliveira. > ********************************************************* > Karina Karla Cavalcante de Oliveira > Computer Science > Universidade Federal de Pernambuco > Centro de Informática - CIn > Caixa Postal 7851 > CEP: 50732-970 > Recife-PE - Brasil > Phone: +55 (081) 3271-8430 > > Email: kkco@cin.ufpe.br > WWW: www.cin.ufpe.br/~kkco > Phones: (HOME) +55 (081) 3338-2054 > ********************************************************* > > -- Xuan Chen USC/ISI From xuanc@ISI.EDU Fri Mar 1 09:50:01 2002 From: xuanc@ISI.EDU (Xuan Chen) Date: Fri Mar 1 09:50:01 2002 Subject: [ns] Queue Limit In-Reply-To: <003101c1a7b5$9c17a900$5d9c598f@resnet.ust.hk> Message-ID: Thanks for pointing out that. We just add a note to ns manual (chapter 7.2 Example: Drop Tail) and hopefully it will help to eliminate confusions. We may move on to fix the implementation in future. Cheers, -chen On Mon, 28 Jan 2002, Jingyi wrote: > I met this problem too. As I understand, "qlim_ = n" actually means a queue size > of n-1. Therefore, when you set "qlim_ = 2", you are working with a queue of size 1, > and setting "qlim_ = 1" will already make the queue disappear. Here is the code for droptail. > > void DropTail::enque(Packet* p) > { > q_.enque(p); > if (q_.length() >= qlim_) { > q_.remove(p); > drop(p); > } > } > > If we want "qlim_ = n" to really mean a queue size of n, "if (q_.length() >= qlim_)" > may need to be modified as "if (q_.length() > qlim_) ". > > Regards, > Jingyi > > ----- Original Message ----- > From: "Xuan Chen" > To: > Cc: > Sent: Monday, January 28, 2002 2:08 AM > Subject: Re: [ns] Queue Limit > > > > Just set queue limit to 0, I did not see an error... > > > > On Tue, 22 Jan 2002, Luiz Henrique Bonani wrote: > > > > > Hi everybody!! > > > I would like to know if anybody have experienced a link without queue... > > > Would have a way to do this??? > > > In my simulation the minimun queue size I could work was 2... This number ( > > > queue size) is ralated to any ns intrinsic time reference??? > > > Thanks in advance! > > > Luiz > > > ___________________________________ > > > Luiz Henrique Bonani do Nascimento > > > Electrical Engineering Ms. Student > > > DSIF/FEEC/Unicamp - CPqD Foundation > > > email: bonani@cpqd.com.br > > > > > > > > > > -- > > Xuan Chen > > USC/ISI > > > > > > > -- Xuan Chen USC/ISI From csysmd@comp.leeds.ac.uk Fri Mar 1 09:50:12 2002 From: csysmd@comp.leeds.ac.uk (Steve) Date: Fri Mar 1 09:50:12 2002 Subject: [ns] Monitoring bytes Message-ID: This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. --193057426-267377565-1015004255=:30408 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, can anyone help me in trying to monitor properties of a simulation. I have the following script which is very simple but when I run it, xgraph does not show the expectd results. It seems to be indicating that there are no bytes flowing into the monitor (tcp sink). I am trying to work out how to do this kind of thing in ns, but I am not really getting anywhere. I have looked at as much documentation as I can find - the manual and others. I realise I am making a simple mistake, but I'm not sure what. Thanks for any help, Steve --193057426-267377565-1015004255=:30408 Content-Type: APPLICATION/x-tcl; name="monitor.tcl" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Content-Disposition: attachment; filename="monitor.tcl" IyBUaGlzIHNjcmlwdCBzdGFydGVkIGFzIGEgc2ltcGxlIHNjcmlwdCB3aXRo IFRDUCB0cnlpbmcgdG8gdHJhY2UgdXNpbmcKIyB0cmFjZSBmaWxlcwojIFNp bXBseSBzZXRzIHVwIGEgVENQIGNvbm5lY3Rpb24sIHJ1bnMgRlRQIG92ZXIg aXQgYW5kIHRyaWVzIHRvIG91dHB1dCB0cmFjZQojIE1vbml0b3JlZAoKc2V0 IG5zIFtuZXcgU2ltdWxhdG9yXSAKCiNmb3IgbmFtCnNldCBuZiBbb3BlbiBt b24ubmFtIHddCiRucyBuYW10cmFjZS1hbGwgJG5mCgojZm9yIHhncmFwaApz ZXQgZjAgW29wZW4gbW9uLnRyIHddCgpwcm9jIGZpbmlzaCB7fSB7CiAgIGds b2JhbCBucyBuZiBmMAogICAkbnMgZmx1c2gtdHJhY2UKICAgY2xvc2UgJG5m CiAgIGNsb3NlICRmMAogICBleGVjIG5hbSBtb24ubmFtICYKICAgZXhlYyB4 Z3JhcGggbW9uLnRyIC1nZW9tZXRyeSA4MDB4NDAwICYKICAgZXhpdCAwCn0K CnNldCBuMCBbJG5zIG5vZGVdCnNldCBuMSBbJG5zIG5vZGVdCnNldCBuMTAg WyRucyBub2RlXQoKJG5zIGR1cGxleC1saW5rICRuMCAkbjEgMU1iIDEwbXMg RHJvcFRhaWwKJG5zIGR1cGxleC1saW5rICRuMSAkbjEwIDFNYiAxMG1zIERy b3BUYWlsCgokbnMgZHVwbGV4LWxpbmstb3AgJG4wICRuMSBvcmllbnQgcmln aHQKJG5zIGR1cGxleC1saW5rLW9wICRuMSAkbjEwIG9yaWVudCByaWdodAoK JG5zIGR1cGxleC1saW5rLW9wICRuMCAkbjEgcXVldWVQb3MgMC41CiRucyBk dXBsZXgtbGluay1vcCAkbjEgJG4xMCBxdWV1ZVBvcyAwLjUKCiRucyBjb2xv ciAxIEJsdWUKCiNTZXR1cCBhIFRDUCBjb25uZWN0aW9uCnNldCB0Y3AgW25l dyBBZ2VudC9UQ1BdCiR0Y3Agc2V0IGNsYXNzXyAyCiRucyBhdHRhY2gtYWdl bnQgJG4wICR0Y3AKCnNldCBzaW5rIFtuZXcgQWdlbnQvVENQU2lua10KJG5z IGF0dGFjaC1hZ2VudCAkbjEwICRzaW5rCiRucyBjb25uZWN0ICR0Y3AgJHNp bmsKJHRjcCBzZXQgZmlkXyAxCgojU2V0dXAgYSBGVFAgb3ZlciBUQ1AgY29u bmVjdGlvbgpzZXQgZnRwIFtuZXcgQXBwbGljYXRpb24vRlRQXQokZnRwIGF0 dGFjaC1hZ2VudCAkdGNwCiRmdHAgc2V0IHR5cGVfIEZUUAoKcHJvYyByZWNv cmQge30gewogICAgICAgICAgZ2xvYmFsIHNpbmswIGYwCiAgICAgICAgICBz ZXQgbnMgW1NpbXVsYXRvciBpbnN0YW5jZV0KCSAgCiAgICAgICAgICAjU2V0 IHRoZSB0aW1lIGFmdGVyIHdoaWNoIHRoZSBwcm9jZWR1cmUgc2hvdWxkIGJl IGNhbGxlZCBhZ2FpbgogICAgICAgICAgc2V0IHRpbWUgMC41CgkgIAogICAg ICAgICAgI0hvdyBtYW55IGJ5dGVzIGhhdmUgYmVlbiByZWNlaXZlZCBieSB0 aGUgdHJhZmZpYyBzaW5rcz8KICAgICAgICAgIHNldCBidzAgWyRzaW5rMCBz ZXQgYnl0ZXNfXQoJICAKICAgICAgICAgICNHZXQgdGhlIGN1cnJlbnQgdGlt ZQogICAgICAgICAgc2V0IG5vdyBbJG5zIG5vd10KICAgICAgICAgICNDYWxj dWxhdGUgdGhlIGJhbmR3aWR0aCAoaW4gTUJpdC9zKSBhbmQgd3JpdGUgaXQg dG8gdGhlIGZpbGVzCiAgICAgICAgICBwdXRzICRmMCAiJG5vdyBbZXhwciAk YncwXSIKCSAgCiAgICAgICAgICAjUmVzZXQgdGhlIGJ5dGVzXyB2YWx1ZXMg b24gdGhlIHRyYWZmaWMgc2lua3MKICAgICAgICAgICRzaW5rMCBzZXQgYnl0 ZXNfIDAKCiAgICAgICAgICAjUmUtc2NoZWR1bGUgdGhlIHByb2NlZHVyZQog ICAgICAgICAgJG5zIGF0IFtleHByICRub3crJHRpbWVdICJyZWNvcmQiCn0K CiNDcmVhdGUgYSB0cmFmZmljIHNpbmsgYW5kIGF0dGFjaCBpdCB0byB0aGUg bm9kZSBuMTAKc2V0IHNpbmswIFtuZXcgQWdlbnQvTG9zc01vbml0b3JdCiRu cyBhdHRhY2gtYWdlbnQgJG4xMCAgJHNpbmswCgokbnMgYXQgMC4xICJyZWNv cmQiCiRucyBhdCAwLjEgIiRmdHAgc3RhcnQiCiRucyBhdCA5LjAgIiRmdHAg c3RvcCIKJG5zIGF0IDEwLjAgImZpbmlzaCIKCiRucyBydW4K --193057426-267377565-1015004255=:30408-- From joubert@vt.edu Fri Mar 1 10:15:02 2002 From: joubert@vt.edu (Laurent Joubert) Date: Fri Mar 1 10:15:02 2002 Subject: [ns] Queue Limit In-Reply-To: References: <003101c1a7b5$9c17a900$5d9c598f@resnet.ust.hk> Message-ID: <20020301130858.3a58df60.joubert@vt.edu> --=.oT'+a4wZ.hUrIM Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Actually, if we considering the link as a client / server model as in queueing theory, a queue limit of 2 is equivalent to a queue-limit in Queueing theory of 0. This is because if the queue-limit is 1 in ns, all packets are dropped. To be sure that a packet is actually serviced, the minimum queue size need to be 2. Regards, Laurent On Fri, 01 Mar 2002 09:45:52 -0800 (PST) Xuan Chen wrote: | | Thanks for pointing out that. | | We just add a note to ns manual (chapter 7.2 Example: Drop Tail) and | hopefully it will help to eliminate confusions. | | We may move on to fix the implementation in future. | | Cheers, | -chen | | On Mon, 28 Jan 2002, Jingyi wrote: | | > I met this problem too. As I understand, "qlim_ = n" actually means a | > queue size of n-1. Therefore, when you set "qlim_ = 2", you are | > working with a queue of size 1, and setting "qlim_ = 1" will already | > make the queue disappear. Here is the code for droptail. | > | > void DropTail::enque(Packet* p) | > { | > q_.enque(p); | > if (q_.length() >= qlim_) { | > q_.remove(p); | > drop(p); | > } | > } | > | > If we want "qlim_ = n" to really mean a queue size of n, "if | > (q_.length() >= qlim_)" may need to be modified as "if (q_.length() > | > qlim_) ". | > | > Regards, | > Jingyi | > | > ----- Original Message ----- | > From: "Xuan Chen" | > To: | > Cc: | > Sent: Monday, January 28, 2002 2:08 AM | > Subject: Re: [ns] Queue Limit | > | > | > > Just set queue limit to 0, I did not see an error... | > > | > > On Tue, 22 Jan 2002, Luiz Henrique Bonani wrote: | > > | > > > Hi everybody!! | > > > I would like to know if anybody have experienced a link without | > > > queue... Would have a way to do this??? | > > > In my simulation the minimun queue size I could work was 2... | > > > This number ( queue size) is ralated to any ns intrinsic time | > > > reference??? Thanks in advance! | > > > Luiz | > > > ___________________________________ | > > > Luiz Henrique Bonani do Nascimento | > > > Electrical Engineering Ms. Student | > > > DSIF/FEEC/Unicamp - CPqD Foundation | > > > email: bonani@cpqd.com.br | > > > | > > > | > > | > > -- | > > Xuan Chen | > > USC/ISI | > > | > > | > > | > | -- Laurent Joubert --=.oT'+a4wZ.hUrIM Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) iD8DBQE8f8O9ZVLvM+iOITkRAiiIAKCVfpQJJH4m/OeipiSxrpvKrz1zUwCfczlb zq0LA9LuibMYiBGeCmKWSrM= =1R38 -----END PGP SIGNATURE----- --=.oT'+a4wZ.hUrIM-- From karambelkarp@hotmail.com Fri Mar 1 10:25:01 2002 From: karambelkarp@hotmail.com (Prasad Karambelkar) Date: Fri Mar 1 10:25:01 2002 Subject: [ns] down-target, up-target definition? Message-ID: Hi all, Can anybody tell me where can I find the definition of down-target, up-target utility/function. It is used in ns-node.tcl to specify interfaces between two objects. Thnaks. Prasad _________________________________________________________________ Send and receive Hotmail on your mobile device: http://mobile.msn.com From najma_ismat@hotmail.com Fri Mar 1 12:55:02 2002 From: najma_ismat@hotmail.com (Najma Ismat) Date: Fri Mar 1 12:55:02 2002 Subject: [ns] Ns2 isntallation problem. help me its urgent!!!!!! Message-ID: hi, i am simulating ad hoc network usig dsr with elfn. i have modified the code of ns2 using the elfn code by g.holland. but during the installation of ns2 i am getting an error: tcp-reno.cc: In method `RenoTcpAgent::RenoTcpAgent ()': tcp-reno.cc:82: class `RenoTcpAgent' does not have any field named `thaw_timer_' tcp-reno.cc:83: class `RenoTcpAgent' does not have any field named `t_thaw_' tcp-reno.cc:86: `t_thaw_' undeclared (first use this function) tcp-reno.cc:86: (Each undeclared identifier is reported only once for each function it appears in.) tcp-reno.cc: At top level: tcp-reno.cc:89: redefinition of `RenoTcpAgent::RenoTcpAgent ()' tcp-reno.cc:82: `RenoTcpAgent::RenoTcpAgent ()' previously defined here tcp-reno.cc:93: `pkt' was not declared in this scope tcp-reno.cc:94: syntax error before `*' tcp-reno.cc: In method `void RenoTcpAgent::recv (Packet *, Handler *)': tcp-reno.cc:102: `hdr_icmp' undeclared (first use this function) tcp-reno.cc:102: `icmp' undeclared (first use this function) tcp-reno.cc:102: parse error before `)' make: *** [tcp-reno.o] Error 1 Ns make failed! Can anyone guide me how to avoid this error .tcl Thanks in advance. Najma Ismat. _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp. From chirra_ns@yahoo.com Fri Mar 1 13:40:02 2002 From: chirra_ns@yahoo.com (sunil ns) Date: Fri Mar 1 13:40:02 2002 Subject: [ns] Reg ZRP(urgent!!!!) Message-ID: <20020301213639.40319.qmail@web21003.mail.yahoo.com> hi, Does any body know whether Zone Routing protocol is implemented in ns-2 simulator before.If it is not then is it possible to do so? Thanks in advance Sunil __________________________________________________ Do You Yahoo!? Yahoo! Greetings - Send FREE e-cards for every occasion! http://greetings.yahoo.com From sshyam10@yahoo.com Fri Mar 1 14:25:02 2002 From: sshyam10@yahoo.com (Shyam Subramanian) Date: Fri Mar 1 14:25:02 2002 Subject: [ns] help me with - ns installation on win nt References: <20020301213639.40319.qmail@web21003.mail.yahoo.com> Message-ID: <000801c1c12c$3031b3d0$dd12d883@EE.UNLV.EDU> Hi, I am a new user to ns. i was trying to install ns on windows and following the instructions given at http://www.isi.edu/nsnam/ns/ns-win32-build.html. I get the following error when i execute c:\progra~1\micros~3\vc98\bin\nmake /f makefile.vc NMAKE : fatal error v1073:don't know hot to make 'random.o'. (i did edit the make file before executing this command) Can someone who has installed ns on windows nt tell me the detailed procedure to install. I would be really thankful if u can do that. Any pointers are welcome. Thanks in advance Shyam _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com From ratancha@uiuc.edu Fri Mar 1 14:40:01 2002 From: ratancha@uiuc.edu (Prashant Ratanchandani) Date: Fri Mar 1 14:40:01 2002 Subject: [ns] Reg ZRP(urgent!!!!) References: <20020301213639.40319.qmail@web21003.mail.yahoo.com> Message-ID: <3C80028D.5C29B89D@uiuc.edu> sunil ns wrote: > > hi, > Does any body know whether Zone Routing protocol is > implemented in ns-2 simulator before.If it is not then > is it possible to do so? > > Thanks in advance > Sunil > Yes, it has..Robin Poss did this implementation for ns-2.1b7a(to the best of my knowledge)..it is not available off the web but you could send him e-mail and he will mail it to you. Best Regards, Prashant -- Prashant Ratanchandani "If I were two-faced, would I be wearing this one?" Abraham Lincoln From lyn555@hotmail.com Fri Mar 1 18:35:06 2002 From: lyn555@hotmail.com (wayne lam) Date: Fri Mar 1 18:35:06 2002 Subject: [ns] Invalid command Message-ID: This is a multi-part message in MIME format. ------=_NextPart_000_0005_01C1C1D6.090E78B0 Content-Type: text/plain; charset="big5" Content-Transfer-Encoding: quoted-printable Hi all, I encoutered a invalid command problem when i write a "for loop" = or "if statement"=20 for {set i 0} {$i < 5} {incr i} { $ns at 1.0 "$cbr($i) start" } something like that in tcl script. could anyone help me ?? Thanks in advances ------=_NextPart_000_0005_01C1C1D6.090E78B0 Content-Type: text/html; charset="big5" Content-Transfer-Encoding: quoted-printable
Hi all, I encoutered a invalid command problem when = i write a=20 "for loop" or "if statement"
for {set i 0} {$i < 5} {incr i} = {
    $ns=20 at 1.0 "$cbr($i) start"
}
something like that in tcl = script.
could anyone help me ??
Thanks in advances
------=_NextPart_000_0005_01C1C1D6.090E78B0-- From madhan80@eth.net Sat Mar 2 05:15:09 2002 From: madhan80@eth.net (Madhan Mohan) Date: Sat Mar 2 05:15:09 2002 Subject: [ns] Packet lost in ll.cc Message-ID: <001601c1c211$1e229520$aab409ca@madhan> I find that certain packets whose arp had not been resolved are lost in the link layer. ( ll.cc ) tx = arptable_->arpresolve(dst, p, this); // arp resolution if (tx == 0) { s.schedule(downtarget_, p, Random::uniform(delay_)); } // If arp is resolved the packet is sent to the downtarget_ But what if tx !=0. I find that this case is not being handled. Many packets are being lost because of this. Please help me to sort out this problem. I am using ns-2.1b6. Thank You Madhan http://annauniv.edu/ece/nrg From 50123604@student.cityu.edu.hk Sat Mar 2 05:15:22 2002 From: 50123604@student.cityu.edu.hk (50123604@student.cityu.edu.hk) Date: Sat Mar 2 05:15:22 2002 Subject: [ns] urgent question!! PLEASE!! Message-ID: <1015061572.mailspinnerdV3.2.5.3@plsmtp.cityu.edu.hk> Dear sir, I encoutered a problem about matrix set i 0.5 set j 0.5 set tran_matrix {{$i $j} {$j,$i} set trans [lindex $tran_matrix 0] puts [lindex $trans 0] The output result was $i !!!! what happened???? The expected result should be 0.5!!!! set i 0.5 set j 0.5 set tran_matrix {{i j} {j,i} set trans [lindex $tran_matrix 0] puts [lindex $trans 0] The output result was i !!! what happened again??? How can I modify the elements in the matrix and make the corresponding change??? Thanks in advance!!! Wayne From 50123604@student.cityu.edu.hk Sat Mar 2 05:15:32 2002 From: 50123604@student.cityu.edu.hk (50123604@student.cityu.edu.hk) Date: Sat Mar 2 05:15:32 2002 Subject: [ns] urgent question!! PLEASE!! Message-ID: <1015061664.mailspinnerdV3.2.5.3@plsmtp.cityu.edu.hk> Dear sir, I encoutered a problem about matrix set i 0.5 set j 0.5 set tran_matrix {{$i $j} {$j,$i} set trans [lindex $tran_matrix 0] puts [lindex $trans 0] The output result was $i !!!! what happened???? The expected result should be 0.5!!!! set i 0.5 set j 0.5 set tran_matrix {{i j} {j,i} set trans [lindex $tran_matrix 0] puts [lindex $trans 0] The output result was i !!! what happened again??? How can I modify the elements in the matrix and make the corresponding change??? Thanks in advance!!! Wayne From jbyfr@263.net Sat Mar 2 05:15:43 2002 From: jbyfr@263.net (fanrui) Date: Sat Mar 2 05:15:43 2002 Subject: [ns] call admission control Message-ID: <200203021004.g22A4Bt17631@tnt.isi.edu> ns-users£¬ who have used ns to do simulation on call admission control in wireless network ¡¡¡¡ ¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡fanrui ¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡jbyfr@263.net ¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡¡2002-03-02 From asaf_shabtai@hotmail.com Sat Mar 2 07:55:02 2002 From: asaf_shabtai@hotmail.com (Asaf Shabtai) Date: Sat Mar 2 07:55:02 2002 Subject: [ns] Counting packets sent by nodes Message-ID: Hi, I'm tring to add a variable in the node, so each time a packet is sent by the node to the next node te variable is change (for example let's say that the each time the node sends a packet its variable is incremented by 7). Does anybody have an idea how to implenet that ? Thank U Asaf _________________________________________________________________ Join the world’s largest e-mail service with MSN Hotmail. http://www.hotmail.com From chirra_ns@yahoo.com Sat Mar 2 11:45:02 2002 From: chirra_ns@yahoo.com (sunil ns) Date: Sat Mar 2 11:45:02 2002 Subject: [ns] Reg simple-wireless.tcl (help required) Message-ID: <20020302194253.27642.qmail@web21007.mail.yahoo.com> hi, I suceessfully installed ns2.But iam getting the following errors when i try to run simple-wireless.tcl in the package.can anybody help me. num_nodes is set 2 WirelessPhy: channel lookup of Channel/WirelessChannel failed (_o17 cmd line 1) invoked from within "_o17 cmd channel Channel/WirelessChannel" invoked from within "catch "$self cmd $args" ret" (procedure "_o17" line 2) (SplitObject unknown line 2) invoked from within "$netif channel $channel" (procedure "_o13" line 98) (Node/MobileNode add-interface line 98) invoked from within "$node add-interface $chan $propInstance_ $llType_ $macType_ $ifqType_ $ifqlen_ $phyType_ $antType_ $errProc_ $FECProc_" (procedure "_o3" line 54) (Simulator create-wireless-node line 54) invoked from within "_o3 create-wireless-node" ("eval" body line 1) invoked from within "eval $self create-wireless-node $args" (procedure "_o3" line 14) (Simulator node line 14) invoked from within "$ns_ node" ("for" body line 2) invoked from within "for {set i 0} {$i < $val(nn) } {incr i} { set node_($i) [$ns_ node] $node_($i) random-motion 0 ;# disable random motion }" (file "simple-wireless.tcl" line 93) Thanks in advance __________________________________________________ Do You Yahoo!? Yahoo! Sports - sign up for Fantasy Baseball http://sports.yahoo.com From jg35@oak.njit.edu Sat Mar 2 12:30:01 2002 From: jg35@oak.njit.edu (Jeu George) Date: Sat Mar 2 12:30:01 2002 Subject: [ns] Different Queues for differnt Routers Message-ID: Hello, Does anyone know if it is possible to simulate a router with multiple queues each having its own queing mechanisms. Is this possible in NS. Or is it that only a sinlge queue can be simulated on a network Thanks Jeu From najma_ismat@hotmail.com Sat Mar 2 15:00:01 2002 From: najma_ismat@hotmail.com (Najma Ismat) Date: Sat Mar 2 15:00:01 2002 Subject: [ns] Getting error in tcp-reno.cc. Urgent help is needed!!!!!!!!!!!!! Message-ID: Hi, I am simulating ad hoc network using dsr with elfn. i have modified the ns2 code using the elfn code by g.holland. the problem i am facing during ns-2.1b8a build up in tcp-reno.cc. i have consulted the ns2 manual but i am not able to solve it.if anyone can guide then it will be really helpful me. i am attaching the modified code of tcp-reno.cc & tcp.h along with the errors I am getting during installation. for tcp.h i checked the code and it is the same as given in the manual. i am thankful in advance. Najma Ismat. ######################Errors########################################## In file included from tcp-reno.cc:30: tcp.h:138: parse error at null character tcp.h:504: warning: This file contains more `}'s than `{'s. tcp-reno.cc:122: ISO C++ forbids declaration of `RenoTcpAgent' with no type tcp-reno.cc:122: no `int TcpAgent::RenoTcpAgent ()' member function declared in class `TcpAgent' tcp-reno.cc: In method `int TcpAgent::RenoTcpAgent ()': tcp-reno.cc:122: only constructors take base initializers tcp-reno.cc:122: class `TcpAgent' does not have any field named `dupwnd_' tcp-reno.cc:124: warning: no return statement in function returning non-void make: *** [tcp-reno.o] Error 1 Ns make failed! ###########################tcp-reno.cc(modified)###################### int RenoTcpAgent::window() { // // reno: inflate the window by dupwnd_ // dupwnd_ will be non-zero during fast recovery, // at which time it contains the number of dup acks // int win = int(cwnd_) + dupwnd_; if (win > int(wnd_)) win = int(wnd_); return (win); } double RenoTcpAgent::windowd() { // // reno: inflate the window by dupwnd_ // dupwnd_ will be non-zero during fast recovery, // at which time it contains the number of dup acks // double win = cwnd_ + dupwnd_; if (win > wnd_) win = wnd_; return (win); } #ifdef ELFN_ENABLE RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0), thaw_timer_(this), t_thaw_(0) { // Defaults for bound variables should be set in ns-default.tcl. bind("t_thaw_", &t_thaw_); } #else ELFN_ENABLE RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0){ [error in this line] } #endif ELFN_ENABLE ##########################tcp.h####################################### class RtxTimer : public TimerHandler { public: RtxTimer(TcpAgent *a) : TimerHandler() { a_ = a; } [error in thsi line] protected: virtual void expire(Event *e); TcpAgent *a_; } }; _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp. From chakra10@msu.edu Sat Mar 2 15:25:01 2002 From: chakra10@msu.edu (Gautam - Chakrabarti) Date: Sat Mar 2 15:25:01 2002 Subject: [ns] Getting error in tcp-reno.cc. Urgent help is needed!!!!!!!!!!!!! References: Message-ID: <3C815EF8.557BC594@msu.edu> there is a parse error in ur tcp.h file, check it up, it seems it has a {} mismatch there. That may lead to some other errors....also check if you have used the flag ELFN_ENABLE everywhere properly, that is, if u have commented out some class methods/member variables by mistake using the flag..., since you have the errors in the ELFN_ENABLE not-defined portion.... also probably u don't need to specify the flag-name after #else --- exact behavior depends on the compiler...and the portion of code that u have sent has an extra '}' at the end, check if it is a problem ---- oh yeah, the last portion of the code u have sent is from tcp.h --- so that is at least 1 of the problems -- an extra '}' hth, gautam Najma Ismat wrote: > > Hi, > I am simulating ad hoc network using dsr with elfn. i have modified the ns2 > code using the elfn code by g.holland. the problem i am facing during > ns-2.1b8a build up in tcp-reno.cc. i have consulted the ns2 manual but i am > not able to solve it.if anyone can guide then it will be really helpful me. > i am attaching the modified code of tcp-reno.cc & tcp.h along with the > errors I am getting during installation. for tcp.h i checked the code and > it is the same as given in the manual. > i am thankful in advance. > Najma Ismat. > > ######################Errors########################################## > In file included from tcp-reno.cc:30: > tcp.h:138: parse error at null character > tcp.h:504: warning: This file contains more `}'s than `{'s. > tcp-reno.cc:122: ISO C++ forbids declaration of `RenoTcpAgent' with no > type > tcp-reno.cc:122: no `int TcpAgent::RenoTcpAgent ()' member function > declared in class `TcpAgent' > tcp-reno.cc: In method `int TcpAgent::RenoTcpAgent ()': > tcp-reno.cc:122: only constructors take base initializers > tcp-reno.cc:122: class `TcpAgent' does not have any field named > `dupwnd_' > tcp-reno.cc:124: warning: no return statement in function returning > non-void > make: *** [tcp-reno.o] Error 1 > Ns make failed! > ###########################tcp-reno.cc(modified)###################### > > int RenoTcpAgent::window() > { > // > // reno: inflate the window by dupwnd_ > // dupwnd_ will be non-zero during fast recovery, > // at which time it contains the number of dup acks > // > int win = int(cwnd_) + dupwnd_; > if (win > int(wnd_)) > win = int(wnd_); > return (win); > } > > double RenoTcpAgent::windowd() > { > // > // reno: inflate the window by dupwnd_ > // dupwnd_ will be non-zero during fast recovery, > // at which time it contains the number of dup acks > // > double win = cwnd_ + dupwnd_; > if (win > wnd_) > win = wnd_; > return (win); > } > #ifdef ELFN_ENABLE > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0), thaw_timer_(this), > t_thaw_(0) > { > // Defaults for bound variables should be set in ns-default.tcl. > bind("t_thaw_", &t_thaw_); > } > #else ELFN_ENABLE > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0){ [error in this line] > } > #endif ELFN_ENABLE > ##########################tcp.h####################################### > class RtxTimer : public TimerHandler { > > public: > > RtxTimer(TcpAgent *a) : TimerHandler() { a_ = a; } [error in thsi line] > > protected: > > virtual void expire(Event *e); > TcpAgent *a_; > > } > > }; > > _________________________________________________________________ > Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp. -- *************************************************************** Gautam Chakrabarti MS Student Computer Sc. and Engg Department Michigan State University Email : chakra10@msu.edu Home Page : http://www.cse.msu.edu/~chakra10 Why is there so much month left at the end of the money? --- John Barrymore *************************************************************** From xuanc@ISI.EDU Sat Mar 2 17:00:06 2002 From: xuanc@ISI.EDU (Xuan Chen) Date: Sat Mar 2 17:00:06 2002 Subject: [ns] Different Queues for differnt Routers In-Reply-To: Message-ID: YOu may want to look at the diffserv model in ns, which allows you to simulate multiple queues for one router. Hope it helps, -chen On Sat, 2 Mar 2002, Jeu George wrote: > > Hello, > Does anyone know if it is possible to simulate a router with multiple > queues each having its own queing mechanisms. Is this possible in NS. Or > is it that only a sinlge queue can be simulated on a network > > Thanks > Jeu > -- Xuan Chen USC/ISI From rlaw@vt.edu Sat Mar 2 17:35:02 2002 From: rlaw@vt.edu (Raymond Law) Date: Sat Mar 2 17:35:02 2002 Subject: [ns] Processing delay Message-ID: <5.1.0.14.0.20020302203917.03c35ae8@mail.vt.edu> The queue_ in an ns node models the queuing delay, and the link_ in an ns node models the transmission delay. Propagation delay can also be specified when creating a link. Is there a way to model the processing delay in the node? Also, is there a good way to measure the end-to-end delay? Thanks. Ray, From c16917@yahoo.com Sat Mar 2 18:35:01 2002 From: c16917@yahoo.com (Lisa Wu) Date: Sat Mar 2 18:35:01 2002 Subject: [ns] Broadcast over satellite link Urgent!! Message-ID: <20020303023425.52580.qmail@web12401.mail.yahoo.com> Hi, there, I am working on MAC protocol over satellite link. I need to set up a system model as the figure below: terminal------geo-------terminal terminal . . terminal Therefore, the geo-satellite node should have the broadcast function. I want to know if that possible? Lisa __________________________________________________ Do You Yahoo!? Yahoo! Sports - sign up for Fantasy Baseball http://sports.yahoo.com From dhiman@cs.bu.edu Sat Mar 2 19:15:02 2002 From: dhiman@cs.bu.edu (Dhiman Barman) Date: Sat Mar 2 19:15:02 2002 Subject: [ns] trace queue Message-ID: <20020302220448.A25202@cs.bu.edu> Hi, I want to trace only the errormodel drops on a link. I am using trace-queue which dumps everything. May I know which files/functions do the dumps for the trace-queue. -Dhiman -- From kopi@stanford.edu Sun Mar 3 05:15:10 2002 From: kopi@stanford.edu (Milind Dilip Kopikare) Date: Sun Mar 3 05:15:10 2002 Subject: [ns] On the mac-802_11 Message-ID: Hi, In mac-802_11.cc, the following code in void Mac802_11::recv(Packet *p, Handler *h) function:- ==================== /* * If the power of the incoming packet is smaller than the * power of the packet currently being received by at least * the capture threshold, then we ignore the new packet. */ if(pktRx_->txinfo_.RxPr / p->txinfo_.RxPr >= p->txinfo_.CPThresh) { capture(p); } else { collision(p); } ==================== Dosen't seem right. Shouldn't we have collision if the condition in the if() condition is true? Thanks, Milind From ongun@ee.bilkent.edu.tr Sun Mar 3 05:15:20 2002 From: ongun@ee.bilkent.edu.tr (Ongun Yucesan) Date: Sun Mar 3 05:15:20 2002 Subject: [ns] Different Queues for differnt Routers References: Message-ID: <010601c1c2a5$64833f80$bc0cb38b@ee> By modifying the queueing mechanisms you can do it. For example take FIFO into hand. You can also check the Diffserv queues. 'Ns By example' site was giving a really good inside about this issue. Ongun Yucesan ----- Original Message ----- From: "Jeu George" To: "NS User's" Sent: Saturday, March 02, 2002 10:25 PM Subject: [ns] Different Queues for differnt Routers > > Hello, > Does anyone know if it is possible to simulate a router with multiple > queues each having its own queing mechanisms. Is this possible in NS. Or > is it that only a sinlge queue can be simulated on a network > > Thanks > Jeu > > > From shivanajaym@yahoo.com Sun Mar 3 06:30:06 2002 From: shivanajaym@yahoo.com (Shivanajay Marwaha) Date: Sun Mar 3 06:30:06 2002 Subject: [ns] Scheduler going backwards in time In-Reply-To: Message-ID: <20020303142827.84957.qmail@web12407.mail.yahoo.com> --0-1307666506-1015165707=:83524 Content-Type: text/plain; charset=us-ascii Hi, I am getting this error sometimes in my simulation. "Scheduler going backwards in time from time 599.33 to 598.23" What's the reason for this. Thanks, Shivanajay --------------------------------- Do You Yahoo!? Yahoo! Sports - Sign up for Fantasy Baseball --0-1307666506-1015165707=:83524 Content-Type: text/html; charset=us-ascii

 Hi,

I am getting this error sometimes in my simulation.

"Scheduler going backwards in time from time 599.33 to 598.23"

What's the reason for this.

Thanks,

Shivanajay


Do You Yahoo!?
Yahoo! Sports - Sign up for Fantasy Baseball --0-1307666506-1015165707=:83524-- From melvin77@singnet.com.sg Sun Mar 3 06:35:02 2002 From: melvin77@singnet.com.sg (meLVin) Date: Sun Mar 3 06:35:02 2002 Subject: [ns] Compiling STL dependent files Message-ID: <000a01c1c2c0$f465b4a0$73bb15a5@omni300000> This is a multi-part message in MIME format. ------=_NextPart_000_0007_01C1C304.01A72020 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hello.... I am doing an installation of NS on windows and I need to build=20 STL-dependent object files (i've have added some cc files that accessed=20 the STL eg ). However these files failed to get compiled and=20 returned errors like "unwind semantics not enabled. Specifg -GX" and=20 other syntax errors due to the STL not being able to be read. I noticed=20 in the ns-2.1b8a-win\conf\makefile.win the following message: # Disable building STL-dependent object files because VC6.x=20 does # not seem to support G++-style (2.8.0 and up) STL. OBJ_STL =3D3D NS_TCL_LIB_STL=3D3D=3D20 How do i actually get the VC++ compiler to compile the STL cc files????? = What code do i have to put into the makefile.vc or makefile.win????? thanks!!!! melvin ------=_NextPart_000_0007_01C1C304.01A72020 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Hello....

    I am doing an = installation=20 of NS on windows and I need to build
STL-dependent object files = (i've have=20 added some cc files that accessed
the STL eg <string>). = However these=20 files failed to get compiled and
returned errors like "unwind = semantics not=20 enabled. Specifg -GX" and
other syntax errors due to the STL not = being able=20 to be read. I noticed
in the ns-2.1b8a-win\conf\makefile.win the = following=20 message:
          &= nbsp; #=20 Disable building STL-dependent object files because VC6.x=20
does

         &nb= sp; =20 # not seem to support G++-style (2.8.0 and up)=20 STL.

          &= nbsp;=20 OBJ_STL=20 =3D3D

          =  =20 NS_TCL_LIB_STL=3D3D=3D20



How do i actually get the VC++ = compiler to=20 compile the STL cc files?????
What code do i have to put into the=20 makefile.vc or=20 makefile.win?????

thanks!!!!

melvin


 ------=_NextPart_000_0007_01C1C304.01A72020-- From shivanajaym@yahoo.com Sun Mar 3 06:35:12 2002 From: shivanajaym@yahoo.com (Shivanajay Marwaha) Date: Sun Mar 3 06:35:12 2002 Subject: [ns] Scheduler going backwards in time In-Reply-To: Message-ID: <20020303143005.86073.qmail@web12401.mail.yahoo.com> --0-170516676-1015165805=:84306 Content-Type: text/plain; charset=us-ascii Hi, I'm getting this error sometimes in my simulation. "Scheduler going backwards in time 599.33 to 598.23" What's the reason for this. Thanks, Shivanajay --------------------------------- Do You Yahoo!? Yahoo! Sports - Sign up for Fantasy Baseball --0-170516676-1015165805=:84306 Content-Type: text/html; charset=us-ascii

Hi,

I'm getting this error sometimes in my simulation.

"Scheduler going backwards in time 599.33 to 598.23"

What's the reason for this.

Thanks,

Shivanajay


Do You Yahoo!?
Yahoo! Sports - Sign up for Fantasy Baseball --0-170516676-1015165805=:84306-- From kopi@stanford.edu Sun Mar 3 08:15:05 2002 From: kopi@stanford.edu (Milind Dilip Kopikare) Date: Sun Mar 3 08:15:05 2002 Subject: [ns] On the mac-802_11 Message-ID: Hi, In mac-802_11.cc, the following code in void Mac802_11::recv(Packet *p, Handler *h) function:- ==================== /* * If the power of the incoming packet is smaller than the * power of the packet currently being received by at least * the capture threshold, then we ignore the new packet. */ if(pktRx_->txinfo_.RxPr / p->txinfo_.RxPr >= p->txinfo_.CPThresh) { capture(p); } else { collision(p); } ==================== Dosen't seem right. Shouldn't we have collision when the if() condition is true? Thanks, Milind From najma_ismat@hotmail.com Sun Mar 3 13:00:01 2002 From: najma_ismat@hotmail.com (Najma Ismat) Date: Sun Mar 3 13:00:01 2002 Subject: [ns] Getting error in tcp-reno.cc. Urgent help is needed!!!!!!!!!!!!! Message-ID: Hi, i have made changes you suggested and able to resolve some of the errors but the now the error i am getting is really b/c i compared that piece of code with the original one and it is the same but still i am getting an error during ns2 buildup. still needs guidence. thanks alot in advance. Najma Ismat. ********************************************************************** tcp-reno.cc:227: ISO C++ forbids declaration of `allow_fast_retransmit' with no type tcp-reno.cc:227: no `int RenoTcpAgent::allow_fast_retransmit (int)' member function declared in class `RenoTcpAgent' tcp-reno.cc:250: no `void RenoTcpAgent::dupack_action ()' member function declared in class `RenoTcpAgent' tcp-reno.cc: In method `void RenoTcpAgent::dupack_action ()': tcp-reno.cc:253: `allow_fast_retransmit' undeclared (first use this function) tcp-reno.cc:253: (Each undeclared identifier is reported only once for each function it appears in.) make: *** [tcp-reno.o] Error 1 Ns make failed! **********************tcp-reno.cc code******************************* RenoTcpAgent::allow_fast_retransmit(int last_cwnd_action_) { return (last_cwnd_action_ == CWND_ACTION_DUPACK); } void RenoTcpAgent::dupack_action() [getting error here] { int recovered = (highest_ack_ > recover_); int allowFastRetransmit=allow_fast_retransmit(last_cwnd_action_);[getting error here] if (recovered || (!bug_fix_ && !ecn_) || allowFastRetransmit) { goto reno_action; } if (ecn_ && last_cwnd_action_ == CWND_ACTION_ECN) { last_cwnd_action_ = CWND_ACTION_DUPACK; reset_rtx_timer(1,0); output(last_ack_ + 1, TCP_REASON_DUPACK); return; } if (bug_fix_) { return; } >From: Gautam - Chakrabarti >To: Najma Ismat , NS Users >Subject: Re: [ns] Getting error in tcp-reno.cc. Urgent help is >needed!!!!!!!!!!!!! >Date: Sat, 02 Mar 2002 18:23:36 -0500 > > >there is a parse error in ur tcp.h file, check it up, it seems it has a >{} mismatch there. That may lead to some other errors....also check if >you have used the flag ELFN_ENABLE everywhere properly, that is, if u >have commented out some class methods/member variables by mistake using >the flag..., since you have the errors in the ELFN_ENABLE not-defined >portion.... also probably u don't need to specify the flag-name after >#else --- exact behavior depends on the compiler...and the portion of >code that u have sent has an extra '}' at the end, check if it is a >problem ---- oh yeah, the last portion of the code u have sent is from >tcp.h --- so that is at least 1 of the problems -- an extra '}' > > >hth, >gautam > >Najma Ismat wrote: > > > > Hi, > > I am simulating ad hoc network using dsr with elfn. i have modified the >ns2 > > code using the elfn code by g.holland. the problem i am facing during > > ns-2.1b8a build up in tcp-reno.cc. i have consulted the ns2 manual but i >am > > not able to solve it.if anyone can guide then it will be really helpful >me. > > i am attaching the modified code of tcp-reno.cc & tcp.h along with the > > errors I am getting during installation. for tcp.h i checked the code >and > > it is the same as given in the manual. > > i am thankful in advance. > > Najma Ismat. > > > > ######################Errors########################################## > > In file included from tcp-reno.cc:30: > > tcp.h:138: parse error at null character > > tcp.h:504: warning: This file contains more `}'s than `{'s. > > tcp-reno.cc:122: ISO C++ forbids declaration of `RenoTcpAgent' with no > > type > > tcp-reno.cc:122: no `int TcpAgent::RenoTcpAgent ()' member function > > declared in class `TcpAgent' > > tcp-reno.cc: In method `int TcpAgent::RenoTcpAgent ()': > > tcp-reno.cc:122: only constructors take base initializers > > tcp-reno.cc:122: class `TcpAgent' does not have any field named > > `dupwnd_' > > tcp-reno.cc:124: warning: no return statement in function returning > > non-void > > make: *** [tcp-reno.o] Error 1 > > Ns make failed! > > ###########################tcp-reno.cc(modified)###################### > > > > int RenoTcpAgent::window() > > { > > // > > // reno: inflate the window by dupwnd_ > > // dupwnd_ will be non-zero during fast recovery, > > // at which time it contains the number of dup acks > > // > > int win = int(cwnd_) + dupwnd_; > > if (win > int(wnd_)) > > win = int(wnd_); > > return (win); > > } > > > > double RenoTcpAgent::windowd() > > { > > // > > // reno: inflate the window by dupwnd_ > > // dupwnd_ will be non-zero during fast recovery, > > // at which time it contains the number of dup acks > > // > > double win = cwnd_ + dupwnd_; > > if (win > wnd_) > > win = wnd_; > > return (win); > > } > > #ifdef ELFN_ENABLE > > > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0), >thaw_timer_(this), > > t_thaw_(0) > > { > > // Defaults for bound variables should be set in ns-default.tcl. > > bind("t_thaw_", &t_thaw_); > > } > > #else ELFN_ENABLE > > > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0){ [error in this >line] > > } > > #endif ELFN_ENABLE > > ##########################tcp.h####################################### > > class RtxTimer : public TimerHandler { > > > > public: > > > > RtxTimer(TcpAgent *a) : TimerHandler() { a_ = a; } [error in >thsi line] > > > > protected: > > > > virtual void expire(Event *e); > > TcpAgent *a_; > > > > } > > > > }; > > > > _________________________________________________________________ > > Get your FREE download of MSN Explorer at >http://explorer.msn.com/intl.asp. > >-- >*************************************************************** >Gautam Chakrabarti >MS Student >Computer Sc. and Engg Department >Michigan State University > >Email : chakra10@msu.edu >Home Page : http://www.cse.msu.edu/~chakra10 > >Why is there so much month left at the end of the money? > --- John Barrymore >*************************************************************** > _________________________________________________________________ MSN Photos is the easiest way to share and print your photos: http://photos.msn.com/support/worldwide.aspx From chakra10@pilot.msu.edu Sun Mar 3 14:45:01 2002 From: chakra10@pilot.msu.edu (Gautam Chakrabarti) Date: Sun Mar 3 14:45:01 2002 Subject: [ns] Getting error in tcp-reno.cc. Urgent help is needed!!!!!!!!!!!!! In-Reply-To: from "Najma Ismat" at Mar 4, 2002 01:59:15 am Message-ID: <200203032243.g23MhJr44478@pilot13.cl.msu.edu> Hi, Assuming you have copied the proper portions of code here, there is a problem with the the function RenoTcpAgent::allow_fast_retransmit, it returns in int, but ur definition does not have any return type, due to which it takes that to be void, so first it says that to be NOTa valid member function, then u get the error in dupack_action() while taking the return value in int. Just add an int as return type to RenoTcpAgent::allow_fast_retransmit, should work. hth, gautam > > **********************tcp-reno.cc code******************************* > RenoTcpAgent::allow_fast_retransmit(int last_cwnd_action_) > { > return (last_cwnd_action_ == CWND_ACTION_DUPACK); > } > void RenoTcpAgent::dupack_action() [getting error here] > { > int recovered = (highest_ack_ > recover_); > int allowFastRetransmit=allow_fast_retransmit(last_cwnd_action_);[getting > error here] > > if (recovered || (!bug_fix_ && !ecn_) || allowFastRetransmit) { > goto reno_action; > } > if (ecn_ && last_cwnd_action_ == CWND_ACTION_ECN) { > last_cwnd_action_ = CWND_ACTION_DUPACK; > reset_rtx_timer(1,0); > output(last_ack_ + 1, TCP_REASON_DUPACK); > return; > } > > if (bug_fix_) { > return; > } > > > > > >From: Gautam - Chakrabarti > >To: Najma Ismat , NS Users > >Subject: Re: [ns] Getting error in tcp-reno.cc. Urgent help is > >needed!!!!!!!!!!!!! > >Date: Sat, 02 Mar 2002 18:23:36 -0500 > > > > > >there is a parse error in ur tcp.h file, check it up, it seems it has a > >{} mismatch there. That may lead to some other errors....also check if > >you have used the flag ELFN_ENABLE everywhere properly, that is, if u > >have commented out some class methods/member variables by mistake using > >the flag..., since you have the errors in the ELFN_ENABLE not-defined > >portion.... also probably u don't need to specify the flag-name after > >#else --- exact behavior depends on the compiler...and the portion of > >code that u have sent has an extra '}' at the end, check if it is a > >problem ---- oh yeah, the last portion of the code u have sent is from > >tcp.h --- so that is at least 1 of the problems -- an extra '}' > > > > > >hth, > >gautam > > > >Najma Ismat wrote: > > > > > > Hi, > > > I am simulating ad hoc network using dsr with elfn. i have modified the > >ns2 > > > code using the elfn code by g.holland. the problem i am facing during > > > ns-2.1b8a build up in tcp-reno.cc. i have consulted the ns2 manual but i > >am > > > not able to solve it.if anyone can guide then it will be really helpful > >me. > > > i am attaching the modified code of tcp-reno.cc & tcp.h along with the > > > errors I am getting during installation. for tcp.h i checked the code > >and > > > it is the same as given in the manual. > > > i am thankful in advance. > > > Najma Ismat. > > > > > > ######################Errors########################################## > > > In file included from tcp-reno.cc:30: > > > tcp.h:138: parse error at null character > > > tcp.h:504: warning: This file contains more `}'s than `{'s. > > > tcp-reno.cc:122: ISO C++ forbids declaration of `RenoTcpAgent' with no > > > type > > > tcp-reno.cc:122: no `int TcpAgent::RenoTcpAgent ()' member function > > > declared in class `TcpAgent' > > > tcp-reno.cc: In method `int TcpAgent::RenoTcpAgent ()': > > > tcp-reno.cc:122: only constructors take base initializers > > > tcp-reno.cc:122: class `TcpAgent' does not have any field named > > > `dupwnd_' > > > tcp-reno.cc:124: warning: no return statement in function returning > > > non-void > > > make: *** [tcp-reno.o] Error 1 > > > Ns make failed! > > > ###########################tcp-reno.cc(modified)###################### > > > > > > int RenoTcpAgent::window() > > > { > > > // > > > // reno: inflate the window by dupwnd_ > > > // dupwnd_ will be non-zero during fast recovery, > > > // at which time it contains the number of dup acks > > > // > > > int win = int(cwnd_) + dupwnd_; > > > if (win > int(wnd_)) > > > win = int(wnd_); > > > return (win); > > > } > > > > > > double RenoTcpAgent::windowd() > > > { > > > // > > > // reno: inflate the window by dupwnd_ > > > // dupwnd_ will be non-zero during fast recovery, > > > // at which time it contains the number of dup acks > > > // > > > double win = cwnd_ + dupwnd_; > > > if (win > wnd_) > > > win = wnd_; > > > return (win); > > > } > > > #ifdef ELFN_ENABLE > > > > > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0), > >thaw_timer_(this), > > > t_thaw_(0) > > > { > > > // Defaults for bound variables should be set in ns-default.tcl. > > > bind("t_thaw_", &t_thaw_); > > > } > > > #else ELFN_ENABLE > > > > > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0){ [error in this > >line] > > > } > > > #endif ELFN_ENABLE > > > ##########################tcp.h####################################### > > > class RtxTimer : public TimerHandler { > > > > > > public: > > > > > > RtxTimer(TcpAgent *a) : TimerHandler() { a_ = a; } [error in > >thsi line] > > > > > > protected: > > > > > > virtual void expire(Event *e); > > > TcpAgent *a_; > > > > > > } > > > > > > }; > > > > > > _________________________________________________________________ > > > Get your FREE download of MSN Explorer at > >http://explorer.msn.com/intl.asp. > > > >-- > >*************************************************************** > >Gautam Chakrabarti > >MS Student > >Computer Sc. and Engg Department > >Michigan State University > > > >Email : chakra10@msu.edu > >Home Page : http://www.cse.msu.edu/~chakra10 > > > >Why is there so much month left at the end of the money? > > --- John Barrymore > >*************************************************************** > > > > > > > _________________________________________________________________ > MSN Photos is the easiest way to share and print your photos: > http://photos.msn.com/support/worldwide.aspx > > -- ------------------------------------------- Gautam Chakrabarti MS Student Computer Sc. & Engineering Dept. Michigan State University Email: chakra10@msu.edu Home Page: http://www.cse.msu.edu/~chakra10 -------------------------------------------- From easy@mail.ustc.edu.cn Sun Mar 3 18:25:02 2002 From: easy@mail.ustc.edu.cn (xu min (easy@mail.ustc.edu.cn)) Date: Sun Mar 3 18:25:02 2002 Subject: [ns] about rwa using ns2 Message-ID: <200203040214.KAA28690@mx1.ustc.edu.cn> Hi, I am going to use ns2 to simulate routing and wavelength assignment algorithm. I have found no informations about it. Would someone give me some advices? Thanks a lot! Regards, Xu Min easy@mail.ustc.edu.cn From ahmedmoustafa01@yahoo.ca Sun Mar 3 19:45:08 2002 From: ahmedmoustafa01@yahoo.ca (Ahmed Moustafa) Date: Sun Mar 3 19:45:08 2002 Subject: [ns] Snoop and wireless problem Message-ID: <20020304033949.63070.qmail@web14101.mail.yahoo.com> HI all, I'm trying to simulate a wireless scenario where I use hierarichal address and snoop agent on the base station, but I'm getting the following errors: num_nodes is set 51 Loading scenario file... Load complete... Starting Simulation... can't read "Node_(4)": no such element in array while executing "set nq $Node_($q)" (procedure "_o4" line 5) (Simulator get-node-id-by-addr line 5) invoked from within "$self get-node-id-by-addr $nh" (procedure "_o4" line 48) (Simulator compute-hier-routes line 48) invoked from within "$self compute-hier-routes " (procedure "_o4" line 3) (Simulator compute-routes line 3) invoked from within "[Simulator instance] compute-routes" (procedure "Agent/rtProto/Static" line 2) (Agent/rtProto/Static init-all line 2) invoked from within "Agent/rtProto/Static init-all" (procedure "_o32" line 8) (RouteLogic configure line 8) invoked from within "[$self get-routelogic] configure" (procedure "_o4" line 4) (Simulator run line 4) invoked from within "$ns_ run" does anyone know why I'm getting these errors ? or maybe have a working example ? Thanks, Ahmed __________________________________________________ Do You Yahoo!? Yahoo! Sports - sign up for Fantasy Baseball http://sports.yahoo.com From shivanajaym@yahoo.com Mon Mar 4 05:15:03 2002 From: shivanajaym@yahoo.com (Shivanajay Marwaha) Date: Mon Mar 4 05:15:03 2002 Subject: [ns] Scheduler going backwards in time In-Reply-To: <20020303143005.86073.qmail@web12401.mail.yahoo.com> Message-ID: <20020304045804.47107.qmail@web12405.mail.yahoo.com> --0-1181048207-1015217884=:45499 Content-Type: text/plain; charset=us-ascii Hi, I'm getting this error sometimes in my simulation. "Scheduler going backwards in time 599.33 to 598.23" What's the reason for this. Thanks, Shivanajay --------------------------------- Do You Yahoo!? Yahoo! Sports - Sign up for Fantasy Baseball --0-1181048207-1015217884=:45499 Content-Type: text/html; charset=us-ascii

Hi,

I'm getting this error sometimes in my simulation.

"Scheduler going backwards in time 599.33 to 598.23"

What's the reason for this.

Thanks,

 Shivanajay


Do You Yahoo!?
Yahoo! Sports - Sign up for Fantasy Baseball --0-1181048207-1015217884=:45499-- From ellich8459@yahoo.com Mon Mar 4 05:15:18 2002 From: ellich8459@yahoo.com (elliot chen) Date: Mon Mar 4 05:15:18 2002 Subject: [ns] how to generate bursty packet losses Message-ID: <20020304055231.44310.qmail@web21404.mail.yahoo.com> Hello, all Does anyone know how to generate bursty pkt loss (consecutive pkt loss) using the current error model in ns? Thank you so much! Elliot __________________________________________________ Do You Yahoo!? Yahoo! Sports - sign up for Fantasy Baseball http://sports.yahoo.com From Marcel.Odena@icn.siemens.de Mon Mar 4 05:15:30 2002 From: Marcel.Odena@icn.siemens.de (Odena Marcel) Date: Mon Mar 4 05:15:30 2002 Subject: [ns] AW: ARP --> Marcel Odena's version of ARP code. Message-ID: <353063C297A9D3118F760008C791E269034C9CC7@MCHH263E> This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_000_01C1C357.48128970 Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: quoted-printable Hi Ahmed, Here you have the code of my version of ARP code. I am using the ns-2.1b8a, the allinone version. I already sent an e-mail where I gave some notes about my version of ARP, but I have seen that it's not yet posted in the mailing list = archive (http://www.isi.edu/nsnam/archive/ns-users/webarch/). Some important details about my version of ARP: it implements a queue = to=20 keep packets waiting for an address resolution which you can set with = the variable 'ARPQUEUE_MAX_LENGTH'. Some improvements that could be = implemented is to set a time out for the packets waiting for an address resolution=20 avoiding long delays. I did not took into account to delete the packets remaining at the = queue at the end of simulation. Robin Poss also implemented his versions of ARP. You can see it at: (http://www.geocities.com/robin_poss/zrp.html), descompress the file in = it and you will see it. There is other interesting information too. -----Urspr=FCngliche Nachricht----- Von: Ahmed Moustafa [mailto:ahmed.moustafa@sympatico.ca] Gesendet: Sonntag, 3. M=E4rz 2002 18:40 An: Marcel.Odena@icn.siemens.de Betreff: ARP Hi Marcel, Is it possible to get copy of your ARP code, and would you please tell me which verion of ns you are using. Thanks ------_=_NextPart_000_01C1C357.48128970 Content-Type: application/octet-stream; name="arp_MarcelOdena.cc" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="arp_MarcelOdena.cc" /*-*- Mode:C++; c-basic-offset:8; tab-width:8; indent-tabs-mode:t = -*-=0A= *=0A= * Copyright (c) 1997 Regents of the University of California.=0A= * All rights reserved.=0A= *=0A= * Redistribution and use in source and binary forms, with or = without=0A= * modification, are permitted provided that the following = conditions=0A= * are met:=0A= * 1. Redistributions of source code must retain the above copyright=0A= * notice, this list of conditions and the following disclaimer.=0A= * 2. Redistributions in binary form must reproduce the above = copyright=0A= * notice, this list of conditions and the following disclaimer in = the=0A= * documentation and/or other materials provided with the = distribution.=0A= * 3. All advertising materials mentioning features or use of this = software=0A= * must display the following acknowledgement:=0A= * This product includes software developed by the Computer Systems=0A= * Engineering Group at Lawrence Berkeley Laboratory.=0A= * 4. Neither the name of the University nor of the Laboratory may be = used=0A= * to endorse or promote products derived from this software = without=0A= * specific prior written permission.=0A= *=0A= * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' = AND=0A= * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, = THE=0A= * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR = PURPOSE=0A= * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE = LIABLE=0A= * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR = CONSEQUENTIAL=0A= * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE = GOODS=0A= * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS = INTERRUPTION)=0A= * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, = STRICT=0A= * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN = ANY WAY=0A= * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY = OF=0A= * SUCH DAMAGE.=0A= *=0A= * $Header: /nfs/jade/vint/CVSROOT/ns-2/arp.cc,v 1.11 2000/09/01 = 03:04:05 haoboy Exp $=0A= */=0A= =0A= =0A= /* =0A= * Ported from CMU/Monarch's code, nov'98 -Padma.=0A= *=0A= * basic arp cache and MAC addr resolution=0A= *=0A= * Note: code in this file violates the convention that addresses of=0A= * type Af_INET stored in nsaddr_t variables are stored in 24/8 = format.=0A= * Many variables in nsaddr_t's in this file store ip addrs as simple = ints.=0A= */=0A= =0A= /***********************************************************************= **************=0A= * Code modified by Marcel Odena on the 18 February 2002.=0A= * The initial code had a queue for keeping packets waiting for an arp = resolution of=0A= * only one packet. I added the capability of keep 'ARPQUEUE_MAX_LENGTH' = packets.=0A= *=0A= * When a packet arrives at the ARP level and it needs an address = resolution, if the queue=0A= * is full, then it is dropped. =0A= * For every destination which the arp protocol must resolve an address = it is created=0A= * a queue of 'ARPQUEUE_MAX_LENGTH' elements. Once the arp protocol has = done the=0A= * resolution of that address it sends all the packets that were waiting = in the queue.=0A= *=0A= * The changes I introduced are marked with my name (Marcel Odena). I = modified the=0A= * functions:=0A= *=0A= * int ARPTable::arpresolve(nsaddr_t dst, Packet *p, LL *ll);=0A= * void ARPTable::arpinput(Packet *p, LL *ll);=0A= *=0A= ************************************************************************= *************/=0A= =0A= =0A= #include =0A= =0A= #include "delay.h"=0A= //#include "debug.h"=0A= #include "mac.h"=0A= #include "arp.h"=0A= #include "topography.h"=0A= #include "cmu-trace.h"=0A= #include "mobilenode.h"=0A= #include "ll.h"=0A= #include "packet.h"=0A= #include =0A= =0A= // #define DEBUG=0A= =0A= static class ARPTableClass : public TclClass {=0A= public:=0A= ARPTableClass() : TclClass("ARPTable") {}=0A= TclObject* create(int, const char*const* argv) {=0A= return (new ARPTable(argv[4], argv[5]));=0A= }=0A= } class_arptable;=0A= =0A= int hdr_arp::offset_;=0A= =0A= static class ARPHeaderClass : public PacketHeaderClass {=0A= public:=0A= ARPHeaderClass() : PacketHeaderClass("PacketHeader/ARP",=0A= sizeof(hdr_arp)) { =0A= bind_offset(&hdr_arp::offset_);=0A= }=0A= } class_arphdr;=0A= =0A= /* = =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=0A= Address Resolution (ARP) Table=0A= = =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D = */=0A= =0A= ARPTable_List ARPTable::athead_ =3D { 0 };=0A= =0A= void=0A= ARPTable::Terminate()=0A= {=0A= ARPEntry *ll;=0A= for(ll =3D arphead_.lh_first; ll; ll =3D ll->arp_link_.le_next) {=0A= if(ll->hold_) {=0A= drop(ll->hold_, DROP_END_OF_SIMULATION);=0A= ll->hold_ =3D 0;=0A= }=0A= }=0A= }=0A= =0A= =0A= ARPTable::ARPTable(const char *tclnode, const char *tclmac) : = LinkDelay() {=0A= LIST_INIT(&arphead_);=0A= =0A= node_ =3D (MobileNode*) TclObject::lookup(tclnode);=0A= assert(node_);=0A= =0A= mac_ =3D (Mac*) TclObject::lookup(tclmac);=0A= assert(mac_);=0A= LIST_INSERT_HEAD(&athead_, this, link_);=0A= }=0A= =0A= int=0A= ARPTable::command(int argc, const char*const* argv)=0A= {=0A= if (argc =3D=3D 2 && strcasecmp(argv[1], "reset") =3D=3D 0) {=0A= Terminate();=0A= //FALL-THROUGH to give parents a chance to reset=0A= }=0A= return LinkDelay::command(argc, argv);=0A= }=0A= =0A= =0A= =0A= // Modified by Marcel Odena. 18/02/2002.=0A= =0A= int=0A= ARPTable::arpresolve(nsaddr_t dst, Packet *p, LL *ll)=0A= {=0A= ARPEntry *llinfo ;=0A= =0A= assert(initialized());=0A= llinfo =3D arplookup(dst);=0A= =0A= #ifdef DEBUG=0A= fprintf(stderr, "%d - %s\n", node_->address(), = __FUNCTION__);=0A= #endif=0A= =0A= if(llinfo && llinfo->up_) {=0A= mac_->hdr_dst((char*) HDR_MAC(p), llinfo->macaddr_);=0A= return 0;=0A= }=0A= =0A= if(llinfo =3D=3D 0) {=0A= /*=0A= * Create a new ARP entry=0A= */=0A= llinfo =3D new ARPEntry(&arphead_, dst);=0A= }=0A= =0A= =0A= // Marcel Odena, 18/2/2002. I disabled this feature. =0A= /************************=0A= if(llinfo->count_ >=3D ARP_MAX_REQUEST_COUNT) {=0A= //=0A= // Because there is not necessarily a scheduled event = between=0A= // this callback and the point where the callback can = return=0A= // to this point in the code, the order of operations = is very=0A= // important here so that we don't get into an infinite = loop.=0A= // - josh=0A= //=0A= Packet *t =3D llinfo->hold_;=0A= =0A= llinfo->count_ =3D 0;=0A= llinfo->hold_ =3D 0;=0A= hdr_cmn* ch;=0A= =0A= if(t) {=0A= ch =3D HDR_CMN(t);=0A= =0A= if (ch->xmit_failure_) {=0A= ch->xmit_reason_ =3D 0;=0A= ch->xmit_failure_(t, = ch->xmit_failure_data_);=0A= }=0A= else {=0A= drop(t, DROP_IFQ_ARP_FULL);=0A= }=0A= }=0A= =0A= ch =3D HDR_CMN(p);=0A= =0A= if (ch->xmit_failure_) {=0A= ch->xmit_reason_ =3D 0;=0A= ch->xmit_failure_(p, = ch->xmit_failure_data_);=0A= }=0A= else {=0A= drop(p, DROP_IFQ_ARP_FULL);=0A= }=0A= =0A= return EADDRNOTAVAIL;=0A= }=0A= =0A= ***********************/=0A= =0A= // Marcel Odena, 18/2/2002.=0A= // If there is not enough space in the buffer we drop the = incoming packet.=0A= if(llinfo->num_packets_>=3DARPQUEUE_MAX_LENGTH){=0A= drop(p, DROP_IFQ_ARP_FULL);=0A= }=0A= else{=0A= //I copy the content of 'p' in the next place free of the = queue (i.e. last_packet),=0A= // and actulalize the pointer 'last_packet' that point to = the next free place in=0A= // the queue. Also increment the number of packets being = held in the queue.=0A= (*llinfo->last_packet_)=3D(*p); =0A= llinfo->last_packet_++;=0A= llinfo->num_packets_++;=0A= =0A= }=0A= =0A= =0A= // Monarch initial code=0A= // llinfo->count_++;=0A= // if(llinfo->hold_)=0A= // drop(llinfo->hold_, DROP_IFQ_ARP_FULL);=0A= // llinfo->hold_ =3D p;=0A= =0A= /*=0A= * We don't have a MAC address for this node. Send an ARP = Request.=0A= *=0A= * XXX: Do I need to worry about the case where I keep ARPing=0A= * for the SAME destination.=0A= */=0A= =0A= int src =3D node_->address(); // this host's IP addr=0A= arprequest(src, dst, ll);=0A= return EADDRNOTAVAIL;=0A= }=0A= =0A= =0A= ARPEntry*=0A= ARPTable::arplookup(nsaddr_t dst)=0A= {=0A= ARPEntry *a;=0A= =0A= for(a =3D arphead_.lh_first; a; a =3D a->nextarp()) {=0A= if(a->ipaddr_ =3D=3D dst)=0A= return a;=0A= }=0A= return 0;=0A= }=0A= =0A= =0A= void=0A= ARPTable::arprequest(nsaddr_t src, nsaddr_t dst, LL *ll)=0A= {=0A= Scheduler& s =3D Scheduler::instance();=0A= Packet *p =3D Packet::alloc();=0A= =0A= hdr_cmn *ch =3D HDR_CMN(p);=0A= char *mh =3D (char*) HDR_MAC(p);=0A= hdr_ll *lh =3D HDR_LL(p);=0A= hdr_arp *ah =3D HDR_ARP(p);=0A= =0A= ch->uid() =3D 0;=0A= ch->ptype() =3D PT_ARP;=0A= ch->size() =3D ARP_HDR_LEN;=0A= ch->iface() =3D -2;=0A= ch->error() =3D 0;=0A= =0A= mac_->hdr_dst(mh, MAC_BROADCAST);=0A= mac_->hdr_src(mh, ll->mac_->addr());=0A= mac_->hdr_type(mh, ETHERTYPE_ARP);=0A= =0A= lh->seqno() =3D 0;=0A= lh->lltype() =3D LL_DATA;=0A= =0A= ch->direction() =3D hdr_cmn::DOWN; // send this pkt down=0A= ah->arp_hrd =3D ARPHRD_ETHER;=0A= ah->arp_pro =3D ETHERTYPE_IP;=0A= ah->arp_hln =3D ETHER_ADDR_LEN;=0A= ah->arp_pln =3D sizeof(nsaddr_t);=0A= ah->arp_op =3D ARPOP_REQUEST;=0A= ah->arp_sha =3D ll->mac_->addr();=0A= ah->arp_spa =3D src;=0A= ah->arp_tha =3D 0; // what were're looking for=0A= ah->arp_tpa =3D dst;=0A= =0A= s.schedule(ll->downtarget_, p, delay_);=0A= }=0A= =0A= =0A= // Modified by Marcel Odena. 18/02/2002.=0A= void=0A= ARPTable::arpinput(Packet *p, LL *ll)=0A= {=0A= Scheduler& s =3D Scheduler::instance();=0A= hdr_arp *ah =3D HDR_ARP(p);=0A= ARPEntry *llinfo;=0A= Packet *i;=0A= =0A= assert(initialized());=0A= =0A= #ifdef DEBUG=0A= fprintf(stderr,=0A= "%d - %s\n\top: %x, sha: %x, tha: %x, spa: %x, tpa: = %x\n",=0A= node_->address(), __FUNCTION__, ah->arp_op,=0A= ah->arp_sha, ah->arp_tha, ah->arp_spa, ah->arp_tpa);=0A= #endif=0A= =0A= if((llinfo =3D arplookup(ah->arp_spa)) =3D=3D 0) {=0A= =0A= /*=0A= * Create a new ARP entry=0A= */=0A= llinfo =3D new ARPEntry(&arphead_, ah->arp_spa);=0A= }=0A= assert(llinfo);=0A= =0A= llinfo->macaddr_ =3D ah->arp_sha;=0A= llinfo->up_ =3D 1;=0A= =0A= /*=0A= * Can we send whatever's being held? =0A= */=0A= =0A= //Marcel Odena, 18/2/2002 =0A= //If we execute this 'if' it is because we have an ADDRESS REPLY =0A= // and we have some packets waiting for this address. So, for all=0A= // the packets waiting for this address resolution we send them = down.=0A= if(llinfo->num_packets_>0) {=0A= = for(i=3Dllinfo->first_packet_;ilast_packet_;i++){=0A= hdr_cmn *ch =3D HDR_CMN(i);=0A= char *mh =3D (char*) HDR_MAC(i);=0A= hdr_ip *ih =3D HDR_IP(i); =0A= =0A= // XXXHACK for now: =0A= // Future work: separate port-id from IP address ??=0A= int dst =3D Address::instance().get_nodeaddr(ih->daddr());=0A= =0A= if((ch->addr_type() =3D=3D NS_AF_NONE &&=0A= dst =3D=3D ah->arp_spa) ||=0A= (NS_AF_INET =3D=3D ch->addr_type() &&=0A= ch->next_hop() =3D=3D ah->arp_spa)) {=0A= #ifdef DEBUG=0A= fprintf(stderr, "\tsending HELD packet.\n");=0A= #endif=0A= mac_->hdr_dst(mh, ah->arp_sha);=0A= s.schedule(ll->downtarget_, i, delay_);=0A= llinfo->num_packets_--;=0A= =0A= // Monarch code=0A= // s.schedule(ll->downtarget_, llinfo->hold_, delay_);=0A= // llinfo->hold_ =3D 0;=0A= =0A= }=0A= else {=0A= fprintf(stderr, "\tfatal ARP error...\n");=0A= exit(1);=0A= }=0A= =0A= } /*end for*/=0A= assert(llinfo->num_packets_=3D=3D0);=0A= llinfo->last_packet_=3Dllinfo->first_packet_;=0A= }/*end if*/=0A= //End of Marcel Odena's modifications., 18/2/2002 =0A= =0A= =0A= if(ah->arp_op =3D=3D ARPOP_REQUEST &&=0A= ah->arp_tpa =3D=3D node_->address()) {=0A= =0A= hdr_cmn *ch =3D HDR_CMN(p);=0A= char *mh =3D (char*)HDR_MAC(p);=0A= hdr_ll *lh =3D HDR_LL(p);=0A= =0A= ch->size() =3D ARP_HDR_LEN;=0A= ch->error() =3D 0;=0A= ch->direction() =3D hdr_cmn::DOWN; // send this pkt down=0A= =0A= mac_->hdr_dst(mh, ah->arp_sha);=0A= mac_->hdr_src(mh, ll->mac_->addr());=0A= mac_->hdr_type(mh, ETHERTYPE_ARP);=0A= =0A= lh->seqno() =3D 0;=0A= lh->lltype() =3D LL_DATA;=0A= =0A= // ah->arp_hrd =3D =0A= // ah->arp_pro =3D=0A= // ah->arp_hln =3D=0A= // ah->arp_pln =3D=0A= =0A= ah->arp_op =3D ARPOP_REPLY;=0A= ah->arp_tha =3D ah->arp_sha;=0A= ah->arp_sha =3D ll->mac_->addr();=0A= =0A= nsaddr_t t =3D ah->arp_spa;=0A= ah->arp_spa =3D ah->arp_tpa;=0A= ah->arp_tpa =3D t;=0A= =0A= s.schedule(ll->downtarget_, p, delay_);=0A= return;=0A= }=0A= Packet::free(p);=0A= }=0A= =0A= ------_=_NextPart_000_01C1C357.48128970 Content-Type: application/octet-stream; name="arp_MarcelOdena.h" Content-Transfer-Encoding: quoted-printable Content-Disposition: attachment; filename="arp_MarcelOdena.h" /*-*- Mode:C++; c-basic-offset:8; tab-width:8; indent-tabs-mode:t -*- = */=0A= /*=0A= * Copyright (c) 1997 Regents of the University of California.=0A= * All rights reserved.=0A= *=0A= * Redistribution and use in source and binary forms, with or = without=0A= * modification, are permitted provided that the following = conditions=0A= * are met:=0A= * 1. Redistributions of source code must retain the above copyright=0A= * notice, this list of conditions and the following disclaimer.=0A= * 2. Redistributions in binary form must reproduce the above = copyright=0A= * notice, this list of conditions and the following disclaimer in = the=0A= * documentation and/or other materials provided with the = distribution.=0A= * 3. All advertising materials mentioning features or use of this = software=0A= * must display the following acknowledgement:=0A= * This product includes software developed by the Computer Systems=0A= * Engineering Group at Lawrence Berkeley Laboratory.=0A= * 4. Neither the name of the University nor of the Laboratory may be = used=0A= * to endorse or promote products derived from this software = without=0A= * specific prior written permission.=0A= *=0A= * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' = AND=0A= * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, = THE=0A= * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR = PURPOSE=0A= * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE = LIABLE=0A= * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR = CONSEQUENTIAL=0A= * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE = GOODS=0A= * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS = INTERRUPTION)=0A= * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, = STRICT=0A= * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN = ANY WAY=0A= * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY = OF=0A= * SUCH DAMAGE.=0A= */=0A= /* Ported from CMU/Monarch's code, nov'98 -Padma.*/=0A= =0A= =0A= /***********************************************************************= **************=0A= * Code modified by Marcel Odena on the 18 February 2002.=0A= * The initial code had a queue for keeping packets waiting for an arp = resolution of=0A= * only one packet. I added the capability of keep 'ARPQUEUE_MAX_LENGTH' = packets.=0A= *=0A= * When a packet arrives at the ARP level and it needs an address = resolution, if the queue=0A= * is full, then it is dropped. =0A= * For every destination which the arp protocol must resolve an address = it is created=0A= * a queue of 'ARPQUEUE_MAX_LENGTH' elements. Once the arp protocol has = done the=0A= * resolution of that address it sends all the packets that were waiting = in the queue.=0A= *=0A= * The changes I introduced are marked with my name (Marcel Odena).=0A= ************************************************************************= *************/=0A= =0A= =0A= =0A= =0A= #ifndef __arp_h__=0A= #define __arp_h__=0A= =0A= #include "scheduler.h"=0A= #include "delay.h"=0A= #include "lib/bsd-list.h"=0A= =0A= #ifndef EADDRNOTAVAIL=0A= #define EADDRNOTAVAIL 125=0A= #endif /* !EADDRNOTAVAIL */=0A= =0A= class LinkDelay;=0A= class ARPEntry;=0A= class ARPTable;=0A= class LL;=0A= class Mac;=0A= =0A= LIST_HEAD(ARPTable_List, ARPTable);=0A= LIST_HEAD(ARPEntry_List, ARPEntry);=0A= =0A= =0A= /* = =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=0A= Address Resolution (ARP) Header=0A= = =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D = */=0A= #define ARPHRD_ETHER 1 /* ethernet hardware format */=0A= =0A= #define ARPOP_REQUEST 1 /* request to resolve address */=0A= #define ARPOP_REPLY 2 /* response to previous request */=0A= #define ARPOP_REVREQUEST 3 /* request protocol address */=0A= #define ARPOP_REVREPLY 4 /* response giving protocol address */=0A= #define ARPOP_INVREQUEST 8 /* request to identify peer */=0A= #define ARPOP_INVREPLY 9 /* response identifying peer */=0A= =0A= #define ARP_HDR_LEN 28=0A= #define ARPQUEUE_MAX_LENGTH 1000 //Marcel Odena, 18/02/2002=0A= =0A= struct hdr_arp {=0A= u_int16_t arp_hrd;=0A= u_int16_t arp_pro;=0A= u_int8_t arp_hln;=0A= u_int8_t arp_pln;=0A= u_int16_t arp_op;=0A= int arp_sha;=0A= u_int16_t pad1; // so offsets are correct=0A= nsaddr_t arp_spa;=0A= int arp_tha;=0A= u_int16_t pad2; // so offsets are correct=0A= nsaddr_t arp_tpa;=0A= =0A= // Header access methods=0A= static int offset_; // required by PacketHeaderManager=0A= inline static int& offset() { return offset_; }=0A= inline static hdr_arp* access(const Packet* p) {=0A= return (hdr_arp*) p->access(offset_);=0A= }=0A= };=0A= =0A= class ARPEntry {=0A= friend class ARPTable;=0A= public:=0A= ARPEntry(ARPEntry_List* head, nsaddr_t dst) {=0A= up_ =3D macaddr_ =3D count_ =3D 0;=0A= ipaddr_ =3D dst;=0A= hold_ =3D 0;=0A= //Marcel Odena, 18/02/2002=0A= num_packets_ =3D 0;=0A= first_packet_ =3D last_packet_ =3D new Packet[ARPQUEUE_MAX_LENGTH]; = =0A= LIST_INSERT_HEAD(head, this, arp_link_);=0A= }=0A= inline ARPEntry* nextarp() { return arp_link_.le_next; }=0A= =0A= private:=0A= LIST_ENTRY(ARPEntry) arp_link_;=0A= =0A= int up_;=0A= nsaddr_t ipaddr_;=0A= int macaddr_;=0A= Packet *hold_;=0A= Packet *first_packet_; //Marcel Odena, 18/02/2002=0A= Packet *last_packet_; //Marcel Odena, 18/02/2002=0A= int num_packets_; //Marcel Odena, 18/02/2002=0A= int count_;=0A= #define ARP_MAX_REQUEST_COUNT 3=0A= };=0A= =0A= =0A= /**************************************************************=0A= * Marcel Odena, 18/02/2002.=0A= *=0A= * SOME NOTES:=0A= *=0A= * NOTE 1:=0A= * The idea is: I define an array of packets (the queue):=0A= * new Packet[ARPQUEUE_MAX_LENGTH];=0A= * =0A= * With 'first_packet_' we point to the first element of the queue,=0A= * and with 'last_packet_' we point to the next free place of the = queue.=0A= * 'num_packets' define the number of packets that we are holding in = =0A= * the queue waiting for an address resolution.=0A= *=0A= *=0A= * NOTE 2:=0A= * I do not use anymore the variables 'hold_' and 'count_', but = anyway,=0A= * I keep them in the program.=0A= ************************************************************************= ****/=0A= =0A= class ARPTable : public LinkDelay {=0A= public:=0A= ARPTable(const char *tclnode, const char *tclmac);=0A= =0A= int command(int argc, const char*const* argv);=0A= int arpresolve(nsaddr_t dst, Packet *p, LL *ll);=0A= void arpinput(Packet *p, LL *ll);=0A= ARPEntry* arplookup(nsaddr_t dst);=0A= void arprequest(nsaddr_t src, nsaddr_t dst, LL *ll);=0A= =0A= void Terminate(void);=0A= =0A= private:=0A= inline int initialized() { return node_ && mac_; }=0A= =0A= ARPEntry_List arphead_;=0A= MobileNode *node_;=0A= Mac *mac_;=0A= =0A= /*=0A= * Used to purge all of the ll->hold packets at the end of the=0A= * simulation.=0A= */=0A= public:=0A= LIST_ENTRY(ARPTable) link_;=0A= static ARPTable_List athead_;=0A= };=0A= =0A= #endif /* __arp_h__ */=0A= =0A= =0A= =0A= =0A= =0A= =0A= =0A= ------_=_NextPart_000_01C1C357.48128970-- From najma_ismat@hotmail.com Mon Mar 4 05:15:50 2002 From: najma_ismat@hotmail.com (Najma Ismat) Date: Mon Mar 4 05:15:50 2002 Subject: [ns] Getting error in tcp-reno.cc. Urgent help is needed!!!!!!!!!!!!! Message-ID: hi, i have made changes in tcp-reno.cc scripts but still getting the same error. don't know how to deal with it. i still needs help in this regard. thanks in advance. Najma Ismat. *****************************error********************************* tcp-reno.cc:229: no `int RenoTcpAgent::allow_fast_retransmit (int)' member function declared in class `RenoTcpAgent' tcp-reno.cc:253: no `void RenoTcpAgent::dupack_action (int)' member function declared in class `RenoTcpAgent' tcp-reno.cc: In method `void RenoTcpAgent::dupack_action (int)': tcp-reno.cc:255: `allow_fast_retransmit' undeclared (first use this function) tcp-reno.cc:255: (Each undeclared identifier is reported only once for each function it appears in.) make: *** [tcp-reno.o] Error 1 Ns make failed! ****************************tcp-reno.cc code************************* int RenoTcpAgent::allow_fast_retransmit(int last_cwnd_action_) { return (last_cwnd_action_ == CWND_ACTION_DUPACK); } void RenoTcpAgent::dupack_action(int) { int recovered = (highest_ack_ > recover_); int allowFastRetransmit=allow_fast_retransmit(last_cwnd_action_); if (recovered || (!bug_fix_ && !ecn_) || allowFastRetransmit) { goto reno_action; } if (ecn_ && last_cwnd_action_ == CWND_ACTION_ECN) { last_cwnd_action_ = CWND_ACTION_DUPACK; reset_rtx_timer(1,0); output(last_ack_ + 1, TCP_REASON_DUPACK); return; } if (bug_fix_) { return; } reno_action: // we are now going to fast-retransmit and will trace that event trace_event("RENO_FAST_RETX"); recover_ = maxseq_; last_cwnd_action_ = CWND_ACTION_DUPACK; slowdown(CLOSE_SSTHRESH_HALF|CLOSE_CWND_HALF); reset_rtx_timer(1,0); //output(last_ack_ + 1, TCP_REASON_DUPACK); // from top return; } >From: "Gautam Chakrabarti" >To: najma_ismat@hotmail.com (Najma Ismat) >CC: chakra10@msu.edu, ns-users@isi.edu >Subject: Re: [ns] Getting error in tcp-reno.cc. Urgent help is >needed!!!!!!!!!!!!! >Date: Sun, 3 Mar 2002 17:43:19 -0500 (EST) > >Hi, > >Assuming you have copied the proper portions of code here, there is a >problem >with the the function RenoTcpAgent::allow_fast_retransmit, it returns in >int, >but ur definition does not have any return type, due to which it takes that >to >be void, so first it says that to be NOTa valid member function, then u get >the >error in dupack_action() while taking the return value in int. > >Just add an int as return type to RenoTcpAgent::allow_fast_retransmit, >should >work. > >hth, >gautam > > > > > **********************tcp-reno.cc code******************************* > > RenoTcpAgent::allow_fast_retransmit(int last_cwnd_action_) > > { > > return (last_cwnd_action_ == CWND_ACTION_DUPACK); > > } > > void RenoTcpAgent::dupack_action() [getting error here] > > { > > int recovered = (highest_ack_ > recover_); > > int >allowFastRetransmit=allow_fast_retransmit(last_cwnd_action_);[getting > > error here] > > > > if (recovered || (!bug_fix_ && !ecn_) || allowFastRetransmit) { > > goto reno_action; > > } > > if (ecn_ && last_cwnd_action_ == CWND_ACTION_ECN) { > > last_cwnd_action_ = CWND_ACTION_DUPACK; > > reset_rtx_timer(1,0); > > output(last_ack_ + 1, TCP_REASON_DUPACK); > > return; > > } > > > > if (bug_fix_) { > > return; > > } > > > > > > > > > > >From: Gautam - Chakrabarti > > >To: Najma Ismat , NS Users > > >Subject: Re: [ns] Getting error in tcp-reno.cc. Urgent help is > > >needed!!!!!!!!!!!!! > > >Date: Sat, 02 Mar 2002 18:23:36 -0500 > > > > > > > > >there is a parse error in ur tcp.h file, check it up, it seems it has a > > >{} mismatch there. That may lead to some other errors....also check if > > >you have used the flag ELFN_ENABLE everywhere properly, that is, if u > > >have commented out some class methods/member variables by mistake using > > >the flag..., since you have the errors in the ELFN_ENABLE not-defined > > >portion.... also probably u don't need to specify the flag-name after > > >#else --- exact behavior depends on the compiler...and the portion of > > >code that u have sent has an extra '}' at the end, check if it is a > > >problem ---- oh yeah, the last portion of the code u have sent is from > > >tcp.h --- so that is at least 1 of the problems -- an extra '}' > > > > > > > > >hth, > > >gautam > > > > > >Najma Ismat wrote: > > > > > > > > Hi, > > > > I am simulating ad hoc network using dsr with elfn. i have modified >the > > >ns2 > > > > code using the elfn code by g.holland. the problem i am facing >during > > > > ns-2.1b8a build up in tcp-reno.cc. i have consulted the ns2 manual >but i > > >am > > > > not able to solve it.if anyone can guide then it will be really >helpful > > >me. > > > > i am attaching the modified code of tcp-reno.cc & tcp.h along with >the > > > > errors I am getting during installation. for tcp.h i checked the >code > > >and > > > > it is the same as given in the manual. > > > > i am thankful in advance. > > > > Najma Ismat. > > > > > > > > >######################Errors########################################## > > > > In file included from tcp-reno.cc:30: > > > > tcp.h:138: parse error at null character > > > > tcp.h:504: warning: This file contains more `}'s than `{'s. > > > > tcp-reno.cc:122: ISO C++ forbids declaration of `RenoTcpAgent' with >no > > > > type > > > > tcp-reno.cc:122: no `int TcpAgent::RenoTcpAgent ()' member function > > > > declared in class `TcpAgent' > > > > tcp-reno.cc: In method `int TcpAgent::RenoTcpAgent ()': > > > > tcp-reno.cc:122: only constructors take base initializers > > > > tcp-reno.cc:122: class `TcpAgent' does not have any field named > > > > `dupwnd_' > > > > tcp-reno.cc:124: warning: no return statement in function returning > > > > non-void > > > > make: *** [tcp-reno.o] Error 1 > > > > Ns make failed! > > > > >###########################tcp-reno.cc(modified)###################### > > > > > > > > int RenoTcpAgent::window() > > > > { > > > > // > > > > // reno: inflate the window by dupwnd_ > > > > // dupwnd_ will be non-zero during fast recovery, > > > > // at which time it contains the number of dup acks > > > > // > > > > int win = int(cwnd_) + dupwnd_; > > > > if (win > int(wnd_)) > > > > win = int(wnd_); > > > > return (win); > > > > } > > > > > > > > double RenoTcpAgent::windowd() > > > > { > > > > // > > > > // reno: inflate the window by dupwnd_ > > > > // dupwnd_ will be non-zero during fast recovery, > > > > // at which time it contains the number of dup acks > > > > // > > > > double win = cwnd_ + dupwnd_; > > > > if (win > wnd_) > > > > win = wnd_; > > > > return (win); > > > > } > > > > #ifdef ELFN_ENABLE > > > > > > > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0), > > >thaw_timer_(this), > > > > t_thaw_(0) > > > > { > > > > // Defaults for bound variables should be set in >ns-default.tcl. > > > > bind("t_thaw_", &t_thaw_); > > > > } > > > > #else ELFN_ENABLE > > > > > > > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0){ [error in >this > > >line] > > > > } > > > > #endif ELFN_ENABLE > > > > >##########################tcp.h####################################### > > > > class RtxTimer : public TimerHandler { > > > > > > > > public: > > > > > > > > RtxTimer(TcpAgent *a) : TimerHandler() { a_ = a; } [error in > > >thsi line] > > > > > > > > protected: > > > > > > > > virtual void expire(Event *e); > > > > TcpAgent *a_; > > > > > > > > } > > > > > > > > }; > > > > > > > > _________________________________________________________________ > > > > Get your FREE download of MSN Explorer at > > >http://explorer.msn.com/intl.asp. > > > > > >-- > > >*************************************************************** > > >Gautam Chakrabarti > > >MS Student > > >Computer Sc. and Engg Department > > >Michigan State University > > > > > >Email : chakra10@msu.edu > > >Home Page : http://www.cse.msu.edu/~chakra10 > > > > > >Why is there so much month left at the end of the money? > > > --- John Barrymore > > >*************************************************************** > > > > > > > > > > > > > _________________________________________________________________ > > MSN Photos is the easiest way to share and print your photos: > > http://photos.msn.com/support/worldwide.aspx > > > > > > >-- >------------------------------------------- >Gautam Chakrabarti >MS Student >Computer Sc. & Engineering Dept. >Michigan State University > >Email: chakra10@msu.edu >Home Page: http://www.cse.msu.edu/~chakra10 >-------------------------------------------- _________________________________________________________________ MSN Photos is the easiest way to share and print your photos: http://photos.msn.com/support/worldwide.aspx From sherubic@21cn.com Mon Mar 4 05:16:02 2002 From: sherubic@21cn.com (ÓÀÉú) Date: Mon Mar 4 05:16:02 2002 Subject: [ns] pls help! Message-ID: Hi,bother to you again.I'm sorry. I'm really exhausted about installing network simulator ,I almost spend two weeks on the event but seems no progressing at all.I really don't understand the reason resulted in this outcome. I really hope and need you can help me out and I seems count on nobody around me. now,I first describe my computer which seems differ from yours.I install win98 in drive c.and install win2000 in drive d.what i can say is just such. when i install otcl-1.0a7,system tell me could not find ,then I change it into,and then tell me it couldn't find files such as ,and so on.and then I replace all these files with their directory like . I don't know how to sovle these problems which seems simple for you,I think. thanks ---------------------------------------------- ¹úÄÚÊÕ·ÑÓÊÏäË­×îÇ¿£¬Ê×´ÎȨÍþÆÀ²â¼û·ÖÏþ http://news.21cn.com/domestic/21CN/2001-11-19/550568.html ÂòÆóÒµÓÊÏ䣬»ñÔùÓòÃû http://mail.21cn.com/corporation/010.html 21cn´øÄú×ß½øÄ¿±êÖ®Âã¬ÏíÊÜÉÌÂÃÀÖȤ http://travel.21cn.com/agency/aimtrip/aimtrip.html ÊÕ·ÑÓÊÏäÕæÄÜ´óÐÐÆäµÀ£¬¸øÎÒÒ»¸öÀíÓÉ http://mail.21cn.com/bijiao/index2-7.html From najma_ismat@hotmail.com Mon Mar 4 05:16:17 2002 From: najma_ismat@hotmail.com (Najma Ismat) Date: Mon Mar 4 05:16:17 2002 Subject: [ns] Getting error in tcp-reno.cc. Urgent help is needed!!!!!!!!!!!!! Message-ID: hi, i have made changes in tcp-reno.cc scripts but still getting the same error. don't know how to deal with it. i still needs help in this regard. thanks in advance. Najma Ismat. *****************************error********************************* tcp-reno.cc:229: no `int RenoTcpAgent::allow_fast_retransmit (int)' member function declared in class `RenoTcpAgent' tcp-reno.cc:253: no `void RenoTcpAgent::dupack_action (int)' member function declared in class `RenoTcpAgent' tcp-reno.cc: In method `void RenoTcpAgent::dupack_action (int)': tcp-reno.cc:255: `allow_fast_retransmit' undeclared (first use this function) tcp-reno.cc:255: (Each undeclared identifier is reported only once for each function it appears in.) make: *** [tcp-reno.o] Error 1 Ns make failed! ****************************tcp-reno.cc code************************* int RenoTcpAgent::allow_fast_retransmit(int last_cwnd_action_) { return (last_cwnd_action_ == CWND_ACTION_DUPACK); } void RenoTcpAgent::dupack_action(int) { int recovered = (highest_ack_ > recover_); int allowFastRetransmit=allow_fast_retransmit(last_cwnd_action_); if (recovered || (!bug_fix_ && !ecn_) || allowFastRetransmit) { goto reno_action; } if (ecn_ && last_cwnd_action_ == CWND_ACTION_ECN) { last_cwnd_action_ = CWND_ACTION_DUPACK; reset_rtx_timer(1,0); output(last_ack_ + 1, TCP_REASON_DUPACK); return; } if (bug_fix_) { return; } reno_action: // we are now going to fast-retransmit and will trace that event trace_event("RENO_FAST_RETX"); recover_ = maxseq_; last_cwnd_action_ = CWND_ACTION_DUPACK; slowdown(CLOSE_SSTHRESH_HALF|CLOSE_CWND_HALF); reset_rtx_timer(1,0); //output(last_ack_ + 1, TCP_REASON_DUPACK); // from top return; } >From: "Gautam Chakrabarti" >To: najma_ismat@hotmail.com (Najma Ismat) >CC: chakra10@msu.edu, ns-users@isi.edu >Subject: Re: [ns] Getting error in tcp-reno.cc. Urgent help is >needed!!!!!!!!!!!!! >Date: Sun, 3 Mar 2002 17:43:19 -0500 (EST) > >Hi, > >Assuming you have copied the proper portions of code here, there is a >problem >with the the function RenoTcpAgent::allow_fast_retransmit, it returns in >int, >but ur definition does not have any return type, due to which it takes that >to >be void, so first it says that to be NOTa valid member function, then u get >the >error in dupack_action() while taking the return value in int. > >Just add an int as return type to RenoTcpAgent::allow_fast_retransmit, >should >work. > >hth, >gautam > > > > > **********************tcp-reno.cc code******************************* > > RenoTcpAgent::allow_fast_retransmit(int last_cwnd_action_) > > { > > return (last_cwnd_action_ == CWND_ACTION_DUPACK); > > } > > void RenoTcpAgent::dupack_action() [getting error here] > > { > > int recovered = (highest_ack_ > recover_); > > int >allowFastRetransmit=allow_fast_retransmit(last_cwnd_action_);[getting > > error here] > > > > if (recovered || (!bug_fix_ && !ecn_) || allowFastRetransmit) { > > goto reno_action; > > } > > if (ecn_ && last_cwnd_action_ == CWND_ACTION_ECN) { > > last_cwnd_action_ = CWND_ACTION_DUPACK; > > reset_rtx_timer(1,0); > > output(last_ack_ + 1, TCP_REASON_DUPACK); > > return; > > } > > > > if (bug_fix_) { > > return; > > } > > > > > > > > > > >From: Gautam - Chakrabarti > > >To: Najma Ismat , NS Users > > >Subject: Re: [ns] Getting error in tcp-reno.cc. Urgent help is > > >needed!!!!!!!!!!!!! > > >Date: Sat, 02 Mar 2002 18:23:36 -0500 > > > > > > > > >there is a parse error in ur tcp.h file, check it up, it seems it has a > > >{} mismatch there. That may lead to some other errors....also check if > > >you have used the flag ELFN_ENABLE everywhere properly, that is, if u > > >have commented out some class methods/member variables by mistake using > > >the flag..., since you have the errors in the ELFN_ENABLE not-defined > > >portion.... also probably u don't need to specify the flag-name after > > >#else --- exact behavior depends on the compiler...and the portion of > > >code that u have sent has an extra '}' at the end, check if it is a > > >problem ---- oh yeah, the last portion of the code u have sent is from > > >tcp.h --- so that is at least 1 of the problems -- an extra '}' > > > > > > > > >hth, > > >gautam > > > > > >Najma Ismat wrote: > > > > > > > > Hi, > > > > I am simulating ad hoc network using dsr with elfn. i have modified >the > > >ns2 > > > > code using the elfn code by g.holland. the problem i am facing >during > > > > ns-2.1b8a build up in tcp-reno.cc. i have consulted the ns2 manual >but i > > >am > > > > not able to solve it.if anyone can guide then it will be really >helpful > > >me. > > > > i am attaching the modified code of tcp-reno.cc & tcp.h along with >the > > > > errors I am getting during installation. for tcp.h i checked the >code > > >and > > > > it is the same as given in the manual. > > > > i am thankful in advance. > > > > Najma Ismat. > > > > > > > > >######################Errors########################################## > > > > In file included from tcp-reno.cc:30: > > > > tcp.h:138: parse error at null character > > > > tcp.h:504: warning: This file contains more `}'s than `{'s. > > > > tcp-reno.cc:122: ISO C++ forbids declaration of `RenoTcpAgent' with >no > > > > type > > > > tcp-reno.cc:122: no `int TcpAgent::RenoTcpAgent ()' member function > > > > declared in class `TcpAgent' > > > > tcp-reno.cc: In method `int TcpAgent::RenoTcpAgent ()': > > > > tcp-reno.cc:122: only constructors take base initializers > > > > tcp-reno.cc:122: class `TcpAgent' does not have any field named > > > > `dupwnd_' > > > > tcp-reno.cc:124: warning: no return statement in function returning > > > > non-void > > > > make: *** [tcp-reno.o] Error 1 > > > > Ns make failed! > > > > >###########################tcp-reno.cc(modified)###################### > > > > > > > > int RenoTcpAgent::window() > > > > { > > > > // > > > > // reno: inflate the window by dupwnd_ > > > > // dupwnd_ will be non-zero during fast recovery, > > > > // at which time it contains the number of dup acks > > > > // > > > > int win = int(cwnd_) + dupwnd_; > > > > if (win > int(wnd_)) > > > > win = int(wnd_); > > > > return (win); > > > > } > > > > > > > > double RenoTcpAgent::windowd() > > > > { > > > > // > > > > // reno: inflate the window by dupwnd_ > > > > // dupwnd_ will be non-zero during fast recovery, > > > > // at which time it contains the number of dup acks > > > > // > > > > double win = cwnd_ + dupwnd_; > > > > if (win > wnd_) > > > > win = wnd_; > > > > return (win); > > > > } > > > > #ifdef ELFN_ENABLE > > > > > > > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0), > > >thaw_timer_(this), > > > > t_thaw_(0) > > > > { > > > > // Defaults for bound variables should be set in >ns-default.tcl. > > > > bind("t_thaw_", &t_thaw_); > > > > } > > > > #else ELFN_ENABLE > > > > > > > > RenoTcpAgent::RenoTcpAgent() : TcpAgent(), dupwnd_(0){ [error in >this > > >line] > > > > } > > > > #endif ELFN_ENABLE > > > > >##########################tcp.h####################################### > > > > class RtxTimer : public TimerHandler { > > > > > > > > public: > > > > > > > > RtxTimer(TcpAgent *a) : TimerHandler() { a_ = a; } [error in > > >thsi line] > > > > > > > > protected: > > > > > > > > virtual void expire(Event *e); > > > > TcpAgent *a_; > > > > > > > > } > > > > > > > > }; > > > > > > > > _________________________________________________________________ > > > > Get your FREE download of MSN Explorer at > > >http://explorer.msn.com/intl.asp. > > > > > >-- > > >*************************************************************** > > >Gautam Chakrabarti > > >MS Student > > >Computer Sc. and Engg Department > > >Michigan State University > > > > > >Email : chakra10@msu.edu > > >Home Page : http://www.cse.msu.edu/~chakra10 > > > > > >Why is there so much month left at the end of the money? > > > --- John Barrymore > > >*************************************************************** > > > > > > > > > > > > > _________________________________________________________________ > > MSN Photos is the easiest way to share and print your photos: > > http://photos.msn.com/support/worldwide.aspx > > > > > > >-- >------------------------------------------- >Gautam Chakrabarti >MS Student >Computer Sc. & Engineering Dept. >Michigan State University > >Email: chakra10@msu.edu >Home Page: http://www.cse.msu.edu/~chakra10 >-------------------------------------------- _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp. From netsim@gmx.at Mon Mar 4 05:16:29 2002 From: netsim@gmx.at (netsim@gmx.at) Date: Mon Mar 4 05:16:29 2002 Subject: [ns] NSE: Scheduler going backwards in time Message-ID: <16071.1015241650@www61.gmx.net> Hi! I am using the emulation mode in ns-2 called nse and ecountered the following problem: scheduler going backwards in time What is the reason for this error and how severe is it? Are all my trace files useless when getting this message? Thanks a lot. Martin Joerg -- GMX - Die Kommunikationsplattform im Internet. http://www.gmx.net From malvarez@dit.upm.es Mon Mar 4 05:16:40 2002 From: malvarez@dit.upm.es (Miguel Angel Alvarez) Date: Mon Mar 4 05:16:40 2002 Subject: [ns] Help with TCP time out Message-ID: <008201c1c371$b881a730$e101048a@dit.upm.es> Hi, I am using Full Tcp Agents with TcpApp Agents. Once all data has been transmitted, i close the tcp connection with the command "close". I have checked that all the data has been transmitted and acked, afterwards the connection is correctly closed, but then this warning appears: (_o292) unexpected timeout 3 in state 0 I know that connection has been closed because state 0 is TCPS_CLOSED, and the warning appears after TCP closing mechanism has been completed. Does anybody know what really happens? what is the matter? Thank you in advance. M.A. Alvarez From netsim@gmx.at Mon Mar 4 05:16:51 2002 From: netsim@gmx.at (netsim@gmx.at) Date: Mon Mar 4 05:16:51 2002 Subject: [ns] Priority in NSE Message-ID: <24328.1015243960@www10.gmx.net> Hi! I am trying to read out the priority/TOS field of IP packets from a live network in NSE with the prio_ variable of an Tap Agent. But it seems that this isn't the right way because I always get the value 0. Has anyone of you an idea how to read out the TOS/Prio field from packets coming from the live network? Thanks, Martin Joerg -- GMX - Die Kommunikationsplattform im Internet. http://www.gmx.net From laoyusg@yahoo.com.sg Mon Mar 4 05:17:02 2002 From: laoyusg@yahoo.com.sg (=?iso-8859-1?q?Lao=20Yu?=) Date: Mon Mar 4 05:17:02 2002 Subject: [ns] Anyone working on emulation over the GPRS network model by Richa Jain? Message-ID: <20020304130652.55899.qmail@web21209.mail.yahoo.com> Can you successfully send the live packets through the simulated GPRS network? Best Regards, Lao Yu __________________________________________________ Do You Yahoo!? Yahoo! Mobile - Jazz up your mobile phone! Get funky ringtones and logos! http://mobile.yahoo.com.sg/ From laoyusg@yahoo.com.sg Mon Mar 4 05:45:11 2002 From: laoyusg@yahoo.com.sg (=?iso-8859-1?q?Lao=20Yu?=) Date: Mon Mar 4 05:45:11 2002 Subject: [ns] Question about the GPRS patch by Richa Jain! Message-ID: <20020304133954.75238.qmail@web21207.mail.yahoo.com> Hi, In the documentation of the GPRS patch, it says that: "The frequency channel a packet is to be transmitted on is stamped onto a new field called 'chan_' in the common header of the packet"; while a physical channel is a slot on a frequency. So for the variable "chan_", it denotes the slot number or the frequency number of the channel? Can anyone help me to answer this question? Thanks in advance! Best Regards, Lao Yu __________________________________________________ Do You Yahoo!? Yahoo! Mobile - Jazz up your mobile phone! Get funky ringtones and logos! http://mobile.yahoo.com.sg/ From ronny.tittoto@exgate.tek.com Mon Mar 4 07:20:02 2002 From: ronny.tittoto@exgate.tek.com (ronny.tittoto@exgate.tek.com) Date: Mon Mar 4 07:20:02 2002 Subject: [ns] new ns-2.1b9 release, when? Message-ID: <355AEBE0872AD411B96A0008C7F377398F0716@eu-pado-m01.pado.tek.com> Dear all, I read in http://www.isi.edu/nsnam/ns/CHANGES.html that the ns-2.1b9 release PENDING will support variable delay links: [floyd] Sat Jun 9 13:44:00 PDT 2001 Added options to ErrorModel for delaying packets. The validation test and example is ``./test-all-links delayPacket'' in ``tcl/test''. does anyone can tell me when this new release will be avaiable? Thank you in advance and best regards Ronny From mkenny@mekb2.sps.mot.com Mon Mar 4 08:05:01 2002 From: mkenny@mekb2.sps.mot.com (Martin McKenny) Date: Mon Mar 4 08:05:01 2002 Subject: [ns] Help: Unable to bind type BOOL? Message-ID: <3C839A97.870A591E@mekb2.sps.mot.com> --------------A2FECC1C6502A241D616303D Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: base64 Hi All, Has anyone else had trouble implementing type bool in NS (Particularly where binding between OTcl and C++ variables is concerned). I have been trying to implement this using bind_bool but it doesn't work. I notice in some examples that char is used where bool might be more appropriate, is there a problem I don't know about??? Thanks, Martin --------------A2FECC1C6502A241D616303D Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit Hi All,

Has anyone else had trouble implementing type bool in NS (Particularly
where binding between OTcl and C++ variables is concerned). I have been
trying to implement this using bind_bool but it doesn't work. I notice
in some examples that char is used where bool might be more appropriate,
is there a problem I don't know about???

Thanks,
Martin 


 

--------------A2FECC1C6502A241D616303D--


From Nidhi.Bansal@prairiecomm.com  Mon Mar  4 11:15:03 2002
From: Nidhi.Bansal@prairiecomm.com (Nidhi Bansal)
Date: Mon Mar  4 11:15:03 2002
Subject: [ns] RE: Unable to recompile ns using make
Message-ID: 

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C1C3B0.9C1FE180
Content-Type: text/plain;
	charset="iso-8859-1"



>  -----Original Message-----
> From: 	Nidhi Bansal  
> Sent:	Monday, March 04, 2002 1:09 PM
> To:	'ns-users@isi.edu'
> Subject:	Unable to recompile ns using make
> 
> I am not able to recompile ns by typing 'make'.
> I am using sparc-sun-solaris2.7
> I have been able to run the example scripts just fine but when I changed a
> C++ file and tried to recompile, I ge the error :
> [nidhik@ontario][ns-allinone-2.1b8a][94] /usr/ccs/bin/make depend
> make: *** No rule to make target `depend'.  Stop.

------_=_NextPart_001_01C1C3B0.9C1FE180
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable






RE: Unable to recompile ns using make







    

     -----Original Message-----

    From:   Nidhi Bansal  

    Sent:   =
Monday, March 04, 2002 1:09 PM

    To:     'ns-users@isi.edu'

    Subject:        Unable to recompile ns using =
make

    


    I am not able to recompile ns by =
typing 'make'.

    I am using =
sparc-sun-solaris2.7

    I have been able to run the example =
scripts just fine but when I changed a C++ file and tried to recompile, =
I ge the error :
    


    [nidhik@ontario][ns-allinone-2.1b8a][94] =
/usr/ccs/bin/make depend

    make: *** No rule to make target =
`depend'.  Stop.

    




------_=_NextPart_001_01C1C3B0.9C1FE180--


From Rongmei Zhang   Mon Mar  4 11:20:02 2002
From: Rongmei Zhang  (Rongmei Zhang)
Date: Mon Mar  4 11:20:02 2002
Subject: [ns] How to get the link layer handle?
Message-ID: <200203041917.g24JHRH25244@schubert.ecn.purdue.edu>

Hi:

DO we have to add a link layer to a node before using it? How to get a reference 
to the link layer of a regular ns node?
Thanks!

rongmei


rongmei@ecn.purdue.edu
(lab) 765-494-3474



From najma_ismat@hotmail.com  Mon Mar  4 11:30:03 2002
From: najma_ismat@hotmail.com (Najma Ismat)
Date: Mon Mar  4 11:30:03 2002
Subject: [ns] Trace file format
Message-ID: 


hi,
i have modified my ns-2.1b8a code with elfn. i have run some scripts but now 
the trace file format has been changed from what i was getting earlier. can 
anyone guide me how to interpret it now. thanks in advance.
Najma Ismat

r 18.756601786 7 (790.96  62.20) RTR  --- 262 tcp 1500 [a2 7 18 800] ------- 
[11:0 12:0 32 7] [85 0] 3 0
f 18.756601786 7 (790.96  62.20) RTR  --- 262 tcp 1500 [a2 7 18 800] ------- 
[11:0 12:0 32 12] [85 0] 3 0
r 18.763832169 12 (954.01 235.99) RTR  --- 261 tcp 1500 [a2 c 7 800] ------- 
[11:0 12:0 32 12] [84 0] 4 0
r 18.763832169 12 (954.01 235.99) RTR  --- 263 ack 40 [0 0 0 0] ------- 
[12:0 11:0 32 0] [84 0] 0 0
r 18.763832169 12 (954.01 235.99) RTR  --- 264 ack 40 [0 0 0 0] ------- 
[12:0 11:0 32 0] [84 0] 0 0
s 18.763832169 12 (954.01 235.99) RTR  --- 263 ack 80 [0 0 0 0] ------- 
[12:0 11:0 32 7] [84 0] 0 0
s 18.763832169 12 (954.01 235.99) RTR  --- 264 ack 80 [0 0 0 0] ------- 
[12:0 11:0 32 7] [84 0] 0 0
r 18.771203346 12 (954.01 235.99) RTR  --- 262 tcp 1500 [a2 c 7 800] ------- 
[11:0 12:0 32 12] [85 0] 4 0

_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com


From rjaising@torrentnet.com  Mon Mar  4 13:55:01 2002
From: rjaising@torrentnet.com (Rahul Jaising)
Date: Mon Mar  4 13:55:01 2002
Subject: [ns] MMPP and FBM models for ns
Message-ID: <3C83ED1B.45CEDFF6@torrentnet.com>

Hello,

I am looking at using MMPP and FBM (Fractional Bernolli Model) traffic
sources for my simulations.
Could anyone let me know if ns supports these and if they are available.

Thanks,
Rahul


From renan_cunha@hotmail.com  Mon Mar  4 17:05:09 2002
From: renan_cunha@hotmail.com (Renan Cunha)
Date: Mon Mar  4 17:05:09 2002
Subject: [ns] Urgent help with TCP Split
Message-ID: 

   Hi all,

    I am facing a problem with TCP Split and I need to solve it urgently.
    I will try to explain my problem with an example.  My first objective 
was creating an Agent Forwarder which is independent of the transport layer, 
so its works with UDP and TCP. With UDP, the funcionality was achieved. But 
with TCP, I got a problem with the ACK.  Suppose that:

   * Node N1 has an Application/[FTP/Telnet| Exponential/CBR ] agent, for 
instance.

   * Node N2 has an Agent Fowarder and an Application, used to generate 
traffic.

   * Node N3 has a TCPSink agent, in the case of TCP applications, and Null 
agent for UDP


     N1                         N2			     N3

   [Telnet|FTP]/TCP <--->SplitTcpSink->App_Forward/TCP<--->TCPSink
  			  ^                       |
  			  |_______________________|


  There are two applications in this topology, one at N1, which generates 
the packets, and one at N2, which generate the packets at App_Forward as 
they arrive at this node. So at N3, the ack is sent back to N2. At this 
time, TCP calls SplitTcpSink, which generates the ack for N1.

  The tcpsplit code I wrote is almost equal to the one posted by
Dimitris in

http://mailman.isi.edu/pipermail/ns-users/2001-November/019312.html

  The difference is that, in my implementation, the class SplitTcpSink 
inherits directly from TCPSink, and not from public virtual DelAckSink.
  When a packet arrive at SplitTcpSink, it call the App_Forward, which calls 
the procedure sendmsg, of the transport agent, passing as parameter the size 
of the packet which has arrived. When the ack arrives at TCP, it makes the 
SplitTcpSink generate the ack to N1.

  My problem is that the ack has size 0, and the number of acks returned is 
incorrect.

  What must be done with the ack packets that are sent back to the N1 agent? 
What must be their behaviour (which path should they follow)?

  Why DelAckSink is used instead of TCPSink? Is there a problem with pure 
TCPSink?

  Can anyone give me any advice or tell me where I can find the code of an 
implementation of TCP Split?

   Thanks in advance,

   Renan Cunha

_________________________________________________________________
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com


From kkco@cin.ufpe.br  Mon Mar  4 17:35:02 2002
From: kkco@cin.ufpe.br (Karina Karla C. de Oliveira)
Date: Mon Mar  4 17:35:02 2002
Subject: [ns] Error: Diffserv.
Message-ID: 

Dear All.

I was configuring my script with Diffserv when the following error
ocurred:

"ERROR: Policy Table size limit exceeded."

The policy table has a maximum limit to entries? Someone had
this problem too?

Thanks in advance,
Karina Oliveira.

*********************************************************
        Karina Karla Cavalcante de Oliveira
	Computer Science
        Universidade Federal de Pernambuco
        Centro de Informática - CIn
        Caixa Postal 7851
        CEP: 50732-970
        Recife-PE - Brasil
        Phone: +55 (081) 3271-8430

        Email: kkco@cin.ufpe.br
	WWW: www.cin.ufpe.br/~kkco
	Phones: (HOME) +55 (081) 3338-2054
*********************************************************


From JZhao_NS@hotmail.com  Mon Mar  4 18:15:04 2002
From: JZhao_NS@hotmail.com (Jun Zhao)
Date: Mon Mar  4 18:15:04 2002
Subject: [ns] problem with Random::integer
References: 
Message-ID: 

I do not know whether we are using the same version of NS, but I encouter
the same problems. The results of the NS are always the same (even no
difference in a statistically manner).
----- Original Message -----
From: "Saravanan Govindan" 
To: 
Sent: Tuesday, February 26, 2002 9:48 AM
Subject: [ns] problem with Random::integer


>
> Dear All,
>
> I need to generate random numbers for my simulation. I included random.h
and
> used the integer function like so;
>
> int x = Random::integer(range)
>
> However I get the same numbers everytime I run my simulations. I think
that
> I have to change the seed. But how do I do this? I noticed that the RED
> implementation in NS uses the above function but couldn't find if a seed
was
> initialized.
>
> I'd greatly appreciate help.
>
> Thank you.
>
> Regards,
>
> Saravanan Govindan
>
>
>


From hai22@it.canterbury.ac.nz  Mon Mar  4 19:15:02 2002
From: hai22@it.canterbury.ac.nz (hai22)
Date: Mon Mar  4 19:15:02 2002
Subject: [ns] pls help!
References: 
Message-ID: <3C843313.D8C21265@student.canterbury.ac.nz>

Hi,
I think that you should swicth to Linux and install ns latest version on
it rather wasting time on windows. In linux things are much much
easier....believe it.....
a.
==============================================
$?I$?I wrote:
> 
> Hi,bother to you again.I'm sorry.
> I'm really exhausted about installing network simulator ,I almost spend two weeks on the event but seems no progressing at all.I really don't understand the reason resulted in this outcome.
> I really hope and need you can help me out and I seems count on nobody around me.
> now,I first describe my computer which seems differ from yours.I install win98 in drive c.and install win2000
> in drive d.what i can say is just such.
> when i install otcl-1.0a7,system tell me could not find ,then I change it into vc98\include\win.mak>,and then tell me it couldn't find files such as ,and so on.and then I
> replace all these files with their directory like .
> I don't know how to sovle these problems which seems simple for you,I think.
> thanks
> 
> ----------------------------------------------
> ¹úÄÚÊÕ·ÑÓÊÏäË­×îÇ¿£¬Ê×´ÎȨÍþÆÀ²â¼û·ÖÏþ
> http://news.21cn.com/domestic/21CN/2001-11-19/550568.html
> ÂòÆóÒµÓÊÏ䣬»ñÔùÓòÃû
> http://mail.21cn.com/corporation/010.html
> 21cn´øÄú×ß½øÄ¿±êÖ®Âã¬ÏíÊÜÉÌÂÃÀÖȤ
> http://travel.21cn.com/agency/aimtrip/aimtrip.html
> ÊÕ·ÑÓÊÏäÕæÄÜ´óÐÐÆäµÀ£¬¸øÎÒÒ»¸öÀíÓÉ
> http://mail.21cn.com/bijiao/index2-7.html


From kfleung@eee.hku.hk  Mon Mar  4 19:15:14 2002
From: kfleung@eee.hku.hk (Gary Leung)
Date: Mon Mar  4 19:15:14 2002
Subject: [ns] TCP new Reno with SACK
Message-ID: <3C8435FA.BD9144F4@eee.hku.hk>

Dear all,

My research is interested in TCP new Reno with SACK. The simulator
I am using is ns2.1b7a. The question I met is how to construct a TCP new

Reno with SACK in tcl script. Does anybody know how to construct it in
tcl?
Many thanks in advance.

regards,
Gary Leung


From cghsh@hotmail.com  Mon Mar  4 19:45:02 2002
From: cghsh@hotmail.com (Guanghai Cai)
Date: Mon Mar  4 19:45:02 2002
Subject: [ns] Problem of installing otcl-1.0a7 in windows2000
Message-ID: 

This is a multi-part message in MIME format.

------=_NextPart_000_002D_01C1C3CA.B7762D30
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi,
       Sorry to disturb you. I have a question as follow:

        I follow the more detailed instructions for building ns on =
Windows ME and Windows 2000.=20
        Each time when I compile otcl-1.0a7 with the folllowing command:
             nmake /I /f makefile.vc
   system displays:
set LIB=3Dc:\program files\microsoft visual studio\vc98\lib
 c:\program files\microsoft visual studio\vc98\bin\link -debug:full =
-debugtype:cv /NODEFAULTLIB /INCREMENTAL:NO /PDB:NONE /RELEASE /NOLOGO =
-subsystem:console,4.0 -out:otclsh.exe  otclAppInit.o otcl.lib  =
..\tcl8.3.2\win\Release\tcl83.lib  msvcrt.lib oldnames.lib kernel32.lib  =
ws2_32.lib mswsock.lib advapi32.lib user32.lib

Microsoft (R) Program Maintenance Utility Version 6.00.8168.0
Copyright (C) Microsoft Corp 1988-1998. All rights reserved.

'c:\program' is not recognized as an internal or external or external =
command,
operable program or batch file.

        run "dir/w", system displays:
 Volume in drive C is 1
 Volume Serial Number is 3953-1C06

 Directory of C:\NetSim\otcl-1.0a7

[.]             [..]            CHANGES.html    Makefile.in     =
README.html
VERSION         [conf]          config.guess    config.sub      =
configure
configure.in    [doc]           install-sh      [lib]           otcl.c
otcl.h          otclAppInit.c   otkAppInit.c    vc60.pdb        =
makefile.vc   =20
 otcl.o          otcl.lib        otclAppInit.o  =20
              18 File(s)        552,042 bytes
               5 Dir(s)   1,873,166,336 bytes free

        If it is convenient for you, please tell me how to correct it. =
Thank you very much.
        Best regards,

guanghai cai
2002/03/04


        Following is my makefile.vc file :
# Generated automatically from Makefile.in by configure.

#
# try ./configure first to fill in all the definitions corresponding
# to your system, but you always can edit the sections below manually.
#
APPVER=3D4.0
TARGETOS=3DBOTH
TOOLS32 =3D       c:\program files\microsoft visual studio\vc98
cc32=3D  $(TOOLS32)\bin\cl
link32=3D         $(TOOLS32)\bin\link
CFLAGS=3D  -Zi -W3
LIB=3D   lib
INSTALL=3D

#modify the lines below if they do not correspond with your system =
configuration
OTCL_DIR=3D c:\NetSim\otcl-1.0a7
!include    

#
# where to find tcl/tk source (for includes) and binaries (for =
libraries)
#

#modify the lines below if they do not correspond with your system =
configuration
TCLINC=3D         -I..\tcl8.3.2\generic
TCLLIB=3D         ..\tcl8.3.2\win\Release\tcl83.lib
TKINC=3D          -I..\tk8.3.2\generic
TKLIB=3D          ..\tk8.3.2\win\Release\tk83.lib

USE_TCL_80 =3D 1
#STATIC_TCLTK =3D 1

!if "$(USE_TCL_80)" =3D=3D "1"
#modify the lines below if they do not correspond with your system =
configuration
TCLINC=3D         -I..\tcl8.3.2\generic
TKINC=3D          -I..\tk8.3.2\generic
#

!ifndef STATIC_TCLTK

#modify the lines below if they do not correspond with your system =
configuration
TCLLIB=3D         ..\tcl8.3.2\win\Release\tcl83.lib
TKLIB=3D          ..\tk8.3.2\win\Release\tk83.lib
#
!else
TCLLIB=3D  ..\tcl8.0\win\stcl80.lib
TKLIB=3D  ..\tk0.0\win\stk80.lib
!endif
!endif

#
# X11 and other libraries from tcl/tk make as needed to link tclsh and =
wish
#

#modify the lines below if they do not correspond with your system =
configuration
XINC=3D-Ic:\NetSim\tcl8.3.2\win\Release\include
#
XLIB=3D
OTHER_TCL_LIBS=3D=20
OTHER_TK_LIBS=3D

#
# where to install shells, libraries, and includes
#

INST_OTCLSH=3D=20
INST_OWISH=3D=20
INST_OLIB=3D=20
INST_OLIBSH=3D=20
INST_OINC=3D=20


#
# ------------ you shouldn't need to configure below here =
-----------------
#
.SUFFIXES : .cc

CINCLUDES=3D -I. $(TCLINC) $(TKINC) $(XINC)
OTCLLIB=3D otcl.lib

.c.o:
 $(CC) $(cdebug) $(cflags) $(cvarsdll) -c $(CFLAGS) $(CINCLUDES) -Fo$@ =
$<

all: $(OTCLLIB) otclsh.exe

#
# compile an appinit with tcl and otcl and link against
# libotcl.a plus required tcl libs to give a standalone binary
#

otclsh.exe: otclAppInit.o  $(OTCLLIB)
        set LIB=3D$(TOOLS32)\lib
 $(link32) $(ldebug) $(conlflags) -out:$@ \
                otclAppInit.o $(OTCLLIB) \
                $(TCLLIB) $(OTHER_TCL_LIBS) $(conlibsdll) user32.lib

#
# compile an appinit with tcl/tk and otcl and link against
# libotcl.a plus required tcl/tk libs to give a standalone binary
# XXX: this is not working in windows yet.
!IF 0
owish: otkAppInit.c libotcl.a
 del -f libotcl$(SHLIB_SUFFIX)
 $(CC) -o owish $(SHLD_FLAGS) $(CFLAGS) $(CINCLUDES) otkAppInit.c \
 $(OTCLLIB) $(TKLIB) $(TCLLIB) $(XLIB) $(DL_LIBS) \
 $(OTHER_TK_LIBS) $(OTHER_TCL_LIBS)
!ENDIF

#
# compile otcl.c and link it into a library archive
# INCLUDES give a path to tclInt.h (plus tk and X11)
#

$(OTCLLIB): otcl.o
 $(LIB) /name:$(OTCLLIB) /OUT:$(OTCLLIB) otcl.o

#
# compile otcl.c and link it into a shared object
# INCLUDES give a path to tclInt.h (plus tk and X11)
#

libotcl$(SHLIB_SUFFIX): otcl.c
 del -f libotcl$(SHLIB_SUFFIX) otcl.o so_locations
 $(CC) -c $(CFLAGS) $(SHLIB_CFLAGS) $(INCLUDES) otcl.c
 $(SHLIB_LD) -o libotcl$(SHLIB_SUFFIX) otcl.o

test: otclsh owish
 ./otclsh lib/test.tcl
 ./owish lib/test.tcl

install: owish otclsh libotcl.a libotcl$(SHLIB_SUFFIX) otcl.h
 $(INSTALL) $(INST_OWISH) owish
 $(INSTALL) $(INST_OTCLSH) otclsh
 $(INSTALL) $(INST_OLIB) libotcl.a
 $(INSTALL) $(INST_OLIBSH) libotcl$(SHLIB_SUFFIX)
 $(INSTALL) $(INST_OINC) otcl.h

clean:
 -@del /q otclsh.exe otcl.lib otcl.o otkAppInit.o otclAppInit.o=20

binclean:
 -@del otcl.o otkAppInit.o otclAppInit.o \
 config.status config.log config.cache=20

srctar:
 @cwd=3D`pwd` ; dir=3D`basename $$cwd` ; \
     name=3Dotcl-`cat VERSION | tr A-Z a-z` ; \
     tar=3Dotcl-`cat VERSION`.tar.gz ; \
     list=3D"" ; \
     for i in `cat FILES` ; do list=3D"$$list $$name/$$i" ; done; \
     echo \
     "(del -f $$tar; cd .. ; ln -s $$dir $$name)" ; \
      (del -f $$tar; cd .. ; ln -s $$dir $$name) ; \
     echo \
     "(cd .. ; tar cfh $$tar [lots of files])" ; \
      (cd .. ; tar cfh - $$list) | gzip -c > $$tar ; \
     echo \
     "del ../$$name; chmod 444 $$tar" ;  \
      del ../$$name; chmod 444 $$tar


=20


------=_NextPart_000_002D_01C1C3CA.B7762D30
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable













Hi,


       =
Sorry to=20
disturb you. I have a question as follow:


 


        I=20
follow the mo=
re=20
detailed instructions for building ns on Windows ME and Windows =
2000. 


     =20
  Each time when I compile =
otcl-1.0a7=20
with the folllowing command:


           =
 =20
nmake /I /f makefile.vc


   system displays:


set=20
LIB=3Dc:\program files\microsoft visual =
studio\vc98\lib
 c:\program=20
files\microsoft visual studio\vc98\bin\link -debug:full -debugtype:cv=20
/NODEFAULTLIB /INCREMENTAL:NO /PDB:NONE /RELEASE /NOLOGO =
-subsystem:console,4.0=20
-out:otclsh.exe  otclAppInit.o otcl.lib =20
..\tcl8.3.2\win\Release\tcl83.lib  msvcrt.lib oldnames.lib=20
kernel32.lib  ws2_32.lib mswsock.lib advapi32.lib=20
user32.lib


 


Microsoft (R)=20
Program Maintenance Utility Version 6.00.8168.0


Copyright (C)=20
Microsoft Corp 1988-1998. All rights reserved.


 


'c:\program'=20
is not recognized as an internal or external or external=20
command,


operable=20
program or batch file.


 


        run "dir/w", =
system=20
displays:


 Volume in drive C is 1
 Volume Serial Number is=20
3953-1C06


 


 Directory of C:\NetSim\otcl-1.0a7


 


[.]          &=
nbsp; =20
[..]           =20
CHANGES.html    Makefile.in    =20
README.html
VERSION        =20
[conf]         =20
config.guess    config.sub     =20
configure
configure.in   =20
[doc]          =20
install-sh     =20
[lib]          =20
otcl.c
otcl.h         =20
otclAppInit.c   otkAppInit.c   =20
vc60.pdb       =20
makefile.vc    


 otcl.o        &nbs=
p;=20
otcl.lib       =20
otclAppInit.o   


          &nbs=
p;   18=20
File(s)        552,042=20
bytes
          &nbs=
p;   =20
5 Dir(s)   1,873,166,336 bytes free


        =
If it=20
is convenient for you, please tell me how to correct it. Thank you very=20
much.


        Best =

regards,


 


guanghai=20
cai


2002/03/04


 


 


        Following is my =

makefile.vc file :


# Generated automatically from Makefile.in by =
configure.


 


#
# try ./configure first to fill in all the definitions=20
corresponding
# to your system, but you always can edit the sections =
below=20
manually.
#
APPVER=3D4.0
TARGETOS=3DBOTH
TOOLS32=20
=3D       c:\program files\microsoft =
visual=20
studio\vc98
cc32=3D  $(TOOLS32)\bin\cl
link32=3D &nb=
sp;      =20
$(TOOLS32)\bin\link
CFLAGS=3D  -Zi -W3
LIB=3D=20
  lib
INSTALL=3D


 


#modify the lines below if they do not correspond with your =
system=20
configuration
OTCL_DIR=3D =
c:\NetSim\otcl-1.0a7
!include   =20
<c:\program files\microsoft visual=20
studio\vc98\include\win32.mak>


 


#
# where to find tcl/tk source (for includes) and binaries =
(for=20
libraries)
#


 


#modify the lines below if they do not correspond with your =
system=20
configuration
TCLINC=3D        =
;=20
-I..\tcl8.3.2\generic
TCLLIB=3D      &nb=
sp; =20
..\tcl8.3.2\win\Release\tcl83.lib
TKINC=3D    &nbs=
p;    =20
-I..\tk8.3.2\generic
TKLIB=3D       =
;  =20
..\tk8.3.2\win\Release\tk83.lib


 


USE_TCL_80 =3D 1
#STATIC_TCLTK =3D 1


 


!if "$(USE_TCL_80)" =3D=3D "1"
#modify the lines below if =
they do not=20
correspond with your system=20
configuration
TCLINC=3D        =
;=20
-I..\tcl8.3.2\generic
TKINC=3D      &nbs=
p;  =20
-I..\tk8.3.2\generic
#


 


!ifndef STATIC_TCLTK


 


#modify the lines below if they do not correspond with your =
system=20
configuration
TCLLIB=3D        =
;=20
..\tcl8.3.2\win\Release\tcl83.lib
TKLIB=3D    &nbs=
p;    =20
..\tk8.3.2\win\Release\tk83.lib
#
!else
TCLLIB=3D  ..\=
tcl8.0\win\stcl80.lib
TKLIB=3D  ..\tk0.0\win\stk80.lib
!e=
ndif
!endif


 


#
# X11 and other libraries from tcl/tk make as needed to =
link tclsh=20
and wish
#


 


#modify the lines below if they do not correspond with your =
system=20
configuration
XINC=3D-Ic:\NetSim\tcl8.3.2\win\Release\include
#
=
XLIB=3D
OTHER_TCL_LIBS=3D=20

OTHER_TK_LIBS=3D


 


#
# where to install shells, libraries, and =
includes
#


 


INST_OTCLSH=3D 
INST_OWISH=3D 
INST_OLIB=3D =
;
INST_OLIBSH=3D 
INST_OINC=3D 


 



#
# ------------ you shouldn't need to configure below =
here=20
-----------------
#
.SUFFIXES : .cc


 


CINCLUDES=3D -I. $(TCLINC) $(TKINC) $(XINC)
OTCLLIB=3D=20
otcl.lib


 


.c.o:
 $(CC) $(cdebug) $(cflags) $(cvarsdll) -c =
$(CFLAGS)=20
$(CINCLUDES) -Fo$@ $<


 


all: $(OTCLLIB) otclsh.exe


 


#
# compile an appinit with tcl and otcl and link =
against
#=20
libotcl.a plus required tcl libs to give a standalone =
binary
#


 


otclsh.exe: otclAppInit.o =20
$(OTCLLIB)
        set=20
LIB=3D$(TOOLS32)\lib
 $(link32) $(ldebug) $(conlflags) -out:$@=20
\
           &n=
bsp;   =20
otclAppInit.o $(OTCLLIB)=20
\
           &n=
bsp;   =20
$(TCLLIB) $(OTHER_TCL_LIBS) $(conlibsdll) user32.lib


 


#
# compile an appinit with tcl/tk and otcl and link =
against
#=20
libotcl.a plus required tcl/tk libs to give a standalone binary
# =
XXX: this=20
is not working in windows yet.
!IF 0
owish: otkAppInit.c=20
libotcl.a
 del -f libotcl$(SHLIB_SUFFIX)
 $(CC) -o owish =

$(SHLD_FLAGS) $(CFLAGS) $(CINCLUDES) otkAppInit.c \
 $(OTCLLIB) =
$(TKLIB)=20
$(TCLLIB) $(XLIB) $(DL_LIBS) \
 $(OTHER_TK_LIBS)=20
$(OTHER_TCL_LIBS)
!ENDIF


 


#
# compile otcl.c and link it into a library archive
# =
INCLUDES=20
give a path to tclInt.h (plus tk and X11)
#


 


$(OTCLLIB): otcl.o
 $(LIB) /name:$(OTCLLIB) =
/OUT:$(OTCLLIB)=20
otcl.o


 


#
# compile otcl.c and link it into a shared object
# =
INCLUDES=20
give a path to tclInt.h (plus tk and X11)
#


 


libotcl$(SHLIB_SUFFIX): otcl.c
 del -f =
libotcl$(SHLIB_SUFFIX)=20
otcl.o so_locations
 $(CC) -c $(CFLAGS) $(SHLIB_CFLAGS) =
$(INCLUDES)=20
otcl.c
 $(SHLIB_LD) -o libotcl$(SHLIB_SUFFIX) otcl.o


 


test: otclsh owish
 ./otclsh =
lib/test.tcl
 ./owish=20
lib/test.tcl


 


install: owish otclsh libotcl.a libotcl$(SHLIB_SUFFIX)=20
otcl.h
 $(INSTALL) $(INST_OWISH) owish
 $(INSTALL)=20
$(INST_OTCLSH) otclsh
 $(INSTALL) $(INST_OLIB)=20
libotcl.a
 $(INSTALL) $(INST_OLIBSH)=20
libotcl$(SHLIB_SUFFIX)
 $(INSTALL) $(INST_OINC) =
otcl.h


 


clean:
 -@del /q=20
otclsh.exe otcl.lib otcl.o otkAppInit.o otclAppInit.o 


 


binclean:
 -@del=20
otcl.o otkAppInit.o otclAppInit.o \
 config.status config.log=20
config.cache 


 


srctar:
 @cwd=3D`pwd` ; dir=3D`basename $$cwd` ;=20
\
     name=3Dotcl-`cat VERSION | tr A-Z a-z` ;=20
\
     tar=3Dotcl-`cat VERSION`.tar.gz ;=20
\
     list=3D"" ; \
     =
for i in=20
`cat FILES` ; do list=3D"$$list $$name/$$i" ; done; =
\
    =20
echo \
     "(del -f $$tar; cd .. ; ln -s $$dir =
$$name)"=20
; \
      (del -f $$tar; cd .. ; ln -s $$dir =
$$name)=20
; \
     echo \
     "(cd =
.. ; tar=20
cfh $$tar [lots of files])" ; \
      (cd .. =
; tar=20
cfh - $$list) | gzip -c > $$tar ; \
     echo=20
\
     "del ../$$name; chmod 444 $$tar" ; =20
\
      del ../$$name; chmod 444 =
$$tar


 



 


 


------=_NextPart_000_002D_01C1C3CA.B7762D30--


From JZhao_NS@hotmail.com  Mon Mar  4 19:45:14 2002
From: JZhao_NS@hotmail.com (Jun Zhao)
Date: Mon Mar  4 19:45:14 2002
Subject: [ns] pls help!
References:  <3C843313.D8C21265@student.canterbury.ac.nz>
Message-ID: 

Quite agree!

Many guys run NS in linux and do other work in Windows. Things become easier
if you follow this method.

Why waste so much time on the system and other things you are not familiar?
Seems that we spend 90% of our time on the system configurations and 10% to
work on our research idea!

----- Original Message -----
From: "hai22" 
To: "$'I$'I" ; 
Sent: Tuesday, March 05, 2002 10:53 AM
Subject: Re: [ns] pls help!


>
> Hi,
> I think that you should swicth to Linux and install ns latest version on
> it rather wasting time on windows. In linux things are much much
> easier....believe it.....
> a.
> ==============================================
> $?I$?I wrote:
> >
> > Hi,bother to you again.I'm sorry.
> > I'm really exhausted about installing network simulator ,I almost spend
two weeks on the event but seems no progressing at all.I really don't
understand the reason resulted in this outcome.
> > I really hope and need you can help me out and I seems count on nobody
around me.
> > now,I first describe my computer which seems differ from yours.I install
win98 in drive c.and install win2000
> > in drive d.what i can say is just such.
> > when i install otcl-1.0a7,system tell me could not find ,then I
change it into > vc98\include\win.mak>,and then tell me it couldn't find files such as
,and so on.and then I
> > replace all these files with their directory like .
> > I don't know how to sovle these problems which seems simple for you,I
think.
> > thanks
> >
> > ----------------------------------------------
> > ¹úÄÚÊÕ·ÑÓÊÏäË­×îÇ¿£¬Ê×´ÎȨÍþÆÀ²â¼û·ÖÏþ
> > http://news.21cn.com/domestic/21CN/2001-11-19/550568.html
> > ÂòÆóÒµÓÊÏ䣬»ñÔùÓòÃû
> > http://mail.21cn.com/corporation/010.html
> > 21cn´øÄú×ß½øÄ¿±êÖ®Âã¬ÏíÊÜÉÌÂÃÀÖȤ
> > http://travel.21cn.com/agency/aimtrip/aimtrip.html
> > ÊÕ·ÑÓÊÏäÕæÄÜ´óÐÐÆäµÀ£¬¸øÎÒÒ»¸öÀíÓÉ
> > http://mail.21cn.com/bijiao/index2-7.html
>
>


From xuanc@ISI.EDU  Tue Mar  5 05:15:04 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Tue Mar  5 05:15:04 2002
Subject: [ns] Error: Diffserv.
In-Reply-To: 
Message-ID: 

Yes, both the policy and policer tables have a limit. But you can work 
arround it by hacking the code. 


On Mon, 4 Mar 2002, Karina Karla C. de Oliveira wrote:

> 
> 
> Dear All.
> 
> I was configuring my script with Diffserv when the following error
> ocurred:
> 
> "ERROR: Policy Table size limit exceeded."
> 
> The policy table has a maximum limit to entries? Someone had
> this problem too?
> 
> Thanks in advance,
> Karina Oliveira.
> 
> *********************************************************
>         Karina Karla Cavalcante de Oliveira
> 	Computer Science
>         Universidade Federal de Pernambuco
>         Centro de Informática - CIn
>         Caixa Postal 7851
>         CEP: 50732-970
>         Recife-PE - Brasil
>         Phone: +55 (081) 3271-8430
> 
>         Email: kkco@cin.ufpe.br
> 	WWW: www.cin.ufpe.br/~kkco
> 	Phones: (HOME) +55 (081) 3338-2054
> *********************************************************
> 

-- 
Xuan Chen
USC/ISI


From easy@mail.ustc.edu.cn  Tue Mar  5 05:15:15 2002
From: easy@mail.ustc.edu.cn (xu min)
Date: Tue Mar  5 05:15:15 2002
Subject: [ns] install ns-allinone-2.1b6 problem
Message-ID: <200203050615.OAA00614@mx1.ustc.edu.cn>

hi, could someone help me?
I cannot installed ns-allinone-2.1b6 in mandrake linux 8.1.
The error is:

make: ***[tclPosixStr.o] Erro1
tcl 8.0.4 make failed! Exiting...

what should I do?

Your assistance is greatly appreciated! 

Xu Min


From easy@mail.ustc.edu.cn  Tue Mar  5 05:15:26 2002
From: easy@mail.ustc.edu.cn (xu min)
Date: Tue Mar  5 05:15:26 2002
Subject: [ns] install different versions of ns2 in different users ?
Message-ID: <200203050719.PAA04945@mx1.ustc.edu.cn>

hi, everyone!
is it ok if I install different versions of ns2 in different users of linux?

Thanks first!

Regards,

xu min


From jill2k@privacy-network.com  Tue Mar  5 05:15:38 2002
From: jill2k@privacy-network.com (jill2k)
Date: Tue Mar  5 05:15:38 2002
Subject: [ns] Trying sample tcl
In-Reply-To: <200203050745.g257jpu19355@gamma.isi.edu>
Message-ID: 

Dear all,

I am new in using ns2.1b7a. When i try tcl script at
../ns-tutorial/example/wireless2.tcl, I get the folowing error (at the
botom). Why i get this message? Help me.....

Thanks
nangwid

===========================================
num_nodes is set 3
warning: Please use -channel as shown in tcl/ex/wireless-mitf.tcl
__FAILED_SHADOW_OBJECT_
    while executing
"error "__FAILED_SHADOW_OBJECT_" """
    (procedure "set-hieraddr" line 4)
    (SplitObject init line 4)
    invoked from within
"set-hieraddr init 1.0.0"
    (Class create line 1)
    invoked from within
"AddrParams create set-hieraddr 1.0.0"
    ("eval" body line 1)
    invoked from within
"eval [list $self] create [list $m] $args         "
    (procedure "AddrParams" line 5)
    (Class unknown line 5)
    invoked from within
"AddrParams set-hieraddr  [$BS(0) node-addr"
    ("for" body line 3)
    invoked from within
"for {set j 0} {$j < $opt(nn)} {incr j} {
    set node_($j) [ $ns_ node [lindex $temp \
            [expr $j+1]] ]
    $node_($j) base-station [AddrParams set..."
    (file "wireless2.tcl" line 138)


From kaiduanx@yahoo.ca  Tue Mar  5 05:15:48 2002
From: kaiduanx@yahoo.ca (kaiduan xie)
Date: Tue Mar  5 05:15:48 2002
Subject: [ns] Entry for Hierarchical routing
Message-ID: <20020305082337.45359.qmail@web14808.mail.yahoo.com>

--0-1870020351-1015316617=:42852
Content-Type: text/plain; charset=us-ascii


hi,

Can anyone tell me the entry point for packet forwarding using hierarchical routing ? I want to study how hierarchical routing work.Normally classify() in classifier-addr class takes care of forwarding data packet. However, in hierarchical node, there is an array of classifier, classifiers_. How the classifiers_ is used to forward the data packet? Thanx for help.

kaiduan xie

 



---------------------------------
Find, Connect, Date! Yahoo! Canada Personals.

--0-1870020351-1015316617=:42852
Content-Type: text/html; charset=us-ascii


hi,


Can anyone tell me the entry point for packet forwarding using hierarchical routing ? I want to study how hierarchical routing work.Normally classify() in classifier-addr class takes care of forwarding data packet. However, in hierarchical node, there is an array of classifier, classifiers_. How the classifiers_ is used to forward the data packet? Thanx for help.


kaiduan xie


 

Find, Connect, Date! Yahoo! Canada Personals.

--0-1870020351-1015316617=:42852--


From tsasaki@sys.i.kyoto-u.ac.jp  Tue Mar  5 05:16:04 2002
From: tsasaki@sys.i.kyoto-u.ac.jp (SASAKI Takashi)
Date: Tue Mar  5 05:16:04 2002
Subject: [ns] deb or rpm package
In-Reply-To: 
References: 
 <3C843313.D8C21265@student.canterbury.ac.nz>
 
Message-ID: <20020305184252W.tsasaki@sys.i.kyoto-u.ac.jp>

Hi, all.

Does anyone have binary package for Debian GNU/Linux or RedHat Linux ?

I'm using ns2 build from all-in-one source package on my Linux box.
It works very well of course. But if there is binary package,
the installation procedure is easier.

If nobody has ever done it, I'm planning to make binary package 
for Debian GNU/Linux (for i386 and later).

Regards,

Takashi SASAKI 


From najma_ismat@hotmail.com  Tue Mar  5 05:16:15 2002
From: najma_ismat@hotmail.com (Najma Ismat)
Date: Tue Mar  5 05:16:15 2002
Subject: [ns] How to get information from the trace file?Urgent!!!!!!!!!!!!!!!!!!!!!
Message-ID: 


hi,
i have modified my ns-2.1b8a code with elfn. i have run some scripts but now

the trace file format has been changed from what i was getting earlier. i 
have found in cmu-trace.cc that there are two trace formats. before 
modifying ns2 source code i was getting trace file in this format:
"%c %.9f _%d_ %3s %4s %d %s %d [%x %x %x %x] "
but now i am getting it in the format below:
"%c %.9f %d (%6.2f %6.2f) %3s %4s %d %s %d [%x %x %x %x] "
the problem i am facing is in counting total number of receive tcp packets 
because in the format mentioned first "tcp" name appears in the seventh 
column ( if i used perl script for counting receive packet)where as in the 
formt defined latter tcp may appear in 9th column. my perl script is not 
recognizing this and when i run it gives 0 packets receive.can anyone guide 
me that my interpretation for  tcp packet (9th column) is right or 
wrong.thanks in advance.
Najma Ismat

r 18.756601786 7 (790.96  62.20) RTR  --- 262 tcp 1500 [a2 7 18 800] -------

[11:0 12:0 32 7] [85 0] 3 0

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp.


From najma_ismat@hotmail.com  Tue Mar  5 05:16:27 2002
From: najma_ismat@hotmail.com (Najma Ismat)
Date: Tue Mar  5 05:16:27 2002
Subject: [ns] How to get information from the trace file?Urgent!!!!!!!!!!!!!!!!!!!!!
Message-ID: 


hi,
i have modified my ns-2.1b8a code with elfn. i have run some scripts but now

the trace file format has been changed from what i was getting earlier. i 
have found in cmu-trace.cc that there are two trace formats. before 
modifying ns2 source code i was getting trace file in this format:
"%c %.9f _%d_ %3s %4s %d %s %d [%x %x %x %x] "
but now i am getting it in the format below:
"%c %.9f %d (%6.2f %6.2f) %3s %4s %d %s %d [%x %x %x %x] "
the problem i am facing is in counting total number of receive tcp packets 
because in the format mentioned first "tcp" name appears in the seventh 
column ( if i used perl script for counting receive packet)where as in the 
formt defined latter tcp may appear in 9th column. my perl script is not 
recognizing this and when i run it gives 0 packets receive.can anyone guide 
me that my interpretation for  tcp packet (9th column) is right or 
wrong.thanks in advance.
Najma Ismat

r 18.756601786 7 (790.96  62.20) RTR  --- 262 tcp 1500 [a2 7 18 800] -------

[11:0 12:0 32 7] [85 0] 3 0

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp.


From Stefan.Rank.extern@icn.siemens.de  Tue Mar  5 06:00:02 2002
From: Stefan.Rank.extern@icn.siemens.de (Rank Stefan ICM N PG U SE D 8)
Date: Tue Mar  5 06:00:02 2002
Subject: [ns] On the mac-802_11
Message-ID: <5B4D0C5BA65ECA46969C1419122317E62D7F97@mchh161e>

 Hi,

I think that there is no collision, because the power difference is too big.

Bye,
Stefan

-----Original Message-----
From: Milind Dilip Kopikare
To: ns-users@ISI.EDU
Sent: 3/3/02 5:12 PM
Subject: [ns] On the mac-802_11


Hi,
	In mac-802_11.cc, the following code in
void Mac802_11::recv(Packet *p, Handler *h) function:-
====================
		/*
		 *  If the power of the incoming packet is smaller than
the
		 *  power of the packet currently being received by at
least
                 *  the capture threshold, then we ignore the new
packet.
		 */
		if(pktRx_->txinfo_.RxPr / p->txinfo_.RxPr >=
p->txinfo_.CPThresh) {
			capture(p);
		} else {
			collision(p);
		}

====================

Dosen't seem right. Shouldn't we have collision when the if() condition
is
true?

Thanks,
Milind



From Stefan.Rank.extern@icn.siemens.de  Tue Mar  5 06:05:01 2002
From: Stefan.Rank.extern@icn.siemens.de (Rank Stefan ICM N PG U SE D 8)
Date: Tue Mar  5 06:05:01 2002
Subject: [ns] Counting packets sent by nodes
Message-ID: <5B4D0C5BA65ECA46969C1419122317E62D7F98@mchh161e>

Hi,

maybe a queue monitor can help.

Bye,
STefan 

-----Original Message-----
From: Asaf Shabtai
To: ns-users@ISI.EDU
Sent: 3/2/02 4:53 PM
Subject: [ns] Counting packets sent by nodes


Hi,
I'm tring to add a variable in the node, so each time a packet is sent
by 
the node to the next node te variable is change (for example let's say
that 
the each time the node sends a packet its variable is incremented by 7).
Does anybody have an idea how to implenet that ?
Thank U
Asaf

_________________________________________________________________
Join the world?s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com


From mkavouri@ics.forth.gr  Tue Mar  5 07:40:02 2002
From: mkavouri@ics.forth.gr (Matina Kavouridou)
Date: Tue Mar  5 07:40:02 2002
Subject: [ns] udp packet size, help!!
Message-ID: 

Hello,
I am running some wireless simulations and when I am using UDP/CBR it
seems that nodes cannot send packets bigger than 1000 bytes- this is what
I have observed at AGT level in trace files. I mean that when I define a
packet of e.g. 1600 bytes - using cbrgen.tcl script- it is fragmented in
1000 and 600bytes.
Could you suggest anything?

Matina.


From wells@ieee.org  Tue Mar  5 08:00:01 2002
From: wells@ieee.org (John Wells)
Date: Tue Mar  5 08:00:01 2002
Subject: [ns] deb or rpm package
In-Reply-To: <20020305184252W.tsasaki@sys.i.kyoto-u.ac.jp>; from tsasaki@sys.i.kyoto-u.ac.jp on Tue, Mar 05, 2002 at 06:42:52PM +0900
References:  <3C843313.D8C21265@student.canterbury.ac.nz>  <20020305184252W.tsasaki@sys.i.kyoto-u.ac.jp>
Message-ID: <20020305105503.A8808@io.irean.vt.edu>

Takashi,

I don't believe so, for the simple reason that most folks on the list need 
to modify the source.  It would, however, be a good idea for students who 
need to install it on their machines for classes and such.  I for one 
would be interested in a binary package.

Thanks,
John

On Tue, Mar 05, 2002 at 06:42:52PM +0900, SASAKI Takashi wrote:
> 
> Hi, all.
> 
> Does anyone have binary package for Debian GNU/Linux or RedHat Linux ?
> 
> I'm using ns2 build from all-in-one source package on my Linux box.
> It works very well of course. But if there is binary package,
> the installation procedure is easier.
> 
> If nobody has ever done it, I'm planning to make binary package 
> for Debian GNU/Linux (for i386 and later).
> 
> Regards,
> 
> Takashi SASAKI 

-- 
John Wells, Virginia Tech Networking Lab
(tel) +1 540 231-8347 (web) http://www.ee.vt.edu/~wells


From wells@ieee.org  Tue Mar  5 08:05:01 2002
From: wells@ieee.org (John Wells)
Date: Tue Mar  5 08:05:01 2002
Subject: [ns] TCP new Reno with SACK
In-Reply-To: <3C8435FA.BD9144F4@eee.hku.hk>; from kfleung@eee.hku.hk on Tue, Mar 05, 2002 at 11:05:30AM +0800
References: <3C8435FA.BD9144F4@eee.hku.hk>
Message-ID: <20020305105817.B8808@io.irean.vt.edu>

Gary,

My guess would be to take one of the TCP agents and hack it to do new Reno 
with SACK.  This is C++ code, not Tcl, though..

-john

On Tue, Mar 05, 2002 at 11:05:30AM +0800, Gary Leung wrote:
> 
> Dear all,
> 
> My research is interested in TCP new Reno with SACK. The simulator
> I am using is ns2.1b7a. The question I met is how to construct a TCP new
> 
> Reno with SACK in tcl script. Does anybody know how to construct it in
> tcl?
> Many thanks in advance.
> 
> regards,
> Gary Leung

-- 
John Wells, Virginia Tech Networking Lab
(tel) +1 540 231-8347 (web) http://www.ee.vt.edu/~wells


From wells@ieee.org  Tue Mar  5 08:10:01 2002
From: wells@ieee.org (John Wells)
Date: Tue Mar  5 08:10:01 2002
Subject: [ns] How to get information from the trace file?Urgent!!!!!!!!!!!!!!!!!!!!!
In-Reply-To: ; from najma_ismat@hotmail.com on Tue, Mar 05, 2002 at 03:27:15PM +0500
References: 
Message-ID: <20020305110408.C8808@io.irean.vt.edu>

Najma,

Try to track the problem in the source code by finding which procedure is
called to print your trace lines using printfs or a debugger, then look at
the print statement used to print out your trace line.  I haven't tried
it, but tcptrace can analyze ns trace files.  You might try that.

-john

On Tue, Mar 05, 2002 at 03:27:15PM +0500, Najma Ismat wrote:
> 
> 
> 
> hi,
> i have modified my ns-2.1b8a code with elfn. i have run some scripts but now
> 
> the trace file format has been changed from what i was getting earlier. i 
> have found in cmu-trace.cc that there are two trace formats. before 
> modifying ns2 source code i was getting trace file in this format:
> "%c %.9f _%d_ %3s %4s %d %s %d [%x %x %x %x] "
> but now i am getting it in the format below:
> "%c %.9f %d (%6.2f %6.2f) %3s %4s %d %s %d [%x %x %x %x] "
> the problem i am facing is in counting total number of receive tcp packets 
> because in the format mentioned first "tcp" name appears in the seventh 
> column ( if i used perl script for counting receive packet)where as in the 
> formt defined latter tcp may appear in 9th column. my perl script is not 
> recognizing this and when i run it gives 0 packets receive.can anyone guide 
> me that my interpretation for  tcp packet (9th column) is right or 
> wrong.thanks in advance.
> Najma Ismat
> 
> r 18.756601786 7 (790.96  62.20) RTR  --- 262 tcp 1500 [a2 7 18 800] -------
> 
> [11:0 12:0 32 7] [85 0] 3 0
> 
> _________________________________________________________________
> Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp.

-- 
John Wells, Virginia Tech Networking Lab
(tel) +1 540 231-8347 (web) http://www.ee.vt.edu/~wells


From haldar@ISI.EDU  Tue Mar  5 08:15:01 2002
From: haldar@ISI.EDU (Padmaparna Haldar)
Date: Tue Mar  5 08:15:01 2002
Subject: [ns] Entry for Hierarchical routing
In-Reply-To: <20020305082337.45359.qmail@web14808.mail.yahoo.com>
Message-ID: 

The hier-classifiers work in a way very similar to the flat classifier. 
The only difference being each node has n level of classifiers denoted by 
clsfr_[n]. The entry point is the clsfr at level 0, which inturn points to 
either the next level clsfr[1] or directly to a link object depending on 
the route entries.
See hierarchicval routing chapter in ns-manual for details.
--Padma

On Tue, 5 Mar 2002, kaiduan xie wrote:


> 
> hi,
> 
> Can anyone tell me the entry point for packet forwarding using hierarchical routing ? I want to study how hierarchical routing work.Normally classify() in classifier-addr class takes care of forwarding data packet. However, in hierarchical node, there is an array of classifier, classifiers_. How the classifiers_ is used to forward the data packet? Thanx for help.
> 
> kaiduan xie
> 
>  
> 
> 
> 
> ---------------------------------
> Find, Connect, Date! Yahoo! Canada Personals.
> 

-- 
------------------------------------------------ 
Be true to your work, your word, and your friend. 
--Thoreau

Padmaparna Haldar


From jelger@clarinet.u-strasbg.fr  Tue Mar  5 08:15:13 2002
From: jelger@clarinet.u-strasbg.fr (Christophe Jelger)
Date: Tue Mar  5 08:15:13 2002
Subject: [ns] Several questions on packet transmission
Message-ID: <3C84EE52.70205@clarinet.u-strasbg.fr>

Hi all,

I have a few questions for NS specialists ... (well I strongly hope that 
there a few out there !!)

First of all, I want to create a new UDP agent. There is no much problem 
with that, everything is in the manual. However, I face a number of 
problems as follows ... I strongly apologize for the messy way I will 
present the different problems ... ;-)

1. In the PING agent example, the sender agent is explicitly connected 
to the receiver agent (with something like $ns connect sender receiver). 
So in this case when one agent wants to send a packet to the other 
(connected) agent, it simply uses the SEND(pkt,hdr) function. This is 
the function related to the CONNECTOR object. In my understanding, this 
can be done because the two agents were explicitly connected ? Correct ?

2. If now I do not connect the agents together, how can I explicitly 
send a packet to a specific node? (something like send(node, pkt, hdr)) 
If I explicitly set the dest address, it is enough to call the send(pkt, 
hdr) function ? (I'm afraid I can't ...)

3. Now a more complex question ... what happens if I want to do the same 
thing as above (2.) but I want to send a packet through a specific link 
?(as it is done in multicast if I just know the output interface) Can I 
use something like : send(link, pkt, hdr) ?

4. And last but not least, where can I find detailed documentation on 
how to use the classifier ?


Thanks a lot. I hope somebody can answer these questions (at least some 
of them !)

Regards,

Christophe



From olemissdiary@yahoo.com  Tue Mar  5 08:50:01 2002
From: olemissdiary@yahoo.com (Yang Li)
Date: Tue Mar  5 08:50:01 2002
Subject: [ns] Re: Questions about target_->recv()
Message-ID: <200203051648.g25GmWt01480@tnt.isi.edu>

This is a multi-part message in MIME format.

--=====002_Dragon651534028382_=====
Content-Type: text/plain;
      charset="us-ascii"
Content-Transfer-Encoding: base64
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--=====002_Dragon651534028382_=====
Content-Type: text/html;
      charset="us-ascii"
Content-Transfer-Encoding: base64
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--=====002_Dragon651534028382_=====--



_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


From xuanc@ISI.EDU  Tue Mar  5 09:10:02 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Tue Mar  5 09:10:02 2002
Subject: [ns] Error: Diffserv.
In-Reply-To: 
Message-ID: 

On Tue, 5 Mar 2002, Karina Karla C. de Oliveira wrote:

> 
> Dear Xuan.
> 
> I already done this modification but I needed of a confirmation if it will
> not cause more problems.
I don't think it will cause further problems.

> 
> On Mon, 4 Mar 2002, Xuan Chen wrote:
> 
> > Yes, both the policy and policer tables have a limit. But you can work
> > arround it by hacking the code.
> 
> 
You basically need to play with those policies. I would suggest you to 
look at the samples cripts and chapter 9 in ns manual. Hope they can help.

-chen

> 
> I am having problems with the configuration to Best Effort Traffic. The
> delay and jitter values aren't decreasing significantly.
> Then I think that I am not configuring the traffic correctly.
> 
> Could you help me then. I would like to know which policy, schedular mode,
> MRED,etc.., I should to use to EF e BE traffics.
> 
> I configured the router to accept more than a traffic to a pair of
> nodes... Yes!!!!
> 
> 
> Best Regards,
> Karina Oliveira.
> 
> 

-- 
Xuan Chen
USC/ISI


From csysmd@comp.leeds.ac.uk  Tue Mar  5 10:05:08 2002
From: csysmd@comp.leeds.ac.uk (Steve)
Date: Tue Mar  5 10:05:08 2002
Subject: [ns] Traceable variables
Message-ID: 

Hi,

for my research I need to be able to trace certain variables from ns.
Could anyone please give me a list, or tell me where there is a complete
list (can't find one in the manual) of all variables in ns packets and
nodes that it is possible to easily trace. This would really help me,

thanks,

Steve


From eehjy@ust.hk  Tue Mar  5 10:15:01 2002
From: eehjy@ust.hk (Jingyi)
Date: Tue Mar  5 10:15:01 2002
Subject: [ns] udp packet size, help!!
References: 
Message-ID: <003101c1c471$453da160$5d9c598f@resnet.ust.hk>

You need also set the UDP packet size. I think adding the following to your tcl script will solve the problem:

Agent/UDP set packetSize_ 1600

Hope it helps.

Jingyi

----- Original Message ----- 
From: "Matina Kavouridou" 
To: 
Sent: Tuesday, March 05, 2002 11:38 PM
Subject: [ns] udp packet size, help!!


> 
> Hello,
> I am running some wireless simulations and when I am using UDP/CBR it
> seems that nodes cannot send packets bigger than 1000 bytes- this is what
> I have observed at AGT level in trace files. I mean that when I define a
> packet of e.g. 1600 bytes - using cbrgen.tcl script- it is fragmented in
> 1000 and 600bytes.
> Could you suggest anything?
> 
> Matina.
> 
> 


From renan_cunha@hotmail.com  Tue Mar  5 11:10:08 2002
From: renan_cunha@hotmail.com (Renan Cunha)
Date: Tue Mar  5 11:10:08 2002
Subject: [ns] Urgent help with TCP Split
Message-ID: 

Hello,

Thanks for answering.

In fact, in my model N2 will work as a router or an access point. Suppose 
that N1 is an application server, and N3 is a client. I want to avaliate the 
connection between N1 and N3 (through N2).
I want to simulate that, because I am supposing that N1 and N3 are not able 
to communicate directly because they have different hardware features, for 
instance.

If a TCP packet from N1 is lost in the link N2-N3, in your model, the node 
N2 will be responsible for the retransmission of that packet. As N2 is just 
a access point, it is not supposed that it can retransmit packets (just 
forward them). In my model, it doesn't matter where the packet is lost, N1 
will always be responsible for the retransmissions.

You said this may be the reason the acks have size 0. Why ??
Am I missing something ?

Bye,

Renan


>hi,
>i'm working on tcpsplit too.
>there's sthing i don't understand. Why are u not sending an ack back to n1
>as soon as n2 receives a packet from n1 and forwards it. That would be the
>way tcp spoofing works (local acks ). In ur case it's like an end2end that
>is performed in 2 steps and ur tcpsplit agent acts just as a router. Is it
>that u want that behavior for specific reasons ? that may be why the ack 
>has
>0 size.
>
>let me know if i missed sthing
>bye


_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com


From whuang@student.egr.unlv.edu  Tue Mar  5 13:55:12 2002
From: whuang@student.egr.unlv.edu (Wei Huang)
Date: Tue Mar  5 13:55:12 2002
Subject: [ns] Help: Synchronous Algorithm on ns
Message-ID: 

hello,

I'd like to simulate the synchronous algorithms, such as BYZANTINE FAILURE
MODEL & BYZANTINE AGREEMENT PROTOCOLS.

since the most of practical protocols are Asynchronous, I'm wonderring if
ns supports Synchronous algorithms.

Help, help ... urgent !!!

Wei


From bulggie@yahoo.com  Tue Mar  5 17:10:01 2002
From: bulggie@yahoo.com (=?iso-8859-1?q?Frank=20Akujobi?=)
Date: Tue Mar  5 17:10:01 2002
Subject: [ns] Re: Bug in NS-2.1b8a for ECN with "gentle_" ??? (fwd)
In-Reply-To: 
Message-ID: <20020306010620.18385.qmail@web12305.mail.yahoo.com>

Hi Chen,
I still have not received any response on this issue. 
Pls endeavour to give a response.

Thanks,
Frank Akujobi.

--- Xuan Chen  wrote: > Hi, Sally:
> 
> Frank (cced above) sent me email about ECN
> implementation in ns. Because I 
> am not familiar with that part of the code, I
> forward his email to you. 
> Thanks.
> 
> Cheers,
> -chen
> 
> -- 
> Xuan Chen
> USC/ISI
> 
> ---------- Forwarded message ----------
> Date: Thu, 28 Feb 2002 21:21:36 +0000 (GMT)
> From: Frank Akujobi 
> To: Xuan Chen 
> Cc: ns-users@ISI.EDU
> Subject: Re: Bug in NS-2.1b8a for ECN with "gentle_"
> ???
> 
> Hi,
> Thanks for your prompt response. The problem is
> thus:
> Within the region of maxth_ < avg_ < 2*maxth_ and
> with
> "gentle_" parameter enabled, the drop probability is
> supposed to vary btw. 10% and 100%. When a drop
> occurs
> under this situation it is supposed to be a FORCED
> drop. 
> In the codes however (function drop_early() in
> red.cc), we see that for ECN the marking
> probability(which should be the drop prob. for RED)
> is
> used only within the region of minth_ < avg_ <
> maxth_.
> If avg_ happens to be above maxth_ but less than
> 2*maxth_ packets are dropped with a probability of
> 100% (as UNFORCED drop) NOT a range btw. 10% and
> 100%
> as required. This is the problem I see. I am hoping
> that you can confirm this or probably show me what I
> don't understand. 
> 
> Hope this helps and looking forward to hearing from
> you.  
> 
> Frank Akujobi.
> 
> 
> 
> 
> 
>  --- Xuan Chen  wrote: > I will try
> to
> find a right person for you
> > (hopefully..:)). but, in 
> > the mean time, could you  please describe the
> > problem (original code, your 
> > fix, and the simulation results) in details for
> > ns-users?
> > 
> > Thanks,
> > -chen
> > 
> > On Thu, 28 Feb 2002, Frank Akujobi wrote:
> > 
> > > Hi,
> > > I am working on some form of ECN and I observe
> > > difference in results when I provide a fix. I
> had
> > > thought there was a reason for writing the codes
> > like
> > > that, otherwise it could be a bug.
> > > If U are not familiar with the area, pls can you
> > refer
> > > to someone amongst your NS "administrators" who
> > might
> > > be able to clear up any doubts we have.
> > > 
> > > Thanks and hoping to hear from U.
> > > Frank.
> > > 
> > > 
> > > 
> > >  --- Xuan Chen  wrote: > I am
> sorry
> > > that I am not familiar with that part of
> > > > the code. Have you got 
> > > > any difference in simulation results with the
> > > > original implementation and 
> > > > your fix?
> > > > 
> > > > -chen
> > > > 
> > > >  On Wed, 27 Feb 2002, Frank Akujobi wrote:
> > > > 
> > > > > Hi,
> > > > > I had posted this on the mailing list 2 days
> > ago
> > > > but
> > > > > till now there's been no reponse. Pls could
> > you
> > > > find
> > > > > time to consider my comments......
> > > > > 
> > > > > With ECN and "gentle_" parameter set, packet
> > > > marking
> > > > > prob. should vary 
> > > > > betw. 10% 
> > > > > and 100% when  maxth_ < avg_ < 2*maxth_. In
> > the
> > > > > NS-2.1b8a, I find the 
> > > > > following 
> > > > > codes in the function drop_early() within
> > red.cc.
> > > > > 
> > > > > if (edp_.setbit && hf->ect() && edv_.v_ave <
> > > > > edp_.th_max) {
> > > > > 			hf->ce() = 1; 	// mark Congestion
> > Experienced
> > > > bit
> > > > > 			return (0);	// no drop
> > > > > 		} else {
> > > > > 			return (1);	// drop
> > > > > 		}
> > > > > 
> > > > > This code follows the computation of marking
> > > > > probability for ECN. If 
> > > > > the avg_ is 
> > > > > between maxth_ and 2*maxth_ and gentle_ is
> > > > enabled,
> > > > > the codes above 
> > > > > would always 
> > > > > cause the arriving packet to experience
> > UNFORCED
> > > > drop
> > > > > irrespective of 
> > > > > the 
> > > > > computed probabilty.
> > > > > Is this a bug or is there something I am
> > missing
> > > > out.
> > > > > 
> > > > > Thanks
> > > > > Frank.
> > > > > 
> > > > > 
> > > > > =====
> > > > > Frank Onyekachi Akujobi
> > > > > Broadband Networks Laboratory 
> > > > > Systems and Computer Engineering
> > > > > Carleton University
> > > > > 1125 Colonel By Drive, Ottawa, Ontario
> > > > > K1S 5B6, Canada
> > > > > Cell:(613)286-3738
> > > > > 
> > > > >
> > __________________________________________________
> > > > > Do You Yahoo!?
> > > > > Everything you'll ever need on one web page
> > > > > from News and Sport to Email and Music
> Charts
> > > > > http://uk.my.yahoo.com
> > > > > 
> > > > 
> > > > -- 
> > > > Xuan Chen
> > > > USC/ISI
> > > >  
> > > 
> > > =====
> > > Frank Onyekachi Akujobi
> > > Broadband Networks Laboratory 
> > > Systems and Computer Engineering
> > > Carleton University
> > > 1125 Colonel By Drive, Ottawa, Ontario
> > > K1S 5B6, Canada
> > > Cell:(613)286-3738
> > > 
> > >
> __________________________________________________
> > > Do You Yahoo!?
> > > Everything you'll ever need on one web page
> > > from News and Sport to Email and Music Charts
> > > http://uk.my.yahoo.com
> > > 
> > 
> > -- 
> > Xuan Chen
> > USC/ISI
> >  
> 
> =====
> Frank Onyekachi Akujobi
> Broadband Networks Laboratory 
> Systems and Computer Engineering
> Carleton University
> 
=== message truncated === 

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com


From HE Jun" 
Message-ID: <002b01c1c4b0$9049d310$c7518489@stu.nus.edu.sg>

This is a multi-part message in MIME format.

------=_NextPart_000_0028_01C1C4F3.9E425990
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

hi:

I found some bugs in god.cc.=20

the problem is that mb_node[i]->energy_model() may return NULL pointer =
which will cause segment error. Before calling node_on() and energy(), =
the return vaule mb_node[i]->energy_model() should be checked against =
NULL.

hejun

bool God::IsNeighbor(int i, int j)
{
  assert(ienergy_model()->node_on() =3D=3D false ||
      mb_node[j]->energy_model()->node_on() =3D=3D false ||
      mb_node[i]->energy_model()->energy() <=3D 0.0 ||
      mb_node[j]->energy_model()->energy() <=3D 0.0 ) {
    return false;
  }


void God::CountAliveNode()
{
  int i;

  num_alive_node =3D 0;

  for (i=3D0; ienergy_model()->energy() > 0.0) {
      num_alive_node++;
    }
  }

}

------=_NextPart_000_0028_01C1C4F3.9E425990
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable









hi:


 


I found some bugs in god.cc. =



 


the problem is that =
mb_node[i]->energy_model()=20
may return NULL pointer which will cause segment error. Before calling =
node_on()=20
and energy(), the return vaule mb_node[i]->energy_model()=20
should be checked against NULL.


 


hejun


 


bool God::IsNeighbor(int i, int =
j)
{
 =20
assert(i<num_nodes && j<num_nodes);


 


  //printf("i=3D%d, j=3D%d\n", =
i,j);
 =20
if (mb_node[i]->energy_model()->node_on() =3D=3D false=20
||
      =
mb_node[j]->energy_model()->node_on()=20
=3D=3D false ||
     =20
mb_node[i]->energy_model()->energy() <=3D 0.0=20
||
      =
mb_node[j]->energy_model()->energy()=20
<=3D 0.0 ) {
    return =
false;
 =20
}



void=20
God::CountAliveNode()
{
  int i;


 


  num_alive_node =3D =
0;


 


  for (i=3D0; i<num_nodes; i++) =

{
    if=20
(mb_node[i]->energy_model()->energy() > 0.0)=20
{
     =20
num_alive_node++;
    }
  }


 


}


------=_NextPart_000_0028_01C1C4F3.9E425990--


From xuanc@ISI.EDU  Wed Mar  6 05:15:02 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Wed Mar  6 05:15:02 2002
Subject: [ns] More Diffserv Examples.
In-Reply-To: <00eb01c1c4be$07656380$4a0e73ca@uestc>
Message-ID: 

do something like:

$ns simplex-link $core1 $core2 5Mb 5ms dsRED/core
$ns simplex-link $core2 $core1 5Mb 5ms dsRED/core

Sorry about the confusion (I should always copy something standard from 
the scripts rather than creating terminologis by myself...:))

-chen

On Wed, 6 Mar 2002, Xiang wrote:

> Hi,
> What do you mean of "config the queue between two core routers as Core/Edge
> Queue"?
> Can you exlain it to me?
> Thanks
> Xiang
> ----- Original Message -----
> From: "Xuan Chen" 
> To: "Karina Karla C. de Oliveira" 
> Cc: 
> Sent: Saturday, March 02, 2002 1:33 AM
> Subject: Re: [ns] More Diffserv Examples.
> 
> 
> >
> > You can basically config the queue between two core routers as Core/Edge
> > Queue. Other configs should be similar to the sample scripts.
> >
> > Hope it helps,
> > -chen
> >
> > On Fri, 1 Mar 2002, Karina Karla C. de Oliveira wrote:
> >
> > >
> > >
> > > Hi,
> > >
> > > I would like to know if someone has some example of a network diffserv
> > > with more than one core router. In the ns examples, in
> > > ns/tcl/ex/diffserv, only a configuration is used with only one core
> > > router.
> > >
> > > My network configuration is:
> > >
> > > edge_router1 ------ core_router1 -------- core_router2 ------- edge_
> > > router2
> > >
> > > How will be the configuration for this case?
> > >
> > > Best Regards,
> > > Karina Oliveira.
> > > *********************************************************
> > >         Karina Karla Cavalcante de Oliveira
> > > Computer Science
> > >         Universidade Federal de Pernambuco
> > >         Centro de Informática - CIn
> > >         Caixa Postal 7851
> > >         CEP: 50732-970
> > >         Recife-PE - Brasil
> > >         Phone: +55 (081) 3271-8430
> > >
> > >         Email: kkco@cin.ufpe.br
> > > WWW: www.cin.ufpe.br/~kkco
> > > Phones: (HOME) +55 (081) 3338-2054
> > > *********************************************************
> > >
> >
> > --
> > Xuan Chen
> > USC/ISI
> >
> >
> 
> 

-- 
Xuan Chen
USC/ISI


From wido   Wed Mar  6 05:15:19 2002
From: wido  (wido)
Date: Wed Mar  6 05:15:19 2002
Subject: [ns] (03/06/2002) Trace graph 1.29b (NS trace files analyser), www.geocities.com/tracegraph/
Message-ID: <101828749.20020306084312@o2.pl>

Hello,
Trace graph 1.29b (NS trace files analyser) executable and C source 
is available to download on:
http://www.geocities.com/tracegraph/
or ftp://156.17.232.145/pub/trgraph/.
My ftp is down sometimes and I shut it down for nights (CET).
Trace graph shows simulation and nodes information, plots 2D and 3D graphs,
histograms, saves results and has another options. 
There are some screenshots from the program on the web site or the ftp site.
Trace graph runs under Windows and Linux. The source can be compiled under most of UNIX systems.

Best regards
Jaroslaw Malek


From francesc@lsihp19.epfl.ch  Wed Mar  6 05:15:29 2002
From: francesc@lsihp19.epfl.ch (Francesc Font (EPFL))
Date: Wed Mar  6 05:15:29 2002
Subject: [ns] Traceable variables
References: 
Message-ID: <3C85CF2E.3DF2DB44@lsihp19.epfl.ch>

Hi Steve,
    In ~ns/tcl/lib/ns-default.tcl you have all the variables that may be
accessed by otcl, but I don't know if all of them are traceable...


Steve wrote:

> Hi,
>
> for my research I need to be able to trace certain variables from ns.
> Could anyone please give me a list, or tell me where there is a complete
> list (can't find one in the manual) of all variables in ns packets and
> nodes that it is possible to easily trace. This would really help me,
>
> thanks,
>
> Steve


From Marcel.Odena@icn.siemens.de  Wed Mar  6 05:15:43 2002
From: Marcel.Odena@icn.siemens.de (Odena Marcel)
Date: Wed Mar  6 05:15:43 2002
Subject: AW: [ns] AW: ARP  --> Marcel Odena's version of ARP code. --> ANS
 WERS and ARP SHORCUT VERSION.
Message-ID: <353063C297A9D3118F760008C791E269034C9CCA@MCHH263E>

Hello Marc,
Answering to your questions:

1) Why have you disabled the counter of requests?
Well I disable it because I did not see the utility. In arp.cc, in function 
int ARPTable::arpsolve, when a new packet arrives the function looks whether
there is already an entry for that destination. Supose that there is already an 
entry and llinfo->count_ is 'ARP_MAX_REQUEST_COUNT'. What the code does then is
to set count_ and hold_ to zero and drop the packet. I do not really see the 
utility of this.


2)Why do you discard the incoming packet if the buffer is full, instead of the 
first packet?
This is a matter of your decision. You can choose the policy of queueing and 
dequeuing the packets. However this is not a problem in the simulations I have 
ran because the queue is set to 1000, and there is no drop at the ARP queue.
But of course, you don't have any drops in the ARP queues but you observe 
some packets with long delays.

In terms of avoiding long delays, you are right, that when a queue is full is
better to delete an old packet that the "new" incoming packet.


A NOTE: using ARP the simulations may be more realistic, but to study the performance
of routing protocols it may difficult their study, because there are some effects
that are not due to the routing protocols but the ARP code. So finally I decided to 
use another version of ARP that does the resolution immediately, without sending 
REQUESTS and waiting REPLIES. This is possible because in ns-2 the addresses of the 
nodes are the same as the MAC addresses. To implement this behavior you only need to
modify the function 'int ARPTable:: arpsolve()' in the original 'arp.cc' (the one you get from ns-2 software)
like this:

=================== ARP shortcut version ======================================================

int
ARPTable::arpresolve(nsaddr_t dst, Packet *p, LL *ll)
{
        ARPEntry *llinfo ;
	
	assert(initialized());
	llinfo = arplookup(dst);

#ifdef DEBUG
        fprintf(stderr, "%d - %s\n", node_->address(), __FUNCTION__);
#endif
	
//	if(llinfo && llinfo->up_) {
//		mac_->hdr_dst((char*) HDR_MAC(p), llinfo->macaddr_);
//		return 0;
//	}
        //Marcel Odena, 4 March 2002
        //Shorcut version of ARP.
        hdr_cmn *ch = HDR_CMN(p);
        if(1){
           mac_->hdr_dst((char*) HDR_MAC(p), ch->next_hop());
           return 0;
        }       



	if(llinfo == 0) {
		/*
		 *  Create a new ARP entry
		 */
		llinfo = new ARPEntry(&arphead_, dst);
	}

        if(llinfo->count_ >= ARP_MAX_REQUEST_COUNT) {
                /*
                 * Because there is not necessarily a scheduled event between
                 * this callback and the point where the callback can return
                 * to this point in the code, the order of operations is very
                 * important here so that we don't get into an infinite loop.
                 *                                      - josh
                 */
                Packet *t = llinfo->hold_;

                llinfo->count_ = 0;
                llinfo->hold_ = 0;
		hdr_cmn* ch;
		
                if(t) {
                        ch = HDR_CMN(t);

                        if (ch->xmit_failure_) {
                                ch->xmit_reason_ = 0;
                                ch->xmit_failure_(t, ch->xmit_failure_data_);
                        }
                        else {
                                drop(t, DROP_IFQ_ARP_FULL);
                        }
                }

                ch = HDR_CMN(p);

		if (ch->xmit_failure_) {
                        ch->xmit_reason_ = 0;
                        ch->xmit_failure_(p, ch->xmit_failure_data_);
                }
                else {
                        drop(p, DROP_IFQ_ARP_FULL);
                }

                return EADDRNOTAVAIL;
        }

	llinfo->count_++;
	if(llinfo->hold_)
		drop(llinfo->hold_, DROP_IFQ_ARP_FULL);
	llinfo->hold_ = p;

	/*
	 *  We don't have a MAC address for this node.  Send an ARP Request.
	 *
	 *  XXX: Do I need to worry about the case where I keep ARPing
	 *	 for the SAME destination.
	 */
	int src = node_->address(); // this host's IP addr
	arprequest(src, dst, ll);
	return EADDRNOTAVAIL;
}

================================== END ============================================================


Well I hope this contributes a little bit more in tha ARP problems. 

Ciao.

MARCEL ODENA.



-----Ursprüngliche Nachricht-----
Von: Marc Torrent Moreno [mailto:marc.torrent@ccrle.nec.de]
Gesendet: Dienstag, 5. März 2002 15:13
An: Odena Marcel
Betreff: Re: [ns] AW: ARP --> Marcel Odena's version of ARP code.


Hi Marcel,

I took a look on your modifications of the ARP code.
I'm quite interested in adding this feature in my simulations, but I don't 
understand a couple of things.
Why have you disabled the counter of requests?
Why do you discard the incoming packet if the buffer is full, instead of the 
first packet?

Thanks a lot,

marc

On Monday 04 March 2002 09:33, you wrote:
> Hi Ahmed,
> Here you have the code of my version of ARP code. I am using the
> ns-2.1b8a, the allinone version.
>
> I already sent an e-mail where I gave some notes about my version of
> ARP, but I have seen that it's not yet posted in the mailing list archive
> (http://www.isi.edu/nsnam/archive/ns-users/webarch/).
>
> Some important details about my version of ARP: it implements a queue to
> keep packets waiting for an address resolution which you can set with the
> variable 'ARPQUEUE_MAX_LENGTH'. Some improvements that could be implemented
> is to set a time out for the packets waiting for an address resolution
> avoiding long delays.
>
> I did not took into account to delete the packets remaining at the queue at
> the end of simulation.
>
> Robin Poss also implemented his versions of ARP. You can see it at:
> (http://www.geocities.com/robin_poss/zrp.html), descompress the file in it
> and you will see it. There is other interesting information too.
>
>
> -----Ursprüngliche Nachricht-----
> Von: Ahmed Moustafa [mailto:ahmed.moustafa@sympatico.ca]
> Gesendet: Sonntag, 3. März 2002 18:40
> An: Marcel.Odena@icn.siemens.de
> Betreff: ARP
>
>
> Hi Marcel,
>
> Is it possible to get copy of your ARP code, and would you please tell
> me which verion of ns you are using.
>
> Thanks


From dbelen@netmode.ece.ntua.gr  Wed Mar  6 05:15:54 2002
From: dbelen@netmode.ece.ntua.gr (Dimitris Velenis)
Date: Wed Mar  6 05:15:54 2002
Subject: [ns] Urgent help with TCP Split
In-Reply-To: ; from renan_cunha@hotmail.com on Mon, Mar 04, 2002 at 10:01:56PM -0300
References: 
Message-ID: <20020306115729.A37048@netmode.ece.ntua.gr>

Hi Renan,

The code in http://mailman.isi.edu/pipermail/ns-users/2001-November/019312.html
does not produce acks the way you describe. Based on it you can create two 
seperate TCP connections (each with its own semantics). Their only relation 
will be that the second forwards the data that have been successfully transfered by
the first. In the example you provided, the SplitTcpSink on N2 acks a segment
as soon as it receives it, it does not wait for N3 to ack it first. That i think,
as pointed out in another reply, would not be much of connection splitting...

The choice of DelAckSink as a base class for SplitTcpSink affects only the frequency
of the acks. Changing to TcpSink will not change the way connections are split.

I cannot say why you're getting 0-sized acks (in my simulations I don't). At any
given time during your simulations, the TcpAgent on N1 is expected to have received more
acks than the forwarder agent on N2. That is because seqno's ack'ed by N2 (to N1) may
still be on their way to N3.

Dimitris.

On Mon, Mar 04, 2002 at 10:01:56PM -0300, Renan Cunha wrote:
> 
> 
>    Hi all,
> 
>     I am facing a problem with TCP Split and I need to solve it urgently.
>     I will try to explain my problem with an example.  My first objective 
> was creating an Agent Forwarder which is independent of the transport layer, 
> so its works with UDP and TCP. With UDP, the funcionality was achieved. But 
> with TCP, I got a problem with the ACK.  Suppose that:
> 
>    * Node N1 has an Application/[FTP/Telnet| Exponential/CBR ] agent, for 
> instance.
> 
>    * Node N2 has an Agent Fowarder and an Application, used to generate 
> traffic.
> 
>    * Node N3 has a TCPSink agent, in the case of TCP applications, and Null 
> agent for UDP
> 
> 
>      N1                         N2			     N3
> 
>    [Telnet|FTP]/TCP <--->SplitTcpSink->App_Forward/TCP<--->TCPSink
>   			  ^                       |
>   			  |_______________________|
> 
> 
>   There are two applications in this topology, one at N1, which generates 
> the packets, and one at N2, which generate the packets at App_Forward as 
> they arrive at this node. So at N3, the ack is sent back to N2. At this 
> time, TCP calls SplitTcpSink, which generates the ack for N1.
> 
>   The tcpsplit code I wrote is almost equal to the one posted by
> Dimitris in
> 
> http://mailman.isi.edu/pipermail/ns-users/2001-November/019312.html
> 
>   The difference is that, in my implementation, the class SplitTcpSink 
> inherits directly from TCPSink, and not from public virtual DelAckSink.
>   When a packet arrive at SplitTcpSink, it call the App_Forward, which calls 
> the procedure sendmsg, of the transport agent, passing as parameter the size 
> of the packet which has arrived. When the ack arrives at TCP, it makes the 
> SplitTcpSink generate the ack to N1.
> 
>   My problem is that the ack has size 0, and the number of acks returned is 
> incorrect.
> 
>   What must be done with the ack packets that are sent back to the N1 agent? 
> What must be their behaviour (which path should they follow)?
> 
>   Why DelAckSink is used instead of TCPSink? Is there a problem with pure 
> TCPSink?
> 
>   Can anyone give me any advice or tell me where I can find the code of an 
> implementation of TCP Split?
> 
>    Thanks in advance,
> 
>    Renan Cunha
> 
> _________________________________________________________________
> Join the world’s largest e-mail service with MSN Hotmail. 
> http://www.hotmail.com

-- 
Dimitris Velenis                              
Network Management & Optimal Design Lab
National Technical University of Athens


From flaw@vt.edu  Wed Mar  6 05:16:05 2002
From: flaw@vt.edu (Raymond Law)
Date: Wed Mar  6 05:16:05 2002
Subject: [ns] MPLS question
Message-ID: <3C8646F6@zathras>

I was looking at an example script using MPLS and have a question.  What does 
the following do and why is it needed?

for {set i 2} {$i < 9} {incr i} {
	set a LSR$i
	set m [eval $$a get-module "MPLS"]
	eval set LSRmpls$i $m
}

Also, why not just use $LSR7 instead of $LSRmpls7, where "set LSR7 [$ns 
mpls-node]" is defined before?

$ns at 0.2 "$LSRmpls7 send-ldp-withdraw-msg  9"

Thanks.
Ray,



From toubix@kom.auc.dk  Wed Mar  6 05:16:18 2002
From: toubix@kom.auc.dk (Thibault RENIER)
Date: Wed Mar  6 05:16:18 2002
Subject: [ns] Re: patches for ns-2.1b8a
References: 
Message-ID: <003301c1c501$17f29ea0$d304080a@toubix>

Hi,

as you told me, I have installed ns2.1b7a to install bluehoc on it. My
platform is a linux. But i face a problem with bluehoc installation. In the
FAQ, step d, ( or in the install file, step 4) there is a note which says :
 In this case you need to manually add the
BLUEHOC C++ files for compilation in OBJ_CC
and ns-btnode.tcl in the NS_TCL_LIB macros
of the Makefile.
I don't know how to do this, could you please help me?
Thank you in advance for your help

Regards
Thibault RENIER


----- Original Message -----
From: "Abhinandan Sharma" 
To: "Thibault RENIER" 
Cc: "Kaushik.R" ; 
Sent: Monday, March 04, 2002 5:05 PM
Subject: Re: patches for ns-2.1b8a


> hi thibault
> basically the patch files provided with bluehoc are written keeping in
> mind the ns-2.1b7a directory structure. hence you need to modify the patch
> files as per the ns-2.1b8a dir structure now. this will be easy if there
> are not much changes in ns-2.1b8a. however a neat idea would be to degrade
> to 7a version if it isn't bothering you a lot.
> hope this will help you.
> bye
> abhi
>
> --
> I can picture in my mind a world without war, a world without hate.
> And I can picture us attacking that world, because they'd never expect it.
>
>
> On Mon, 4 Mar 2002, Thibault RENIER wrote:
>
> > Hello,
> >
> > we are installing bluehoc on ns2.1b8a. We acheived the first steps
adviced in the bluehoc FAQ : How to install bluehoc.
> > But we had problem with the application of patches to the ns files which
have been modified for bluehoc (step d in the installation process).
> >
> > We typed :
> >     patch -p0 -b < bt_patch
> >
> > We obtained :
> > can't find file to patch at input line 3
> > Perhaps you used the wrong -p or --strip option?
> > The text leading up to this was:
> > --------------------------
> > |*** ns-2.1b7a/god.cc    Thu Oct 19 01:40:23 2000
> > |--- god.cc    Sun Feb 25 04:15:39 2001
> > --------------------------
> > File to patch:
> >
> > Could you please help us to know how we can install bluehoc on ns-2.1b8a
? Do we actually need to install the patches provided to ns version 2.1b8a
??
> >
> > Thank you in advance
> >
> > Regards
> > Thibault RENIER
> >
> >
> >
> >
>
>
>
>



From gdemarco@unisa.it  Wed Mar  6 05:16:29 2002
From: gdemarco@unisa.it (giuseppe de marco)
Date: Wed Mar  6 05:16:29 2002
Subject: [ns] debug_ variable?
Message-ID: <20020306130841.662852ed.gdemarco@unisa.it>

Who know the reason of this problem?

ns: _o16 start: can't read "debug_": no such variable
    (Object set line 1)
    invoked from within
"Classifier/Hash set debug_"
    invoked from within
"catch "$c set $var" val"
    (procedure "_o16" line 1)
    (SplitObject unknown line 1)
    invoked from within
"_o16 start"                       

-- 
Giuseppe De Marco
Department of Electrical and Information Engineering (DIIIE)
University of Salerno
tel . +39 089 964012


From aam11@rediffmail.com  Wed Mar  6 05:16:40 2002
From: aam11@rediffmail.com (Amuthamozhi)
Date: Wed Mar  6 05:16:40 2002
Subject: [ns] [bug] Mac problem in the wired-cum-wireless-sim.tcl
Message-ID: <200203061210.MAA17766@www.isi.edu>

[Bug Report]

-----------------------------
Category:  Run Time - Segmentation Fault
Package:   ns 2.1b8
OS:        Linux redHat release 7.1
Environment Variables:
LD_LIBRARY_PATH=
TCL_LIBRARY=
TK_LIBRARY=


-----------------------------
Description:

  As I run the wired-cum-wireless-sim.tcl program the mac access
out of range error occurs.I have already changed the Mac-803-11.cc
and tcl script programs by seeing the mailing list.Has anyone
please fixed my bug problem?


From pangbo@sina.com  Wed Mar  6 05:16:56 2002
From: pangbo@sina.com (Poby Pang)
Date: Wed Mar  6 05:16:56 2002
Subject: [ns] [bug] Interesting Bug in mobilenode.cc
Message-ID: <200203061300.NAA22791@www.isi.edu>

[Bug Report]

-----------------------------
Category:  Other
Package:   ns 2.1b8a
OS:        Linux 2.4.7-10 (RedHat 7.2)
Environment Variables:
LD_LIBRARY_PATH=
TCL_LIBRARY=
TK_LIBRARY=


-----------------------------
Description:

Description of Problem:
in ~ns/mobilenode.cc, there is a code bug, but it does not generate errors when running.

in mobilenode.cc:
  ... ...
  } else if(argc == 3) {
    if(strcmp(argv[1], "addif") == 0) {
      WiredPhy* phyp = (WiredPhy*)TclObject::lookup(argv[2]);
      if(phyp == 0)
        return TCL_ERROR;
      phyp->insertnode(&ifhead_);
      phyp->setnode(this);
      return TCL_OK;
    } else if (strcmp(argv[1], "setsleeptime") == 0) {
    ... ... 
    } else if(strcmp(argv[1], "addif") == 0) {
      WirelessPhy *n = (WirelessPhy*)TclObject::lookup(argv[2]);
      if(n == 0)
        return TCL_ERROR;
      n->insertnode(&ifhead_);
      n->setnode(this);
      return TCL_OK;
    } ... ...

I think it should be changed to:
  ... ...
  } else if(argc == 3) {
    if(strcmp(argv[1], "addif") == 0) {
      WirelessPhy* phyp = (WirelessPhy*)TclObject::lookup(argv[2]);
      if(phyp == 0)
        return TCL_ERROR;
      phyp->insertnode(&ifhead_);
      phyp->setnode(this);
      return TCL_OK;
    } else if (strcmp(argv[1], "setsleeptime") == 0) {
    ... ... 
 
This bug does not generate error because insertnode() and setnode() both are the methods of base class Phy.


How Easily Reproducible:
(e.g. every time, intermittent, once only, etc.)


Steps to Reproduce:
(describe the minimal set of steps necessary to trigger the bug)
1. 
2. 
3. 


Actual Results:
(describe what the application did after performing the above steps)


Expected Results:
(describe what the application should have done, were the bug not present)


Additional Information:
(the following infomation is helpful to debug:
 1. simulation script, detailed output files, packet trace
 2. patch file if you modify some source code
 3. a backtrace from gdb if you get a segment fault
 If they are big files, PLEASE put them in your web space and
 include the URL here.)



From sbchen@iipc.zju.edu.cn  Wed Mar  6 05:45:08 2002
From: sbchen@iipc.zju.edu.cn (=?ISO-8859-1?Q?=B3=C2=C9=D0=B1=F8?=)
Date: Wed Mar  6 05:45:08 2002
Subject: [ns] About the Maximum Window Size of TCP
Message-ID: <200203061349.VAA00948@iipc.zju.edu.cn>

Hi, everyone,

In ns-doc, page 97 about the TCP object, it says: 

	MWS The Maximum Window Size in packets for a TCP connection. MWS determines the size of an array in tcpsink.cc. The default for MWS is 1024 packets. For Tahoe TCP, the "window" parameter, representing the receiver's advertised window, should be less than MWS-1. For Reno TCP, the window" parameter should be less than (MWS-1)/2.

I can not understand the expression of 'MWS-1' and '(MWS-1)/2'. Can you give me a detailed interpretation? 

Thank you in advance.

sbchen@iipc.zju.edu.cn


From tasdemir@ata.cs.hun.edu.tr  Wed Mar  6 06:35:01 2002
From: tasdemir@ata.cs.hun.edu.tr (Mustafa Tasdemir)
Date: Wed Mar  6 06:35:01 2002
Subject: [ns] What is the difference between target_->recv(p) & send(p,0)?
Message-ID: <001401c1c51c$660e64a0$1fec8cc1@cs.hacettepe.edu.tr>

This is a multi-part message in MIME format.

------=_NextPart_000_0011_01C1C52D.29805140
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

hello,

can someone tell what is the difference between these commands when used =
in agent:

Application---->UDP------->Node
                           |
                           |
                        commands are used in UDP agents
  1.. target_->recv(p);    // target defined as Process *target_;=20
  2.. send(p,0);
I think, the first one is a direct call and the second one is an =
indirect call. Can we also call the send() method as a SAP (Service =
Access Point)?

Mustafa TASDEMIR




------=_NextPart_000_0011_01C1C52D.29805140
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable









hello,


 


can someone tell what is the difference =
between=20
these commands when used in agent:


 


Application---->UDP------->Node

          &nbs=
p;            =
;    |


          &nbs=
p;            =
;  =20
 |


    =
   =20
            =
   =20
commands are used in UDP agents


    
  
  1. target_->recv(p);    //=20
  target defined as Process *target_; =

    2. 
  
  2. send(p,0);


I think, the first one is a direct call =
and the=20
second one is an indirect call. Can we also call the =
send()=20
method as a SAP (Service Access=20
Point)?


 


Mustafa TASDEMIR


 


 


 


------=_NextPart_000_0011_01C1C52D.29805140--


From sbchen@iipc.zju.edu.cn  Wed Mar  6 07:25:01 2002
From: sbchen@iipc.zju.edu.cn (=?ISO-8859-1?Q?=B3=C2=C9=D0=B1=F8?=)
Date: Wed Mar  6 07:25:01 2002
Subject: [ns] Why there is no packet drop in my RED simulation?
Message-ID: <200203061537.XAA02205@iipc.zju.edu.cn>

Hi,ns-users,

I am doing a simple RED simulation as below:

	set ns [new Simulator]

	# Define core nodes and other nodes
	set nc0 [$ns node]
	set nc1 [$ns node]
	set i 1
	while {$i <= 5} {
		set src($i) [$ns node]
		set des($i) [$ns node]
		incr i
	}

	Queue/DropTail set drop_front_ 1
	Queue set limit_ 100

	# Define the bottleneck link 
	$ns simplex-link $nc0 $nc1 10Mb 10ms RED
	$ns simplex-link $nc1 $nc0 1000Mb 10ms DropTail

	# RED Parameters
	set link1 [$ns get-link $nc0 $nc1]
	set qlink1 [$link1 queue]
	$qlink1 set limit_ 150
	$qlink1 set thresh_ 75
	$qlink1 set maxthresh_ 150
	$qlink1 set mean_pktsize_ 1000 
	$qlink1 set drop_front_ true


	#Attach the access nodes with the core nodes
	$ns duplex-link $src(1) $nc0 30Mb 25ms DropTail
	$ns duplex-link $des(1) $nc1 30Mb 25ms DropTail
	$ns duplex-link $src(2) $nc0 30Mb 30ms DropTail
	$ns duplex-link $des(2) $nc1 30Mb 30ms DropTail
	...
	$ns duplex-link $src(5) $nc0 30Mb 45ms DropTail
	$ns duplex-link $des(5) $nc1 30Mb 45ms DropTail

	set i 1
	while {$i <= 5} {
		set p($i) [new Agent/TCP/Reno]
		$ns attach-agent $src($i) $p($i)
		$p($i) set fid_ [expr $i + 100]
		set q($i) [new Agent/TCPSink]
		$ns attach-agent $des($i) $q($i)
		$ns connect $p($i) $q($i)
		set f($i) [new Application/FTP]
		$f($i) attach-agent $p($i)
		incr i
	}

	set i 1
	while {$i <= 5} {
		set starttime [expr $i * 0.1]
		$ns at $starttime "$f($i) start"
		incr i
	}

	$ns at 20.0 "finish"
	$ns run

I think there has been enough traffic load, and the average queue length should be between 75 and 150.
But the graph result shows that the queue length is around 50, and there were no drops.

What's wrong with my simulation?

Thank you!

            sbchen@iipc.zju.edu.cn


From mkenny@mekb2.sps.mot.com  Wed Mar  6 07:55:01 2002
From: mkenny@mekb2.sps.mot.com (Martin McKenny)
Date: Wed Mar  6 07:55:01 2002
Subject: [ns] Routing packets in a link protocol
Message-ID: <3C863BAC.AE65C697@mekb2.sps.mot.com>

--------------D0DC042626DE199C6A01A0DD
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: base64

Hi,

I am attempting to describe a topology based upon a link protocol. That
is, once an interface (agent) acknowledges a packet, it takes full
responsibility for the next hop that packet must make. I have created an
agent that sends and acknowledges packets. I now wish to update the
routing function associated with the central node so that all packets
are sent to a particular intelligent node (all the time obeying the link
protocol that I have implemented). My question is: how do I access the
routing table  associated with the central node so that packets are
forwarded to a articular end-point that I can specify.

Thanks for any help,
MMK

--



--------------D0DC042626DE199C6A01A0DD
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit



Hi,

I am attempting to describe a topology based upon a link protocol. That
is, once an interface (agent) acknowledges a packet, it takes full responsibility
for the next hop that packet must make. I have created an agent that sends
and acknowledges packets. I now wish to update the routing function associated
with the central node so that all packets are sent to a particular intelligent
node (all the time obeying the link protocol that I have implemented).
My question is: how do I access the routing table  associated with
the central node so that packets are forwarded to a articular end-point
that I can specify.

Thanks for any help,

MMK

-- 


 

--------------D0DC042626DE199C6A01A0DD--


From armenb@static.net  Wed Mar  6 09:25:01 2002
From: armenb@static.net (Armen Babikyan)
Date: Wed Mar  6 09:25:01 2002
Subject: [ns] error running nam on MacOSX
Message-ID: 

Hi,

I've compiled ns and nam for MacOSX, however, I get the following error
when attempting to run nam:

  % nam out.nam
  nam: invalid command name "tcl_findLibrary"
  Trying to push call frame for dead namespace
  Abort
  %

Does anyone else have nam successfully running on MacOSX? I am
using tcl 8.3.4 and tk 8.3.4, for use with X11 on MacOSX (as opposed to
MacOSX carbon graphical environment).

Any and all information is appreciated. Thanks.

  - Armen



From haldar@ISI.EDU  Wed Mar  6 09:45:05 2002
From: haldar@ISI.EDU (Padmaparna Haldar)
Date: Wed Mar  6 09:45:05 2002
Subject: [ns] What is the difference between target_->recv(p) & send(p,0)?
In-Reply-To: <001401c1c51c$660e64a0$1fec8cc1@cs.hacettepe.edu.tr>
Message-ID: 


Most objects in ns are Connector objects having a target_ to SEND pkts to. 
It can also RECV pkts. recv and send functions in the Connector class are 
defined as follows:

void recv(Packet*, Handler* callback = 0);
inline void send(Packet* p, Handler* h) { target_->recv(p, h); }

so as you can see, send(p,h) and target_->recv(p) are basically doing the 
same thing.

--Padma

On Wed, 6 Mar 2002, Mustafa Tasdemir wrote:


> hello,
> 
> can someone tell what is the difference between these commands when used in agent:
> 
> Application---->UDP------->Node
>                            |
>                            |
>                         commands are used in UDP agents
>   1.. target_->recv(p);    // target defined as Process *target_; 
>   2.. send(p,0);
> I think, the first one is a direct call and the second one is an indirect call. Can we also call the send() method as a SAP (Service Access Point)?
> 
> Mustafa TASDEMIR
> 
> 
> 
> 

-- 
------------------------------------------------ 
Be true to your work, your word, and your friend. 
--Thoreau

Padmaparna Haldar


From andrea.palmieri@unile.it  Wed Mar  6 11:10:02 2002
From: andrea.palmieri@unile.it (Andrea Palmieri)
Date: Wed Mar  6 11:10:02 2002
Subject: [ns] rng problem
Message-ID: <000c01c1c542$00c7ebe0$694dccc1@unile.it>

hi all,
i'm trying to use rng.cc in the "raw" mode but my simulation attempts always
fail.
what's the reason?
i give an input seed which is less than MAXINT specified in that file.
can you hel me?

thanks in advance, andrea
_________________________________
Andrea Palmieri
University of Lecce
Department of Innovation Engineering
Via Monteroni
73100 LECCE - ITALY
Tel. +39 - 832 - 320222
Fax. +39 - 832 - 320279
e-mail: andrea.palmieri@unile.it


From dipti@ee.iitb.ac.in  Wed Mar  6 11:30:03 2002
From: dipti@ee.iitb.ac.in (Dipti Rani Taur _98d07001_)
Date: Wed Mar  6 11:30:03 2002
Subject: [ns] ns -help
Message-ID: 


Hello all,

I'm having trouble getting the Tracefile traffic source application to work in NS2.
The warning is:

warning: no class variable Tracefile::debug_
        see tcl-object.tcl in tclcl for info about this warning

 Does anyone know what may be the problem?  Any help
would be appreciated.i am using ns-2.1b8a

thanks in advance!!!
dipti




From cs152ct@cs.ucr.edu  Wed Mar  6 12:50:02 2002
From: cs152ct@cs.ucr.edu (cs152 Student)
Date: Wed Mar  6 12:50:02 2002
Subject: [ns] Help with Trace files from DEC
Message-ID: 


I am running a webserver simulation using NS and I need to use the
trace files I obtained from DEC.

Do anyone know if there is an automated way of converting these trace
files into the format needed by the NS? I have noticed the NS uses 2 files
(1 for request and 1 for page)  I found a website
http://www.research.att.com/~breslau/vint/trace.html but it not
explanatory.

I appreciate any assistance. Thank you.

Stephanie



From jmangan@tamu.edu  Wed Mar  6 13:35:03 2002
From: jmangan@tamu.edu (Trey Mangan)
Date: Wed Mar  6 13:35:03 2002
Subject: [ns] Help with Trace files from DEC
In-Reply-To: 
Message-ID: 

Stephanie,

I'm assuming that when you say trace files you are talking about using an
external trace file as a traffic source.  If so, this should help if you are
willing to recompile NS2.

It seems that most people have problems with this binary file format.  I've
never been able to get it to work right.  However, I found a modification to
the traffictrace.cc file on the web and made a few additional changes.  I
can't find the link now, so I will describe the changes here.  The source is
included below.  All the changes are in TraceFile::setup().  Note that I
also changed TraceFile::setup() to return 0.  This will start the file
playback at the beginning of the trace file.  If you want it to start in a
random place, then replace this line with the original line from
traffictrace.cc.  Once recompiled it will now accept ASCII test files where
the first column is  spacing between packets in 10^-6 seconds and the second
is packet size in bytes.  For instance if you are using UDP and you start
your trace file source at 2.0 seconds and the file looks like this:

0 60
100000 60

then two 60 byte packets will be sent, one at 2.0 seconds and the next at
2.1 seconds.

I hope this helps!

Regards,

John Mangan
jmangan@tamu.edu



/* -*-	Mode:C++; c-basic-offset:8; tab-width:8; indent-tabs-mode:t -*- */
/*
 * Copyright (c) Xerox Corporation 1997. All rights reserved.
 *
 * License is granted to copy, to use, and to make and to use derivative
 * works for research and evaluation purposes, provided that Xerox is
 * acknowledged in all documentation pertaining to any such copy or
derivative
 * work. Xerox grants no other licenses expressed or implied. The Xerox
trade
 * name should not be used in any advertising without its written
permission.
 *
 * XEROX CORPORATION MAKES NO REPRESENTATIONS CONCERNING EITHER THE
 * MERCHANTABILITY OF THIS SOFTWARE OR THE SUITABILITY OF THIS SOFTWARE
 * FOR ANY PARTICULAR PURPOSE.  The software is provided "as is" without
 * express or implied warranty of any kind.
 *
 * These notices must be retained in any copies of any part of this
software.
 */

/* Some other modifications 2/2/02

#ifndef lint
static const char rcsid[] =
    "@(#) $Header: /nfs/jade/vint/CVSROOT/ns-2/traffictrace.cc,v 1.13
1999/11/13 23:44:35 heideman Exp $ (Xerox)";
#endif

/* XXX: have not dealt with errors.  e.g., if something fails during
 * TraceFile::setup(), or TrafficTrace is not pointing to a TraceFile,
 * no guarantee about results.
 */

#include 
#include 
#include 

#include "config.h"
#ifdef HAVE_NETINET_IN_H
#include 
#endif /* HAVE_NETINET_IN_H */


#include "random.h"
/* module to implement a traffic generator in ns driven by a trace
 * file.  records in the trace file consist of 2 32 bit fields, the
 * first indicating the inter-packet time in microseconds, and the
 * second indicating the packet length in bytes.  multiple TraffficTrace
 * objects can use the same trace file, and different TrafficTrace
 * objects can use different trace files.  For each TrafficTrace, a
 * random starting point within the trace file is selected.
 */

#include "object.h"
#include "trafgen.h"

struct tracerec {
        u_int32_t trec_time; /* inter-packet time (usec) */
	u_int32_t trec_size; /* size of packet (bytes */
};


/* object to hold a single trace file */

class TraceFile : public NsObject {
 public:
	TraceFile();
	void get_next(int&, struct tracerec&); /* called by TrafficGenerator
						* to get next record in trace.
						*/
	int setup();  /* initialize the trace file */
	int command(int argc, const char*const* argv);
 private:
	void recv(Packet*, Handler*); /* must be defined for NsObject */
        int status_;
	char *name_;  /* name of the file in which the trace is stored */
	int nrec_;    /* number of records in the trace file */
	struct tracerec *trace_; /* array holding the trace */
};

/* instance of a traffic generator.  has a pointer to the TraceFile
 * object and implements the interval() function.
 */

class TrafficTrace : public TrafficGenerator {
 public:
	TrafficTrace();
	int command(int argc, const char*const* argv);
	virtual double next_interval(int &);
 protected:
	void timeout();
	TraceFile *tfile_;
	struct tracerec trec_;
	int ndx_;
	void init();
};


static class TraceFileClass : public TclClass {
 public:
	TraceFileClass() : TclClass("Tracefile") {}
	TclObject* create(int, const char*const*) {
		return (new TraceFile());
	}
} class_tracefile;

TraceFile::TraceFile() : status_(0)
{
}

int TraceFile::command(int argc, const char*const* argv)
{

	if (argc == 3) {
		if (strcmp(argv[1], "filename") == 0) {
			name_ = new char[strlen(argv[2])+1];
			strcpy(name_, argv[2]);
			return(TCL_OK);
		}
	}
	return (NsObject::command(argc, argv));
}

void TraceFile::get_next(int& ndx, struct tracerec& t)
{
	t.trec_time = trace_[ndx].trec_time;
	t.trec_size = trace_[ndx].trec_size;

	if (++ndx == nrec_)
		ndx = 0;

}

int TraceFile::setup()
{
   tracerec* t;
   struct stat buf;
   int i;
   FILE *fp;
   //ofstream of("readlog");

   /* only open/read the file once (could be shared by multiple
    * SourceModel's
    */
   if (! status_)
   {
      status_ = 1;

      if (stat(name_, (struct stat *)&buf))
      {
         printf("could not stat %s\n", name_);
         return -1;
      }
      nrec_ = buf.st_size/sizeof(tracerec);
      unsigned nrecplus = nrec_ * sizeof(tracerec);
      unsigned bufst = buf.st_size;
      trace_ = new struct tracerec[nrec_];

      if ((fp = fopen(name_, "rb")) == NULL)
      {
         printf("can't open file %s\n", name_);
         return -1;
      }

      t = trace_;
      char buf[4*1024];
      while (!feof(fp)  )
      {

         if (fgets(buf,4*1024,fp)==0)
            break;
         if (sscanf(buf,"%d %d",&t->trec_time,&t->trec_size)!=2)
         {
            printf("read failed\n");
            return -1 ;
         };
         //of<trec_time<<" "<trec_size<setup();
}

int TrafficTrace::command(int argc, const char*const* argv)
{
	Tcl& tcl = Tcl::instance();

	if (argc == 3) {
		if (strcmp(argv[1], "attach-tracefile") == 0) {
			tfile_ = (TraceFile *)TclObject::lookup(argv[2]);
			if (tfile_ == 0) {
				tcl.resultf("no such node %s", argv[2]);
				return(TCL_ERROR);
			}
			return(TCL_OK);
		}
	}

	return (TrafficGenerator::command(argc, argv));

}

void TrafficTrace::timeout()
{
        if (! running_)
                return;

        /* send a packet */
	// Note:  May need to set "NEW_BURST" flag in sendmsg() for
	// signifying a new talkspurt when using vat traces.
	// (see expoo.cc, tcl/ex/test-rcvr.tcl)
	agent_->sendmsg(size_);
        /* figure out when to send the next one */
        nextPkttime_ = next_interval(size_);
        /* schedule it */
        timer_.resched(nextPkttime_);
}


double TrafficTrace::next_interval(int& size)
{
        tfile_->get_next(ndx_, trec_);
	size = trec_.trec_size;
	return(((double)trec_.trec_time)/1000000.0); /* usecs->secs */
}




-----Original Message-----
From: ns-users-admin@ISI.EDU [mailto:ns-users-admin@ISI.EDU]On Behalf Of
cs152 Student
Sent: Wednesday, March 06, 2002 2:49 PM
To: ns-users@ISI.EDU
Subject: [ns] Help with Trace files from DEC





I am running a webserver simulation using NS and I need to use the
trace files I obtained from DEC.

Do anyone know if there is an automated way of converting these trace
files into the format needed by the NS? I have noticed the NS uses 2 files
(1 for request and 1 for page)  I found a website
http://www.research.att.com/~breslau/vint/trace.html but it not
explanatory.

I appreciate any assistance. Thank you.

Stephanie




From jw@erg.sri.com  Wed Mar  6 13:35:18 2002
From: jw@erg.sri.com (Julie Wong)
Date: Wed Mar  6 13:35:18 2002
Subject: [ns] bandwidth question for mobilenodes
Message-ID: <3C868B77.6D516A32@erg.sri.com>

Hello,

I have been trying to set the bandwidth of a mobile radio node to 11e6
without any success.  My wireless network uses interface
Phy/WirelessPhy.  Do I need to do anything else besides

    Phy/WirelessPhy set bandwidth_  11e6

in my .tcl file?

Thanks in advance,

Julie Wong

P.S.  I saw the default bandwidth in the ns-default.tcl file.
I even changed that to 11e6 but it did not seem to help.


From chakra10@msu.edu  Wed Mar  6 13:45:02 2002
From: chakra10@msu.edu (Gautam - Chakrabarti)
Date: Wed Mar  6 13:45:02 2002
Subject: [ns] bandwidth question for mobilenodes
References: <3C868B77.6D516A32@erg.sri.com>
Message-ID: <3C868D01.92B231BB@msu.edu>

You need to do

Mac/802_11 set bandwidth_ 11e6

--- this would change the bandwidth from the default value of 2, but if
I remember
right it cannot be increased above 10 or something like that. Try
increasing it from 2 to 4 or 5 ... that should work.

rgds,
gautam

Julie Wong wrote:
> 
> Hello,
> 
> I have been trying to set the bandwidth of a mobile radio node to 11e6
> without any success.  My wireless network uses interface
> Phy/WirelessPhy.  Do I need to do anything else besides
> 
>     Phy/WirelessPhy set bandwidth_  11e6
> 
> in my .tcl file?
> 
> Thanks in advance,
> 
> Julie Wong
> 
> P.S.  I saw the default bandwidth in the ns-default.tcl file.
> I even changed that to 11e6 but it did not seem to help.

-- 
***************************************************************
Gautam Chakrabarti
MS Student
Computer Sc. and Engg Department
Michigan State University

Email : chakra10@msu.edu
Home Page : http://www.cse.msu.edu/~chakra10

Why is there so much month left at the end of the money?
		--- John Barrymore
***************************************************************


From jw@erg.sri.com  Wed Mar  6 14:05:01 2002
From: jw@erg.sri.com (Julie Wong)
Date: Wed Mar  6 14:05:01 2002
Subject: [ns] bandwidth question for mobilenodes
References: <3C868B77.6D516A32@erg.sri.com> <3C868D01.92B231BB@msu.edu>
Message-ID: <3C869253.5C845895@erg.sri.com>

Thanks Gautam for coming to the rescue again.

I will try it out and verify it is indeed transmitting at 11mbs.

Sort of curious as to why I have to set it at the Mac level too.

Julie Wong

Gautam - Chakrabarti wrote:

> You need to do
>
> Mac/802_11 set bandwidth_ 11e6
>
> --- this would change the bandwidth from the default value of 2, but if
> I remember
> right it cannot be increased above 10 or something like that. Try
> increasing it from 2 to 4 or 5 ... that should work.
>
> rgds,
> gautam
>
> Julie Wong wrote:
> >
> > Hello,
> >
> > I have been trying to set the bandwidth of a mobile radio node to 11e6
> > without any success.  My wireless network uses interface
> > Phy/WirelessPhy.  Do I need to do anything else besides
> >
> >     Phy/WirelessPhy set bandwidth_  11e6
> >
> > in my .tcl file?
> >
> > Thanks in advance,
> >
> > Julie Wong
> >
> > P.S.  I saw the default bandwidth in the ns-default.tcl file.
> > I even changed that to 11e6 but it did not seem to help.
>
> --
> ***************************************************************
> Gautam Chakrabarti
> MS Student
> Computer Sc. and Engg Department
> Michigan State University
>
> Email : chakra10@msu.edu
> Home Page : http://www.cse.msu.edu/~chakra10
>
> Why is there so much month left at the end of the money?
>                 --- John Barrymore
> ***************************************************************


From mhammer@iagora.com  Wed Mar  6 15:45:02 2002
From: mhammer@iagora.com (mike hammer)
Date: Wed Mar  6 15:45:02 2002
Subject: [ns] RED + ECN problem
Message-ID: <1015458272.3c86a9e04f873@webmail.iagora.com>

Hello,

I'm using a RED queue with ECN. In order to track the 
packets marked in the nam, I change the flowid 
value of the ECN-marked packets in the queue in the 
red.cc file. I associate a different color with this 
flowid value. The packets that have the ECN bits set 
are displayed with the defined color but they don't 
dissapear with the queue. Has anybody out there had 
similar problems and maybe can help me?
Thanks in advance.
Cheers,
mike



_________________________________________________________________
iAgora -- Jobs, Information and Contacts for Young Internationals
http://www.iagora.com


From buchheim@ISI.EDU  Wed Mar  6 17:20:02 2002
From: buchheim@ISI.EDU (Tim Buchheim)
Date: Wed Mar  6 17:20:02 2002
Subject: [ns] Monitoring bytes
In-Reply-To: 
References: 
Message-ID: <1015463612.936.105.camel@ski.isi.edu>

On Fri, 2002-03-01 at 09:37, Steve wrote:

> can anyone help me in trying to monitor properties of a simulation.
> I have the following script which is very simple but when I run it,
> xgraph does not show the expectd results. It seems to be indicating that
> there are no bytes flowing into the monitor (tcp sink).
> I am trying to work out how to do this kind of thing in ns, but I am not
> really getting anywhere.

Your script is printing correct results.  Your Agent/LossMonitor object
($sink0) receives no packets because it is not connected to anything. 
The only packets sent in your simulation are received by $sink, not
$sink0.

Note that Agent/LossMonitor does not work with TCP.  The following
thread gives some ideas on modifying Agent/TCPSink to keep some stats.

http://mailman.isi.edu/pipermail/ns-users/2001-February/013735.html


-- 
Tim Buchheim


From sdarisa@yahoo.co.uk  Wed Mar  6 18:00:02 2002
From: sdarisa@yahoo.co.uk (=?iso-8859-1?q?Sandeep=20Darisala?=)
Date: Wed Mar  6 18:00:02 2002
Subject: [ns] MPLS question
In-Reply-To: <3C8646F6@zathras>
Message-ID: <20020307015136.74929.qmail@web10701.mail.yahoo.com>

 --- Raymond Law  wrote: > 
> I was looking at an example script using MPLS and
> have a question.  What does 
> the following do and why is it needed?
$LSR7 refers to an MPLS node, while $LSRmpls7 refers
to the MPLS module ( or MPLS object ) "embedded" in
the MPLS node. Functions that are specific to MPLS
have been implemented in the MPLS module. So 
set m [eval $$a get-module "MPLS"] returns the MPLS
module of the corresponding MPLS node to the variable
m.

Thanks,
Sandeep

> 
> for {set i 2} {$i < 9} {incr i} {
> 	set a LSR$i
> 	set m [eval $$a get-module "MPLS"]
> 	eval set LSRmpls$i $m
> }
> 
> Also, why not just use $LSR7 instead of $LSRmpls7,
> where "set LSR7 [$ns 
> mpls-node]" is defined before?
> 
> $ns at 0.2 "$LSRmpls7 send-ldp-withdraw-msg  9"
> 
> Thanks.
> Ray,
> 
>  

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com


From nauzads@cs.tamu.edu  Wed Mar  6 19:15:02 2002
From: nauzads@cs.tamu.edu (Nauzad Erach Sadry)
Date: Wed Mar  6 19:15:02 2002
Subject: [ns] CHANGE BS - WIRELESS NODE BANDWIDTH -- URGENT
Message-ID: 

Hello

Can some one tell me how do I change the bandwidth between a basestation
and a wireless node.

I tried to use

Phy/Wireless set bandwith 1e6 ... but this does not work at all ... 

Can someone please suggust me something urgently

Thank you

Nauzad E. Sadry



From kopi@stanford.edu  Wed Mar  6 19:45:01 2002
From: kopi@stanford.edu (Milind Dilip Kopikare)
Date: Wed Mar  6 19:45:01 2002
Subject: [ns] Queer behaviour on 802.11
Message-ID: 

Hi All,
	I am trying to simulate a wireless LAN. The problem is no matter
how aggressively I pump in packets, there dosen't seem to be any
congestion at all. I say this because I am monitoring the cw_ (the
collission window) of each mobile station. It barely ever rises more than
CWmin. Note that by the concept of exponential backoff, the cw_ should
increase exponentially each time a collission happens. Even scenarios of
30 mobile stations, each pumping in at a CBR rate of 10 Mbps dosent seem
to create enough congestion. Would anyone know what is going wrong here?

Thanks,
Milind


From rlaw@vt.edu  Thu Mar  7 05:15:01 2002
From: rlaw@vt.edu (Raymond Law)
Date: Thu Mar  7 05:15:01 2002
Subject: [ns] MPLS question
In-Reply-To: <20020307015136.74929.qmail@web10701.mail.yahoo.com>
References: <3C8646F6@zathras>
Message-ID: <5.1.0.14.0.20020306225824.034c1800@mail.vt.edu>

Are all tcl commands for mns v2.0 mpls-specific and thus require the mpls 
module to call?

Thanks.
Ray,

At 01:51 AM 3/7/2002 +0000, Sandeep Darisala wrote:


>  --- Raymond Law  wrote: >
> > I was looking at an example script using MPLS and
> > have a question.  What does
> > the following do and why is it needed?
>$LSR7 refers to an MPLS node, while $LSRmpls7 refers
>to the MPLS module ( or MPLS object ) "embedded" in
>the MPLS node. Functions that are specific to MPLS
>have been implemented in the MPLS module. So
>set m [eval $$a get-module "MPLS"] returns the MPLS
>module of the corresponding MPLS node to the variable
>m.
>
>Thanks,
>Sandeep
>
> >
> > for {set i 2} {$i < 9} {incr i} {
> >       set a LSR$i
> >       set m [eval $$a get-module "MPLS"]
> >       eval set LSRmpls$i $m
> > }
> >
> > Also, why not just use $LSR7 instead of $LSRmpls7,
> > where "set LSR7 [$ns
> > mpls-node]" is defined before?
> >
> > $ns at 0.2 "$LSRmpls7 send-ldp-withdraw-msg  9"
> >
> > Thanks.
> > Ray,
> >
> >
>
>__________________________________________________
>Do You Yahoo!?
>Everything you'll ever need on one web page
>from News and Sport to Email and Music Charts
>http://uk.my.yahoo.com


From shantanu@ee.eng.ohio-state.edu  Thu Mar  7 05:15:15 2002
From: shantanu@ee.eng.ohio-state.edu (Shantanu Bhardwaj)
Date: Thu Mar  7 05:15:15 2002
Subject: [ns] audio & video packet headers
Message-ID: <3C86E807.AF369231@ee.eng.ohio-state.edu>

what are the audio and video packets headers for. 
these are listed in the packets.h file.

enum packet_t {
	PT_TCP,
	PT_UDP,
	PT_CBR,
	PT_AUDIO,
	PT_VIDEO,
	.
	.
	.


class p_info {
public:
	p_info() {
		name_[PT_TCP]= "tcp";
		name_[PT_UDP]= "udp";
		name_[PT_CBR]= "cbr";
		name_[PT_AUDIO]= "audio";
		name_[PT_VIDEO]= "video";


-Shantanu


From zhaihongqiang@msrchina.research.microsoft.com  Thu Mar  7 05:15:33 2002
From: zhaihongqiang@msrchina.research.microsoft.com (Zhai Hongqiang)
Date: Thu Mar  7 05:15:33 2002
Subject: [ns] Why are there so many unexpected warning when I use NAM
Message-ID: <85256B849596A5429983B497A65508A5C60A7E@msrcnx.msrchina.research.microsoft.com>

This is a multi-part message in MIME format.

--------------InterScan_NT_MIME_Boundary
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1C59D.C52403F6"

------_=_NextPart_001_01C1C59D.C52403F6
Content-Type: text/plain;
	charset="US-ASCII"
Content-Transfer-Encoding: quoted-printable

C:\NetSim\genapps>nam out.nam

Cannot connect to existing nam instance. Starting a new one...

Unexpected end of line in: n -t * -a 0 -s 0 -S UP -v circle -c black -i
black

Perhaps you are missing a value following the attribute flag.

Last parsed attribute flag was -i

=20

Unexpected end of line in: n -t * -a 1 -s 1 -S UP -v circle -c black -i
black

Perhaps you are missing a value following the attribute flag.

Last parsed attribute flag was -i

=20

Unexpected end of line in: + -t 0.1 -s 0 -d 1 -p cbr -e 250 -c 0 -i 0 -a
0 -x {0

.0 1.0 0 ------- null}

Perhaps you are missing a value following the attribute flag.

Last parsed attribute flag was -x

=20

...

...

...

=20

=20


------_=_NextPart_001_01C1C59D.C52403F6
Content-Type: text/html;
	charset="US-ASCII"
Content-Transfer-Encoding: quoted-printable





















C:\NetSim\genapps>nam =
out.nam



Cannot connect =
to existing
nam instance. Starting a new one...



Unexpected end =
of line in:
n -t * -a 0 -s 0 -S UP -v circle -c black -i black



Perhaps you are =
missing a
value following the attribute flag.



Last parsed =
attribute flag
was -i



 



Unexpected end =
of line in:
n -t * -a 1 -s 1 -S UP -v circle -c black -i black



Perhaps you are =
missing a
value following the attribute flag.



Last parsed =
attribute flag
was -i



 



Unexpected end =
of line in:
+ -t 0.1 -s 0 -d 1 -p cbr -e 250 -c 0 -i 0 -a 0 -x {0



.0 1.0 0 =
------- null}



Perhaps you are =
missing a
value following the attribute flag.



Last parsed =
attribute flag
was –x



 



…



…



…



 



 









------_=_NextPart_001_01C1C59D.C52403F6--

--------------InterScan_NT_MIME_Boundary--


From Nazih.KHADDAJMALLAT@enst-bretagne.fr  Thu Mar  7 05:16:26 2002
From: Nazih.KHADDAJMALLAT@enst-bretagne.fr (KHADDAJ MALLAT Nazih)
Date: Thu Mar  7 05:16:26 2002
Subject: [ns] question about ns , urgent please
Message-ID: <20020307110843.A5575@p3ainfo2.enst-bretagne.fr>

Hi,

Please can you help me for :

             * write a command line in TCL script in order to have a link
between two nodes during the simulation 
               for example, the command line for link is : $ns duplex-link
$n7 $n4 10Mb 10ms DropTail , but it is running 
               at the beginning of simulation . I want to run this command
line , for example , after 2 seconds what should 
		i write .
		i say link two nodes , but i have the same question for
adding a node during the simulation or any event i want 
		to happen but after a time not exactly at the start of the
simulation.

		* how can i get the routing table for a network ?, for
example in static routing , a routing table is created but how can i view
		this table.

		* What is the difference between the session routing and
dynamic routing in NS ?

                Thank you and please i am waiting your answer ( urgently )


-- 
Nazih  KHADDAJ  MALLAT
Ecole Nationale Superieure des Telecommunications 
de   BRETAGNE
ENST BP832  29285 BREST   CEDEX
portable :   06 16 38 96 99


From sbchen@iipc.zju.edu.cn  Thu Mar  7 05:16:37 2002
From: sbchen@iipc.zju.edu.cn (Chen Shangbing)
Date: Thu Mar  7 05:16:37 2002
Subject: [ns] Ask again: About the Maximum Window Size of TCP
Message-ID: <200203071052.SAA15400@iipc.zju.edu.cn>

Hi, everyone,

In ns-doc, page 97 about the TCP object, it says: 

	MWS The Maximum Window Size in packets for a TCP connection. MWS determines the size of an array in tcpsink.cc. The default for MWS is 1024 packets. For Tahoe TCP, the "window" parameter, representing the receiver's advertised window, should be less than MWS-1. For Reno TCP, the window" parameter should be less than (MWS-1)/2.

I can not understand the expression of 'MWS-1' and '(MWS-1)/2'. Can you give me a detailed interpretation? 

Thank you in advance.


            Chen Shangbing
            sbchen@iipc.zju.edu.cn


From sbchen@iipc.zju.edu.cn  Thu Mar  7 05:16:48 2002
From: sbchen@iipc.zju.edu.cn (Chen Shangbing)
Date: Thu Mar  7 05:16:48 2002
Subject: [ns] Ask again: Why there is no packet drop in my RED simulation?
Message-ID: <200203071101.TAA15516@iipc.zju.edu.cn>

Hi, ns-users£¬

I am doing a simple RED simulation as below:

	set ns [new Simulator]

	# Define core nodes and other nodes
	set nc0 [$ns node]
	set nc1 [$ns node]
	set i 1
	while {$i <= 5} {
		set src($i) [$ns node]
		set des($i) [$ns node]
		incr i
	}

	Queue/DropTail set drop_front_ 1
	Queue set limit_ 100

	# Define the bottleneck link 
	$ns simplex-link $nc0 $nc1 10Mb 10ms RED
	$ns simplex-link $nc1 $nc0 1000Mb 10ms DropTail

	# RED Parameters
	set link1 [$ns get-link $nc0 $nc1]
	set qlink1 [$link1 queue]
	$qlink1 set limit_ 150
	$qlink1 set thresh_ 75
	$qlink1 set maxthresh_ 150
	$qlink1 set mean_pktsize_ 1000 
	$qlink1 set drop_front_ true


	#Attach the access nodes with the core nodes
	$ns duplex-link $src(1) $nc0 30Mb 25ms DropTail
	$ns duplex-link $des(1) $nc1 30Mb 25ms DropTail
	$ns duplex-link $src(2) $nc0 30Mb 30ms DropTail
	$ns duplex-link $des(2) $nc1 30Mb 30ms DropTail
	...
	$ns duplex-link $src(5) $nc0 30Mb 45ms DropTail
	$ns duplex-link $des(5) $nc1 30Mb 45ms DropTail

	set i 1
	while {$i <= 5} {
		set p($i) [new Agent/TCP/Reno]
		$ns attach-agent $src($i) $p($i)
		$p($i) set fid_ [expr $i + 100]
		set q($i) [new Agent/TCPSink]
		$ns attach-agent $des($i) $q($i)
		$ns connect $p($i) $q($i)
		set f($i) [new Application/FTP]
		$f($i) attach-agent $p($i)
		incr i
	}

	set i 1
	while {$i <= 5} {
		set starttime [expr $i * 0.1]
		$ns at $starttime "$f($i) start"
		incr i
	}

	$ns at 20.0 "finish"
	$ns run

I think there has been enough traffic load, and the average queue length should be 
between 75 and 150.
But the graph result shows that the queue length is around 50, and there were no 
drops.

What's wrong with my simulation?


            Chen Shangbing
            sbchen@iipc.zju.edu.cn


From csysmd@comp.leeds.ac.uk  Thu Mar  7 05:17:01 2002
From: csysmd@comp.leeds.ac.uk (Steve)
Date: Thu Mar  7 05:17:01 2002
Subject: [ns] Monitoring bytes
In-Reply-To: <1015463612.936.105.camel@ski.isi.edu>
Message-ID: 

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

--193059218-130954919-1015498741=:26527
Content-Type: TEXT/PLAIN; charset=US-ASCII

Thanks for replying,


On 6 Mar 2002, Tim Buchheim wrote:

> Your script is printing correct results.  Your Agent/LossMonitor object
> ($sink0) receives no packets because it is not connected to anything.
> The only packets sent in your simulation are received by $sink, not
> $sink0.

This was a mistake - I got the names wrong for the sink. I have
corrected this and changed the type of connection to UDP. I still get
the same results though.

> Note that Agent/LossMonitor does not work with TCP.  The following
> thread gives some ideas on modifying Agent/TCPSink to keep some stats.
>
> http://mailman.isi.edu/pipermail/ns-users/2001-February/013735.html

Thanks for providing this link, I was hoping that I would be able to do
everything I need to without modifying the ns code. But maybe I will
have to!


I have provided the script if you want to look,

Steve

--193059218-130954919-1015498741=:26527
Content-Type: APPLICATION/x-tcl; name="loss.tcl"
Content-Transfer-Encoding: BASE64
Content-ID: 
Content-Description: 
Content-Disposition: attachment; filename="loss.tcl"
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--193059218-130954919-1015498741=:26527--


From non-nil@gmx.net  Thu Mar  7 06:20:01 2002
From: non-nil@gmx.net (Martin Schmidt)
Date: Thu Mar  7 06:20:01 2002
Subject: [ns] xgraph segfaults if invoked in batch mode: xgraph -device ps -o out.ps foo.xgraph
Message-ID: <20020307133624.GA5912@schmidt.dnsalias.org>

Hello,

xgraph will segfault when invoked with -device  -o 

spice$ xgraph -device ps -o foo.ps v10.xgraph
Segmentation fault

The segfault is probably caused by the function call XFlush(disp) in 
DrawData at draw.c
But I don't know that for sure, cause I rebuilt it with the 12.1
sources.

I'm no X hacker, so I don't know exactly what is to be done; I think  
XFlush is called on a display which is not existent.
I enclosed the XFlush in an if clause which checks wether there is an
disp.

thanks.

regards,
Martin


From kkco@cin.ufpe.br  Thu Mar  7 07:45:02 2002
From: kkco@cin.ufpe.br (Karina Karla C. de Oliveira)
Date: Thu Mar  7 07:45:02 2002
Subject: [ns] PHB EF.
Message-ID: 

Dear All.

I would like to know if the diffserv version in ns-2.1b8 implements PHB
EF? Or only PHB AF?

In chapter 9 of ns manual, there is the following statement:
"In ns, packets are defaulted to a code point of zero. Therefore, user
must add a PHB entry for the zero code point in order to handle best
effort traffic."

But if a don't specify an entry in policy table to a specific traffic,
then "segmentation fault" ocurrs. I thought that in this case, code point
0 will be attributed by default. Am I wrong?

I don't want to use PHB AF to my voice traffic. And the others
traffics will be BE. There isn't AF traffic in my scenario. What I have to
do to configure them?

Best Regards,
Karina Oliveira.

*********************************************************
        Karina Karla Cavalcante de Oliveira
	Computer Science
        Universidade Federal de Pernambuco
        Centro de Informática - CIn
        Caixa Postal 7851
        CEP: 50732-970
        Recife-PE - Brasil
        Phone: +55 (081) 3271-8430

        Email: kkco@cin.ufpe.br
	WWW: www.cin.ufpe.br/~kkco
	Phones: (HOME) +55 (081) 3338-2054
*********************************************************








From geunkim@postech.ac.kr  Thu Mar  7 07:50:03 2002
From: geunkim@postech.ac.kr (Geunhyung Kim)
Date: Thu Mar  7 07:50:03 2002
Subject: [ns] Update by Snapshot
Message-ID: 

Hello! All,

I would like to update source code using snapshot.
Could you tell me how to updat source code though snapshot smartly ?

Thanks in advance,

Geunhyung

None of us is as smart as all of us
==========================================
Geunhyung Kim

E-mail: geunkim@postech.edu

Tel: +82-54-279-5655
Fax: +82-54-279-5699

Networking & Distributed Systems Lab.
CSE
POSTECH
===========================================


From jasonohcs@yahoo.com  Thu Mar  7 09:00:01 2002
From: jasonohcs@yahoo.com (oh jason)
Date: Thu Mar  7 09:00:01 2002
Subject: [ns] rate_ = transmission rate?
Message-ID: <20020307165610.88630.qmail@web20605.mail.yahoo.com>

Hi, I would like to know whether setting rate_ for a
cbr traffic means setting the transmission rate. For
example, in my simulation, I want a packet of 64 bytes
transmitted every 30ms, so the transmission rate
should be 64x8/30e-3 = 17.066kbps. So would it be
correct to have "$cbr set rate_ 17.066kb" to realize
the situation above in ns?

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From dbelen@netmode.ece.ntua.gr  Thu Mar  7 09:10:05 2002
From: dbelen@netmode.ece.ntua.gr (Dimitris Velenis)
Date: Thu Mar  7 09:10:05 2002
Subject: [ns] Ask again: About the Maximum Window Size of TCP
In-Reply-To: <200203071052.SAA15400@iipc.zju.edu.cn>; from sbchen@iipc.zju.edu.cn on Thu, Mar 07, 2002 at 06:42:08PM +0800
References: <200203071052.SAA15400@iipc.zju.edu.cn>
Message-ID: <20020307190948.B48881@netmode.ece.ntua.gr>

MWS is actually the size of the array (int seen_[MWS]) that represents the receiver's 
(TcpSink) buffer. This buffer must be able to accomodate a window's worth of segments.
That is why Tahoe's maximum window should be less than MWS-1. 

Reno TCP inflates its cwnd during recovery by the amount of dupACKs received.
If cwnd0 is the window value when Fast Retransmit is triggered, the Reno sender may 
receive up to cwnd0 dupACKs. Therefore, at the end of recovery the sender's window 
may have reached a value up to 2*cwnd0. In this case 2*cwnd0 must be less than MWS-1.

Dimitris.

On Thu, Mar 07, 2002 at 06:42:08PM +0800, Chen Shangbing wrote:
> 
> Hi, everyone,
> 
> In ns-doc, page 97 about the TCP object, it says: 
> 
> 	MWS The Maximum Window Size in packets for a TCP connection. MWS determines the size of an array in tcpsink.cc. The default for MWS is 1024 packets. For Tahoe TCP, the "window" parameter, representing the receiver's advertised window, should be less than MWS-1. For Reno TCP, the window" parameter should be less than (MWS-1)/2.
> 
> I can not understand the expression of 'MWS-1' and '(MWS-1)/2'. Can you give me a detailed interpretation? 
> 
> Thank you in advance.
> 
> 
>             Chen Shangbing
>             sbchen@iipc.zju.edu.cn

-- 
Dimitris Velenis                              
Network Management & Optimal Design Lab
National Technical University of Athens


From alefiyah@catarina.usc.edu  Thu Mar  7 09:40:03 2002
From: alefiyah@catarina.usc.edu (Alefiya Hussain)
Date: Thu Mar  7 09:40:03 2002
Subject: [ns] NSE: Scheduler going backwards in time
In-Reply-To: <16071.1015241650@www61.gmx.net>
Message-ID: <20020307093354.W50226-100000@rumi.usc.edu>

It indicates that the emulator cannot keep up with real time. Using a
faster computer will elevate this problem.

The results are not affeted drastically from this error.
Though you most prbably will be experiencing high packet losses  in bpf.

alefiya
On Mon, 4 Mar 2002 netsim@gmx.at wrote:

>
> Hi!
>
> I am using the emulation mode in ns-2 called nse and ecountered the
> following problem:
>
> scheduler going backwards in time
>
> What is the reason for this error and how severe is it? Are all my trace
> files useless when getting this message?
>
> Thanks a lot.
>
> Martin Joerg
>
> --
> GMX - Die Kommunikationsplattform im Internet.
> http://www.gmx.net
>
>


From buchheim@ISI.EDU  Thu Mar  7 11:30:01 2002
From: buchheim@ISI.EDU (Tim Buchheim)
Date: Thu Mar  7 11:30:01 2002
Subject: [ns] Monitoring bytes
In-Reply-To: 
References: 
Message-ID: <1015529158.936.123.camel@ski.isi.edu>

--=-vqWmkDUl6ps5RMly9NQC
Content-Type: text/plain
Content-Transfer-Encoding: 7bit

On Thu, 2002-03-07 at 02:59, Steve wrote:
> Thanks for replying,
> 
> 
> On 6 Mar 2002, Tim Buchheim wrote:
> 
> > Your script is printing correct results.  Your Agent/LossMonitor object
> > ($sink0) receives no packets because it is not connected to anything.
> > The only packets sent in your simulation are received by $sink, not
> > $sink0.
> 
> This was a mistake - I got the names wrong for the sink. I have
> corrected this and changed the type of connection to UDP. I still get
> the same results though.

You still aren't connecting your traffic source to the loss monitor. 
Instead, you have connected it to a null agent.

Try the attached script (a modified version of your last one).  Note
that the traffic source is now sending its data to the loss monitor.



-- 
Tim Buchheim

--=-vqWmkDUl6ps5RMly9NQC
Content-Disposition: attachment; filename=loss.tcl
Content-Transfer-Encoding: quoted-printable
Content-Type: text/x-tcl; charset=ISO-8859-1

# This is a simple script to monitor packets lost

set ns [new Simulator]=20

#for nam
set nf [open loss.nam w]
$ns namtrace-all $nf

#for xgraph
set f0 [open loss.tr w]

proc finish {} {
   global ns nf f0
   $ns flush-trace
   close $nf
   close $f0
   exec nam loss.nam &
   exec xgraph loss.tr -geometry 800x400 &
   exit 0
}

set n0 [$ns node]
set n1 [$ns node]
set n10 [$ns node]

$ns duplex-link $n0 $n1 1Mb 10ms DropTail
$ns duplex-link $n1 $n10 0.1Mb 10ms DropTail

$ns duplex-link-op $n0 $n1 orient right
$ns duplex-link-op $n1 $n10 orient right

$ns duplex-link-op $n0 $n1 queuePos 0.5
$ns duplex-link-op $n1 $n10 queuePos 0.5

#CBR/UDP
$ns color 1 Red

#Setup UDP
set udp [new Agent/UDP]
$ns attach-agent $n0 $udp
$udp set fid_ 1

#Setup CBR
set cbr [new Application/Traffic/CBR]
$cbr attach-agent $udp
$cbr set type_ CBR
$cbr set packet_size_ 1000
$cbr set rate_ 1mb
$cbr set random_ false

#Create sink for UDP
set sink [new Agent/LossMonitor]
$ns attach-agent $n10 $sink
$ns connect $udp $sink

proc record {} {
          global sink f0
          set ns [Simulator instance]
	 =20
          #Set the time after which the procedure should be called again
          set time 0.5
	 =20
          #How many bytes have been received by the traffic sinks?
          set bw0 [$sink set bytes_]
	 =20
          #Get the current time
          set now [$ns now]
          #Calculate the bandwidth (in MBit/s) and write it to the files
          puts $f0 "$now [expr $bw0]"
	 =20
          #Reset the bytes_ values on the traffic sinks
          $sink set bytes_ 0

          #Re-schedule the procedure
          $ns at [expr $now+$time] "record"
}

$ns at 0.0 "record"
$ns at 0.1 "$cbr start"
$ns at 10.0 "$cbr stop"
$ns at 10.0 "finish"

$ns run

--=-vqWmkDUl6ps5RMly9NQC--


From wido   Thu Mar  7 11:40:01 2002
From: wido  (wido)
Date: Thu Mar  7 11:40:01 2002
Subject: [ns] (03/07/2002) Trace graph 1.30b (NS trace files analyser), www.geocities.com/tracegraph/
Message-ID: <192828719.20020307203728@o2.pl>



From tmaneesh@cedt.iisc.ernet.in  Thu Mar  7 12:05:12 2002
From: tmaneesh@cedt.iisc.ernet.in (Maneesh Tewari)
Date: Thu Mar  7 12:05:12 2002
Subject: [ns] invalid packet type :exp in wireless simulation while using expon
 ential traffic
Message-ID: 

hello ns-experts ,
 
i'm also facing the same problem what 've been posted on the ns-user list on
6th Feb 2001 regarding the invalid packet type : exp while defining the
exponential traffic on the wireless.

now a solution has been also provided on the next day 7th feb to do some
changes in cmu-trace.cc/.h

sir can u let me know about the function in cmu-trace.cc what must have been
added to sort out that problem :

void CMUTrace::format_ping(Packet *p, int offset) 
{        
    if (newtrace_)
                   { ... } 
             else { ... }
       } 

i'm not able to think that what should i put in the body of this function.

thanks 
regards 
maneesh


From ellich8459@yahoo.com  Thu Mar  7 12:10:05 2002
From: ellich8459@yahoo.com (elliot chen)
Date: Thu Mar  7 12:10:05 2002
Subject: [ns] how to generate bursty packet losses
Message-ID: <20020307200932.44283.qmail@web21402.mail.yahoo.com>

Hello, all

Does anyone know how to generate bursty pkt loss
(consecutive pkt loss)
using the current error model in ns?

Thank you so much!

Elliot



__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From rahul b jain cs student   Thu Mar  7 12:25:02 2002
From: rahul b jain cs student  (rahul b jain cs student)
Date: Thu Mar  7 12:25:02 2002
Subject: [ns] Problems while installing otcl for windows platform
Message-ID: <200203072021.g27KLQT09784@goode.njit.edu>

hi,

my name is rahul and i am graduate student in NJIT. I was trying to
install the windows version of ns2. i was able to install "tcl" and
"tk" without any problem. but i ran into problem when i was trying to
install "otcl". I am getting the following kinds of error

1.otcl.c(21) : fatal error C1083: Cannot open include file: 'tclInt.h': No
  such file or directory

2.LIB : fatal error LNK1181: cannot open input file "otcl.o"

3.otclAppInit.c(17) : fatal error C1083: Cannot open include
  file: 'tcl.h': No such file or directory

4.otkAppInit.c(18) : fatal error C1083: Cannot open include
  file: 'tk.h': No such file or directory 

5.LINK : fatal error LNK1181: cannot open input file "otkAppInit.o"

I checked my classpath and path and both of them contain the directory in
which tcl was installed.

I would really appreciate if you could give me some guidance,

Thanks,
Rahul Jain.


From bali_n@yahoo.com  Thu Mar  7 13:10:01 2002
From: bali_n@yahoo.com (Sudarshan Narasimha)
Date: Thu Mar  7 13:10:01 2002
Subject: [ns] NS Make Failed - URGENT
Message-ID: <20020307210549.28782.qmail@web20505.mail.yahoo.com>

Hi,
I am trying to install the dialy snapshot of NS and
NAM on my Solaris machine.
(SunOS 5.7 Generic_106541-15,Ultra-5_10)

The NS installation exits with the following error:
c++ -c -O2 -DTCP_DELAY_BIND_ALL -DNO_TK -DNIXVECTOR 
-DTCLCL_CLASSINSTVAR  -DNDEBUG -DHAVE_STL -mv8
-msupersparc -D__svr4__ -DUSE_SHM 
-DHAVE_LIBTCLCL -DHAVE_TCLCL_H -DHAVE_LIBOTCL1_0A7
-DHAVE_OTCL_H -DHAVE_LIBTK8_3 
-DHAVE_TK_H -DHAVE_LIBTCL8_3 -DHAVE_TCL_H 
-DHAVE_CONFIG_H -DPGM 
-DPGM_DEBUG -DNS_DIFFUSION -I. 
-I/afs/bp.ncsu.edu/dist/netsim/ns-2/tclcl-1.0b11
-I/afs/bp.ncsu.edu/dist/netsim/ns-2/otcl-1.0a7 
-I/afs/bp.ncsu.edu/dist/netsim/ns-2/include
-I/afs/bp.ncsu.edu/dist/netsim/ns-2/include 
-I/usr/include/pcap -I./tcp -I./common -I./link
-I./queue -I./adc -I./apps 
-I./mac -I./mobile -I./trace -I./routing -I./tools
-I./classifier 
-I./mcast -I./diffusion3/main -I./diffusion3/lib
-I./diffusion3/nr 
-I./diffusion3/ns -o tools/rng.o tools/rng.cc
tools/rng.cc:486: sorry, not implemented: namespace
make: *** [tools/rng.o] Error 1
Ns make failed!
See http://www.isi.edu/nsnam/ns/ns-problems.html for
problems

I have installed the latest version of tcl/tk (8.3.2) 
I have made sure that the TCLSH variable in the
Makefile points to 
tclsh8.3

Please help.

Thanks in advance
Sudarshan

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From buchheim@ISI.EDU  Thu Mar  7 13:40:02 2002
From: buchheim@ISI.EDU (Tim Buchheim)
Date: Thu Mar  7 13:40:02 2002
Subject: [ns] Processing delay
In-Reply-To: <5.1.0.14.0.20020302203917.03c35ae8@mail.vt.edu>
References: <5.1.0.14.0.20020302203917.03c35ae8@mail.vt.edu>
Message-ID: <1015537113.936.256.camel@ski.isi.edu>

On Sat, 2002-03-02 at 17:39, Raymond Law wrote:
> 
> The queue_ in an ns node models the queuing delay, and the link_ in an ns 
> node models the transmission delay. Propagation delay can also be specified 
> when creating a link. Is there a way to model the processing delay in the node?

I don't think there is any easy way to do this without modifying
existing code.


> Also, is there a good way to measure the end-to-end delay?

Lloyd Wood posted a script to extract end-to-end delay from trace files:

  http://www.isi.edu/nsnam/archive/ns-users/webarch/1999/msg02093.html



-- 
Tim Buchheim


From hli@navini.com  Thu Mar  7 15:40:01 2002
From: hli@navini.com (hang li)
Date: Thu Mar  7 15:40:01 2002
Subject: [ns] installing ns2 on NT
Message-ID: <1F003D27978ABD419A4CC06B6F89CE66116460@navexch001.navini.com>

Hi,

I try to install ns2-2.1b8a-win on my NT desktop but cannot continue due
to the following message:

NMAKE : fatal error U1073: don't know how to make '.\Release\panic.obj'
Stop.

I used sample makefiles downloaded from
http://www.isi.edu/nsnam/ns/ns-win32-build.html.

Could you give some suggestion?

Thanks.

Hang               


From gudipati@naur.csee.wvu.edu  Thu Mar  7 19:45:08 2002
From: gudipati@naur.csee.wvu.edu (Neeraja Gudipati)
Date: Thu Mar  7 19:45:08 2002
Subject: [ns] ns installation error
Message-ID: 

Hi,

When I try to install ns-allinone-2.1b8a on a unix machine, while compiling
ns-2.1b8a it gives the following error:
----------------------------------------------------------------
In file included from scheduler.cc:49:
mem-trace.h: In method `void MemInfo::checkpoint()':
mem-trace.h:67: implicit declaration of function `int getrusage(...)'
*** Error code 1
make: Fatal error: Command failed for target `scheduler.o'
Ns make failed!
-----------------------------------------------------------------

Can someone tell me what's causing this error.

Thanks

-Neeraja



From svarma@cedar.Buffalo.EDU  Fri Mar  8 05:15:04 2002
From: svarma@cedar.Buffalo.EDU (Saket Varma)
Date: Fri Mar  8 05:15:04 2002
Subject: [ns] ns installation error
Message-ID: 

Did u try a make depend before the make??
maybe that would help...

Saket

On Thu, 7 Mar 2002, Neeraja Gudipati wrote:

>
>
> Hi,
>
> When I try to install ns-allinone-2.1b8a on a unix machine, while compiling
> ns-2.1b8a it gives the following error:
> ----------------------------------------------------------------
> In file included from scheduler.cc:49:
> mem-trace.h: In method `void MemInfo::checkpoint()':
> mem-trace.h:67: implicit declaration of function `int getrusage(...)'
> *** Error code 1
> make: Fatal error: Command failed for target `scheduler.o'
> Ns make failed!
> -----------------------------------------------------------------
>
> Can someone tell me what's causing this error.
>
> Thanks
>
> -Neeraja
>
>


From JZhao_NS@hotmail.com  Fri Mar  8 05:15:19 2002
From: JZhao_NS@hotmail.com (Jun Zhao)
Date: Fri Mar  8 05:15:19 2002
Subject: [ns] installing ns2 on NT
References: <1F003D27978ABD419A4CC06B6F89CE66116460@navexch001.navini.com>
Message-ID: 

Try to install NS on a linux machine. It is fairly easy and usually one
command line is OK. For windows version installation, you have to modify
many settings which may be a frustrated experience.
----- Original Message -----
From: "hang li" 
To: 
Sent: Friday, March 08, 2002 7:35 AM
Subject: [ns] installing ns2 on NT


>
> Hi,
>
> I try to install ns2-2.1b8a-win on my NT desktop but cannot continue due
> to the following message:
>
> NMAKE : fatal error U1073: don't know how to make '.\Release\panic.obj'
> Stop.
>
> I used sample makefiles downloaded from
> http://www.isi.edu/nsnam/ns/ns-win32-build.html.
>
> Could you give some suggestion?
>
> Thanks.
>
> Hang
>
>


From gudipati@naur.csee.wvu.edu  Fri Mar  8 05:15:30 2002
From: gudipati@naur.csee.wvu.edu (Neeraja Gudipati)
Date: Fri Mar  8 05:15:30 2002
Subject: [ns] ns installation error
In-Reply-To: 
Message-ID: 

Thanks for the response.
"make depend" of ns-2.1b8a stops with this information:

emulate/ether.cc:46: net/ethernet.h: No such file or directory
emulate/arp.cc:50: net/ethernet.h: No such file or directory

Thanks

Neeraja


On Thu, 7 Mar 2002, Saket Varma wrote:

> Did u try a make depend before the make??
> maybe that would help...
> 
> Saket
> 
> On Thu, 7 Mar 2002, Neeraja Gudipati wrote:
> 
> >
> >
> > Hi,
> >
> > When I try to install ns-allinone-2.1b8a on a unix machine, while compiling
> > ns-2.1b8a it gives the following error:
> > ----------------------------------------------------------------
> > In file included from scheduler.cc:49:
> > mem-trace.h: In method `void MemInfo::checkpoint()':
> > mem-trace.h:67: implicit declaration of function `int getrusage(...)'
> > *** Error code 1
> > make: Fatal error: Command failed for target `scheduler.o'
> > Ns make failed!
> > -----------------------------------------------------------------
> >
> > Can someone tell me what's causing this error.
> >
> > Thanks
> >
> > -Neeraja
> >
> >
> 
> 


From asaf_shabtai@hotmail.com  Fri Mar  8 05:15:42 2002
From: asaf_shabtai@hotmail.com (Asaf Shabtai)
Date: Fri Mar  8 05:15:42 2002
Subject: [ns] Does anyone knows ? - Counting packets sent by nodes
Message-ID: 

Hi,
I'm tring to add a variable in the node, so each time a packet is sent by 
the node to the next node te variable is change (for example let's say that 
the each time the node sends a packet its variable is incremented by 7).
Does anybody have an idea how to implenet that ?
Thank U
Asaf

_________________________________________________________________
MSN Photos is the easiest way to share and print your photos: 
http://photos.msn.com/support/worldwide.aspx


From arshahid@yahoo.com  Fri Mar  8 05:15:54 2002
From: arshahid@yahoo.com (Ahmad R Shahid)
Date: Fri Mar  8 05:15:54 2002
Subject: [ns] $ns_ node-config
Message-ID: <20020308072129.4353.qmail@web10505.mail.yahoo.com>

Hi all,

in my tcl file i have written the following lines:

for {set j 0} {$j < 5} {incr j} {
    set node_($j) [ $opt(rp)-create-mobile-node $j
[lindex $temp $j] ]
    $node_($j) base-station [AddrParams addr2id
[$BS(0) node-addr]]
}

now i want to add/change certain features in the
mobile nodes created by [ $opt(rp)-create-mobile-node
$j [lindex $temp $j] ] command. how do i do that using
$ns_ node_config command?

Ahmad

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From aam11@rediffmail.com  Fri Mar  8 05:16:08 2002
From: aam11@rediffmail.com (Amuthamozhi)
Date: Fri Mar  8 05:16:08 2002
Subject: [ns] [bug] Nam problem in the wireless tcl program
Message-ID: <200203080737.HAA00063@www.isi.edu>

[Bug Report]

-----------------------------
Category:  Other
Package:   ns 2.1b8
OS:        Linux redHat release 7.1
Environment Variables:
LD_LIBRARY_PATH=
TCL_LIBRARY=
TK_LIBRARY=


-----------------------------
Description:

I found than none of the nam file generated by the wireless related tcl
scripts can be displayed correctly by nam-1.0a10.

Besides the following warnings
"perhaps you have extra space at the end of the line
 last parsed attribute flag was -k"

I go through the ns mailing list.In order to fix the problem,ns list 
suggested to attach the patch file.But that patch file is not found in 
the mailing list.

  Anyone please sent the patch file to my email address
      "aam11@rediffmail.com"



From Rajeh.ElAli@uts.edu.au  Fri Mar  8 05:16:22 2002
From: Rajeh.ElAli@uts.edu.au (Rajeh El Ali)
Date: Fri Mar  8 05:16:22 2002
Subject: [ns] NS-2.18b - installation problems
Message-ID: <1983bc197c50.197c501983bc@uts.edu.au>

(this is my original mail resent)

 To Whom It May Concern,

 I wonder if someone can help me with this problem. I am tyring to
 install the NS ver2.1b8a software package. However it keeps on giving
 the error that TCLSH could not be found. So I went to the TCL8.3.2 and
 tried to compile that package by itself. It seemed to go Ok (inspite of
 some Pointer misuse warnings) and created the TCLSH executable. However
 when I try to run the file I get an error saying that there is no such
 file. I hope you can shed some light on this issue, as I need to start
 working on my Thesis soon.

 Rajeh El-Ali
 phone: 8900 2615
 mobile: (0425) 219599
 UTS Telecommunications Engineering

Rajeh El-Ali
phone: 8900 2615
mobile: (0425) 219599
UTS Telecommunications Engineering



UTS CRICOS Provider Code:  00099F

DISCLAIMER
=====================================================================
This email message and any accompanying attachments may contain
confidential information.  If you are not the intended recipient, do not
read, use, disseminate, distribute or copy this message or attachments.
If you have received this message in error, please notify the sender
immediately and delete this message. Any views expressed in this message
are those of the individual sender, except where the sender expressly,
and with authority, states them to be the views the University of
Technology Sydney. Before opening any attachments, please check them for
viruses and defects.
=====================================================================



From arijitg@iitg.ernet.in  Fri Mar  8 05:16:38 2002
From: arijitg@iitg.ernet.in (Arijit Ganguly)
Date: Fri Mar  8 05:16:38 2002
Subject: [ns] Pragmatic Generic Multicast
Message-ID: 

Hello friends
Can you please let me know some links where I can get some useful papers 
on Pragmatic Generic Multicast (PGM)?
Thanking you.

Regards
Arijit




-- 
------------------------------------------------------------------------------
                          Arijit Ganguly
                        Final Year Student
            Department of Computer Science & Engineering
              Indian Institute of Technology, Guwahati
                Guwahati - 781039, District Kamrup
                          Assam, India
          ------------------------------------------------
                Phone : +00-91-0361-690763 (room #203)
                Fax   : +91 361 690762
                email : ariganguly@yahoo.co.in
                CC    : ariganguly@hotmail.com
          ------------------------------------------------




From gdemarco@unisa.it  Fri Mar  8 05:16:54 2002
From: gdemarco@unisa.it (giuseppe de marco)
Date: Fri Mar  8 05:16:54 2002
Subject: [ns] bad manual
Message-ID: <20020308130648.3dc23d02.gdemarco@unisa.it>

THe ns-manual is the worst manual I never read in my life



-- 
Giuseppe De Marco
Department of Electrical and Information Engineering (DIIIE)
University of Salerno
tel . +39 089 964012


From ramya4_a@rediffmail.com  Fri Mar  8 05:17:08 2002
From: ramya4_a@rediffmail.com (Ramya)
Date: Fri Mar  8 05:17:08 2002
Subject: [ns] [bug] Problem with dynamic routing with LAN
Message-ID: <200203081211.MAA14791@www.isi.edu>

[Bug Report]

-----------------------------
Category:  Other
Package:   ns 2.1b8
OS:        Linux redHat 7.1 
Environment Variables:
LD_LIBRARY_PATH=
TCL_LIBRARY=
TK_LIBRARY=


-----------------------------
Description:

Dear NS users,
          I have created the tcl script program for the network topography that includes the agent
PING,TCP,UDP,RSVP.I want to specify the node 2,3,4 connection to the local area network.The error
occurs for this program.I have also attached the error code below the tcl script.Please give the
solution for this problem
 
 
 
 
#   0 --- 1 -- 2
#   |          |
#   |          |
#   4 ________ 3
#   |
#   |
#   5
#  | |
#  | |
#  6 7

#Create a simulator object
set ns [new Simulator]
 
 
#Define different colors for data flows(for NAM)
$ns color 0  Red
$ns color 1  Blue
$ns color 2  Yellow
$ns color 46 purple
 
 
#Open the nam trace file
set f0 [open out0.tr w]
set f1 [open out1.tr w]
set f  [open out.tr w]
$ns trace-all $f
set nf [open out.nam w]
$ns namtrace-all $nf
 
#Define a 'finish' procedure
proc finish {} {
        global ns f  nf
        $ns flush-trace
        #Close the trace file
        close $f
        close $nf
        #Execute nam on the trace file
        exec nam out.nam &
        exit 0
}
$ns rtproto DV
 
#create eight nodes
for {set i 0} {$i < 8} {incr i} {
      set n($i) [$ns node]
}
 
 
#create lan connection for the node n(2) n(3) n(4)
set lan0 [$ns make-lan "$n(2) $n(3) $n(4)" 10Mb 10ms LL Queue/DropTail Mac/802_3 Channel]
 
#create link between the nodes
$ns duplex-link $n(0) $n(1) 1Mb 10ms DropTail
$ns duplex-link $n(1) $n(2) 1Mb 20ms RED
$ns queue-limit $n(1) $n(2) 25
$ns queue-limit $n(2) $n(1) 25
$ns duplex-link $n(2) $n(3) 1Mb 5ms DropTail
$ns duplex-link $n(3) $n(4) 1Mb 10ms SFQ
$ns duplex-link $n(4) $n(0) 1Mb 10ms FQ
$ns duplex-rsvp-link $n(4) $n(5) 1Mb 1s 0.5 200 5000 Param Null
$ns duplex-rsvp-link $n(5) $n(6) 1Mb 1s 0.5 200 5000 Param Null
$ns duplex-rsvp-link $n(5) $n(7) 1Mb 1s 0.5 200 5000 Param Null
 
 
 
#attach the udp agent to the node n(0)
set udp0 [new Agent/UDP]
$ns attach-agent $n(0) $udp0
#attach a cbr over udp connection
set cbr0 [new Application/Traffic/CBR]
$cbr0 attach-agent $udp0
 
 
#Define a 'recv' function for the class 'Agent/Ping'
Agent/Ping instproc recv {from rtt} {
        $self instvar node_
        puts "node [$node_ id] received ping answer from \
            $from with round-trip-time $rtt ms."
}
 
#attach the ping agent to the node n(0)
set p0 [new Agent/Ping]
$ns attach-agent $n(0) $p0
 
 
#attach the tcp agent to the node n(1)
set tcp0 [new Agent/TCP]
$tcp0 set class_ 2
$ns attach-agent $n(1) $tcp0
$tcp0 set fid_ 1
#attach the ftp over tcp connection
set ftp0 [new Application/FTP]
$ftp0 attach-agent $tcp0
$ftp0 set type_ FTP
 
#attach the ping agent to the node n(2)
set p1 [new Agent/Ping]
$ns attach-agent $n(2) $p1
 
 
#attach the agent sink and null to the node n(3)
set null0 [new Agent/Null]
$ns attach-agent $n(3) $null0
set sink0 [new Agent/TCPSink]
$ns attach-agent $n(3) $sink0
 
#attach the agent rsvp to the node n(4) , n(5) , n(6) , n(7)
Agent/RSVP set noisy_ 255
#create two traffic sinks
set rsvp4 [$n(4) add-rsvp-agent]
set sink6 [new Agent/LossMonitor]
$ns attach-agent $n(4) $sink6
set sink7 [new Agent/LossMonitor]
$ns attach-agent $n(4) $sink7
 
set rsvp5 [$n(5) add-rsvp-agent]
#create two traffic sources
set rsvp6 [$n(6) add-rsvp-agent]
set udp6 [new Agent/UDP]
$ns attach-agent $n(6) $udp6
$udp6 set packetSize_ 500
$udp6 set fid_ 1
set cbr6 [new Application/Traffic/CBR]
$cbr6 set interval_ 0.008
$cbr6 set rate_ 500k
$cbr6 set random_ 1
$cbr6 set packetSize_ 500
$cbr6 attach-agent $udp6
 
set rsvp7 [$n(7) add-rsvp-agent]
set udp7 [new Agent/UDP]
$ns attach-agent $n(7) $udp7
$udp7 set packetSize- 500
$udp7 set fid_ 1
set cbr7 [new Application/Traffic/CBR]
$cbr7 set interval_ 0.008
$cbr7 set rate_ 500k
$cbr7 set random_ 1
$cbr7 set packetSize_ 500
$cbr7 attach-agent $udp7
 
 
 
 
 
#Connection between the node n(0)-n(3), n(1)-n(3)and n(0) - n(2)
$ns connect $tcp0 $sink0
$ns connect $udp0 $null0
$ns connect $p0 $p1
$ns connect $udp6 $sink6
$ns connect $udp7 $sink7
 
#create the same session on the sender node - n(4)
$rsvp6 session $n(4) 1
$rsvp7 session $n(4) 1
#start the procedure record
set counter 0
$ns at 0.0 "record"
 
#path message from the sender
$ns at 1.0 "$rsvp6 sender 0 +500000 5000 32"
$ns at 3.0 "$rsvp7 sender 0 +500000 5000 32"
 
#Reserve message from the node n(6) and n(7)
$ns at 300.0  "$rsvp4 reserve 0 ff +100000 100000 $n(6)"
$ns at 600.0  "$rsvp4 reserve 0 ff +500000 100000 $n(7)"
$ns at 900.0  "$rsvp4 reserve 0 ff +400000 100000 $n(6)"
$ns at 1200.0 "$rsvp4 reserve 0 ff +100000 100000 $n(7)"
$ns at 1500.0 "$rsvp4 reserve 0 ff +300000 100000 $n(7)"
 
#Schedule events
$ns at 1.0 "$p0 send"
$ns at 1.5 "$cbr0 start"
$ns rtmodel-at 2.0 down $n(3) $n(4)
$ns rtmodel-at 2.5 up $n(3) $n(4)
$ns at 3.0 "$cbr0 stop"
$ns at 3.0 "$ftp0 start"
$ns at 4.0 "$ftp0 stop"
$ns at 10.0 "$cbr6 start"
$ns at 1500.0 "$cbr6 stop"
$ns at 10.0 "$cbr7 start"
$ns at 1500.0 "$cbr7 stop"
 
 
#call the finish procedure after 5 seconds simulation time
$ns at 1500.0  "finish"
 
proc record {} {
        global sink6 sink7 ns f0 f1 counter
        set time 10.0
        set bw0 [$sink6 set bytes_]
        set bw1 [$sink7 set bytes_]
        set now [$ns now]
        puts $f0 "$now [expr $bw0/$time*8]"
        $sink6 set bytes_ 0
        $sink7 set bytes_ 0
        $ns at [expr $now+$time] "record"
        if { $counter == 10 } {
              puts "Time: $now"
              set counter 0
        }
        incr counter
}
 
 
#Run the simulation
$ns run
 
 
The following error occur :-
 
 
    (_o35 cmd line 1)
    invoked from within
"_o35 cmd init-routing _o353"
    invoked from within
"catch "$self cmd $args" ret"
    (procedure "_o35" line 2)
    (SplitObject unknown line 2)
    invoked from within
"$node init-routing $self"
    (procedure "_o353" line 9)
    (rtObject init line 9)
    invoked from within
"_o353 init _o35"
    (Class create line 1)
    invoked from within
"rtObject create _o353 _o35"
    invoked from within
"catch "$className create $o $args" msg"
    (procedure "new" line 3)
    invoked from within
"new rtObject $node"
  (procedure "rtObject" line 4)
    (rtObject init-all line 4)
    invoked from within
"rtObject init-all _o11 _o14 _o17 _o20 _o23 _o26 _o29 _o32 _o35"
    ("eval" body line 1)
    invoked from within
"eval rtObject init-all $nodeslist"
    (procedure "Agent/rtProto/DV" line 8)
    (Agent/rtProto/DV init-all line 8)
    invoked from within
"Agent/rtProto/DV init-all"
    ("eval" body line 1)
    invoked from within
"eval Agent/rtProto/$proto init-all $rtprotos_($proto)"
    (procedure "_o10" line 5)
    (RouteLogic configure line 5)
    invoked from within
"[$self get-routelogic] configure"
    (procedure "_o3" line 4)
    (Simulator run line 4)
    invoked from within
"$ns run"
    (file "template1.tcl" line 208)


From mkenny@mekb2.sps.mot.com  Fri Mar  8 05:45:11 2002
From: mkenny@mekb2.sps.mot.com (Martin McKenny)
Date: Fri Mar  8 05:45:11 2002
Subject: [ns] Multiple agents, One node
Message-ID: <3C88BE4E.4588AFD9@mekb2.sps.mot.com>

--------------147A8142414A33AA9FB32AEA
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: base64

Hi,

If I have multiple agents attached to a single node. If agents A and B
are attached to a single node, how can I use agent A to call agent B
after A receives and acknowledges a packet that has a final destination
on a link serviced by agent B. Each agent is responsible for maintaining
the variables associated with it's single link. When an agent receives a
packet, it should call another agent to forward the packet to the
destination.  I am using command line arguments to call the agents from
the OTcl code.

Am I approaching this in the correct way? Does know of any examples?

MMK



--------------147A8142414A33AA9FB32AEA
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit



Hi,

If I have multiple agents attached to a single node. If agents A and
B are attached to a single node, how can I use agent A to call agent B
after A receives and acknowledges a packet that has a final destination
on a link serviced by agent B. Each agent is responsible for maintaining
the variables associated with it's single link. When an agent receives
a packet, it should call another agent to forward the packet to the destination. 
I am using command line arguments to call the agents from the OTcl code.

Am I approaching this in the correct way? Does know of any examples? 

MMK




 

--------------147A8142414A33AA9FB32AEA--


From kopi@stanford.edu  Fri Mar  8 06:00:12 2002
From: kopi@stanford.edu (Milind Dilip Kopikare)
Date: Fri Mar  8 06:00:12 2002
Subject: [ns] Does anyone knows ? - Counting packets sent by nodes
In-Reply-To: 
Message-ID: 

I had and still have a similar problem. I am trying to maintain a queue at
each node in a wireless network. I can offer a few solutions.

1>Maintining list of how many packets went out: In the mac.cc (or
mac-802_11.cc in my case) code, whenever a packet is sent using the recv()
function, I decrement a counter i've created by one.

2> To track packets coming in: Here I am in a problem. I thought I'd
increment my counter whenever a packet comes into the mac layer from teh
link layer of a mobile node. But it dosent seem to work.

3> Does anyone know how one can maintain a queue for a mobile node? The
problem is that the normal queue features, I believe, work only when a
queue is attached to a link. But in case of a mobile node, there is no
link.

Thanks,
milind


On Fri, 8 Mar 2002, Asaf Shabtai wrote:

>
> Hi,
> I'm tring to add a variable in the node, so each time a packet is sent by
> the node to the next node te variable is change (for example let's say that
> the each time the node sends a packet its variable is incremented by 7).
> Does anybody have an idea how to implenet that ?
> Thank U
> Asaf
>
> _________________________________________________________________
> MSN Photos is the easiest way to share and print your photos:
> http://photos.msn.com/support/worldwide.aspx
>


From jg35@oak.njit.edu  Fri Mar  8 07:05:03 2002
From: jg35@oak.njit.edu (Jeu George)
Date: Fri Mar  8 07:05:03 2002
Subject: [ns] bad manual
In-Reply-To: <20020308130648.3dc23d02.gdemarco@unisa.it>
Message-ID: 

what is in the manual that makes it bad?? or worst??

On Fri, 8 Mar 2002, giuseppe de marco wrote:

> 
> THe ns-manual is the worst manual I never read in my life
> 
> 
> 
> -- 
> Giuseppe De Marco
> Department of Electrical and Information Engineering (DIIIE)
> University of Salerno
> tel . +39 089 964012
> 
> 


From tkt_astu@cs.Helsinki.FI  Fri Mar  8 07:15:01 2002
From: tkt_astu@cs.Helsinki.FI (Davide Astuti)
Date: Fri Mar  8 07:15:01 2002
Subject: [ns] Generating Gaussian Random Numbers under NS
Message-ID: <3C88D462.25B78775@cs.helsinki.fi>

Does anyone know how generating Gaussian Random Numbers under NS?

At http://www.taygeta.com/random/gaussian.html

there is a transformation to get two new independent random numbers
which have a Gaussian distribution with zero mean and a standard
deviation of one. 

How can I change the mean and the standard deviation values?

Thanks, Davide


From billtj@Glue.umd.edu  Fri Mar  8 08:35:02 2002
From: billtj@Glue.umd.edu (William Djaja Tjokroaminata)
Date: Fri Mar  8 08:35:02 2002
Subject: [ns] bad manual
Message-ID: 

Probably it is not the worst manual.  However, there are things that can
be improved.  To me, it is not about the details, but about the selection
and/or order in the manual.

For example, the first time I used ns, I followed the Marc Greis's
tutorial.  Things are fine until I got to Section VII which gives a
complicated example on how to create a custom ping application.  Reading
the corresponding ns manual Section 3 (OTcl Linkage), which contains a lot
of details, it was still not clear why we do certain things.  In other
words, the learning curve is pretty steep.

I have used C++ and Tcl/Tk before.  The following things are the order
that I would like to see explained in the ns manual or tutorial:

1) How to Create Your Own Agent.
This includes how to write the C++ code, the Tcl code, binding between the
two, and run the simulation.  The agent does not need to do anything in
this case; we just need it to be able to compile and run.  Also no
variable binding between C++ and Tcl needs to be covered here.

2) How to Use Timers
In this case the agent will have a timer that is, say, exponentially
distributed and it will continuously print, say, "hello" every now and
then.  This section will cover to some extent how to use the random number
generator in ns.

3) How to Send and Receive Packet
In this case the relation between agent and node will be explained.  The
ns packet and link concepts are also described.  With the knowledge of 1),
2) and 3), hopefully a novice should be able to simulate a simple M/M/1
queue.

4) How to Generate Output
This will include nam and traces, and how to do post-processing on the
trace output.  With these, hopefully a person can verify that the output
delay distribution in the M/M/1 queue is correct.

5) How Routing is Modeled in ns
In all the previous steps, we can involve only two nodes.  With multiple
nodes and multiple paths, the (default) routing should be explained.

I think 1), 2), 3), 4) and 5) form the foundation of any network
simulation.  Armed with this knowledge, hopefully a person can simulate
any network from scratch.  Of course, to save time he or she will use the
available library, in which case the rest of ns manual comes into play.

Regards,

Bill

P.S. For my own Bluetooth network simulation, I created my own simulation
entirely from scratch, where I followed the steps above, because those are
all the things that I need, and nothing else.

======================================================================
At Friday, 8 March 2002, giuseppe de marco  wrote:

>THe ns-manual is the worst manual I never read in my life
>
>-- 
>Giuseppe De Marco
>Department of Electrical and Information Engineering (DIIIE)
>University of Salerno
>tel . +39 089 964012
>


From francesc@lsihp19.epfl.ch  Fri Mar  8 09:25:02 2002
From: francesc@lsihp19.epfl.ch (Francesc Font (EPFL))
Date: Fri Mar  8 09:25:02 2002
Subject: [ns] Pragmatic Generic Multicast
References: 
Message-ID: <3C88F2F2.7BD3E16E@lsihp19.epfl.ch>

Have a look on the internet draft:

http://www.talarian.com/products/smartpgm/draft-speakman-pgm-spec-06.txt


Arijit Ganguly wrote:

> Hello friends
> Can you please let me know some links where I can get some useful papers
> on Pragmatic Generic Multicast (PGM)?
> Thanking you.
>
> Regards
> Arijit
>
> --
> ------------------------------------------------------------------------------
>                           Arijit Ganguly
>                         Final Year Student
>             Department of Computer Science & Engineering
>               Indian Institute of Technology, Guwahati
>                 Guwahati - 781039, District Kamrup
>                           Assam, India
>           ------------------------------------------------
>                 Phone : +00-91-0361-690763 (room #203)
>                 Fax   : +91 361 690762
>                 email : ariganguly@yahoo.co.in
>                 CC    : ariganguly@hotmail.com
>           ------------------------------------------------


From svarma@cedar.Buffalo.EDU  Fri Mar  8 09:30:01 2002
From: svarma@cedar.Buffalo.EDU (Saket Varma)
Date: Fri Mar  8 09:30:01 2002
Subject: [ns] Object type from Object ID
Message-ID: 

Hi all!

I have created a network with 2 nodes, n0 and n1.

i have attached a constant bit rate UDP agent at n0 and a null agent at
n1 and sending packets from n0 to n1.

How do i know which is the classifier object that is being used to
"classify()" and then recv object?

I can trace the classify function used by inserting printf statements, but
is there any other elegant way of knowing the object type classify
statements being used?

i tried "[[$node entry] name]", But name does not exist for the $node
entry... are there anything on these lines..


A more general question:
How can we know what is the object *type* if we know the *object id* ???


thanks in advance..


Saket




From xuanc@ISI.EDU  Fri Mar  8 09:40:02 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Fri Mar  8 09:40:02 2002
Subject: [ns] Interfacing tcl script with c/c++ (fwd)
Message-ID: 

---------- Forwarded message ----------
Date: Fri, 8 Mar 2002 03:57:41 -0800 (PST)
From: ashokan J 
Subject: Interfacing tcl script with c/c++

Hi 

I'm working on ns-2 simulator, we are able to use your
sample script to perform ns simulator with nam and
able to visualize the functionality. Now we wanted to
have our own application to be attached with UDP or
TCP for simulation.

Can you let me know how to interface tcl scripts with
either c or c++ code or how to call c/c++ functions
from tcl scripts.

Thanks in Advance
Ashok

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From xuanc@ISI.EDU  Fri Mar  8 09:45:05 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Fri Mar  8 09:45:05 2002
Subject: [ns] help (fwd)
Message-ID: 

---------- Forwarded message ----------
Date: Fri,  8 Mar 2002 17:25:56 +0100
From: "[utf-8] stream1@voila.fr" 
Subject: help


i began installing ns-allinone-2.1b8 on sun solaris5.8. but while installing ns-2.1b8a the folowing message appears:

simulator.o: In function `Simulator::populate_hier_classifiers()':
simulator.o(.text+0xa10): undefined reference to `RouteLogic::elements_in_level(int*, int)'
collect2: ld returned 1 exit status
*** Error code 1
make: Fatal error: Command failed for target `ns'

i would be very pleased if you send me a solution to that problem.
thanks in advance.

____________________________________________________________
Faites un voeu et puis Voila ! www.voila.fr 
Avec Voila Mail, consultez vos e-mails sur votre mobile Wap. 


From xuanc@ISI.EDU  Fri Mar  8 09:55:09 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Fri Mar  8 09:55:09 2002
Subject: [ns] Interfacing tcl script with c/c++ (fwd)
In-Reply-To: 
Message-ID: 

You may find the first a few chapter in ns manual are helpful. 

On Fri, 8 Mar 2002, Xuan Chen wrote:

> 
> ---------- Forwarded message ----------
> Date: Fri, 8 Mar 2002 03:57:41 -0800 (PST)
> From: ashokan J 
> Subject: Interfacing tcl script with c/c++
> 
> Hi 
> 
> I'm working on ns-2 simulator, we are able to use your
> sample script to perform ns simulator with nam and
> able to visualize the functionality. Now we wanted to
> have our own application to be attached with UDP or
> TCP for simulation.
> 
> Can you let me know how to interface tcl scripts with
> either c or c++ code or how to call c/c++ functions
> from tcl scripts.
> 
> Thanks in Advance
> Ashok
> 
> __________________________________________________
> Do You Yahoo!?
> Try FREE Yahoo! Mail - the world's greatest free email!
> http://mail.yahoo.com/
> 

-- 
Xuan Chen
USC/ISI


From chakra10@msu.edu  Fri Mar  8 10:15:05 2002
From: chakra10@msu.edu (Gautam - Chakrabarti)
Date: Fri Mar  8 10:15:05 2002
Subject: [ns] help (fwd)
References: 
Message-ID: <3C88FE97.1187E9ED@msu.edu>

Xuan Chen wrote:
> 
> ---------- Forwarded message ----------
> Date: Fri,  8 Mar 2002 17:25:56 +0100
> From: "[utf-8] stream1@voila.fr" 
> Subject: help
> 
> i began installing ns-allinone-2.1b8 on sun solaris5.8. but while installing ns-2.1b8a the folowing message appears:
> 
> simulator.o: In function `Simulator::populate_hier_classifiers()':
> simulator.o(.text+0xa10): undefined reference to `RouteLogic::elements_in_level(int*, int)'

U must have seen RouteLogic::elements_in_level(int*, int) is defined as
inline, remove "inline" from both the declaration and the definition.

hope this helps,
gautam


> collect2: ld returned 1 exit status
> *** Error code 1
> make: Fatal error: Command failed for target `ns'
> 
> i would be very pleased if you send me a solution to that problem.
> thanks in advance.
> 
> ____________________________________________________________
> Faites un voeu et puis Voila ! www.voila.fr
> Avec Voila Mail, consultez vos e-mails sur votre mobile Wap.

-- 
***************************************************************
Gautam Chakrabarti
MS Student
Computer Sc. and Engg Department
Michigan State University

Email : chakra10@msu.edu
Home Page : http://www.cse.msu.edu/~chakra10

Why is there so much month left at the end of the money?
		--- John Barrymore
***************************************************************


From hai22@it.canterbury.ac.nz  Fri Mar  8 10:35:05 2002
From: hai22@it.canterbury.ac.nz (hai22)
Date: Fri Mar  8 10:35:05 2002
Subject: [ns] bad manual
References: 
Message-ID: <3C89033B.F26E12C9@student.canterbury.ac.nz>

By the way who advised giuseppe de marco  to try to
read it !!!!!! 
Manual is not so bad i think.
a.
===
Lloyd Wood wrote:
> 
> On Fri, 8 Mar 2002, giuseppe de marco wrote:
> 
> > THe ns-manual is the worst manual I never read in my life
> 
> Since you've never read it, how do you know?
> 
> L.
> 
> PGP


From netsim@gmx.at  Fri Mar  8 10:45:03 2002
From: netsim@gmx.at (netsim@gmx.at)
Date: Fri Mar  8 10:45:03 2002
Subject: [ns] NSE: Scheduler going backwards in time
References: <28271.1015612955@www39.gmx.net>
Message-ID: <31135.1015613069@www39.gmx.net>


 One more question concerning this topic because I got this messages even
 with the simple ping demo on a PIII-800.
 
 There are to different messages:
 
 1. a warning message: scheduler going backwards in time and
 2. a error message: time limit eceeded, scheduler going backwards in time
 
 Does the first message (warning) mean that the scheduler puts the packets
 onto the live network but with a slight delay and that there is no packet
 loss?
 
 Does the second message indicate a packet loss?
 
 martin
 
 
> > 
> > It indicates that the emulator cannot keep up with real time. Using a
> > faster computer will elevate this problem.
> > 
> > The results are not affeted drastically from this error.
> > Though you most prbably will be experiencing high packet losses  in bpf.
> > 
> > alefiya
> > On Mon, 4 Mar 2002 netsim@gmx.at wrote:
> > 
> > >
> > > Hi!
> > >
> > > I am using the emulation mode in ns-2 called nse and ecountered the
> > > following problem:
> > >
> > > scheduler going backwards in time
> > >
> > > What is the reason for this error and how severe is it? Are all my
> trace
> > > files useless when getting this message?
> > >
> > > Thanks a lot.
> > >
> > > Martin Joerg
> > >
> > > --
> > > GMX - Die Kommunikationsplattform im Internet.
> > > http://www.gmx.net
> > >
> > >
> > 
> 
> -- 
> GMX - Die Kommunikationsplattform im Internet.
> http://www.gmx.net
> 
> 

-- 
GMX - Die Kommunikationsplattform im Internet.
http://www.gmx.net


From flaw@vt.edu  Fri Mar  8 12:00:01 2002
From: flaw@vt.edu (Raymond Law)
Date: Fri Mar  8 12:00:01 2002
Subject: [ns] mns v2.0
Message-ID: <3C8DFA7F@zathras>

I have two questions regarding mns v2.0.

In '$ns at 0.9 "$LSRmpls2 bind-flow-erlsp 9 -1 3000"', what does -1 mean?  
Does it mean all flows, or is -1 the default fid value?

What is the meanings of LIBptr, SLIBptr, and aPATHptr in teh ERB table?

Thanks.
Ray,



From tmaneesh@cedt.iisc.ernet.in  Fri Mar  8 12:20:02 2002
From: tmaneesh@cedt.iisc.ernet.in (Maneesh Tewari)
Date: Fri Mar  8 12:20:02 2002
Subject: [ns] Exponential traffic on wireless
Message-ID: 

hello ns-experts,

i'm facing some problems in passing exponential ON-OFF traffic on wireless
while trying to model Voice over wireless, let me know if anyone is working
in same thing ??

thanks.

                    
regards
maneesh


From yacolic@hotmail.com  Fri Mar  8 12:50:01 2002
From: yacolic@hotmail.com (EL IDRISSI MEHDI)
Date: Fri Mar  8 12:50:01 2002
Subject: [ns] bad manual
References:  <3C89033B.F26E12C9@student.canterbury.ac.nz>
Message-ID: 

hi all,

I have just read those few mails concerning the manual and just wanted to
add this:

When u begin to understand how does ns work and where to get the
information, ok the manual is useful, but for ns beginners it's a bit tough.
The Marc greig's tutorial should really be included in the manual in some
way. At least it presents the first steps and how to make basic scripts to
get started. In the manual there are only parts of scripts concerning the
topic of each chapter. There should be an extra chapter between the first
chapters which explain split objects and the following ones which
immediately start to describe specific classes ( link, nodes...)
In fact i have to say that the best manual for ns is this mailing list.

well i just wanted to say that to be a little fair with giuseppe who was a
bit extreme in his judgment.
bye

M.
----- Original Message -----
From: "hai22" 
To: 
Sent: Friday, March 08, 2002 1:30 PM
Subject: Re: [ns] bad manual


>
> By the way who advised giuseppe de marco  to try to
> read it !!!!!!
> Manual is not so bad i think.
> a.
> ===
> Lloyd Wood wrote:
> >
> > On Fri, 8 Mar 2002, giuseppe de marco wrote:
> >
> > > THe ns-manual is the worst manual I never read in my life
> >
> > Since you've never read it, how do you know?
> >
> > L.
> >
> > PGP
>
>


From c16917@yahoo.com  Fri Mar  8 14:55:02 2002
From: c16917@yahoo.com (Lisa Wu)
Date: Fri Mar  8 14:55:02 2002
Subject: [ns] Question about MAC/TDMA
Message-ID: <20020308225246.49123.qmail@web12403.mail.yahoo.com>

Hi, there, 

I have a few questions about Mac/Tdma. According to
"NS Manual" 16.1.4, it said the preamble is divided in
subslot for each node to write data. But after I view
the code(mac-tdma.cc), it seems that the preamble is
not subslotted. All nodes write its destination
address in one slot at the same time. Will this cause
contention problem? 
Another question is who will call function
SlotTdmaTimer::handle()?


Thanks,

Lisa
 

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From xuanc@ISI.EDU  Fri Mar  8 15:35:02 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Fri Mar  8 15:35:02 2002
Subject: [ns] Question about MAC/TDMA
In-Reply-To: <20020308225246.49123.qmail@web12403.mail.yahoo.com>
Message-ID: 

On Fri, 8 Mar 2002, Lisa Wu wrote:

> 
> Hi, there, 
> 
> I have a few questions about Mac/Tdma. According to
> "NS Manual" 16.1.4, it said the preamble is divided in
> subslot for each node to write data. But after I view
> the code(mac-tdma.cc), it seems that the preamble is
> not subslotted. All nodes write its destination
> address in one slot at the same time. Will this cause
> contention problem? 
The preamble is an array and each node has a dedicated sub-slot. So, I 
don't think there will be any congestion. Does the description look 
confused? Please let me know. Thanks.

> Another question is who will call function
> SlotTdmaTimer::handle()?
It is called by ns when the corresponding timer expires. I think you can 
find details in ns manual.

Hope it helps,
-chen

> 
> 
> Thanks,
> 
> Lisa
>  
> 
> __________________________________________________
> Do You Yahoo!?
> Try FREE Yahoo! Mail - the world's greatest free email!
> http://mail.yahoo.com/
> 

-- 
Xuan Chen
USC/ISI


From JZhao_NS@hotmail.com  Fri Mar  8 18:20:02 2002
From: JZhao_NS@hotmail.com (Jun Zhao)
Date: Fri Mar  8 18:20:02 2002
Subject: [ns] bad manual
References: 
Message-ID: 

William's suggestion is very constructive. The questions listed in the
archives may be an good entry for improvement.

1)How to creat your own agent(the architecture, variable bounding and etc);
2) Timers, random generators (shall we set the seed each time or it is done
automatically); 3) Besides the things about sending and receiving packets, I
think more description is needed for the link and bandwidth concept in
NS(there are many questions due to misunderstanding the bandwidth and/or
delay in NS, especially how to set the bandwidth of IEEE 802.11. It may be
one of the questions mentioned most frequently)/ 4) How to trace your own
variables and postprocessing (I think some scripts are OK. Since many guys
asked the same question, why not put some examples in the manual).


----- Original Message -----
From: "William Djaja Tjokroaminata" 
To: 
Sent: Saturday, March 09, 2002 12:32 AM
Subject: Re: [ns] bad manual


>
>
> Probably it is not the worst manual.  However, there are things that can
> be improved.  To me, it is not about the details, but about the selection
> and/or order in the manual.
>
> For example, the first time I used ns, I followed the Marc Greis's
> tutorial.  Things are fine until I got to Section VII which gives a
> complicated example on how to create a custom ping application.  Reading
> the corresponding ns manual Section 3 (OTcl Linkage), which contains a lot
> of details, it was still not clear why we do certain things.  In other
> words, the learning curve is pretty steep.
>
> I have used C++ and Tcl/Tk before.  The following things are the order
> that I would like to see explained in the ns manual or tutorial:
>
> 1) How to Create Your Own Agent.
> This includes how to write the C++ code, the Tcl code, binding between the
> two, and run the simulation.  The agent does not need to do anything in
> this case; we just need it to be able to compile and run.  Also no
> variable binding between C++ and Tcl needs to be covered here.
>
> 2) How to Use Timers
> In this case the agent will have a timer that is, say, exponentially
> distributed and it will continuously print, say, "hello" every now and
> then.  This section will cover to some extent how to use the random number
> generator in ns.
>
> 3) How to Send and Receive Packet
> In this case the relation between agent and node will be explained.  The
> ns packet and link concepts are also described.  With the knowledge of 1),
> 2) and 3), hopefully a novice should be able to simulate a simple M/M/1
> queue.
>
> 4) How to Generate Output
> This will include nam and traces, and how to do post-processing on the
> trace output.  With these, hopefully a person can verify that the output
> delay distribution in the M/M/1 queue is correct.
>
> 5) How Routing is Modeled in ns
> In all the previous steps, we can involve only two nodes.  With multiple
> nodes and multiple paths, the (default) routing should be explained.
>
> I think 1), 2), 3), 4) and 5) form the foundation of any network
> simulation.  Armed with this knowledge, hopefully a person can simulate
> any network from scratch.  Of course, to save time he or she will use the
> available library, in which case the rest of ns manual comes into play.
>
> Regards,
>
> Bill
>
> P.S. For my own Bluetooth network simulation, I created my own simulation
> entirely from scratch, where I followed the steps above, because those are
> all the things that I need, and nothing else.
>
> ======================================================================
> At Friday, 8 March 2002, giuseppe de marco  wrote:
>
> >THe ns-manual is the worst manual I never read in my life
> >
> >--
> >Giuseppe De Marco
> >Department of Electrical and Information Engineering (DIIIE)
> >University of Salerno
> >tel . +39 089 964012
> >
>
>


From giuliastro@tin.it  Sat Mar  9 05:15:16 2002
From: giuliastro@tin.it (Giulio Ardoino)
Date: Sat Mar  9 05:15:16 2002
Subject: [ns] bad manual
In-Reply-To: 
References: <3C89033B.F26E12C9@student.canterbury.ac.nz> 
Message-ID: <20020309100132.AB86.GIULIASTRO@tin.it>

EL IDRISSI MEHDI wrote:

> When u begin to understand how does ns work and where to get the
> information, ok the manual is useful, but for ns beginners it's a bit tough.
> The Marc greig's tutorial should really be included in the manual in some
> way.

I completely agree with this. The best way to learn how to use NS is
Marc Greig's tutorial together with the NS' search page (a great thing!).

This doesn't mean that who wrote that manual did something completely
unuseful, but, in my opinion, it could be arranged a little better,
including more properties for ns' objects and classes and pointing to
the web for those things that change more frequently, they don't need to
be all in the manual.

Giulio




From flaw@vt.edu  Sat Mar  9 05:15:33 2002
From: flaw@vt.edu (Raymond Law)
Date: Sat Mar  9 05:15:33 2002
Subject: [ns] Error when running mns v2.0 example script test-crlsp.tcl
Message-ID: <3C8A5BB8@zathras>

There is an error when running mns v2.0 example script.  Based on the trace 
file, it happens just after 11.044872; based on the script, between $ns at 
11.0  "$RT2 start" and $ns at 30.0  "$RT2 stop"


0.14188800000000001 nodeid=1 : lspid=1000
0.14238400000000001 nodeid=1 : lspid=1100
0.14288000000000001 nodeid=1 : lspid=1200
10.041888 nodeid=1 : lspid=1300
Attempting to cancel timer at 0094C408 which is not scheduled

abnormal program termination


Any help will be appreciated.

Thanks.
Ray,



From madhan80@eth.net  Sat Mar  9 07:40:02 2002
From: madhan80@eth.net (Madhan Mohan)
Date: Sat Mar  9 07:40:02 2002
Subject: [ns] Debugging segmentation fault
Message-ID: <001001c1c7e9$82273700$d69509ca@madhan>

 Hi

 I get this segmentation fault error. But I find it difficult to find the
 error in my code.  Is there a method to find the part of the code which
 gives this error or the function that is being accessed when this error
 occurs. I normally use printf statements. But this method is not that
useful
 for large codes.

 Thank You
 Madhan




From arshahid@yahoo.com  Sat Mar  9 09:30:11 2002
From: arshahid@yahoo.com (Ahmad R Shahid)
Date: Sat Mar  9 09:30:11 2002
Subject: Fwd: [ns] $ns_ node-config
Message-ID: <20020309172659.58618.qmail@web10503.mail.yahoo.com>

--0-734739575-1015694819=:56825
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

I asked this question a few days back but didn't get a
reply. I am putting it again.

Ahmad


__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/
--0-734739575-1015694819=:56825
Content-Type: message/rfc822

X-Apparently-To: arshahid@yahoo.com via web10507; 08 Mar 2002 05:16:37 -0800 (PST)
Return-Path: 
X-Track: 1: 40
Return-Path: 
Received: from gamma.isi.edu (128.9.144.145)
  by mta459.mail.yahoo.com with SMTP; 08 Mar 2002 05:16:30 -0800 (PST)
Received: from gamma.isi.edu (localhost [127.0.0.1])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g28DG2R05371;
	Fri, 8 Mar 2002 05:16:02 -0800 (PST)
Received: (from daemon@localhost)
	by gamma.isi.edu (8.11.6/8.11.2) id g287LUD28263
	for ns-users-mailman@isi.edu; Thu, 7 Mar 2002 23:21:30 -0800 (PST)
Received: from tnt.isi.edu (tnt.isi.edu [128.9.128.128])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g287LUR28251
	for ; Thu, 7 Mar 2002 23:21:30 -0800 (PST)
Received: from web10505.mail.yahoo.com (web10505.mail.yahoo.com [216.136.130.155])
	by tnt.isi.edu (8.11.6/8.11.2) with SMTP id g287LUt24754
	for ; Thu, 7 Mar 2002 23:21:30 -0800 (PST)
Received: from [35.9.50.165] by web10505.mail.yahoo.com via HTTP; Thu, 07 Mar 2002 23:21:29 PST
From: Ahmad R Shahid 
To: ns-users@ISI.EDU
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: [ns] $ns_ node-config
Sender: ns-users-admin@ISI.EDU
Errors-To: ns-users-admin@ISI.EDU
X-BeenThere: ns-users@isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 
X-Original-Date: Thu, 7 Mar 2002 23:21:29 -0800 (PST)
Date: Thu, 7 Mar 2002 23:21:29 -0800 (PST)
Content-Length: 575


Hi all,

in my tcl file i have written the following lines:

for {set j 0} {$j < 5} {incr j} {
    set node_($j) [ $opt(rp)-create-mobile-node $j
[lindex $temp $j] ]
    $node_($j) base-station [AddrParams addr2id
[$BS(0) node-addr]]
}

now i want to add/change certain features in the
mobile nodes created by [ $opt(rp)-create-mobile-node
$j [lindex $temp $j] ] command. how do i do that using
$ns_ node_config command?

Ahmad

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/

--0-734739575-1015694819=:56825--


From ratancha@uiuc.edu  Sat Mar  9 09:50:03 2002
From: ratancha@uiuc.edu (Prashant Ratanchandani)
Date: Sat Mar  9 09:50:03 2002
Subject: [ns] Exponential traffic on wireless
References: 
Message-ID: <3C8A4B17.74E82994@uiuc.edu>

Maneesh Tewari wrote:
> 
> hello ns-experts,
> 
> i'm facing some problems in passing exponential ON-OFF traffic on wireless
> while trying to model Voice over wireless, let me know if anyone is working
> in same thing ??
> 
> thanks.
> 
> 
> regards
> maneesh
The problem that I faced while doing this is that the CMU Trace object
does not understand the packet type PT_EXP. A simple addition in the
case statements that distinguish between packet types in cmu-trace.cc
should do the trick.

Prashant
-- 
Prashant Ratanchandani							

"If I were two-faced, would I be wearing this one?" Abraham Lincoln


From tmaneesh@cedt.iisc.ernet.in  Sat Mar  9 10:00:01 2002
From: tmaneesh@cedt.iisc.ernet.in (Maneesh Tewari)
Date: Sat Mar  9 10:00:01 2002
Subject: [ns] Exponential traffic on wireless
Message-ID: 

hello Prasant,

thanks for u'r help, but its not so simple, in the case statement u've to
add the function void format_exp(Packet *p, int offset ), now this function
u've to write in the cmu-trace.cc and then u've to make an entry
corresponding to it in the cmu-trace.h file

now my problem is that what should be written in the body of the function

CMUTrace::format_exp(Packet *p, int offset) in the file cmu-trace.cc

i hope now u can get my problem.

thanks for u'r effort to help me.

bye

                             ---maneesh



-----Original Message-----
From: Prashant Ratanchandani [mailto:ratancha@uiuc.edu]
Sent: Saturday, March 09, 2002 11:19 PM
To: Maneesh Tewari
Cc: ns-users@ISI.EDU
Subject: Re: [ns] Exponential traffic on wireless


Maneesh Tewari wrote:
> 
> hello ns-experts,
> 
> i'm facing some problems in passing exponential ON-OFF traffic on wireless
> while trying to model Voice over wireless, let me know if anyone is
working
> in same thing ??
> 
> thanks.
> 
> 
> regards
> maneesh
The problem that I faced while doing this is that the CMU Trace object
does not understand the packet type PT_EXP. A simple addition in the
case statements that distinguish between packet types in cmu-trace.cc
should do the trick.

Prashant
-- 
Prashant Ratanchandani							

"If I were two-faced, would I be wearing this one?" Abraham Lincoln


From rbj2@oak.njit.edu  Sat Mar  9 11:20:01 2002
From: rbj2@oak.njit.edu (rahul b jain cs student)
Date: Sat Mar  9 11:20:01 2002
Subject: [ns] Problems while installing NAM
In-Reply-To: 
Message-ID: 

Hi,

I have been trying to install ns on windows. i was trying to install nam
and i got the following error.

        mkdir gen
        rm -f gen\\version.c
'rm' is not recognized as an internal or external command,
operable program or batch file.
NMAKE : fatal error U1077: 'rm' : return code '0x1'

in addition to this, after i had run the nmake command, there were lot
of warnings. Would these warnings affect the running of the working of ns
and nam later on ? 

Can someone tell me what is the solution to this problem. Plz reply,

Thanks,
Rahul.


From mportnoi@ieee.org  Sat Mar  9 11:50:01 2002
From: mportnoi@ieee.org (Marcos "Locksmith" Portnoi)
Date: Sat Mar  9 11:50:01 2002
Subject: [ns] bad manual
References: <3C89033B.F26E12C9@student.canterbury.ac.nz>  <20020309100132.AB86.GIULIASTRO@tin.it>
Message-ID: <004701c1c7a2$db304e50$83f2dfc8@nebula>

Just a comment on the manual.  The pdf version is never paginated right!  :)
Just compare the page numbers in the Contents with the page numbers the
chapters actually begin...

Portnoi

----- Original Message -----
From: "Giulio Ardoino" 
To: 
Sent: Saturday, March 09, 2002 6:09 AM
Subject: Re: [ns] bad manual


>
> EL IDRISSI MEHDI wrote:
>
> > When u begin to understand how does ns work and where to get the
> > information, ok the manual is useful, but for ns beginners it's a bit
tough.
> > The Marc greig's tutorial should really be included in the manual in
some
> > way.
>
> I completely agree with this. The best way to learn how to use NS is
> Marc Greig's tutorial together with the NS' search page (a great thing!).
>
> This doesn't mean that who wrote that manual did something completely
> unuseful, but, in my opinion, it could be arranged a little better,
> including more properties for ns' objects and classes and pointing to
> the web for those things that change more frequently, they don't need to
> be all in the manual.
>
> Giulio
>
>
>
>



From c16917@yahoo.com  Sat Mar  9 12:35:02 2002
From: c16917@yahoo.com (Lisa Wu)
Date: Sat Mar  9 12:35:02 2002
Subject: [ns] Question about MAC/TDMA
In-Reply-To: 
Message-ID: <20020309203102.91017.qmail@web12407.mail.yahoo.com>

Hi, there,

Thank you very much. It is very helpful.
But I still have a question about the function
SlotTdmaTimer::handle(). If it is called by the
corresponding timer expiration, why do not use the
expire() function. What is the difference to use
handle() and expire() at this moment?

Thanks, 
Lisa


--- Xuan Chen  wrote:
> On Fri, 8 Mar 2002, Lisa Wu wrote:
> 
> > 
> > Hi, there, 
> > 
> > I have a few questions about Mac/Tdma. According
> to
> > "NS Manual" 16.1.4, it said the preamble is
> divided in
> > subslot for each node to write data. But after I
> view
> > the code(mac-tdma.cc), it seems that the preamble
> is
> > not subslotted. All nodes write its destination
> > address in one slot at the same time. Will this
> cause
> > contention problem? 
> The preamble is an array and each node has a
> dedicated sub-slot. So, I 
> don't think there will be any congestion. Does the
> description look 
> confused? Please let me know. Thanks.
> 
> > Another question is who will call function
> > SlotTdmaTimer::handle()?
> It is called by ns when the corresponding timer
> expires. I think you can 
> find details in ns manual.
> 
> Hope it helps,
> -chen
> 
> > 
> > 
> > Thanks,
> > 
> > Lisa
> >  
> > 
> > __________________________________________________
> > Do You Yahoo!?
> > Try FREE Yahoo! Mail - the world's greatest free
> email!
> > http://mail.yahoo.com/
> > 
> 
> -- 
> Xuan Chen
> USC/ISI
> 


__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From chakra10@msu.edu  Sat Mar  9 14:40:01 2002
From: chakra10@msu.edu (Gautam - Chakrabarti)
Date: Sat Mar  9 14:40:01 2002
Subject: [ns] Debugging segmentation fault
References: <001001c1c7e9$82273700$d69509ca@madhan>
Message-ID: <3C8A8EE5.7F8C8262@msu.edu>

when you use printf to find seg-faults, i hope u end the printf with a
"\n", for this you can use something like gdb to find out where its
giving the error.

hth,
gautam

Madhan Mohan wrote:
> 
>  Hi
> 
>  I get this segmentation fault error. But I find it difficult to find the
>  error in my code.  Is there a method to find the part of the code which
>  gives this error or the function that is being accessed when this error
>  occurs. I normally use printf statements. But this method is not that
> useful
>  for large codes.
> 
>  Thank You
>  Madhan

-- 
***************************************************************
Gautam Chakrabarti
MS Student
Computer Sc. and Engg Department
Michigan State University

Email : chakra10@msu.edu
Home Page : http://www.cse.msu.edu/~chakra10

Why is there so much month left at the end of the money?
		--- John Barrymore
***************************************************************


From cs152ct@cs.ucr.edu  Sat Mar  9 15:30:01 2002
From: cs152ct@cs.ucr.edu (cs152 Student)
Date: Sat Mar  9 15:30:01 2002
Subject: [ns] Trace converter - Berkeley website invalid?
Message-ID: 

Hi,

The NS manual said there is a converter that converts a real trace to NS
format trace and it is located at
http://mash.cs.berkeley.edu/dist/vint/webcache-trace-conv.tar.gz.

However, the UCB website is not accessible. If anyone knows what the new
website is for this converter, please let me know.

NOTE: I alread tried using the Traffic Trace converter from
www.research.att.com/!breslau/vint/trace.html but couldn't get it to do
anything.

Thank you, Stephanie




From dhiman@cs.bu.edu  Sat Mar  9 18:25:02 2002
From: dhiman@cs.bu.edu (Dhiman Barman)
Date: Sat Mar  9 18:25:02 2002
Subject: [ns] perfect TCP
Message-ID: <20020309212010.A12011@cs.bu.edu>

Hi, 
  I want to see the performance of the perfect TCP which will 
allow the sender to know the exact reason of a packet loss, congestion loss
or random loss.
  How can I implement that ? Has anyone already done that? Any suggestions ? 


Thanks,
Dhiman 


From flaw@vt.edu  Sun Mar 10 05:15:49 2002
From: flaw@vt.edu (Raymond Law)
Date: Sun Mar 10 05:15:49 2002
Subject: [ns] Error - Attempting to cancel timer
Message-ID: <3C8B6443@zathras>

When I ran the sample test-crlsp.tcl script for mns v2.0, the error 
"Attempting to cancel timer at 0094C408 which is not scheduled" still occurs 
even after the following patch is applied to trafgen.cc.

TrafficGenerator::TrafficGenerator() : nextPkttime_(-1), running_(0),
timer_(this)

All the starts are scheduled before the stops in the tcl script.  Any 
suggestions?

Thanks.
Ray,



From cwchiang@eagle.ee.ntu.edu.tw  Sun Mar 10 05:16:00 2002
From: cwchiang@eagle.ee.ntu.edu.tw (Chiang Chih-Wei)
Date: Sun Mar 10 05:16:00 2002
Subject: [ns] Debugging segmentation fault
In-Reply-To: <3C8A8EE5.7F8C8262@msu.edu> from "Gautam - Chakrabarti" at Mar 9, 2 05:38:29 pm
Message-ID: <200203100849.QAA04616@eagle.ee.ntu.edu.tw>

For someone who would ask this question,
it's better to suggest ddd rather than gdb or xxgdb.

> 
> 
> when you use printf to find seg-faults, i hope u end the printf with a
> "\n", for this you can use something like gdb to find out where its
> giving the error.
> 
> hth,
> gautam
> 
> Madhan Mohan wrote:
> > 
> >  Hi
> > 
> >  I get this segmentation fault error. But I find it difficult to find the
> >  error in my code.  Is there a method to find the part of the code which
> >  gives this error or the function that is being accessed when this error
> >  occurs. I normally use printf statements. But this method is not that
> > useful
> >  for large codes.
> > 
> >  Thank You
> >  Madhan
> 
> -- 
> ***************************************************************
> Gautam Chakrabarti
> MS Student
> Computer Sc. and Engg Department
> Michigan State University
> 
> Email : chakra10@msu.edu
> Home Page : http://www.cse.msu.edu/~chakra10
> 
> Why is there so much month left at the end of the money?
> 		--- John Barrymore
> ***************************************************************
> 
> 


From aladakem@kfupm.edu.sa  Sun Mar 10 05:16:13 2002
From: aladakem@kfupm.edu.sa (SALAM AMISU OLUWAKEMI)
Date: Sun Mar 10 05:16:13 2002
Subject: [ns] Setting up FEC in dynamic network
Message-ID: 

Hello,
I want to use ns to simulate a network where the effect of Forward Error
Correction is to be tested. I have tried out some known configuration but
it is not working. 

 Please can anybody help in successfully setting up this 
configuration.

Regards.


Salam-Alada, A.



From badinar@yahoo.com  Sun Mar 10 05:16:25 2002
From: badinar@yahoo.com (Badri Kasthuri)
Date: Sun Mar 10 05:16:25 2002
Subject: [ns] Mac level, RTS-CTS-DS-DATA-ACK  implementation
Message-ID: <20020310114606.81703.qmail@web12208.mail.yahoo.com>

Hi all,
          does anyone have the RTS-CTS-DS-DATA-ACK 
implemented in ns. can somone kindly mail the codes if
anybody has the code. i will be extremely greateful if
someone can help.

thanks,
badri

=====


__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From tasdemir@ata.cs.hun.edu.tr  Sun Mar 10 05:16:37 2002
From: tasdemir@ata.cs.hun.edu.tr (Mustafa Tasdemir)
Date: Sun Mar 10 05:16:37 2002
Subject: [ns] send packet size not equal to received packet size
Message-ID: <000b01c1c832$5e039e60$1fec8cc1@cs.hacettepe.edu.tr>

This is a multi-part message in MIME format.

------=_NextPart_000_0007_01C1C843.217D2C20
Content-Type: multipart/alternative;
	boundary="----=_NextPart_001_0008_01C1C843.217D2C20"


------=_NextPart_001_0008_01C1C843.217D2C20
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

hello,

I implemented a simple application which sends packets with a size of =
100 bytes.
I also calculate the total send and received bytes. And there is always =
100 bytes difference between send and received.

(total send bytes - total received bytes) =3D 100 bytes

I'm sending the C++ and Tcl codes. If you have time, can you have a look =
at it.

bye.

Mustafa TASDEMIR


------=_NextPart_001_0008_01C1C843.217D2C20
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable









hello,


 


I implemented a simple application which sends =
packets=20
with a size of 100 bytes.


I also calculate the total send and received =
bytes. And=20
there is always 100 bytes difference between send and =
received.


 


(total send bytes - total received =
bytes) =3D 100=20
bytes


 


I'm sending the C++ and Tcl codes. If you have =
time, can=20
you have a look at it.


 


bye.


 


Mustafa TASDEMIR


 


------=_NextPart_001_0008_01C1C843.217D2C20--

------=_NextPart_000_0007_01C1C843.217D2C20
Content-Type: application/octet-stream;
	name="SimpleApp.h"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="SimpleApp.h"

#ifndef simple_app_h=0A=
#define simple_app_h=0A=
=0A=
#include "timer-handler.h"=0A=
#include "app.h"=0A=
#include "agent.h"=0A=
=0A=
class SimpleApp;=0A=
=0A=
class SendTimer : public TimerHandler {=0A=
public:=0A=
        SendTimer( SimpleApp *a);   // constructor=0A=
        void expire(Event *e);  // event handler when timer expires=0A=
=0A=
protected:=0A=
        SimpleApp *app;=0A=
=0A=
}; // class SendTimer=0A=
/*----------------------------------------------------------------*/=0A=
class SimpleApp : public Application {=0A=
public:=0A=
        SimpleApp();=0A=
        void recv(int);=0A=
        void sendmsg();=0A=
        void start();=0A=
        void stop();=0A=
=0A=
private:=0A=
        double timeInterval;=0A=
        int running;=0A=
        int packetSize;=0A=
        SendTimer *timer;=0A=
        int tsend, trecv;  // total trasmited and received bytes=0A=
=0A=
}; // class SimpleApp=0A=
=0A=
#endif=0A=

------=_NextPart_000_0007_01C1C843.217D2C20
Content-Type: application/octet-stream;
	name="SimpleApp.cc"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="SimpleApp.cc"

#include "SimpleApp.h"=0A=
=0A=
SendTimer::SendTimer(SimpleApp *a) : TimerHandler() {=0A=
        app =3D a;=0A=
} // constructor=0A=
/*-----------------------------------------------------------*/=0A=
void SendTimer::expire(Event *e) {=0A=
        app->sendmsg();=0A=
} // expire=0A=
/*-----------------------------------------------------------*/=0A=
SimpleApp::SimpleApp() : Application(), tsend(0), trecv(0) {=0A=
        running =3D 0;=0A=
        timer =3D new SendTimer(this);=0A=
        bind("interval", &timeInterval);=0A=
        bind("size", &packetSize);=0A=
} // consructor=0A=
/*-----------------------------------------------------------*/=0A=
void SimpleApp::start() {=0A=
        packetSize =3D 100;=0A=
        timeInterval =3D 0.02;=0A=
        running =3D 1;=0A=
        sendmsg();=0A=
}=0A=
/*-----------------------------------------------------------*/=0A=
void SimpleApp::stop() {=0A=
        running =3D 0;=0A=
}=0A=
/*-----------------------------------------------------------*/=0A=
void SimpleApp::sendmsg() {=0A=
        Tcl &tcl =3D Tcl::instance();=0A=
=0A=
        if (running) {=0A=
                send(packetSize);=0A=
                tsend +=3D packetSize;=0A=
                tcl.evalf( " puts \"%d bytes send. \" ", tsend );=0A=
                timer->resched(timeInterval);=0A=
        } // else=0A=
        else if (tsend > 0)=0A=
                agent_->sendmsg(0, "MSG_EOF");=0A=
=0A=
}=0A=
/*-----------------------------------------------------------*/=0A=
void SimpleApp::recv(int nbytes) {=0A=
        Tcl &tcl =3D Tcl::instance();=0A=
=0A=
        trecv +=3D nbytes;=0A=
        tcl.evalf( " puts \"%d bytes received. \" ", trecv );=0A=
=0A=
}=0A=
/*-----------------------------------------------------------*/=0A=
static class SimpleAppClass : TclClass {=0A=
public:=0A=
        SimpleAppClass() : TclClass( "Application/SmplApp" ) {}   =
//constructor=0A=
        TclObject *create(int, const char*const* ) {=0A=
                return (new SimpleApp() );=0A=
        }=0A=
=0A=
} class_simple_app;=0A=

------=_NextPart_000_0007_01C1C843.217D2C20
Content-Type: application/octet-stream;
	name="test.tcl"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
	filename="test.tcl"

set ns [new Simulator]=0A=
=0A=
set ofile [open out.nam w ]=0A=
$ns namtrace-all $ofile=0A=
=0A=
set srcnode [$ns node]=0A=
set dstnode [$ns node]=0A=
=0A=
$ns duplex-link $srcnode $dstnode 1Mb 10ms DropTail=0A=
=0A=
set tcp [new Agent/TCP/FullTcp]=0A=
set tcpsink [new Agent/TCP/FullTcp]=0A=
=0A=
$ns attach-agent $srcnode $tcp=0A=
$ns attach-agent $dstnode $tcpsink=0A=
=0A=
$ns connect $tcp $tcpsink=0A=
=0A=
$tcpsink listen=0A=
=0A=
set app1 [new Application/SmplApp]=0A=
set app2 [new Application/SmplApp]=0A=
=0A=
$app1 attach-agent $tcp=0A=
$app2 attach-agent $tcpsink=0A=
=0A=
proc finish {} {=0A=
        global ns ofile=0A=
        $ns flush-trace=0A=
        close $ofile=0A=
        exec nam out.nam &=0A=
        exit 0=0A=
}=0A=
=0A=
#initial packet size is 100 bytes, and time interval 0.02s=0A=
#Later, I set tem to 200 bytes and 0.05s=0A=
=0A=
$ns at 0.6  "$app1 start"=0A=
$ns at 1.2  "$app1 set size 200"=0A=
$ns at 1.2  "$app1 set interval 0.05"=0A=
$ns at 2.2  "$app1 stop"=0A=
=0A=
$ns at 2.5 "finish"=0A=
=0A=
$ns run=0A=

------=_NextPart_000_0007_01C1C843.217D2C20--


From kopi@stanford.edu  Sun Mar 10 09:35:01 2002
From: kopi@stanford.edu (Milind Dilip Kopikare)
Date: Sun Mar 10 09:35:01 2002
Subject: [ns] Mac level, RTS-CTS-DS-DATA-ACK  implementation
In-Reply-To: <20020310114606.81703.qmail@web12208.mail.yahoo.com>
Message-ID: 

THe RTS-CTS-DS-DATA-ACK code is a part of the mac-802_11.h and .cc code
and is very much present in NS. If you want to hack into it, the above two
files are the ones you should look at.

Milind




On Sun, 10 Mar 2002, Badri Kasthuri wrote:

>
> Hi all,
>           does anyone have the RTS-CTS-DS-DATA-ACK
> implemented in ns. can somone kindly mail the codes if
> anybody has the code. i will be extremely greateful if
> someone can help.
>
> thanks,
> badri
>
> =====
>
>
> __________________________________________________
> Do You Yahoo!?
> Try FREE Yahoo! Mail - the world's greatest free email!
> http://mail.yahoo.com/
>


From xuanc@ISI.EDU  Sun Mar 10 16:45:03 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Sun Mar 10 16:45:03 2002
Subject: [ns] Question about MAC/TDMA
In-Reply-To: <20020309203102.91017.qmail@web12407.mail.yahoo.com>
Message-ID: 

Cited from ns manual:

virtual void expire =0 this method must be filled in by the 
timer client 

virtual void handle consumes an event; invokes 
expire() and sets status_ of the timer appropriately

I guess I just followed the style in mac-timers when I wrote the code. 
Maybe I missed something here, if you have any specific comments, please 
let me know. Thanks.

-chen 

 On Sat, 9 Mar 2002, Lisa Wu wrote:

> 
> Hi, there,
> 
> Thank you very much. It is very helpful.
> But I still have a question about the function
> SlotTdmaTimer::handle(). If it is called by the
> corresponding timer expiration, why do not use the
> expire() function. What is the difference to use
> handle() and expire() at this moment?
> 
> Thanks, 
> Lisa
> 
> 
> --- Xuan Chen  wrote:
> > On Fri, 8 Mar 2002, Lisa Wu wrote:
> > 
> > > 
> > > Hi, there, 
> > > 
> > > I have a few questions about Mac/Tdma. According
> > to
> > > "NS Manual" 16.1.4, it said the preamble is
> > divided in
> > > subslot for each node to write data. But after I
> > view
> > > the code(mac-tdma.cc), it seems that the preamble
> > is
> > > not subslotted. All nodes write its destination
> > > address in one slot at the same time. Will this
> > cause
> > > contention problem? 
> > The preamble is an array and each node has a
> > dedicated sub-slot. So, I 
> > don't think there will be any congestion. Does the
> > description look 
> > confused? Please let me know. Thanks.
> > 
> > > Another question is who will call function
> > > SlotTdmaTimer::handle()?
> > It is called by ns when the corresponding timer
> > expires. I think you can 
> > find details in ns manual.
> > 
> > Hope it helps,
> > -chen
> > 
> > > 
> > > 
> > > Thanks,
> > > 
> > > Lisa
> > >  
> > > 
> > > __________________________________________________
> > > Do You Yahoo!?
> > > Try FREE Yahoo! Mail - the world's greatest free
> > email!
> > > http://mail.yahoo.com/
> > > 
> > 
> > -- 
> > Xuan Chen
> > USC/ISI
> > 
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Try FREE Yahoo! Mail - the world's greatest free email!
> http://mail.yahoo.com/
> 

-- 
Xuan Chen
USC/ISI


From luorui999@sohu.com  Sun Mar 10 18:10:07 2002
From: luorui999@sohu.com (R.Luo)
Date: Sun Mar 10 18:10:07 2002
Subject: [ns] how to recompile a new agent
Message-ID: <000801c1c8a1$c5ccaea0$60011eac@123>

This is a multi-part message in MIME format.

------=_NextPart_000_0005_01C1C8E4.CA92C6C0
Content-Type: text/plain;
	charset="gb2312"
Content-Transfer-Encoding: base64

SGksDQogIGV4Y3VzZSBtZSwgbm93IEkgaGF2ZSBnb3R0ZW4gYSBuZXcgYWdlbnQsIHR3by1zdGF0
ZSBlcnJvciBtb2RlbCwgIHRoZSBwcm9ncmFtbWVzIGFyZSBlcnJtb2RlbC5jYywgZXJybW9kZWwu
aCwgbnMtZXJybW9kZWwudGNsLCBub3cgSSB3YW50IHRvIHJlY29tcGlsZSB0aGUgbmV3IGFnbmV0
IGluIG5zLTIsIGhvdyB0byBkbyBpdD8NCg==

------=_NextPart_000_0005_01C1C8E4.CA92C6C0
Content-Type: text/html;
	charset="gb2312"
Content-Transfer-Encoding: base64
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------=_NextPart_000_0005_01C1C8E4.CA92C6C0--


From ns_questions@yahoo.ca  Sun Mar 10 18:45:01 2002
From: ns_questions@yahoo.ca (Li Li)
Date: Sun Mar 10 18:45:01 2002
Subject: [ns] Connection vs. Flow
Message-ID: <20020311024103.58926.qmail@web20804.mail.yahoo.com>

Hi,

I want to obtain drop numbers of different connection
by flowmon, but it can only recognize different fid.
If I set the same fid for packets from different
connections, it can only calculate drops of all
connections. Please tell me how I could do.

Thanks

______________________________________________________________________ 
Find, Connect, Date! http://personals.yahoo.ca


From ns_questions@yahoo.ca  Sun Mar 10 18:45:14 2002
From: ns_questions@yahoo.ca (Li Li)
Date: Sun Mar 10 18:45:14 2002
Subject: [ns] Src and Dst vs. Fid
Message-ID: <20020311024251.59676.qmail@web20804.mail.yahoo.com>

To clarify the problems, I want to calculate packet
drops with the same src and Dst address not based on
Fid. 

Thanks

______________________________________________________________________ 
Find, Connect, Date! http://personals.yahoo.ca


From mahesh@cs.iitm.ernet.in  Mon Mar 11 05:15:01 2002
From: mahesh@cs.iitm.ernet.in (Mahesh)
Date: Mon Mar 11 05:15:01 2002
Subject: [ns] how to recompile a new agent
In-Reply-To: <000801c1c8a1$c5ccaea0$60011eac@123>
Message-ID: 

Hi,

Add the agent file entries in the makefile and recompile the NS using 
make. For more info on this see the nsbyexample site.

Hope this info will be of some use.

regards,
Mahesh. 

On Mon, 11 Mar 2002, R.Luo wrote:

> Hi,
>   excuse me, now I have gotten a new agent, two-state error model,  the programmes are errmodel.cc, errmodel.h, ns-errmodel.tcl, now I want to recompile the new agnet in ns-2, how to do it?
> 

-- 
Mahesh D			Mahesh D
Project Associate Gr.I		Cauvery#312
ERNET				IITMadras
IITMadras			Ph#445-9071
Ph#445-8355


From xuanc@ISI.EDU  Mon Mar 11 05:15:14 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Mon Mar 11 05:15:14 2002
Subject: [ns] Re: how to compile in ns-2
In-Reply-To: <355809.1015810583530.JavaMail.postfix@mailsrv1.mail.sohu.com>
Message-ID: 

just do a make, if you add new files, you need to modify Makefile.

Hope it helps,
-chen

On Mon, 11 Mar 2002 luorui999@sohu.com wrote:

> excuse me, if I have chang a module in the ns-2.1b7a, for example, errmodel.cc, errmodel.h, ns-errmodel.tcl, how can I compile these files and replace the old module with new module?
> 
> 
> 
>  thank your help very much!
> 

-- 
Xuan Chen
USC/ISI


From xuanc@ISI.EDU  Mon Mar 11 05:15:25 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Mon Mar 11 05:15:25 2002
Subject: [ns] Re: your mail
In-Reply-To: <20020311105525.46147.qmail@eyou.com>
Message-ID: 

You can find the answer (if there is anyg) by browsing the posts on the 
same thread. Or, just post the question to the list.

On 11 Mar 2002, ºúÍ¿ wrote:

> hi,
>    how can  I get the answers to the questions asked in archives?
>    thanks!
> 
> 
> 
> 
> 
> --http://www.eyou.com
> --Îȶ¨¿É¿¿µÄÃâ·Ñµç×ÓÐÅÏä  ÓïÒôÓʼþ  Òƶ¯ÊéÇ©  ÈÕÀú·þÎñ  ÍøÂç´æ´¢...ÒÚÓÊδ¾¡
> 
> 

-- 
Xuan Chen
USC/ISI


From geunkim@postech.ac.kr  Mon Mar 11 05:15:37 2002
From: geunkim@postech.ac.kr (Geunhyung Kim)
Date: Mon Mar 11 05:15:37 2002
Subject: [ns] Validate failure  in ns-2.1b8a
Message-ID: 

Hello All,

I validated my installation of ns-2.1b8a on solaris, but my validation results result in some failures in some test.
In the "validate.out" file, there is one failure.

Please tell me what is wrong. 


Thanks in advance


None of us is as smart as all of us
==========================================
Geunhyung Kim

E-mail: geunkim@postech.edu

Tel: +82-54-279-5655
Fax: +82-54-279-5699

Networking & Distributed Systems Lab.
CSE
POSTECH
===========================================


From 90106054@ccmail.npic.edu.tw  Mon Mar 11 05:15:49 2002
From: 90106054@ccmail.npic.edu.tw (C.G)
Date: Mon Mar 11 05:15:49 2002
Subject: [ns] How to edit and run [file.tcl] in ns environment
Message-ID: <002f01c1c8c7$20118230$447f40cb@VERITON>

This is a multi-part message in MIME format.

------=_NextPart_000_002C_01C1C90A.2E0C52A0
Content-Type: text/plain;
	charset="big5"
Content-Transfer-Encoding: quoted-printable

Dear Sir:
       We are students from Taiwan,we do have some problems for ns =
recently. There are questions that how to compile directly the program =
we written  and how to coding it in the editor(like emacs). We need an =
environment for coding and compiling .For now, we just can coding =
something we wanted sentance by sentance. May you answer the qusetion as =
soon? We will very appreciate you .
   You can mail us with the e-mail:jasonchien99@hotmail.com or =
gragchen@hotmail.com

                                                      Sincerely=20

------=_NextPart_000_002C_01C1C90A.2E0C52A0
Content-Type: text/html;
	charset="big5"
Content-Transfer-Encoding: quoted-printable











Dear =
Sir:


       We are=20
students from Taiwan,we do have some problems for ns =
recently.=20
There are questions that how to compile directly the program we =
written =20
and how to coding it in the editor(like emacs). We need an environment =
for=20
coding and compiling .For now, we just can coding something we wanted =
sentance=20
by sentance. May you answer the qusetion as soon? We will very =
appreciate=20
you .


   You can =
mail us with the e-mail:jasonchien99@hotma=
il.com=20
or gragchen@hotmail.com

 


          &nbs=
p;            =
;            =
    =20
            &=
nbsp;=20
Sincerely 


------=_NextPart_000_002C_01C1C90A.2E0C52A0--


From ramya4_a@rediffmail.com  Mon Mar 11 05:16:03 2002
From: ramya4_a@rediffmail.com (ramya)
Date: Mon Mar 11 05:16:03 2002
Subject: [ns] [bug] TCL command problem
Message-ID: <200203110652.GAA06345@www.isi.edu>

[Bug Report]

-----------------------------
Category:  Other
Package:   ns ns-2.1b8
OS:        Linux redHat release 7.1
Environment Variables:
LD_LIBRARY_PATH=
TCL_LIBRARY=
TK_LIBRARY=


-----------------------------
Description:

  When i want to specify the dynamic routing to particular node, the
following problem can occur.Please give some useful suggestion

 "$ns rtproto DV $n(1) $n(2)"

Error :-
   Wrong # args : should be set variable ? newValue?


From arshahid@yahoo.com  Mon Mar 11 05:16:16 2002
From: arshahid@yahoo.com (Ahmad R Shahid)
Date: Mon Mar 11 05:16:16 2002
Subject: [ns] Trace(Energy Level)
Message-ID: <20020311065455.38923.qmail@web10505.mail.yahoo.com>

how to see energy level of a node in a trace file?

Ahmad

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From andrealosacco@yahoo.it  Mon Mar 11 05:16:30 2002
From: andrealosacco@yahoo.it (Andrea Losacco)
Date: Mon Mar 11 05:16:30 2002
Subject: [ns] Clustering in NS
Message-ID: <002001c1c8d4$9e016cc0$c94dccc1@unile.it>

This is a multi-part message in MIME format.

------=_NextPart_000_001D_01C1C8DC.FF798980
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi everybody,
did anyone deal with clustering in NS? Any suggestions about the =
subject?
I'm working on clustering in ad hoc networks and I would like to get =
informations about classes that need to be implemented.

Thanks for your help.

Andrea Losacco


------=_NextPart_000_001D_01C1C8DC.FF798980
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable









Hi everybody,


did anyone deal with clustering in NS? =
Any=20
suggestions about the subject?


I'm working on clustering in ad hoc =
networks and I=20
would like to get informations about classes that need to be=20
implemented.


 


Thanks for your help.


 


Andrea Losacco


 


------=_NextPart_000_001D_01C1C8DC.FF798980--


From arshahid@yahoo.com  Mon Mar 11 05:16:43 2002
From: arshahid@yahoo.com (Ahmad R Shahid)
Date: Mon Mar 11 05:16:43 2002
Subject: [ns] node number
Message-ID: <20020311083710.17505.qmail@web10501.mail.yahoo.com>

hi all,

where do we define the node number while using the
command:

$ns_ node-config ..., since i dont find anything which
has to do anything with the node id!

Thanks,

Ahmad

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From fabrega@silver.udg.es  Mon Mar 11 05:16:56 2002
From: fabrega@silver.udg.es (=?iso-8859-1?Q?Llu=EDs=20F=E0brega?= i Soler)
Date: Mon Mar 11 05:16:56 2002
Subject: [ns] Update by Snapshot
References: 
Message-ID: <3C8CA515.C7C4C537@eia.udg.es>

Download the daily snapshot, and do:

./configure
make

Maybe you will also need to use the options ./configure
--with-tcl=your-path-to-tcl  --with-tk=tour-patht-to-tk ...
(see http://mailman.isi.edu/pipermail/ns-users/2002-January/020169.html)

Lluís



Geunhyung Kim wrote:

> Hello! All,
>
> I would like to update source code using snapshot.
> Could you tell me how to updat source code though snapshot smartly ?
>
> Thanks in advance,
>
> Geunhyung
>
> None of us is as smart as all of us
> ==========================================
> Geunhyung Kim
>
> E-mail: geunkim@postech.edu
>
> Tel: +82-54-279-5655
> Fax: +82-54-279-5699
>
> Networking & Distributed Systems Lab.
> CSE
> POSTECH
> ===========================================


From shivanajaym@yahoo.com  Mon Mar 11 05:45:03 2002
From: shivanajaym@yahoo.com (Shivanajay Marwaha)
Date: Mon Mar 11 05:45:03 2002
Subject: [ns] Scheduler going backwards in time
In-Reply-To: <20020307093354.W50226-100000@rumi.usc.edu>
Message-ID: <20020311131734.72561.qmail@web12402.mail.yahoo.com>

--0-2057841853-1015852654=:71110
Content-Type: text/plain; charset=us-ascii


 Hi,
 I am also getting this error but I am not doing emulation. I am only doing simulation. Is it because of my PC's speed or I have done some mistake in coding.
Thanks,
Shivanajay
  Alefiya Hussain  wrote: 
It indicates that the emulator cannot keep up with real time. Using a
faster computer will elevate this problem.

The results are not affeted drastically from this error.
Though you most prbably will be experiencing high packet losses in bpf.

alefiya
On Mon, 4 Mar 2002 netsim@gmx.at wrote:

>
> Hi!
>
> I am using the emulation mode in ns-2 called nse and ecountered the
> following problem:
>
> scheduler going backwards in time
>
> What is the reason for this error and how severe is it? Are all my trace
> files useless when getting this message?
>
> Thanks a lot.
>
> Martin Joerg
>
> --
> GMX - Die Kommunikationsplattform im Internet.
> http://www.gmx.net
>
>



---------------------------------
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
--0-2057841853-1015852654=:71110
Content-Type: text/html; charset=us-ascii


 Hi,

 I am also getting this error but I am not doing emulation. I am only doing simulation. Is it because of my PC's speed or I have done some mistake in coding.

Thanks,

Shivanajay

  Alefiya Hussain <alefiyah@catarina.usc.edu> wrote: 


It indicates that the emulator cannot keep up with real time. Using a
faster computer will elevate this problem.

The results are not affeted drastically from this error.
Though you most prbably will be experiencing high packet losses in bpf.

alefiya
On Mon, 4 Mar 2002 netsim@gmx.at wrote:

>
> Hi!
>
> I am using the emulation mode in ns-2 called nse and ecountered the
> following problem:
>
> scheduler going backwards in time
>
> What is the reason for this error and how severe is it? Are all my trace
> files useless when getting this message?
>
> Thanks a lot.
>
> Martin Joerg
>
> --
> GMX - Die Kommunikationsplattform im Internet.
> http://www.gmx.net
>
>

Do You Yahoo!?

Try FREE Yahoo! Mail - the world's greatest free email!
--0-2057841853-1015852654=:71110--


From fabrega@silver.udg.es  Mon Mar 11 05:45:29 2002
From: fabrega@silver.udg.es (=?iso-8859-1?Q?Llu=EDs=20F=E0brega?= i Soler)
Date: Mon Mar 11 05:45:29 2002
Subject: [ns] PHB EF.
References: 
Message-ID: <3C8CB224.DCA5E21E@eia.udg.es>

This is a multi-part message in MIME format.
--------------4E2A73FD9AC4597F40F70050
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

To implement PHB EF you need to choose a suitable queue discipline (prior=
ity,
WRR...) and then assign your codepoints.
I send you an example where there are 2 physical queues with priority
scheduling:
         - the high priority queue uses DROP and PHBs 10, 11; there is al=
so a
limit on the maximum bandwidth it can get. This would be EF.
         - the low priority queue uses RIO-C and PHBs 20, 21. This would =
be
AF.

Hope it helps!

Llu=EDs


"Karina Karla C. de Oliveira" wrote:

> Dear All.
>
> I would like to know if the diffserv version in ns-2.1b8 implements PHB=

> EF? Or only PHB AF?
>
> In chapter 9 of ns manual, there is the following statement:
> "In ns, packets are defaulted to a code point of zero. Therefore, user
> must add a PHB entry for the zero code point in order to handle best
> effort traffic."
>
> But if a don't specify an entry in policy table to a specific traffic,
> then "segmentation fault" ocurrs. I thought that in this case, code poi=
nt
> 0 will be attributed by default. Am I wrong?
>
> I don't want to use PHB AF to my voice traffic. And the others
> traffics will be BE. There isn't AF traffic in my scenario. What I have=
 to
> do to configure them?
>
> Best Regards,
> Karina Oliveira.
>
> *********************************************************
>         Karina Karla Cavalcante de Oliveira
>         Computer Science
>         Universidade Federal de Pernambuco
>         Centro de Inform=E1tica - CIn
>         Caixa Postal 7851
>         CEP: 50732-970
>         Recife-PE - Brasil
>         Phone: +55 (081) 3271-8430
>
>         Email: kkco@cin.ufpe.br
>         WWW: www.cin.ufpe.br/~kkco
>         Phones: (HOME) +55 (081) 3338-2054
> *********************************************************

--------------4E2A73FD9AC4597F40F70050
Content-Type: application/x-tcl;
 name="example7.tcl"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
 filename="example7.tcl"

#--------------------------------------------------------------------------------
# example7.tcl (based on ds-cbr-tb-PRI.tcl)
# Author: Lluis Fabrega
# Dates: 9 gener - 30 gener, 2002
# Notes: 
#       CBR traffic and token bucket conditioners:
#            1) s0-dest: PHB 10, 11
#            2) s1-dest: PHB 20, 21
#       2 physical queues:
#            1) high priority + DROP (PHB 10, 11) and rate-limited,
#            2) low priority + RIO (PHB 20,21)
#
#    ----
#    |s0|-----------
#    ----   10 Mb   \
#            5 ms    \
#                     \----           ------          ----           ------
#                      |e1|-----------|core|----------|e2|-----------|dest|
#                     /----   10 Mb   ------   5 Mb   ----   10 Mb   ------
#                    /         5 ms            5 ms           5 ms
#    ----           /
#    |s1|-----------
#    ----   10 Mb
#            5 ms
#
#--------------------------------------------------------------------------------


# cir[bps], cbs[bytes], rate [bps]
set cir0  1000000
set cbs0    10000
set rate0 4000000

set cir1  1000000
set cbs1    10000
set rate1 6000000

# tesTime[s] and paquet length [bytes]
set testTime 85.0
set packetSize 1000

# maximum bw in the high priority queue
set maxbw 3000000



set ns [new Simulator]
set nf [open out.nam w]
$ns namtrace-all $nf


# Set up the network topology shown at the top of this file:
set s0 [$ns node]
set s1 [$ns node]
set e1 [$ns node]
set core [$ns node]
set e2 [$ns node]
set dest [$ns node]

$ns duplex-link $s0 $e1 10Mb 5ms DropTail
$ns duplex-link $s1 $e1 10Mb 5ms DropTail

$ns simplex-link $e1 $core 10Mb 5ms dsRED/edge
$ns simplex-link $core $e1 10Mb 5ms dsRED/core

$ns simplex-link $core $e2 5Mb 5ms dsRED/core
$ns simplex-link $e2 $core 5Mb 5ms dsRED/edge

$ns duplex-link $e2 $dest 10Mb 5ms DropTail


$ns duplex-link-op $s0 $e1 orient down-right
$ns duplex-link-op $s1 $e1 orient up-right
$ns duplex-link-op $e1 $core orient right
$ns duplex-link-op $core $e2 orient right
$ns duplex-link-op $e2 $dest orient right


set qE1C [[$ns link $e1 $core] queue]
set qE2C [[$ns link $e2 $core] queue]
set qCE1 [[$ns link $core $e1] queue]
set qCE2 [[$ns link $core $e2] queue]


# Set DS RED parameters from Edge1 to Core:
$qE1C set numQueues_ 2
$qE1C setNumPrec 2
$qE1C meanPktSize $packetSize

$qE1C addPHBEntry 10 0 0
$qE1C addPHBEntry 11 0 1
$qE1C addPHBEntry 20 1 0
$qE1C addPHBEntry 21 1 1

$qE1C setSchedularMode PRI
$qE1C addQueueRate 0 $maxbw

$qE1C setMREDMode DROP 0
$qE1C configQ 0 0 20 40 0.02
$qE1C configQ 0 1 10 20 0.10

$qE1C setMREDMode RIO-C 1
$qE1C configQ 1 0 20 40 0.02
$qE1C configQ 1 1 10 20 0.10

$qE1C addPolicerEntry TokenBucket 10 11
$qE1C addPolicerEntry TokenBucket 20 21
$qE1C addPolicyEntry [$s0 id] [$dest id] TokenBucket 10 $cir0 $cbs0
$qE1C addPolicyEntry [$s1 id] [$dest id] TokenBucket 20 $cir1 $cbs1


# Set DS RED parameters from Edge2 to Core:
$qE2C set numQueues_ 2
$qE2C setNumPrec 2
$qE2C meanPktSize $packetSize

$qE2C addPHBEntry 10 0 0
$qE2C addPHBEntry 11 0 1
$qE2C addPHBEntry 20 1 0
$qE2C addPHBEntry 21 1 1

$qE2C setSchedularMode PRI
$qE2C addQueueRate 0 $maxbw

$qE2C setMREDMode DROP 0
$qE2C configQ 0 0 20 40 0.02
$qE2C configQ 0 1 10 20 0.10

$qE2C setMREDMode RIO-C 1
$qE2C configQ 1 0 20 40 0.02
$qE2C configQ 1 1 10 20 0.10

$qE2C addPolicerEntry TokenBucket 10 11
$qE2C addPolicerEntry TokenBucket 20 21
$qE2C addPolicyEntry [$dest id] [$s0 id] TokenBucket 10 $cir0 $cbs0
$qE2C addPolicyEntry [$dest id] [$s1 id] TokenBucket 20 $cir1 $cbs1


# Set DS RED parameters from Core to Edge1:
$qCE1 set numQueues_ 2
$qCE1 setNumPrec 2
$qCE1 meanPktSize $packetSize

$qCE1 addPHBEntry 10 0 0
$qCE1 addPHBEntry 11 0 1
$qCE1 addPHBEntry 20 1 0
$qCE1 addPHBEntry 21 1 1

$qCE1 setSchedularMode PRI
$qCE1 addQueueRate 0 $maxbw

$qCE1 setMREDMode DROP 0
$qCE1 configQ 0 0 20 40 0.02
$qCE1 configQ 0 1 10 20 0.10

$qCE1 setMREDMode RIO-C 1
$qCE1 configQ 1 0 20 40 0.02
$qCE1 configQ 1 1 10 20 0.10



# Set DS RED parameters from Core to Edge2:
$qCE2 set numQueues_ 2
$qCE2 setNumPrec 2
$qCE2 meanPktSize $packetSize

$qCE2 addPHBEntry 10 0 0
$qCE2 addPHBEntry 11 0 1
$qCE2 addPHBEntry 20 1 0
$qCE2 addPHBEntry 21 1 1

$qCE2 setSchedularMode PRI
$qCE2 addQueueRate 0 $maxbw

$qCE2 setMREDMode DROP 0
$qCE2 configQ 0 0 20 40 0.02
$qCE2 configQ 0 1 10 20 0.10

$qCE2 setMREDMode RIO-C 1
$qCE2 configQ 1 0 20 40 0.02
$qCE2 configQ 1 1 10 20 0.10


# Set up one CBR connection between each source and the destination:
set udp0 [new Agent/UDP]
$ns attach-agent $s0 $udp0
set cbr0 [new Application/Traffic/CBR]
$cbr0 attach-agent $udp0
$cbr0 set packet_size_ $packetSize
$udp0 set packetSize_ $packetSize
$cbr0 set rate_ $rate0

set null0 [new Agent/LossMonitor]
$ns attach-agent $dest $null0

$ns connect $udp0 $null0

set udp1 [new Agent/UDP]
$ns attach-agent $s1 $udp1
set cbr1 [new Application/Traffic/CBR]
$cbr1 attach-agent $udp1
$cbr1 set packet_size_ $packetSize
$udp1 set packetSize_ $packetSize
$cbr1 set rate_ $rate1

set null1 [new Agent/LossMonitor]
$ns attach-agent $dest $null1

$ns connect $udp1 $null1

$udp0 set class_ 0
$udp1 set class_ 1
$ns color 0 Blue
$ns color 1 Red


# Trace file for queue qCE2 and data file for xgraph for flows s0-dest and s1-dest
set nc [open cuaCE2.tr w]
$ns trace-queue $core $e2 $nc

set f0 [open outs0.tr w]
set f1 [open outs1.tr w]


proc finish {} {
        global f0 f1 ns nf
        #Close the output files
        close $f0
	close $f1
        close $nf
	exec nam out.nam &
        #Call xgraph to display the results
        exec xgraph outs0.tr outs1.tr -geometry 800x400 &
        exit 0
}


# Procedure that writes each flow throughput in a file for xgraph
proc record {} {
        global null0 null1 f0 f1
        #Get an instance of the simulator
        set ns [Simulator instance]
        #Set the time after which the procedure should be called again
        set time 0.1
        #How many bytes have been received by the traffic sinks?
        set bw0 [$null0 set bytes_]
	set bw1 [$null1 set bytes_]
        #Get the current time
        set now [$ns now]
        #Calculate the bandwidth (in MBit/s) and write it to the files
        puts $f0 "$now [expr $bw0/$time*8/1000000]"
        puts $f1 "$now [expr $bw1/$time*8/1000000]"
	#Reset the bytes_ values on the traffic sinks
        $null0 set bytes_ 0
        $null1 set bytes_ 0
	#Re-schedule the procedure
        $ns at [expr $now+$time] "record"
}



$qE1C printPolicyTable
$qE1C printPolicerTable
$qE1C printPHBTable

$ns at 0.0 "record"
$ns at 0.0 "$cbr0 start"
$ns at 0.0 "$cbr1 start"
$ns at 20.0 "$qCE2 printStats"
$ns at 40.0 "$qCE2 printStats"
$ns at 60.0 "$qCE2 printStats"
$ns at 80.0 "$qCE2 printStats"
$ns at $testTime "$cbr0 stop"
$ns at $testTime "$cbr1 stop"
$ns at [expr $testTime + 1.0] "finish"

$ns run

--------------4E2A73FD9AC4597F40F70050--


From engp1130@nus.edu.sg  Mon Mar 11 05:45:44 2002
From: engp1130@nus.edu.sg (liu yong)
Date: Mon Mar 11 05:45:44 2002
Subject: [ns] node number
Message-ID: <200203111337.g2BDbFk28193@leonis.nus.edu.sg>

Hi,
Accually the node id is the sequence number when you creat the new node in your tcl file.So the first node created will have the node id 0, and so on....You can use nam to verify it.
Rgds,
Liu Yong
>hi all,
>
>where do we define the node number while using the
>command:
>
>$ns_ node-config ..., since i dont find anything which
>has to do anything with the node id!
>
>Thanks,
>
>Ahmad
>
>__________________________________________________
>Do You Yahoo!?
>Try FREE Yahoo! Mail - the world's greatest free email!
>http://mail.yahoo.com/


From fabrega@silver.udg.es  Mon Mar 11 06:20:02 2002
From: fabrega@silver.udg.es (=?iso-8859-1?Q?Llu=EDs=20F=E0brega?= i Soler)
Date: Mon Mar 11 06:20:02 2002
Subject: [ns] How to know the PHB codepoint from the trace files?
Message-ID: <3C8CBD0E.4B29EB1B@eia.udg.es>

Hi all,

I am using the Diffserv module, and I want to know the PHB codepoint
that is assigned to a packet.
Is this information written in the trace files? And if not, what should
I modify in order to see this information in the trace files?

Thanks!

Lluís



From kkco@cin.ufpe.br  Mon Mar 11 06:55:01 2002
From: kkco@cin.ufpe.br (Karina Karla C. de Oliveira)
Date: Mon Mar 11 06:55:01 2002
Subject: [ns] How to know the PHB codepoint from the trace files?
In-Reply-To: <3C8CBD0E.4B29EB1B@eia.udg.es>
Message-ID: 

Hello All,

The PHB codepoint of a packet isn't write in the trace files. The second
question, I would like to know too.

Karina Oliveira.

On Mon, 11 Mar 2002, Lluís Fàbrega i Soler wrote:

>
> Hi all,
>
> I am using the Diffserv module, and I want to know the PHB codepoint
> that is assigned to a packet.
> Is this information written in the trace files? And if not, what should
> I modify in order to see this information in the trace files?
>
> Thanks!
>
> Lluís
>
>
>


From ahmedmoustafa01@yahoo.ca  Mon Mar 11 07:40:02 2002
From: ahmedmoustafa01@yahoo.ca (Ahmed Moustafa)
Date: Mon Mar 11 07:40:02 2002
Subject: [ns] DSR in wired-cum-wireless
Message-ID: <20020311153841.18543.qmail@web14101.mail.yahoo.com>

HI all

I'm trying to simulate DSR in wired-cum-wireless
scenario. When I did that I fot the following error:

Can't read ragent_

I invistigated the problem, and found that DSR works
fine in Wireless (Ad-Hoc Netowrk) but if you try to
attach the Ad-Hoc Network to a wired netowrk through
base station it fails. I looked at the code in the
ns-lib.tcl and found the following:


Simulator instproc set-dsr-nodetype {} {
        $self instvar wiredRouting_
        set nodetype SRNodeNew
        # MIP mobilenode
        if [Simulator set mobile_ip_] {
                set nodetype SRNodeNew/MIPMH
        }
        # basestation dsr node
        if { [info exists wiredRouting_] &&
$wiredRouting_ == "ON"} {
                set nodetype
Node/MobileNode/BaseStationNode

        }
        return $nodetype
}


As you can see if wired routing is off which means the
node is a mobile node the node type is SRnodeNew which
will instaniate the ragent_ inside the init procedure
for the SRNodeNew in the ns-mobilenode.tcl. But if the
wired routing is on which means it is a base station,
the node type is assigned to
Node/MobileNode/BaseStationNode, which doesn't not
instaniate the ragent_ in the init procedure. Is that
the correct thing and if it isn't as I assume what is
the right thing to do. I need your help urgently.

Thanks,
Ahmed

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From jelger@clarinet.u-strasbg.fr  Mon Mar 11 08:30:02 2002
From: jelger@clarinet.u-strasbg.fr (Christophe Jelger)
Date: Mon Mar 11 08:30:02 2002
Subject: [ns] What is the difference between target_->recv(p) & send(p,0)?
References: 
Message-ID: <3C8CDA86.30609@clarinet.u-strasbg.fr>

Hi,

And what happens when objects are not connectors ? For example if an 
agent simply wants to send packets to a given dest address ? (assuming 
that there is a routing entry in the classifier) ? For example, I want 
my agent to send a packet to the well known ALL-PIM-ROUTERS address ... 
how shall I proceed ? (the agent is not connected so there is no target_)

Thanks

Christophe


Padmaparna Haldar wrote:

>
>
>Most objects in ns are Connector objects having a target_ to SEND pkts to. 
>It can also RECV pkts. recv and send functions in the Connector class are 
>defined as follows:
>
>void recv(Packet*, Handler* callback = 0);
>inline void send(Packet* p, Handler* h) { target_->recv(p, h); }
>
>so as you can see, send(p,h) and target_->recv(p) are basically doing the 
>same thing.
>
>--Padma
>
>On Wed, 6 Mar 2002, Mustafa Tasdemir wrote:
>
>
>>hello,
>>
>>can someone tell what is the difference between these commands when used in agent:
>>
>>Application---->UDP------->Node
>>                           |
>>                           |
>>                        commands are used in UDP agents
>>  1.. target_->recv(p);    // target defined as Process *target_; 
>>  2.. send(p,0);
>>I think, the first one is a direct call and the second one is an indirect call. Can we also call the send() method as a SAP (Service Access Point)?
>>
>>Mustafa TASDEMIR
>>
>>
>>
>>
>



From mkenny@mekb2.sps.mot.com  Mon Mar 11 08:30:25 2002
From: mkenny@mekb2.sps.mot.com (Martin McKenny)
Date: Mon Mar 11 08:30:25 2002
Subject: [ns] recv(); method in a classifier.h
Message-ID: <3C8CDB6E.9A04BCC3@mekb2.sps.mot.com>

--------------8B6015F17FF89E0AA505BBC7
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: base64

Hi NS,

I am trying to call the a classifier's recv(); method from within my
code. I have passed a packet for the 'Packet*' argument; what do I pass
for the 'Handler*' arument. I am getting the following error.

rapidio.cc:180: cannot call member function `Classifier::recv(Packet *,
Handler *)' without object

Thanks

MMcKenny



--------------8B6015F17FF89E0AA505BBC7
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit



Hi NS,

I am trying to call the a classifier's recv(); method from within my
code. I have passed a packet for the 'Packet*' argument; what do I pass
for the 'Handler*' arument. I am getting the following error.

rapidio.cc:180: cannot call member function `Classifier::recv(Packet
*, Handler *)' without object

Thanks



MMcKenny

 

--------------8B6015F17FF89E0AA505BBC7--


From sarma_vvrkv@yahoo.com  Mon Mar 11 09:00:03 2002
From: sarma_vvrkv@yahoo.com (vangala sarma)
Date: Mon Mar 11 09:00:03 2002
Subject: [ns] Help required urgently...
Message-ID: <20020311165637.39077.qmail@web12307.mail.yahoo.com>

Hi,
 I am trying to introduce a two State error model in
to the file using Snoop protocol. It is in
/tcl/ex/snoop/MySnoop.tcl. I am getting the following
error..

sourcing ../../lan/vlan.tcl...
In the errormodel constructor
In the two state error model constructor::
--- Classfier::no-slot{} default handler
(tcl/lib/ns-lib.tcl) ---
        _o13: no target for slot -1
        _o13 type: Classifier/Hash/Dest
content dump:
classifier _o13
        0 offset
        0 shift
        2147483647 mask
        0 slots
        -1 default
---------- Finished standard no-slot{} default handler
----------


Can anybody please help me out. Also this file uses
the 802.3 standard. The wireless part must use the
802.11 standard. It is not specified in the file
enywhere. 
Please help me as I am stuck on this point for a long
time.

Thanks in advance,
Sarma Vangala



__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From paul@cse.nsysu.edu.tw  Mon Mar 11 09:05:05 2002
From: paul@cse.nsysu.edu.tw (paul)
Date: Mon Mar 11 09:05:05 2002
Subject: [ns] Performace of  error models over wireless networks ?
Message-ID: <015401c1c91e$408bf340$16c0758c@KNOT>

This is a multi-part message in MIME format.

------=_NextPart_000_0151_01C1C961.4E85D950
Content-Type: text/plain;
	charset="big5"
Content-Transfer-Encoding: quoted-printable

Hi ns-users,

I am doing an emulation experiment, and I want to introduce some error =
model
into a wired-cum-wireless scenario.
The topology is like the following:

MH(0)  BS(0)<--->W(0)<--------------->W(1)<--->BS(1)  MH(1)

Introducing real traffic into mobile nodes (MH(0) and MH(1)), and there =
is a
Tap agent connection between MH(0) and MH(1).

And then, I have the following tests.
(ns-2-snapshot-20020311, Linux RedHat 7.1, PIII 600MHz)
(1) When I introduce the error model into wired links, such as
      $ns_ lossmodel $em $W(0) $W(1)
      Eveything is ok. (with tcpdump check)
(2) When I introduce the error model into mobile nodes, such as
      $ns_ node-config -IncomingErrProc UniformErr -OutgoingErrProc =
UniformErr
      The error rate seems ok, but there is extra about 1 second (or =
above)
      delay in ouput packets!!!! What is the reason leading to the =
serious=20
      delay ??

Very Thanks for your help,

paul


------=_NextPart_000_0151_01C1C961.4E85D950
Content-Type: text/html;
	charset="big5"
Content-Transfer-Encoding: quoted-printable









Hi ns-users,

I am doing an =
emulation=20
experiment, and I want to introduce some error model
into a=20
wired-cum-wireless scenario.
The topology is like the=20
following:

MH(0) =20
BS(0)<--->W(0)<--------------->W(1)<--->BS(1) =20
MH(1)

Introducing real traffic into mobile nodes (MH(0) and =
MH(1)), and=20
there is a
Tap agent connection between MH(0) and MH(1).

And =
then, I=20
have the following tests.
(ns-2-snapshot-20020311, Linux RedHat 7.1, =
PIII=20
600MHz)
(1) When I introduce the error model into wired links, such=20
as
      $ns_ lossmodel $em $W(0)=20
$W(1)
      Eveything is ok. (with tcpdump=20
check)
(2) When I introduce the error model into mobile nodes, such=20
as
      $ns_ node-config -IncomingErrProc=20
UniformErr -OutgoingErrProc UniformErr
      =
The=20
error rate seems ok, but there is extra about 1 second (or=20
above)


      delay in=20
ouput packets!!!! What is the reason leading to the serious =



      delay=20
??

Very Thanks for your=20
help,

paul


------=_NextPart_000_0151_01C1C961.4E85D950--


From rlaw@vt.edu  Mon Mar 11 09:15:02 2002
From: rlaw@vt.edu (Raymond Law)
Date: Mon Mar 11 09:15:02 2002
Subject: [ns] Error - Attempting to cancel timer
In-Reply-To: <3C8B6443@zathras>
Message-ID: <5.1.0.14.0.20020311121628.039b2100@mail.vt.edu>

Solved.  The problem turned out to be my modified code.

Ray,

At 02:46 AM 3/10/2002 -0500, Raymond Law wrote:

>When I ran the sample test-crlsp.tcl script for mns v2.0, the error
>"Attempting to cancel timer at 0094C408 which is not scheduled" still occurs
>even after the following patch is applied to trafgen.cc.
>
>TrafficGenerator::TrafficGenerator() : nextPkttime_(-1), running_(0),
>timer_(this)
>
>All the starts are scheduled before the stops in the tcl script.  Any
>suggestions?
>
>Thanks.
>Ray,


From mkenny@mekb2.sps.mot.com  Mon Mar 11 09:20:01 2002
From: mkenny@mekb2.sps.mot.com (Martin McKenny)
Date: Mon Mar 11 09:20:01 2002
Subject: [ns] Help required urgently...
References: <20020311165637.39077.qmail@web12307.mail.yahoo.com>
Message-ID: <3C8CE6A6.3023EF@mekb2.sps.mot.com>

--------------DC9EBD5D643D85D1FCF14916
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: base64

Sounds as though you are trying to send packets between nodes that are
not connected or where the destination does not have a slot reference in
the classifier. I am quite new to this myself therefore maybe some of
the experts may be able to help. You could maybe try using add-route?
(NS manual section 5.6).



vangala sarma wrote:

> Hi,
>  I am trying to introduce a two State error model in
> to the file using Snoop protocol. It is in
> /tcl/ex/snoop/MySnoop.tcl. I am getting the following
> error..
>
> sourcing ../../lan/vlan.tcl...
> In the errormodel constructor
> In the two state error model constructor::
> --- Classfier::no-slot{} default handler
> (tcl/lib/ns-lib.tcl) ---
>         _o13: no target for slot -1
>         _o13 type: Classifier/Hash/Dest
> content dump:
> classifier _o13
>         0 offset
>         0 shift
>         2147483647 mask
>         0 slots
>         -1 default
> ---------- Finished standard no-slot{} default handler
> ----------
>
> Can anybody please help me out. Also this file uses
> the 802.3 standard. The wireless part must use the
> 802.11 standard. It is not specified in the file
> enywhere.
> Please help me as I am stuck on this point for a long
> time.
>
> Thanks in advance,
> Sarma Vangala
>
> __________________________________________________
> Do You Yahoo!?
> Try FREE Yahoo! Mail - the world's greatest free email!
> http://mail.yahoo.com/

--
==============================================================
Martin McKenny
Research Engineer ISLI/Motorola
Phone  : +44 1506473378
Email  : mkenny@mekb2.sps.mot.com             (Mot)
       : martin.mckenny@sli-institute.ac.uk   (ISLI)
==============================================================



--------------DC9EBD5D643D85D1FCF14916
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit



Sounds as though you are trying to send packets between nodes that are
not connected or where the destination does not have a slot reference in
the classifier. I am quite new to this myself therefore maybe some of the
experts may be able to help. You could maybe try using add-route? (NS manual
section 5.6).

 

 

vangala sarma wrote:

Hi,

 I am trying to introduce a two State error model in

to the file using Snoop protocol. It is in

/tcl/ex/snoop/MySnoop.tcl. I am getting the following

error..

sourcing ../../lan/vlan.tcl...

In the errormodel constructor

In the two state error model constructor::

--- Classfier::no-slot{} default handler

(tcl/lib/ns-lib.tcl) ---

        _o13: no target for slot
-1

        _o13 type: Classifier/Hash/Dest

content dump:

classifier _o13

        0 offset

        0 shift

        2147483647 mask

        0 slots

        -1 default

---------- Finished standard no-slot{} default handler

----------

Can anybody please help me out. Also this file uses

the 802.3 standard. The wireless part must use the

802.11 standard. It is not specified in the file

enywhere.

Please help me as I am stuck on this point for a long

time.

Thanks in advance,

Sarma Vangala

__________________________________________________

Do You Yahoo!?

Try FREE Yahoo! Mail - the world's greatest free email!

http://mail.yahoo.com/



-- 
==============================================================
Martin McKenny
Research Engineer ISLI/Motorola
Phone  : +44 1506473378
Email  : mkenny@mekb2.sps.mot.com             (Mot)
       : martin.mckenny@sli-institute.ac.uk   (ISLI)
==============================================================

 

--------------DC9EBD5D643D85D1FCF14916--


From xuanc@ISI.EDU  Mon Mar 11 09:50:02 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Mon Mar 11 09:50:02 2002
Subject: [ns] RE: ns mailing list on the web is not working
In-Reply-To: <000001c1c8d9$d885f280$0d039a83@pccris>
Message-ID: 

After we switched mailing lists to mailman, all the archives can be found 
at:

http://mailman.isi.edu/pipermail/ns-users/

Maybe, we should add something to redirect people from the old place to 
the new one.

Cheers,
-chen

On Mon, 11 Mar 2002, Sergio Andreozzi wrote:

> Hi,
> 
> http://www.isi.edu/nsnam/archive/ns-users/webarch/current/
> 
> last message I can read is dated 10 Jan 2002. I guess there is some
> later messege. Maybe I'm wrong.
> 
> Have a nice day
> 
> Sergio
> 
> -----Original Message-----
> From: Xuan Chen [mailto:xuanc@isi.edu] 
> Sent: Friday, March 08, 2002 6:41 PM
> To: Sergio Andreozzi
> Subject: Re: ns mailing list on the web is not working
> 
> which mailing list? I just checked ns-users, ns-announce, and ns-edu.
> Both 
> the information page and the archives seem ok....
> 
> On Fri, 8 Mar 2002, Sergio Andreozzi wrote:
> 
> > Hello,
> >  
> > I noticed the web mailing list about ns is not working. Can you check
> > it, please?
> >  
> > Thanks,
> > 
> > Bye
> > 
> > Sergio
> > 
> 
> 

-- 
Xuan Chen
USC/ISI


From xuanc@ISI.EDU  Mon Mar 11 10:00:02 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Mon Mar 11 10:00:02 2002
Subject: [ns] How to know the PHB codepoint from the trace files?
In-Reply-To: 
Message-ID: 

Karina, 

You are right. The diffserv module does not use the standard trace 
function in ns. If you want to see the codepoint assigned to EACH packet, 
you need to add trace function by yourself. Please refer to ns manual for 
trace support. Once you figure it out, I think we will be happy to 
integrate it into ns.

If what you are interested is actually some statistics on packet marking, 
say how many packets are marked as IN. You can find such function from the 
current diffserv module, please refer to the ns manual and the sample 
scripts under ns/tcl/ex/diffserv/ for details.

Hope it helps,
-chen

 On Mon, 11 Mar 2002, Karina Karla C. de Oliveira wrote:

> 
> 
> Hello All,
> 
> The PHB codepoint of a packet isn't write in the trace files. The second
> question, I would like to know too.
> 
> Karina Oliveira.
> 
> On Mon, 11 Mar 2002, Lluís Fàbrega i Soler wrote:
> 
> >
> > Hi all,
> >
> > I am using the Diffserv module, and I want to know the PHB codepoint
> > that is assigned to a packet.
> > Is this information written in the trace files? And if not, what should
> > I modify in order to see this information in the trace files?
> >
> > Thanks!
> >
> > Lluís
> >
> >
> >
> 

-- 
Xuan Chen
USC/ISI


From haldar@ISI.EDU  Mon Mar 11 11:40:01 2002
From: haldar@ISI.EDU (Padmaparna Haldar)
Date: Mon Mar 11 11:40:01 2002
Subject: [ns] What is the difference between target_->recv(p) & send(p,0)?
In-Reply-To: <3C8CDA86.30609@clarinet.u-strasbg.fr>
Message-ID: 

when an agent gets attached to a node, the target_ points to the node 
entry_ which in default condition points to the the classifier.
when you connect two agents you are basically setting up the destination 
to which the agent will send pkts to. 
so assumimg yr agent was connected to a node, it can send 
pkts to a given dest address, provided there are corr. routing entries for 
that address in the classifier.
--Padma

On Mon, 11 Mar 2002, Christophe Jelger wrote:


> 
> Hi,
> 
> And what happens when objects are not connectors ? For example if an 
> agent simply wants to send packets to a given dest address ? (assuming 
> that there is a routing entry in the classifier) ? For example, I want 
> my agent to send a packet to the well known ALL-PIM-ROUTERS address ... 
> how shall I proceed ? (the agent is not connected so there is no target_)
> 
> Thanks
> 
> Christophe
> 
> 
> Padmaparna Haldar wrote:
> 
> >
> >
> >Most objects in ns are Connector objects having a target_ to SEND pkts to. 
> >It can also RECV pkts. recv and send functions in the Connector class are 
> >defined as follows:
> >
> >void recv(Packet*, Handler* callback = 0);
> >inline void send(Packet* p, Handler* h) { target_->recv(p, h); }
> >
> >so as you can see, send(p,h) and target_->recv(p) are basically doing the 
> >same thing.
> >
> >--Padma
> >
> >On Wed, 6 Mar 2002, Mustafa Tasdemir wrote:
> >
> >
> >>hello,
> >>
> >>can someone tell what is the difference between these commands when used in agent:
> >>
> >>Application---->UDP------->Node
> >>                           |
> >>                           |
> >>                        commands are used in UDP agents
> >>  1.. target_->recv(p);    // target defined as Process *target_; 
> >>  2.. send(p,0);
> >>I think, the first one is a direct call and the second one is an indirect call. Can we also call the send() method as a SAP (Service Access Point)?
> >>
> >>Mustafa TASDEMIR
> >>
> >>
> >>
> >>
> >
> 
> 

-- 
------------------------------------------------ 
Be true to your work, your word, and your friend. 
--Thoreau

Padmaparna Haldar


From jw@erg.sri.com  Mon Mar 11 14:30:03 2002
From: jw@erg.sri.com (Julie Wong)
Date: Mon Mar 11 14:30:03 2002
Subject: [ns] bandwidth question for mobilenodes
References: <3C868B77.6D516A32@erg.sri.com> <3C868D01.92B231BB@msu.edu> <000f01c1c59f$b8a41ec0$b108708c@pegasus>
Message-ID: <3C8D2DB2.D0892DA0@erg.sri.com>

Zhang Zhe-Wei,

By now you might have already figured things out yourself.
But just in case, I set bandwidth as follows and it works for me.


Phy/WirelessPhy set bandwidth_ 11e6
Mac/802_11 set bandwidth_ 11e6

Julie Wong


Zhang Zhe-Wei wrote:

> hello.
>
>     I add  "Mac/802_11 set bandwidth_ 11e6" to my tcl file.
>     And the result is the same as default.
>
>     Could you tell me what's wrong?
>     Thanks in advance!
>
>                          Che-Wei
>
> ----- Original Message -----
> From: "Gautam - Chakrabarti" 
> To: "Julie Wong" 
> Cc: 
> Sent: Thursday, March 07, 2002 5:41 AM
> Subject: Re: [ns] bandwidth question for mobilenodes
>
> >
> > You need to do
> >
> > Mac/802_11 set bandwidth_ 11e6
> >
> > --- this would change the bandwidth from the default value of 2, but if
> > I remember
> > right it cannot be increased above 10 or something like that. Try
> > increasing it from 2 to 4 or 5 ... that should work.
> >
> > rgds,
> > gautam
> >
> > Julie Wong wrote:
> > >
> > > Hello,
> > >
> > > I have been trying to set the bandwidth of a mobile radio node to 11e6
> > > without any success.  My wireless network uses interface
> > > Phy/WirelessPhy.  Do I need to do anything else besides
> > >
> > >     Phy/WirelessPhy set bandwidth_  11e6
> > >
> > > in my .tcl file?
> > >
> > > Thanks in advance,
> > >
> > > Julie Wong
> > >
> > > P.S.  I saw the default bandwidth in the ns-default.tcl file.
> > > I even changed that to 11e6 but it did not seem to help.
> >
> > --
> > ***************************************************************
> > Gautam Chakrabarti
> > MS Student
> > Computer Sc. and Engg Department
> > Michigan State University
> >
> > Email : chakra10@msu.edu
> > Home Page : http://www.cse.msu.edu/~chakra10
> >
> > Why is there so much month left at the end of the money?
> > --- John Barrymore
> > ***************************************************************
> >
> >
> >


From john_smith1179@yahoo.com  Mon Mar 11 14:50:02 2002
From: john_smith1179@yahoo.com (John Smith)
Date: Mon Mar 11 14:50:02 2002
Subject: [ns] Wireless scenario generation in windows
Message-ID: <20020311224649.55767.qmail@web20201.mail.yahoo.com>

Hi,
I want to generate wireless node movement scenarios in
windows, but I did not find any support for making the
configuration. The tutorial also says it for
Unix/Linux but nothing about generating scenarios on
windows.

could anyone who has tried it please enlighten me on
the same

Thanks in advance



__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From john_smith1179@yahoo.com  Mon Mar 11 14:50:18 2002
From: john_smith1179@yahoo.com (John Smith)
Date: Mon Mar 11 14:50:18 2002
Subject: [ns] Wireless scenario generation in windows
Message-ID: <20020311224657.51349.qmail@web20205.mail.yahoo.com>

Hi,
I want to generate wireless node movement scenarios in
windows, but I did not find any support for making the
configuration. The tutorial also says it for
Unix/Linux but nothing about generating scenarios on
windows.

could anyone who has tried it please enlighten me on
the same

Thanks in advance



__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From john_smith1179@yahoo.com  Mon Mar 11 14:50:38 2002
From: john_smith1179@yahoo.com (John Smith)
Date: Mon Mar 11 14:50:38 2002
Subject: [ns] Wireless scenario generation in windows
Message-ID: <20020311224657.21381.qmail@web20207.mail.yahoo.com>

Hi,
I want to generate wireless node movement scenarios in
windows, but I did not find any support for making the
configuration. The tutorial also says it for
Unix/Linux but nothing about generating scenarios on
windows.

could anyone who has tried it please enlighten me on
the same

Thanks in advance



__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From chakra10@msu.edu  Mon Mar 11 14:55:01 2002
From: chakra10@msu.edu (Gautam - Chakrabarti)
Date: Mon Mar 11 14:55:01 2002
Subject: [ns] Trace(Energy Level)
References: <20020311065455.38923.qmail@web10505.mail.yahoo.com>
Message-ID: <3C8D351C.B35B7DB3@msu.edu>

if you are doing wireless simulation, and using the new trace format,
the number after -Ne would give the node's energy level.

rgds,
gautam

Ahmad R Shahid wrote:
> 
> how to see energy level of a node in a trace file?
> 
> Ahmad
> 
> __________________________________________________
> Do You Yahoo!?
> Try FREE Yahoo! Mail - the world's greatest free email!
> http://mail.yahoo.com/

-- 
***************************************************************
Gautam Chakrabarti
MS Student
Computer Sc. and Engg Department
Michigan State University

Email : chakra10@msu.edu
Home Page : http://www.cse.msu.edu/~chakra10

Why is there so much month left at the end of the money?
		--- John Barrymore
***************************************************************


From jw@erg.sri.com  Mon Mar 11 18:35:02 2002
From: jw@erg.sri.com (Julie Wong)
Date: Mon Mar 11 18:35:02 2002
Subject: [ns] Re: (no subject)
References: <3C869384.1F3F178@msu.edu> <3C8D2F64.508A6B98@erg.sri.com> <3C8D35F1.8C9B20E0@msu.edu>
Message-ID: <3C8D6939.71FE31AE@erg.sri.com>

Gautam,

I was a bit rushed to say that it works.  I do see that CTS and RTS
messages get through with a link speed of 11mbs.  However, when
it comes to data packets, for some reason, if a packet has more than
1 hop, it is dropped at the second hop.  I think the dropping occurs
in mac-802_11.cc line 1290 discard(p, DROP_MAC_BUSY) in the module called
recvDATA().  So I take back what I said before that
setting the Mac bandwidth to 11mbs works.

BTW, when the speed is 9mbs or less, things are ok.  Only when it
gets up to 10mbs or more, then the above happens.  Also, data
packets on the first hop is ok.   Any idea why this might be the
case?

Thanks,

Julie Wong



Gautam - Chakrabarti wrote:

> Julie Wong wrote:
> >
> > Thanks Gautam for the extra information.  I set the bandwidth at
> > the Mac level and it works now.  By the way, setting it to 11mbs
> > is ok.  I got something very close (close enough) to a 11mbs link.
>
> Thanks for clarifying, actually I am not exactly working on it for
> sometime,
> but I was told by someone that the throughput is not getting to the
> value,
> although the bandwidth has the value 11. You might want to look at that.
> If
> that is true, that probably means there are some other controlling
> parameters.
>
> rgds,
>
> Gautam
>
> >
> > Julie Wong
> >
> > Gautam - Chakrabarti wrote:
> >
> > > Hi,
> > >
> > > You may look into the following messages which appear to be relevant.
> > >
> > > Hth,
> > > gautam
> > >
> > >  -------------------------
> > > >
> > > > Hi,
> > > > I think that 'bandwidth' variable of Phy/WirelessPhy object is useful
> > > > only
> > > > if you enable the Errormodel: it is used to compute txtime when decrease
> > > > node energy (check wirelessphy.cc).
> > > > The variable you have to change is the 'bandwidth' of Mac object! This
> > > > is
> > > > the variable that real affect txtime of a packet.
> > > > By default, it is set as 2Mb (as the bandwidth of Phy/WirelessPhy; check
> > > > ~ns2/tcl/lib/ns-default.tcl, ~ns2/tcl/lan/ns-mac.tcl). You can change it
> > > > and
> > > > see what happens in the tracefile using
> > > >
> > > > Mac/802_11 set bandwidth_ 
> > > >
> > > > Federico Bertocchi
> > > >
> > > > The above is from  "Federico Bertocchi" 
> > > >
> > > > Hi!
> > > >
> > > > I do not use a recent snapshot, but ns2.1b8a.
> > > > In my opinion the bandwidth_ variable of Wireless Phy influences only
> > > > the
> > > > energy related part of communication. For example, when using IEEE
> > > > 802.11, the
> > > > mac-802_11 has its own bandwidth_ variable, which controls the
> > > > transmission
> > > > data rate.
> > > > If I am wrong, please let me know!
> > > >
> > > > Regards,
> > > > Patrick
> > > >
> > > >
> > > > --- from  Patrick Peschlow 
> > > >
> > > >
> > > >
>
> --
> ***************************************************************
> Gautam Chakrabarti
> MS Student
> Computer Sc. and Engg Department
> Michigan State University
>
> Email : chakra10@msu.edu
> Home Page : http://www.cse.msu.edu/~chakra10
>
> Why is there so much month left at the end of the money?
>                 --- John Barrymore
> ***************************************************************


From vamsiparuchuri@hotmail.com  Mon Mar 11 18:50:01 2002
From: vamsiparuchuri@hotmail.com (vamsi paruchuri)
Date: Mon Mar 11 18:50:01 2002
Subject: [ns] prune packets as feedback packets??
Message-ID: 

Hi,

   Can anyone one tell me if I can modify prune/graft agents so as to send 
some feedback(of one byte) from the receivers to the multicast source. If 
yes, how?
Thanks,
Vamsi.



_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp.


From vamsiparuchuri@hotmail.com  Mon Mar 11 19:15:03 2002
From: vamsiparuchuri@hotmail.com (vamsi paruchuri)
Date: Mon Mar 11 19:15:03 2002
Subject: [ns] target and send methods
Message-ID: 

Hi,

   Can I send a packet from Multicast receiver to its multicast source using 
Target or Send methods.
Thanks,
Vamsi.



_________________________________________________________________
Chat with friends online, try MSN Messenger: http://messenger.msn.com


From ramya4_a@rediffmail.com  Tue Mar 12 05:15:05 2002
From: ramya4_a@rediffmail.com (Ramya)
Date: Tue Mar 12 05:15:05 2002
Subject: [ns] [bug] Wireless Characteristics
Message-ID: <200203120438.EAA23603@www.isi.edu>

[Bug Report]

-----------------------------
Category:  Other
Package:   ns 2.1b8
OS:        Linux redHat release 7.1
Environment Variables:
LD_LIBRARY_PATH=
TCL_LIBRARY=
TK_LIBRARY=


-----------------------------
Description:

  The examples of the wireless network are implemented for the mobile node.
Is any ordinary node which contains the interface characteristics of the
wireless network ?.I tried by creating 2 ordinary nodes with wireless
network characteristics data flow.The program does not give any error 
message and run with infinite loop.Please clarify my details.














From arshahid@yahoo.com  Tue Mar 12 05:15:20 2002
From: arshahid@yahoo.com (Ahmad R Shahid)
Date: Tue Mar 12 05:15:20 2002
Subject: [ns] node number
In-Reply-To: <200203111337.g2BDbFk28193@leonis.nus.edu.sg>
Message-ID: <20020312062637.24441.qmail@web10506.mail.yahoo.com>

Hi,

actually in my tcl file i have written the following
lines:

for {set j 0} {$j < 5} {incr j} {
    set node_($j) [ $opt(rp)-create-mobile-node $j
[lindex $temp $j] ]
    $node_($j) base-station [AddrParams addr2id
[$BS(0) node-addr]]
}

now i want to add/change certain features in the
mobile nodes created by [ $opt(rp)-create-mobile-node
$j [lindex $temp $j] ] command. now if i change the
above to:

for {set j 0} {$j < 5} {incr j} {
    set node_($j) [ $opt(rp)-create-mobile-node $j
[lindex $temp $j] ]
    $node_($j) base-station [AddrParams addr2id
[$BS(0) node-addr]]

    $ns_ node-config -energyModel EneryModel
                     -initialEnergy 0.01
                     -rxPower     0.3
                     -txPower     0.4
}

would these characteristics be given to the newly
created node by create-mobile-node? 


Ahmad

--- liu yong  wrote:
> 
> Hi,
> Accually the node id is the sequence number when you
> creat the new node in your tcl file.So the first
> node created will have the node id 0, and so
> on....You can use nam to verify it.
> Rgds,
> Liu Yong
> >hi all,
> >
> >where do we define the node number while using the
> >command:
> >
> >$ns_ node-config ..., since i dont find anything
> which
> >has to do anything with the node id!
> >
> >Thanks,
> >
> >Ahmad
> >
> >__________________________________________________
> >Do You Yahoo!?
> >Try FREE Yahoo! Mail - the world's greatest free
> email!
> >http://mail.yahoo.com/
> 


__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From giuliastro@tin.it  Tue Mar 12 05:15:34 2002
From: giuliastro@tin.it (Giulio Ardoino)
Date: Tue Mar 12 05:15:34 2002
Subject: [ns] recv(); method in a classifier.h
In-Reply-To: <3C8CDB6E.9A04BCC3@mekb2.sps.mot.com>
References: <3C8CDB6E.9A04BCC3@mekb2.sps.mot.com>
Message-ID: <02031208413203.21699@olmo.reti.dist.unige.it>

On Monday 11 March 2002 17:29, Martin McKenny wrote:

> I am trying to call the a classifier's recv(); method from within my code. 

I suggest you to take a look under ns' webcache/ folder for tcpapp.*
That's the way I used to comunicate by sending packets.

Giulio


From arshahid@yahoo.com  Tue Mar 12 05:15:49 2002
From: arshahid@yahoo.com (Ahmad R Shahid)
Date: Tue Mar 12 05:15:49 2002
Subject: [ns] portion in NS
Message-ID: <20020312082639.46724.qmail@web10503.mail.yahoo.com>

Hi all,

What this portion in wireless-newnode-energy.tcl is
achieving:

proc usage { argv0 }  {
	puts "Usage: $argv0"
	puts "\tmandatory arguments:"
	puts "\t\t\[-x MAXX\] \[-y MAXY\]"
	puts "\toptional arguments:"
	puts "\t\t\[-cp conn pattern\] \[-sc scenario\] \[-nn
nodes\]"
	puts "\t\t\[-seed seed\] \[-stop sec\] \[-tr
tracefile\]\n"
}


proc getopt {argc argv} {
	global opt
	lappend optlist cp nn seed sc stop tr x y

	for {set i 0} {$i < $argc} {incr i} {
		set arg [lindex $argv $i]
		if {[string range $arg 0 0] != "-"} continue

		set name [string range $arg 1 end]
		set opt($name) [lindex $argv [expr $i+1]]
	}
}

Thanks,

Ahmad

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From kopi@stanford.edu  Tue Mar 12 05:16:02 2002
From: kopi@stanford.edu (Milind Dilip Kopikare)
Date: Tue Mar 12 05:16:02 2002
Subject: [ns] Exponential traffic in wireless
Message-ID: 

Attaching Manish's question which I dug up from the archive. Does anyone
know how to start exponential traffic on a wireless system?

thx,
Milind

============================

hello Prasant,

thanks for u'r help, but its not so simple, in the case statement u've to
add the function void format_exp(Packet *p, int offset ), now this
function
u've to write in the cmu-trace.cc and then u've to make an entry
corresponding to it in the cmu-trace.h file

now my problem is that what should be written in the body of the function

CMUTrace::format_exp(Packet *p, int offset) in the file cmu-trace.cc

i hope now u can get my problem.

thanks for u'r effort to help me.

bye


From digitallh@yahoo.com  Tue Mar 12 05:16:15 2002
From: digitallh@yahoo.com (Michael Y)
Date: Tue Mar 12 05:16:15 2002
Subject: [ns] Random Number Generator
Message-ID: <20020312084336.90514.qmail@web14906.mail.yahoo.com>

Dear all,
  I noticed NS implemented its own random number
generator. May I ask why NS didn't use the randon
number generator from ? What is the difference
between them? 
  Can we use the pseudo-random generator from DEK in
NS?

Thanks for your help
michael


__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


From gdemarco@unisa.it  Tue Mar 12 07:45:01 2002
From: gdemarco@unisa.it (giuseppe de marco)
Date: Tue Mar 12 07:45:01 2002
Subject: [ns] more application data over a node
Message-ID: <20020312164104.1f04f015.gdemarco@unisa.it>

A simple question?

If i wnat to attach a lot of data sources (i.e. Traffic source) to a node (i.e. DropTail or Red) how can I do it? Can I attach two or more application data over two or more agent, which are attached on a node?

Giuseppe


-- 
Giuseppe De Marco
Department of Electrical and Information Engineering
University of Salerno - Italy
tel: +39 089 964012
mail: gdemarco@unisa.it
personal: inggdm@tiscalinet.it


From tmaneesh@cedt.iisc.ernet.in  Tue Mar 12 08:35:02 2002
From: tmaneesh@cedt.iisc.ernet.in (Maneesh Tewari)
Date: Tue Mar 12 08:35:02 2002
Subject: [ns] RE: Exponential traffic in wireless
Message-ID: 

hello Milind,

i've kept those functions in the cmu-trace.cc blank just to proceed ahead
till the time any other well-knowledged person doesn't gives its proper
solution, but this is surely not correct.

bye

                  
                        ---maneesh


-----Original Message-----
From: Milind Dilip Kopikare [mailto:kopi@Stanford.EDU]
Sent: Tuesday, March 12, 2002 2:00 PM
To: tmaneesh@cedt.iisc.ernet.in
Cc: ratancha@uiuc.edu; ns-users@isi.edu
Subject: Exponential traffic in wireless


Attaching Manish's question which I dug up from the archive. Does anyone
know how to start exponential traffic on a wireless system?

thx,
Milind

============================

hello Prasant,

thanks for u'r help, but its not so simple, in the case statement u've to
add the function void format_exp(Packet *p, int offset ), now this
function
u've to write in the cmu-trace.cc and then u've to make an entry
corresponding to it in the cmu-trace.h file

now my problem is that what should be written in the body of the function

CMUTrace::format_exp(Packet *p, int offset) in the file cmu-trace.cc

i hope now u can get my problem.

thanks for u'r effort to help me.

bye


From haldar@ISI.EDU  Tue Mar 12 09:10:03 2002
From: haldar@ISI.EDU (Padmaparna Haldar)
Date: Tue Mar 12 09:10:03 2002
Subject: [ns] RE: Exponential traffic in wireless
In-Reply-To: 
Message-ID: 

The pkt-hdr specific format functions provide methods for dumping 
additional info that is stored in that specific pkt-hdr.
If you see the cases for GAF or diffusion, for which no extra info is  
written in the tracefile, they donot have any format_* methods defined.
I guess the same could be done for expo pkt-hdrs, unless you want some 
specific data, in which case they should be defined under a format_expo 
method.

Otherwise as Milind suggested, simply adding the foll line for 
CMUTrace::format(..) should work:

...
case PT_DIFF:
case PT_GAF:
case PT_EXPO:
	break;
default:
 ...

--Padma

On Tue, 12 Mar 2002, Maneesh Tewari wrote:


> 
> hello Milind,
> 
> i've kept those functions in the cmu-trace.cc blank just to proceed ahead
> till the time any other well-knowledged person doesn't gives its proper
> solution, but this is surely not correct.
> 
> bye
> 
>                   
>                         ---maneesh
> 
> 
> -----Original Message-----
> From: Milind Dilip Kopikare [mailto:kopi@Stanford.EDU]
> Sent: Tuesday, March 12, 2002 2:00 PM
> To: tmaneesh@cedt.iisc.ernet.in
> Cc: ratancha@uiuc.edu; ns-users@isi.edu
> Subject: Exponential traffic in wireless
> 
> 
> Attaching Manish's question which I dug up from the archive. Does anyone
> know how to start exponential traffic on a wireless system?
> 
> thx,
> Milind
> 
> ============================
> 
> hello Prasant,
> 
> thanks for u'r help, but its not so simple, in the case statement u've to
> add the function void format_exp(Packet *p, int offset ), now this
> function
> u've to write in the cmu-trace.cc and then u've to make an entry
> corresponding to it in the cmu-trace.h file
> 
> now my problem is that what should be written in the body of the function
> 
> CMUTrace::format_exp(Packet *p, int offset) in the file cmu-trace.cc
> 
> i hope now u can get my problem.
> 
> thanks for u'r effort to help me.
> 
> bye
> 

-- 
------------------------------------------------ 
Be true to your work, your word, and your friend. 
--Thoreau

Padmaparna Haldar


From haldar@ISI.EDU  Tue Mar 12 09:20:02 2002
From: haldar@ISI.EDU (Padmaparna Haldar)
Date: Tue Mar 12 09:20:02 2002
Subject: [ns] What is the difference between target_->recv(p) & send(p
 ,0)?
Message-ID: 

On Tue, 12 Mar 2002, Martin McKenny wrote:


> Hi Padmaparna,
> 
> I am having difficulties with the classifier in NS-2. My simulation is set
> up to describe a link protocol with a central crossbar switch. The switch
> has multiple agents attached to it and each link has two agents associated
> with it. When one of the agents attached to the switch recieves a packet, it
> must acknowledge that packet on the same link. It must then forward that
> packet to an outgoing link which is serviced by another agent attached to
> THE SAME NODE. How can I set up the central switching node so that it
> forwards packets between agents of the same node? I think I need to change
> the classifier but I am unsure of exactly what needs altered.

Since your agents are going to be connected to different 
ports (hopefully) in the given node, the port-classifier will 
automatically take care of demux'ing pkts addressed to multiple 
agents. 
so all you need to do is connect each agent pair up. It shouldnot matter 
if they are connected to the same node or not.
Hth,
--Padma

> 
> Hope I have explained this clearly enough and thanks for any help you can
> offer.
> 
> Martin McKenny
> 
> -----Original Message-----
> From: Padmaparna Haldar
> To: Christophe Jelger
> Cc: NS
> Sent: 3/11/02 7:35 PM
> Subject: Re: [ns] What is the difference between target_->recv(p) &
> send(p,0)?
> 
> 
> 
> when an agent gets attached to a node, the target_ points to the node 
> entry_ which in default condition points to the the classifier.
> when you connect two agents you are basically setting up the destination
> 
> to which the agent will send pkts to. 
> so assumimg yr agent was connected to a node, it can send 
> pkts to a given dest address, provided there are corr. routing entries
> for 
> that address in the classifier.
> --Padma
> 
> On Mon, 11 Mar 2002, Christophe Jelger wrote:
> 
> 
> > 
> > Hi,
> > 
> > And what happens when objects are not connectors ? For example if an 
> > agent simply wants to send packets to a given dest address ? (assuming
> 
> > that there is a routing entry in the classifier) ? For example, I want
> 
> > my agent to send a packet to the well known ALL-PIM-ROUTERS address
> ... 
> > how shall I proceed ? (the agent is not connected so there is no
> target_)
> > 
> > Thanks
> > 
> > Christophe
> > 
> > 
> > Padmaparna Haldar wrote:
> > 
> > >
> > >
> > >Most objects in ns are Connector objects having a target_ to SEND
> pkts to. 
> > >It can also RECV pkts. recv and send functions in the Connector class
> are 
> > >defined as follows:
> > >
> > >void recv(Packet*, Handler* callback = 0);
> > >inline void send(Packet* p, Handler* h) { target_->recv(p, h); }
> > >
> > >so as you can see, send(p,h) and target_->recv(p) are basically doing
> the 
> > >same thing.
> > >
> > >--Padma
> > >
> > >On Wed, 6 Mar 2002, Mustafa Tasdemir wrote:
> > >
> > >
> > >>hello,
> > >>
> > >>can someone tell what is the difference between these commands when
> used in agent:
> > >>
> > >>Application---->UDP------->Node
> > >>                           |
> > >>                           |
> > >>                        commands are used in UDP agents
> > >>  1.. target_->recv(p);    // target defined as Process *target_; 
> > >>  2.. send(p,0);
> > >>I think, the first one is a direct call and the second one is an
> indirect call. Can we also call the send() method as a SAP (Service
> Access Point)?
> > >>
> > >>Mustafa TASDEMIR
> > >>
> > >>
> > >>
> > >>
> > >
> > 
> > 
> 
> 

-- 
------------------------------------------------ 
Be true to your work, your word, and your friend. 
--Thoreau

Padmaparna Haldar



From francesc@lsihp19.epfl.ch  Tue Mar 12 09:35:03 2002
From: francesc@lsihp19.epfl.ch (Francesc Font (EPFL))
Date: Tue Mar 12 09:35:03 2002
Subject: [ns] How to know the PHB codepoint from the trace files?
References: 
Message-ID: <3C8E3C20.298F1A82@lsihp19.epfl.ch>

Do you know if the trace file, when applying diffserv
or intserv is different from the "normal" trace file format? It means,
Is there in ns some trace-diffserv or trace-intserv support?

Thanks in advance

"Karina Karla C. de Oliveira" wrote:

> Hello All,
>
> The PHB codepoint of a packet isn't write in the trace files. The second
> question, I would like to know too.
>
> Karina Oliveira.
>
> On Mon, 11 Mar 2002, Lluís Fàbrega i Soler wrote:
>
> >
> > Hi all,
> >
> > I am using the Diffserv module, and I want to know the PHB codepoint
> > that is assigned to a packet.
> > Is this information written in the trace files? And if not, what should
> > I modify in order to see this information in the trace files?
> >
> > Thanks!
> >
> > Lluís
> >
> >
> >


From francesc@lsihp19.epfl.ch  Tue Mar 12 09:45:02 2002
From: francesc@lsihp19.epfl.ch (Francesc Font (EPFL))
Date: Tue Mar 12 09:45:02 2002
Subject: [ns] more application data over a node
References: <20020312164104.1f04f015.gdemarco@unisa.it>
Message-ID: <3C8E3E58.57F155E8@lsihp19.epfl.ch>

Do you know if it is supported some kind of input
queueing in a node. As I know from the node and
link ns models only output queueing is possible.
Can packets really be dropped in the entry of a node
due to a high bit rate (because p.e too much applications
are attached to the same node)?

Thanks in advance!


From kopi@stanford.edu  Tue Mar 12 11:40:09 2002
From: kopi@stanford.edu (Milind Dilip Kopikare)
Date: Tue Mar 12 11:40:09 2002
Subject: [ns] RE: Exponential traffic in wireless
In-Reply-To: 
Message-ID: 

Yeah, I did that too. But it dosen't work. I guess by not correct you mean
it dosent work either?


On Tue, 12 Mar 2002, Maneesh Tewari wrote:

> hello Milind,
>
> i've kept those functions in the cmu-trace.cc blank just to proceed ahead
> till the time any other well-knowledged person doesn't gives its proper
> solution, but this is surely not correct.
>
> bye
>
>
>                         ---maneesh
>
>
> -----Original Message-----
> From: Milind Dilip Kopikare [mailto:kopi@Stanford.EDU]
> Sent: Tuesday, March 12, 2002 2:00 PM
> To: tmaneesh@cedt.iisc.ernet.in
> Cc: ratancha@uiuc.edu; ns-users@isi.edu
> Subject: Exponential traffic in wireless
>
>
> Attaching Manish's question which I dug up from the archive. Does anyone
> know how to start exponential traffic on a wireless system?
>
> thx,
> Milind
>
> ============================
>
> hello Prasant,
>
> thanks for u'r help, but its not so simple, in the case statement u've to
> add the function void format_exp(Packet *p, int offset ), now this
> function
> u've to write in the cmu-trace.cc and then u've to make an entry
> corresponding to it in the cmu-trace.h file
>
> now my problem is that what should be written in the body of the function
>
> CMUTrace::format_exp(Packet *p, int offset) in the file cmu-trace.cc
>
> i hope now u can get my problem.
>
> thanks for u'r effort to help me.
>
> bye
>


From tmaneesh@cedt.iisc.ernet.in  Tue Mar 12 11:55:01 2002
From: tmaneesh@cedt.iisc.ernet.in (Maneesh Tewari)
Date: Tue Mar 12 11:55:01 2002
Subject: [ns] RE: Exponential traffic in wireless
Message-ID: 

hello,

no its working with me, by "not correct" i meant that this is not the
correct way, but as in one mail Padma Haldar said that this is the right
method to proceed so we can go ahead without worring too much.

u just do like this in cmu-trace.cc file

CMUTrace::format_exp(Packet *p, int offset) 
 {
   // leave it blank

          }

pay attention that u've to also make the corresponding entries in the
cmu-trace.h file and then only u should recompile.

it must work.

bye

                                    ---maneesh
  


-----Original Message-----
From: Milind Dilip Kopikare [mailto:kopi@Stanford.EDU]
Sent: Wednesday, March 13, 2002 1:05 AM
To: Maneesh Tewari
Cc: 'ns-users@isi.edu'
Subject: RE: Exponential traffic in wireless


Yeah, I did that too. But it dosen't work. I guess by not correct you mean
it dosent work either?


On Tue, 12 Mar 2002, Maneesh Tewari wrote:

> hello Milind,
>
> i've kept those functions in the cmu-trace.cc blank just to proceed ahead
> till the time any other well-knowledged person doesn't gives its proper
> solution, but this is surely not correct.
>
> bye
>
>
>                         ---maneesh
>
>
> -----Original Message-----
> From: Milind Dilip Kopikare [mailto:kopi@Stanford.EDU]
> Sent: Tuesday, March 12, 2002 2:00 PM
> To: tmaneesh@cedt.iisc.ernet.in
> Cc: ratancha@uiuc.edu; ns-users@isi.edu
> Subject: Exponential traffic in wireless
>
>
> Attaching Manish's question which I dug up from the archive. Does anyone
> know how to start exponential traffic on a wireless system?
>
> thx,
> Milind
>
> ============================
>
> hello Prasant,
>
> thanks for u'r help, but its not so simple, in the case statement u've to
> add the function void format_exp(Packet *p, int offset ), now this
> function
> u've to write in the cmu-trace.cc and then u've to make an entry
> corresponding to it in the cmu-trace.h file
>
> now my problem is that what should be written in the body of the function
>
> CMUTrace::format_exp(Packet *p, int offset) in the file cmu-trace.cc
>
> i hope now u can get my problem.
>
> thanks for u'r effort to help me.
>
> bye
>


From najma_ismat@hotmail.com  Tue Mar 12 13:50:02 2002
From: najma_ismat@hotmail.com (Najma Ismat)
Date: Tue Mar 12 13:50:02 2002
Subject: [ns] Getting Low TCP throughput when elfn Technique is used. Urgent help Needed!!!
Message-ID: 

hi,
I am simulating ad hoc network using dsr. i have already calculated tcp 
throughput using dsr for 50 mobility patterns moving at four different 
speeds. now i have modified my ns2 code by adding elfn technique. the 
results i m getting by running the script on modified code are very 
differnt. according to G.Holland & N.Vaidya mobicom 99 paper when elfn 
technique is added with TCP TCP throughput increases for all mobility 
patterns . But in my case I am getting fluctuating results. Some mobility 
patterns gives high tcp throughput and some are showing low tcp throughput. 
I have used same patterns for modified code as I had used earlier for my 
simulation. Can anyone guide me why I am getting fluctuating result or what 
else should I do to get the higher value of TCP throughput for all the 
mobilty patterns.
Thanks in advance.
Najma Ismat

_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com


From cghsh@hotmail.com  Tue Mar 12 15:10:01 2002
From: cghsh@hotmail.com (Guanghai Cai)
Date: Tue Mar 12 15:10:01 2002
Subject: [ns] How to install ns-2.1b8 on redhat7.2?
Message-ID: 

This is a multi-part message in MIME format.

------=_NextPart_000_0012_01C1C9F0.6E030430
Content-Type: text/plain;
	charset="gb2312"
Content-Transfer-Encoding: quoted-printable

Hi,
    I am a newer of ns. Could you please tell me the detail step by step =
procedure to install ns-2.1b8 on redhat7.2 platform.
    Kindly give me the step by step procedure after downloading the =
installable files...
    Thank you very much.

Guanghai cai
20020312

------=_NextPart_000_0012_01C1C9F0.6E030430
Content-Type: text/html;
	charset="gb2312"
Content-Transfer-Encoding: quoted-printable









Hi,
    I am a newer of ns. Could you =
please=20
tell me the detail step by step procedure to install ns-2.1b8=20
on redhat7.2 platform.
    Kindly give me the =
step by=20
step procedure after downloading the installable=20
files...
    Thank you very much.

Guanghai =

cai


20020312


------=_NextPart_000_0012_01C1C9F0.6E030430--


From ARISBCN@terra.es  Tue Mar 12 15:10:17 2002
From: ARISBCN@terra.es (Aristides Rib)
Date: Tue Mar 12 15:10:17 2002
Subject: [ns] Problems attaching Taffic TRACEFILES
Message-ID: <003001c1ca1a$edb4e3e0$3ab298c1@pc01>

Hi all,

I'm trying to use a trace file as a Traffic agent but I'm getting the
following error when running ns:

    warning: no class variable Tracefile::debug_
      see tcl-object.tcl in tclcl for info about this warning.

The code I'm using to attach the tracefile is listed below and the tracefile
is called traza. This tracefile is located on the same folder I have the
.tcl file and where I'm executing ns.

set tfile [new Tracefile]
$tfile filename traza
set t1 [new Application/Traffic/Trace]
$t1 attach-tracefile $tfile

Please, is very urgent for me to obtain a way to solve this problem.

I hope somebody can help me.
Thanks in advance.

Aris.


From rlaw@vt.edu  Tue Mar 12 15:50:02 2002
From: rlaw@vt.edu (Raymond Law)
Date: Tue Mar 12 15:50:02 2002
Subject: [ns] Get e2e delay in tcl script
Message-ID: <5.1.0.14.0.20020312185130.039d3238@mail.vt.edu>

I know a script can be used to calculate the end-to-end delay from the 
trace file.  But is it possible to get the end-to-end delay in the tcl 
script itself?  For example, is there a callback when receiving a 
packet?  That would be useful if the simulation needs to respond 
differently depending on the end-to-end delay.

Thanks.
Ray.


From mhefeeda@cs.purdue.edu  Tue Mar 12 17:15:01 2002
From: mhefeeda@cs.purdue.edu (Mohamed M. Hefeeda)
Date: Tue Mar 12 17:15:01 2002
Subject: [ns] rtp examples
Message-ID: <3C8EA6F0.86686147@cs.purdue.edu>

Hi;

I am trying to use RTP in some ns2 simulation.
Are there any examples on how to use RTP in
unicast mode?


Appreciate your help.

Mohamed


From ji@pa.dec.com  Tue Mar 12 17:30:04 2002
From: ji@pa.dec.com (Minwen JI)
Date: Tue Mar 12 17:30:04 2002
Subject: [ns] Re: license?
References: <3C8EAA35.72D67865@pa.dec.com>
Message-ID: <3C8EAADC.A5BFF81F@pa.dec.com>

> 
> Hi,
> 
> I am considering contributing code to ns-2 that simulates an Internet
> cloud and overlay networks (i.e. tunnels) over the cloud. Could you
> please let me know which type of license (e.g. BSD or GNU) applies to
> ns-2?
> 
> Thanks,
> Minwen
> 
> P.S. I am not sure if this is the right email address to send my
> question to, so I don't include much detail here. But I will be happy to
> answer any questions you may have regarding the code contribution.


From jw@erg.sri.com  Tue Mar 12 17:50:01 2002
From: jw@erg.sri.com (Julie Wong)
Date: Tue Mar 12 17:50:01 2002
Subject: [ns] Re: (no subject)
References: <3C869384.1F3F178@msu.edu> <3C8D2F64.508A6B98@erg.sri.com> <3C8D35F1.8C9B20E0@msu.edu> <3C8D6939.71FE31AE@erg.sri.com>
Message-ID: <3C8EAF15.6513C7A3@erg.sri.com>

Hello Gautam and other ns-users,

Just want to follow up on this thread a little in case this should help
others who might be trying to do the same thing.  I looked into the
mac-802_11 implementation (ns2 version 2.1b8)
and saw that my problem of not getting any data pkts through is
because the timeout value set for mac state CTS is too small in
the case of 11Mbps.  My raw data pkt size is 512 bytes and
the CTS window is not large enough for the receiver to stay in the
CTS state to begin reception.  Anyhow, I see 2 ways of handling
this:

1.  Turn off RTS/CTS by changing the RTSThreshold
2.  Adjust the CTS timeout value.  I checked that in the 2Mbps
    case it is 0.002661 and I suppose one could use that instead
    of making the timeout value a function of the link bandwidth.

Also there is a paper by Mineo Takai et. al. from UCLA that talks
about the incompleteness of the ns2 physical model for wireless
simulations.

    ftp://pcl.cs.ucla.edu/pub/papers/mobihoc.ps

I think it is worthwhile to read the paper so we know what exactly
we are using or not using for our simulations.

Julie Wong

Julie Wong wrote:

> Gautam,
>
> I was a bit rushed to say that it works.  I do see that CTS and RTS
> messages get through with a link speed of 11mbs.  However, when
> it comes to data packets, for some reason, if a packet has more than
> 1 hop, it is dropped at the second hop.  I think the dropping occurs
> in mac-802_11.cc line 1290 discard(p, DROP_MAC_BUSY) in the module called
> recvDATA().  So I take back what I said before that
> setting the Mac bandwidth to 11mbs works.
>
> BTW, when the speed is 9mbs or less, things are ok.  Only when it
> gets up to 10mbs or more, then the above happens.  Also, data
> packets on the first hop is ok.   Any idea why this might be the
> case?
>
> Thanks,
>
> Julie Wong
>
> Gautam - Chakrabarti wrote:
>
> > Julie Wong wrote:
> > >
> > > Thanks Gautam for the extra information.  I set the bandwidth at
> > > the Mac level and it works now.  By the way, setting it to 11mbs
> > > is ok.  I got something very close (close enough) to a 11mbs link.
> >
> > Thanks for clarifying, actually I am not exactly working on it for
> > sometime,
> > but I was told by someone that the throughput is not getting to the
> > value,
> > although the bandwidth has the value 11. You might want to look at that.
> > If
> > that is true, that probably means there are some other controlling
> > parameters.
> >
> > rgds,
> >
> > Gautam
> >
> > >
> > > Julie Wong
> > >
> > > Gautam - Chakrabarti wrote:
> > >
> > > > Hi,
> > > >
> > > > You may look into the following messages which appear to be relevant.
> > > >
> > > > Hth,
> > > > gautam
> > > >
> > > >  -------------------------
> > > > >
> > > > > Hi,
> > > > > I think that 'bandwidth' variable of Phy/WirelessPhy object is useful
> > > > > only
> > > > > if you enable the Errormodel: it is used to compute txtime when decrease
> > > > > node energy (check wirelessphy.cc).
> > > > > The variable you have to change is the 'bandwidth' of Mac object! This
> > > > > is
> > > > > the variable that real affect txtime of a packet.
> > > > > By default, it is set as 2Mb (as the bandwidth of Phy/WirelessPhy; check
> > > > > ~ns2/tcl/lib/ns-default.tcl, ~ns2/tcl/lan/ns-mac.tcl). You can change it
> > > > > and
> > > > > see what happens in the tracefile using
> > > > >
> > > > > Mac/802_11 set bandwidth_ 
> > > > >
> > > > > Federico Bertocchi
> > > > >
> > > > > The above is from  "Federico Bertocchi" 
> > > > >
> > > > > Hi!
> > > > >
> > > > > I do not use a recent snapshot, but ns2.1b8a.
> > > > > In my opinion the bandwidth_ variable of Wireless Phy influences only
> > > > > the
> > > > > energy related part of communication. For example, when using IEEE
> > > > > 802.11, the
> > > > > mac-802_11 has its own bandwidth_ variable, which controls the
> > > > > transmission
> > > > > data rate.
> > > > > If I am wrong, please let me know!
> > > > >
> > > > > Regards,
> > > > > Patrick
> > > > >
> > > > >
> > > > > --- from  Patrick Peschlow 
> > > > >
> > > > >
> > > > >
> >
> > --
> > ***************************************************************
> > Gautam Chakrabarti
> > MS Student
> > Computer Sc. and Engg Department
> > Michigan State University
> >
> > Email : chakra10@msu.edu
> > Home Page : http://www.cse.msu.edu/~chakra10
> >
> > Why is there so much month left at the end of the money?
> >                 --- John Barrymore
> > ***************************************************************


From engp1882@nus.edu.sg  Wed Mar 13 05:15:02 2002
From: engp1882@nus.edu.sg (Govindan Saravanan)
Date: Wed Mar 13 05:15:02 2002
Subject: [ns] how to find node address
Message-ID: <9C4C56CDF89E0440A6BD571E76D2387F04B7F690@exs23.ex.nus.edu.sg>

Dear All,

I am working with an ICMP agent. Given a packet's source address I need to
find the node corresponding to it and the agent at that node. I read
somewhere that I have to use a function in the Address class. But I don't
understand how to go about doing this.

I hope someone can help me with this.

Thanks you.

Saravanana


From xuanc@ISI.EDU  Wed Mar 13 05:15:23 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Wed Mar 13 05:15:23 2002
Subject: [ns] Re: help again.. urgent
In-Reply-To: <200203130614070930.0022A6B2@smtp.idola.net.id>
Message-ID: 

Have you specified the scenario file in your script? Please refer to the 
tutorial for further information. Thanks.

-chen

On Wed, 13 Mar 2002, Lukman Hakim wrote:

> Hello again...
> 
> I made this script.. (attached) but there's somthing wrong.. and I dont know what was that.. I hope you could help me ..
> 
> My topology is :
> 
>   			      0		
> 			
> 			    O	
> 			   /
> 	O----------------------O
>    	 |  \                   / |
> 	 |    \	         /	 |
> 	 |      \	       /	 |
> 	 |        \       /	 |
> 	 |	O	 |
> 	 |        /      \       |
> 	 |      /          \     |
> 	 |    /              \   |
> 	 |  /                 \  |
> 	 O--------------------O
>             /                         \
>          O       		   O	
> 
>    0        0
> 
> 
> 
> An error message occured when I run it :
> 
> [lhakim@iglo scenario]$ ns simulasi.tcl
> num_nodes is set 3
> warning: Please use -channel as shown in tcl/ex/wireless-mitf.tcl
> *** NOTE: no connection pattern specified.
> *** NOTE: no scenario file specified.
> Starting Simulation...
> warning: Route to base_stn not known: dropping pkt
> warning: Route to base_stn not known: dropping pkt
> warning: Route to base_stn not known: dropping pkt
> warning: Route to base_stn not known: dropping pkt
> warning: Route to base_stn not known: dropping pkt
> warning: Route to base_stn not known: dropping pkt
> warning: Route to base_stn not known: dropping pkt
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> MAC_802_11: accessing MAC cache_ array out of range (src 8, dst 2, size 3)!
> [suppressing additional MAC cache_ warnings]
> ns: _o208 tunnel-exit 12582912: can't read "TunnelExit_(12582912)": no such
> element in array
>     while executing
> "subst $[subst $var"
>     (procedure "_o205" line 5)
>     (Object next line 5)
>     invoked from within
> "_o205 next TunnelExit_(12582912)"
>     ("eval" body line 1)
>     invoked from within
> "eval $self next $args"
>     (procedure "_o205" line 18)
>     (Agent set line 18)
>     invoked from within
> "[$node_ set regagent_] set TunnelExit_($mhaddr)"
>     (procedure "_o208" line 3)
>     (MIPEncapsulator tunnel-exit line 3)
>     invoked from within
> "_o208 tunnel-exit 12582912"
> 
> 
> Could you tell me whats wrong withmy script... pls...
> 
> Regards,
> 
> -lukman-
> 

-- 
Xuan Chen
USC/ISI


From engp1882@nus.edu.sg  Wed Mar 13 05:15:39 2002
From: engp1882@nus.edu.sg (Govindan Saravanan)
Date: Wed Mar 13 05:15:39 2002
Subject: [ns] Working of "Address::instance()"
Message-ID: <9C4C56CDF89E0440A6BD571E76D2387F04B7F691@exs23.ex.nus.edu.sg>

Dear All,

This is in continuation of my last query. I found that to convert an address
to a node number, the following has been used,

int srcnode = hdrip->src_ >> Address::instance().NodeShift_[1];

However I don't really understand how it works. I get the following error

no match for 'ns_addr_t& >> int &'

Since I don't understand how Address::instance() works I can't to anything
about it. 

Could someone give me some insight into this please?

Thank you.

Saravanan Govindan


From engg_sarah@yahoo.com  Wed Mar 13 05:15:52 2002
From: engg_sarah@yahoo.com (Syeda Noor-ul-ain)
Date: Wed Mar 13 05:15:52 2002
Subject: [ns] a piece of code
Message-ID: <20020313070113.62480.qmail@web20310.mail.yahoo.com>

--0-1122431551-1016002873=:62467
Content-Type: text/plain; charset=us-ascii


hi all,

there is this strange piece of code in wireless-newnode-energy.tcl in ~/tcl/ex:

could any one tell me how it works and what does it achieve?

Thanks,

Syeda Noor



---------------------------------
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
--0-1122431551-1016002873=:62467
Content-Type: text/html; charset=us-ascii


hi all,


there is this strange piece of code in wireless-newnode-energy.tcl in ~/tcl/ex:


could any one tell me how it works and what does it achieve?


Thanks,


Syeda Noor

Do You Yahoo!?

Try FREE Yahoo! Mail - the world's greatest free email!
--0-1122431551-1016002873=:62467--


From engg_sarah@yahoo.com  Wed Mar 13 05:16:08 2002
From: engg_sarah@yahoo.com (Syeda Noor-ul-ain)
Date: Wed Mar 13 05:16:08 2002
Subject: [ns] energy trace
Message-ID: <20020313073501.6537.qmail@web20301.mail.yahoo.com>

--0-1019570762-1016004901=:6153
Content-Type: text/plain; charset=us-ascii


hi all,

i am trying to implement energy model in my tcl file, this is the output trace file generated by my code but i dont see any mention of the energy levels of nodes in the trace file or am i missing something?

Thanks,

Syeda Noor



---------------------------------
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
--0-1019570762-1016004901=:6153
Content-Type: text/html; charset=us-ascii


hi all,


i am trying to implement energy model in my tcl file, this is the output trace file generated by my code but i dont see any mention of the energy levels of nodes in the trace file or am i missing something?


Thanks,


Syeda Noor

Do You Yahoo!?

Try FREE Yahoo! Mail - the world's greatest free email!
--0-1019570762-1016004901=:6153--


From engg_sarah@yahoo.com  Wed Mar 13 05:16:22 2002
From: engg_sarah@yahoo.com (Syeda Noor-ul-ain)
Date: Wed Mar 13 05:16:22 2002
Subject: [ns] energy model archives
Message-ID: <20020313082926.28567.qmail@web20303.mail.yahoo.com>

--0-983428731-1016008166=:27424
Content-Type: text/plain; charset=us-ascii

hi all,

how can i look for the energy model archives?

Syeda Noor



---------------------------------
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
--0-983428731-1016008166=:27424
Content-Type: text/html; charset=us-ascii

hi all,

how can i look for the energy model archives?

Syeda Noor

Do You Yahoo!?

Try FREE Yahoo! Mail - the world's greatest free email!
--0-983428731-1016008166=:27424--


From ns@cdnet.edu.cn  Wed Mar 13 05:16:35 2002
From: ns@cdnet.edu.cn (Xiang)
Date: Wed Mar 13 05:16:35 2002
Subject: [ns] About self-similar source
Message-ID: <001001c1ca70$997034a0$4a0e73ca@uestc>

This is a multi-part message in MIME format.

------=_NextPart_000_000D_01C1CAB3.A782ABC0
Content-Type: text/plain;
	charset="gb2312"
Content-Transfer-Encoding: base64

SGkNCkkgd291bGQgbGlrZSB0byBrbm93IGlmIHRoZXJlIGFyZSBzb21lIG1vZHVsZXMgb24gZ2Vu
ZXJhdGUgc2VsZi1zaW1pbGFyIHRyYWZmaWM/DQpUaGFuayB5b3UgdmVyeSBtdWNoIHRvIGhlbHAg
bWUhDQpYaWFuZw0K

------=_NextPart_000_000D_01C1CAB3.A782ABC0
Content-Type: text/html;
	charset="gb2312"
Content-Transfer-Encoding: base64
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------=_NextPart_000_000D_01C1CAB3.A782ABC0--


From ºúÍ¿" 

hi,everyone
    In order to access node in link ,i add following codes in ns-link.lib
 
SimpleLink instproc  init { }
   
                           .
                           .
                           .
                $queue_ set-link [$self set toNode_]
                           .
                           .
                           .
with modified codes, ns can be compiled successfully,but when the tcl
program
is running,the following err was reported:

    (_o66 cmd line 1)
    invoked from within
"_o66 cmd set-link _o19"
    invoked from within
"catch "$self cmd $args" ret"
    (procedure "_o66" line 2)
    (SplitObject unknown line 2)
    invoked from within
"$queue_ set-link [$self set toNode_]"
    (procedure "_o67" line 19)
    (SimpleLink init line 19)
    invoked from within
"_o67 init _o10 _o19 10Mb 5ms _o66"
    (Class create line 1)
    invoked from within
"SimpleLink create _o67 _o10 _o19 10Mb 5ms _o66"
    invoked from within
"catch "$className create $o $args" msg"
    (procedure "new" line 3)
    invoked from within
"new SimpleLink     $n1 $n2 $bw $delay $q"
    ("default" arm line 2)
    invoked from within
"switch -exact $qtypeOrig {
RTM {
set c [lindex $args 1]
set link_($sid:$did) [new CBQLink        $n1 $n2 $bw $delay $q $c]
}
CBQ -
CBQ/WRR {
if {[llen..."
    (procedure "_o3" line 33)
    (Simulator simplex-link line 33)
    invoked from within
"_o3 simplex-link _o10 _o19 10Mb 5ms RED"
    ("eval" body line 1)
    invoked from within
"eval $self simplex-link $n1 $n2 $bw $delay $type $args"
    (procedure "_o3" line 8)
    (Simulator duplex-link line 8)
    invoked from within
"$ns duplex-link $s0 $LSR3 10Mb 5ms RED"
    (file "example.tcl" line 50)


help me !
thanks in advance








--http://www.eyou.com
--Îȶ¨¿É¿¿µÄÃâ·Ñµç×ÓÐÅÏä  ÓïÒôÓʼþ  Òƶ¯ÊéÇ©  ÈÕÀú·þÎñ  ÍøÂç´æ´¢...ÒÚÓÊδ¾¡



From jlores@mat.upc.es  Wed Mar 13 05:17:04 2002
From: jlores@mat.upc.es (Jorge Lores)
Date: Wed Mar 13 05:17:04 2002
Subject: [ns] Video traces over SRM, MFTP and PGM agents
Message-ID: <003b01c1ca76$b9ac8f10$ed275393@educa3>

This is a multi-part message in MIME format.

------=_NextPart_000_0038_01C1CA7F.1B557FD0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Hi everybody,
As far now I've used the=20
set s [new Application/Traffic/Trace]

with UDP agents and everything has work all right. Now I would like to =
add this kind of traffic to a SRM, MFTP or PGM agent. I don't know if =
it'spossible. I've tried to do it with SRM but the video packets appears =
as "undefined" in the NAM and in the out.tr, and the files srmEvents.tr =
and the SrmStats.tr are empty. Does anybody know what can I do???=20

Thanks in advance for your support

Best Regards

Jorge Lores






------=_NextPart_000_0038_01C1CA7F.1B557FD0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable









Hi everybody,


As far now I've used the 

set s [new Application/Traffic/Trace]


with UDP agents and everything has work all right. Now I would like =
to add=20
this kind of traffic to a SRM, MFTP or PGM agent. I don't know if =
it'spossible.=20
I've tried to do it with SRM but the video packets appears as =
"undefined" in the=20
NAM and in the out.tr, and the files srmEvents.tr and the SrmStats.tr =
are empty.=20
Does anybody know what can I do??? 


Thanks in advance for your support


Best Regards


Jorge Lores


 


 


------=_NextPart_000_0038_01C1CA7F.1B557FD0--


From Sergio.Herreria@det.uvigo.es  Wed Mar 13 05:17:21 2002
From: Sergio.Herreria@det.uvigo.es (Sergio.Herreria@det.uvigo.es)
Date: Wed Mar 13 05:17:21 2002
Subject: [ns] link with two queues
Message-ID: 

How could I create a link with two sequenced queues?

head_
 ----> enqT_ ----> queue1_ ----> queue2_ ----> deqT_ ----> link_ ....
			|             |
			------------------>drophead_ ----> drpT_
Thank you very much
Sergio Herreria


From jonasn@diku.dk  Wed Mar 13 05:17:35 2002
From: jonasn@diku.dk (Jonas Nielsen)
Date: Wed Mar 13 05:17:35 2002
Subject: [ns] Printing out a node's buffers
Message-ID: 

Are there other ways print to out a specified nodes buffers during the
simulation than inserting code that write to a file/tty in
the dsragent.cc file.

I'm particularly interested in showing different buffers (e.g. the route cache) in the DSR protocol.


From studyns" 

hi,everyone:
     In order to access node via link,I add following codes in ns-link.tcl:

SimpleLink instproc init { src dst bw delay q {lltype "DelayLink"} } {
                 :
                 :
             $queue_ set-link [$self set toNode_]      
                 :
                 :
}
             
    I have define set-link method in  for dsred.cc

    With modified codes,ns can be compiled successfully,but the following
errors are reported if the tcl program is running:

    (_o66 cmd line 1)
    invoked from within
"_o66 cmd set-link _o19"
    invoked from within
"catch "$self cmd $args" ret"
    (procedure "_o66" line 2)
    (SplitObject unknown line 2)
    invoked from within
"$queue_ set-link [$self set toNode_]"
    (procedure "_o67" line 19)
    (SimpleLink init line 19)
    invoked from within
"_o67 init _o10 _o19 10Mb 5ms _o66"
    (Class create line 1)
    invoked from within
"SimpleLink create _o67 _o10 _o19 10Mb 5ms _o66"
    invoked from within
"catch "$className create $o $args" msg"
    (procedure "new" line 3)
    invoked from within
"new SimpleLink     $n1 $n2 $bw $delay $q"
    ("default" arm line 2)
    invoked from within
"switch -exact $qtypeOrig {
RTM {
set c [lindex $args 1]
set link_($sid:$did) [new CBQLink        $n1 $n2 $bw $delay $q $c]
}
CBQ -
CBQ/WRR {
if {[llen..."
    (procedure "_o3" line 33)
    (Simulator simplex-link line 33)
    invoked from within
"_o3 simplex-link _o10 _o19 10Mb 5ms RED"
    ("eval" body line 1)
    invoked from within
"eval $self simplex-link $n1 $n2 $bw $delay $type $args"
    (procedure "_o3" line 8)
    (Simulator duplex-link line 8)
    invoked from within
"$ns duplex-link $s0 $LSR3 10Mb 5ms RED"
    (file "example.tcl" line 50)


why? please help me!
thanks in advance
 








--http://www.eyou.com
--Îȶ¨¿É¿¿µÄÃâ·Ñµç×ÓÐÅÏä  ÓïÒôÓʼþ  Òƶ¯ÊéÇ©  ÈÕÀú·þÎñ  ÍøÂç´æ´¢...ÒÚÓÊδ¾¡



From koyabe@erg.abdn.ac.uk  Wed Mar 13 05:18:05 2002
From: koyabe@erg.abdn.ac.uk (Martin Koyabe)
Date: Wed Mar 13 05:18:05 2002
Subject: [ns] Research Contribution
Message-ID: 

Hi,

Could you kindly include this research work under the Satellite Area or
Multicast in the Research Using NS section:

M.W. Koyabe, G. Fairhurst, Performance of Reliable Multicast Protocols via
Satellite at EHF with Persistent Fades, presented at the 7th Ka-Band
Utilization Conference, 26-28th September 2001, Santa Margherita Ligure,
Genoa, Italy.


Thanks,

-- Martin


From jelger@clarinet.u-strasbg.fr  Wed Mar 13 05:45:14 2002
From: jelger@clarinet.u-strasbg.fr (Christophe Jelger)
Date: Wed Mar 13 05:45:14 2002
Subject: [ns] mrtObject : getWellKnownGroup procedure
Message-ID: <3C8F55CB.2070401@clarinet.u-strasbg.fr>

Hello,

I'm trying to use the procedure getWellKnownGroup (from 
tcl/mcast/McastProto.tcl) defined for an object of type "mrtObject".

 From a node I can access the mrtObject (with something like : set 
dummy  [$node getArbiter]), but when I do : $dummy getWellKnownGroup 
ALL_PIM_ROUTERS (for example) I get the error message :

_o16: unable to dispatch method getWellKnownGroup
    while executing
"$dummy getWellKnownGroup ALL_PIM_ROUTERS"
    invoked from within
"set pim [$dummy getWellKnownGroup ALL_PIM_ROUTERS]  "
    (file "ssm_trace.tcl" line 165)

_o16 is the handler ($dummy) I get from the [$node getArbiter] call ...

What is very strange is that I can call other procedures related to the 
mrtObject (like start, stop, ...) but with a few functions it always 
return the same error message (with getWellKnownGroup, expandaddr etc ...).

Does anybody know what's wrong ?

Thanks

Christophe



From hvardhan@ee.duke.edu  Wed Mar 13 06:45:03 2002
From: hvardhan@ee.duke.edu (harshavardhan)
Date: Wed Mar 13 06:45:03 2002
Subject: [ns] error when calling OldSim: ex1.tcl
Message-ID: <001b01c1cab6$db157640$ddc40398@em2>

Hi all,

I am getting this error when I tried to run ns on windows

error when calling OldSim: ex1.tcl

when I browsed through the archives I found this posted twice but couldn't
find the fix.

anybody has any idea?

thanks
harsha



From hsh125@psu.edu  Wed Mar 13 06:50:01 2002
From: hsh125@psu.edu (Harshal Haridas)
Date: Wed Mar 13 06:50:01 2002
Subject: [ns] Documentation for Wireless Protocol Coding
Message-ID: <000001c1cab7$08208760$8ea7cb82@mumbai>

Hi,

I wanted to know if there is any Readme for the wireless protocols
implemented in ad hoc environment for DSDV, TORA, DSR or AODV. I found
some information in the tex files. However, Is there any design document
available for DSR specifically

-Harshal

Harshal S. Haridas
Graduate Student,
Computer Science and Engineering Department,
Penn State University, University Park




From rajac6@hotmail.com  Wed Mar 13 06:50:15 2002
From: rajac6@hotmail.com (Raja Chatterjee)
Date: Wed Mar 13 06:50:15 2002
Subject: [ns] ns2 SS7 simulation
Message-ID: 

Hi,

Has anybody tried to simulate SS7 protocol in ns2. Has the effort to
simulate VoIP in ns2 limited to generating SS7 traffic in the simulation run 
and assuming minimal impact to overall QoS from the ckt based PSTN network.
Thanks much,

Raja

P.S - As an aside, has anyone tried MPLS-diffserv integration in the 2.1b8 
ns release( I think the above integration was done by SriHari using MNS and 
Sean Murphy's diffserv patches but that(Sean Murphy's patch) seem to be lost 
and possibly incompatible with the current release that has the Nortel 
Networks implementation)


_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com


From mportnoi@ieee.org  Wed Mar 13 07:25:01 2002
From: mportnoi@ieee.org (Marcos "Locksmith" Portnoi)
Date: Wed Mar 13 07:25:01 2002
Subject: [ns] How to install ns-2.1b8 on redhat7.2?
References: 
Message-ID: <008d01c1caa3$0f1ad020$83f2dfc8@nebula>

This is a multi-part message in MIME format.

------=_NextPart_000_0070_01C1CA89.C714BD40
Content-Type: text/plain;
	charset="gb2312"
Content-Transfer-Encoding: quoted-printable

What I did was:

- extracted ns-allinone-2.1b8a to /usr/local using KDE's Archiver
- cd'ed to /usr/local/ns-allinone-2.1b8a and ran ./install
- after installation, modified PATH, LD_LIBRARY_PATH and TCL_LIBRARY in =
.bash_profile for all users that would use ns (as instructed in the =
message after installation):
    included the lines in .bash_profile, after the last PATH definition
    =
PATH=3D$PATH:/usr/local/ns-allinone-2.1b8a/bin:/usr/local/ns-allinone-2.1=
b8a/tcl8.3.2/unix:/usr/local/ns-allinone-2.1b8a/tk8.3.2/unix
    =
LD_LIBRARY_PATH=3D/usr/local/ns-allinone-2.1b8a/otcl-1.0a7:/usr/local/ns-=
allinone-2.1b8a/lib
    TCL_LIBRARY=3D/usr/local/ns-allinone-2.1b8a/tcl8.3.2/library

- and, at the last line of .bash_profile, included LD_LIBRARY_PATH and =
TCL_LIBRARY to resemble something like this:
    export USERNAME BASH_ENV PATH LD_LIBRARY_PATH TCL_LIBRARY

- cd'ed to /usr/local/ns-allinone-2.1b8a/ns-2.1b8a  and ran ./validate

- done!

M. Portnoi


   ----- Original Message -----=20
  From: Guanghai Cai=20
  To: ns-users@ISI.EDU=20
  Sent: Tuesday, March 12, 2002 8:05 PM
  Subject: [ns] How to install ns-2.1b8 on redhat7.2?


  Hi,
      I am a newer of ns. Could you please tell me the detail step by =
step procedure to install ns-2.1b8 on redhat7.2 platform.
      Kindly give me the step by step procedure after downloading the =
installable files...
      Thank you very much.

  Guanghai cai
  20020312

------=_NextPart_000_0070_01C1CA89.C714BD40
Content-Type: text/html;
	charset="gb2312"
Content-Transfer-Encoding: quoted-printable









What I did was:


 


- extracted ns-allinone-2.1b8a to =
/usr/local using=20
KDE's Archiver


- cd'ed to =
/usr/local/ns-allinone-2.1b8a and ran=20
./install


- after installation, modified PATH,=20
LD_LIBRARY_PATH and TCL_LIBRARY in .bash_profile for all users that =
would use ns=20
(as instructed in the message after installation):


    included the lines =
in=20
.bash_profile, after the last PATH definition


   =20
PATH=3D$PATH:/usr/local/ns-allinone-2.1b8a/bin:/usr/local/ns-allinone-2.1=
b8a/tcl8.3.2/unix:/usr/local/ns-allinone-2.1b8a/tk8.3.2/unix
=


   =20
LD_LIBRARY_PATH=3D/usr/local/ns-allinone-2.1b8a/otcl-1.0a7:/usr/local/ns-=
allinone-2.1b8a/lib
   =20
TCL_LIBRARY=3D/usr/local/ns-allinone-2.1b8a/tcl8.3.2/library
=


 


- and, at the last line of =
.bash_profile, included=20
LD_LIBRARY_PATH and TCL_LIBRARY to resemble something like =
this:


    export USERNAME =
BASH_ENV PATH=20
LD_LIBRARY_PATH TCL_LIBRARY


 


- cd'ed=20
to /usr/local/ns-allinone-2.1b8a/ns-2.1b8a  and ran=20
./validate


 


- done!


 


M. Portnoi






  
 ----- Original Message ----- =


  From:=20
  Guanghai Cai=20
  

  
  
Sent: Tuesday, March 12, 2002 =
8:05=20
  PM

  
Subject: [ns] How to install =
ns-2.1b8 on=20
  redhat7.2?

  


  
Hi,
    I am a newer of ns. Could you =
please=20
  tell me the detail step by step procedure to install ns-2.1b8=20
  on redhat7.2 platform.
    Kindly give me the =
step by=20
  step procedure after downloading the installable=20
  files...
    Thank you very =
much.

Guanghai=20
  cai

  
20020312


------=_NextPart_000_0070_01C1CA89.C714BD40--



From jelger@clarinet.u-strasbg.fr  Wed Mar 13 07:40:02 2002
From: jelger@clarinet.u-strasbg.fr (Christophe Jelger)
Date: Wed Mar 13 07:40:02 2002
Subject: [ns] What is the difference between target_->recv(p) & send(p,0)?
References: 
Message-ID: <3C8F71D1.8090609@clarinet.u-strasbg.fr>

Hello,

Well I finally managed to get it to work !!! I have two agents that are 
NOT connected, and when putting correct addresses and ports, packets are 
sent correctly from one agent to the other. For interested readers, I 
want to stress the fact that the port number must also be set and (in my 
specific case) agents are attached to node with : $node attach $agent 
$port (in order for the agent to be correctly bound to the appropriate 
port number).

Christophe

Padmaparna Haldar wrote:

>when an agent gets attached to a node, the target_ points to the node 
>entry_ which in default condition points to the the classifier.
>when you connect two agents you are basically setting up the destination 
>to which the agent will send pkts to. 
>so assumimg yr agent was connected to a node, it can send 
>pkts to a given dest address, provided there are corr. routing entries for 
>that address in the classifier.
>--Padma
>
>On Mon, 11 Mar 2002, Christophe Jelger wrote:
>
>
>>Hi,
>>
>>And what happens when objects are not connectors ? For example if an 
>>agent simply wants to send packets to a given dest address ? (assuming 
>>that there is a routing entry in the classifier) ? For example, I want 
>>my agent to send a packet to the well known ALL-PIM-ROUTERS address ... 
>>how shall I proceed ? (the agent is not connected so there is no target_)
>>
>>Thanks
>>
>>Christophe
>>
>>
>>Padmaparna Haldar wrote:
>>
>>>
>>>Most objects in ns are Connector objects having a target_ to SEND pkts to. 
>>>It can also RECV pkts. recv and send functions in the Connector class are 
>>>defined as follows:
>>>
>>>void recv(Packet*, Handler* callback = 0);
>>>inline void send(Packet* p, Handler* h) { target_->recv(p, h); }
>>>
>>>so as you can see, send(p,h) and target_->recv(p) are basically doing the 
>>>same thing.
>>>
>>>--Padma
>>>
>>>On Wed, 6 Mar 2002, Mustafa Tasdemir wrote:
>>>
>>>
>>>>hello,
>>>>
>>>>can someone tell what is the difference between these commands when used in agent:
>>>>
>>>>Application---->UDP------->Node
>>>>                          |
>>>>                          |
>>>>                       commands are used in UDP agents
>>>> 1.. target_->recv(p);    // target defined as Process *target_; 
>>>> 2.. send(p,0);
>>>>I think, the first one is a direct call and the second one is an indirect call. Can we also call the send() method as a SAP (Service Access Point)?
>>>>
>>>>Mustafa TASDEMIR
>>>>
>>>>
>>>>
>>>>
>>
>



From jain@cs.umn.edu  Wed Mar 13 08:05:05 2002
From: jain@cs.umn.edu (Priyanka Jain)
Date: Wed Mar 13 08:05:05 2002
Subject: [ns] linking problem on extending ns-2.1b5
Message-ID: 

Hi All,
I have been struggling with this for some time now, and I do not
understand this. I have some .cc, .h files which I am trying to compile
with ns2.1b5.
Here is my make.local
**********************************************

 ################################################################
#                        Local Changes
################################################################

# change this to your ns-2.1b5/directory
NSb5_DIR = /project/condor06/usr/jain/ns-allinone-2.1b5/ns-2.1b5

# change this to the respective directories where tcl.h and otcl.h are
# and possibly more if compiler doesn't file the include files
MORE_INCLUDES = -I/project/condor06/usr/jain/ns-allinone-2.1b5/tcl8.0$
                -I/project/condor06/usr/jain/ns-allinone-2.1b5/tclcl-$
                -I/project/condor06/usr/jain/ns-allinone-2.1b5/tclbox$
                -I/project/condor06/usr/jain/ns-allinone-2.1b5/tkbox/$

MORE_LIBS =  -lstdc++
CPP = g++
## No need to change anything below

NS_DIR = $(NSb5_DIR)
.cc.o:
        @rm -f $@
        $(CPP) -O3 -c -DNSANTS -I. -I$(NS_DIR) $(MORE_INCLUDES) $(INC$
                -o $@ $<
LOCAL_OBJ_CC = packet.o trace.o rtProtoAntNet.o ants.o
#LOCAL_OBJ_CC = packet.o trace.o rtProtoAntNet.o ants.o gen/ns_tcl.o
packet.o trace.o rtProtoAntNet.o : packet.h
rtProtoAntNet.o ants.o: ants.h

LOCAL_OBJ = $(LOCAL_OBJ_CC)
ORIGINAL_OBJ = $(filter-out $(LOCAL_OBJ), $(OBJ) ) tclAppInit.o
NEW_OBJ = $(addprefix $(NS_DIR)/,$(ORIGINAL_OBJ) ) $(LOCAL_OBJ)

nsants : $(NEW_OBJ)
        $(LINK) $(LDFLAGS) $(LDOUT)$@ \
                $(NEW_OBJ) $(LIB)

TEST_LOCAL_OBJ = ants.o rtProtoAntNet.o test_ants.o
test_ants.o : rtProtoAntNet.h ants.h

TEST_ORIGINAL_OBJ = $(filter-out $(TEST_LOCAL_OBJ), $(OBJ) )
TEST_NEW_OBJ = $(addprefix $(NS_DIR)/,$(TEST_ORIGINAL_OBJ) ) $(TEST_L$

test_ants : $(TEST_LOCAL_OBJ)
        $(LINK) $(LDFLAGS) $(LDOUT)$@ \
                 $(TEST_LOCAL_OBJ) $(LIB) $(MORE_LIBS)
*******************************************
Now, I get errors like this
******************************************

..........
r/jain/ns-allinone-2.1b5/otcl-1.0a4
-L/project/condor06/usr/jain/ns-allinone-2.1b5/otcl-1.0a4 -lotcl
-R/project/condor06/usr/jain/ns-allinone-2.1b5/tkbox/lib
-L/project/condor06/usr/jain/ns-allinone-2.1b5/tkbox/lib -ltk8.0
-R/project/condor06/usr/jain/ns-allinone-2.1b5/tclbox/lib
-L/project/condor06/usr/jain/ns-allinone-2.1b5/tclbox/lib -ltcl8.0 -lXext
-lX11 -lsocket -lnsl -lintl -ldl -ldl -lm -ldl
packet.o: In function
`PacketHeaderManagerClass::~PacketHeaderManagerClass(void)':
/project/condor06/usr/jain/rtProtoAnts_b2/packet.cc(.gnu.linkonce.t.__tf17PacketHeaderClass+0x18):
undefined reference to `TclClass type_info function'
/project/condor06/usr/jain/rtProtoAnts_b2/packet.cc(.gnu.linkonce.t.__tf17PacketHeaderClass+0x24):
undefined reference to `TclClass type_info node'
.......
In packet.h, I have included the necessary files for TclClass, but still
it crashes on linking...Does anybody, what I am doing wrong..pls help..
thanks
priyanka


From chakra10@msu.edu  Wed Mar 13 10:00:08 2002
From: chakra10@msu.edu (Gautam - Chakrabarti)
Date: Wed Mar 13 10:00:08 2002
Subject: [ns] Documentation for Wireless Protocol Coding
References: <000001c1cab7$08208760$8ea7cb82@mumbai>
Message-ID: <3C8F926A.19CFD1C2@msu.edu>

I assume you have the relevant papers, the original papers for each
protocol, and the CMU Mobicom ('99 if I remember right) paper giving
quite a bit of details about the implementation.

Gautam

Harshal Haridas wrote:
> 
> Hi,
> 
> I wanted to know if there is any Readme for the wireless protocols
> implemented in ad hoc environment for DSDV, TORA, DSR or AODV. I found
> some information in the tex files. However, Is there any design document
> available for DSR specifically
> 
> -Harshal
> 
> Harshal S. Haridas
> Graduate Student,
> Computer Science and Engineering Department,
> Penn State University, University Park

-- 
***************************************************************
Gautam Chakrabarti
MS Student
Computer Sc. and Engg Department
Michigan State University

Email : chakra10@msu.edu
Home Page : http://www.cse.msu.edu/~chakra10

Why is there so much month left at the end of the money?
		--- John Barrymore
***************************************************************


From xuanc@ISI.EDU  Wed Mar 13 10:10:02 2002
From: xuanc@ISI.EDU (Xuan Chen)
Date: Wed Mar 13 10:10:02 2002
Subject: [ns] link with two queues
In-Reply-To: 
Message-ID: 

You may want to look into the diffserv module and refer to their 
implementation on physical queues.

-chen

On Wed, 13 Mar 2002 Sergio.Herreria@det.uvigo.es wrote:

> 
> How could I create a link with two sequenced queues?
> 
> head_
>  ----> enqT_ ----> queue1_ ----> queue2_ ----> deqT_ ----> link_ ....
> 			|             |
> 			------------------>drophead_ ----> drpT_
> Thank you very much
> Sergio Herreria
> 

-- 
Xuan Chen
USC/ISI


From cheprosh@yahoo.com  Wed Mar 13 11:15:02 2002
From: cheprosh@yahoo.com (Prashant)
Date: Wed Mar 13 11:15:02 2002
Subject: [ns] need help compiling tclcl-1.0b11 - weird tcl2c++ problem
Message-ID: <3C8F5E4D.5000803@yahoo.com>

Dear experts,

I compiled the ns2 package (ver 2.1b8) on my Mandrake 8.1 Linux box at 
home (all-in-one) with no problem. So I am not so incompetent ;)

I then attempted to install on an UltraSPARC 80 at school the 
all-in-one, but failed, and then the packages seperately. Both have 
failed no matter what I tried, including the suggestions I found on the 
help page and in the mail archives. I shall show below how I tried to 
compile the packages separately one at a time.

I am compiling in my own account only, not system-wide. So I modified 
the configure flags to reflect this, ie --prefix=$HOME/ns2_pieces.

Finally, it seems to hang while compiling tclcl-1.0b11.

Any help is appreciated, thanks in advance!

-Prashant

===============================================================================
This is how I compiled each package:

% cd ~/ns2_pieces/tcl8.4a3/unix
% ./configure --prefix=$HOME/ns2_pieces --enable-gcc --disable-shared

% cd ~/ns2_pieces/tk8.4a3/unix
% ./configure --prefix=$HOME/ns2_pieces --enable-gcc --disable-shared

% cd ~/ns2_pieces/otcl-1.0a7
% ./configure --prefix=$HOME/ns2_pieces --enable-gcc --with-tcl-ver=8.4 
--with-tk-ver=8.4
% make
% make install [had to force install of libotcl.so into
                 $HOME/ns2_pieces/lib]

% ~/ns2_pieces/tclcl-1.0b11
% ./configure --prefix=$HOME/ns2_pieces --enable-gcc --with-tcl-ver=8.4 
--with-tk-ver=8.4
% cd ~/ns2_pieces/tclcl-1.0b11
% vi Makefile   [change CPP=c++ to CPP=g++]
% make
gcc -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__ 
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H 
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4 
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1 
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1 
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o tcl2c++.o 
tcl2c++.c
rm -f tcl2c++
g++   -o tcl2c++ tcl2c++.o
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__ 
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H 
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4 
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1 
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1 
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o Tcl.o Tcl.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__ 
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H 
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4 
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1 
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1 
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o Tcl2.o Tcl2.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__ 
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H 
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4 
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1 
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1 
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o 
idlecallback.o idlecallback.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__ 
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H 
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4 
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1 
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1 
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o iohandler.o 
iohandler.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__ 
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H 
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4 
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1 
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1 
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o timer.o timer.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__ 
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H 
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4 
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1 
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1 
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o 
rate-variable.o rate-variable.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__ 
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H 
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1 
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4 
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1 
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1 
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include 
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o tracedvar.o 
tracedvar.cc
rm -f embedded-tcl.cc
./tcl2c++ et_tcl  > embedded-tcl.cc

# here it hangs forever
# if I look at embedded-tcl.cc it is empty

# running ./tcl2c++ et_tcl gives

% ./tcl2c++ et_tcl
static const char code[] = {

# and it is waiting for input! hitting 1 enter
# and then 2 enter and then ctrl-D looks like this:

% ./tcl2c++ et_tcl
static const char code[] = {
1
49,10,2
50,10,0 };
#include "tclcl.h"
EmbeddedTcl et_tcl(code);
%



==========================================================================



[other misc. information about my system]

% uname -a
SunOS gazebo 5.8 Generic_108528-12 sun4u sparc SUNW,Ultra-80

% echo $LD_LIBRARY_PATH
/home/users01/prashant/ns2_pieces/tcl8.4/unix:/home/users01/prashant/ns2_pieces/tk8.4/unix:/usr/local/lib/

% whereis ld
ld: /usr/ccs/bin/ld /usr/lib/ld.so /usr/lib/ld /usr/ucb/ld

% whereis gcc
gcc: /usr/local/bin/gcc

% whereis g++
g++: /usr/local/bin/g++

% whereis as
as: /usr/ccs/bin/as

% ls ~/ns2_pieces  # show you versions I am using
nam-1.0a10/
ns-2.1b8/
olibsh/
otcl-1.0a7/
tcl8.4a3/
tclcl-1.0b11/
tk8.4a3/








_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


From nauzads@cs.tamu.edu  Wed Mar 13 11:25:02 2002
From: nauzads@cs.tamu.edu (Nauzad Erach Sadry)
Date: Wed Mar 13 11:25:02 2002
Subject: [ns] HELP !!! HOW TO SET BS & WIRELESS NODE BANDWIDTH
Message-ID: 

Hello

Can someone please advise me how set the bandwidth between the basestation
and a wireless node ...

Apparently the bandwidth variable in wireless.cc file is NOT binded in the
ns2.1b6 version ... 

So there is no sense setting Phy/Wireless set bandwidth_ 1e6

Can someone pleasehelp me out ... i m stuck in this since a long time

Nauzad E. Sadry

Res: 601, Cross Street, Apt# 59        Office: Graduate Teaching Assistant
     College Station, TX - 77840               Dept. of Computer Science
     USA                                       Texas A&M University
     Tel:(979)691-5762                         Off : (979)845-4094


From nauzads@cs.tamu.edu  Wed Mar 13 11:25:37 2002
From: nauzads@cs.tamu.edu (Nauzad Erach Sadry)
Date: Wed Mar 13 11:25:37 2002
Subject: [ns] HELP !!! HOW TO SET BS & WIRELESS NODE BANDWIDTH (fwd)
Message-ID: 

Hello

Can someone please advise me how set the bandwidth between the basestation
and a wireless node ...

Apparently the bandwidth variable in wireless.cc file is NOT binded in the
ns2.1b6 version ... 

So there is no sense setting Phy/Wireless set bandwidth_ 1e6

Can someone pleasehelp me out ... i m stuck in this since a long time

Nauzad E. Sadry

Res: 601, Cross Street, Apt# 59        Office: Graduate Teaching Assistant
     College Station, TX - 77840               Dept. of Computer Science
     USA                                       Texas A&M University
     Tel:(979)691-5762                         Off : (979)845-4094



From nauzads@cs.tamu.edu  Wed Mar 13 11:26:01 2002
From: nauzads@cs.tamu.edu (Nauzad Erach Sadry)
Date: Wed Mar 13 11:26:01 2002
Subject: [ns] HELP !!! HOW TO SET BS & WIRELESS NODE BANDWIDTH (fwd)
Message-ID: 

Hello

Can someone please advise me how set the bandwidth between the basestation
and a wireless node ...

Apparently the bandwidth variable in wireless.cc file is NOT binded in the
ns2.1b6 version ... 

So there is no sense setting Phy/Wireless set bandwidth_ 1e6

Can someone pleasehelp me out ... i m stuck in this since a long time

Nauzad E. Sadry

Res: 601, Cross Street, Apt# 59        Office: Graduate Teaching Assistant
     College Station, TX - 77840               Dept. of Computer Science
     USA                                       Texas A&M University
     Tel:(979)691-5762                         Off : (979)845-4094




From cheprosh@yahoo.com  Wed Mar 13 11:30:03 2002
From: cheprosh@yahoo.com (Prashant)
Date: Wed Mar 13 11:30:03 2002
Subject: [ns] test - ignore
Message-ID: <3C8F6200.2010904@yahoo.com>


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


From cheprosh@yahoo.com  Wed Mar 13 11:35:06 2002
From: cheprosh@yahoo.com (Prashant)
Date: Wed Mar 13 11:35:06 2002
Subject: [ns] need help compiling tclcl-1.0b11 - weird tcl2c++ problem
Message-ID: <3C8F6371.6050602@yahoo.com>

Dear experts,

I compiled the ns2 package (ver 2.1b8) on my Mandrake 8.1 Linux box at
home (all-in-one) with no problem. So I am not so incompetent ;)

I then attempted to install on an UltraSPARC 80 at school the
all-in-one, but failed, and then the packages seperately. Both have
failed no matter what I tried, including the suggestions I found on the
help page and in the mail archives. I shall show below how I tried to
compile the packages separately one at a time.

I am compiling in my own account only, not system-wide. So I modified
the configure flags to reflect this, ie --prefix=$HOME/ns2_pieces.

Finally, it seems to hang while compiling tclcl-1.0b11.

Any help is appreciated, thanks in advance!

-Prashant

===============================================================================
This is how I compiled each package:

% cd ~/ns2_pieces/tcl8.4a3/unix
% ./configure --prefix=$HOME/ns2_pieces --enable-gcc --disable-shared

% cd ~/ns2_pieces/tk8.4a3/unix
% ./configure --prefix=$HOME/ns2_pieces --enable-gcc --disable-shared

% cd ~/ns2_pieces/otcl-1.0a7
% ./configure --prefix=$HOME/ns2_pieces --enable-gcc --with-tcl-ver=8.4
--with-tk-ver=8.4
% make
% make install [had to force install of libotcl.so into
                  $HOME/ns2_pieces/lib]

% ~/ns2_pieces/tclcl-1.0b11
% ./configure --prefix=$HOME/ns2_pieces --enable-gcc --with-tcl-ver=8.4
--with-tk-ver=8.4
% cd ~/ns2_pieces/tclcl-1.0b11
% vi Makefile   [change CPP=c++ to CPP=g++]
% make
gcc -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o tcl2c++.o
tcl2c++.c
rm -f tcl2c++
g++   -o tcl2c++ tcl2c++.o
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o Tcl.o Tcl.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o Tcl2.o Tcl2.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o
idlecallback.o idlecallback.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o iohandler.o
iohandler.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o timer.o timer.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o
rate-variable.o rate-variable.cc
g++ -c -O2 -DNO_TK  -DNDEBUG -DHAVE_STL -mv8 -msupersparc -D__svr4__
-DUSE_SHM -DHAVE_LIBOTCL1_0A7 -DHAVE_OTCL_H -DHAVE_LIBTK8_4 -DHAVE_TK_H
-DHAVE_LIBTCL8_4 -DHAVE_TCL_H   -DSTDC_HEADERS=1 -DHAVE_STRING_H=1
-DHAVE_SNPRINTF=1 -DSTDC_HEADERS=1 -DHAVE_STRTOLL=1 -DSIZEOF_LONG=4
-DHAVE_INT64=1 -DHAVE_TCL_H=1 -DHAVE_LIBTCL8_4=1 -DHAVE_TK_H=1
-DHAVE_LIBTK8_4=1 -DHAVE_OTCL_H=1 -DHAVE_LIBOTCL1_0A7=1 -DHAVE_LIBDL=1
-I. -I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/otcl-1.0a7
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include
-I/tmp_mnt/vol/vol0/users01/prashant/ns2_pieces/include -o tracedvar.o
tracedvar.cc
rm -f embedded-tcl.cc
./tcl2c++ et_tcl  > embedded-tcl.cc

# here it hangs forever
# if I look at embedded-tcl.cc it is empty

# running ./tcl2c++ et_tcl gives

% ./tcl2c++ et_tcl
static const char code[] = {

# and it is waiting for input! hitting 1 enter
# and then 2 enter and then ctrl-D looks like this:

% ./tcl2c++ et_tcl
static const char code[] = {
1
49,10,2
50,10,0 };
#include "tclcl.h"
EmbeddedTcl et_tcl(code);
%



==========================================================================



[other misc. information about my system]

% uname -a
SunOS gazebo 5.8 Generic_108528-12 sun4u sparc SUNW,Ultra-80

% echo $LD_LIBRARY_PATH
/home/users01/prashant/ns2_pieces/tcl8.4/unix:/home/users01/prashant/ns2_pieces/tk8.4/unix:/usr/local/lib/

% whereis ld
ld: /usr/ccs/bin/ld /usr/lib/ld.so /usr/lib/ld /usr/ucb/ld

% whereis gcc
gcc: /usr/local/bin/gcc

% whereis g++
g++: /usr/local/bin/g++

% whereis as
as: /usr/ccs/bin/as

% ls ~/ns2_pieces  # show you versions I am using
nam-1.0a10/
ns-2.1b8/
olibsh/
otcl-1.0a7/
tcl8.4a3/
tclcl-1.0b11/
tk8.4a3/









_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


From leob@dcc.ufmg.br  Wed Mar 13 11:50:05 2002
From: leob@dcc.ufmg.br (Leonardo Barbosa e Oliveira)
Date: Wed Mar 13 11:50:05 2002
Subject: [ns] P2P application
Message-ID: 


Hello, everybody!

    I gonna implement a p2p application in ns and I would like
to know if someone already tried sth like that.

Thanks! 


[]'s!
     
     Leonardo

      "You are what you eat"
                 Garfield 
         
-------------------- 		 
Leonardo Barbosa e Oliveira	
Lab. ATM/DCC/UFMG/Brasil - leob@dcc.ufmg.br
Projeto ReMAV BH2 - http://www.dcc.ufmg.br/~admbh2
Grupo de Gerencia, Operacao e Manutencao da Rede


From lu_bisa@yahoo.com.br  Wed Mar 13 12:00:04 2002
From: lu_bisa@yahoo.com.br (=?iso-8859-1?q?Luise=20Fabrisia?=)
Date: Wed Mar 13 12:00:04 2002
Subject: [ns] running tests
In-Reply-To: <000001c1cab7$08208760$8ea7cb82@mumbai>
Message-ID: <20020313195630.63463.qmail@web21310.mail.yahoo.com>

 Hello
How can I use the scripts using the command 'ns