Your e-mail messages will be slightly edited before being posted here :-) Ljiljana. ******************************************************************************* Date: Thu, 20 Jan 2000 14:21:07 -0800 (PST) From: Nazy Alborz Subject: ATM switch model in Opnet To: ensc894-g1@sfu.ca Hi, For those of you who are intrested in modeling an ATM switch, I just checked Opnet library. Apparantly it is possible to model such a switch in Opnet. Regards, Nazy Alborz ******************************************************************************* From crosbie@rm-rstar.sfu.ca Thu Jan 20 08:17:31 2000 To: ljilja@cs.sfu.ca From: Catherine Crosbie Subject: Re: Getting logins for my graduate class(Gripe#8943)ENSC894/CMPT885 Cc: ensc-help@sfu.ca, chan@cs.sfu.ca Dear Ljiljana, Students must complete an appropriate application form for any computer account. Marilyn in the ENSC General Office provides all forms for ENSC students. The computers in ASB9884 are connected to the CSSNet so to acquire an account students must fill out a CSSNet Account Application form. Most ENSC grad students already have CSSNet accounts. I don't know the status CMPT grads. There is a place on the application form to state the purpose of the account. The applications could thus state "ENSC894/CMPT885-using ASB9884 lab". You will have to sign the application as the advisor/sponsor so you could check that. The procedure is to drop the completed forms off at ASB10856. CSSNet staff create accounts on the CSSNet. The issue of whether CMPT students may access ENSC labs is not a matter that computer support staff can answer. In the case of ASB9884 it is rather a question of whether the department will grant security access cards to students. I don't know the answer to that; but if true, this would imply that the students have departmental authority to use ENSC resources. My advisor, Shahram, suggests you apply for that. Perhaps you might address the question to John Jones who is currently the head of the Lab Committee. Catherine. ******************************************************************************* From bwei@home.com Sat Jan 22 13:27:16 2000 From: Bernard Wei Reply-To: bwei@home.com To: Ljiljana Trajkovic Subject: About Telecommunication Acronyms Hi Dr. Trajkovic, I found this site on Communication Acronyms: http://www.csrstds.com/acro-a-d.html P/S: Is it possible to add "bwei@home.com" to the class ENSC-894 mailing list. I'm having trouble accessing gwei@sfu.ca from home. -- Bernard Wei -o) bwei@home.com o) o) /\\ ______________________________/\__/\_____>V/__ ******************************************************************************* From bwei@home.com Mon Jan 24 21:38:09 2000 From: Bernard Wei To: Ljiljana Trajkovic Subject: BT Cellnet - new & promotions Hi Dr. Trajkovic, I came across this link that talk about BT's GPRS network. The article doesn't give a lot of details, but if you have insight to this type of network, hope you can touch base with us in class on how they function. Thanks. http://www.cellnet.co.uk/ins_BT/pre_rel/2000/jan/210100_gpr.html -- Bernard Wei -o) bwei@home.com o) o) /\\ ______________________________/\__/\_____>V/__ ******************************************************************************* From fwlau@cs.sfu.ca Wed Jan 26 12:29:38 2000 From: Felix Lau Subject: Info about Voice over ATM To: ljilja@cs.sfu.ca (Ljiljana Trajkovic) Hi Ljiljana, I did a web search for Voice over ATM, I found some literature at Ohio State. You mentioned some work being done at Columbia? Would you have the link to this info? Thanks, Felix ******************************************************************************* From ljilja@cs.sfu.ca Thu Jan 27 11:36:03 2000 From: Ljiljana Trajkovic To: fwlau@cs.sfu.ca Subject: Re: Info about Voice over ATM Felix, The site is: http://www.cs.columbia.edu/~hgs Regards, Ljiljana ******************************************************************************* From ljilja@cs.sfu.ca Thu Jan 27 11:32:09 2000 From: Ljiljana Trajkovic To: armand@juxta.com Subject: Re: Assignment for ENSC 894 Thank you Danesh. The INDEX WWW site is: http://www.INDEX.Berkeley.EDU/public/index.phtml Regards, Ljiljana ******************************************************************************* From ljilja@cs.sfu.ca Thu Jan 27 11:56:06 2000 From: Ljiljana Trajkovic To: mlian@sfu.ca Subject: Re:project topic Thank you for the proposal. I suggest you choose one topic: security or pricing. To work on both is a bit too much for the project. For pricing, check WWW sites: Prof. Jeffrey MacKie-Mason site linked to: http://china.si.umich.edu/telecom/telecom-info.html and http://www.INDEX.Berkeley.EDU/public/index.phtml Thanks, Ljiljana ******************************************************************************* From ljilja@cs.sfu.ca Thu Jan 27 22:09:26 2000 From: Ljiljana Trajkovic To: William.Ressl@VLSI.com Subject: RE: ENSC 894 Project Topic Cc: ljilja@cs.sfu.ca Bill, Here is the description of the traces: You can see it on the bottom of my home page: http://www.ensc.sfu.ca/people/faculty/ljilja/ ******************************************************************************* The file name is starwars.frames It contains the frame sizes of the Star Wars movie (in bits). The size of the file is around 1MB. The inter-frame time of the StarWars trace is 1/24 sec. That means that the inter-arrival time for the arriving bits is 1/24 seconds. Hence, the time series is: Time (sec) Packets (bits) 0 9537 1/24 9558 2/24 9573 3/24 9576 4/24 10592 5/24 11963 . . . ******************************************************************************* The files named: Size (in bytes): news.B 109461 news.I 16350 news.IPB 173064 news.P 47253 The file is encoded MPEG-1 sequence of German TV news. The main details for all the traces is: GOP pattern: IBBPBBPBBPBB (12 frames) Encoder input: 384x288 pels with 12 bit color information, Number of frames per sequence: 40000 frames (around half an hour with 1/24 sec inter-frame time), All of the above traces are around 150K (in tar and zipped format) and contain the frame sizes (in bits) for the whole trace (I,P and B frames together), for only I, only P and only B frame sizes. You may start with the smallest trace, if you cannot handle the larger ones. Best wishes, Ljiljana ******************************************************************************* From ljilja@cs.sfu.ca Fri Jan 28 18:29:53 2000 From: Ljiljana Trajkovic To: cklui@sfu.ca Subject: Re: ENSC894 Project from CKLui Thank you for the proposal CK. If I have not already suggested, please check sites for IP Telephony where you may find some related work: http://www.cs.columbia.edu/~hgs Best regards, Ljiljana ******************************************************************************* From scong@menkar.cs.sfu.ca Wed Feb 2 11:22:34 2000 From: Shi Cong To: ljilja@cs.sfu.ca Subject: Re: Something about your book (fwd) Dear Ljiljana, Maybe we have one mistake in the "High Performance Communication Network" 2000. I have written to the write and here is the answer. Hope this will be helpful. Best Regards, Shi Cong Ph.D Computing Science Department ---------- Forwarded message ---------- Date: Wed, 02 Feb 2000 11:10:51 -0800 From: Pravin Varaiya To: Shi Cong Cc: wlr@eecs.berkeley.edu Subject: Re: Something about your book You are quite right. Thanks! We'll correct it the next printing. Shi Cong wrote: > > Dear Jean and Pravin, > > I am reading your book "High-performance communication network" 2000. At > page 29 line 11: the advantage of ATM over IP. I wonder if it should be IP > over ATM. If I am wrong, sorry for your time. Have a nice day! > > Best Regards, > > Shi Cong > Ph.D student > Computing Science Department > Simon Fraser University ******************************************************************************* From scong@scutum.cs.sfu.ca Wed Feb 9 15:46:39 2000 Date: Wed, 9 Feb 2000 15:46:36 -0800 (PST) From: Shi Cong To: Ljiljana Trajkovic Subject: Good RFC sites Dear Ljiljana, Since a lot of students will use the Internet RFC, so I think the following website may be useful for us. The website give a very good summary and index on all of the RFCs. http://www.cis.ohio-state.edu/hypertext/information/rfc.html Have a nice day! Best Regards, Shi Cong ******************************************************************************* From ktpang@rm-rstar.sfu.ca Wed Feb 9 21:04:52 2000 To: nalborz@cs.sfu.ca From: Thomas Pang Cc: ljilja@cs.sfu.ca, ktpang@cs.sfu.ca Nazy, Here is the link to SS7 tutorial prepared by International Engineering Consortium (IEC). http://www.webproforum.com/ss7/index.html Please note that SS7 protocol stack consists only four levels and does not perfectly align with the OSI model. This is due in part to the fact that SS7 was developed before the OSI model. I cc Ljiljana so that she can include the link into our webpage. Hope this helps. - Thomas Pang ******************************************************************************* From zguoa@cs.sfu.ca Mon Feb 14 01:10:50 2000 From: Zhifeng Guo Subject: Project plan To: ljilja@cs.sfu.ca Dr. Trajkovic, As I mentioned last time, I will be unable to attend next lecture for ENSC 894 due to time conflict. Sorry about that. There're two web sites that I think probably be of someone's interest: http://www.psc.edu/networking/research.html http://www.monarch.cs.cmu.edu/ The former contains intensive network tools, software and information about TCP research. And the other is famous on mobile IP research and wireless extension for NS. I managed to install the new NS version (2.1b6) on RedHat Linux 6.1 in my home PC (I encounted some difficulties during the procedure). If anyone wants to run NS on home Linux box and has problems on installation, I'll be pleased to help. I will use NS to do my final project: Simulation of versions of TCP over Lossy links (Wireless Links) All versions of TCP apply a common window adaptation procedure for congest control. If a packet is lost, however, in TCP-OldTahoe, retransmission will start only upon timeout, and newer versions of TCP - Tahoe, Reno, NewReno, and Vegas, implement different fast-retransmit and fast-recover schemes. I would like to use NS to simulate the performance impact of such schemes over mobile radio link. Also I am interested in the effect of varying wireless link conditions. I noted that the new NS 2.1b6 version had included CMU wireless extension. It will be of great help to do the simulations. With best regards, Jeff Zhifeng Guo ******************************************************************************* From owner-ensc894-g1@rm-rstar.sfu.ca Mon Feb 14 22:42:10 2000 Sender: bwei@home.com From: Bernard Wei To: ensc894-g1@sfu.ca Subject: NS 2.1b5 I am successful in building NS 2.1b5 with gcc 2.95.2 and glibc2.1 from Mandrake 7.0 distribution on my home machine. I'm running custom kernel 2.2.15pre5, but I don't believe the kernel make any difference. There aren't any special thing you need to do, here's some simple instructions: Unpack the ns-allinone-2.1b5 tar ball, cd to ns-2.1b5 and run "configure --prefix=/usr/local", do a "make" and then do a "make install" to place the ns executable in /usr/local/bin directory. Let me know if you have problem building NS on similar system, I'll be happy to help. Some possible problem may be that you are missing the following package which are all available in Mandrake 7.0 distribution: tcl 8.0.5 tclx 8.0.5 tk 8.0.5 perl 5.00503 Cheers! Installing is only half the fun, using it is the other half :P -- Bernard Wei -o) bwei@home.com o) o) /\\ ______________________________/\__/\_____>V/__ ******************************************************************************* From jliud@sfu.ca Wed Feb 16 15:18:00 2000 Date: Wed, 16 Feb 2000 15:17:59 -0800 (PST) From: Jingyu Liu To: Ljiljana Trajkovic Hi Ljiljana, Could you add IETF's website to the course web page? I think it would be helpful if someone wants to lookup the latest RFCs. I am using it to check IP over ATM related RFCs(RFC1755/1754/1626/1577/1483). Here is the http address: http://www.ietf.cnri.reston.va.us/home.html Thanks. -- Jingyu Liu ******************************************************************************* From milan@cs.sfu.ca Thu Feb 17 00:16:52 2000 Date: Thu, 17 Feb 2000 00:16:52 -0800 (PST) From: Milan Nikolic Subject: Telecommunications Institutions and Forums To: ljilja@cs.sfu.ca Dear Prof. Ljiljana, here is a list of various Telecommunications Institutions and Forums that may be of interest to the class. (I omitted those important ones that are already listed on our course web page). All URLs are correct and alive. American National Standards Institute (ANSI) http://www.ansi.org/ European Telecommunications Standards Institute (ETSI) http://www.etsi.org/ ADSL Forum http://www.adsl.com/ Frame Relay Forum (FRF) http://www.frforum.com/ Intelligent Network Forum (IN Forum) http://www.inf.org/ Internet Telephony Consortium (ITC) http://itel.mit.edu/ Multiservice Switching Forum (MSF) http://www.msforum.org/ Gigabit Ethernet Alliance http://www.gigabit-ethernet.org/ Cell Relay Retreat http://cell-relay.indiana.edu/ Digital Audio Visual Council (DAVIC) http://www.davic.org/ European ISDN User Forum http://www.eiuf.org/ North American ISDN Users Forum (NIUF) http://www.niuf.nist.gov/ Vendors ISDN Association (VIA) http://www.via-isdn.org/ CDMA Development Group http://www.cdg.org/ Wireless LAN Interoperability Forum (WLIF) http://www.wlif.com/ Best regards, Milan. ******************************************************************************* From scong@indus.cs.sfu.ca Thu Feb 17 14:27:19 2000 From: Shi Cong To: Ljiljana Trajkovic Subject: Re: Your presentations on February 23. Dear Ljiljana, Here is some interesting website on Network Management: The first one is a very simple introduction to beginner on Network management: http://www.inforamp.net/~kjvallil/t/snmp.html The second one provides very good links and information on network management, including software, RFCs and tutorials. http://snmp.cs.utwente.nl/ The third one is a little bit slower, it is called THE WORLD OF SNMP AND NETWORK MANAGEMENT: http://www.snmpworld.com/ Hope this will be helpful. Have a nice day! Best Regards, Shi Cong ******************************************************************************* From tlie@grus.cs.sfu.ca Fri Feb 18 14:55:17 2000 From: Tianyi Li To: Ljiljana Trajkovic Subject: Useful website Hi, http://www.cisco.com/public/solution_overviews.shtml Above is the solution part in Cisco website, and it contains lots of resourses on nearly all kinds of high speed network essential knowledge, together with a not-bad 'terms and acronyms' dictionary, and Cisco's own product general information. However, I think it is a good idea that taking a look from the viewpoint of the industry leader in networking while Cisco has got the 2nd chair in market value in the world, only after Microsoft. Please evaluate it. Rds Tianyi ******************************************************************************* From rafrasht@cs.sfu.ca Thu Mar 2 14:57:41 2000 From: Reza Afrashteh Date: Thu, 2 Mar 2000 14:57:38 -0800 (PST) To: ljilja@cs.sfu.ca Subject: Some problems and their answers Cc: xzhong@cs.sfu.ca Question: Isn't it possible to save Opnet model,ceated by students on hard disk? Answer: 1. the user's op_admin dir should be rx by others as the system needs to read it when starting opnet. please chmod -R 755 op_admin under his home dir. 2. in the env_db6.0 under mod_dirs, the first entry has been user's own model dir, e.g.: /ensc/grad1/gchenc/op_models 3. ~gchenc/op_admin/env_db6.0 was set to /ensc/dept1/chao/op_models, thus he can not save the files. I modified his entry to /ensc/grad1/gchenc/op_models. please mkdir op_models under his home dir. Hope, this helps. Chao Cheng, Network Support Group, Centre for Systems Science ****************************************************************************** Question When I am simulating one example of tuturial, it gives me the below errors: How can Ifix this? Hi When I am simulating one example of tuturial, it gives me the below errors: How can Ifix this? Reza |-----------------------------------------------------------------------------| | <<< Recoverable Error >>> | | Object repository construction failed | | due to errors encountered by the binder program (bind_so_unix) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Errors reported by the binder program follow | | (these messages have been saved in /op_admin/tmp/bind_err): | |-----------------------------------------------------------------------------| ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found mv: cannot access /tmp/bso12345.o ld: Command not found <<< Program Abort >>> * Time: 14:30:48 Mon Feb 07 2000 * Program: op_runsim (Version 6.0.L PL3) * Error: Error encountered rebuilding repository -- unable to proceed |-----------------------------------------------------------------------------| | <<< Program Abort >>> | | Error encountered rebuilding repository -- unable to proceed | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Simulation Completed - Collating Results. | | Simulated: Time (-1 sec.), Events (0) | | Time: Elapsed (0 sec.) | | Simulation Log: 2 entries | |-----------------------------------------------------------------------------| Answer: make sure you has /usr/ucb in you path then try it again. Chao Cheng, Network Support Group, Centre for Systems Science ****************************************************************************** Question: When I ran the first example, 'Small Internetworks', after adding the 2nd floor into the scenario, then ran the simulation, I got the following message, which says that process model 'eigrp' cannot be read. Do you know the reason? Thanks. |-----------------------------------------------------------------------------| | Reading network model (jliud_Sm_Int-extension) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Reading result collection file: (jliud_Sm_Int-extension) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Verifying model consistency. | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | <<< Recoverable Error >>> | | Unable to read process model: eigrp | | The specified file does not exist. | |-----------------------------------------------------------------------------| <<< Program Abort >>> * Time: 16:56:57 Fri Feb 11 2000 * Program: op_runsim (Version 6.0.L PL9) * Error: Unable to construct lists of constituent models |-----------------------------------------------------------------------------| | <<< Program Abort >>> | | Unable to construct lists of constituent models | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Simulation Completed - Collating Results. | | Simulated: Time (-1sec.), Events (0) | | Time: Elapsed (0 sec.) | |-----------------------------------------------------------------------------| Answer: Make sure that the /usr/local-bell/opnet6/6.0.L/models/std/eigrp directory is included in the mod_dirs preference in ~/op_admin/env_db6.0 Sincerely, Jerome Technical Support OPNET Technologies, Inc. ******************************************************************************* From rafrasht@cs.sfu.ca Fri Mar 3 09:03:32 2000 Date: Fri, 3 Mar 2000 09:03:31 -0800 (PST) From: Reza Afrashteh Subject: (TSC 60658) To: ljilja@cs.sfu.ca, xzhong@cs.sfu.ca ------------- Begin Forwarded Message ------------- From: opnet@mil3.com Mime-Version: 1.0 Date: 03 Mar 00 10:13 -0500 To: rafrasht@cs.sfu.ca Subject: (TSC 60658) Dear Reza, The reason the students do not see the given error correction model for the link is that that 1. the model is not present in any of your model directories, or 2. if the model is present, the corresponding object file si not in the model corresponding directory. 1. Please, make sure that the environment database (file typically located under user home directory, op_admin subdirectory, and called env_db5.1 (for 5.1 release), or env_db6.0 (for 6.0 release), has a directory //models/links included under "mod_dirs". If the directory is not present, insert it. 2. Make sure that the above directory includes file dpt_ecc. If the file is not there, get it from the CD. If the file is there, check whether the corresponding object file is there. If is it is not there, compile is using >op_mko -type ps -m dpt_ecc Hopefully, the above should solve the problem. I am closing the case. Please, let us know if the problem persists, and we will re-open the case. Please, when contacting us again, let us also know the OPNET release, patch level and platform (Unix, NT) you are using. Thank you. Best regards, Dana ******************************* OPNET Technical Support opnet@mil3.com (202) 364.8889 ******************************** I get several errors by students(grad students) they try to run some example of tutorial and they get the errors which m3_vuerr command doesn't help. Below are 2 complaints that I would appreciate if you could help me to answer them. (I am sending this email from Simon Fraser University, BC, Canada and our ID is 7384) Thanks Reza **************************************************************************** Hi Reza, It seems we'll have to constantly bug you about opnet bugs. I got over the initial error I told you about in lesson Packet switched network 1. Eventually it turned out to be a typo. But still it doesn't woark, there is another problem. In the beginning of the lesson we are asked to specify a new link type that will use our own packet format, in that we have to set some attributes, for the attribute ecc_model we have to select dpt_ecc from the selection menu according to the lesson. But such a model is not available in the menu. This gives rise to an error latter on when we try to run the simulation. Another student in our class Kathri Balasubramanium also had this same problem. May be this is something you'll have to ask the Opnet people about. ------------- End Forwarded Message ------------- ******************************************************************************* From xzhong@cs.sfu.ca Tue Apr 4 22:45:53 2000 Date: Tue, 04 Apr 2000 22:45:51 -0700 From: Xiang Zhong To: Ljiljana Trajkovic Subject: Error again in Opnet simulation Date: Fri, 24 Mar 2000 14:16:24 -0800 (PST) From: Tianyi Li To: Xiang Zhong , Nazy Alborz CC: Ljiljana Trajkovic Since last evening, I have encounter many strange error messages in simulation. And each time it appeared to be different. To conclude what they are, I start a simulation all over again. After I launch Opnet, New a project, put a ATM_uni_server_ini node in the scenario, and simulate, the following message appears: I tried it again, the same error message given out. I tried it again while replace the server by ATM_server_ini, the samething happened. I tried it with a ATM switch which I can simulate successfully this morning, the same thing happened. I tried LOGGING IN FROM DIFFERENT USER ID, the same thing happened, also said 'relocation error' Now I cannot get other error messages, but I really encounter them this morning, including the one William and I both encountered last evening. And, my colleague Lily Wang has encountered the following message last evening, but I didn't. And today she made up a new folder of op_admin, and all problems solved, but I cannot solve any error doing the same way. The situation is so complicated, Please help! I believe there must be some after the following error has been settled. And, last evening, I guessed Bell has something wrong, Wenmin had encountered an error that cannot open opnet, but this error turned to be ok after 1/2 hour without doing anything. And I found sometimes I cannot even use 'ls' command after log in bell (error: segmentation error). |-----------------------------------------------------------------------------| | || || ||| ||| |||| Network Simulation of: | | ||| ||| ||| ||| | ||| project1-scenario1 | | |||| |||| ||| ||| ||| | | ||||||||| ||| ||| ||| Simulation and Model Library Copyright | | |||| |||| ||| ||| | ||| 1987-1999 by MIL 3, Incorporated as a | | |||| |||| ||| |||||| |||| part of OPNET Release 6.0.L | |-----------------------------------------------------------------------------| | Modeling Technologies for the Third Millennium | |-----------------------------------------------------------------------------| | MIL 3, Inc. / 3400 International Drive / Washington, DC, USA | | TEL: +1.202.364.4700 / FAX: +1.202.364.6182 | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Reading network model (project1-scenario1) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Verifying model consistency. | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Rebuilding scenario model library. | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Loading scenario model library. | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | <<< Recoverable Error >>> | | error loading object file (/cs/grad2/tlie/op_admin/project1-scenario1.s1.nt.so) | | ld.so.1: /usr/local-bell/opnet6/6.0.L/sys/sun_sparc_solaris/bin/op_runsim: fatal: relocation error: | | symbol not found: app_mgr_mta_params_info_obtain: referenced in /cs/grad2/tlie/op_admin/project1-sc | |-----------------------------------------------------------------------------| <<< Program Abort >>> * Time: 14:01:37 Fri Mar 24 2000 * Program: op_runsim (Version 6.0.L PL9) * Error: Error loading scenario model library -- unable to proceed |-----------------------------------------------------------------------------| | <<< Program Abort >>> | | Error loading scenario model library -- unable to proceed | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Simulation Completed - Collating Results. | | Simulated: Time (-1 sec.), Events (0) | | Time: Elapsed (0 sec.) | | Simulation Log: 1 entries | |-----------------------------------------------------------------------------| Fix: Clear all the file in op_admin directory. restart opnet. ****************************************************************************** Subject: Another error occurs--about memory segmentation error Date: Fri, 24 Mar 2000 19:39:59 -0800 (PST) From: Tianyi Li To: Reza Afrashteh , Xiang Zhong CC: Ljiljana Trajkovic Hi, Sorry for interruption. I just wanna report another error I encountered just now. And I am quite sure this error would be disappearing like it did today. I had this error on different users at different workstations at different time (yesterday and now), and I guess it might be caused by the insufficient memory on Bell. Below is the message: |-----------------------------------------------------------------------------| | || || ||| ||| |||| Network Simulation of: | | ||| ||| ||| ||| | ||| hn_0324_atmvsppp-scenario1 | | |||| |||| ||| ||| ||| | | ||||||||| ||| ||| ||| Simulation and Model Library Copyright | | |||| |||| ||| ||| | ||| 1987-1999 by MIL 3, Incorporated as a | | |||| |||| ||| |||||| |||| part of OPNET Release 6.0.L | |-----------------------------------------------------------------------------| | Modeling Technologies for the Third Millennium | |-----------------------------------------------------------------------------| | MIL 3, Inc. / 3400 International Drive / Washington, DC, USA | | TEL: +1.202.364.4700 / FAX: +1.202.364.6182 | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Reading network model (hn_0324_atmvsppp-scenario1) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Reading result collection file: (hn_0324_atmvsppp-scenario1) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Verifying model consistency. | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Rebuilding scenario model library. | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | <<< Recoverable Error >>> | | Object repository construction failed | | due to errors encountered by the binder program (bind_so_unix) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Errors reported by the binder program follow | | (these messages have been saved in /op_admin/tmp/bind_err): | |-----------------------------------------------------------------------------| ld: fatal: file /tmp/bso6419.o: cannot msync file; errno=12 mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /tmp/bso6419.o mv: cannot access /tmp/bso6419.o ld: fatal: file /tmp/bso6419a.o: cannot open file; errno=2 ld: fatal: File processing errors. No output written to /cs/grad1/hzhangb/op_admin/hn_0324_atmvsppp-scenario1.s1.nt.so <<< Program Abort >>> * Time: 17:35:29 Fri Mar 24 2000 * Program: op_runsim (Version 6.0.L PL9) * Error: Error encountered rebuilding repository -- unable to proceed |-----------------------------------------------------------------------------| | <<< Program Abort >>> | | Error encountered rebuilding repository -- unable to proceed | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Simulation Completed - Collating Results. | | Simulated: Time (-1 sec.), Events (0) | | Time: Elapsed (0 sec.) | | Simulation Log: 2 entries | |-----------------------------------------------------------------------------| Fix: Probably caused by too much files in /tmp directory. Try to delete them. ****************************************************************************** Subject: Another error in Opnet Date: Tue, 22 Feb 2000 21:23:22 -0800 (PST) From: Tianyi Li To: Xiang Zhong When I was trying to use opnet to construct project with a router, opnet always give me an error message "unable to load process model: eigrp". I also got error message in console as following: |-----------------------------------------------------------------------------| | Reading network model (ty_test5-scenario2) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Verifying model consistency. | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | <<< Recoverable Error >>> | | Unable to read process model: eigrp | | The specified file does not exist. | |-----------------------------------------------------------------------------| <<< Program Abort >>> * Time: 19:58:36 Wed Mar 01 2000 * Program: op_runsim (Version 6.0.L PL9) * Error: Unable to construct lists of constituent models |-----------------------------------------------------------------------------| | <<< Program Abort >>> | | Unable to construct lists of constituent models | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | Simulation Completed - Collating Results. | | Simulated: Time (-1 sec.), Events (0) | | Time: Elapsed (0 sec.) | | Simulation Log: 1 entries | |-----------------------------------------------------------------------------| Fix: (This is the most frequently asked question. Quite a lot students asked similiar question. The only difference is the file that can not found.) It is caused by fail to include the diretory that hold the file needed. Just locate the file and add its path in under the item "mod_dir" in file "env_db6.0" ****************************************************************************** Subject: opnet problem Date: Tue, 22 Feb 2000 11:17:45 -0800 From: wli To: rafrasht@cs.sfu.ca, xzhong@cs.sfu.ca, ljilja@cs.sfu.ca Hi Reza and William, When I ran lesson 1 of opnet tutorial, the first part: first_floor's simulation works just fine. Then, I move to the second part: expansin's simulation. But, it doesn't work. The error message is the following: Reading result collection file: (expansion-expansion) --- verifying model consistency. << Recoverable Error>> Unable to read process model: eigrp The specified file does not exist. <> Unable to construct lists of constituent models --- Simulation completed - Collating Results. Did you have this kind problem before? Could you tell me how to fix it? Thanks! Wenmin Fix: 1. First step is same as the previous Fix. (This is the most frequently asked question. Quite a lot students asked similiar question. The only difference is the file that can not found.) It is caused by fail to include the diretory that hold the file needed. Just locate the file and add its path in under the item "mod_dir" in file "env_db6.0" 2. make sure op_admin directory is writable by group users. ****************************************************************************** Date: Thu, 9 Mar 2000 18:49:09 -0800 (PST) From: Nazy Alborz Subject: Opnet problem To: rafrasht@cs.sfu.ca Cc: ljilja@cs.sfu.ca Hi Reza, I'm receiveing the following error while running my ATM network. It is related to the Kernel function op_pro_invoke() of the ATM.mgmt module of atm_uni_scr_adv node model in Opnet. The troubleshooting section says that it is due to a malformed prohandle but I've checked the enter and exit executives of all states of my process model. All the prothandles are defined correctly and nothing seems to be malformed. Can you just pass this e-mail to MIL-3 maybe thay have a solution. Thanks in advance, Nazy _______________________________________________________________________________ The Error message is: <<< Program Fault >>> * Time: 17:58:33 Tue Mar 07 2000 * Program: op_runsim (Version 6.0.L PL9) * Error: program abort -- segmentation violation T (1.32568e-05), EV (364), MOD (top.Campus Network.node_9.ATM_mgmt), KP (op_pro_invoke) |-----------------------------------------------------------------------------| | <<< Program Fault >>> | | program abort -- segmentation violation | | T (1.32568e-05), EV (364), MOD (top.Campus Network.node_9.ATM_mgmt), KP (op_pro_invoke) | |-----------------------------------------------------------------------------| Answer: From: opnet@mil3.com Date: 16 Mar 00 15:19 -0500 To: rafrasht@cs.sfu.ca Subject: RE: Opnet problem (TSC 61080) Dear Reza, From the forwarded message in your email, it looks like your colleague is using a process model that he/she has built from scratch or one that is a modified version of one our standard process models. In either case, segmentation violations cannot always be resolved by just looking at the code. Often, you have to use a C/C++ debugger in conjunction with an OPNET simulation to find and fix the problem. Therefore, I'd recommend that your colleague use a C/C++ debugger (such as gdb or xxgdb on UNIX platforms) to resolve this problem. Here is a Q&A from our Tech. Support FAQ database which is accessible from our website: Q: How do I use xxgdb with a dynamically linked OPNET simulation? A: On UNIX platforms: 0. Compile all model code with " -g" option. This is done by specifying " -g" (note leading space after the opening double quotes) as the value of the "comp_flags" OPNET environment attribute. 1. Run the simulation in debug mode, i.e. choose the debug environment file in the Simulation Sequence editor. 2. From the command line, figure out the process id of the op_runsim process: % ps -ef | grep op_runsim 3. Run "xxgdb". 4. Attach the dynamic simulation with the it -- from "(xxgdb)" prompt, type: (xxgdb) attach 5. At this point, xxgdb controls the simulation. You can set breakpoints any time by typing: (xxgdb) break or by opening the file in the code area of xxgdb and graphically setting the break point using the "break" option button. 6. In order to pass control to "ODB", type: (xxgdb) cont 7. Now ODB controls the simulation execution and you can use ODB commands to move on. 8. If you had set breakpoints in "xxgdb", control will go back to xxgdb if that is hit. Otherwise, simulation control remains with ODB. 9. If you want to defer setting breakpoints in xxgdb (e.g., after a certain event number), navigate using ODB to go to the event of interest. Then, in order to pass control to "xxgdb", hit +; "xxgdb" should now control the simulation execution. 10. Special note for process model code debugging: In order to check the values of model "state variables" use the following convention: (xxgdb) print op_sv_ptr-> Sincerely, Alfred S. Jones Technical Support OPNET Technologies, Inc. ******************************************************************************** Date: Tue, 14 Mar 2000 18:08:29 -0800 From: Prabha Ranaweera To: reza afrashteh , Xiang Zhong , Ljiljana Trajkovic Subject: OPNET problem Hello, I want to use one of the external c files available in the directory /usr/local-bell/opnet6/6.0.L/models/std/lane in one of my process models. The files I want to include are lane_notify_log_support_v2.ex.c and lane_v2.ex.c. In order to be able to include them, the object files of the above source files should be available. I looked at the reference manual but could not find a way of compiling them so that I would be able to produce the object code. Would you be able to get them compiled for me, so that the object code ( a file with ex.o extension) is available. Thank you Prabha Answer: From: opnet@mil3.com Date: 17 Mar 00 12:20 -0500 To: rafrasht@cs.sfu.ca Subject: RE: opnet problem (TSC 61123) Dear Reza, Given a file named X.ex.c in directory D, you would 1) make sure that D is in the mod_dirs preference in the env_db6.0, and 2) compile using 'op_mko -m X -type ex' Sincerely, Jerome Technical Support OPNET Technologies, Inc. ************************************************************************ Date: Wed, 22 Mar 2000 20:03:04 -0800 From: Prabha Ranaweera To: Ljiljana Trajkovic , reza afrashteh , Xiang Zhong Subject: opnet problem Hello, I tried to run the simulation of the already existing library model of LANE. But since some object files are not available and I don't have the execution right to theses directories, I got the following error. I attempted copying every thing into my home directory and compiling them, but even that did not work. Then I got a binding error. I think that if all the non compiled process models (with .pr.m extension) and all the non compiled external files (with .ex.c extension) in the following directories are complied by the system administator I will be able to over come this problem. Or if possible I'd like to get the execution permission for the following directories. /usr/local-bell/opnet6/6.0.L/models/std/lane /usr/local-bell/opnet6/6.0.L/models/std/rip /ospf /eigrp /igrp /bgp /utilities/oms I'd like to get this done as soon as possible, since there is not much time left to finish the project. Thank you Compiling process model (ospf_interface_v2) | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | <<< Recoverable Error >>> | | Unable to open file | | (: Permission denied) | | File Path: /usr/local-bell/opnet6/6.0.L/models/std/ospf/ospf_interface_v2.pr.c | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | <<< Recoverable Error >>> | | Process model compilation failed | | Source code file couldn't be generated | |-----------------------------------------------------------------------------| |-----------------------------------------------------------------------------| | <<< Recoverable Error >>> | | Object repository construction failed | | Process model (/usr/local-bell/opnet6/6.0.L/models/std/ospf/ospf_interface_v2.pr.c) compilation fai | | ed, see /cs/grad2/prabha/op_admin/tmp/cc_err. Answer: done with /usr/local-bell/opnet6/6.0.L/models/std/ospf/ospf_interface_v2.pr.c, please try it. All user have rx permission to the model directories. But, they can not write or create any files there. In the past and with normal practice, the users would copy the source (*.c file) to his/her own model directory to complete the projects. Since model directory is a source directory that we can not give the write permission to the users thus staff have to do the compile or we will be glad to give permission to a designate person who will be responsible for the integrity of the files. Make sure put your own model dir as the first entry of the model dir entries in your env file. Try "which bin_so_unix" command to see if you have the path set correctly. Also, you can give me the list of the *.c files that you need to compile. Chao Cheng, Network Support Group, Centre for Systems Science ******************************************************************************** Date: Thu, 30 Mar 2000 17:29:37 -0800 (PST) From: Nazy Alborz Subject: Opnet Problem To: ensc894-g1@sfu.ca Hi, I ran into an Opnet Problem which might also has happened to you. There is a file in your op_admin directory named, session_log which keeps track of whatever you have done in Opnet. This file might get very large and cause your simulation not be run. If you get an error message saying "disk quota exceeded" it can be related to this file. Just keep in mind that you can't delete this file entirly and you'd better decrease its load. Good luck with your projects, Nazy *****************************************************************************