From bharathi@usc.edu Wed Jan 16 22:32:49 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g0H6WnR16009 for ; Wed, 16 Jan 2002 22:32:49 -0800 (PST) Received: from aludra.usc.edu (bharathi@aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id WAA16150 for ; Wed, 16 Jan 2002 22:33:21 -0800 (PST) Received: from localhost (bharathi@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g0H6XKY26964 for ; Wed, 16 Jan 2002 22:33:20 -0800 (PST) Date: Wed, 16 Jan 2002 22:33:20 -0800 (PST) From: Shishir B S To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Want to swap to Wed's class ? Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi Everyone, If anyone is interested in swapping into Wednesday's class, pls let me know asap. Thanks and regards, Shishir From vkhanna@qualcomm.com Fri Jan 18 18:44:48 2002 Received: from illyana.qualcomm.com (illyana.qualcomm.com [129.46.65.100]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g0J2imR13484 for ; Fri, 18 Jan 2002 18:44:48 -0800 (PST) Received: from VKHANNA.qualcomm.com (vkhanna.qualcomm.com [129.46.21.96]) by illyana.qualcomm.com (8.12.1/8.12.1/1.0) with ESMTP id g0J2ij5M011240 for ; Fri, 18 Jan 2002 18:44:45 -0800 (PST) Message-Id: <5.1.0.14.2.20020118184420.00b0cb68@illyana> X-Sender: vkhanna@illyana X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Fri, 18 Jan 2002 18:45:04 -0800 To: csci551-talk@mailman.isi.edu From: Vivek Khanna In-Reply-To: <200201172005.g0HK56R22268@gamma.isi.edu> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Subject: [Csci551-talk] Unable to view HW1 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Unable to view HW 1 on web-site --Vivek From tiesongz@usc.edu Sun Jan 20 09:38:07 2002 Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g0KHc7R24165 for ; Sun, 20 Jan 2002 09:38:07 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id JAA12034 for ; Sun, 20 Jan 2002 09:37:25 -0800 (PST) Received: from janny (ppp-227-089.usc.edu [128.125.227.89]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g0KHamq21419 for ; Sun, 20 Jan 2002 09:36:48 -0800 (PST) Message-ID: <004f01c1a1d8$9fc54cc0$59e37d80@janny> From: "tiesongz" To: Date: Sun, 20 Jan 2002 09:33:42 -0800 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_004C_01C1A195.8CCB0940" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Subject: [Csci551-talk] Is there anyone who have downloaded HW1? Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: This is a multi-part message in MIME format. ------=_NextPart_000_004C_01C1A195.8CCB0940 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I can't download HW1 pdf file and see the webpage of HW1. Is there = anyone who have downloaded HW1? Please tell me how to do it. Thanks ------=_NextPart_000_004C_01C1A195.8CCB0940 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I can't download HW1 pdf file and see = the webpage=20 of HW1. Is there anyone who have downloaded HW1? Please tell me how to = do=20 it.
 
Thanks
------=_NextPart_000_004C_01C1A195.8CCB0940-- From srihari@USC.Edu Mon Jan 21 12:48:39 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g0LKmcR15865 for ; Mon, 21 Jan 2002 12:48:38 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id MAA28604 for ; Mon, 21 Jan 2002 12:49:11 -0800 (PST) Received: from aludra.usc.edu (srihari@aludra.usc.edu [128.125.253.184]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g0LKmaq29673 for ; Mon, 21 Jan 2002 12:48:36 -0800 (PST) Date: Mon, 21 Jan 2002 12:49:10 -0800 (PST) From: Prithvi Srihari To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] HW 1 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Can anyone access HW 1? Please let me know how to do it. Thanks. Prithvi Srihari. Student, The University of Southern California. From johnh@ISI.EDU Mon Jan 21 14:13:46 2002 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 g0LMDkR05088 for ; Mon, 21 Jan 2002 14:13:46 -0800 (PST) Received: from dash.isi.edu (hlessil.isi.edu [128.9.97.35]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g0LMDjg19273 for ; Mon, 21 Jan 2002 14:13:45 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.2) with ESMTP id g0LMBWg21767 for ; Mon, 21 Jan 2002 14:11:32 -0800 Message-Id: <200201212211.g0LMBWg21767@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: csci551-talk@ISI.EDU MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Mon, 21 Jan 2002 14:11:32 -0800 From: John Heidemann Subject: [Csci551-talk] homework 1 now readable Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Homework one should now be readable on the web page. -John Heidemann From tiesongz@usc.edu Mon Jan 21 16:34:06 2002 Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g0M0Y6R06517 for ; Mon, 21 Jan 2002 16:34:06 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id QAA18871 for ; Mon, 21 Jan 2002 16:34:39 -0800 (PST) Received: from janny (ppp-229-184.usc.edu [128.125.229.184]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g0M0Y2q18796 for ; Mon, 21 Jan 2002 16:34:02 -0800 (PST) Message-ID: <001501c1a2dc$137abc00$b8e57d80@janny> From: "tiesongz" To: Date: Mon, 21 Jan 2002 16:30:56 -0800 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0012_01C1A299.0055BF00" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Subject: [Csci551-talk] About HW1 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: This is a multi-part message in MIME format. ------=_NextPart_000_0012_01C1A299.0055BF00 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Is there still a limit that no more than 20 words per question? Thanks ------=_NextPart_000_0012_01C1A299.0055BF00 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Is there still a limit that no more = than 20 words=20 per question?
 
Thanks
 
 
------=_NextPart_000_0012_01C1A299.0055BF00-- From johnh@ISI.EDU Mon Jan 21 17:14:31 2002 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 g0M1EVR16117 for ; Mon, 21 Jan 2002 17:14:31 -0800 (PST) Received: from dash.isi.edu (hlessil.isi.edu [128.9.97.35]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g0M1EUg15815 for ; Mon, 21 Jan 2002 17:14:30 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.2) with ESMTP id g0M1CHZ22876; Mon, 21 Jan 2002 17:12:17 -0800 Message-Id: <200201220112.g0M1CHZ22876@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: "tiesongz" Cc: Csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] About HW1 In-reply-to: <001501c1a2dc$137abc00$b8e57d80@janny> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Mon, 21 Jan 2002 17:12:17 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Mon, 21 Jan 2002 16:30:56 PST, "tiesongz" wrote: >Is there still a limit that no more than 20 words per question? > Correct. (Any changes are already listed on the homework 1 web page.) -John Heidemann From vkhanna@qualcomm.com Tue Jan 22 19:14:28 2002 Received: from illyana.qualcomm.com (illyana.qualcomm.com [129.46.65.100]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g0N3ERR06069 for ; Tue, 22 Jan 2002 19:14:28 -0800 (PST) Received: from VKHANNA.qualcomm.com (vkhanna.qualcomm.com [129.46.21.96]) by illyana.qualcomm.com (8.12.1/8.12.1/1.0) with ESMTP id g0N3ENe6013341 for ; Tue, 22 Jan 2002 19:14:23 -0800 (PST) Message-Id: <5.1.0.14.2.20020122191154.00b0f360@illyana> X-Sender: vkhanna@illyana X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Tue, 22 Jan 2002 19:14:52 -0800 To: csci551-talk@mailman.isi.edu From: Vivek Khanna Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Subject: [Csci551-talk] Unable to print papers from outside USC Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hello Professor, I am a remote student based in San Diego and I am unable to print the papers meant to be read as they are USC Internal Network. Can you put them on the DEN web-site for CSCS-551 so that I can print them. Regards Vivek From johnh@ISI.EDU Tue Jan 22 22:42:46 2002 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 g0N6gjR20124 for ; Tue, 22 Jan 2002 22:42:46 -0800 (PST) Received: from dash.isi.edu (hlessil.isi.edu [128.9.97.35]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g0N6ghg22349 for ; Tue, 22 Jan 2002 22:42:43 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.2) with ESMTP id g0N6eRF02224; Tue, 22 Jan 2002 22:40:29 -0800 Message-Id: <200201230640.g0N6eRF02224@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Vivek Khanna Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Unable to print papers from outside USC In-reply-to: <5.1.0.14.2.20020122191154.00b0f360@illyana> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Tue, 22 Jan 2002 22:40:26 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Tue, 22 Jan 2002 19:14:52 PST, Vivek Khanna wrote: >Hello Professor, > >I am a remote student based in San Diego and I am unable to print the >papers meant to be read as they are USC Internal Network. Can you put them >on the DEN web-site for CSCS-551 so that I can print them. That is not not the case, all the papers are on the public web. (In fact, they are on a server at ISI, not at USC.) They ARE password protected with the password given to you on the first day of class. Perhaps you're not using the password. In general, by the way, if you have problems, please report the exact error message you're getting, and where you encounter that error. -John Heidemann From johnh@ISI.EDU Wed Jan 23 21:20:18 2002 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 g0O5KIR11769 for ; Wed, 23 Jan 2002 21:20:18 -0800 (PST) Received: from dash.isi.edu (hlessil.isi.edu [128.9.97.35]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g0O5KEg10004; Wed, 23 Jan 2002 21:20:14 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.2) with ESMTP id g0O5HxU05767; Wed, 23 Jan 2002 21:18:00 -0800 Message-Id: <200201240518.g0O5HxU05767@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: csci551-talk@ISI.EDU Cc: Alefiya Hussain MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Wed, 23 Jan 2002 21:17:59 -0800 From: John Heidemann Subject: [Csci551-talk] hw1 deadline extended Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Because of some problems distributing the diag exam I am extending the deadline for HW1 until 3pm Wed. Jan. 30 (during the TA's office hours). Details are on the class web page. -John Heidemann From mtabibia@usc.edu Sun Jan 27 14:51:57 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g0RMpvR12174 for ; Sun, 27 Jan 2002 14:51:57 -0800 (PST) Received: from scf-fs.usc.edu (scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id OAA10945 for ; Sun, 27 Jan 2002 14:52:32 -0800 (PST) Received: from D715H111 (adsl-64-166-98-50.dsl.lsan03.pacbell.net [64.166.98.50]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g0RMpPq02547 for ; Sun, 27 Jan 2002 14:51:25 -0800 (PST) From: "Mohammad Tabibian" To: Date: Sun, 27 Jan 2002 14:49:54 -0800 Message-ID: <000601c1a784$f63f26a0$6a01a8c0@D715H111> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0007_01C1A741.E81BE6A0" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.2627 Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Subject: [Csci551-talk] I can't print the papers! Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: This is a multi-part message in MIME format. ------=_NextPart_000_0007_01C1A741.E81BE6A0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hi all, I am unable to print some of class papers such as Clark88a.pdf (except first two pages!!) or Saltzer81a.pdf. I tried from USC computers or even from my home printer. I think there is an error in the pdf files. The message which I receive is "Failed to print". If anyone can tell me what I have to do, that will be a great help! Bests, Mohammad ------=_NextPart_000_0007_01C1A741.E81BE6A0 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hi all,

 

I am unable to print some of class papers such as = Clark88a.pdf  = (except first two pages!!) or Saltzer81a.pdf. I tried from USC computers or even = from my home printer. I think there is an error in the pdf files. The message which I receive is “Failed to print”. =

 

If anyone can tell me what I have to do, that will be = a great help!

 

Bests,

Mohammad

 

------=_NextPart_000_0007_01C1A741.E81BE6A0-- From johnh@ISI.EDU Mon Jan 28 22:19:13 2002 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 g0T6JDR11382 for ; Mon, 28 Jan 2002 22:19:13 -0800 (PST) Received: from dash.isi.edu (12-224-252-187.client.attbi.com [12.224.252.187]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g0T6JCg14018 for ; Mon, 28 Jan 2002 22:19:12 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.2) with ESMTP id g0T6Ghg15181; Mon, 28 Jan 2002 22:16:44 -0800 Message-Id: <200201290616.g0T6Ghg15181@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: "Mohammad Tabibian" Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] I can't print the papers! In-reply-to: <000601c1a784$f63f26a0$6a01a8c0@D715H111> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=X-UNKNOWN Date: Mon, 28 Jan 2002 22:16:43 -0800 From: John Heidemann Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by gamma.isi.edu id g0T6JDR11382 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 27 Jan 2002 14:49:54 PST, "Mohammad Tabibian" wrote: >I am unable to print some of class papers such as Clark88a.pdf (except first >two pages!!) or Saltzer81a.pdf. I tried from USC computers or even from my >home printer. I think there is an error in the pdf files. The message which I >receive is Failed to print. Can you view them in acroread or whatever you're using? I believe all the pdf files work for me. -John Heidemann From hussain@ISI.EDU Wed Jan 30 14:33:21 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g0UMXKR05281 for ; Wed, 30 Jan 2002 14:33:20 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g0UMXKi00726 for ; Wed, 30 Jan 2002 14:33:20 -0800 (PST) Date: Wed, 30 Jan 2002 14:33:20 -0800 (PST) From: Alefiya Hussain To: csci551-talk@ISI.EDU Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] TA office hours and location Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: My office hours and location information is present on the administrative page located at http://www.isi.edu/~johnh/WORK/CS551/SP2002/BROCHURE/index.html Hw is due today by 4:30pm in SAL 235. Thank you, alefiya From mhr@usc.edu Tue Feb 12 21:35:25 2002 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 g1D5ZPR01734 for ; Tue, 12 Feb 2002 21:35:25 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1D5ZPO21060 for ; Tue, 12 Feb 2002 21:35:25 -0800 (PST) Received: from aludra.usc.edu (mhr@aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id VAA08410 for ; Tue, 12 Feb 2002 21:36:02 -0800 (PST) Received: from localhost (mhr@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1D5a3X10011 for ; Tue, 12 Feb 2002 21:36:03 -0800 (PST) Date: Tue, 12 Feb 2002 21:36:02 -0800 (PST) From: mhr To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] question Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: does anybody knows how i can get rid of zombie processes at the end of a program -m From bharathi@usc.edu Wed Feb 13 08:49:50 2002 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 g1DGnoR28542 for ; Wed, 13 Feb 2002 08:49:50 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1DGnnO06146 for ; Wed, 13 Feb 2002 08:49:49 -0800 (PST) Received: from aludra.usc.edu (bharathi@aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id IAA17902; Wed, 13 Feb 2002 08:50:25 -0800 (PST) Received: from localhost (bharathi@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1DGoPp02084; Wed, 13 Feb 2002 08:50:25 -0800 (PST) Date: Wed, 13 Feb 2002 08:50:25 -0800 (PST) From: Shishir B S To: mhr cc: Subject: Re: [Csci551-talk] question In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: use wait() in your signal handler for SIGCHLD. The man pages for wait() and signal() will prove helpful. On Tue, 12 Feb 2002, mhr wrote: > > does anybody knows how i can get rid of zombie processes > at the end of a program > > -m > > From rohanbhindwale@yahoo.com Sun Feb 17 23:24:29 2002 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 g1I7OSR19898 for ; Sun, 17 Feb 2002 23:24:29 -0800 (PST) Received: from web12302.mail.yahoo.com (web12302.mail.yahoo.com [216.136.173.100]) by tnt.isi.edu (8.11.6/8.11.2) with SMTP id g1I7OSO28505 for ; Sun, 17 Feb 2002 23:24:28 -0800 (PST) Message-ID: <20020218072428.84772.qmail@web12302.mail.yahoo.com> Received: from [128.125.5.191] by web12302.mail.yahoo.com via HTTP; Sun, 17 Feb 2002 23:24:28 PST Date: Sun, 17 Feb 2002 23:24:28 -0800 (PST) From: rohan bhindwale To: csci551-talk@ISI.EDU MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: [Csci551-talk] Project Queries Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: A few queries about the project 1. The routers and the manager are to communicate using TCP connection . Are we to hardcode the ports that they communicate over. Ofcourse the manager will have one port and the routers can also all communicate via one port. Is there any other way we are to assign these ports ?? 2. The routers are to communicate with each other using UPD . This info will be passed from the router to the manager on the TCP connection and then the manager will pass this info to that routers neighbour. Again do we hardcode this port number in each router ?? 3. Stevens uses read and write to send data whereas Beejs guide uses send and recv . which one is preferred ?? Thanks for the responses , rohan __________________________________________________ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.yahoo.com From rohanbhindwale@yahoo.com Sun Feb 17 23:54:49 2002 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 g1I7snR24211 for ; Sun, 17 Feb 2002 23:54:49 -0800 (PST) Received: from web12306.mail.yahoo.com (web12306.mail.yahoo.com [216.136.173.104]) by tnt.isi.edu (8.11.6/8.11.2) with SMTP id g1I7smO02198 for ; Sun, 17 Feb 2002 23:54:48 -0800 (PST) Message-ID: <20020218075448.1297.qmail@web12306.mail.yahoo.com> Received: from [128.125.5.191] by web12306.mail.yahoo.com via HTTP; Sun, 17 Feb 2002 23:54:48 PST Date: Sun, 17 Feb 2002 23:54:48 -0800 (PST) From: rohan bhindwale To: usc_cs551@yahoogroups.com Cc: csci551-talk@ISI.EDU In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: [Csci551-talk] Data Format Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: When we send an object onto the network do we have to convert it into network byte order. No where in Stevens do they do that . But they also use char strings exclusively for sending data. Will it suffice if we cast a packet to (char * ) before sending it ?? --- fundooz_2000 wrote: > Hi guys, > Remember the problem that Prof. gave us during the > first sub > homework#1.... > > 1 Mbps line > Each use uses 0.1 Mbps only > and each user is active 10% of the time > > 10 users can use it if Circuit switching is used > and 35 users can use it if Packet switching is used > Prove it... > and something else abt 0.004% probability... > > Well did any of you solve that problem ? > If you cud, can you plz post the solution here... > > Thanks > Krishna > > __________________________________________________ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.yahoo.com From hussain@ISI.EDU Mon Feb 18 07:20:55 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g1IFKtR11815 for ; Mon, 18 Feb 2002 07:20:55 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g1IFKoT24606; Mon, 18 Feb 2002 07:20:50 -0800 (PST) Date: Mon, 18 Feb 2002 07:20:50 -0800 (PST) From: Alefiya Hussain To: rohan bhindwale cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Project Queries In-Reply-To: <20020218072428.84772.qmail@web12302.mail.yahoo.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 17 Feb 2002, rohan bhindwale wrote: > A few queries about the project > > 1. The routers and the manager are to communicate > using TCP connection . Are we to hardcode the ports > that they communicate over. Ofcourse the manager will > have one port and the routers can also all communicate > via one port. Is there any other way we are to assign > these ports ?? There is a way to get ports assgined by the system dynamically. I suggest using that method. Refer to Stevens/Beej's Guide. > 2. The routers are to communicate with each other > using UPD . This info will be passed from the router > to the manager on the TCP connection and then the > manager will pass this info to that routers neighbour. > Again do we hardcode this port number in each router > ?? no. Assign them dynamically. If the port is unavailable, your program will fail. > 3. Stevens uses read and write to send data whereas > Beejs guide uses send and recv . which one is > preferred ?? You need to use sendto and recvfrom for UDP sockets. For TCP, is you use read and write make sure it read the number od bytes correctly. alefiya From johnh@ISI.EDU Mon Feb 18 11:09:08 2002 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 g1IJ98R27892 for ; Mon, 18 Feb 2002 11:09:08 -0800 (PST) Received: from dash.isi.edu (station-21.events.itd.umich.edu [141.211.252.150]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1IJ97O23635 for ; Mon, 18 Feb 2002 11:09:07 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g1IJ66517991; Mon, 18 Feb 2002 11:06:06 -0800 Message-Id: <200202181906.g1IJ66517991@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: rohan bhindwale Cc: usc_cs551@yahoogroups.com, csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Data Format In-reply-to: <20020218075448.1297.qmail@web12306.mail.yahoo.com> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Mon, 18 Feb 2002 11:06:06 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 17 Feb 2002 23:54:48 PST, rohan bhindwale wrote: >When we send an object onto the network do we have to >convert it into network byte order. No where in >Stevens do they do that . But they also use char >strings exclusively for sending data. > >Will it suffice if we cast a packet to (char * ) >before sending it ?? The format of most of the messages is completely up to you. The format of the LSA announcement is specified in the assignment, but it does not specify endianness, so you can do either. -John Heidemann From johnh@ISI.EDU Mon Feb 18 11:09:12 2002 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 g1IJ9CR27899 for ; Mon, 18 Feb 2002 11:09:12 -0800 (PST) Received: from dash.isi.edu (station-21.events.itd.umich.edu [141.211.252.150]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1IJ9BO23642 for ; Mon, 18 Feb 2002 11:09:11 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g1IJ6Bv18041; Mon, 18 Feb 2002 11:06:11 -0800 Message-Id: <200202181906.g1IJ6Bv18041@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: rohan bhindwale Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Project Queries In-reply-to: <20020218072428.84772.qmail@web12302.mail.yahoo.com> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Mon, 18 Feb 2002 11:06:11 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 17 Feb 2002 23:24:28 PST, rohan bhindwale wrote: >A few queries about the project > >1. The routers and the manager are to communicate >using TCP connection . Are we to hardcode the ports >that they communicate over. Ofcourse the manager will >have one port and the routers can also all communicate >via one port. Is there any other way we are to assign >these ports ?? How the master picks its port is up to it. It must pass this info to the routers. It can either use a static port or find its own. >2. The routers are to communicate with each other >using UPD . This info will be passed from the router >to the manager on the TCP connection and then the >manager will pass this info to that routers neighbour. >Again do we hardcode this port number in each router >?? It can either use a static port or find its own. You cannot assume that all the routers know each others ports a priori, though---they must figure this out based on the config info from the manager. > >3. Stevens uses read and write to send data whereas >Beejs guide uses send and recv . which one is >preferred ?? You should decide this, perhaps after reading the man pages. -John Heidemann From johnh@ISI.EDU Mon Feb 18 11:09:21 2002 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 g1IJ9LR27906 for ; Mon, 18 Feb 2002 11:09:21 -0800 (PST) Received: from dash.isi.edu (station-21.events.itd.umich.edu [141.211.252.150]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1IJ9KO23649 for ; Mon, 18 Feb 2002 11:09:20 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g1IJ6Mx18056 for ; Mon, 18 Feb 2002 11:06:22 -0800 Message-Id: <200202181906.g1IJ6Mx18056@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: csci551-talk@ISI.EDU MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Mon, 18 Feb 2002 11:06:22 -0800 From: John Heidemann Subject: [Csci551-talk] another update to project A Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: The spec of Project A has been updated in several important ways. The most important changes are: - a different (and hopefully easier to use) LSA message format - some additional requirements in the README In addition there are a number of clarifications. All changes are marked with "18-Feb-02". -John Heidemann From hussain@ISI.EDU Tue Feb 19 17:23:58 2002 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 g1K1NwR10927 for ; Tue, 19 Feb 2002 17:23:58 -0800 (PST) Received: from mat.isi.edu (mat.isi.edu [128.9.160.109]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1K1NvO20841 for ; Tue, 19 Feb 2002 17:23:57 -0800 (PST) Received: from localhost (hussain@localhost) by mat.isi.edu (8.11.6/8.11.6) with ESMTP id g1JAPiV03478 for ; Tue, 19 Feb 2002 05:25:44 -0500 Date: Tue, 19 Feb 2002 05:25:44 -0500 (EST) From: Alefiya Hussain To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Timer API code updated Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: FYI: We made a couple of changes to the timer API over to weekend. Please download a new version from http://www.isi.edu/~hussain/cs551_hints.html all the best, Alefiya From hussain@ISI.EDU Wed Feb 20 12:46:26 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g1KKkQR05986 for ; Wed, 20 Feb 2002 12:46:26 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g1KKkPZ24202 for ; Wed, 20 Feb 2002 12:46:25 -0800 (PST) Date: Wed, 20 Feb 2002 12:46:25 -0800 (PST) From: Alefiya Hussain Reply-To: Alefiya Hussain To: csci551-talk@ISI.EDU In-Reply-To: <200202201725.g1KHP3o31339@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] CS551 Hints page Updated Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Please refer to the CS551 hints page for the following updates: - Timer API were updated on 02-19-02 and the following changes were made to the code: - Changed the TimerCallback class and moved the member variable p_ to the Timer1 and Timer2 class in test-app.h This mean that you will no longer need to modify and timer code to use the Timer API :-)) - rearranged timers.hh and documented the classes extensively. -removed some bugs in the TestApp::start function. The select loop now check for all timers that have expired and executes them. It also make sure that the select timeout value is set to the first timer in the event queue. - Added more test cases against which you can test your project - Added a more detailed description of the LSA message fields. all the best, Alefiya From johnh@ISI.EDU Fri Feb 22 15:33:39 2002 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 g1MNXdR29842 for ; Fri, 22 Feb 2002 15:33:39 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1MNXcO00025 for ; Fri, 22 Feb 2002 15:33:38 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g1MNUYt13706 for ; Fri, 22 Feb 2002 15:30:34 -0800 Message-Id: <200202222330.g1MNUYt13706@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Project Queries In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 22 Feb 2002 15:30:34 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Mon, 18 Feb 2002 11:06:11 PST, John Heidemann wrote: >On Sun, 17 Feb 2002 23:24:28 PST, rohan bhindwale wrote: >>A few queries about the project >> >>1. The routers and the manager are to communicate >>using TCP connection . Are we to hardcode the ports >>that they communicate over. Ofcourse the manager will >>have one port and the routers can also all communicate >>via one port. Is there any other way we are to assign >>these ports ?? > >How the master picks its port is up to it. >It must pass this info to the routers. >It can either use a static port or find its own. > >>2. The routers are to communicate with each other >>using UPD . This info will be passed from the router >>to the manager on the TCP connection and then the >>manager will pass this info to that routers neighbour. >>Again do we hardcode this port number in each router >>?? > >It can either use a static port or find its own. >You cannot assume that all the routers know each others ports >a priori, though---they must figure this out based on the config >info from the manager. Just to clarify one thing here, in the 2nd question, I assumed that "info" refers to port numbers. You should NOT send data (i.e., LSA messages) through the master. -John Heidemann From samdadiy@usc.edu Sun Feb 24 00:21:57 2002 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 g1O8LvR29363 for ; Sun, 24 Feb 2002 00:21:57 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1O8Lvt05089 for ; Sun, 24 Feb 2002 00:21:57 -0800 (PST) Received: from sal-sun023.usc.edu (sal-sun023.usc.edu [128.125.5.93]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id AAA08944 for ; Sun, 24 Feb 2002 00:22:37 -0800 (PST) Received: from localhost (samdadiy@localhost) by sal-sun023.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1O8MZC06018 for ; Sun, 24 Feb 2002 00:22:35 -0800 (PST) Date: Sun, 24 Feb 2002 00:22:35 -0800 (PST) From: Parag Samdadiya To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] LSA format Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, I have a basic question. Why does the LSA message format deals with router id's of 16 bits ?? ( e.g. original router id, last-hop id ) What is the significance of this id ?? Is this the number assigned to the router ?? Dont we have to use ip address ( 32 bits ) to identify router information, like origin router id, last hop router etc. May be I am missing something. If someone could please clarify on the fields of LSA message and their significance it would be great. Thanks, Parag Samdadiya From rishisha@usc.edu Sun Feb 24 12:34:33 2002 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 g1OKYXR09893 for ; Sun, 24 Feb 2002 12:34:33 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1OKYXt14215 for ; Sun, 24 Feb 2002 12:34:33 -0800 (PST) Received: from aludra.usc.edu (rishisha@aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id MAA17299; Sun, 24 Feb 2002 12:35:13 -0800 (PST) Received: from localhost (rishisha@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1OKZE517550; Sun, 24 Feb 2002 12:35:14 -0800 (PST) Date: Sun, 24 Feb 2002 12:35:14 -0800 (PST) From: rishisha To: Parag Samdadiya cc: Subject: Re: [Csci551-talk] LSA format In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, You can check the hyperlink to the LSA message format,given on www.isi.edu/~hussain/cs551_hints.html. Rishi On Sun, 24 Feb 2002, Parag Samdadiya wrote: > > Hi, > > I have a basic question. > > Why does the LSA message format deals with router id's of 16 bits ?? > ( e.g. original router id, last-hop id ) What is the significance > of this id ?? Is this the number assigned to the router ?? > > Dont we have to use ip address ( 32 bits ) to identify router > information, like origin router id, last hop router etc. > > May be I am missing something. If someone could please > clarify on the fields of LSA message and their significance > it would be great. > > Thanks, > > Parag Samdadiya > From samdadiy@usc.edu Sun Feb 24 15:31:09 2002 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 g1ONV9R04857 for ; Sun, 24 Feb 2002 15:31:09 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1ONV8t10135 for ; Sun, 24 Feb 2002 15:31:08 -0800 (PST) Received: from sal-sun023.usc.edu (sal-sun023.usc.edu [128.125.5.93]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id PAA21624; Sun, 24 Feb 2002 15:31:49 -0800 (PST) Received: from localhost (samdadiy@localhost) by sal-sun023.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1ONVl510749; Sun, 24 Feb 2002 15:31:47 -0800 (PST) Date: Sun, 24 Feb 2002 15:31:46 -0800 (PST) From: Parag Samdadiya To: rishisha cc: Subject: Re: [Csci551-talk] LSA format In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Thanks, But my question is if we identify routers by id (16bits) then why do we have the 32 bits IP addresses assigned to them ??? Parag On Sun, 24 Feb 2002, rishisha wrote: > Hi, > You can check the hyperlink to the LSA message format,given on > www.isi.edu/~hussain/cs551_hints.html. > Rishi > > On Sun, 24 Feb 2002, Parag Samdadiya wrote: > > > > > Hi, > > > > I have a basic question. > > > > Why does the LSA message format deals with router id's of 16 bits ?? > > ( e.g. original router id, last-hop id ) What is the significance > > of this id ?? Is this the number assigned to the router ?? > > > > Dont we have to use ip address ( 32 bits ) to identify router > > information, like origin router id, last hop router etc. > > > > May be I am missing something. If someone could please > > clarify on the fields of LSA message and their significance > > it would be great. > > > > Thanks, > > > > Parag Samdadiya > > > From anshulpa@usc.edu Sun Feb 24 16:09:36 2002 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 g1P09aR11244 for ; Sun, 24 Feb 2002 16:09:36 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1P09Zt16728 for ; Sun, 24 Feb 2002 16:09:35 -0800 (PST) Received: from koh-sun003.usc.edu (koh-sun003.usc.edu [128.125.5.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id QAA04078; Sun, 24 Feb 2002 16:10:16 -0800 (PST) Received: from localhost (anshulpa@localhost) by koh-sun003.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1P09wx06867; Sun, 24 Feb 2002 16:09:58 -0800 (PST) Date: Sun, 24 Feb 2002 16:09:58 -0800 (PST) From: anshulpa To: Parag Samdadiya cc: rishisha , Subject: Re: [Csci551-talk] LSA format In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: In OSPF the routing is done on the basis of router id and IP addresses are used to communicate with other routers. Router id serves as identification parameter and IP address as communicatiing parameter. The IP address is only used when real implementation is done. For communication in our simulation we are going to use only UDP ports and same loopback IP address. Anshul On Sun, 24 Feb 2002, Parag Samdadiya wrote: > > Thanks, > But my question is if we identify routers by id (16bits) then why do > we have the 32 bits IP addresses assigned to them ??? > > Parag > > On Sun, 24 Feb 2002, rishisha wrote: > > > Hi, > > You can check the hyperlink to the LSA message format,given on > > www.isi.edu/~hussain/cs551_hints.html. > > Rishi > > > > On Sun, 24 Feb 2002, Parag Samdadiya wrote: > > > > > > > > Hi, > > > > > > I have a basic question. > > > > > > Why does the LSA message format deals with router id's of 16 bits ?? > > > ( e.g. original router id, last-hop id ) What is the significance > > > of this id ?? Is this the number assigned to the router ?? > > > > > > Dont we have to use ip address ( 32 bits ) to identify router > > > information, like origin router id, last hop router etc. > > > > > > May be I am missing something. If someone could please > > > clarify on the fields of LSA message and their significance > > > it would be great. > > > > > > Thanks, > > > > > > Parag Samdadiya > > > > > > From hussain@ISI.EDU Sun Feb 24 17:29:36 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g1P1TZR22602 for ; Sun, 24 Feb 2002 17:29:35 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g1P1TZ629018 for ; Sun, 24 Feb 2002 17:29:35 -0800 (PST) Date: Sun, 24 Feb 2002 17:29:35 -0800 (PST) From: Alefiya Hussain To: csci551-talk@ISI.EDU Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Hints page updated Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, I have updated the CS551 project hints page located at http://www.isi.edu/~hussain/cs551_hints.html - The Sample output for manager.conf in the project description has been modified significantly. - Added a FAQ section. all the best, Alefiya From johnh@ISI.EDU Mon Feb 25 20:51:28 2002 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 g1Q4pRR04890 for ; Mon, 25 Feb 2002 20:51:27 -0800 (PST) Received: from gull.prod.itd.earthlink.net (gull.mail.pas.earthlink.net [207.217.120.84]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1Q4pRt20402 for ; Mon, 25 Feb 2002 20:51:27 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by gull.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16fZaI-0000i8-00; Mon, 25 Feb 2002 20:51:26 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g1Q4m6E14338; Mon, 25 Feb 2002 20:48:06 -0800 Message-Id: <200202260448.g1Q4m6E14338@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: anshulpa Cc: Parag Samdadiya , rishisha , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] LSA format In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Mon, 25 Feb 2002 20:48:06 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 24 Feb 2002 16:09:58 PST, anshulpa wrote: > >In OSPF the routing is done on the basis of router id and IP addresses are >used to communicate with other routers. Router id serves as identification >parameter and IP address as communicatiing parameter. Good description of the difference. >On Sun, 24 Feb 2002, Parag Samdadiya wrote: >> Thanks, >> But my question is if we identify routers by id (16bits) then why do >> we have the 32 bits IP addresses assigned to them ??? The IP prefixes are used to identify the routes for which a router is reponsible. (Logically these represent something like a LAN attached to the router with a bunch of hosts hanging off of it.) -John Heidemann From hussain@ISI.EDU Mon Feb 25 23:08:58 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g1Q78wR01812 for ; Mon, 25 Feb 2002 23:08:58 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g1Q78wd22439 for ; Mon, 25 Feb 2002 23:08:58 -0800 (PST) Date: Mon, 25 Feb 2002 23:08:57 -0800 (PST) From: Alefiya Hussain To: csci551-talk@ISI.EDU Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Hints page updated Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, I have updated the CS551 hints page at http://www.isi.edu/~hussain/cs551_hints.html - Clarified the difference between number of links and number of routes in the LSA. - Added a list of approved code that can be used for the project. Please check the hints page periodically as i will regularly update the FAQ. all the best, Alefiya From mhr@usc.edu Wed Feb 27 15:41:01 2002 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 g1RNf1R02126 for ; Wed, 27 Feb 2002 15:41:01 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1RNf1t06384 for ; Wed, 27 Feb 2002 15:41:01 -0800 (PST) Received: from aludra.usc.edu (mhr@aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id PAA22710 for ; Wed, 27 Feb 2002 15:41:35 -0800 (PST) Received: from localhost (mhr@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1RNfe808207 for ; Wed, 27 Feb 2002 15:41:40 -0800 (PST) Date: Wed, 27 Feb 2002 15:41:39 -0800 (PST) From: mhr To: In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] ddd Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: I am using xxgdb inmy linux machine and i can debug the functions that are in main process but can not debug in child processes, It gives me transport error and comes out. anybody had the same problem. tanx -m From johnh@ISI.EDU Wed Feb 27 16:34:26 2002 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 g1S0YOR21942 for ; Wed, 27 Feb 2002 16:34:24 -0800 (PST) Received: from dash.isi.edu (dash.isi.edu [128.9.160.184]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1S0YNt29977 for ; Wed, 27 Feb 2002 16:34:23 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g1S0V8c15175; Wed, 27 Feb 2002 16:31:08 -0800 Message-Id: <200202280031.g1S0V8c15175@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: mhr Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] ddd In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Wed, 27 Feb 2002 16:31:08 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 27 Feb 2002 15:41:39 PST, mhr wrote: > >I am using xxgdb inmy linux machine and i can debug the functions >that are in main process but can not debug in child processes, >It gives me transport error and comes out. >anybody had the same problem. > >tanx >-m Look at the attach command in gdb. -John Heidemann From samdadiy@usc.edu Wed Feb 27 17:34:29 2002 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 g1S1YSR13536 for ; Wed, 27 Feb 2002 17:34:28 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1S1YSt27090 for ; Wed, 27 Feb 2002 17:34:28 -0800 (PST) Received: from aludra.usc.edu (samdadiy@aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id RAA21187; Wed, 27 Feb 2002 17:35:01 -0800 (PST) Received: from localhost (samdadiy@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1S1Z5V00537; Wed, 27 Feb 2002 17:35:05 -0800 (PST) Date: Wed, 27 Feb 2002 17:35:05 -0800 (PST) From: Parag Samdadiya To: , mhr Subject: Re: [Csci551-talk] ddd In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Setting break point at the first line of code of new thread of execution used to work in debugging multithreaded application (I tried it in CS-402) Not sure if same trick works when you fork new process. Parag On Wed, 27 Feb 2002, mhr wrote: > > I am using xxgdb inmy linux machine and i can debug the functions > that are in main process but can not debug in child processes, > It gives me transport error and comes out. > anybody had the same problem. > > tanx > -m > From kaiyuxia@usc.edu Wed Feb 27 19:36:06 2002 Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g1S3a6R21999 for ; Wed, 27 Feb 2002 19:36:06 -0800 (PST) Received: from aludra.usc.edu (kaiyuxia@aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id TAA01108; Wed, 27 Feb 2002 19:36:43 -0800 (PST) Received: from localhost (kaiyuxia@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1S3anv12277; Wed, 27 Feb 2002 19:36:49 -0800 (PST) Date: Wed, 27 Feb 2002 19:36:49 -0800 (PST) From: kaiyuxia To: cc: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] How to handle lost packet in Proj A? Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, Here is what is stated in the assignment: "The message must be sent via reliable flooding. The receiver should ACK (via UDP) the message and sequence number when it gets it. (Clarification 18-Feb-02: you may wish to use the type field to distinguish between ACKs and LSA announcements.) The sender must sent a timer and resend the message if it was not ACKed. For simplicity we'll assume 5 second fixed timers." My question is: When the sender times out, its (the sender) link state information may have been changed dramatically. ( Assume that the sender is able to get LSAs from other routers.) Should the sender resend the old message which has not been received by a particular router (but it is a stale message) ? Or should the sender just send the new LSA message? Could TA post some sample output with non-zeor loss rate ? (By the way, is "sent a timer" just a typo of "set a timer"? ) Thanks very much From mhr@usc.edu Wed Feb 27 19:40:23 2002 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 g1S3eNR22738 for ; Wed, 27 Feb 2002 19:40:23 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1S3eMt05863 for ; Wed, 27 Feb 2002 19:40:22 -0800 (PST) Received: from aludra.usc.edu (mhr@aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id TAA04429; Wed, 27 Feb 2002 19:41:00 -0800 (PST) Received: from localhost (mhr@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1S3f6h15609; Wed, 27 Feb 2002 19:41:06 -0800 (PST) Date: Wed, 27 Feb 2002 19:41:06 -0800 (PST) From: mhr To: Parag Samdadiya cc: Subject: Re: Re: [Csci551-talk] ddd In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: yes i think u do not have problem with multithreaded userspace programm but with multiprocess programs the main process does not have access to the child processes so u have to explicitly attach it i assume fro previous mail.I am trying to understand though. -m >Setting break point at the first line of code >of new thread of execution used to work in >debugging multithreaded application (I tried it in CS-402) >Not sure if same trick works when you fork new process. >Parag >>On Wed, 27 Feb 2002, mhr wrote: >> >> I am using xxgdb inmy linux machine and i can debug the functions >> that are in main process but can not debug in child processes, >> It gives me transport error and comes out. >> anybody had the same problem. >> >> tanx >> -m >> >> From ragarwal@usc.edu Wed Feb 27 20:01:35 2002 Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g1S41YR28173 for ; Wed, 27 Feb 2002 20:01:34 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id UAA18854; Wed, 27 Feb 2002 20:02:12 -0800 (PST) Received: from sal-sun025.usc.edu (sal-sun025.usc.edu [128.125.5.95]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g1S41Xc01724; Wed, 27 Feb 2002 20:01:33 -0800 (PST) Message-Id: <200202280401.g1S41Xc01724@scf-fs.usc.edu> Date: Wed, 27 Feb 2002 20:02:10 -0800 (PST) From: ragarwal Reply-To: ragarwal Subject: Re: [Csci551-talk] How to handle lost packet in Proj A? To: csci551-talk@mailman.isi.edu, kaiyuxia@usc.edu Cc: csci551@phakt.usc.edu MIME-Version: 1.0 Content-Type: TEXT/plain; charset=us-ascii Content-MD5: SPj03fOWLAjf9A8Tx1roWg== X-Mailer: dtmail 1.3.0 @(#)CDE Version 1.4.2 SunOS 5.8 sun4u sparc Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: I think that sender has no option other than to send the new LSA message. The reasoning is like this: when the sender's timer timeouts, its LS information would have changed. Now when sender prepares itself to resend the LSA, it has to dig into its databank to get the LSA, which is(the databank) is kept updated. If sender were to send the old LSA .. it would have to store it somewhere , which is a useless overhead. Rohit. >Hi, > >Here is what is stated in the assignment: >"The message must be sent via reliable flooding. The receiver should ACK >(via UDP) the message and sequence >number when it gets it. (Clarification 18-Feb-02: you may wish to use the >type field to distinguish between ACKs and LSA announcements.) The >sender must sent a timer and resend the message if it was not ACKed. For >simplicity we'll assume 5 second fixed timers." > >My question is: >When the sender times out, its (the sender) link state information may >have been changed dramatically. ( Assume that the sender is able to get >LSAs from other routers.) Should the sender resend the old message which >has not been received by a particular router (but it is a stale message) ? >Or should the sender just send the new LSA message? > >Could TA post some sample output with non-zeor loss rate ? > >(By the way, is "sent a timer" just a typo of "set a timer"? ) > >Thanks very much From johnh@ISI.EDU Wed Feb 27 21:14:52 2002 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 g1S5EpR12894 for ; Wed, 27 Feb 2002 21:14:51 -0800 (PST) Received: from harrier.prod.itd.earthlink.net (harrier.mail.pas.earthlink.net [207.217.120.12]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1S5Ept25450; Wed, 27 Feb 2002 21:14:51 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by harrier.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16gIu2-0004Iq-00; Wed, 27 Feb 2002 21:14:50 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g1S5BZm16961; Wed, 27 Feb 2002 21:11:35 -0800 Message-Id: <200202280511.g1S5BZm16961@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: kaiyuxia Cc: csci551-talk@ISI.EDU, Alefiya Hussain In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Wed, 27 Feb 2002 21:11:35 -0800 From: John Heidemann Subject: [Csci551-talk] Re: How to handle lost packet in Proj A? Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 27 Feb 2002 19:36:49 PST, kaiyuxia wrote: >Hi, > >Here is what is stated in the assignment: >"The message must be sent via reliable flooding. The receiver should ACK >(via UDP) the message and sequence >number when it gets it. (Clarification 18-Feb-02: you may wish to use the >type field to distinguish between ACKs and LSA announcements.) The >sender must sent a timer and resend the message if it was not ACKed. For >simplicity we'll assume 5 second fixed timers." > >My question is: >When the sender times out, its (the sender) link state information may >have been changed dramatically. ( Assume that the sender is able to get >LSAs from other routers.) Should the sender resend the old message which >has not been received by a particular router (but it is a stale message) ? >Or should the sender just send the new LSA message? What do you think? Is it important that the sender guarantee that every LSA message get there, or just the final one? I could see implementations where they all matter, but it's also possible (and actually probably preferable) that only the final one matters. >Could TA post some sample output with non-zeor loss rate ? It doesn't look that much different. Here's a portion: 1 10.1.0.0/16 1 0 --- 1 10.1.0.0/16 1 0 3 192.168.123.0/24 3 1 --- 1 10.1.0.0/16 1 0 2 10.2.0.0/16 2 1 3 192.168.123.0/24 3 1 --- 1 10.1.0.0/16 1 0 2 10.2.0.0/16 2 1 3 192.168.123.0/24 3 1 5 192.168.200.128/25 3 2 --- ... --- 1 10.1.0.0/16 1 0 2 10.2.0.0/16 2 1 3 192.168.123.0/24 3 1 4 192.168.101.0/24 2 2 5 192.168.200.128/25 3 2 --- messages-sent 60 messages-received 48 messages-dropped 13 (Not very exciting :-) > >(By the way, is "sent a timer" just a typo of "set a timer"? ) Yes, that's a typo and should be "set a timer". -John Heidemann From johnh@ISI.EDU Wed Feb 27 21:20:16 2002 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 g1S5KFR13575 for ; Wed, 27 Feb 2002 21:20:15 -0800 (PST) Received: from goose.prod.itd.earthlink.net (goose.mail.pas.earthlink.net [207.217.120.18]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g1S5KFt26230 for ; Wed, 27 Feb 2002 21:20:15 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by goose.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16gIzG-00028b-00; Wed, 27 Feb 2002 21:20:14 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g1S5Gx116998; Wed, 27 Feb 2002 21:16:59 -0800 Message-Id: <200202280516.g1S5Gx116998@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: mhr Cc: Parag Samdadiya , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] ddd In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Wed, 27 Feb 2002 21:16:59 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 27 Feb 2002 19:41:06 PST, mhr wrote: > >yes i think u do not have problem with multithreaded userspace >programm but with multiprocess programs the main process does >not have access to the child processes so u have to explicitly >attach it i assume fro previous mail.I am trying to understand >though. > >-m > >>Setting break point at the first line of code >>of new thread of execution used to work in >>debugging multithreaded application (I tried it in CS-402) > >>Not sure if same trick works when you fork new process. Processes and threads are very different... but this is CSci555 material :-). If you're using gdb, I recommend looking at the "Debugging programs with multiple processes" section of its manual. Unfortunately, I could not get it to trace into child processes for me (under Linux, YMMV on Solaris). -John Heidemann From anshulpa@usc.edu Thu Feb 28 03:26:32 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g1SBQVR24147 for ; Thu, 28 Feb 2002 03:26:31 -0800 (PST) Received: from koh-sun016.usc.edu (koh-sun016.usc.edu [128.125.5.124]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id DAA27335; Thu, 28 Feb 2002 03:27:09 -0800 (PST) Received: from localhost (anshulpa@localhost) by koh-sun016.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g1SBRAC02899; Thu, 28 Feb 2002 03:27:10 -0800 (PST) Date: Thu, 28 Feb 2002 03:27:09 -0800 (PST) From: anshulpa To: kaiyuxia cc: , Subject: Re: [Csci551-talk] How to handle lost packet in Proj A? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: I think so it does not make sense to transmit a stale packet. if a message is not received by the neighbour and it does not ack it then one has to resend the LSA. But if u send the old LSA which is of no use then u are waisting n/w resources. Anshul On Wed, 27 Feb 2002, kaiyuxia wrote: > Hi, > > Here is what is stated in the assignment: > "The message must be sent via reliable flooding. The receiver should ACK > (via UDP) the message and sequence > number when it gets it. (Clarification 18-Feb-02: you may wish to use the > type field to distinguish between ACKs and LSA announcements.) The > sender must sent a timer and resend the message if it was not ACKed. For > simplicity we'll assume 5 second fixed timers." > > My question is: > When the sender times out, its (the sender) link state information may > have been changed dramatically. ( Assume that the sender is able to get > LSAs from other routers.) Should the sender resend the old message which > has not been received by a particular router (but it is a stale message) ? > Or should the sender just send the new LSA message? > > Could TA post some sample output with non-zeor loss rate ? > > (By the way, is "sent a timer" just a typo of "set a timer"? ) > > Thanks very much > From srangwal@usc.edu Thu Feb 28 17:19:48 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g211JmR17783 for ; Thu, 28 Feb 2002 17:19:48 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id RAA10540; Thu, 28 Feb 2002 17:20:22 -0800 (PST) Received: from localhost (srangwal@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g211KQO20602; Thu, 28 Feb 2002 17:20:26 -0800 (PST) Date: Thu, 28 Feb 2002 17:20:26 -0800 (PST) From: Sumit Rangwala To: anshulpa cc: kaiyuxia , , Subject: Re: [Csci551-talk] How to handle lost packet in Proj A? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: If efficiency is the concern that why one need to send the LSA when the router finds that the information has changed. Whenever the link information of a router changes it will send a LSA for this new change. Now when it gets a time out and find that for the LSA which timed out a new LSA ( one with a greater seq. no.) has been sent, it just need to ignore it and take care that then next LSA ( with new information) reaches the neighbour reliably. Sumit On Thu, 28 Feb 2002, anshulpa wrote: > > I think so it does not make sense to transmit a stale packet. > > if a message is not received by the neighbour and it does not ack it then > one has to resend the LSA. But if u send the old LSA which is of no use > then u are waisting n/w resources. > > > Anshul > On Wed, 27 Feb 2002, kaiyuxia wrote: > > > Hi, > > > > Here is what is stated in the assignment: > > "The message must be sent via reliable flooding. The receiver should ACK > > (via UDP) the message and sequence > > number when it gets it. (Clarification 18-Feb-02: you may wish to use the > > type field to distinguish between ACKs and LSA announcements.) The > > sender must sent a timer and resend the message if it was not ACKed. For > > simplicity we'll assume 5 second fixed timers." > > > > My question is: > > When the sender times out, its (the sender) link state information may > > have been changed dramatically. ( Assume that the sender is able to get > > LSAs from other routers.) Should the sender resend the old message which > > has not been received by a particular router (but it is a stale message) ? > > Or should the sender just send the new LSA message? > > > > Could TA post some sample output with non-zeor loss rate ? > > > > (By the way, is "sent a timer" just a typo of "set a timer"? ) > > > > Thanks very much > > > From qcao@usc.edu Thu Feb 28 21:32:57 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g215WvR13519 for ; Thu, 28 Feb 2002 21:32:57 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id VAA17287 for ; Thu, 28 Feb 2002 21:33:35 -0800 (PST) Received: from localhost (qcao@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g215Xfc20583 for ; Thu, 28 Feb 2002 21:33:41 -0800 (PST) Date: Thu, 28 Feb 2002 21:33:41 -0800 (PST) From: qcao To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] about listen() and other questions Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hello everyone, I just joined this email-list so I don't know if you guys have talked about the project. Sorry if somebody asked these before. My questions are about the flow of manager: I think that what the manager do is first, open a socket and listening on routers, then go through the manager conf file to fork all the routers, but 1. I remember the listen() is not blocking, so it could communicate with one child router (listen, accepting, send) while still forking others? I don't understand why it could do this at the same time in the same process? 2. How long will listen() stays alive until timeout? Does the manager program just idles infinately until the routing table is built? 3. Because manager and routers all communicate use UNIX IPC, so they all use Socket file instead of IP and port pairs, right? But do we need to set a unique socket file for each communicate pair or only needed for each process? For example, for the manager TCP socket, only one socket file is enough? 4. Is anybody out there almost finish the project? I will appreciate it very much if you could give some idea of how many hours did you spend on it? I have a very tight schedule due to other exams and projects. :( Thanks very much for reading, -Ellen From hussain@ISI.EDU Fri Mar 1 01:09:01 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g21990R25311 for ; Fri, 1 Mar 2002 01:09:00 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g21990Y24648 for ; Fri, 1 Mar 2002 01:09:00 -0800 (PST) Date: Fri, 1 Mar 2002 01:09:00 -0800 (PST) From: Alefiya Hussain To: csci551-talk@ISI.EDU Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] project questions (pls forward to class group) Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: ---------- Forwarded message ---------- Date: Thu, 28 Feb 2002 23:44:25 -0800 (PST) From: qcao To: csci551@phakt.usc.edu Subject: project questions (pls forward to class group) Sorry dear TA, I have problem access the emaillist, could you help me forward this email to the class? Thanks a lot. Hello everyone, I just joined this email-list so I don't know if you guys have talked about the project. Sorry if somebody asked these before. My questions are about the flow of manager: I think that what the manager do is first, open a socket and listening on routers, then go through the manager conf file to fork all the routers, but 1. I remember the listen() is not blocking, so it could communicate with one child router (listen, accepting, send) while still forking others? I don't understand why it could do this at the same time in the same process? 2. How long will listen() stays alive until timeout? Does the manager program just idles infinately until the routing table is built? 3. Because manager and routers all communicate use UNIX IPC, so they all use Socket file instead of IP and port pairs, right? But do we need to set a unique socket file for each communicate pair or only needed for each process? For example, for the manager TCP socket, only one socket file is enough? 4. The handout mentioned we need to send the UDP ports of each routers' neighbors to them. But I could not find how we should generate them in the handout? Do we let manager generate them randomly? 5. Is anybody out there almost finish the project? I will appreciate it very much if you could give some idea of how many hours did you spend on it? I have a very tight schedule due to other exams and projects. :( Thanks very much for reading, -Ellen From amilinen@usc.edu Fri Mar 1 06:21:08 2002 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 g21EL8R26180 for ; Fri, 1 Mar 2002 06:21:08 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g21EL7t15263 for ; Fri, 1 Mar 2002 06:21:07 -0800 (PST) Received: from sal-sun045.usc.edu (sal-sun045.usc.edu [128.125.5.144]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id GAA11830 for ; Fri, 1 Mar 2002 06:21:44 -0800 (PST) Received: from localhost (amilinen@localhost) by sal-sun045.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g21ELgO29179 for ; Fri, 1 Mar 2002 06:21:42 -0800 (PST) Date: Fri, 1 Mar 2002 06:21:42 -0800 (PST) From: Jagadish Amilineni To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Connection Error Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, When I am trying to 'connect' the router processes to manager process, I am getting the following error: connect: Cannot assign requested address I am passing the same address structure of the manager process to each router process. Could someone clarify me on this. Thank You, Jagadish From alsaeed@usc.edu Fri Mar 1 12:55:20 2002 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 g21KtKR03000 for ; Fri, 1 Mar 2002 12:55:20 -0800 (PST) Received: from lsmls02.we.mediaone.net (lsmls02.we.mediaone.net [24.130.1.15]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g21KtJt28523 for ; Fri, 1 Mar 2002 12:55:19 -0800 (PST) Received: from mss (24-130-200-13.we.client2.attbi.com [24.130.200.13]) by lsmls02.we.mediaone.net (8.11.4/8.11.3) with SMTP id g21KrJm16935 for ; Fri, 1 Mar 2002 12:53:19 -0800 (PST) Message-ID: <003301c1c163$ae8242e0$0dc88218@we.mediaone.net> From: "MSS" To: Date: Fri, 1 Mar 2002 12:57:21 -0800 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0030_01C1C120.A02CA840" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Subject: [Csci551-talk] way to go about reliable flooding Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: This is a multi-part message in MIME format. ------=_NextPart_000_0030_01C1C120.A02CA840 Content-Type: text/plain; charset="windows-1256" Content-Transfer-Encoding: quoted-printable Hi! There are many ways to do reliable flooding. These come to my mind: 1- (hop-to-hop chain) When B receives an LSA-Ad from A, it immediately = Acks it and keeps the most recent LSA-Ad from A. It then floods it to = all its neighbors (but A) keeping timers and waiting for LSA-Acks from = those neighbors (and reflood it in case these timers expire): - requires storage of the most recent LSA-Ad from every node in = the system. - requires managing timers for every LSA-Ad received (at least = for those not Acked yet). - may reflood stale LSA-Ads. 2- (hop-to-hop stack) When B receives an LSA-Ad from A, it does NOT Ack = it immediately. Instead, it refloods it to all its neighbors (but A) and = wait for LSA-Acks from those neighbors. We have two options here: 2a- keep timers and reflood it when these timers expire: - suffers the same problems as in 1 above. - does extra work since A itself keeps a timer = waiting for an Ack from B. It will eventually timeout and retx. 2b- B does not keep timers; it just keeps a list of = not-yet-Acked LSA-Ad sequence #s from each node. When it accumulates enough Acks = for one of these LSA-Ads it Acks it to A. - seems to be the right way to go. - still has to keep LSA-Ad sequence #s from each = node. Another thing is that an origin router sending an LSA-Ad does not really = have to keep timers for old LSA-Ads it previously generated since it = will send the most recent LSA-Ad in the event it times out waiting for = an Ack. This changes option 2b; we have now to keep only the most recent = sequence # from each node (which we have to do anyways to see if an = LSA-Ad is new or not). Is option 2b with timers for the most recent LSA-Ad generated the right = way to go? if not, what could be a better way? Sorry for the long Q. Have a great weekend. -Alsaeed ------=_NextPart_000_0030_01C1C120.A02CA840 Content-Type: text/html; charset="windows-1256" Content-Transfer-Encoding: quoted-printable
Hi!
 
There are many ways to do reliable flooding. These = come to my=20 mind:
 
1- (hop-to-hop chain) When B receives an LSA-Ad from = A, it=20 immediately Acks it and keeps the most recent LSA-Ad from A. = It then=20 floods it to all its neighbors (but A) keeping timers and waiting for = LSA-Acks=20 from those neighbors (and reflood it in case these timers = expire):
        - requires = storage of=20 the most recent LSA-Ad from every node in the system.
        - requires = managing=20 timers for every LSA-Ad received (at least for those not Acked=20 yet).
        - may reflood = stale=20 LSA-Ads.
 
2- (hop-to-hop stack) When B receives an LSA-Ad from = A, it=20 does NOT Ack it immediately. Instead, it refloods it to all its = neighbors (but=20 A) and wait for LSA-Acks from those neighbors. We have two options=20 here:
 
        2a- keep = timers and=20 reflood it when these timers expire:
        =    =20         - suffers the same problems as in = 1=20 above.
        =    =20         - does extra work since A itself = keeps a=20 timer waiting for an Ack
          &nbs= p;           from = B. It will eventually timeout and retx.
 
        2b- B does not = keep=20 timers; it just keeps a list of not-yet-Acked LSA-Ad
        =    =20  sequence #s from each node. When it accumulates enough Acks=20 for
        =    =20  one of these LSA-Ads it Acks it to A.
        =    =20         - seems to be the right way to=20 go.
        =    =20         - still has to keep LSA-Ad = sequence #s=20 from each node.
 
 
Another thing is that an origin router sending an = LSA-Ad does=20 not really have to keep timers for old LSA-Ads it previously generated = since it=20 will send the most recent LSA-Ad in the event it times out waiting for = an Ack.=20 This changes option 2b; we have now to keep only the most recent = sequence # from=20 each node (which we have to do anyways to see if an LSA-Ad is new or=20 not).
 
Is option 2b with timers for the most recent LSA-Ad = generated=20 the right way to go? if not, what could be a better way?
 
Sorry for the long Q. Have a great = weekend.
 
-Alsaeed
------=_NextPart_000_0030_01C1C120.A02CA840-- From hussain@ISI.EDU Fri Mar 1 13:31:28 2002 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 g21LVSR17260 for ; Fri, 1 Mar 2002 13:31:28 -0800 (PST) Received: from mat.isi.edu (mat.isi.edu [128.9.160.109]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g21LVSt19464 for ; Fri, 1 Mar 2002 13:31:28 -0800 (PST) Received: from localhost (hussain@localhost) by mat.isi.edu (8.11.6/8.11.6) with ESMTP id g21LVRr01517; Fri, 1 Mar 2002 13:31:27 -0800 Date: Fri, 1 Mar 2002 13:31:27 -0800 (PST) From: Alefiya Hussain To: Jagadish Amilineni cc: Subject: Re: [Csci551-talk] Connection Error In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: The port is most probably already used. Use dynamic assignment by the OS for ports, or use the setsockopt to allow reuse of ports. alefiya On Fri, 1 Mar 2002, Jagadish Amilineni wrote: > Hi, > > When I am trying to 'connect' the router processes to manager process, > I am getting the following error: > > connect: Cannot assign requested address > > I am passing the same address structure of the manager process to each > router process. > > Could someone clarify me on this. > > Thank You, > > Jagadish > From hussain@ISI.EDU Fri Mar 1 13:36:51 2002 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 g21LapR19801 for ; Fri, 1 Mar 2002 13:36:51 -0800 (PST) Received: from mat.isi.edu (mat.isi.edu [128.9.160.109]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g21Lapt21792 for ; Fri, 1 Mar 2002 13:36:51 -0800 (PST) Received: from localhost (hussain@localhost) by mat.isi.edu (8.11.6/8.11.6) with ESMTP id g21LaoX01548; Fri, 1 Mar 2002 13:36:50 -0800 Date: Fri, 1 Mar 2002 13:36:50 -0800 (PST) From: Alefiya Hussain To: MSS cc: Subject: Re: [Csci551-talk] way to go about reliable flooding In-Reply-To: <003301c1c163$ae8242e0$0dc88218@we.mediaone.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: You need to keep a timer for each message you send. Hence this statement "does extra work since A itself keeps a timer waiting for an Ack" is incorrect. This is also true for the hop-to-hop chain. When you move on the the packet loss stage, keeping timers is even more crucial for every packet, since any packet can be lost. Please use the Timer API provided on the hints page or develope your own method to mentain timers correctly. alefiya On Fri, 1 Mar 2002, MSS wrote: > Hi! > > There are many ways to do reliable flooding. These come to my mind: > > 1- (hop-to-hop chain) When B receives an LSA-Ad from A, it immediately Acks it and keeps the most recent LSA-Ad from A. It then floods it to all its neighbors (but A) keeping timers and waiting for LSA-Acks from those neighbors (and reflood it in case these timers expire): > - requires storage of the most recent LSA-Ad from every node in the system. > - requires managing timers for every LSA-Ad received (at least for those not Acked yet). > - may reflood stale LSA-Ads. > 2- (hop-to-hop stack) When B receives an LSA-Ad from A, it does NOT Ack it immediately. Instead, it refloods it to all its neighbors (but A) and wait for LSA-Acks from those neighbors. We have two options here: > > 2a- keep timers and reflood it when these timers expire: > - suffers the same problems as in 1 above. > - does extra work since A itself keeps a timer waiting for an Ack > from B. It will eventually timeout and retx. > > 2b- B does not keep timers; it just keeps a list of not-yet-Acked LSA-Ad > sequence #s from each node. When it accumulates enough Acks for > one of these LSA-Ads it Acks it to A. > - seems to be the right way to go. > - still has to keep LSA-Ad sequence #s from each node. > > > Another thing is that an origin router sending an LSA-Ad does not really have to keep timers for old LSA-Ads it previously generated since it will send the most recent LSA-Ad in the event it times out waiting for an Ack. This changes option 2b; we have now to keep only the most recent sequence # from each node (which we have to do anyways to see if an LSA-Ad is new or not). > > Is option 2b with timers for the most recent LSA-Ad generated the right way to go? if not, what could be a better way? > > Sorry for the long Q. Have a great weekend. > > -Alsaeed > From alsaeed@usc.edu Fri Mar 1 15:54:17 2002 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 g21NsFR11520 for ; Fri, 1 Mar 2002 15:54:15 -0800 (PST) Received: from lsmls01.we.mediaone.net (lsmls01.we.mediaone.net [24.130.1.20]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g21NsEt06712 for ; Fri, 1 Mar 2002 15:54:14 -0800 (PST) Received: from mobilemoe (24-130-200-13.we.client2.attbi.com [24.130.200.13]) by lsmls01.we.mediaone.net (8.11.4/8.11.3) with SMTP id g21NqFI17699 for ; Fri, 1 Mar 2002 15:52:15 -0800 (PST) Message-ID: <002101c1c17c$54ba8560$e400a8c0@mobilemoe> From: "M. AlSaeed" To: References: Subject: Re: [Csci551-talk] way to go about reliable flooding Date: Fri, 1 Mar 2002 15:53:48 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Does this mean that each router is going to keep timers for all LSAs? those it generates and those it forwards for other routers? That's a lot of timers, may grow into thousands of timers with a moderate size network and a high loss rate (and our 5-seconds fixed timeouts). All I'm saying is that it does not make any sense to keep timers for the LSAa that others generate since they do that themselves. And it also does not make sense to keep a timer for my LSA # n when my LSA # n+1 is already Acked by all my neighbors (and their neighbors) or my LSA # n is out of date and I already sent a new and I have a timer for it. regards, -alsaeed ----- Original Message ----- From: "Alefiya Hussain" To: "MSS" Cc: Sent: Friday, March 01, 2002 1:36 PM Subject: Re: [Csci551-talk] way to go about reliable flooding > > You need to keep a timer for each message you send. Hence this statement > "does extra work since A itself keeps a timer waiting for an Ack" is > incorrect. This is also true for the hop-to-hop chain. > > When you move on the the packet loss stage, keeping timers is even more > crucial for every packet, since any packet can be lost. Please use the > Timer API provided on the hints page or develope your own method to mentain timers correctly. > > alefiya > > On Fri, 1 Mar 2002, MSS wrote: > > > Hi! > > > > There are many ways to do reliable flooding. These come to my mind: > > > > 1- (hop-to-hop chain) When B receives an LSA-Ad from A, it immediately Acks it and keeps the most recent LSA-Ad from A. It then floods it to all its neighbors (but A) keeping timers and waiting for LSA-Acks from those neighbors (and reflood it in case these timers expire): > > - requires storage of the most recent LSA-Ad from every node in the system. > > - requires managing timers for every LSA-Ad received (at least for those not Acked yet). > > - may reflood stale LSA-Ads. > > > > 2- (hop-to-hop stack) When B receives an LSA-Ad from A, it does NOT Ack it immediately. Instead, it refloods it to all its neighbors (but A) and wait for LSA-Acks from those neighbors. We have two options here: > > > > 2a- keep timers and reflood it when these timers expire: > > - suffers the same problems as in 1 above. > > - does extra work since A itself keeps a timer waiting for an Ack > > from B. It will eventually timeout and retx. > > > > 2b- B does not keep timers; it just keeps a list of not-yet-Acked LSA-Ad > > sequence #s from each node. When it accumulates enough Acks for > > one of these LSA-Ads it Acks it to A. > > - seems to be the right way to go. > > - still has to keep LSA-Ad sequence #s from each node. > > > > > > Another thing is that an origin router sending an LSA-Ad does not really have to keep timers for old LSA-Ads it previously generated since it will send the most recent LSA-Ad in the event it times out waiting for an Ack. This changes option 2b; we have now to keep only the most recent sequence # from each node (which we have to do anyways to see if an LSA-Ad is new or not). > > > > Is option 2b with timers for the most recent LSA-Ad generated the right way to go? if not, what could be a better way? > > > > Sorry for the long Q. Have a great weekend. > > > > -Alsaeed > > > From qcao@usc.edu Fri Mar 1 16:13:46 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g220DkR19215 for ; Fri, 1 Mar 2002 16:13:46 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id QAA08889 for ; Fri, 1 Mar 2002 16:14:23 -0800 (PST) Received: from localhost (qcao@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g220ERW12436 for ; Fri, 1 Mar 2002 16:14:27 -0800 (PST) Date: Fri, 1 Mar 2002 16:14:27 -0800 (PST) From: qcao To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] UNIX Domain IPC? Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, all As far as I did in my previous network programming, the processes in the same machine use UNIX domain IPC, and all the communications are done by socket file instead of IP&PORT, that is: s=socket(AF_UNIX, SOCK_STREAM,0); name.sun_family=AF_UNIX; strcpy(name.sun_path, "Socket"); bind(..)... But it seems to me in this project we need to use tcp and udp port to do communication. But how we could use INTERNET DOMAIN here? Which IP address should we use? Any help is greatly appreciated. -Qun From ganglu@usc.edu Fri Mar 1 16:15:18 2002 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 g220FIR19457 for ; Fri, 1 Mar 2002 16:15:18 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g220FHt19170 for ; Fri, 1 Mar 2002 16:15:17 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id QAA10355; Fri, 1 Mar 2002 16:15:54 -0800 (PST) Received: from godavari (eebtux.usc.edu [128.125.111.218]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g220FGl04713; Fri, 1 Mar 2002 16:15:16 -0800 (PST) Message-ID: <003a01c1c17f$6a466450$032aa8c0@usc.edu> From: "Gang Lu" To: "M. AlSaeed" Cc: References: <002101c1c17c$54ba8560$e400a8c0@mobilemoe> Subject: Re: [Csci551-talk] way to go about reliable flooding Date: Fri, 1 Mar 2002 16:15:46 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: ----- Original Message ----- From: "M. AlSaeed" To: Sent: Friday, March 01, 2002 3:53 PM Subject: Re: [Csci551-talk] way to go about reliable flooding > Does this mean that each router is going to keep timers for all LSAs? those > it generates and those it forwards for other routers? > That's a lot of timers, may grow into thousands of timers with a moderate > size network and a high loss rate (and our 5-seconds fixed timeouts). > > All I'm saying is that it does not make any sense to keep timers for the > LSAa that others generate since they do that themselves. And it also does But how can the origina router guarantee its LSA is received by all routers in the network? It can only use timer to make sure its neighbors receive the LSA. But for reliable flooding, its neighbor's neighbors must also receive this LSA. > not make sense to keep a timer for my LSA # n when my LSA # n+1 is already > Acked by all my neighbors (and their neighbors) or my LSA # n is out of date > and I already sent a new and I have a timer for it. I think so. If a new LSA#n+1 is generated or a new LSA#n+1 is received, the timers for the old LSA#n should be cleared. Or when the timer is out, if the sqn is older, then do not restranmit. Gang Lu > > regards, > > -alsaeed > > ----- Original Message ----- > From: "Alefiya Hussain" > To: "MSS" > Cc: > Sent: Friday, March 01, 2002 1:36 PM > Subject: Re: [Csci551-talk] way to go about reliable flooding > > > > > > You need to keep a timer for each message you send. Hence this statement > > "does extra work since A itself keeps a timer waiting for an Ack" is > > incorrect. This is also true for the hop-to-hop chain. > > > > When you move on the the packet loss stage, keeping timers is even more > > crucial for every packet, since any packet can be lost. Please use the > > Timer API provided on the hints page or develope your own method to > mentain timers correctly. > > > > alefiya > > > > On Fri, 1 Mar 2002, MSS wrote: > > > > > Hi! > > > > > > There are many ways to do reliable flooding. These come to my mind: > > > > > > 1- (hop-to-hop chain) When B receives an LSA-Ad from A, it immediately > Acks it and keeps the most recent LSA-Ad from A. It then floods it to all > its neighbors (but A) keeping timers and waiting for LSA-Acks from those > neighbors (and reflood it in case these timers expire): > > > - requires storage of the most recent LSA-Ad from every node in > the system. > > > - requires managing timers for every LSA-Ad received (at least > for those not Acked yet). > > > - may reflood stale LSA-Ads. > > > > > > > 2- (hop-to-hop stack) When B receives an LSA-Ad from A, it does NOT Ack > it immediately. Instead, it refloods it to all its neighbors (but A) and > wait for LSA-Acks from those neighbors. We have two options here: > > > > > > 2a- keep timers and reflood it when these timers expire: > > > - suffers the same problems as in 1 above. > > > - does extra work since A itself keeps a timer > waiting for an Ack > > > from B. It will eventually timeout and retx. > > > > > > 2b- B does not keep timers; it just keeps a list of > not-yet-Acked LSA-Ad > > > sequence #s from each node. When it accumulates enough Acks > for > > > one of these LSA-Ads it Acks it to A. > > > - seems to be the right way to go. > > > - still has to keep LSA-Ad sequence #s from each > node. > > > > > > > > > Another thing is that an origin router sending an LSA-Ad does not really > have to keep timers for old LSA-Ads it previously generated since it will > send the most recent LSA-Ad in the event it times out waiting for an Ack. > This changes option 2b; we have now to keep only the most recent sequence # > from each node (which we have to do anyways to see if an LSA-Ad is new or > not). > > > > > > Is option 2b with timers for the most recent LSA-Ad generated the right > way to go? if not, what could be a better way? > > > > > > Sorry for the long Q. Have a great weekend. > > > > > > -Alsaeed > > > > > > From johnh@ISI.EDU Fri Mar 1 16:29:06 2002 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 g220T6R24734 for ; Fri, 1 Mar 2002 16:29:06 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g220T5t28248; Fri, 1 Mar 2002 16:29:05 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g220PlL03752; Fri, 1 Mar 2002 16:25:47 -0800 Message-Id: <200203020025.g220PlL03752@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Alefiya Hussain Cc: MSS , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] way to go about reliable flooding In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 01 Mar 2002 16:25:47 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Fri, 01 Mar 2002 13:36:50 PST, Alefiya Hussain wrote: > >You need to keep a timer for each message you send. Hence this statement >"does extra work since A itself keeps a timer waiting for an Ack" is >incorrect. This is also true for the hop-to-hop chain. > >When you move on the the packet loss stage, keeping timers is even more >crucial for every packet, since any packet can be lost. Please use the >Timer API provided on the hints page or develope your own method to mentain timers correctly. Also, I recommend against the stack approach since in very large networks it might take a long time for all your neighbors (and their neighbors, etc.) to ACK, so the source would (falsely) retransmit. Hmm...sounds like a form of congestion collapse. -John Heidemann > >alefiya > >On Fri, 1 Mar 2002, MSS wrote: > >> Hi! >> >> There are many ways to do reliable flooding. These come to my mind: >> >> 1- (hop-to-hop chain) When B receives an LSA-Ad from A, it immediately Acks it and keeps the most recent LSA-Ad from A. It then floods it to all its neighbors (but A) keeping timers and waiting for LSA-Acks from those neighbors (and reflood it in case these timers expire): >> - requires storage of the most recent LSA-Ad from every node in the system. >> - requires managing timers for every LSA-Ad received (at least for those not Acked yet). >> - may reflood stale LSA-Ads. > > >> 2- (hop-to-hop stack) When B receives an LSA-Ad from A, it does NOT Ack it immediately. Instead, it refloods it to all its neighbors (but A) and wait for LSA-Acks from those neighbors. We have two options here: >> >> 2a- keep timers and reflood it when these timers expire: >> - suffers the same problems as in 1 above. >> - does extra work since A itself keeps a timer waiting for an Ack >> from B. It will eventually timeout and retx. >> >> 2b- B does not keep timers; it just keeps a list of not-yet-Acked LSA-Ad >> sequence #s from each node. When it accumulates enough Acks for >> one of these LSA-Ads it Acks it to A. >> - seems to be the right way to go. >> - still has to keep LSA-Ad sequence #s from each node. >> >> >> Another thing is that an origin router sending an LSA-Ad does not really have to keep timers for old LSA-Ads it previously generated since it will send the most recent LSA-Ad in the event it times out waiting for an Ack. This changes option 2b; we have now to keep only the most recent sequence # from each node (which we have to do anyways to see if an LSA-Ad is new or not). >> >> Is option 2b with timers for the most recent LSA-Ad generated the right way to go? if not, what could be a better way? >> >> Sorry for the long Q. Have a great weekend. >> >> -Alsaeed >> From johnh@ISI.EDU Fri Mar 1 16:32:54 2002 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 g220WsR26037 for ; Fri, 1 Mar 2002 16:32:54 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g220Wrt02009 for ; Fri, 1 Mar 2002 16:32:53 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g220RPZ03769; Fri, 1 Mar 2002 16:27:25 -0800 Message-Id: <200203020027.g220RPZ03769@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: qcao Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] UNIX Domain IPC? In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 01 Mar 2002 16:27:25 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Fri, 01 Mar 2002 16:14:27 PST, qcao wrote: >Hi, all > >As far as I did in my previous network programming, the processes in the >same machine use UNIX domain IPC, and all the communications are done >by socket file instead of IP&PORT, that is: > >s=socket(AF_UNIX, SOCK_STREAM,0); >name.sun_family=AF_UNIX; >strcpy(name.sun_path, "Socket"); >bind(..)... > >But it seems to me in this project we need to use tcp and udp port to do >communication. But how we could use INTERNET DOMAIN here? Which IP address >should we use? > >Any help is greatly appreciated. That is correct, you need to use Internet sockets and not Unix domain IPC for this assignment. I suggest you consult Stevens for how to set up Internet sockets. -John Heidemann From johnh@ISI.EDU Fri Mar 1 16:34:07 2002 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 g220Y7R26068 for ; Fri, 1 Mar 2002 16:34:07 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g220Y7t02606 for ; Fri, 1 Mar 2002 16:34:07 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g220UnW03798; Fri, 1 Mar 2002 16:30:49 -0800 Message-Id: <200203020030.g220UnW03798@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: "M. AlSaeed" Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] way to go about reliable flooding In-reply-to: <002101c1c17c$54ba8560$e400a8c0@mobilemoe> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 01 Mar 2002 16:30:49 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Fri, 01 Mar 2002 15:53:48 PST, "M. AlSaeed" wrote: >Does this mean that each router is going to keep timers for all LSAs? those >it generates and those it forwards for other routers? >That's a lot of timers, may grow into thousands of timers with a moderate >size network and a high loss rate (and our 5-seconds fixed timeouts). > >All I'm saying is that it does not make any sense to keep timers for the >LSAa that others generate since they do that themselves. And it also does >not make sense to keep a timer for my LSA # n when my LSA # n+1 is already >Acked by all my neighbors (and their neighbors) or my LSA # n is out of date >and I already sent a new and I have a timer for it. If one LSA superceeds another and so there is no purpose in sending the original, than I agree you don't need to guarantee the original's transmission and so you don't need to keep a timer. However, if you only keep timers that you generate (and not forward), then how do guarantee that reliable flooding succeeds? For the purposes of the project you should not feel contstrained by the number of active timers, although in the real world it is an issue. (And there are techniques that allow a large number of concurrent timers for hosts such as web servers.) -John Heidemann > >regards, > >-alsaeed > >----- Original Message ----- >From: "Alefiya Hussain" >To: "MSS" >Cc: >Sent: Friday, March 01, 2002 1:36 PM >Subject: Re: [Csci551-talk] way to go about reliable flooding > > >> >> You need to keep a timer for each message you send. Hence this statement >> "does extra work since A itself keeps a timer waiting for an Ack" is >> incorrect. This is also true for the hop-to-hop chain. >> >> When you move on the the packet loss stage, keeping timers is even more >> crucial for every packet, since any packet can be lost. Please use the >> Timer API provided on the hints page or develope your own method to >mentain timers correctly. >> >> alefiya >> >> On Fri, 1 Mar 2002, MSS wrote: >> >> > Hi! >> > >> > There are many ways to do reliable flooding. These come to my mind: >> > >> > 1- (hop-to-hop chain) When B receives an LSA-Ad from A, it immediately >Acks it and keeps the most recent LSA-Ad from A. It then floods it to all >its neighbors (but A) keeping timers and waiting for LSA-Acks from those >neighbors (and reflood it in case these timers expire): >> > - requires storage of the most recent LSA-Ad from every node in >the system. >> > - requires managing timers for every LSA-Ad received (at least >for those not Acked yet). >> > - may reflood stale LSA-Ads. >> >> >> > 2- (hop-to-hop stack) When B receives an LSA-Ad from A, it does NOT Ack >it immediately. Instead, it refloods it to all its neighbors (but A) and >wait for LSA-Acks from those neighbors. We have two options here: >> > >> > 2a- keep timers and reflood it when these timers expire: >> > - suffers the same problems as in 1 above. >> > - does extra work since A itself keeps a timer >waiting for an Ack >> > from B. It will eventually timeout and retx. >> > >> > 2b- B does not keep timers; it just keeps a list of >not-yet-Acked LSA-Ad >> > sequence #s from each node. When it accumulates enough Acks >for >> > one of these LSA-Ads it Acks it to A. >> > - seems to be the right way to go. >> > - still has to keep LSA-Ad sequence #s from each >node. >> > >> > >> > Another thing is that an origin router sending an LSA-Ad does not really >have to keep timers for old LSA-Ads it previously generated since it will >send the most recent LSA-Ad in the event it times out waiting for an Ack. >This changes option 2b; we have now to keep only the most recent sequence # >from each node (which we have to do anyways to see if an LSA-Ad is new or >not). >> > >> > Is option 2b with timers for the most recent LSA-Ad generated the right >way to go? if not, what could be a better way? >> > >> > Sorry for the long Q. Have a great weekend. >> > >> > -Alsaeed >> > >> From rohanbhindwale@yahoo.com Fri Mar 1 18:26:49 2002 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 g222QnR00729 for ; Fri, 1 Mar 2002 18:26:49 -0800 (PST) Received: from web12304.mail.yahoo.com (web12304.mail.yahoo.com [216.136.173.102]) by tnt.isi.edu (8.11.6/8.11.2) with SMTP id g222Qmt28246 for ; Fri, 1 Mar 2002 18:26:48 -0800 (PST) Message-ID: <20020302022648.47182.qmail@web12304.mail.yahoo.com> Received: from [128.125.228.42] by web12304.mail.yahoo.com via HTTP; Fri, 01 Mar 2002 18:26:48 PST Date: Fri, 1 Mar 2002 18:26:48 -0800 (PST) From: rohan bhindwale To: csci551-talk@ISI.EDU In-Reply-To: <200203020030.g220UnW03798@dash.isi.edu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: [Csci551-talk] Link State Update Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: The project spec says that each router is to send a LSA " Whenever it receives a link-state update from a neighbor that causes a change in its own table " . I am unclear on what Link State Update changing your table means ? Does it mean that there is a change in the neighbour topology of that router ( a neighbour has been added or deleted ) or does it mean that some link some where in the network has been added or deleted . It seems illogical to send an LSA when your neighbour topology has not changed. So I assume that link Update means that a neighbour has been added or deleted . If that is the case then for our project each router will have to send LSA either when a neighbour is added or deleted. Handling the former case is easy but how will we know when a neighbour has been deleted ??? Do we have to provide the functionality that if X number of LSAs are not ACKed then assume that the neighbour is gone , create a new LSA and flood ?? Thanks , rohan __________________________________________________ Do You Yahoo!? Yahoo! Sports - sign up for Fantasy Baseball http://sports.yahoo.com From johnh@ISI.EDU Fri Mar 1 19:19:52 2002 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 g223JpR10388 for ; Fri, 1 Mar 2002 19:19:51 -0800 (PST) Received: from falcon.prod.itd.earthlink.net (falcon.mail.pas.earthlink.net [207.217.120.74]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g223Jpt09321 for ; Fri, 1 Mar 2002 19:19:51 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by falcon.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16h03r-0006yG-00; Fri, 01 Mar 2002 19:19:51 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g223GXF06412; Fri, 1 Mar 2002 19:16:33 -0800 Message-Id: <200203020316.g223GXF06412@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: rohan bhindwale Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Link State Update In-reply-to: <20020302022648.47182.qmail@web12304.mail.yahoo.com> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 01 Mar 2002 19:16:32 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Fri, 01 Mar 2002 18:26:48 PST, rohan bhindwale wrote: >The project spec says that each router is to send a >LSA " Whenever it receives a link-state update from a >neighbor that causes a change in its own table " . > >I am unclear on what Link State Update changing your >table means ? Does it mean that there is a change in >the neighbour topology of that router ( a neighbour >has been added or deleted ) or does it mean that some >link some where in the network has been added or >deleted . It seems illogical to send an LSA when your >neighbour topology has not changed. So I assume that >link Update means that a neighbour has been added or >deleted . You are correct that there is no need to send an LSA packet that is identical to the previous one; it will have no effect. >If that is the case then for our project each router >will have to send LSA either when a neighbour is added >or deleted. Handling the former case is easy but how >will we know when a neighbour has been deleted ??? Do >we have to provide the functionality that if X number >of LSAs are not ACKed then assume that the neighbour >is gone , create a new LSA and flood ?? I don't beleive the assignment says anything about taking down routers once tehy're up, so you don't need to handle that case (unless you want to :-). -John Heidemann From hussain@ISI.EDU Fri Mar 1 19:21:14 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g223LER10438 for ; Fri, 1 Mar 2002 19:21:14 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g223LDv05549; Fri, 1 Mar 2002 19:21:13 -0800 (PST) Date: Fri, 1 Mar 2002 19:21:13 -0800 (PST) From: Alefiya Hussain To: rohan bhindwale cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Link State Update In-Reply-To: <20020302022648.47182.qmail@web12304.mail.yahoo.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Link-state update is defined as when the state on a directly connected link changes. Assume all links are down unless you have explicit information indicating otherwise. ie: refering to the sample topology, 1 does not know it has a link to 2 or 3 when it starts. Only when it receives an LSA from 2 it registers 2 as a far-end router id. Similarly for 3. I hope that clarifies link-state update. alefiya On Fri, 1 Mar 2002, rohan bhindwale wrote: > The project spec says that each router is to send a > LSA " Whenever it receives a link-state update from a > neighbor that causes a change in its own table " . > > I am unclear on what Link State Update changing your > table means ? Does it mean that there is a change in > the neighbour topology of that router ( a neighbour > has been added or deleted ) or does it mean that some > link some where in the network has been added or > deleted . It seems illogical to send an LSA when your > neighbour topology has not changed. So I assume that > link Update means that a neighbour has been added or > deleted . > > If that is the case then for our project each router > will have to send LSA either when a neighbour is added > or deleted. Handling the former case is easy but how > will we know when a neighbour has been deleted ??? Do > we have to provide the functionality that if X number > of LSAs are not ACKed then assume that the neighbour > is gone , create a new LSA and flood ?? > > > Thanks , > > > rohan > > > > __________________________________________________ > Do You Yahoo!? > Yahoo! Sports - sign up for Fantasy Baseball > http://sports.yahoo.com > From ragarwal@usc.edu Fri Mar 1 20:37:35 2002 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 g224bZR24230 for ; Fri, 1 Mar 2002 20:37:35 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g224bZt26281 for ; Fri, 1 Mar 2002 20:37:35 -0800 (PST) Received: from sal-sun044.usc.edu (sal-sun044.usc.edu [128.125.5.143]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id UAA15722 for ; Fri, 1 Mar 2002 20:38:11 -0800 (PST) Received: from localhost (ragarwal@localhost) by sal-sun044.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g224c8h11004 for ; Fri, 1 Mar 2002 20:38:08 -0800 (PST) Date: Fri, 1 Mar 2002 20:38:08 -0800 (PST) From: ragarwal To: In-Reply-To: <200203020316.g223GXF06412@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] stage 1 output Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: In the stage 1 output, how does a router know the IP addresses of other routers? There is no field in the LSA which lists the IP addresses of its neighbours. There is this responsible routes field, but the routers in its own AS do not come under this. So in responsible routes field it can only list its own IP address (in stage 1 and 2, atleast). So suppose router 1 recieves LSA from 2, it knows now that 4 is a neighbour of 2. So it adds the line for 4 in its routing table, as hops=2 and nexthop = router 2. But it still doesnt know the IP address of 4!! since 2 doesnt have it in its LSA. it only has its own IP address !! 1 will only come to know of IP address of 4 after it has recieved a LSA from 4. I know I am missing something here, but what? rohit. From vivekkh@hotmail.com Sat Mar 2 10:02:27 2002 Received: from hotmail.com (f33.law3.hotmail.com [209.185.241.33]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g22I2RR22996 for ; Sat, 2 Mar 2002 10:02:27 -0800 (PST) Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Sat, 2 Mar 2002 10:02:22 -0800 Received: from 204.210.37.4 by lw3fd.law3.hotmail.msn.com with HTTP; Sat, 02 Mar 2002 18:02:20 GMT X-Originating-IP: [204.210.37.4] From: "vivek khanna" To: csci551-talk@mailman.isi.edu Date: Sat, 02 Mar 2002 18:02:20 Mime-Version: 1.0 Content-Type: text/plain; format=flowed Message-ID: X-OriginalArrivalTime: 02 Mar 2002 18:02:22.0104 (UTC) FILETIME=[66B02980:01C1C214] Subject: [Csci551-talk] Link State Routing implementation query Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hello, I am using recvfrom(...) system call in the router child process to wait for any LSA from neighbouring routers. Once this router gets a message from another router, it processes the existing message. In the meantime if another message comes for the same router, then it will be lost unless queued. Is this the only way of UDP implementation at router side or you guys know of a better way of implementation. Thnx Vivek _________________________________________________________________ Chat with friends online, try MSN Messenger: http://messenger.msn.com From vkhanna@qualcomm.com Sat Mar 2 18:19:35 2002 Received: from illyana.qualcomm.com (illyana.qualcomm.com [129.46.65.100]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g232JWR03815 for ; Sat, 2 Mar 2002 18:19:34 -0800 (PST) Received: from VKHANNA.qualcomm.com (vkhanna.qualcomm.com [129.46.21.96]) by illyana.qualcomm.com (8.12.1/8.12.1/1.0) with ESMTP id g232JT0c012011; Sat, 2 Mar 2002 18:19:29 -0800 (PST) Message-Id: <5.1.0.14.2.20020302162340.01f885b8@illyana> X-Sender: vkhanna@illyana X-Mailer: QUALCOMM Windows Eudora Version 5.1 Date: Sat, 02 Mar 2002 18:20:17 -0800 To: hussain@ISI.EDU From: Vivek Khanna Cc: csci551-talk@mailman.isi.edu, vivekkh@hotmail.com Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Subject: [Csci551-talk] Link State Routing implementation query Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hello, The writeup says to use select(..) system call in Routers process to listen to traffic from other routers. If yes, can you explain why? I think any child process does not know the socket id of the other child process [They only know the port number]. So how can one use select(..) system call. I think select(..) should be used here only in the child router process to indicate a time out or if some packet has been received. Kindly elaborate. Vivek From anshulpa@usc.edu Sat Mar 2 21:43:39 2002 Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g235hdR00857 for ; Sat, 2 Mar 2002 21:43:39 -0800 (PST) Received: from lvl-sun696.usc.edu (lvl-sun696.usc.edu [128.125.140.174]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id VAA18854; Sat, 2 Mar 2002 21:44:10 -0800 (PST) Received: from localhost (anshulpa@localhost) by lvl-sun696.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g235iEb00048; Sat, 2 Mar 2002 21:44:14 -0800 (PST) Date: Sat, 2 Mar 2002 21:44:14 -0800 (PST) From: anshulpa To: Vivek Khanna cc: , , Subject: Re: [Csci551-talk] Link State Routing implementation query In-Reply-To: <5.1.0.14.2.20020302162340.01f885b8@illyana> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: U are right in saying that select is neccessary for timeout or if some packet is received. I think so you are confusing listen with listen() call. listen in the project specification means the state when you are waiting to receive the message from other nodes. Anshul Pandey. On Sat, 2 Mar 2002, Vivek Khanna wrote: > Hello, > > The writeup says to use select(..) system call in Routers process to listen > to traffic from other routers. If yes, can you explain why? I think any > child process does not know the socket id of the other child process [They > only know the port number]. So how can one use select(..) system call. I > think select(..) should be used here only in the child router process to > indicate a time out or if some packet has been received. Kindly elaborate. > > Vivek > > From vivekkh@hotmail.com Sun Mar 3 09:53:57 2002 Received: from hotmail.com (f12.law3.hotmail.com [209.185.241.12]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g23HrvR27558 for ; Sun, 3 Mar 2002 09:53:57 -0800 (PST) Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Sun, 3 Mar 2002 09:53:50 -0800 Received: from 204.210.37.4 by lw3fd.law3.hotmail.msn.com with HTTP; Sun, 03 Mar 2002 17:53:50 GMT X-Originating-IP: [204.210.37.4] From: "vivek khanna" To: anshulpa@usc.edu, vkhanna@qualcomm.com Cc: hussain@ISI.EDU, csci551-talk@mailman.isi.edu Subject: Re: [Csci551-talk] Link State Routing implementation query Date: Sun, 03 Mar 2002 17:53:50 Mime-Version: 1.0 Content-Type: text/plain; format=flowed Message-ID: X-OriginalArrivalTime: 03 Mar 2002 17:53:50.0439 (UTC) FILETIME=[60200770:01C1C2DC] Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: I think listen(..) is only used in manager process for waiting for TCP type requests from the child processes i.e. routers. Are you suggesting that the child process which waits on a UDP port also does a listen(..). That is not possible. One further clarification I need is that do we need to implement our own queues with select(...) system call to queue multiple LSA's coming from its neighbours. Would appreciate an early reply. --Vivek >From: anshulpa >To: Vivek Khanna >CC: , , > >Subject: Re: [Csci551-talk] Link State Routing implementation query >Date: Sat, 2 Mar 2002 21:44:14 -0800 (PST) > >U are right in saying that select is neccessary for timeout or if some >packet is received. I think so you are confusing listen with listen() >call. listen in the project specification means the state when you are >waiting to receive the message from other nodes. >Anshul Pandey. > >On Sat, 2 Mar 2002, Vivek Khanna wrote: > > > Hello, > > > > The writeup says to use select(..) system call in Routers process to >listen > > to traffic from other routers. If yes, can you explain why? I think any > > child process does not know the socket id of the other child process >[They > > only know the port number]. So how can one use select(..) system call. I > > think select(..) should be used here only in the child router process to > > indicate a time out or if some packet has been received. Kindly >elaborate. > > > > Vivek > > > > > _________________________________________________________________ Chat with friends online, try MSN Messenger: http://messenger.msn.com From anshulpa@usc.edu Sun Mar 3 12:31:36 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g23KVaR21499 for ; Sun, 3 Mar 2002 12:31:36 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id MAA17073; Sun, 3 Mar 2002 12:30:25 -0800 (PST) Received: from localhost (anshulpa@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g23KUXj17504; Sun, 3 Mar 2002 12:30:33 -0800 (PST) Date: Sun, 3 Mar 2002 12:30:33 -0800 (PST) From: anshulpa To: vivek khanna cc: , , Subject: Re: [Csci551-talk] Link State Routing implementation query In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: No we don't need to implement our own queue for the requests. i think so UDp socket is going to buffer the multiple LSA's and give to the application one by one. Am i missing something here??? And we surely don't need to listen on udp sockets. Anshul. On Sun, 3 Mar 2002, vivek khanna wrote: > I think listen(..) is only used in manager process for waiting for TCP type > requests from the child processes i.e. routers. Are you suggesting that the > child process which waits on a UDP port also does a listen(..). That is not > possible. > > One further clarification I need is that do we need to implement our own > queues with select(...) system call to queue multiple LSA's coming from its > neighbours. > > Would appreciate an early reply. > > --Vivek > > >From: anshulpa > >To: Vivek Khanna > >CC: , , > > > >Subject: Re: [Csci551-talk] Link State Routing implementation query > >Date: Sat, 2 Mar 2002 21:44:14 -0800 (PST) > > > >U are right in saying that select is neccessary for timeout or if some > >packet is received. I think so you are confusing listen with listen() > >call. listen in the project specification means the state when you are > >waiting to receive the message from other nodes. > >Anshul Pandey. > > > >On Sat, 2 Mar 2002, Vivek Khanna wrote: > > > > > Hello, > > > > > > The writeup says to use select(..) system call in Routers process to > >listen > > > to traffic from other routers. If yes, can you explain why? I think any > > > child process does not know the socket id of the other child process > >[They > > > only know the port number]. So how can one use select(..) system call. I > > > think select(..) should be used here only in the child router process to > > > indicate a time out or if some packet has been received. Kindly > >elaborate. > > > > > > Vivek > > > > > > > > > > > > > _________________________________________________________________ > Chat with friends online, try MSN Messenger: http://messenger.msn.com > From kaiyuxia@usc.edu Sun Mar 3 13:35:09 2002 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 g23LZ9R29440 for ; Sun, 3 Mar 2002 13:35:09 -0800 (PST) Received: from usc-fs1.usc.edu (usc-fs1.usc.edu [128.125.150.4]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g23LZ8t25204 for ; Sun, 3 Mar 2002 13:35:08 -0800 (PST) Received: from lvl-sun701.usc.edu (lvl-sun701.usc.edu [128.125.140.179]) by usc-fs1.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g23LZif02018; Sun, 3 Mar 2002 13:35:44 -0800 (PST) Received: from localhost (kaiyuxia@localhost) by lvl-sun701.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g23LVZd10252; Sun, 3 Mar 2002 13:31:35 -0800 (PST) Date: Sun, 3 Mar 2002 13:31:35 -0800 (PST) From: kaiyuxia To: ragarwal cc: Subject: Could TA clarify this? Re: [Csci551-talk] stage 1 output In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: I am also stuck on this problem. It would be very appreciated if TA could clarify this ASAP. Regarding the LSA packet format, can we change the length of type field and prefix to 16 bits? This could save us some time to manage word alignment. Thanks and best regards, On Fri, 1 Mar 2002, ragarwal wrote: > > In the stage 1 output, how does a router know the IP addresses of other > routers? > There is no field in the LSA which lists the IP addresses of its > neighbours. There is this responsible routes field, but the routers in its > own AS do not come under this. So in responsible routes field it can only > list its own IP address (in stage 1 and 2, atleast). > > So suppose router 1 recieves LSA from 2, it knows now that 4 is a > neighbour of 2. So it adds the line for 4 in its routing table, as hops=2 > and nexthop = router 2. But it still doesnt know the IP address of 4!! > since 2 doesnt have it in its LSA. it only has its own IP address !! > > 1 will only come to know of IP address of 4 after it has recieved a LSA > from 4. > > I know I am missing something here, but what? > > rohit. > > > From hussain@ISI.EDU Sun Mar 3 19:23:27 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g243NRR21520 for ; Sun, 3 Mar 2002 19:23:27 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g243NQP14649; Sun, 3 Mar 2002 19:23:26 -0800 (PST) Date: Sun, 3 Mar 2002 19:23:26 -0800 (PST) From: Alefiya Hussain To: ragarwal cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] stage 1 output In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Fri, 1 Mar 2002, ragarwal wrote: > > In the stage 1 output, how does a router know the IP addresses of other > routers? > There is no field in the LSA which lists the IP addresses of its > neighbours. There is this responsible routes field, but the routers in its > own AS do not come under this. So in responsible routes field it can only > list its own IP address (in stage 1 and 2, atleast). Each router that generates a LSA will include the routes that it is responsible for. ie: its own IP address. Hence for s1-X.out you will know the IP addresses for all the routers in the network. > So suppose router 1 recieves LSA from 2, it knows now that 4 is a > neighbour of 2. So it adds the line for 4 in its routing table, as hops=2 > and nexthop = router 2. But it still doesnt know the IP address of 4!! > since 2 doesnt have it in its LSA. it only has its own IP address !! 1 will know 4's IP address when it receives an LSA from it. all the best alefiya From hussain@ISI.EDU Sun Mar 3 19:27:53 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g243RrR22258 for ; Sun, 3 Mar 2002 19:27:53 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g243RkE15245; Sun, 3 Mar 2002 19:27:46 -0800 (PST) Date: Sun, 3 Mar 2002 19:27:46 -0800 (PST) From: Alefiya Hussain To: anshulpa cc: vivek khanna , vkhanna@qualcomm.com, csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Link State Routing implementation query In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 3 Mar 2002, anshulpa wrote: > No we don't need to implement our own queue for the requests. i think so > UDp socket is going to buffer the multiple LSA's and give to the > application one by one. Am i missing something here??? yes, Anshul you are correct. Just use sendto and recvfrom to send/receive LSA on UDP socket. Listen(..) call DOES NOT apply to UDP socket. alefiya > And we surely don't need to listen on udp sockets. > > Anshul. > > On Sun, 3 Mar 2002, vivek khanna wrote: > > > I think listen(..) is only used in manager process for waiting for TCP type > > requests from the child processes i.e. routers. Are you suggesting that the > > child process which waits on a UDP port also does a listen(..). That is not > > possible. > > > > One further clarification I need is that do we need to implement our own > > queues with select(...) system call to queue multiple LSA's coming from its > > neighbours. > > > > Would appreciate an early reply. > > > > --Vivek > > > > >From: anshulpa > > >To: Vivek Khanna > > >CC: , , > > > > > >Subject: Re: [Csci551-talk] Link State Routing implementation query > > >Date: Sat, 2 Mar 2002 21:44:14 -0800 (PST) > > > > > >U are right in saying that select is neccessary for timeout or if some > > >packet is received. I think so you are confusing listen with listen() > > >call. listen in the project specification means the state when you are > > >waiting to receive the message from other nodes. > > >Anshul Pandey. > > > > > >On Sat, 2 Mar 2002, Vivek Khanna wrote: > > > > > > > Hello, > > > > > > > > The writeup says to use select(..) system call in Routers process to > > >listen > > > > to traffic from other routers. If yes, can you explain why? I think any > > > > child process does not know the socket id of the other child process > > >[They > > > > only know the port number]. So how can one use select(..) system call. I > > > > think select(..) should be used here only in the child router process to > > > > indicate a time out or if some packet has been received. Kindly > > >elaborate. > > > > > > > > Vivek > > > > > > > > > > > > > > > > > > > > > _________________________________________________________________ > > Chat with friends online, try MSN Messenger: http://messenger.msn.com > > > From hussain@ISI.EDU Sun Mar 3 19:30:04 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g243U3R22851 for ; Sun, 3 Mar 2002 19:30:03 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g243U3B15492; Sun, 3 Mar 2002 19:30:03 -0800 (PST) Date: Sun, 3 Mar 2002 19:30:03 -0800 (PST) From: Alefiya Hussain To: kaiyuxia cc: ragarwal , csci551-talk@ISI.EDU Subject: Re: Could TA clarify this? Re: [Csci551-talk] stage 1 output In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 3 Mar 2002, kaiyuxia wrote: > I am also stuck on this problem. It would be very appreciated if TA could > clarify this ASAP. I clarified this on the newsgroup on an earlier mail. > Regarding the LSA packet format, can we change the length of type field > and prefix to 16 bits? This could save us some time to manage word > alignment. Please adhere to the LSA field formats. Do not change their sizes. We will be verifying the format used based on the last message output. alefiya > On Fri, 1 Mar 2002, ragarwal wrote: > > > > > In the stage 1 output, how does a router know the IP addresses of other > > routers? > > There is no field in the LSA which lists the IP addresses of its > > neighbours. There is this responsible routes field, but the routers in its > > own AS do not come under this. So in responsible routes field it can only > > list its own IP address (in stage 1 and 2, atleast). > > > > So suppose router 1 recieves LSA from 2, it knows now that 4 is a > > neighbour of 2. So it adds the line for 4 in its routing table, as hops=2 > > and nexthop = router 2. But it still doesnt know the IP address of 4!! > > since 2 doesnt have it in its LSA. it only has its own IP address !! > > > > 1 will only come to know of IP address of 4 after it has recieved a LSA > > from 4. > > > > I know I am missing something here, but what? > > > > rohit. > > > > > > > From hussain@ISI.EDU Sun Mar 3 21:47:00 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g245l0R05165 for ; Sun, 3 Mar 2002 21:47:00 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g245l0j16651 for ; Sun, 3 Mar 2002 21:47:00 -0800 (PST) Date: Sun, 3 Mar 2002 21:47:00 -0800 (PST) From: Alefiya Hussain To: csci551-talk@ISI.EDU Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Hints Page Updated Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, I have updated the hints page and added the following: -C language binding to the Timer API. To run the test application using the C language binding type ./test-app-c I have also added the following project clarifications on the webpage. Project Clarifications: 1. If you extend the LSA with additional you must justify your extensions in the readme file. While you ARE allowed to do so, your justification must be sufficient to receive full credit. The current (18-Feb-02) LSA format can be used to implement the algorithms, so finding a "sufficient" justification may be difficult. 2. Please note that the LSA format must follow the exact format given in the assignment (plus with any additional fields as you document). Thus the type must take only 1 byte, sequence number 2 bytes, etc. An entire message with no routes and links should consume exactly 13 bytes. Each addition route adds 7 bytes, each link 2 bytes. Hint: I recommend you have a simple to manage data structure for LSAs in your program, and then you have *marshaling* and *unmarshaling* routines that encode and decode from this internal format to the official external format. all the best Alefiya From bhat@usc.edu Mon Mar 4 02:35:32 2002 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 g24AZWR21246 for ; Mon, 4 Mar 2002 02:35:32 -0800 (PST) Received: from usc-fs1.usc.edu (usc-fs1.usc.edu [128.125.150.4]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g24AZVt18375; Mon, 4 Mar 2002 02:35:31 -0800 (PST) Received: from sal-sun059.usc.edu (sal-sun059.usc.edu [128.125.5.158]) by usc-fs1.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g24Aa6f08051; Mon, 4 Mar 2002 02:36:06 -0800 (PST) Received: from localhost (bhat@localhost) by sal-sun059.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g24AVhh14951; Mon, 4 Mar 2002 02:31:43 -0800 (PST) Date: Mon, 4 Mar 2002 02:31:42 -0800 (PST) From: Gajanan Bhat To: Alefiya Hussain cc: In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Mistake in the sample output. Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, I was just comparing the output of my files for the second stage with the sample output given on the web site and I noticed that the next hop for router 2 in the file s1-4.out is given as 1, which is wrong. Thought I will let you know.. Thanx ------------------------------------------------------------------------------ Gajanan Bhat Masters in Electrical Engineering University of Southern California, Los Angeles ------------------------------------------------------------------------------ From ganglu@usc.edu Mon Mar 4 09:21:43 2002 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 g24HLgR26515 for ; Mon, 4 Mar 2002 09:21:42 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g24HLgt08622 for ; Mon, 4 Mar 2002 09:21:42 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id JAA29960 for ; Mon, 4 Mar 2002 09:22:17 -0800 (PST) Received: from godavari (eebtux.usc.edu [128.125.111.218]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g24HLel26656 for ; Mon, 4 Mar 2002 09:21:40 -0800 (PST) Message-ID: <009a01c1c3a1$225c4d70$032aa8c0@usc.edu> From: "Gang Lu" To: References: Subject: Re: [Csci551-talk] stage 1 output Date: Mon, 4 Mar 2002 09:22:12 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: I still have some questions. 1. The first LSA a router advertise is with a LinkNum=0. For example 1 receive 2's first LSA message, should this be forward to 1's neighbors? I think it should not, because 2's LSA is not correct: 2 should at least have a neighbor 1. 2. Suppose such a topology. 1------2-------3. 2 sent to 1 a LSA message, said 3 is its neighbor. But at that time, 1 have not received 3's LSA. So although 1 know it can reach 3 through 2, but it did not know 3's IP address. How can 1 then build its routing table? 3. I still need a little more clarification about the number of routes and number of links. The number of links is the number of direct connected neighbor that is up. right? In the website, it is said that: Number of routes = addresses of the routes that the origin router is responsible for. A router is NOT responsible for routes of other neighbors in its own AS.(although it may be responsible for routes in external ASes for which it is peering). My question is: Is it the sum of the number of up links and the number of external routes? thanks, Gang Lu ----- Original Message ----- From: "Alefiya Hussain" To: "ragarwal" Cc: Sent: Sunday, March 03, 2002 7:23 PM Subject: Re: [Csci551-talk] stage 1 output > > On Fri, 1 Mar 2002, ragarwal wrote: > > > > > In the stage 1 output, how does a router know the IP addresses of other > > routers? > > There is no field in the LSA which lists the IP addresses of its > > neighbours. There is this responsible routes field, but the routers in its > > own AS do not come under this. So in responsible routes field it can only > > list its own IP address (in stage 1 and 2, atleast). > > > Each router that generates a LSA will include the routes that it is > responsible for. ie: its own IP address. Hence for s1-X.out you will know > the IP addresses for all the routers in the network. > > > > > So suppose router 1 recieves LSA from 2, it knows now that 4 is a > > neighbour of 2. So it adds the line for 4 in its routing table, as hops=2 > > and nexthop = router 2. But it still doesnt know the IP address of 4!! > > since 2 doesnt have it in its LSA. it only has its own IP address !! > > 1 will know 4's IP address when it receives an LSA from it. > > all the best > alefiya > From hussain@ISI.EDU Mon Mar 4 11:59:08 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g24Jx8R06605 for ; Mon, 4 Mar 2002 11:59:08 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g24Jx7o10818; Mon, 4 Mar 2002 11:59:07 -0800 (PST) Date: Mon, 4 Mar 2002 11:59:07 -0800 (PST) From: Alefiya Hussain To: Gang Lu cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] stage 1 output In-Reply-To: <009a01c1c3a1$225c4d70$032aa8c0@usc.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Mon, 4 Mar 2002, Gang Lu wrote: > > I still have some questions. > > 1. The first LSA a router advertise is with a LinkNum=0. For example 1 > receive 2's first LSA message, > should this be forward to 1's neighbors? I think it should not, because 2's > LSA is not correct: 2 should > at least have a neighbor 1. Forward every LSA you receive to all your neighbors. Why do you want to add complexity to the project by doing selective forwarding ?? > 2. Suppose such a topology. > 1------2-------3. > > 2 sent to 1 a LSA message, said 3 is its neighbor. But at that time, 1 have > not received 3's LSA. > So although 1 know it can reach 3 through 2, but it did not know 3's IP > address. How can 1 > then build its routing table? Leave the entry empty and enter it when you receive 3's LSA. > 3. I still need a little more clarification about the number of routes and > number of links. > The number of links is the number of direct connected neighbor that is up. > right? yes. > In the website, it is said that: > Number of routes = addresses of the routes that the origin router is > responsible for. > A router is NOT responsible for routes of other neighbors in its own > AS.(although it may be responsible for routes in external ASes for which it > is peering). > My question is: Is it the sum of the number of up links and the number of > external routes? I am not sure i understand your question. What sum are we talking about. alefiya > From: "Alefiya Hussain" > To: "ragarwal" > Cc: > Sent: Sunday, March 03, 2002 7:23 PM > Subject: Re: [Csci551-talk] stage 1 output > > > > > > On Fri, 1 Mar 2002, ragarwal wrote: > > > > > > > > In the stage 1 output, how does a router know the IP addresses of other > > > routers? > > > There is no field in the LSA which lists the IP addresses of its > > > neighbours. There is this responsible routes field, but the routers in > its > > > own AS do not come under this. So in responsible routes field it can > only > > > list its own IP address (in stage 1 and 2, atleast). > > > > > > Each router that generates a LSA will include the routes that it is > > responsible for. ie: its own IP address. Hence for s1-X.out you will know > > the IP addresses for all the routers in the network. > > > > > > > > > So suppose router 1 recieves LSA from 2, it knows now that 4 is a > > > neighbour of 2. So it adds the line for 4 in its routing table, as > hops=2 > > > and nexthop = router 2. But it still doesnt know the IP address of 4!! > > > since 2 doesnt have it in its LSA. it only has its own IP address !! > > > > 1 will know 4's IP address when it receives an LSA from it. > > > > all the best > > alefiya > > > From samdadiy@usc.edu Mon Mar 4 13:12:24 2002 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 g24LCOR04170 for ; Mon, 4 Mar 2002 13:12:24 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g24LCNt01511 for ; Mon, 4 Mar 2002 13:12:23 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id NAA11906 for ; Mon, 4 Mar 2002 13:12:58 -0800 (PST) Received: from localhost (samdadiy@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g24LD8M19956 for ; Mon, 4 Mar 2002 13:13:08 -0800 (PST) Date: Mon, 4 Mar 2002 13:13:08 -0800 (PST) From: Parag Samdadiya To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Confusion in output files to generate Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Are we supposed to generate the stage 1 ( s1-X.out ) output files and stage2 (s2-X.out ) files simultaneously. One considers error rate and other does not. So do we do flooding twice based on two scenarios ???? OR is it that depending on error rate just generate one set of files ??? Also are stage 3 files ( s3-X.out) also generated simultanously ???? Or is it after we are done with s1-X.out or stage 1 ?? If TA could please clarify the sequence, then it would be great. Thanks, Parag From hussain@ISI.EDU Mon Mar 4 14:09:28 2002 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 g24M9RR00289 for ; Mon, 4 Mar 2002 14:09:27 -0800 (PST) Received: from mat.isi.edu (mat.isi.edu [128.9.160.109]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g24M9Rt01623 for ; Mon, 4 Mar 2002 14:09:27 -0800 (PST) Received: from localhost (hussain@localhost) by mat.isi.edu (8.11.6/8.11.6) with ESMTP id g24M9Pa27963; Mon, 4 Mar 2002 14:09:25 -0800 Date: Mon, 4 Mar 2002 14:09:25 -0800 (PST) From: Alefiya Hussain To: Parag Samdadiya cc: Subject: Re: [Csci551-talk] Confusion in output files to generate In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: s0.out will be created by the manager. The routers will create the following file: -First create s1-X.out based on how the routing converges. -Then recalculate the routes by executing the routing protocol with packet losses for stage 2. Create s2-X.out -Then assume the presence of external routers and use policy routing to create s3-X.out Do not create s1-X.out and s2-X.out simultaneously. The transient routing tables/ numbers of message sent / number of message received will be different for these two stage. alefiya On Mon, 4 Mar 2002, Parag Samdadiya wrote: > > > Are we supposed to generate the stage 1 ( s1-X.out ) output files > and stage2 (s2-X.out ) files simultaneously. > > One considers error rate and other does not. So do we do > flooding twice based on two scenarios ???? > > OR is it that depending on error rate just generate one set of files ??? > > Also are stage 3 files ( s3-X.out) also generated > simultanously ???? > Or is it after we are done with s1-X.out or stage 1 ?? > > If TA could please clarify the sequence, then it would be great. > > Thanks, > > Parag > From johnh@ISI.EDU Tue Mar 5 10:50:20 2002 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 g25IoGR13281 for ; Tue, 5 Mar 2002 10:50:16 -0800 (PST) Received: from dash.isi.edu (dash.isi.edu [128.9.160.184]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g25IoFt21490; Tue, 5 Mar 2002 10:50:15 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g25Ikor03843; Tue, 5 Mar 2002 10:46:50 -0800 Message-Id: <200203051846.g25Ikor03843@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Alefiya Hussain Cc: anshulpa , vivek khanna , vkhanna@qualcomm.com, csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Link State Routing implementation query In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Tue, 05 Mar 2002 10:46:48 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 03 Mar 2002 19:27:46 PST, Alefiya Hussain wrote: > > >On Sun, 3 Mar 2002, anshulpa wrote: > >> No we don't need to implement our own queue for the requests. i think so >> UDp socket is going to buffer the multiple LSA's and give to the >> application one by one. Am i missing something here??? > >yes, Anshul you are correct. Just use sendto and recvfrom to send/receive >LSA on UDP socket. Listen(..) call DOES NOT apply to UDP socket. And you are correct that there will be some buffering of UDP messages in the socket buffer. Therefore, to answer: >> > One further clarification I need is that do we need to implement our own >> > queues with select(...) system call to queue multiple LSA's coming from its >> > neighbours. You do not need to explicitly queue data yourself in the app. But to get back to Vivek's initial question: >> > >On Sat, 2 Mar 2002, Vivek Khanna wrote: >> > > >> > > > Hello, >> > > > >> > > > The writeup says to use select(..) system call in Routers process to >> > >listen >> > > > to traffic from other routers. If yes, can you explain why? I think any >> > > > child process does not know the socket id of the other child process >> > >[They >> > > > only know the port number]. So how can one use select(..) system call. I >> > > > think select(..) should be used here only in the child router process to >> > > > indicate a time out or if some packet has been received. Kindly >> > >elaborate. Yes, Vivek, you need to use select so that you can both (1) get incoming messages and (2) have timers that expire at the same time. Although there are some other ways to do this (say, with signals), select (plus a timers package) is by FAR the cleanest. -John Heidemann From johnh@ISI.EDU Tue Mar 5 10:58:41 2002 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 g25IweR17560 for ; Tue, 5 Mar 2002 10:58:40 -0800 (PST) Received: from dash.isi.edu (dash.isi.edu [128.9.160.184]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g25Iwet26030; Tue, 5 Mar 2002 10:58:40 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g25ItGX03931; Tue, 5 Mar 2002 10:55:16 -0800 Message-Id: <200203051855.g25ItGX03931@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Alefiya Hussain Cc: ragarwal , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] stage 1 output In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Tue, 05 Mar 2002 10:55:16 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 03 Mar 2002 19:23:26 PST, Alefiya Hussain wrote: > >On Fri, 1 Mar 2002, ragarwal wrote: > >> >> In the stage 1 output, how does a router know the IP addresses of other >> routers? >> There is no field in the LSA which lists the IP addresses of its >> neighbours. There is this responsible routes field, but the routers in its >> own AS do not come under this. So in responsible routes field it can only >> list its own IP address (in stage 1 and 2, atleast). > > >Each router that generates a LSA will include the routes that it is >responsible for. ie: its own IP address. Hence for s1-X.out you will know >the IP addresses for all the routers in the network. Just to clarify, the LSA message does NOT contain an IP address for the router ("ie: its own IP address") , but for the networks that is responsible for. >> So suppose router 1 recieves LSA from 2, it knows now that 4 is a >> neighbour of 2. So it adds the line for 4 in its routing table, as hops=2 >> and nexthop = router 2. But it still doesnt know the IP address of 4!! >> since 2 doesnt have it in its LSA. it only has its own IP address !! > >1 will know 4's IP address when it receives an LSA from it. And you are correct that there will be a time when you know how to get to 4, but not what networks 4 is reponsible for. -John Heidemann From johnh@ISI.EDU Tue Mar 5 11:54:17 2002 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 g25JsHR16185 for ; Tue, 5 Mar 2002 11:54:17 -0800 (PST) Received: from dash.isi.edu (dash.isi.edu [128.9.160.184]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g25JsGt27078; Tue, 5 Mar 2002 11:54:16 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g25Jorf04435; Tue, 5 Mar 2002 11:50:53 -0800 Message-Id: <200203051950.g25Jorf04435@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Alefiya Hussain Cc: Parag Samdadiya , csci551-talk@ISI.EDU, Quamrul Hasan Tipu Subject: Re: [Csci551-talk] Confusion in output files to generate In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Tue, 05 Mar 2002 11:50:53 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Mon, 04 Mar 2002 14:09:25 PST, Alefiya Hussain wrote: >s0.out will be created by the manager. The routers will create the >following file: > -First create s1-X.out based on how the routing converges. > -Then recalculate the routes by executing the routing protocol > with packet losses for stage 2. Create s2-X.out > -Then assume the presence of external routers and use policy routing to > create s3-X.out > >Do not create s1-X.out and s2-X.out simultaneously. The transient routing >tables/ numbers of message sent / number of message received will be >different for these two stage. Actually, Alefiya's statement at the top is not correct. 1. always generate s0 2. then determine which of s1, s2, s3 you should write based on the config file: a. if there are any external routes, write only s3 b. if the loss model is non-zero, write only s2 c. if neither a or b holds, write only s1 You should always produce s0 and ONLY ONE of s1/s2/s3. I previously clarified this on the mailing list. (Alefiya, please add this to your hints page.) -John Heidemann From hussain@ISI.EDU Tue Mar 5 13:51:07 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g25Lp7R12864 for ; Tue, 5 Mar 2002 13:51:07 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g25Lp7C06667; Tue, 5 Mar 2002 13:51:07 -0800 (PST) Date: Tue, 5 Mar 2002 13:51:06 -0800 (PST) From: Alefiya Hussain To: csci551-talk@ISI.EDU cc: johnh@ISI.EDU Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Testing and Debugging of Projects Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: ISD requires CS551 project to run either on the Sun machines in the labs or on nunki.usc.edu using "cs551P1" as the process name. all the best, Alefiya From arsalank@usc.edu Tue Mar 5 15:26:40 2002 Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g25NQeR05006 for ; Tue, 5 Mar 2002 15:26:40 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id PAA24699 for ; Tue, 5 Mar 2002 15:27:13 -0800 (PST) Received: from gnubird (adsl-66-122-133-74.dsl.lsan03.pacbell.net [66.122.133.74]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g25NQWl08326 for ; Tue, 5 Mar 2002 15:26:34 -0800 (PST) Message-ID: <000e01c1c4e0$18191560$0202a8c0@usc.edu> From: "Arsalan Khan" To: Date: Tue, 5 Mar 2002 23:25:11 -0800 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_000B_01C1C49C.FF383320" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Subject: [Csci551-talk] External Routes? Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: This is a multi-part message in MIME format. ------=_NextPart_000_000B_01C1C49C.FF383320 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Prof, TA, I have a couple of questions/requests regarding the external routes: - Why are the external IDs negative and how to use them properly? The = sample output files give little information. - A little bit detail on the exact use of MED in the project would be = very helpful. Thanks - Arsalan M Khan ------=_NextPart_000_000B_01C1C49C.FF383320 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Prof, TA,
 
I have a couple of questions/requests = regarding the=20 external routes:
- Why are the external IDs negative and = how to use=20 them properly? The sample output files give little = information.
- A little bit detail on the exact use = of MED in=20 the project would be very helpful.
 
Thanks
 
- Arsalan M Khan
 
------=_NextPart_000_000B_01C1C49C.FF383320-- From alsaeed@usc.edu Tue Mar 5 15:40:25 2002 Received: from lsmls01.we.mediaone.net (lsmls01.we.mediaone.net [24.130.1.20]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g25NeOR19075 for ; Tue, 5 Mar 2002 15:40:24 -0800 (PST) Received: from mss (c-24-130-200-13.we.mediaone.net [24.130.200.13]) by lsmls01.we.mediaone.net (8.11.4/8.11.3) with SMTP id g25NcMI01520 for ; Tue, 5 Mar 2002 15:38:22 -0800 (PST) Message-ID: <003901c1c49f$67e77230$0100a8c0@we.mediaone.net> From: "MSS" To: Subject: Re: [Csci551-talk] External Routes? Date: Tue, 5 Mar 2002 15:42:26 -0800 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0036_01C1C45C.59829550" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: This is a multi-part message in MIME format. ------=_NextPart_000_0036_01C1C45C.59829550 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I have the same Q about the external routers. Quoting the project = handout: "Clarification 19-Feb-02: When generate the output for external = routes, the id portion can correspond to either the external router's id or the peer it = talks to, and the distance field can be include the hop to the destination router or not at your = discretion." So I used the internal router id of the router responsible for these = routes when printing the routing tables. And the cost includes the hop = count and MED value. Is this correct? -Alsaeed ----- Original Message -----=20 From: Arsalan Khan=20 To: csci551-talk@mailman.isi.edu=20 Sent: Tuesday, March 05, 2002 11:25 PM Subject: [Csci551-talk] External Routes? Prof, TA, I have a couple of questions/requests regarding the external routes: - Why are the external IDs negative and how to use them properly? The = sample output files give little information. - A little bit detail on the exact use of MED in the project would be = very helpful. Thanks - Arsalan M Khan ------=_NextPart_000_0036_01C1C45C.59829550 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I have the same Q about the external routers. = Quoting the=20 project handout:
 
        =  "Clarification=20 19-Feb-02: When generate the output for external routes, the = id=20 portion can
        =  correspond to=20 either the external router's id or the peer it talks to, and the = distance=20 field
         can be = include the=20 hop to the destination router or not at your discretion."
 
So I used the internal router id of the router = responsible for=20 these routes when printing the routing tables. And the cost includes the = hop=20 count and MED value.
 
Is this correct?
 
-Alsaeed
 
----- Original Message -----=20
From: Arsalan = Khan
Sent: Tuesday, March 05, 2002 11:25 PM
Subject: [Csci551-talk] External Routes?

Prof, TA,
 
I have a couple of questions/requests = regarding the=20 external routes:
- Why are the external IDs negative and = how to use=20 them properly? The sample output files give little = information.
- A little bit detail on the exact use = of MED in=20 the project would be very helpful.
 
Thanks
 
- Arsalan M Khan
 
------=_NextPart_000_0036_01C1C45C.59829550-- From johnh@ISI.EDU Tue Mar 5 17:39:27 2002 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 g261dQR08222 for ; Tue, 5 Mar 2002 17:39:26 -0800 (PST) Received: from dash.isi.edu (dash.isi.edu [128.9.160.184]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g261dQt22928 for ; Tue, 5 Mar 2002 17:39:26 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g261a2T02712; Tue, 5 Mar 2002 17:36:02 -0800 Message-Id: <200203060136.g261a2T02712@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: "Arsalan Khan" Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] External Routes? In-reply-to: <000e01c1c4e0$18191560$0202a8c0@usc.edu> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Tue, 05 Mar 2002 17:36:02 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Tue, 05 Mar 2002 23:25:11 PST, "Arsalan Khan" wrote: >I have a couple of questions/requests regarding the external routes: > >- Why are the external IDs negative and how to use them properly? The sample >output files give little information. The sign is mostly irrelevant for the assignment, just treat them all as "ids". >- A little bit detail on the exact use of MED in the project would be very >helpful. See the definition of routing protority in the assignment. For more details, see the discussion of MED in BGP from class (and the optional Stewart book if you want more detail). -John Heidemann From arsalank@usc.edu Tue Mar 5 17:56:29 2002 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 g261uTR12292 for ; Tue, 5 Mar 2002 17:56:29 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g261uTt27751; Tue, 5 Mar 2002 17:56:29 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id RAA27467; Tue, 5 Mar 2002 17:57:02 -0800 (PST) Received: from gnubird (adsl-66-122-133-74.dsl.lsan03.pacbell.net [66.122.133.74]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g261uLl22738; Tue, 5 Mar 2002 17:56:21 -0800 (PST) Message-ID: <010b01c1c4f5$051b6e80$0202a8c0@usc.edu> From: "Arsalan Khan" To: "John Heidemann" Cc: References: <200203060136.g261a2T02712@dash.isi.edu> Subject: Re: [Csci551-talk] External Routes? Date: Wed, 6 Mar 2002 01:55:03 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Prof, TA, One little bit that I want to clarify. From the sample output it looks as if a node sourcing external routes advertises them as internal routes to the internal routers. Is this observation correct? The LSA format supports my observation though. - Arsalan Khan ----- Original Message ----- From: "John Heidemann" To: "Arsalan Khan" Cc: Sent: Tuesday, March 05, 2002 5:36 PM Subject: Re: [Csci551-talk] External Routes? > On Tue, 05 Mar 2002 23:25:11 PST, "Arsalan Khan" wrote: > >I have a couple of questions/requests regarding the external routes: > > > >- Why are the external IDs negative and how to use them properly? The sample > >output files give little information. > > The sign is mostly irrelevant for the assignment, just treat them all > as "ids". > > >- A little bit detail on the exact use of MED in the project would be very > >helpful. > > See the definition of routing protority in the assignment. > For more details, see the discussion of MED in BGP from class (and the > optional Stewart book if you want more detail). > > -John Heidemann > From arsalank@usc.edu Tue Mar 5 17:57:02 2002 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 g261v1R12446 for ; Tue, 5 Mar 2002 17:57:01 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g261v1t27903; Tue, 5 Mar 2002 17:57:01 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id RAA27854; Tue, 5 Mar 2002 17:57:34 -0800 (PST) Received: from gnubird (adsl-66-122-133-74.dsl.lsan03.pacbell.net [66.122.133.74]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g261usl23162; Tue, 5 Mar 2002 17:56:56 -0800 (PST) Message-ID: <011001c1c4f5$18d92480$0202a8c0@usc.edu> From: "Arsalan Khan" To: "John Heidemann" Cc: References: <200203060136.g261a2T02712@dash.isi.edu> Subject: Re: [Csci551-talk] External Routes? Date: Wed, 6 Mar 2002 01:55:16 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Prof, TA, One little bit that I want to clarify. From the sample output it looks as if a node sourcing external routes advertises them as internal routes to the internal routers. Is this observation correct? The LSA format supports my observation though. - Arsalan Khan ----- Original Message ----- From: "John Heidemann" To: "Arsalan Khan" Cc: Sent: Tuesday, March 05, 2002 5:36 PM Subject: Re: [Csci551-talk] External Routes? > On Tue, 05 Mar 2002 23:25:11 PST, "Arsalan Khan" wrote: > >I have a couple of questions/requests regarding the external routes: > > > >- Why are the external IDs negative and how to use them properly? The sample > >output files give little information. > > The sign is mostly irrelevant for the assignment, just treat them all > as "ids". > > >- A little bit detail on the exact use of MED in the project would be very > >helpful. > > See the definition of routing protority in the assignment. > For more details, see the discussion of MED in BGP from class (and the > optional Stewart book if you want more detail). > > -John Heidemann > From qcao@usc.edu Wed Mar 6 08:36:30 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g26GaUR17109 for ; Wed, 6 Mar 2002 08:36:30 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id IAA05418 for ; Wed, 6 Mar 2002 08:37:04 -0800 (PST) Received: from localhost (qcao@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g26GbBH20295 for ; Wed, 6 Mar 2002 08:37:11 -0800 (PST) Date: Wed, 6 Mar 2002 08:37:11 -0800 (PST) From: qcao To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] still on LSA message Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, Sorry if someone mentioned that, but I am wondering that when routers send LSA message, could we use fixed length struct instead of varial length string? It could be much easier if we just pass the struct pointer in sendto(). Or if we need to copy the LSA to a buffer, how to copy a 16 bits short int to a string, I tried some ways but all failed. And for the 8 bits prefix and 32 bits IP address, I think use hex value (like FFF0) will be good, but I could not find on books how to manipulate a hex value, like which type to use and how to assign the value, etc. Sorry for asking the details, but I could not find them on my reference books or the web. :( Thanks, -Qun From qcao@usc.edu Wed Mar 6 09:03:49 2002 Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g26H3nR29553 for ; Wed, 6 Mar 2002 09:03:49 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id JAA00794 for ; Wed, 6 Mar 2002 09:04:23 -0800 (PST) Received: from localhost (qcao@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g26H4Ui21380 for ; Wed, 6 Mar 2002 09:04:30 -0800 (PST) Date: Wed, 6 Mar 2002 09:04:30 -0800 (PST) From: qcao To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] stage 1 output match Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, In stage 1, we output the converging process of routing tables. Do we need to match the nubmer of message send/receive with the sample output? I think different implementations will make this different. Or all we need to match is the final route table? Thanks, -Qun From johnh@ISI.EDU Wed Mar 6 09:32:44 2002 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 g26HWiR10927 for ; Wed, 6 Mar 2002 09:32:44 -0800 (PST) Received: from dash.isi.edu (dash.isi.edu [128.9.160.184]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g26HWit08118; Wed, 6 Mar 2002 09:32:44 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g26HTHC08139; Wed, 6 Mar 2002 09:29:18 -0800 Message-Id: <200203061729.g26HTHC08139@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Gajanan Bhat Cc: Alefiya Hussain , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Mistake in the sample output. In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Wed, 06 Mar 2002 09:29:17 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Mon, 04 Mar 2002 02:31:42 PST, Gajanan Bhat wrote: >Hi, > >I was just comparing the output of my files for the second stage with the >sample output given on the web site and I noticed that the next hop for >router 2 in the file s1-4.out is given as 1, which is wrong. >Thought I will let you know.. You are correct that that output is NOT correct. Thanks for catching that. -John Heidemann From johnh@ISI.EDU Wed Mar 6 09:40:15 2002 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 g26HeER14654 for ; Wed, 6 Mar 2002 09:40:14 -0800 (PST) Received: from dash.isi.edu (dash.isi.edu [128.9.160.184]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g26HeEt13871; Wed, 6 Mar 2002 09:40:14 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g26Han008167; Wed, 6 Mar 2002 09:36:49 -0800 Message-Id: <200203061736.g26Han008167@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: "MSS" Cc: Alefiya Hussain , Quamrul Hasan Tipu , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] External Routes? In-reply-to: <003901c1c49f$67e77230$0100a8c0@we.mediaone.net> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=X-UNKNOWN Date: Wed, 06 Mar 2002 09:36:49 -0800 From: John Heidemann Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by gamma.isi.edu id g26HeER14654 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Tue, 05 Mar 2002 15:42:26 PST, "MSS" wrote: >[1 ] > >[2 ] > >I have the same Q about the external routers. Quoting the project handout: > > > > "Clarification 19-Feb-02: When generate the output for external >routes, the id portion can > > correspond to either the external router's id or the peer it talks >to, and the distance field > > can be include the hop to the destination router or not at your >discretion." > > > >So I used the internal router id of the router responsible for these routes >when printing the routing tables. And the cost includes the hop count and MED >value. > > > >Is this correct? > That sounds consistent with the clarification. -John Heidemann From hussain@ISI.EDU Wed Mar 6 11:46:54 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g26JkrR26026 for ; Wed, 6 Mar 2002 11:46:53 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g26Jkhx00971; Wed, 6 Mar 2002 11:46:43 -0800 (PST) Date: Wed, 6 Mar 2002 11:46:43 -0800 (PST) From: Alefiya Hussain To: qcao cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] still on LSA message In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 6 Mar 2002, qcao wrote: > Hi, > > Sorry if someone mentioned that, but I am wondering that when routers send > LSA message, could we use fixed length struct instead of varial length > string ? It could be much easier if we just pass the struct pointer in > sendto(). depending on how many local routes/neighbouring routers are present, the LSA is going to beof variable length. I am not sure how you could define it as a fixed length structure ? > Or if we need to copy the LSA to a buffer, how to copy a 16 bits short int > to a string, I tried some ways but all failed. did you try uding sprintf ? > And for the 8 bits prefix and 32 bits IP address, I think use hex value > (like FFF0) will be good, but I could not find on books how to manipulate > a hex value, like which type to use and how to assign the value, etc. Hex values can be printed using "%x" and assigned using "0xffff" where ffff is the value you need to assign. alefiya From hussain@ISI.EDU Wed Mar 6 11:59:05 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g26Jx5R02173 for ; Wed, 6 Mar 2002 11:59:05 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g26Jx4E09871; Wed, 6 Mar 2002 11:59:04 -0800 (PST) Date: Wed, 6 Mar 2002 11:59:04 -0800 (PST) From: Alefiya Hussain To: qcao cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] stage 1 output match In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: The number of message sent/received can be calculated offline as follows: Refer to sample topology in project discription: Each router sends an update only when it receives a link-state update we will get the following number of messages generated: router 1: 3 messages ( one message at the start with no links, one with link 2 and one with link 3) router 2: 3 mesgs router 3: 4 mesgs router 4: 4 mesgs router 5: 3 mesgs Hence the maximum number of LSA messages generated in topology = 17 ie: sum of all LSA generated Router 1 will receive these messages from 2 links: Messages received by router 1 = 2*17 + 2*17 The first term 2*17 = for each link attached to router 1 ie: 2 and 3 Second term 2*17 = for the ACK's it receives for the messages sent from 2 and 3. Hence total message received = 68 Total messages sent = 68 ( 17 messages on two links, 17 acks on two links) Hence if your message sent/received does not match this number, please specifiy in your README file, why it does not. alefiya On Wed, 6 Mar 2002, qcao wrote: > Hi, > > In stage 1, we output the converging process of routing tables. Do we need > to match the nubmer of message send/receive with the sample output? I > think different implementations will make this different. Or all we need > to match is the final route table? > > Thanks, > -Qun > From ganglu@usc.edu Wed Mar 6 12:31:30 2002 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 g26KVTR15463 for ; Wed, 6 Mar 2002 12:31:29 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g26KVSt14591 for ; Wed, 6 Mar 2002 12:31:28 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id MAA24283 for ; Wed, 6 Mar 2002 12:31:51 -0800 (PST) Received: from godavari (eebtux.usc.edu [128.125.111.218]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g26KVGl27475 for ; Wed, 6 Mar 2002 12:31:17 -0800 (PST) Message-ID: <002f01c1c54d$f4f75c90$032aa8c0@usc.edu> From: "Gang Lu" To: References: Subject: Re: [Csci551-talk] stage 1 output match Date: Wed, 6 Mar 2002 12:31:49 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, I agree with you on the number of message a router will generate. But about the number of messages received, here is an example on this problem: 1-------3 | | | | 2------4 4 will generate 3 LSA messages. But 1 probably will not received 2*3 LSA messages originated from 4. Think a case that 4 generate a LSA message with sequence number 1. 3 first receive 4's message then forwared to 1, 1 then reforward to 2. Only fter that wil 2 receive 4's message from 4 and 2 will not forward to 1 but drop this message because it is an old one. So node 1 actually will receive only one copy of node 4's LSA message with sequence number 1. But node 2 received twice. Also 1 will receive its own LSA message only once from 2. 3 will not forward the message back to 1. The total number of messages received for each node varies with each run of my program. If this is wrong, please notify me ASAP and I have to fix it. Thanks very much, Gang Lu ----- Original Message ----- From: "Alefiya Hussain" To: "qcao" Cc: Sent: Wednesday, March 06, 2002 11:59 AM Subject: Re: [Csci551-talk] stage 1 output match > The number of message sent/received can be calculated offline as follows: > > Refer to sample topology in project discription: > Each router sends an update only when it receives a link-state > update we will get the following number of messages generated: > > router 1: 3 messages ( one message at the start with no links, one with > link 2 and one with link 3) > router 2: 3 mesgs > router 3: 4 mesgs > router 4: 4 mesgs > router 5: 3 mesgs > > Hence the maximum number of LSA messages generated in topology = 17 > ie: sum of all LSA generated > > Router 1 will receive these messages from 2 links: > Messages received by router 1 = 2*17 + 2*17 > The first term 2*17 = for each link attached to router 1 ie: 2 and 3 > Second term 2*17 = for the ACK's it receives for the messages sent from 2 > and 3. > > Hence total message received = 68 > Total messages sent = 68 ( 17 messages on two links, 17 acks on two links) > > Hence if your message sent/received does not match this number, please > specifiy in your README file, why it does not. > > alefiya > > > On Wed, 6 Mar 2002, qcao wrote: > > > Hi, > > > > In stage 1, we output the converging process of routing tables. Do we need > > to match the nubmer of message send/receive with the sample output? I > > think different implementations will make this different. Or all we need > > to match is the final route table? > > > > Thanks, > > -Qun > > > ----- Original Message ----- From: "Alefiya Hussain" To: "qcao" Cc: Sent: Wednesday, March 06, 2002 11:59 AM Subject: Re: [Csci551-talk] stage 1 output match > The number of message sent/received can be calculated offline as follows: > > Refer to sample topology in project discription: > Each router sends an update only when it receives a link-state > update we will get the following number of messages generated: > > router 1: 3 messages ( one message at the start with no links, one with > link 2 and one with link 3) > router 2: 3 mesgs > router 3: 4 mesgs > router 4: 4 mesgs > router 5: 3 mesgs > > Hence the maximum number of LSA messages generated in topology = 17 > ie: sum of all LSA generated > > Router 1 will receive these messages from 2 links: > Messages received by router 1 = 2*17 + 2*17 > The first term 2*17 = for each link attached to router 1 ie: 2 and 3 > Second term 2*17 = for the ACK's it receives for the messages sent from 2 > and 3. > > Hence total message received = 68 > Total messages sent = 68 ( 17 messages on two links, 17 acks on two links) > > Hence if your message sent/received does not match this number, please > specifiy in your README file, why it does not. > > alefiya > > > On Wed, 6 Mar 2002, qcao wrote: > > > Hi, > > > > In stage 1, we output the converging process of routing tables. Do we need > > to match the nubmer of message send/receive with the sample output? I > > think different implementations will make this different. Or all we need > > to match is the final route table? > > > > Thanks, > > -Qun > > > From samdadiy@usc.edu Wed Mar 6 16:34:19 2002 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 g270YJR27480 for ; Wed, 6 Mar 2002 16:34:19 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g270YIt21966 for ; Wed, 6 Mar 2002 16:34:18 -0800 (PST) Received: from sal-sun023.usc.edu (sal-sun023.usc.edu [128.125.5.93]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id QAA28926 for ; Wed, 6 Mar 2002 16:34:51 -0800 (PST) Received: from localhost (samdadiy@localhost) by sal-sun023.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g270Ypo06105 for ; Wed, 6 Mar 2002 16:34:51 -0800 (PST) Date: Wed, 6 Mar 2002 16:34:51 -0800 (PST) From: Parag Samdadiya To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Error (????) in Test input files Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi TA, Please correct me if I am wrong. But I think the external routes specified in the test case 1,2,3 on the hints page dont have the CIDR length (mask) So my scanner failed to read the file correctly. I am sorry if this had been brought to notice before on the mailing group and I missed that mail. Thanks, Parag From jngai618@yahoo.com Wed Mar 6 16:41:37 2002 Received: from web14104.mail.yahoo.com (web14104.mail.yahoo.com [216.136.172.134]) by gamma.isi.edu (8.11.6/8.11.2) with SMTP id g270fbR01506 for ; Wed, 6 Mar 2002 16:41:37 -0800 (PST) Message-ID: <20020307004131.72560.qmail@web14104.mail.yahoo.com> Received: from [192.59.177.5] by web14104.mail.yahoo.com via HTTP; Wed, 06 Mar 2002 16:41:31 PST Date: Wed, 6 Mar 2002 16:41:31 -0800 (PST) From: Jeffrey Ngai To: csci551-talk@mailman.isi.edu MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: [Csci551-talk] Dropped UDP Packets Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, I am just wondering if anyone runs into a problem that when a router sends out an ACK to it's neighbor (eg. r2), r2 never receives it even though there isn't any "packet-dropping" mechanism implemented? According to some previous posts, the UDP socket is supposed to buffer all incoming packets, but it SEEMS like that's not the case with my program. Somehow, some packets will got "dropped" and select() will just time out instead of signaling that the socket is ready. I am pretty sure I implement the sendto and recvfrom logic correctly, and I just can't figure out why some packets are never received after they are sent out. If you know what causes this problem, please let me know ASAP. Thanks! Jeffrey __________________________________________________ Do You Yahoo!? Try FREE Yahoo! Mail - the world's greatest free email! http://mail.yahoo.com/ From hussain@ISI.EDU Wed Mar 6 16:52:10 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g270q9R05682 for ; Wed, 6 Mar 2002 16:52:09 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g270q8D16926; Wed, 6 Mar 2002 16:52:08 -0800 (PST) Date: Wed, 6 Mar 2002 16:52:08 -0800 (PST) From: Alefiya Hussain To: Parag Samdadiya cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Error (????) in Test input files In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Thanks, i will fix it alefiya On Wed, 6 Mar 2002, Parag Samdadiya wrote: > > Hi TA, > Please correct me if I am wrong. > > But I think the external routes specified in the test case 1,2,3 > on the hints page dont have the CIDR length (mask) > So my scanner failed to read the file correctly. > > I am sorry if this had been brought to notice before on > the mailing group and I missed that mail. > > Thanks, > Parag > From ragarwal@usc.edu Wed Mar 6 17:22:53 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g271MrR16497 for ; Wed, 6 Mar 2002 17:22:53 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id RAA17703; Wed, 6 Mar 2002 17:23:26 -0800 (PST) Received: from kohpc43 (koh-pc43.usc.edu [128.125.19.43]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g271Mpl03258; Wed, 6 Mar 2002 17:22:51 -0800 (PST) Message-ID: <006f01c1c576$99b281b0$2b137d80@usc.edu> From: "Rohit Agarwal" To: "Jeffrey Ngai" , References: <20020307004131.72560.qmail@web14104.mail.yahoo.com> Subject: Re: [Csci551-talk] Dropped UDP Packets Date: Wed, 6 Mar 2002 17:21:31 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4807.1700 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4807.1700 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Jeffrey: I was having the same problem.... I think you r not setting the router's udpsocket to the readfd in select() every time you go back to select. You have to call FD_SET(etc, etc) every time...because I think select() clears the file discriptor list whenever it comes out from it. Rohit. ----- Original Message ----- From: "Jeffrey Ngai" To: Sent: Wednesday, March 06, 2002 4:41 PM Subject: [Csci551-talk] Dropped UDP Packets > Hi, > > I am just wondering if anyone runs into a problem that > when a router sends out an ACK to it's neighbor (eg. > r2), r2 never receives it even though there isn't any > "packet-dropping" mechanism implemented? According to > some previous posts, the UDP socket is supposed to > buffer all incoming packets, but it SEEMS like that's > not the case with my program. Somehow, some packets > will got "dropped" and select() will just time out > instead of signaling that the socket is ready. I am > pretty sure I implement the sendto and recvfrom logic > correctly, and I just can't figure out why some > packets are never received after they are sent out. > > If you know what causes this problem, please let me > know ASAP. Thanks! > > Jeffrey > > > __________________________________________________ > Do You Yahoo!? > Try FREE Yahoo! Mail - the world's greatest free email! > http://mail.yahoo.com/ > From kaiyuxia@usc.edu Wed Mar 6 23:03:30 2002 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 g2773UR02717 for ; Wed, 6 Mar 2002 23:03:30 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2773Tt11513; Wed, 6 Mar 2002 23:03:29 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id XAA15348; Wed, 6 Mar 2002 23:04:03 -0800 (PST) Received: from localhost (kaiyuxia@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2774Ed07811; Wed, 6 Mar 2002 23:04:14 -0800 (PST) Date: Wed, 6 Mar 2002 23:04:14 -0800 (PST) From: kaiyuxia To: Alefiya Hussain cc: qcao , Subject: Re: [Csci551-talk] stage 1 output match In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, there, I have two questions regarding the number of messages: (1) If router 2 received a message from router 1, why should it send the same message back to router 1? (2) Suppose router 2 has already got a message from router 1, then it gets the same message from router 4, why not drop the message? Forwarding the same message will not add any new information. It is my understanding that flood means sending packet received to every neighbor EXCEPT the neighbor from which the packet comes from. Maybe I am wrong. I hope that both TA and professor will clarify this at their earliest convenience. Thanks. On Wed, 6 Mar 2002, Alefiya Hussain wrote: > The number of message sent/received can be calculated offline as follows: > > Refer to sample topology in project discription: > Each router sends an update only when it receives a link-state > update we will get the following number of messages generated: > > router 1: 3 messages ( one message at the start with no links, one with > link 2 and one with link 3) > router 2: 3 mesgs > router 3: 4 mesgs > router 4: 4 mesgs > router 5: 3 mesgs > > Hence the maximum number of LSA messages generated in topology = 17 > ie: sum of all LSA generated > > Router 1 will receive these messages from 2 links: > Messages received by router 1 = 2*17 + 2*17 > The first term 2*17 = for each link attached to router 1 ie: 2 and 3 > Second term 2*17 = for the ACK's it receives for the messages sent from 2 > and 3. > > Hence total message received = 68 > Total messages sent = 68 ( 17 messages on two links, 17 acks on two links) > > Hence if your message sent/received does not match this number, please > specifiy in your README file, why it does not. > > alefiya > > > On Wed, 6 Mar 2002, qcao wrote: > > > Hi, > > > > In stage 1, we output the converging process of routing tables. Do we need > > to match the nubmer of message send/receive with the sample output? I > > think different implementations will make this different. Or all we need > > to match is the final route table? > > > > Thanks, > > -Qun > > > From kaiyuxia@usc.edu Thu Mar 7 00:54:38 2002 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 g278scR26137 for ; Thu, 7 Mar 2002 00:54:38 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g278sbt06675; Thu, 7 Mar 2002 00:54:37 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id AAA25955; Thu, 7 Mar 2002 00:55:11 -0800 (PST) Received: from localhost (kaiyuxia@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g278tNs03268; Thu, 7 Mar 2002 00:55:23 -0800 (PST) Date: Thu, 7 Mar 2002 00:55:23 -0800 (PST) From: kaiyuxia To: John Heidemann cc: Alefiya Hussain , In-Reply-To: <200203051855.g25ItGX03931@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Need clarification about LSA & sample output Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, Prof. & TA, Accordint to the LSA format, ====================== Number of links: 16 bits The number of routers directly connected to origin router. Count only the links that are up. Far-end router id: 16 bits The id of the router connected directly to origin router. The Far-end router id is only added to the LSA after the origin router confirms that the link between them is up. ie: it receives a LSA from the far-end router. ====================== But in the sample output (based on the topology given in project A), router 1 has 3 links, one to router 2, one to router 3, and yet another one to itself. According to sample output, a router always counts itself as a far end router and a self-connected link. Is this the expected behaviour? If this is true, there may be some inconsistency with the following statement: ( Because there will always be one link is up ). =========== Even if routers know the port numbers and/or ids of their peers, they are not allowed to assume the link is up until they've actually heard a UDP message from that peer. Thus, the first message a router can send to its peers is a message with 0 neighboring links. (Hopefully it will be able to send another more positive message quickly. :-) =========== Thanks and regards, From johnh@ISI.EDU Thu Mar 7 08:19:11 2002 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 g27GJBR25318 for ; Thu, 7 Mar 2002 08:19:11 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27GJAt23892 for ; Thu, 7 Mar 2002 08:19:10 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g27GFJb15362; Thu, 7 Mar 2002 08:15:19 -0800 Message-Id: <200203071615.g27GFJb15362@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: "Arsalan Khan" Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] External Routes? In-reply-to: <010b01c1c4f5$051b6e80$0202a8c0@usc.edu> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Thu, 07 Mar 2002 08:15:19 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 06 Mar 2002 01:55:03 PST, "Arsalan Khan" wrote: >Prof, TA, > >One little bit that I want to clarify. From the sample output it looks as if >a node sourcing external routes advertises them as internal routes to the >internal routers. Is this observation correct? The LSA format supports my >observation though. The short answer---yes, this is OK, a node can spread all the routes it is responsible for. The longer answer---you should think about WHY this is OK. When would it cause problems to forward other people's routes for them? When would it be necessary? In this case it's necessary, since the internal routers need to know how to route traffic to external routes. But when would it cause problems? Why would it cause problems? How can those problems be avoided? We talked about these things in class wrt BGP. -John Heidemann From johnh@ISI.EDU Thu Mar 7 08:35:53 2002 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 g27GZrR28870 for ; Thu, 7 Mar 2002 08:35:53 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27GZqt01979 for ; Thu, 7 Mar 2002 08:35:52 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g27GWP515437; Thu, 7 Mar 2002 08:32:25 -0800 Message-Id: <200203071632.g27GWP515437@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: qcao , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] still on LSA message In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Thu, 07 Mar 2002 08:32:25 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 06 Mar 2002 11:46:43 PST, Alefiya Hussain wrote: > > >On Wed, 6 Mar 2002, qcao wrote: > >> Hi, >> >> Sorry if someone mentioned that, but I am wondering that when routers send >> LSA message, could we use fixed length struct instead of varial length >> string ? It could be much easier if we just pass the struct pointer in >> sendto(). > >depending on how many local routes/neighbouring routers are present, the >LSA is going to beof variable length. I am not sure how you could define >it as a fixed length structure ? More specifically, I don't see how you could possibly do this as a fixed-length structure and meet the message formatting requirements. (And even if there wasn't two variable length fields in the LSA message, you STILL can't count on a static C structure to get the right encoding. It is completely within the C standard to allow the compiler to re-order or add invisible padding fields to the structure. Almost ALL compilers will pad structure fields to 32-bit boundries because all modern CPUs either require it or work much much faster with aligned data.) Part of the goal of the assignment is that you review how data is encoded when it's sent in fixed-format messages like in other network protocols! >> Or if we need to copy the LSA to a buffer, how to copy a 16 bits short int >> to a string, I tried some ways but all failed. > >did you try uding sprintf ? So sprintf will convert int of varying sizes to strings, but the LSA message is NOT a string, it's binary. You need to identify the relevant bits and copy them. See the C operaters about bit-manipulation (&, |, >>, <<, etc). I assume that all graduate students studied data representations in their undergraduate computer architecture classes. > > >> And for the 8 bits prefix and 32 bits IP address, I think use hex value >> (like FFF0) will be good, but I could not find on books how to manipulate >> a hex value, like which type to use and how to assign the value, etc. > >Hex values can be printed using "%x" and assigned using "0xffff" where >ffff is the value you need to assign. Again, %x and sprintf works good for printing to strings, but is NOT what you should use to copy data into a binary encoding. -John Heidemann From johnh@ISI.EDU Thu Mar 7 08:42:22 2002 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 g27GgKR01943 for ; Thu, 7 Mar 2002 08:42:20 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27GgJt04329; Thu, 7 Mar 2002 08:42:19 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g27GcrQ15481; Thu, 7 Mar 2002 08:38:53 -0800 Message-Id: <200203071638.g27GcrQ15481@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Alefiya Hussain Cc: qcao , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] stage 1 output match In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Thu, 07 Mar 2002 08:38:53 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 06 Mar 2002 11:59:04 PST, Alefiya Hussain wrote: >The number of message sent/received can be calculated offline as follows: > >Refer to sample topology in project discription: >Each router sends an update only when it receives a link-state >update we will get the following number of messages generated: > >router 1: 3 messages ( one message at the start with no links, one with >link 2 and one with link 3) >... > >Hence if your message sent/received does not match this number, please >specifiy in your README file, why it does not. Alefiya described a very good back of the envelope way to judge if you're seeing the right number of messages. However, to emphasize her last comment: There ARE other ways you could get different numbers of messages, so if you do get different numbers, you need to describe what you get, why you think your count is correct, and why your count is different than Alefiya's sketch. That said, if you're seeing a VERY different number of messages, you might ask yourself if you have a bug in your code :-( . -John Heidemann From johnh@ISI.EDU Thu Mar 7 08:49:16 2002 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 g27GnFR05302 for ; Thu, 7 Mar 2002 08:49:16 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27GnFt07727 for ; Thu, 7 Mar 2002 08:49:15 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g27Gjef15526; Thu, 7 Mar 2002 08:45:40 -0800 Message-Id: <200203071645.g27Gjef15526@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: "Rohit Agarwal" Cc: "Jeffrey Ngai" , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Dropped UDP Packets In-reply-to: <006f01c1c576$99b281b0$2b137d80@usc.edu> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Thu, 07 Mar 2002 08:45:40 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 06 Mar 2002 17:21:31 PST, "Rohit Agarwal" wrote: >Jeffrey: >I was having the same problem.... >I think you r not setting the router's udpsocket to the readfd in select() >every time you go back to select. You have to call FD_SET(etc, etc) every >time...because I think select() clears the file discriptor list whenever it >comes out from it. Good suggestion of something to check. Another thing is that it IS possible to drop packets if the socket buffer fills up. But if that happens (given how small LSA messages are) you probably have a bug in your program... (As at least an existance proof: I have code that does not exhibit UDP drops.) -John Heidemann From johnh@ISI.EDU Thu Mar 7 08:58:23 2002 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 g27GwNR09561 for ; Thu, 7 Mar 2002 08:58:23 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27GwNt12049; Thu, 7 Mar 2002 08:58:23 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g27GsuU15592; Thu, 7 Mar 2002 08:54:56 -0800 Message-Id: <200203071654.g27GsuU15592@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: kaiyuxia Cc: Alefiya Hussain , qcao , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] stage 1 output match In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Thu, 07 Mar 2002 08:54:56 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Wed, 06 Mar 2002 23:04:14 PST, kaiyuxia wrote: >Hi, there, > >I have two questions regarding the number of messages: >(1) If router 2 received a message from router 1, why should it send the >same message back to router 1? You're correct that flooding a message back where it came from does not add new info, so an optimization would be to suppress it. But forwarding it back doesn't cause harm, right? Therefore this optimization is not manditory. (In fact, it would be interesting to treat the message as the ACK... then you wouldn't need an explicit ACK field.) >(2) Suppose router 2 has already got a message from router 1, then it gets >the same message from router 4, why not drop the message? Forwarding the >same message will not add any new information. Again, yes, there's no need to forward a message you've already seen before. Assuing you really have seen it before, i.e., it does not have new information. In fact, what would happen to the number of messages in the system if you re-forwarded duplicate messages and did NOT suppress them? Hmmm... maybe this is why some people are observing overflowing UDP socket buffers... -John Heidemann From johnh@ISI.EDU Thu Mar 7 09:05:23 2002 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 g27H5NR13496 for ; Thu, 7 Mar 2002 09:05:23 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27H5Mt15827; Thu, 7 Mar 2002 09:05:22 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g27H1tp15667; Thu, 7 Mar 2002 09:01:55 -0800 Message-Id: <200203071701.g27H1tp15667@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: kaiyuxia Cc: Alefiya Hussain , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Need clarification about LSA & sample output In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Thu, 07 Mar 2002 09:01:55 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Thu, 07 Mar 2002 00:55:23 PST, kaiyuxia wrote: >Hi, Prof. & TA, > >Accordint to the LSA format, > >====================== >Number of links: 16 bits >The number of routers directly connected to origin router. Count only the >links that are up. > >Far-end router id: 16 bits >The id of the router connected directly to origin router. >The Far-end router id is only added to the LSA after the >origin router confirms that the link between them is up. >ie: it receives a LSA from the far-end router. >====================== > >But in the sample output (based on the topology given in project A), >router 1 has 3 links, one to router 2, one to router 3, and yet another >one to itself. Why do you say that? I.e., can you be more specific about what in the sample output indicates 1 has three links? (I don't see anything in the output that indicates the number of links a router has, other than encoded in the LSA message.) >According to sample output, a router always counts itself as a far end >router and a self-connected link. Again, why do you say that? Yes, a router always has a route entry to its own network. The first entry of s1-1.out is therefore: 1 10.1.0.0/16 1 0 --- But this is a routing table entry, not a "link". Or am I misunderstanding your question? >Is this the expected behaviour? > >If this is true, there may be some inconsistency with the following >statement: ( Because there will always be one link is up ). > >=========== >Even if routers know the port numbers and/or ids of their peers, they are >not allowed to assume the link is up until they've >actually heard a UDP message from that peer. Thus, the first message a >router can send to its peers is a message with 0 neighboring links. >(Hopefully it >will be able to send another more positive message quickly. :-) >=========== I don't see an inconsistency there. I think perhaps you're confusing routes (i.e., IP address prefixes and their next-hops) and links (lists of router ids that routers are directly connected to). -John Heidemann From johnh@ISI.EDU Thu Mar 7 09:11:05 2002 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 g27HB4R16130 for ; Thu, 7 Mar 2002 09:11:04 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27HB3t17783 for ; Thu, 7 Mar 2002 09:11:03 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g27H7bZ15725 for ; Thu, 7 Mar 2002 09:07:37 -0800 Message-Id: <200203071707.g27H7bZ15725@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: csci551-talk@ISI.EDU MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Thu, 07 Mar 2002 09:07:37 -0800 From: John Heidemann Subject: [Csci551-talk] Amit Kumar networking talk today Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: By the way, USC has a number of good networking researchers visiting campus and giving talks. If you're interested in networking you might check some of these out. Amit Kumar's VPN work ties DIRECTLY in to the IntServe/DiffServe work we've been talking about, and his BGP work also sounds interesting. Talks from current researchers are particularly good to find out what's really current in networking research, as opposed to class which necessarily lags the state-of-the-art by a year or more. (Rumors that he'll be answering questions after his talk about link-state routing in 551 projects are unfortunately unfounded. :-) -John Heidemann ---------------------------------------------------------------------- "Algorithms for Network Management" Amit Kumar Cornell University Thursday, March 7, 2002 EEB 248 12:00 1:00pm Abstract: Communication networks have witnessed phenomenal growth in recent years. The issues raised by the introduction of new technology and protocols have led to problems which are both theoretically fundamental and relevant to current applications. In this talk, I shall focus on two key themes, namely, virtual private networks (VPN) and the Border Gateway Protocol (BGP). VPNs provide customers with predictable and secure network connections over a shared network. Provisioning a VPN over a set of terminals gives rise to the following general network design problem: we have bounds on the cumulative amount of traffic each terminal can send and receive; we must reserve enough bandwidth on the network so that any pair-wise traffic matrix consistent with these bounds can be routed. BGP is the standard protocol for exchanging information between border routers of autonomous systems(AS) in the Internet. Within an AS, border routers exchange externally learned route information. Naive solutions for this peering between the border routers simply cannot scale to the sizes of modern AS networks. Carefully designed route reflector configurations can drastically reduce the communication costs needed by such peering sessions. We give the first algorithmic approach towards designing such configurations. We show connections of these problems with facility location problems and give efficient algorithms with provable performance guarantees. Biography: Amit Kumar is a graduate student at the department of computer science in Cornell University. He is expected to receive his PhD degree in May, 2002. He received his B. Tech. degree in computer science and engineering from Indian Institute of Technology, Kanpur. He is interested in algorithms, with a particular emphasis on the algorithmic foundations of network management protocols. Host: Ashish Goel From jngai618@yahoo.com Thu Mar 7 09:53:42 2002 Received: from web14104.mail.yahoo.com (web14104.mail.yahoo.com [216.136.172.134]) by gamma.isi.edu (8.11.6/8.11.2) with SMTP id g27HrfR06873 for ; Thu, 7 Mar 2002 09:53:41 -0800 (PST) Message-ID: <20020307175341.1803.qmail@web14104.mail.yahoo.com> Received: from [192.59.177.5] by web14104.mail.yahoo.com via HTTP; Thu, 07 Mar 2002 09:53:41 PST Date: Thu, 7 Mar 2002 09:53:41 -0800 (PST) From: Jeffrey Ngai Subject: Re: [Csci551-talk] Dropped UDP Packets To: csci551-talk@mailman.isi.edu In-Reply-To: <200203071645.g27Gjef15526@dash.isi.edu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, Any idea why big the socket buffer is? Just want to know so that I can double check if anything goes wrong. Thanks! Jeffrey --- John Heidemann wrote: > Good suggestion of something to check. > > Another thing is that it IS possible to drop packets > if the socket > buffer fills up. But if that happens (given how > small LSA messages > are) you probably have a bug in your program... > > (As at least an existance proof: I have code that > does not exhibit UDP > drops.) > > -John Heidemann __________________________________________________ Do You Yahoo!? Try FREE Yahoo! Mail - the world's greatest free email! http://mail.yahoo.com/ From ganglu@usc.edu Thu Mar 7 12:06:35 2002 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 g27K6ZR20081 for ; Thu, 7 Mar 2002 12:06:35 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27K6Zt28986 for ; Thu, 7 Mar 2002 12:06:35 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id MAA16771 for ; Thu, 7 Mar 2002 12:07:09 -0800 (PST) Received: from godavari (eebtux.usc.edu [128.125.111.218]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with SMTP id g27K6Xl15263 for ; Thu, 7 Mar 2002 12:06:33 -0800 (PST) Message-ID: <000f01c1c613$ab7abaf0$032aa8c0@usc.edu> From: "Gang Lu" To: Date: Thu, 7 Mar 2002 12:07:07 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Subject: [Csci551-talk] den passwd Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, Who knows the passwd for the den webcast for CSCi551? thanks, Gang Lu From kaiyuxia@usc.edu Thu Mar 7 14:01:01 2002 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 g27M10R04639 for ; Thu, 7 Mar 2002 14:01:00 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27M10t06087; Thu, 7 Mar 2002 14:01:00 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id OAA22889; Thu, 7 Mar 2002 14:01:24 -0800 (PST) Received: from localhost (kaiyuxia@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g27M1Zj16212; Thu, 7 Mar 2002 14:01:35 -0800 (PST) Date: Thu, 7 Mar 2002 14:01:35 -0800 (PST) From: kaiyuxia To: John Heidemann cc: Alefiya Hussain , Subject: Re: [Csci551-talk] Need clarification about LSA & sample output In-Reply-To: <200203071701.g27H1tp15667@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, Prof., Sorry that I did not state the problem clear enough. I was referring to the last-lsa-msg-sent: For router 1: "last-lsa-msg-sent 000002000f00010001000100030a010000100000000100020003" We can extract: number of links = 0003 far-end router id = 0000100020003 For me, that means router 1 also treat itself as a far end. That is what I want a clarification. Thanks very much On Thu, 7 Mar 2002, John Heidemann wrote: > On Thu, 07 Mar 2002 00:55:23 PST, kaiyuxia wrote: > >Hi, Prof. & TA, > > > >Accordint to the LSA format, > > > >==================== > >Number of links: 16 bits > >The number of routers directly connected to origin router. Count only the > >links that are up. > > > >Far-end router id: 16 bits > >The id of the router connected directly to origin router. > >The Far-end router id is only added to the LSA after the > >origin router confirms that the link between them is up. > >ie: it receives a LSA from the far-end router. > >====================== > > > >But in the sample output (based on the topology given in project A), > >router 1 has 3 links, one to router 2, one to router 3, and yet another > >one to itself. > > Why do you say that? I.e., can you be more specific about what in the > sample output indicates 1 has three links? > > (I don't see anything in the output that indicates the number of links > a router has, other than encoded in the LSA message.) > > >According to sample output, a router always counts itself as a far end > >router and a self-connected link. > > Again, why do you say that? > > Yes, a router always has a route entry to its own network. > The first entry of s1-1.out is therefore: > > 1 10.1.0.0/16 1 0 > --- > > But this is a routing table entry, not a "link". > > Or am I misunderstanding your question? > > >Is this the expected behaviour? > > > >If this is true, there may be some inconsistency with the following > >statement: ( Because there will always be one link is up ). > > > >=========== > >Even if routers know the port numbers and/or ids of their peers, they are > >not allowed to assume the link is up until they've > >actually heard a UDP message from that peer. Thus, the first message a > >router can send to its peers is a message with 0 neighboring links. > >(Hopefully it > >will be able to send another more positive message quickly. :-) > >=========== > > I don't see an inconsistency there. > > I think perhaps you're confusing routes (i.e., IP address prefixes and > their next-hops) and links (lists of router ids that routers are > directly connected to). > > -John Heidemann > From kaiyuxia@usc.edu Thu Mar 7 14:14:42 2002 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 g27MEgR10146 for ; Thu, 7 Mar 2002 14:14:42 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g27MEft13911; Thu, 7 Mar 2002 14:14:41 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id OAA08850; Thu, 7 Mar 2002 14:15:15 -0800 (PST) Received: from localhost (kaiyuxia@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g27MFQi27699; Thu, 7 Mar 2002 14:15:26 -0800 (PST) Date: Thu, 7 Mar 2002 14:15:25 -0800 (PST) From: kaiyuxia To: John Heidemann cc: Alefiya Hussain , Subject: Re: [Csci551-talk] stage 1 output match In-Reply-To: <200203071654.g27GsuU15592@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, Prof., Thanks very much for your response. Please see my comment inline. One question: According to your previous response, it is OK to have different number of messages, but need to justify it. Is this right? ( Less number of messages should be better provided that correct routing table is reached. ) On Thu, 7 Mar 2002, John Heidemann wrote: > On Wed, 06 Mar 2002 23:04:14 PST, kaiyuxia wrote: > >Hi, there, > > > >I have two questions regarding the number of messages: > >(1) If router 2 received a message from router 1, why should it send the > >same message back to router 1? > > You're correct that flooding a message back where it came from does > not add new info, so an optimization would be to suppress it. > But forwarding it back doesn't cause harm, right? > Therefore this optimization is not manditory. According to one text book (Andrew S. Tanenbaum, Computer Network, page 351, third edition ), the definition of flooding is "Another static algorithm is flooding, in which every incoming packet is sent out on every outgoing line except the one it arrived on". > > (In fact, it would be interesting to treat the message as the ACK... > then you wouldn't need an explicit ACK field.) > > >(2) Suppose router 2 has already got a message from router 1, then it gets > >the same message from router 4, why not drop the message? Forwarding the > >same message will not add any new information. > > Again, yes, there's no need to forward a message you've already seen > before. Assuing you really have seen it before, i.e., it does not > have new information. > That is why there is a sequence number field in LSA. Comparing the sequence number to make sure that we does not pass any stale information. > In fact, what would happen to the number of messages in the system if > you re-forwarded duplicate messages and did NOT suppress them? > > Hmmm... maybe this is why some people are observing overflowing UDP > socket buffers... > > -John Heidemann > From hussain@ISI.EDU Thu Mar 7 16:21:17 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g280LHR08024 for ; Thu, 7 Mar 2002 16:21:17 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g280LHJ16475; Thu, 7 Mar 2002 16:21:17 -0800 (PST) Date: Thu, 7 Mar 2002 16:21:16 -0800 (PST) From: Alefiya Hussain To: Gang Lu cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] den passwd In-Reply-To: <000f01c1c613$ab7abaf0$032aa8c0@usc.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Use the same passwd as the class website. all the best alefiya On Thu, 7 Mar 2002, Gang Lu wrote: > Hi, > > Who knows the passwd for the den webcast for CSCi551? > > thanks, > > Gang Lu > From mhr@usc.edu Thu Mar 7 20:05:18 2002 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 g2845HR22509 for ; Thu, 7 Mar 2002 20:05:17 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2845Ht05315; Thu, 7 Mar 2002 20:05:17 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id UAA14999; Thu, 7 Mar 2002 20:05:51 -0800 (PST) Received: from localhost (mhr@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2845xp21692; Thu, 7 Mar 2002 20:05:59 -0800 (PST) Date: Thu, 7 Mar 2002 20:05:59 -0800 (PST) From: mhr To: John Heidemann cc: In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Question about BGP Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Dear Prof or class, I had two basic question about BGP which may sound stupid though: 1- Why in I-BGP routers simply do not advertise their routing table like any other distant vector protocol whcich then needs full mesh or (confederation or reflecting router).Is it beacuse of loop and count to infinity problem? 2- I still do not understand the difference between Local-Pref and MED, tanx -m From johnh@ISI.EDU Fri Mar 8 08:08:47 2002 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 g28G8lR17905 for ; Fri, 8 Mar 2002 08:08:47 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g28G8lt22354 for ; Fri, 8 Mar 2002 08:08:47 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g28G5Ir27476; Fri, 8 Mar 2002 08:05:18 -0800 Message-Id: <200203081605.g28G5Ir27476@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Jeffrey Ngai Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Dropped UDP Packets In-reply-to: <20020307175341.1803.qmail@web14104.mail.yahoo.com> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 08 Mar 2002 08:05:18 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Thu, 07 Mar 2002 09:53:41 PST, Jeffrey Ngai wrote: >Hi, > >Any idea why big the socket buffer is? Just want to >know so that I can double check if anything goes >wrong. Thanks! It is implementation dependent, but you can assume at least 8KB. See getsockopt(2) and SO_SNDBUF/SO_RCVBUF if you want to know the real answer that is specific to your system. -John Heidemann From johnh@ISI.EDU Fri Mar 8 08:14:16 2002 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 g28GEGR20130 for ; Fri, 8 Mar 2002 08:14:16 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g28GEGt23550; Fri, 8 Mar 2002 08:14:16 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g28GAli27513; Fri, 8 Mar 2002 08:10:47 -0800 Message-Id: <200203081610.g28GAli27513@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: kaiyuxia Cc: Alefiya Hussain , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] stage 1 output match In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 08 Mar 2002 08:10:47 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Thu, 07 Mar 2002 14:15:25 PST, kaiyuxia wrote: >Hi, Prof., > >Thanks very much for your response. >Please see my comment inline. >One question: >According to your previous response, it is OK to have different >number of messages, but need to justify it. Is this right? >( Less number of messages should be better provided that correct routing >table is reached. ) Yes, that is correct. > >On Thu, 7 Mar 2002, John Heidemann wrote: > >> On Wed, 06 Mar 2002 23:04:14 PST, kaiyuxia wrote: >> >Hi, there, >> > >> >I have two questions regarding the number of messages: >> >(1) If router 2 received a message from router 1, why should it send the >> >same message back to router 1? >> >> You're correct that flooding a message back where it came from does >> not add new info, so an optimization would be to suppress it. >> But forwarding it back doesn't cause harm, right? >> Therefore this optimization is not manditory. > >According to one text book (Andrew S. Tanenbaum, Computer Network, page >351, third edition ), the definition of flooding is "Another static >algorithm is flooding, in which every incoming packet is sent out on >every outgoing line except the one it arrived on". Yes, as I said, that optimization is allowed but not manditory. But see the comment below (in the >>'ed text) about why you need to do more than just not send it back on the incoming link to make sure flooding terminates! -John Heidemann >> (In fact, it would be interesting to treat the message as the ACK... >> then you wouldn't need an explicit ACK field.) >> >> >(2) Suppose router 2 has already got a message from router 1, then it gets >> >the same message from router 4, why not drop the message? Forwarding the >> >same message will not add any new information. >> >> Again, yes, there's no need to forward a message you've already seen >> before. Assuing you really have seen it before, i.e., it does not >> have new information. >> >That is why there is a sequence number field in LSA. Comparing the >sequence number to make sure that we does not pass any stale information. > >> In fact, what would happen to the number of messages in the system if >> you re-forwarded duplicate messages and did NOT suppress them? >> >> Hmmm... maybe this is why some people are observing overflowing UDP >> socket buffers... >> >> -John Heidemann >> From johnh@ISI.EDU Fri Mar 8 08:20:31 2002 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 g28GKVR21779 for ; Fri, 8 Mar 2002 08:20:31 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g28GKVt25879 for ; Fri, 8 Mar 2002 08:20:31 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g28GH2g27557; Fri, 8 Mar 2002 08:17:02 -0800 Message-Id: <200203081617.g28GH2g27557@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: mhr Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Question about BGP In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 08 Mar 2002 08:17:02 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Thu, 07 Mar 2002 20:05:59 PST, mhr wrote: > >Dear Prof or class, > >I had two basic question about BGP which may sound stupid though: > >1- Why in I-BGP routers simply do not advertise their routing table > like any other distant vector protocol whcich then needs full mesh or > (confederation or reflecting router).Is it beacuse of loop and > count to infinity problem? That's a good question, and you're on the right track with loop detection. You should ask yourself what mechanisms E-BGP uses to avoid looping. (We talked about this in class.) You're correct that basic I-BGP would have looping troubles if it forwarded routes. So a good follow-up question is, what mechanisms do the I-BGP scaling mechanisms (Confederations and Route Reflectors) use to avoid looping? >2- I still do not understand the difference between Local-Pref and MED, For that general a question I encourage you to look at the class notes, the Stewart book or Gao and Rexford discussion, or talk to me or the TA in office hours. -John Heidemann From johnh@ISI.EDU Fri Mar 8 08:25:36 2002 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 g28GPaR23290 for ; Fri, 8 Mar 2002 08:25:36 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g28GPat28942; Fri, 8 Mar 2002 08:25:36 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g28GM8J27655; Fri, 8 Mar 2002 08:22:08 -0800 Message-Id: <200203081622.g28GM8J27655@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: kaiyuxia Cc: Alefiya Hussain , csci551-talk@ISI.EDU, Quamrul Hasan Tipu Subject: Re: [Csci551-talk] Need clarification about LSA & sample output In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 08 Mar 2002 08:22:08 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Thu, 07 Mar 2002 14:01:35 PST, kaiyuxia wrote: >Hi, Prof., > >Sorry that I did not state the problem clear enough. >I was referring to the last-lsa-msg-sent: >For router 1: >"last-lsa-msg-sent 000002000f00010001000100030a010000100000000100020003" >We can extract: number of links = 0003 > far-end router id = 0000100020003 >For me, that means router 1 also treat itself as a far end. > >That is what I want a clarification. Thank you for clarifing your question. You are correct in your interpretation of that message. A router is allowed to list a link to itself, but this is not required, nor is it preferred. (You're right that there's no reason to send this link because it can be inferred that each router can talk to itself.) -John Heidemann > >Thanks very much > >On Thu, 7 Mar 2002, John Heidemann wrote: > >> On Thu, 07 Mar 2002 00:55:23 PST, kaiyuxia wrote: >> >Hi, Prof. & TA, >> > >> >Accordint to the LSA format, >> > >> >==================== >> >Number of links: 16 bits >> >The number of routers directly connected to origin router. Count only the >> >links that are up. >> > >> >Far-end router id: 16 bits >> >The id of the router connected directly to origin router. >> >The Far-end router id is only added to the LSA after the >> >origin router confirms that the link between them is up. >> >ie: it receives a LSA from the far-end router. >> >====================== >> > >> >But in the sample output (based on the topology given in project A), >> >router 1 has 3 links, one to router 2, one to router 3, and yet another >> >one to itself. >> >> Why do you say that? I.e., can you be more specific about what in the >> sample output indicates 1 has three links? >> >> (I don't see anything in the output that indicates the number of links >> a router has, other than encoded in the LSA message.) >> >> >According to sample output, a router always counts itself as a far end >> >router and a self-connected link. >> >> Again, why do you say that? >> >> Yes, a router always has a route entry to its own network. >> The first entry of s1-1.out is therefore: >> >> 1 10.1.0.0/16 1 0 >> --- >> >> But this is a routing table entry, not a "link". >> >> Or am I misunderstanding your question? >> >> >Is this the expected behaviour? >> > >> >If this is true, there may be some inconsistency with the following >> >statement: ( Because there will always be one link is up ). >> > >> >=========== >> >Even if routers know the port numbers and/or ids of their peers, they are >> >not allowed to assume the link is up until they've >> >actually heard a UDP message from that peer. Thus, the first message a >> >router can send to its peers is a message with 0 neighboring links. >> >(Hopefully it >> >will be able to send another more positive message quickly. :-) >> >=========== >> >> I don't see an inconsistency there. >> >> I think perhaps you're confusing routes (i.e., IP address prefixes and >> their next-hops) and links (lists of router ids that routers are >> directly connected to). >> >> -John Heidemann >> From johnh@ISI.EDU Fri Mar 8 13:59:53 2002 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 g28LxrR15783 for ; Fri, 8 Mar 2002 13:59:53 -0800 (PST) Received: from dash.isi.edu (dash.usc.edu [128.125.187.56]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g28Lxrt00193 for ; Fri, 8 Mar 2002 13:59:53 -0800 (PST) Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g28LuPj02749 for ; Fri, 8 Mar 2002 13:56:25 -0800 Message-Id: <200203082156.g28LuPj02749@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: csci551-talk@ISI.EDU MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Fri, 08 Mar 2002 13:56:25 -0800 From: John Heidemann Subject: [Csci551-talk] additional libraries for CS551 project A Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: As said in class, all libraries (i.e., code you didn't write) for CS551 projects need to be approved. For example, the class timers library. Alefiya has a list of libraries that have been approved on her web page. If you haven't yet talked to me about your library, please know that you MUST have your libraries approved by noon March 9. After that time I will reject any new libraries. -John Heidemann From qcao@usc.edu Fri Mar 8 15:58:23 2002 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 g28NwMR29275 for ; Fri, 8 Mar 2002 15:58:22 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g28NwMt23534; Fri, 8 Mar 2002 15:58:22 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id PAA08471; Fri, 8 Mar 2002 15:58:54 -0800 (PST) Received: from localhost (qcao@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g28Nx4i08213; Fri, 8 Mar 2002 15:59:04 -0800 (PST) Date: Fri, 8 Mar 2002 15:59:04 -0800 (PST) From: qcao To: Alefiya Hussain cc: Subject: Re: [Csci551-talk] stage 1 output match In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Thanks very much for your specification. But I am thinking an optimization here: for the first LSA each router send, it only indicates itself is up, no any information for the SPF algorithm actually; so I think this message should not be flood forwarding by it's neighbors because other nodes in the graph don't need this information. Any comments on it? -Qun On Wed, 6 Mar 2002, Alefiya Hussain wrote: > The number of message sent/received can be calculated offline as follows: > > Refer to sample topology in project discription: > Each router sends an update only when it receives a link-state > update we will get the following number of messages generated: > > router 1: 3 messages ( one message at the start with no links, one with > link 2 and one with link 3) > router 2: 3 mesgs > router 3: 4 mesgs > router 4: 4 mesgs > router 5: 3 mesgs > > Hence the maximum number of LSA messages generated in topology = 17 > ie: sum of all LSA generated > > Router 1 will receive these messages from 2 links: > Messages received by router 1 = 2*17 + 2*17 > The first term 2*17 = for each link attached to router 1 ie: 2 and 3 > Second term 2*17 = for the ACK's it receives for the messages sent from 2 > and 3. > > Hence total message received = 68 > Total messages sent = 68 ( 17 messages on two links, 17 acks on two links) > > Hence if your message sent/received does not match this number, please > specifiy in your README file, why it does not. > > alefiya > > > On Wed, 6 Mar 2002, qcao wrote: > > > Hi, > > > > In stage 1, we output the converging process of routing tables. Do we need > > to match the nubmer of message send/receive with the sample output? I > > think different implementations will make this different. Or all we need > > to match is the final route table? > > > > Thanks, > > -Qun > > > From qcao@usc.edu Fri Mar 8 16:07:15 2002 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 g2907FR03023 for ; Fri, 8 Mar 2002 16:07:15 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2907Ft28357; Fri, 8 Mar 2002 16:07:15 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id QAA16910; Fri, 8 Mar 2002 16:07:47 -0800 (PST) Received: from localhost (qcao@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2907vc08905; Fri, 8 Mar 2002 16:07:57 -0800 (PST) Date: Fri, 8 Mar 2002 16:07:57 -0800 (PST) From: qcao To: John Heidemann cc: Subject: Re: [Csci551-talk] still on LSA message In-Reply-To: <200203071632.g27GWP515437@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Thu, 7 Mar 2002, John Heidemann wrote: > On Wed, 06 Mar 2002 11:46:43 PST, Alefiya Hussain wrote: > > > > > >On Wed, 6 Mar 2002, qcao wrote: > > > >> Hi, > >> > >> Sorry if someone mentioned that, but I am wondering that when routers send > >> LSA message, could we use fixed length struct instead of varial length > >> string ? It could be much easier if we just pass the struct pointer in > >> sendto(). > > > >depending on how many local routes/neighbouring routers are present, the > >LSA is going to beof variable length. I am not sure how you could define > >it as a fixed length structure ? > > More specifically, I don't see how you could possibly do this as a > fixed-length structure and meet the message formatting requirements. > > (And even if there wasn't two variable length fields in the LSA > message, you STILL can't count on a static C structure to get the > right encoding. It is completely within the C standard to allow the > compiler to re-order or add invisible padding fields to the > structure. Almost ALL compilers will pad structure fields to 32-bit > boundries because all modern CPUs either require it or work much much > faster with aligned data.) > > Part of the goal of the assignment is that you review how data is > encoded when it's sent in fixed-format messages like in other network > protocols! Sorry this whole problem may seems silly. What I was thinking is having a struct LSA in my program, which has two fixed arrays having the size of a pre-defined MAX_NEBRS ( max prossible neighbors in the graph), so the LSA will be fixed size structure padded with zeros, and I could send the structure to the other routers, then they could figure out the data based on the layout of the structure. > > >> Or if we need to copy the LSA to a buffer, how to copy a 16 bits short int > >> to a string, I tried some ways but all failed. > > > >did you try uding sprintf ? > > So sprintf will convert int of varying sizes to strings, > but the LSA message is NOT a string, it's binary. I tried sprintf and it doesn't work, just because it is string? So we are not using sendto (socket, char * , length, , , ), but which structure other than string could accomodate vary-sized binary bits? From samdadiy@usc.edu Fri Mar 8 20:54:15 2002 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 g294sFR00076 for ; Fri, 8 Mar 2002 20:54:15 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g294sFt16214; Fri, 8 Mar 2002 20:54:15 -0800 (PST) Received: from sal-sun023.usc.edu (sal-sun023.usc.edu [128.125.5.93]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id UAA07525; Fri, 8 Mar 2002 20:54:37 -0800 (PST) Received: from localhost (samdadiy@localhost) by sal-sun023.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g294seJ28440; Fri, 8 Mar 2002 20:54:40 -0800 (PST) Date: Fri, 8 Mar 2002 20:54:40 -0800 (PST) From: Parag Samdadiya To: , John Heidemann Subject: Re: [Csci551-talk] Confusion in output files to generate In-Reply-To: <200203051950.g25Jorf04435@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Does this imply that if there are external routes then the error rate would be zero ?? Otherwise we would have to generate both s2 and s3 ?? Thanks, Parag > Actually, Alefiya's statement at the top is not correct. > > 1. always generate s0 > 2. then determine which of s1, s2, s3 you should write > based on the config file: > a. if there are any external routes, write only s3 > b. if the loss model is non-zero, write only s2 > c. if neither a or b holds, write only s1 > > You should always produce s0 and ONLY ONE of s1/s2/s3. > From hussain@ISI.EDU Fri Mar 8 21:37:24 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g295bOR08066 for ; Fri, 8 Mar 2002 21:37:24 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g295bOS28840; Fri, 8 Mar 2002 21:37:24 -0800 (PST) Date: Fri, 8 Mar 2002 21:37:24 -0800 (PST) From: Alefiya Hussain To: qcao cc: csci551-talk@ISI.EDU In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] LSA with 0 links Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: FYI: I am sending this message to the news group, since we keep getting the same question. It IS acceptable to not forward LSAs with 0 links, since they only are relevant to neighbor nodes. HOWEVER, this is just an optimization and is unnecessary. In fact, it's a BAD optimization because even though YOU have no links to others, others will have links to you. In that sense it is incorrect. However, I'm willing to allow it for the assignment. -John Heidemann On Fri, 8 Mar 2002, qcao wrote: > Thanks very much for your specification. > > But I am thinking an optimization here: for the first LSA each router > send, it only indicates itself is up, no any information for the SPF > algorithm actually; so I think this message should not be flood forwarding > by it's neighbors because other nodes in the graph don't need this > information. > Any comments on it? > > -Qun > > On Wed, 6 Mar 2002, Alefiya Hussain wrote: > > > The number of message sent/received can be calculated offline as follows: > > > > Refer to sample topology in project discription: > > Each router sends an update only when it receives a link-state > > update we will get the following number of messages generated: > > > > router 1: 3 messages ( one message at the start with no links, one with > > link 2 and one with link 3) > > router 2: 3 mesgs > > router 3: 4 mesgs > > router 4: 4 mesgs > > router 5: 3 mesgs > > > > Hence the maximum number of LSA messages generated in topology = 17 > > ie: sum of all LSA generated > > > > Router 1 will receive these messages from 2 links: > > Messages received by router 1 = 2*17 + 2*17 > > The first term 2*17 = for each link attached to router 1 ie: 2 and 3 > > Second term 2*17 = for the ACK's it receives for the messages sent from 2 > > and 3. > > > > Hence total message received = 68 > > Total messages sent = 68 ( 17 messages on two links, 17 acks on two links) > > > > Hence if your message sent/received does not match this number, please > > specifiy in your README file, why it does not. > > > > alefiya > > > > > > On Wed, 6 Mar 2002, qcao wrote: > > > > > Hi, > > > > > > In stage 1, we output the converging process of routing tables. Do we need > > > to match the nubmer of message send/receive with the sample output? I > > > think different implementations will make this different. Or all we need > > > to match is the final route table? > > > > > > Thanks, > > > -Qun > > > > > > From johnh@ISI.EDU Sat Mar 9 08:57:24 2002 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 g29GvOR28286 for ; Sat, 9 Mar 2002 08:57:24 -0800 (PST) Received: from avocet.prod.itd.earthlink.net (avocet.mail.pas.earthlink.net [207.217.120.50]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g29GvOt17500; Sat, 9 Mar 2002 08:57:24 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by avocet.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16jk9r-0005je-00; Sat, 09 Mar 2002 08:57:23 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g29Grn506428; Sat, 9 Mar 2002 08:53:49 -0800 Message-Id: <200203091653.g29Grn506428@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: qcao Cc: Alefiya Hussain , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] stage 1 output match In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Sat, 09 Mar 2002 08:53:49 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Fri, 08 Mar 2002 15:59:04 PST, qcao wrote: >Thanks very much for your specification. > >But I am thinking an optimization here: for the first LSA each router >send, it only indicates itself is up, no any information for the SPF >algorithm actually; so I think this message should not be flood forwarding >by it's neighbors because other nodes in the graph don't need this >information. >Any comments on it? That optimization is allowed, but not at all required. While the the LINK information in a zero-link LSA message is only of use to adjacent routers, the ROUTE information is potentially of use to all routers, so there are reasons to forward these messages. -John Heidemann From johnh@ISI.EDU Sat Mar 9 09:19:57 2002 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 g29HJvR01966 for ; Sat, 9 Mar 2002 09:19:57 -0800 (PST) Received: from snipe.prod.itd.earthlink.net (snipe.mail.pas.earthlink.net [207.217.120.62]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g29HJvt21665 for ; Sat, 9 Mar 2002 09:19:57 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by snipe.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16jkVg-0002jD-00; Sat, 09 Mar 2002 09:19:56 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g29HGPm06524; Sat, 9 Mar 2002 09:16:25 -0800 Message-Id: <200203091716.g29HGPm06524@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: qcao Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] still on LSA message In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Sat, 09 Mar 2002 09:16:25 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Fri, 08 Mar 2002 16:07:57 PST, qcao wrote: > > >On Thu, 7 Mar 2002, John Heidemann wrote: > >> On Wed, 06 Mar 2002 11:46:43 PST, Alefiya Hussain wrote: >> > >> > >> >On Wed, 6 Mar 2002, qcao wrote: >> > >> >> Hi, >> >> >> >> Sorry if someone mentioned that, but I am wondering that when routers send >> >> LSA message, could we use fixed length struct instead of varial length >> >> string ? It could be much easier if we just pass the struct pointer in >> >> sendto(). >> > >> >depending on how many local routes/neighbouring routers are present, the >> >LSA is going to beof variable length. I am not sure how you could define >> >it as a fixed length structure ? >> >> More specifically, I don't see how you could possibly do this as a >> fixed-length structure and meet the message formatting requirements. >> >> (And even if there wasn't two variable length fields in the LSA >> message, you STILL can't count on a static C structure to get the >> right encoding. It is completely within the C standard to allow the >> compiler to re-order or add invisible padding fields to the >> structure. Almost ALL compilers will pad structure fields to 32-bit >> boundries because all modern CPUs either require it or work much much >> faster with aligned data.) >> >> Part of the goal of the assignment is that you review how data is >> encoded when it's sent in fixed-format messages like in other network >> protocols! > > >Sorry this whole problem may seems silly. What I was thinking is having a >struct LSA in my program, which has two fixed arrays having the size of a >pre-defined MAX_NEBRS ( max prossible neighbors in the graph), so the LSA >will be fixed size structure padded with zeros, and I could send the >structure to the other routers, then they could figure out the data based >on the layout of the structure. That representation is acceptable for an INTERNAL data structure. (Although it is not ideal... how do you know how to set MAX_NEBRS? In C++ w/STL auto-sizing vectors or lists are better for that reason.) However, that representation is NOT appropriate for the external (on-the-wire) format of an LSA message, since it would not match the format given in the assignment for the reasons described agove. >> >> Or if we need to copy the LSA to a buffer, how to copy a 16 bits short int >> >> to a string, I tried some ways but all failed. >> > >> >did you try uding sprintf ? >> >> So sprintf will convert int of varying sizes to strings, >> but the LSA message is NOT a string, it's binary. > >I tried sprintf and it doesn't work, just because it is string? >So we are not using sendto (socket, char * , length, , , ), but which >structure other than string could accomodate vary-sized binary bits? "char*" can mean two things, either a string of characters, OR an array of 8-bit bytes of binary data. Consider the number 258 (base 10). In hex, this is 0x0102, and as a binary number in the computer occupies memory as either: byte 0: 0x01 byte 1: 0x02 OR byte 0: 0x02 byte 1: 0x01 depending on what kind of processor you're using (big-endian or little-endian). As a STRING it might be rendered "258\0" and would occupy memory: byte 0: 0x32 byte 1: 0x35 byte 2: 0x38 byte 0: 0x00 sprintf will map numbers into a string representation. The LSA format should be a binary representation, so sprintf is not appliciable. Instead you need to use C's bitwise operators &, |, <<, >>, etc. If this is not clear, you may want to consult your computer architecture textbook about data representation. -John Heidemann From johnh@ISI.EDU Sat Mar 9 09:22:11 2002 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 g29HMBR02085 for ; Sat, 9 Mar 2002 09:22:11 -0800 (PST) Received: from harrier.prod.itd.earthlink.net (harrier.mail.pas.earthlink.net [207.217.120.12]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g29HMAt21963 for ; Sat, 9 Mar 2002 09:22:11 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by harrier.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16jkXq-0002yR-00; Sat, 09 Mar 2002 09:22:10 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g29HIek06552; Sat, 9 Mar 2002 09:18:40 -0800 Message-Id: <200203091718.g29HIek06552@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Parag Samdadiya Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Confusion in output files to generate In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Sat, 09 Mar 2002 09:18:40 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Fri, 08 Mar 2002 20:54:40 PST, Parag Samdadiya wrote: > >Does this imply that if there are external routes then >the error rate would be zero ?? No, I explicitly wrote it to NOT imply that. >Otherwise we would have to generate both s2 and s3 ?? No, I explicitly wrote it to NOT imply that. Any given run only outputs two files: s0 and one of s1/s2/s3. If you follow the rules a, b, c in order it should be clear which to produce. -John Heidemann >> 1. always generate s0 >> 2. then determine which of s1, s2, s3 you should write >> based on the config file: >> a. if there are any external routes, write only s3 >> b. if the loss model is non-zero, write only s2 >> c. if neither a or b holds, write only s1 >> >> You should always produce s0 and ONLY ONE of s1/s2/s3. From Srihari@USC.Edu Sat Mar 9 12:05:35 2002 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 g29K5YR27223 for ; Sat, 9 Mar 2002 12:05:35 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g29K5Yt15375; Sat, 9 Mar 2002 12:05:34 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id MAA24182; Sat, 9 Mar 2002 12:05:56 -0800 (PST) Received: from USC.Edu (koh-sun006.usc.edu [128.125.5.186]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g29K5Nl06810; Sat, 9 Mar 2002 12:05:23 -0800 (PST) Message-ID: <3C8A6B2E.5000508@USC.Edu> Date: Sat, 09 Mar 2002 12:06:06 -0800 From: Prithvi Srihari Organization: The University of Southern California User-Agent: Mozilla/5.0 (X11; U; SunOS 5.8 sun4u; en-US; m18) Gecko/20010307 Netscape6/6.01 X-Accept-Language: en MIME-Version: 1.0 To: John Heidemann CC: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] still on LSA message References: <200203091716.g29HGPm06524@dash.isi.edu> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: But if we use sprintf %x, 258 will be rendered as "102\0" which is what we want. We can ignore (skip) the final null. John Heidemann wrote: > Consider the number 258 (base 10). > In hex, this is 0x0102, > and as a binary number in the computer occupies memory as either: > byte 0: 0x01 > byte 1: 0x02 > OR > byte 0: 0x02 > byte 1: 0x01 > depending on what kind of processor you're using (big-endian > or little-endian). > > As a STRING it might be rendered "258\0" > and would occupy memory: > byte 0: 0x32 > byte 1: 0x35 > byte 2: 0x38 > byte 0: 0x00 > > sprintf will map numbers into a string representation. > > The LSA format should be a binary representation, > so sprintf is not appliciable. > > Instead you need to use C's bitwise operators &, |, <<, >>, etc. > > If this is not clear, you may want to consult your computer > architecture textbook about data representation. > > -John Heidemann. From alsaeed@usc.edu Sat Mar 9 13:45:43 2002 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 g29LjhR12697 for ; Sat, 9 Mar 2002 13:45:43 -0800 (PST) Received: from lsmls02.we.mediaone.net (lsmls02.we.mediaone.net [24.130.1.15]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g29Ljht28796 for ; Sat, 9 Mar 2002 13:45:43 -0800 (PST) Received: from mss (c-24-130-200-13.we.mediaone.net [24.130.200.13]) by lsmls02.we.mediaone.net (8.11.4/8.11.3) with SMTP id g29Lhcm12945 for ; Sat, 9 Mar 2002 13:43:38 -0800 (PST) Message-ID: <002201c1c7b4$0ef1c190$0dc88218@we.mediaone.net> From: "MSS" To: References: <200203091716.g29HGPm06524@dash.isi.edu> Subject: Re: [Csci551-talk] still on LSA message Date: Sat, 9 Mar 2002 13:47:48 -0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: just for the record, if you don't want to use bitwise ops, you have other options. 1. typecasting: if you want to put a short int y (2 bytes) into buffer[x] (where buffer has at least x bytes reserved) then you can do this: (short int &) buffer[x] = y and you can take it back the same way using typecasting. However, this might be slower than bitwise ops and would create bus errors (byte alignment problems) on some processors. 2. direct memory copying: if you want to do the same as in 1, do it this way: memcpy(buffer+x, &y, 2) and you can take it back also using memcpy. This worked on the lab machines and I guess would work everywhere. It also might be slower than bitwise ops. Hope I helped. -alsaeed ----- Original Message ----- From: "John Heidemann" To: "qcao" Cc: Sent: Saturday, March 09, 2002 9:16 AM Subject: Re: [Csci551-talk] still on LSA message ..... > > "char*" can mean two things, either a string of characters, > OR an array of 8-bit bytes of binary data. > > Consider the number 258 (base 10). > In hex, this is 0x0102, > and as a binary number in the computer occupies memory as either: > byte 0: 0x01 > byte 1: 0x02 > OR > byte 0: 0x02 > byte 1: 0x01 > depending on what kind of processor you're using (big-endian > or little-endian). > > As a STRING it might be rendered "258\0" > and would occupy memory: > byte 0: 0x32 > byte 1: 0x35 > byte 2: 0x38 > byte 0: 0x00 > > sprintf will map numbers into a string representation. > > The LSA format should be a binary representation, > so sprintf is not appliciable. > > Instead you need to use C's bitwise operators &, |, <<, >>, etc. > > If this is not clear, you may want to consult your computer > architecture textbook about data representation. > > -John Heidemann From samdadiy@usc.edu Sat Mar 9 15:26:55 2002 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 g29NQsR27251 for ; Sat, 9 Mar 2002 15:26:55 -0800 (PST) Received: from usc-fs1.usc.edu (usc-fs1.usc.edu [128.125.150.4]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g29NQst12830 for ; Sat, 9 Mar 2002 15:26:54 -0800 (PST) Received: from sal-sun023.usc.edu (sal-sun023.usc.edu [128.125.5.93]) by usc-fs1.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g29NRUD09207 for ; Sat, 9 Mar 2002 15:27:30 -0800 (PST) Received: from localhost (samdadiy@localhost) by sal-sun023.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g29NOOx07003 for ; Sat, 9 Mar 2002 15:24:24 -0800 (PST) Date: Sat, 9 Mar 2002 15:24:24 -0800 (PST) From: Parag Samdadiya To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] Message count. Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: I am sorry if this is a repeat question. But does acks also account for messages sent ???? Parag From gina@lanl.gov Sat Mar 9 15:46:32 2002 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 g29NkVR29864 for ; Sat, 9 Mar 2002 15:46:31 -0800 (PST) Received: from mailrelay1.lanl.gov (mailrelay1.lanl.gov [128.165.4.101]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g29NkVt14440 for ; Sat, 9 Mar 2002 15:46:31 -0800 (PST) Received: from merilahna.lanl.gov (localhost.localdomain [127.0.0.1]) by mailrelay1.lanl.gov (8.11.6/8.11.6/(ccn-5)) with ESMTP id g29NkTQ00375 for ; Sat, 9 Mar 2002 16:46:29 -0700 Received: by merilahna.lanl.gov (Postfix, from userid 1000) id C57674BBCF; Tue, 13 Mar 2001 22:55:20 -0700 (MST) Received: from localhost (localhost [127.0.0.1]) by merilahna.lanl.gov (Postfix) with ESMTP id 4EBB643A84; Tue, 13 Mar 2001 22:55:20 -0700 (MST) Date: Tue, 13 Mar 2001 22:55:20 -0700 (MST) From: Gina Fisk To: Parag Samdadiya Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] Message count. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: > But does acks also account for messages sent ???? The project says that this field is "any message sent in any way, including forwarding", so I would say that yes, and ACK does count. Gina _______________________________________ Gina Fisk Los Alamos National Laboratory CCS-1 Advanced Computing Group, MS D451 Los Alamos, New Mexico 87545 (work) 505-667-6769, (fax) 505-665-4934 From johnh@ISI.EDU Sat Mar 9 16:41:47 2002 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 g2A0flR08004 for ; Sat, 9 Mar 2002 16:41:47 -0800 (PST) Received: from avocet.prod.itd.earthlink.net (avocet.mail.pas.earthlink.net [207.217.120.50]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A0flt20867 for ; Sat, 9 Mar 2002 16:41:47 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by avocet.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16jrPG-0007Py-00; Sat, 09 Mar 2002 16:41:47 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g2A0cG308020; Sat, 9 Mar 2002 16:38:16 -0800 Message-Id: <200203100038.g2A0cG308020@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Prithvi Srihari Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] still on LSA message In-reply-to: <3C8A6B2E.5000508@USC.Edu> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Sat, 09 Mar 2002 16:38:16 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sat, 09 Mar 2002 12:06:06 PST, Prithvi Srihari wrote: >But if we use sprintf %x, 258 will be rendered as "102\0" which is what >we want. We can ignore (skip) the final null. No, that's not correct. You've put the string in hex rather than decimal, but it's still a string of characters, not 16-bits of binary data. Sprintf %x creates the STRING "102\0", which in memory is: byte 0: 0x31 byte 1: 0x30 byte 2: 0x32 byte 3: 0x00 (4 bytes, long, each byte the ASCII character for 1, 0, and 2, and a null). Even if you ingore the null, (a) it's too long, 3 bytes, not 2, (b) it's not the right format: it's ASCII characters and not binary data. Sprintf is not the right tool to use. When you want the bytes to contain 0x01 and 0x02, the binary data, consuming two bytes as described below. One way to think about this is by analogy to source code and executables. A .c file has source code in it. An executable as machine instructions. On one hand, they're the same content...both are instructions about how to do something. On the other hand, they're NOT the same at all. A human can read the .c file, a computer can run the executable, and not the other way around. Similarly, strings are printable and binary data is directly usable by the computer. This assignment requires that you encode the message in binary. -John Heidemann > >John Heidemann wrote: > >> Consider the number 258 (base 10). >> In hex, this is 0x0102, >> and as a binary number in the computer occupies memory as either: >> byte 0: 0x01 >> byte 1: 0x02 >> OR >> byte 0: 0x02 >> byte 1: 0x01 >> depending on what kind of processor you're using (big-endian >> or little-endian). >> >> As a STRING it might be rendered "258\0" >> and would occupy memory: >> byte 0: 0x32 >> byte 1: 0x35 >> byte 2: 0x38 >> byte 0: 0x00 >> >> sprintf will map numbers into a string representation. >> >> The LSA format should be a binary representation, >> so sprintf is not appliciable. >> >> Instead you need to use C's bitwise operators &, |, <<, >>, etc. >> >> If this is not clear, you may want to consult your computer >> architecture textbook about data representation. >> >> -John Heidemann. > From johnh@ISI.EDU Sat Mar 9 16:46:46 2002 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 g2A0kkR08637 for ; Sat, 9 Mar 2002 16:46:46 -0800 (PST) Received: from pintail.mail.pas.earthlink.net (pintail.mail.pas.earthlink.net [207.217.120.122]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A0kjt21435 for ; Sat, 9 Mar 2002 16:46:45 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by pintail.mail.pas.earthlink.net with esmtp (Exim 3.33 #1) id 16jrU5-0005yn-00; Sat, 09 Mar 2002 16:46:45 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g2A0hEY08038; Sat, 9 Mar 2002 16:43:14 -0800 Message-Id: <200203100043.g2A0hEY08038@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: "MSS" Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] still on LSA message In-reply-to: <002201c1c7b4$0ef1c190$0dc88218@we.mediaone.net> MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Sat, 09 Mar 2002 16:43:14 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sat, 09 Mar 2002 13:47:48 PST, "MSS" wrote: >just for the record, if you don't want to use bitwise ops, you have other >options. > >1. typecasting: >if you want to put a short int y (2 bytes) into buffer[x] (where buffer has >at least x bytes reserved) then you can do this: > >(short int &) buffer[x] = y > >and you can take it back the same way using typecasting. However, this might >be slower than bitwise ops and would create bus errors (byte alignment >problems) on some processors. While this may work, this will NOT produce portable C code and WILL produce bus errors on many computers. I discourage this particular approach. >2. direct memory copying: >if you want to do the same as in 1, do it this way: > >memcpy(buffer+x, &y, 2) > >and you can take it back also using memcpy. This worked on the lab machines >and I guess would work everywhere. It also might be slower than bitwise ops. This approach is much better in that it IS portable. The downside (relative to using C's bitwise operators) is that it writes the data out in processor-specific order, thus your program would not work on architectures that have different encodings (for example, in a mixed network of SPARCs and x86 PCs). However, the assignment does NOT require that your app run on mixed architectures, so at least for the assignment, this disadvantage is not a problem. -John Heidemann From gina@lanl.gov Sat Mar 9 17:36:29 2002 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 g2A1aSR16657 for ; Sat, 9 Mar 2002 17:36:29 -0800 (PST) Received: from mailrelay3.lanl.gov (mailrelay3.lanl.gov [128.165.3.1]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A1aSt27126; Sat, 9 Mar 2002 17:36:28 -0800 (PST) Received: from merilahna.lanl.gov (localhost.localdomain [127.0.0.1]) by mailrelay3.lanl.gov (8.11.6/8.11.6/(ccn-5)) with ESMTP id g2A1aRv07196; Sat, 9 Mar 2002 18:36:27 -0700 Received: by merilahna.lanl.gov (Postfix, from userid 1000) id E27CF4BBCF; Sat, 9 Mar 2002 18:36:24 -0700 (MST) Received: from localhost (localhost [127.0.0.1]) by merilahna.lanl.gov (Postfix) with ESMTP id BCA5A43A84; Sat, 9 Mar 2002 18:36:24 -0700 (MST) Date: Sat, 9 Mar 2002 18:36:24 -0700 (MST) From: Gina Fisk To: John Heidemann Cc: csci551-talk@ISI.EDU In-Reply-To: <200203100043.g2A0hEY08038@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] shortest path using Dijkstra? Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: In section 3 of the project description it says that the first priority for routes is the MED rather than path length. But, Dijkstra's algorithm works because path lengths are cumulative. It doesn't appear that MED discriminator are cumulative. Therefore, it doesn't seem like you can use Dijkstra's algorithm to compute the routes. So, my question is, can we really used Dijkstra's algorithm for this? I don't think it is possible, but if I am wrong, please let me know why. Are we supposed to use Dijkstra only after checking for a dominating MED value? Or, can we use a different algorithm to do this that may be simpler? Also, it is not mentioned if the MED is prioritized by highest or lowest values. Please clarify. Thanks... Gina _______________________________________ Gina Fisk Los Alamos National Laboratory CCS-1 Advanced Computing Group, MS D451 Los Alamos, New Mexico 87545 (work) 505-667-6769, (fax) 505-665-4934 From ragarwal@usc.edu Sat Mar 9 17:53:08 2002 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 g2A1r8R18979 for ; Sat, 9 Mar 2002 17:53:08 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A1r7t28667 for ; Sat, 9 Mar 2002 17:53:08 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id RAA25619; Sat, 9 Mar 2002 17:51:19 -0800 (PST) Received: from localhost (ragarwal@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2A1pYF07068; Sat, 9 Mar 2002 17:51:34 -0800 (PST) Date: Sat, 9 Mar 2002 17:51:34 -0800 (PST) From: ragarwal To: Gina Fisk cc: Subject: Re: [Csci551-talk] shortest path using Dijkstra? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Gina: I think you have to use Dijkstra's algo for finding shortest path to 'internal' routers. Use MED values for external routes. This means you dont have to use Dijkstra for external routes ... just some simple if else statements will do. Also MED is prioritized by lowest values... (refer class notes) Rohit. On Sat, 9 Mar 2002, Gina Fisk wrote: > > In section 3 of the project description it says that the first priority > for routes is the MED rather than path length. But, Dijkstra's algorithm > works because path lengths are cumulative. It doesn't appear that MED > discriminator are cumulative. Therefore, it doesn't seem like you can use > Dijkstra's algorithm to compute the routes. So, my question is, can we > really used Dijkstra's algorithm for this? I don't think it is possible, > but if I am wrong, please let me know why. Are we supposed to use > Dijkstra only after checking for a dominating MED value? Or, can we use a > different algorithm to do this that may be simpler? > > Also, it is not mentioned if the MED is prioritized by highest or lowest > values. Please clarify. > > Thanks... > > Gina > _______________________________________ > Gina Fisk > Los Alamos National Laboratory > CCS-1 Advanced Computing Group, MS D451 > Los Alamos, New Mexico 87545 > (work) 505-667-6769, (fax) 505-665-4934 > From gina@lanl.gov Sat Mar 9 17:58:22 2002 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 g2A1wMR19598 for ; Sat, 9 Mar 2002 17:58:22 -0800 (PST) Received: from mailrelay3.lanl.gov (mailrelay3.lanl.gov [128.165.3.1]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A1wLt29090 for ; Sat, 9 Mar 2002 17:58:22 -0800 (PST) Received: from merilahna.lanl.gov (localhost.localdomain [127.0.0.1]) by mailrelay3.lanl.gov (8.11.6/8.11.6/(ccn-5)) with ESMTP id g2A1wKv07936 for ; Sat, 9 Mar 2002 18:58:20 -0700 Received: by merilahna.lanl.gov (Postfix, from userid 1000) id F39AB4BBCF; Sat, 9 Mar 2002 18:58:18 -0700 (MST) Received: from localhost (localhost [127.0.0.1]) by merilahna.lanl.gov (Postfix) with ESMTP id 8238943A84; Sat, 9 Mar 2002 18:58:18 -0700 (MST) Date: Sat, 9 Mar 2002 18:58:18 -0700 (MST) From: Gina Fisk To: ragarwal Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] shortest path using Dijkstra? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: You have to use some sort of shortest path algorithm after looking at the MED values in part 2 or you won't have the shortest path, right? So, I believe you have to use Dijkstra after you look at the MED values. But, it says to just use Dijkstra, which isn't possible. Gina On Sat, 9 Mar 2002, ragarwal wrote: > Gina: > I think you have to use Dijkstra's algo for finding shortest path to > 'internal' routers. Use MED values for external routes. This means you > dont have to use Dijkstra for external routes ... just some simple if else > statements will do. > Also MED is prioritized by lowest values... (refer class notes) > > Rohit. > > > > > On Sat, 9 Mar 2002, Gina Fisk wrote: > > > > > In section 3 of the project description it says that the first priority > > for routes is the MED rather than path length. But, Dijkstra's algorithm > > works because path lengths are cumulative. It doesn't appear that MED > > discriminator are cumulative. Therefore, it doesn't seem like you can use > > Dijkstra's algorithm to compute the routes. So, my question is, can we > > really used Dijkstra's algorithm for this? I don't think it is possible, > > but if I am wrong, please let me know why. Are we supposed to use > > Dijkstra only after checking for a dominating MED value? Or, can we use a > > different algorithm to do this that may be simpler? > > > > Also, it is not mentioned if the MED is prioritized by highest or lowest > > values. Please clarify. > > > > Thanks... > > > > Gina > > _______________________________________ > > Gina Fisk > > Los Alamos National Laboratory > > CCS-1 Advanced Computing Group, MS D451 > > Los Alamos, New Mexico 87545 > > (work) 505-667-6769, (fax) 505-665-4934 > > > _______________________________________ Gina Fisk Los Alamos National Laboratory CCS-1 Advanced Computing Group, MS D451 Los Alamos, New Mexico 87545 (work) 505-667-6769, (fax) 505-665-4934 From samdadiy@usc.edu Sat Mar 9 18:00:32 2002 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 g2A20WR20014 for ; Sat, 9 Mar 2002 18:00:32 -0800 (PST) Received: from usc-fs1.usc.edu (usc-fs1.usc.edu [128.125.150.4]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A20Vt29250; Sat, 9 Mar 2002 18:00:31 -0800 (PST) Received: from sal-sun023.usc.edu (sal-sun023.usc.edu [128.125.5.93]) by usc-fs1.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2A217D09733; Sat, 9 Mar 2002 18:01:07 -0800 (PST) Received: from localhost (samdadiy@localhost) by sal-sun023.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2A1vU508271; Sat, 9 Mar 2002 17:57:30 -0800 (PST) Date: Sat, 9 Mar 2002 17:57:30 -0800 (PST) From: Parag Samdadiya To: Gina Fisk cc: John Heidemann , Subject: Re: [Csci551-talk] shortest path using Dijkstra? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sat, 9 Mar 2002, Gina Fisk wrote: > but if I am wrong, please let me know why. Are we supposed to use > Dijkstra only after checking for a dominating MED value? Or, can we use a > different algorithm to do this that may be simpler? Well, if you want to run a single function to resolve both internal and external routes, then set med value for internal nodes to 0. Automatically they will tie at this value and and your algo will select the shortest path. To figure out the shortest path, you need the Dijkstras Algo. > > Also, it is not mentioned if the MED is prioritized by highest or lowest > values. Please clarify. Lower MED values are prefered. TA, please clarify if otherwise. Parag From qcao@usc.edu Sat Mar 9 18:28:56 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2A2SuR23933 for ; Sat, 9 Mar 2002 18:28:56 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id SAA06242 for ; Sat, 9 Mar 2002 18:29:29 -0800 (PST) Received: from localhost (qcao@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2A2Thf20293 for ; Sat, 9 Mar 2002 18:29:43 -0800 (PST) Date: Sat, 9 Mar 2002 18:29:43 -0800 (PST) From: qcao To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] strange problem (wondering packets? ) Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, I got some wierd output when debugging flooding. Unfortuntely, I forgot to set the if (ttl=0), stop flooding condition, maybe it makes some packets wondering on the network as professor said in the class. The output is wierd, say I set some sequence number 100, 200,300,.. for routers, but my routers get packets with sequence number 43555, 98593, etc. Basically I could not use my problem anymore! But I am not sure if this is caused by packets in network, because I tried quit all processes, log off, even change into aludra, all failed. Also it is obviously that my UDP ports are keep changing, and my program will not forward out-dated packets. So I am wondering how could these packets still in the network, and how could they get into my program again? I am totally confused here. Or there are other reasons for those strange sequence numbers? Regards, -Qun From qcao@usc.edu Sat Mar 9 18:40:18 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2A2eHR25301 for ; Sat, 9 Mar 2002 18:40:17 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id SAA09785 for ; Sat, 9 Mar 2002 18:40:51 -0800 (PST) Received: from localhost (qcao@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2A2exi03162 for ; Sat, 9 Mar 2002 18:40:59 -0800 (PST) Date: Sat, 9 Mar 2002 18:40:59 -0800 (PST) From: qcao To: Subject: Re: [Csci551-talk] strange problem (wondering packets? ) In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Sorry, I meant wandering, maybe I was just wondering too much. :) On Sat, 9 Mar 2002, qcao wrote: > Hi, > > I got some wierd output when debugging flooding. Unfortuntely, I forgot to > set the if (ttl=0), stop flooding condition, maybe it makes some > packets wondering on the network as professor said in the class. The > output is wierd, say I set some sequence number 100, 200,300,.. for > routers, but my routers get packets with sequence number 43555, 98593, > etc. Basically I could not use my problem anymore! > > But I am not sure if this is caused by packets in network, because I tried > quit all processes, log off, even change into aludra, all failed. Also it > is obviously that my UDP ports are keep changing, and my program will > not forward out-dated packets. So I am wondering how could these packets > still in the network, and how could they get into my program again? > I am totally confused here. > > Or there are other reasons for those strange sequence numbers? > > Regards, > -Qun > From ganglu@usc.edu Sat Mar 9 18:49:17 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2A2nHR27014 for ; Sat, 9 Mar 2002 18:49:17 -0800 (PST) Received: from aludra.usc.edu (aludra.usc.edu [128.125.253.184]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id SAA11657; Sat, 9 Mar 2002 18:47:34 -0800 (PST) Received: from localhost (ganglu@localhost) by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2A2lk726771; Sat, 9 Mar 2002 18:47:46 -0800 (PST) Date: Sat, 9 Mar 2002 18:47:46 -0800 (PST) From: ganglu To: qcao cc: Subject: Re: [Csci551-talk] strange problem (wondering packets? ) In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: It maybe caused by Byte alignment, order in the "struct" data type. You had better make all data in your "struct" occupy 4 bytes. On Sat, 9 Mar 2002, qcao wrote: > Hi, > > I got some wierd output when debugging flooding. Unfortuntely, I forgot to > set the if (ttl=0), stop flooding condition, maybe it makes some > packets wondering on the network as professor said in the class. The > output is wierd, say I set some sequence number 100, 200,300,.. for > routers, but my routers get packets with sequence number 43555, 98593, > etc. Basically I could not use my problem anymore! > > But I am not sure if this is caused by packets in network, because I tried > quit all processes, log off, even change into aludra, all failed. Also it > is obviously that my UDP ports are keep changing, and my program will > not forward out-dated packets. So I am wondering how could these packets > still in the network, and how could they get into my program again? > I am totally confused here. > > Or there are other reasons for those strange sequence numbers? > > Regards, > -Qun > From hussain@ISI.EDU Sat Mar 9 20:10:31 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2A4AUR08309 for ; Sat, 9 Mar 2002 20:10:30 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g2A4AQ000071; Sat, 9 Mar 2002 20:10:26 -0800 (PST) Date: Sat, 9 Mar 2002 20:10:26 -0800 (PST) From: Alefiya Hussain To: Parag Samdadiya cc: Gina Fisk , John Heidemann , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] shortest path using Dijkstra? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sat, 9 Mar 2002, Parag Samdadiya wrote: > > > On Sat, 9 Mar 2002, Gina Fisk wrote: > > > but if I am wrong, please let me know why. Are we supposed to use > > Dijkstra only after checking for a dominating MED value? Or, can we use a > > different algorithm to do this that may be simpler? > > Well, if you want to run a single function to resolve both > internal and external routes, then set med value for internal > nodes to 0. Automatically they will tie at this value and > and your algo will select the shortest path. To figure out > the shortest path, you need the Dijkstras Algo. > > > > > Also, it is not mentioned if the MED is prioritized by highest or lowest > > values. Please clarify. > > Lower MED values are prefered. TA, please clarify if otherwise. > > Parag > Like Rohit and Parag have already mentioned, lower MED values are prefered. alefiya From gina@lanl.gov Sat Mar 9 20:12:45 2002 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 g2A4CjR08703 for ; Sat, 9 Mar 2002 20:12:45 -0800 (PST) Received: from mailrelay1.lanl.gov (mailrelay1.lanl.gov [128.165.4.101]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A4Cit14033; Sat, 9 Mar 2002 20:12:44 -0800 (PST) Received: from merilahna.lanl.gov (localhost.localdomain [127.0.0.1]) by mailrelay1.lanl.gov (8.11.6/8.11.6/(ccn-5)) with ESMTP id g2A4CgQ09382; Sat, 9 Mar 2002 21:12:42 -0700 Received: by merilahna.lanl.gov (Postfix, from userid 1000) id CE0404BBCF; Sat, 9 Mar 2002 21:12:30 -0700 (MST) Received: from localhost (localhost [127.0.0.1]) by merilahna.lanl.gov (Postfix) with ESMTP id 59DF443A84; Sat, 9 Mar 2002 21:12:30 -0700 (MST) Date: Sat, 9 Mar 2002 21:12:30 -0700 (MST) From: Gina Fisk To: Alefiya Hussain Cc: Parag Samdadiya , John Heidemann , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] shortest path using Dijkstra? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: > Like Rohit and Parag have already mentioned, lower MED values are > prefered. Yes, but you can't use Dijkstra to do what is asked of us in Part 2 (with external routes). How do you want us to handle this? Gina _______________________________________ Gina Fisk Los Alamos National Laboratory CCS-1 Advanced Computing Group, MS D451 Los Alamos, New Mexico 87545 (work) 505-667-6769, (fax) 505-665-4934 From hussain@ISI.EDU Sat Mar 9 20:35:51 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2A4ZoR11911 for ; Sat, 9 Mar 2002 20:35:50 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g2A4ZlQ06709; Sat, 9 Mar 2002 20:35:47 -0800 (PST) Date: Sat, 9 Mar 2002 20:35:47 -0800 (PST) From: Alefiya Hussain Reply-To: Alefiya Hussain To: Gina Fisk cc: Parag Samdadiya , John Heidemann , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] shortest path using Dijkstra? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi Gina, one way do it is the following: When you receive a LSA with a route ( say route A) that has a MED value, find a path to the origin router that is responsible for that route using Djkstra's algo, provided the MED value for A is lower than what you have seen before. i hope this helps, alefiya On Sat, 9 Mar 2002, Gina Fisk wrote: > > > Like Rohit and Parag have already mentioned, lower MED values are > > prefered. > > Yes, but you can't use Dijkstra to do what is asked of us in Part 2 (with > external routes). How do you want us to handle this? > > Gina > _______________________________________ > Gina Fisk > Los Alamos National Laboratory > CCS-1 Advanced Computing Group, MS D451 > Los Alamos, New Mexico 87545 > (work) 505-667-6769, (fax) 505-665-4934 > From samdadiy@usc.edu Sat Mar 9 20:43:22 2002 Received: from usc.edu (root@usc.edu [128.125.253.136]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2A4hMR12897 for ; Sat, 9 Mar 2002 20:43:22 -0800 (PST) Received: from sal-sun023.usc.edu (sal-sun023.usc.edu [128.125.5.93]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id UAA12464; Sat, 9 Mar 2002 20:43:54 -0800 (PST) Received: from localhost (samdadiy@localhost) by sal-sun023.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2A4hww09760; Sat, 9 Mar 2002 20:43:58 -0800 (PST) Date: Sat, 9 Mar 2002 20:43:58 -0800 (PST) From: Parag Samdadiya To: qcao cc: Subject: Re: [Csci551-talk] strange problem (wondering packets? ) In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Couple of places where you might want to check. See if you are reading the right location from the buffer. Means if you are receiving high sequence numbers, see if you are also actually sending those values. Just to ensure that you are not missing a byte here and there while transfering. Another thing is, where exactly are you checking for ttl ?? That should be the first thing to check once you receive a packet. May be you are first flooding the packet to your neighbours and then checking the ttl before you process it yourself. That way you will keep receiving the same packets again and again. Also you are not supposed to increment the sequence numbers of LSA messages received from others. You just need to change the ttl and last hop id before flooding it. May you are always increasing the seq number. That may be one reason. Also see if you are sending int and reading unsigned int. Because that might make a difference. Parag On Sat, 9 Mar 2002, qcao wrote: > Hi, > > I got some wierd output when debugging flooding. Unfortuntely, I forgot to > set the if (ttl=0), stop flooding condition, maybe it makes some > packets wondering on the network as professor said in the class. The > output is wierd, say I set some sequence number 100, 200,300,.. for > routers, but my routers get packets with sequence number 43555, 98593, > etc. Basically I could not use my problem anymore! > > But I am not sure if this is caused by packets in network, because I tried > quit all processes, log off, even change into aludra, all failed. Also it > is obviously that my UDP ports are keep changing, and my program will > not forward out-dated packets. So I am wondering how could these packets > still in the network, and how could they get into my program again? > I am totally confused here. > > Or there are other reasons for those strange sequence numbers? > > Regards, > -Qun > From gina@lanl.gov Sat Mar 9 20:44:11 2002 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 g2A4iBR12965 for ; Sat, 9 Mar 2002 20:44:11 -0800 (PST) Received: from mailrelay1.lanl.gov (mailrelay1.lanl.gov [128.165.4.101]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A4iAt17122; Sat, 9 Mar 2002 20:44:10 -0800 (PST) Received: from merilahna.lanl.gov (localhost.localdomain [127.0.0.1]) by mailrelay1.lanl.gov (8.11.6/8.11.6/(ccn-5)) with ESMTP id g2A4i8Q10683; Sat, 9 Mar 2002 21:44:09 -0700 Received: by merilahna.lanl.gov (Postfix, from userid 1000) id 948C44BBCF; Sat, 9 Mar 2002 21:43:56 -0700 (MST) Received: from localhost (localhost [127.0.0.1]) by merilahna.lanl.gov (Postfix) with ESMTP id 1D61543A84; Sat, 9 Mar 2002 21:43:56 -0700 (MST) Date: Sat, 9 Mar 2002 21:43:56 -0700 (MST) From: Gina Fisk To: Alefiya Hussain Cc: Parag Samdadiya , John Heidemann , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] shortest path using Dijkstra? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Okay, so we first need to sort the routes (based on MED), and then search with Dijkstra for the shortest path. (This is not clear in the project description since it says to use Dijkstra to find the shortest path based on three criteria, which isn't actually possible). Thanks for the clarification... Gina On Sat, 9 Mar 2002, Alefiya Hussain wrote: > Hi Gina, > > one way do it is the following: > When you receive a LSA with a route ( say route A) that has a MED value, > find a path to the origin router that is responsible for that route using > Djkstra's algo, provided the MED value for A is lower than what you have > seen before. > > i hope this helps, > alefiya > > On Sat, 9 Mar 2002, Gina Fisk wrote: > > > > > > Like Rohit and Parag have already mentioned, lower MED values are > > > prefered. > > > > Yes, but you can't use Dijkstra to do what is asked of us in Part 2 (with > > external routes). How do you want us to handle this? > > > > Gina > > _______________________________________ > > Gina Fisk > > Los Alamos National Laboratory > > CCS-1 Advanced Computing Group, MS D451 > > Los Alamos, New Mexico 87545 > > (work) 505-667-6769, (fax) 505-665-4934 > > > _______________________________________ Gina Fisk Los Alamos National Laboratory CCS-1 Advanced Computing Group, MS D451 Los Alamos, New Mexico 87545 (work) 505-667-6769, (fax) 505-665-4934 From johnh@ISI.EDU Sat Mar 9 22:09:58 2002 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 g2A69vR24673 for ; Sat, 9 Mar 2002 22:09:57 -0800 (PST) Received: from goose.prod.itd.earthlink.net (goose.mail.pas.earthlink.net [207.217.120.18]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A69vt27530; Sat, 9 Mar 2002 22:09:57 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by goose.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16jwWq-00008p-00; Sat, 09 Mar 2002 22:09:56 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g2A66P709202; Sat, 9 Mar 2002 22:06:25 -0800 Message-Id: <200203100606.g2A66P709202@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: Gina Fisk Cc: Alefiya Hussain , Parag Samdadiya , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] shortest path using Dijkstra? In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Sat, 09 Mar 2002 22:06:24 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sat, 09 Mar 2002 21:43:56 MST, Gina Fisk wrote: > >Okay, so we first need to sort the routes (based on MED), and then search >with Dijkstra for the shortest path. (This is not clear in the project >description since it says to use Dijkstra to find the shortest path based >on three criteria, which isn't actually possible). The assignment says you need to run Dijkstra's algorithm, and what the route prioritization is. It does not say that "Dijkstra's algorithm finds the shortest path based on the three criteria". As Alefiya suggested, you will need to do something more than just run Dijkstra's algorithm. -John Heidemann > >Thanks for the clarification... > >Gina > >On Sat, 9 Mar 2002, Alefiya Hussain wrote: > >> Hi Gina, >> >> one way do it is the following: >> When you receive a LSA with a route ( say route A) that has a MED value, >> find a path to the origin router that is responsible for that route using >> Djkstra's algo, provided the MED value for A is lower than what you have >> seen before. >> >> i hope this helps, >> alefiya >> >> On Sat, 9 Mar 2002, Gina Fisk wrote: >> >> > >> > > Like Rohit and Parag have already mentioned, lower MED values are >> > > prefered. >> > >> > Yes, but you can't use Dijkstra to do what is asked of us in Part 2 (with >> > external routes). How do you want us to handle this? >> > >> > Gina >> > _______________________________________ >> > Gina Fisk >> > Los Alamos National Laboratory >> > CCS-1 Advanced Computing Group, MS D451 >> > Los Alamos, New Mexico 87545 >> > (work) 505-667-6769, (fax) 505-665-4934 >> > >> > >_______________________________________ >Gina Fisk >Los Alamos National Laboratory >CCS-1 Advanced Computing Group, MS D451 >Los Alamos, New Mexico 87545 >(work) 505-667-6769, (fax) 505-665-4934 From johnh@ISI.EDU Sat Mar 9 22:14:17 2002 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 g2A6EHR25099 for ; Sat, 9 Mar 2002 22:14:17 -0800 (PST) Received: from gull.prod.itd.earthlink.net (gull.mail.pas.earthlink.net [207.217.120.84]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2A6EHt27761 for ; Sat, 9 Mar 2002 22:14:17 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by gull.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16jwb2-0006UD-00; Sat, 09 Mar 2002 22:14:16 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g2A6Air09239; Sat, 9 Mar 2002 22:10:44 -0800 Message-Id: <200203100610.g2A6Air09239@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: ganglu Cc: qcao , csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] strange problem (wondering packets? ) In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Sat, 09 Mar 2002 22:10:44 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sat, 09 Mar 2002 18:47:46 PST, ganglu wrote: > >It maybe caused by Byte alignment, order in the "struct" data type. >You had better make all data in your "struct" occupy 4 bytes. Although the assignment specifies the data "on-the-wire", and it is NOT aligned on 4-byte boundries. (Changing the LSA structure as it appears in your message to align it on 4-byte boundries will not get you full credit for the assignment.) But Ganglu has identified a likely problem. I'd suggest you check your marshalling (packing/unpacking) code. -John Heidemann > > >On Sat, 9 Mar 2002, qcao wrote: > >> Hi, >> >> I got some wierd output when debugging flooding. Unfortuntely, I forgot to >> set the if (ttl=0), stop flooding condition, maybe it makes some >> packets wondering on the network as professor said in the class. The >> output is wierd, say I set some sequence number 100, 200,300,.. for >> routers, but my routers get packets with sequence number 43555, 98593, >> etc. Basically I could not use my problem anymore! >> >> But I am not sure if this is caused by packets in network, because I tried >> quit all processes, log off, even change into aludra, all failed. Also it >> is obviously that my UDP ports are keep changing, and my program will >> not forward out-dated packets. So I am wondering how could these packets >> still in the network, and how could they get into my program again? >> I am totally confused here. >> >> Or there are other reasons for those strange sequence numbers? >> >> Regards, >> -Qun >> From qcao@usc.edu Sun Mar 10 09:19:07 2002 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 g2AHJ7R27739 for ; Sun, 10 Mar 2002 09:19:07 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2AHJ7t16497; Sun, 10 Mar 2002 09:19:07 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id JAA05694; Sun, 10 Mar 2002 09:19:38 -0800 (PST) Received: from localhost (qcao@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2AHJnu14528; Sun, 10 Mar 2002 09:19:49 -0800 (PST) Date: Sun, 10 Mar 2002 09:19:49 -0800 (PST) From: qcao To: John Heidemann cc: ganglu , Subject: Re: [Csci551-talk] strange problem (wondering packets? ) In-Reply-To: <200203100610.g2A6Air09239@dash.isi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: It turns out to be a small mistake in memcpy(). :) Thanks very much for professor and fellow students for helping people on weekend nights. -Qun On Sat, 9 Mar 2002, John Heidemann wrote: > On Sat, 09 Mar 2002 18:47:46 PST, ganglu wrote: > > > >It maybe caused by Byte alignment, order in the "struct" data type. > >You had better make all data in your "struct" occupy 4 bytes. > > Although the assignment specifies the data "on-the-wire", and it is > NOT aligned on 4-byte boundries. (Changing the LSA structure as it > appears in your message to align it on 4-byte boundries will not get > you full credit for the assignment.) > > But Ganglu has identified a likely problem. I'd suggest you check > your marshalling (packing/unpacking) code. > > -John Heidemann > > > > > > >On Sat, 9 Mar 2002, qcao wrote: > > > >> Hi, > >> > >> I got some wierd output when debugging flooding. Unfortuntely, I forgot to > >> set the if (ttl=0), stop flooding condition, maybe it makes some > >> packets wondering on the network as professor said in the class. The > >> output is wierd, say I set some sequence number 100, 200,300,.. for > >> routers, but my routers get packets with sequence number 43555, 98593, > >> etc. Basically I could not use my problem anymore! > >> > >> But I am not sure if this is caused by packets in network, because I tried > >> quit all processes, log off, even change into aludra, all failed. Also it > >> is obviously that my UDP ports are keep changing, and my program will > >> not forward out-dated packets. So I am wondering how could these packets > >> still in the network, and how could they get into my program again? > >> I am totally confused here. > >> > >> Or there are other reasons for those strange sequence numbers? > >> > >> Regards, > >> -Qun > >> > From qcao@usc.edu Sun Mar 10 09:26:36 2002 Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2AHQaR29059 for ; Sun, 10 Mar 2002 09:26:36 -0800 (PST) Received: from nunki.usc.edu (nunki.usc.edu [128.125.5.168]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id JAA08628 for ; Sun, 10 Mar 2002 09:27:08 -0800 (PST) Received: from localhost (qcao@localhost) by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2AHRI114688 for ; Sun, 10 Mar 2002 09:27:19 -0800 (PST) Date: Sun, 10 Mar 2002 09:27:18 -0800 (PST) From: qcao To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Subject: [Csci551-talk] is this assumption ok? Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: Hi, In the LSA info, I don't think routers could figure out which links mapping to which routes in the LSA received based on layout below: then for each local route: +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | address of route | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | prefix size | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | MED value | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ then for each link: +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | far-end router id | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Hence it is difficult to know the far-end routers' IP address based on the far-end router id, but the IP address is needed out the S1 output file. So I just assume the routers know the IP address-id mapping in the network by passing this info from manager, is that OK? -Qun From hussain@ISI.EDU Sun Mar 10 10:06:11 2002 Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2AI6AR04867 for ; Sun, 10 Mar 2002 10:06:10 -0800 (PST) Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g2AI6Ag13589; Sun, 10 Mar 2002 10:06:10 -0800 (PST) Date: Sun, 10 Mar 2002 10:06:10 -0800 (PST) From: Alefiya Hussain To: qcao cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] is this assumption ok? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: If you mean the manager is sending routerid-prefix pair information regarding all routers, to all other routers in the network, then thats totally incorrect. Please if needed, use an extra field in the LSA for the association of prefix to far-end router id. alefiya On Sun, 10 Mar 2002, qcao wrote: > Hi, > > In the LSA info, I don't think routers could figure out which > links mapping to which routes in the LSA received based on layout below: > > then for each local route: > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > | address of route | > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > | prefix size | > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > | MED value | > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > then for each link: > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > | far-end router id | > +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > > Hence it is difficult to know the far-end routers' IP address based on the > far-end router id, but the IP address is needed out the S1 output file. So > I just assume the routers know the IP address-id mapping in the network by > passing this info from manager, is that OK? > > -Qun > From johnh@ISI.EDU Sun Mar 10 18:36:03 2002 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 g2B2a2R19538 for ; Sun, 10 Mar 2002 18:36:02 -0800 (PST) Received: from snipe.prod.itd.earthlink.net (snipe.mail.pas.earthlink.net [207.217.120.62]) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2B2a2t02401 for ; Sun, 10 Mar 2002 18:36:02 -0800 (PST) Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu) by snipe.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16kFfN-0002lg-00; Sun, 10 Mar 2002 18:36:01 -0800 Received: from dash.isi.edu (johnh@localhost) by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g2B2WRi19271; Sun, 10 Mar 2002 18:32:28 -0800 Message-Id: <200203110232.g2B2WRi19271@dash.isi.edu> X-url: http://www.isi.edu/~johnh/ To: qcao Cc: csci551-talk@ISI.EDU Subject: Re: [Csci551-talk] is this assumption ok? In-reply-to: MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya") Content-Type: text/plain; charset=US-ASCII Date: Sun, 10 Mar 2002 18:32:26 -0800 From: John Heidemann Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: On Sun, 10 Mar 2002 09:27:18 PST, qcao wrote: >Hi, > >In the LSA info, I don't think routers could figure out which >links mapping to which routes in the LSA received based on layout below: > >then for each local route: >+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ >| address of route | >+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ >| prefix size | >+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ >| MED value | >+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ >then for each link: >+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ >| far-end router id | >+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > >Hence it is difficult to know the far-end routers' IP address based on the >far-end router id, but the IP address is needed out the S1 output file. So >I just assume the routers know the IP address-id mapping in the network by >passing this info from manager, is that OK? For this assignment, routers are not given IP addresses. (They do have routes they are responsible for, but these are not a single IP address.) It is NOT ok to pass additional information from the manager. The only info you're permitted to pass to each router is the ports and ids of the neighbors. I suggest you look at the part of the LSA you didn't quote, and think about how it relates to the part you did quote. Specifically, putting the two parts together, combined with flooding these messages to all routers, gives all routers the information you're asking for. -John Heidemann From Srihari@USC.Edu Sun Mar 17 01:27:18 2002 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 g2H9RIR22386 for ; Sun, 17 Mar 2002 01:27:18 -0800 (PST) Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged)) by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2H9RHt06694 for ; Sun, 17 Mar 2002 01:27:17 -0800 (PST) Received: from scf-fs.usc.edu (root@scf-fs.usc.edu [128.125.253.183]) by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP id BAA18905 for ; Sun, 17 Mar 2002 01:27:45 -0800 (PST) Received: from USC.Edu (srihari@aludra.usc.edu [128.125.253.184]) by scf-fs.usc.edu (8.10.1/8.10.1/usc) with ESMTP id g2H9RHl15194 for ; Sun, 17 Mar 2002 01:27:17 -0800 (PST) Message-ID: <3C9461A8.1020106@USC.Edu> Date: Sun, 17 Mar 2002 01:28:08 -0800 From: Prithvi Srihari Organization: The University of Southern California User-Agent: Mozilla/5.0 (X11; U; SunOS 5.8 sun4u; en-US; m18) Gecko/20010307 Netscape6/6.01 X-Accept-Language: en MIME-Version: 1.0 To: csci551-talk@ISI.EDU Content-Type: multipart/alternative; boundary="------------040807070208030009070407" Subject: [Csci551-talk] Seminar Sender: csci551-talk-admin@mailman.isi.edu Errors-To: csci551-talk-admin@mailman.isi.edu X-BeenThere: csci551-talk@mailman.isi.edu X-Mailman-Version: 2.0.5 Precedence: bulk List-Help: List-Archive: --------------040807070208030009070407 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Hi All, This will also be interesting: SC Networks Seminar Characterizing the Internet Hierarchy from Multiple Vantage Points Jennifer Rexford AT&T Research Labs Tuesday, March 19, 2002 2:00 PM EEB 248 -------- Original Message -------- Subject: [Csci551-talk] Amit Kumar networking talk today Date: Thu, 07 Mar 2002 09:07:37 -0800 From: John Heidemann To: csci551-talk@ISI.EDU By the way, USC has a number of good networking researchers visiting campus and giving talks. If you're interested in networking you might check some of these out. Amit Kumar's VPN work ties DIRECTLY in to the IntServe/DiffServe work we've been talking about, and his BGP work also sounds interesting. Talks from current researchers are particularly good to find out what's really current in networking research, as opposed to class which necessarily lags the state-of-the-art by a year or more. (Rumors that he'll be answering questions after his talk about link-state routing in 551 projects are unfortunately unfounded. :-) -John Heidemann --------------040807070208030009070407 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit 
Hi All,
This will also be interesting:

SC Networks Seminar


Characterizing the Internet Hierarchy from Multiple Vantage Points


Jennifer Rexford
AT&T Research Labs

Tuesday, March 19, 2002
2:00 PM
EEB 248
-------- Original Message --------
Subject: [Csci551-talk] Amit Kumar networking talk today
Date: Thu, 07 Mar 2002 09:07:37 -0800
From: John Heidemann <johnh@ISI.EDU>
To: csci551-talk@ISI.EDU
 


By the way, USC has a number of good networking researchers visiting
campus and giving talks.  If you're interested in networking you might
check some of these out.

Amit Kumar's VPN work ties DIRECTLY in to the IntServe/DiffServe work
we've been talking about, and his BGP work also sounds interesting.

Talks from current researchers are particularly good to find out
what's really current in networking research, as opposed to class
which necessarily lags the state-of-the-art by a year or more.

(Rumors that he'll be answering questions after his talk about
link-state routing in 551 projects are unfortunately unfounded. :-)

   -John Heidemann



--------------040807070208030009070407--


From tipu@usc.edu Thu Mar 21 01:27:01 2002
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 g2L9R0R11975
	for ; Thu, 21 Mar 2002 01:27:00 -0800 (PST)
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2L9R0t19532
	for ; Thu, 21 Mar 2002 01:27:00 -0800 (PST)
Received: from aludra.usc.edu (tipu@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id BAA23259 for ; Thu, 21 Mar 2002 01:27:26 -0800 (PST)
Received: from localhost (tipu@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g2L9RqM14987 for ; Thu, 21 Mar 2002 01:27:52 -0800 (PST)
Date: Thu, 21 Mar 2002 01:27:52 -0800 (PST)
From: tipu 
To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Attention!!!!   Grading project 1
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 


hi,

Please see the list...Those who have some kind of error with his/her name
please come to me immediately.....

The Time schedule is

9    - 11am    (SAL) --- 3/21/2002
9:30 - 11:30pm (SAL) --- 3/21/2002
5:30 - 8:30pm  (SAL) --- 3/22/2002

here is the list -------

Mar  9 2002 09:43:20 ganglu        done		C
Mar  9 2002 11:29:57 sundarra      mk pr	C
Mar  9 2002 16:38:34 jitaekli      done		C++
Mar  9 2002 22:04:11 rishisha      done 	C++/ program is actually C
Mar 10 2002 00:14:52 bhat	   done		C
Mar 10 2002 01:15:16 kaiyuxia      done		C
Mar 10 2002 01:42:29 chaoyuwa      mk pr	C++
Mar 10 2002 02:18:25 jngai 	   done		C
Mar 10 2002 02:39:12 alsaeed 	   done		C++
Mar 10 2002 05:56:05 aaj 	   mk pr	C
Mar 10 2002 06:51:35 davidmwe 	   done		C
Mar 10 2002 06:47:57 bhanot        mk er	C
Mar 10 2002 08:30:27 srangwal 	   done		C
Mar 10 2002 08:55:15 rkothari 	   seg flt	C++
Mar 10 2002 08:50:06 mprasann 	   done		C
Mar 10 2002 08:49:46 samdadiy 	   done		C++
Mar 10 2002 08:49:30 gurmeets 	   done		C
Mar 10 2002 08:47:39 kasat 	   parse flt	C
Mar 10 2002 09:16:27 anshulpa 	   parse flt	C
Mar 10 2002 21:13:35 ragarwal 	   seg flt	C
Mar 10 2002 22:42:42 ggoel 	   done		C
Mar 10 2002 23:24:13 nangunoo 	   pro hangs	C
Mar 11 2002 00:35:58 vbhargav 	   done		C
Mar 11 2002 00:30:58 phadke 	   seg flt	C
Mar 11 2002 03:48:14 bhave 	   mk pr	C
Mar 11 2002 04:53:01 vsj 	   seg flt	C
Mar 11 2002 05:03:27 renukaku 	   seg flt	C
Mar 11 2002 05:32:33 kagarwal 	   mk pr	C
Mar 11 2002 06:24:14 sameerme 	   mk pr	C
Mar 11 2002 06:17:27 jeongimp 	   done		C
Mar 11 2002 07:41:09 srihari 	   done		C
Mar 11 2002 07:36:13 sumitbha      err+seg flt  C
Mar 11 2002 07:58:04 sson 	   err		C
Mar 11 2002 07:47:45 bhuptani	   Berr+seg flt C
Mar 11 2002 07:59:46 woojinch	   Berr         C
Mar 11 2002 08:35:26 fisk          err		C
Mar 11 2002 08:56:58 rohera 	   seg flt	C


done   = ok
mk     = make file is not running
seg    = segmentation fault
Berr   = bus error
parse  = cannot parse input file

Thanks.........

Quamrul Hasan Tipu
Graduate Student
Dept. of Computer Science
USC.



From sundarra@usc.edu Thu Mar 21 13:16:43 2002
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2LLGgR29696
	for ; Thu, 21 Mar 2002 13:16:42 -0800 (PST)
Received: from aludra.usc.edu (sundarra@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id NAA08621 for ; Thu, 21 Mar 2002 13:17:08 -0800 (PST)
Received: (from sundarra@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc)
	id g2LLHZl07319 for csci551-talk@mailman.isi.edu; Thu, 21 Mar 2002 13:17:35 -0800 (PST)
Message-Id: <200203212117.g2LLHZl07319@aludra.usc.edu>
To: csci551-talk@mailman.isi.edu
Date: Thu, 21 Mar 2002 13:17:35 -0800 (PST)
From: "Srikanth Sundarrajan" 
In-Reply-To: <200203212005.g2LK51R04962@gamma.isi.edu> from "csci551-talk-request@mailman.isi.edu" at Mar 21, 2 12:05:02 pm
X-Mailer: ELM [version 2.4 PL25]
Content-Type: text
Subject: [Csci551-talk] Re: Csci551-talk digest, Vol 1 #33 - 1 msg
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,
	Is there a particular workstation in SAL, where I can meet you. Please
let me know. I will be there between 9:30 and 11pm.

	I used to compile my project using the follwing command line 

    make -f CSCI551.mak all

Thanks

Srikanth Sundarrrajan.



> 
> Send Csci551-talk mailing list submissions to
> 	csci551-talk@mailman.isi.edu
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	http://mailman.isi.edu/mailman/listinfo/csci551-talk
> or, via email, send a message with subject or body 'help' to
> 	csci551-talk-request@mailman.isi.edu
> 
> You can reach the person managing the list at
> 	csci551-talk-admin@mailman.isi.edu
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Csci551-talk digest..."
> 
> 
> Today's Topics:
> 
>    1. Attention!!!!   Grading project 1 (tipu)
> 
> --__--__--
> 
> Message: 1
> Date: Thu, 21 Mar 2002 01:27:52 -0800 (PST)
> From: tipu 
> To: 
> Subject: [Csci551-talk] Attention!!!!   Grading project 1
> 
> 
> 
> hi,
> 
> Please see the list...Those who have some kind of error with his/her name
> please come to me immediately.....
> 
> The Time schedule is
> 
> 9    - 11am    (SAL) --- 3/21/2002
> 9:30 - 11:30pm (SAL) --- 3/21/2002
> 5:30 - 8:30pm  (SAL) --- 3/22/2002
> 
> here is the list -------
> 
> Mar  9 2002 09:43:20 ganglu        done		C
> Mar  9 2002 11:29:57 sundarra      mk pr	C
> Mar  9 2002 16:38:34 jitaekli      done		C++
> Mar  9 2002 22:04:11 rishisha      done 	C++/ program is actually C
> Mar 10 2002 00:14:52 bhat	   done		C
> Mar 10 2002 01:15:16 kaiyuxia      done		C
> Mar 10 2002 01:42:29 chaoyuwa      mk pr	C++
> Mar 10 2002 02:18:25 jngai 	   done		C
> Mar 10 2002 02:39:12 alsaeed 	   done		C++
> Mar 10 2002 05:56:05 aaj 	   mk pr	C
> Mar 10 2002 06:51:35 davidmwe 	   done		C
> Mar 10 2002 06:47:57 bhanot        mk er	C
> Mar 10 2002 08:30:27 srangwal 	   done		C
> Mar 10 2002 08:55:15 rkothari 	   seg flt	C++
> Mar 10 2002 08:50:06 mprasann 	   done		C
> Mar 10 2002 08:49:46 samdadiy 	   done		C++
> Mar 10 2002 08:49:30 gurmeets 	   done		C
> Mar 10 2002 08:47:39 kasat 	   parse flt	C
> Mar 10 2002 09:16:27 anshulpa 	   parse flt	C
> Mar 10 2002 21:13:35 ragarwal 	   seg flt	C
> Mar 10 2002 22:42:42 ggoel 	   done		C
> Mar 10 2002 23:24:13 nangunoo 	   pro hangs	C
> Mar 11 2002 00:35:58 vbhargav 	   done		C
> Mar 11 2002 00:30:58 phadke 	   seg flt	C
> Mar 11 2002 03:48:14 bhave 	   mk pr	C
> Mar 11 2002 04:53:01 vsj 	   seg flt	C
> Mar 11 2002 05:03:27 renukaku 	   seg flt	C
> Mar 11 2002 05:32:33 kagarwal 	   mk pr	C
> Mar 11 2002 06:24:14 sameerme 	   mk pr	C
> Mar 11 2002 06:17:27 jeongimp 	   done		C
> Mar 11 2002 07:41:09 srihari 	   done		C
> Mar 11 2002 07:36:13 sumitbha      err+seg flt  C
> Mar 11 2002 07:58:04 sson 	   err		C
> Mar 11 2002 07:47:45 bhuptani	   Berr+seg flt C
> Mar 11 2002 07:59:46 woojinch	   Berr         C
> Mar 11 2002 08:35:26 fisk          err		C
> Mar 11 2002 08:56:58 rohera 	   seg flt	C
> 
> 
> done   = ok
> mk     = make file is not running
> seg    = segmentation fault
> Berr   = bus error
> parse  = cannot parse input file
> 
> Thanks.........
> 
> Quamrul Hasan Tipu
> Graduate Student
> Dept. of Computer Science
> USC.
> 
> 
> 
> 
> --__--__--
> 
> _______________________________________________
> Csci551-talk mailing list
> Csci551-talk@mailman.isi.edu
> http://mailman.isi.edu/mailman/listinfo/csci551-talk
> 
> 
> End of Csci551-talk Digest
> 


From sundarra@usc.edu Thu Mar 21 21:45:49 2002
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g2M5jnR15752
	for ; Thu, 21 Mar 2002 21:45:49 -0800 (PST)
Received: from aludra.usc.edu (sundarra@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id VAA01597 for ; Thu, 21 Mar 2002 21:45:57 -0800 (PST)
Received: (from sundarra@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc)
	id g2M5kQi20368 for csci551-talk@mailman.isi.edu; Thu, 21 Mar 2002 21:46:26 -0800 (PST)
Message-Id: <200203220546.g2M5kQi20368@aludra.usc.edu>
To: csci551-talk@mailman.isi.edu
Date: Thu, 21 Mar 2002 21:46:26 -0800 (PST)
From: "Srikanth Sundarrajan" 
In-Reply-To: <200203212005.g2LK51R04962@gamma.isi.edu> from "csci551-talk-request@mailman.isi.edu" at Mar 21, 2 12:05:02 pm
X-Mailer: ELM [version 2.4 PL25]
Content-Type: text
Subject: [Csci551-talk] Re: Grading Project 1
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

	Does anyone where I could possibly meet Mr. Tipu in SAL ?

Thanks

Srikanth Sundarrajan

> 
> Message: 1
> Date: Thu, 21 Mar 2002 01:27:52 -0800 (PST)
> From: tipu 
> To: 
> Subject: [Csci551-talk] Attention!!!!   Grading project 1
> 
> 
> 
> hi,
> 
> Please see the list...Those who have some kind of error with his/her name
> please come to me immediately.....
> 
> The Time schedule is
> 
> 9    - 11am    (SAL) --- 3/21/2002
> 9:30 - 11:30pm (SAL) --- 3/21/2002
> 5:30 - 8:30pm  (SAL) --- 3/22/2002
> 
> here is the list -------
> 
> Thanks.........
> 
> Quamrul Hasan Tipu
> Graduate Student
> Dept. of Computer Science
> USC.

From amilinen@usc.edu Wed Mar 27 02:47:29 2002
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 g2RAlTR13558
	for ; Wed, 27 Mar 2002 02:47:29 -0800 (PST)
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g2RAlSt25784
	for ; Wed, 27 Mar 2002 02:47:29 -0800 (PST)
Received: from aludra.usc.edu (amilinen@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id CAA26236 for ; Wed, 27 Mar 2002 02:47:50 -0800 (PST)
Received: from localhost (amilinen@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g2RAmOC24594 for ; Wed, 27 Mar 2002 02:48:24 -0800 (PST)
Date: Wed, 27 Mar 2002 02:48:24 -0800 (PST)
From: Jagadish Amilineni 
To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Duplicate Packets
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,

In Paxson Paper [Paxson 99b], we talked about duplicate packets.
Why duplication occurs...?
I think one reason for duplication is,

	Sender didnt get ack for a packet which eventually arrived at
receiver, and so sender sends the packet again because of timeout or
duplicate acks..(duplicate acks are sent by receiver before the packet
actually arrived..and the ack for this (which might be cumulative) has not
yet arrived at sender..)

Is there any other reason for duplication....? If not, is it(duplicate
packtes) really a surprise...and why?

please clarify me on this.

Thank you,
Jagadish


From johnh@ISI.EDU Sun Mar 31 20:40:31 2002
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 g314eVm18813
	for ; Sun, 31 Mar 2002 20:40:31 -0800 (PST)
Received: from snipe.prod.itd.earthlink.net (snipe.mail.pas.earthlink.net [207.217.120.62])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g314eTp28318
	for ; Sun, 31 Mar 2002 20:40:29 -0800 (PST)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by snipe.prod.itd.earthlink.net with esmtp (Exim 3.33 #1)
	id 16rtcL-0003A1-00; Sun, 31 Mar 2002 20:40:29 -0800
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g314eJK23655;
	Sun, 31 Mar 2002 20:40:20 -0800
Message-Id: <200204010440.g314eJK23655@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: Jagadish Amilineni 
Cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Duplicate Packets 
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Sun, 31 Mar 2002 20:40:19 -0800
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Wed, 27 Mar 2002 02:48:24 PST, Jagadish Amilineni wrote: 
>Hi,
>
>In Paxson Paper [Paxson 99b], we talked about duplicate packets.
>Why duplication occurs...?
>I think one reason for duplication is,
>
>	Sender didnt get ack for a packet which eventually arrived at
>receiver, and so sender sends the packet again because of timeout or
>duplicate acks..(duplicate acks are sent by receiver before the packet
>actually arrived..and the ack for this (which might be cumulative) has not
>yet arrived at sender..)
>
>Is there any other reason for duplication....? If not, is it(duplicate
>packtes) really a surprise...and why?

Yes, that would be an example of duplicate packets.

But recall that Paxson measured both the sender AND receiver,
so he knew how many pkts the sender sent, and if there were
retransmissions, and factored those out.

His definition of duplicate packets was cases where the sender sent
one packet but the receiver got two.

One possible explanation is an error at the MAC level.  Think about
our discussions about 802.11 and MACAW.

   -John Heidemann


From vkhanna@qualcomm.com Mon Apr  1 20:19:50 2002
Received: from illyana.qualcomm.com (illyana.qualcomm.com [129.46.65.100])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g324Jnm04183
	for ; Mon, 1 Apr 2002 20:19:49 -0800 (PST)
Received: from VKHANNA.qualcomm.com (vkhanna.qualcomm.com [129.46.21.96])
	by illyana.qualcomm.com (8.12.1/8.12.1/1.0) with ESMTP id g324JkWg017929
	for ; Mon, 1 Apr 2002 20:19:47 -0800 (PST)
Message-Id: <5.1.0.14.2.20020401201736.0197ec18@illyana>
X-Sender: vkhanna@illyana
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Mon, 01 Apr 2002 20:21:24 -0800
To: csci551-talk@mailman.isi.edu
From: Vivek Khanna 
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_495036064==_.ALT"
Subject: [Csci551-talk] HW2: Q3(a) query
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

--=====================_495036064==_.ALT
Content-Type: text/plain; charset="us-ascii"; format=flowed

In Q3(a), can you tell what is the meaning of "When does your TCP ow complete".

--Vivek

--=====================_495036064==_.ALT
Content-Type: text/html; charset="us-ascii"


In Q3(a), can you tell what is the meaning of "When does your TCP
ow complete".


--Vivek



--=====================_495036064==_.ALT--


From vkhanna@qualcomm.com Mon Apr  1 20:30:14 2002
Received: from illyana.qualcomm.com (illyana.qualcomm.com [129.46.65.100])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g324UDm06781
	for ; Mon, 1 Apr 2002 20:30:13 -0800 (PST)
Received: from VKHANNA.qualcomm.com (vkhanna.qualcomm.com [129.46.21.96])
	by illyana.qualcomm.com (8.12.1/8.12.1/1.0) with ESMTP id g324UBWg018504
	for ; Mon, 1 Apr 2002 20:30:11 -0800 (PST)
Message-Id: <5.1.0.14.2.20020401203003.01971058@illyana>
X-Sender: vkhanna@illyana
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Mon, 01 Apr 2002 20:31:49 -0800
To: csci551-talk@mailman.isi.edu
From: Vivek Khanna 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Csci551-talk] HW2: Q3(a) query
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

In Q3, to observe the animation behavior, I click on play but the animation 
does not show packets flowing. Isn't it suppose to  show the flow of all 
the packets. Am I missing something? Please clarify

--Vivek


From jngai618@yahoo.com Tue Apr  2 10:08:44 2002
Received: from web14101.mail.yahoo.com (web14101.mail.yahoo.com [216.136.172.131])
	by gamma.isi.edu (8.11.6/8.11.2) with SMTP id g32I8im12366
	for ; Tue, 2 Apr 2002 10:08:44 -0800 (PST)
Message-ID: <20020402180843.42092.qmail@web14101.mail.yahoo.com>
Received: from [192.59.177.5] by web14101.mail.yahoo.com via HTTP; Tue, 02 Apr 2002 10:08:43 PST
Date: Tue, 2 Apr 2002 10:08:43 -0800 (PST)
From: Jeffrey Ngai 
To: csci551-talk@mailman.isi.edu
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: [Csci551-talk] HW2 3d Question
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

In Q.3d), do we drop both packet #6 and #3 (we drop
packet #3 in Q.3b)? or do we just drop packet #6 only?

Jeffrey

__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://http://taxes.yahoo.com/

From jngai618@yahoo.com Tue Apr  2 10:16:07 2002
Received: from web14106.mail.yahoo.com (web14106.mail.yahoo.com [216.136.172.136])
	by gamma.isi.edu (8.11.6/8.11.2) with SMTP id g32IG7m15190
	for ; Tue, 2 Apr 2002 10:16:07 -0800 (PST)
Message-ID: <20020402181607.80098.qmail@web14106.mail.yahoo.com>
Received: from [192.59.177.5] by web14106.mail.yahoo.com via HTTP; Tue, 02 Apr 2002 10:16:07 PST
Date: Tue, 2 Apr 2002 10:16:07 -0800 (PST)
From: Jeffrey Ngai 
To: csci551-talk@mailman.isi.edu
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: [Csci551-talk] RE: HW2: Q3(a) query
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

The packet will start flowing at time 1 second. The
time keep incrementing by certain milliseconds after
you click the play button, depeding on your settings.
So, it will take a while REAL time to actually see the
animation.

Hope this helps.

Jeffrey

-----------------------------------------------------
In Q3, to observe the animation behavior, I click on
play but the animation 
does not show packets flowing. Isn't it suppose to 
show the flow of all 
the packets. Am I missing something? Please clarify

--Vivek



__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://http://taxes.yahoo.com/

From hussain@ISI.EDU Tue Apr  2 10:45:43 2002
Received: from vapor.isi.edu (vapor.isi.edu [128.9.64.64])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g32Ijhm25970
	for ; Tue, 2 Apr 2002 10:45:43 -0800 (PST)
Received: from vin.isi.edu (vin.isi.edu [128.9.168.27])
	by vapor.isi.edu (8.11.6/8.11.2) with ESMTP id g32IjfH01480;
	Tue, 2 Apr 2002 10:45:41 -0800 (PST)
Date: Tue, 2 Apr 2002 10:45:41 -0800 (PST)
From: Alefiya Hussain 
To: Vivek Khanna 
cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] HW2: Q3(a) query
In-Reply-To: <5.1.0.14.2.20020401201736.0197ec18@illyana>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi Vivek,

That is "TCP flow" 

alefiya 

On Mon, 1 Apr 2002, Vivek Khanna wrote:

> In Q3(a), can you tell what is the meaning of "When does your TCP ow complete".
> 
> --Vivek
> 


From hussain@ISI.EDU Tue Apr  2 10:45:15 2002
Received: from vapor.isi.edu (vapor.isi.edu [128.9.64.64])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g32IjEm25568
	for ; Tue, 2 Apr 2002 10:45:14 -0800 (PST)
Received: from vin.isi.edu (vin.isi.edu [128.9.168.27])
	by vapor.isi.edu (8.11.6/8.11.2) with ESMTP id g32Ij4H01338;
	Tue, 2 Apr 2002 10:45:04 -0800 (PST)
Date: Tue, 2 Apr 2002 10:45:04 -0800 (PST)
From: Alefiya Hussain 
To: Jeffrey Ngai 
cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] HW2 3d Question
In-Reply-To: <20020402180843.42092.qmail@web14101.mail.yahoo.com>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi Jeffrey,

Please drop both packets 3 and 6 . 

alefiya 

On Tue, 2 Apr 2002, Jeffrey Ngai wrote:

> In Q.3d), do we drop both packet #6 and #3 (we drop
> packet #3 in Q.3b)? or do we just drop packet #6 only?
> 
> Jeffrey
> 
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Tax Center - online filing with TurboTax
> http://http://taxes.yahoo.com/
> 


From johnh@ISI.EDU Tue Apr  2 11:03:13 2002
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 g32J3Cm05207
	for ; Tue, 2 Apr 2002 11:03:12 -0800 (PST)
Received: from dash.isi.edu (dash.isi.edu [128.9.160.184])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g32J3Cp16466;
	Tue, 2 Apr 2002 11:03:12 -0800 (PST)
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g32J3Bf19328;
	Tue, 2 Apr 2002 11:03:11 -0800
Message-Id: <200204021903.g32J3Bf19328@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: Alefiya Hussain 
Cc: Jeffrey Ngai , csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] HW2 3d Question 
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Tue, 02 Apr 2002 11:03:11 -0800
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Tue, 02 Apr 2002 10:45:04 PST, Alefiya Hussain wrote: 
>Hi Jeffrey,
>
>Please drop both packets 3 and 6 . 
>
>alefiya 
>
>On Tue, 2 Apr 2002, Jeffrey Ngai wrote:
>
>> In Q.3d), do we drop both packet #6 and #3 (we drop
>> packet #3 in Q.3b)? or do we just drop packet #6 only?
>> 
>> Jeffrey



NO, this is not what is specified.

For parts b and c, drop ONLY packet #3.

For parts d and e, drop ONLY packet #6 (NOT both).

I will update the homework.

   -John Heidemann

From kaiyuxia@usc.edu Tue Apr  2 23:23:09 2002
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 g337N9m11486
	for ; Tue, 2 Apr 2002 23:23:09 -0800 (PST)
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g337N8p24188;
	Tue, 2 Apr 2002 23:23:08 -0800 (PST)
Received: from aludra.usc.edu (kaiyuxia@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id XAA22035; Tue, 2 Apr 2002 23:23:26 -0800 (PST)
Received: from localhost (kaiyuxia@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g337O6a22326; Tue, 2 Apr 2002 23:24:06 -0800 (PST)
Date: Tue, 2 Apr 2002 23:24:06 -0800 (PST)
From: kaiyuxia 
To: John Heidemann 
cc: Alefiya Hussain , 
In-Reply-To: <200204021903.g32J3Bf19328@dash.isi.edu>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] problem with class web
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi, Prof. & TA,

I tried to download the latest version of homework2, yet it still contains
the same kind of typo like Fle, Confge etc.

For TA's clarification page,  the syntax of ns and nam are not there.
(Very strange thing, I did remember seeing it last week. )

Could you please have a check of both files?


Thanks very much,





From s_nangunoori@trillium.com Tue Apr  2 09:34:30 2002
Received: from ganymede.or.intel.com (jffdns01.or.intel.com [134.134.248.3])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g32HYUm25917
	for ; Tue, 2 Apr 2002 09:34:30 -0800 (PST)
Received: from orsmsxvs040.jf.intel.com (orsmsxvs040.jf.intel.com [192.168.65.206])
	by ganymede.or.intel.com (8.11.6/8.11.6/d: solo.mc,v 1.32 2002/04/01 18:53:51 root Exp $) with SMTP id g32HYLF18056
	for ; Tue, 2 Apr 2002 17:34:21 GMT
Received: from orsmsx26.jf.intel.com ([192.168.65.26])
 by orsmsxvs040.jf.intel.com (NAVGW 2.5.1.16) with SMTP id M2002040209405521086
 ; Tue, 02 Apr 2002 09:40:55 -0800
Received: by orsmsx26.jf.intel.com with Internet Mail Service (5.5.2653.19)
	id ; Tue, 2 Apr 2002 09:34:20 -0800
Message-ID: 
From: "Nangunoori, Srikanth" 
To: "'Vivek Khanna'" , csci551-talk@mailman.isi.edu
Subject: RE: [Csci551-talk] HW2: Q3(a) query
Date: Tue, 2 Apr 2002 09:34:16 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="ISO-8859-1"
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

I think .tr file is more helpful than the animation. You can try that.
It contains the complete info -- events, time, src, dst, etc.

Srikanth

-----Original Message-----
From: Vivek Khanna [mailto:vkhanna@qualcomm.com]
Sent: Monday, April 01, 2002 8:32 PM
To: csci551-talk@mailman.isi.edu
Subject: [Csci551-talk] HW2: Q3(a) query


In Q3, to observe the animation behavior, I click on play but the animation 
does not show packets flowing. Isn't it suppose to  show the flow of all 
the packets. Am I missing something? Please clarify

--Vivek

From johnh@ISI.EDU Wed Apr  3 09:53:52 2002
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 g33Hrqm06830
	for ; Wed, 3 Apr 2002 09:53:52 -0800 (PST)
Received: from avocet.prod.itd.earthlink.net (avocet.mail.pas.earthlink.net [207.217.120.50])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g33Hrpp27858;
	Wed, 3 Apr 2002 09:53:51 -0800 (PST)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by avocet.prod.itd.earthlink.net with esmtp (Exim 3.33 #1)
	id 16soxD-0005h7-00; Wed, 03 Apr 2002 09:53:51 -0800
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g33Hrbl02846;
	Wed, 3 Apr 2002 09:53:37 -0800
Message-Id: <200204031753.g33Hrbl02846@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: kaiyuxia 
Cc: Alefiya Hussain , csci551-talk@ISI.EDU
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Wed, 03 Apr 2002 09:53:37 -0800
From: John Heidemann 
Subject: [Csci551-talk] Re: problem with class web
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Tue, 02 Apr 2002 23:24:06 PST, kaiyuxia wrote: 
>Hi, Prof. & TA,
>
>I tried to download the latest version of homework2, yet it still contains
>the same kind of typo like Fle, Confge etc.

Thank you for catching this.  There was a font encoding problem
that should now be fixed.

>For TA's clarification page,  the syntax of ns and nam are not there.
>(Very strange thing, I did remember seeing it last week. )

Alefiya, can you check on this part?

   -John Heidemann

>
>Could you please have a check of both files?
>
>
>Thanks very much,
>
>
>

From vkhanna@qualcomm.com Wed Apr  3 11:01:45 2002
Received: from illyana.qualcomm.com (illyana.qualcomm.com [129.46.65.100])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g33J1hm06630
	for ; Wed, 3 Apr 2002 11:01:45 -0800 (PST)
Received: from VKHANNA.qualcomm.com (vkhanna.qualcomm.com [129.46.21.96])
	by illyana.qualcomm.com (8.12.1/8.12.1/1.0) with ESMTP id g33J1eWg022910
	for ; Wed, 3 Apr 2002 11:01:40 -0800 (PST)
Message-Id: <5.1.0.14.2.20020403110259.04607608@illyana>
X-Sender: vkhanna@illyana
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Wed, 03 Apr 2002 11:03:14 -0800
To: csci551-talk@mailman.isi.edu
From: Vivek Khanna 
Mime-Version: 1.0
Content-Type: multipart/alternative;
	boundary="=====================_634345741==_.ALT"
Subject: [Csci551-talk] HW2: Q3(c) query
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

--=====================_634345741==_.ALT
Content-Type: text/plain; charset="us-ascii"; format=flowed

Another one. This is one of the lines in output

r 1.010032 0 1 tcp 40 ------- 0 0.0 3.0 0 0

Can you tell what is the meaning of above text highlighted in bold. Also I 
saw that in the first letter + or - being used.

On running ns hw2.tcl, an output file out.nam and out.tr is produced. On 
running sequence [PLAY], this file (out.tr) does not get updated. Is it 
because everytime we run the same seq again, the output is the same.

--Vivek


--=====================_634345741==_.ALT
Content-Type: text/html; charset="us-ascii"


Another one. This is one of the lines in output


r 1.010032 0 1 tcp 40 ------- 0 0.0 3.0 0 0


Can you tell what is the meaning of above text highlighted in bold.
Also I saw that in the first letter + or - being used.


On running ns hw2.tcl, an output file out.nam and out.tr is produced. On
running sequence [PLAY], this file (out.tr) does not get updated. Is it
because everytime we run the same seq again, the output is the
same.


--Vivek





--=====================_634345741==_.ALT--


From s_nangunoori@trillium.com Wed Apr  3 11:09:44 2002
Received: from hermes.fm.intel.com (fmr01.intel.com [192.55.52.18])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g33J9im18974
	for ; Wed, 3 Apr 2002 11:09:44 -0800 (PST)
Received: from talaria.fm.intel.com (talaria.fm.intel.com [10.1.192.39])
	by hermes.fm.intel.com (8.11.6/8.11.6/d: outer.mc,v 1.37 2002/04/01 17:50:55 root Exp $) with ESMTP id g33JB0J20493
	for ; Wed, 3 Apr 2002 19:11:00 GMT
Received: from fmsmsxvs042.fm.intel.com (fmsmsxv042-1.fm.intel.com [132.233.48.110])
	by talaria.fm.intel.com (8.11.6/8.11.6/d: inner.mc,v 1.15 2002/04/01 17:51:48 root Exp $) with SMTP id g33JAu405526
	for ; Wed, 3 Apr 2002 19:10:56 GMT
Received: from fmsmsx29.FM.INTEL.COM ([132.233.42.29])
 by fmsmsxvs042.fm.intel.com (NAVGW 2.5.1.16) with SMTP id M2002040311132531293
 ; Wed, 03 Apr 2002 11:13:25 -0800
Received: by fmsmsx29.fm.intel.com with Internet Mail Service (5.5.2653.19)
	id <2B43D4CQ>; Wed, 3 Apr 2002 11:09:37 -0800
Message-ID: 
From: "Nangunoori, Srikanth" 
To: "'Vivek Khanna'" , csci551-talk@mailman.isi.edu
Subject: RE: [Csci551-talk] HW2: Q3(c) query
Date: Wed, 3 Apr 2002 11:09:29 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C1DB43.14398410"
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

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_01C1DB43.14398410
Content-Type: text/plain;
	charset="ISO-8859-1"

 

-----Original Message-----
From: Vivek Khanna [mailto:vkhanna@qualcomm.com]
Sent: Wednesday, April 03, 2002 11:03 AM
To: csci551-talk@mailman.isi.edu
Subject: [Csci551-talk] HW2: Q3(c) query


Another one. This is one of the lines in output

r 1.010032 0 1 tcp 40 ------- 0 0.0 3.0 0 0

Can you tell what is the meaning of above text highlighted in bold. Also I
saw that in the first letter + or - being used.
[Nangunoori, Srikanth] + and - indicates that the packet is queued and
dequeued respectively. 

On running ns hw2.tcl, an output file out.nam and out.tr is produced. On
running sequence [PLAY], this file (out.tr) does not get updated. Is it
because everytime we run the same seq again, the output is the same.
[Nangunoori, Srikanth] Pl. correct me if I'm wrong on this one. Everytime
you replay, nam wont simulate all the actions again instead it just shows
the sequence of actions created by the ns output. So thats why you will see
the same data everytime you replay.

--Vivek




------_=_NextPart_001_01C1DB43.14398410
Content-Type: text/html;
	charset="ISO-8859-1"









 



  
-----Original Message-----
From: Vivek Khanna 
  [mailto:vkhanna@qualcomm.com]
Sent: Wednesday, April 03, 2002 11:03 
  AM
To: csci551-talk@mailman.isi.edu
Subject: 
  [Csci551-talk] HW2: Q3(c) query

Another one. This is one 
  of the lines in output

r 1.010032 0 1 tcp 40 ------- 0 0.0 
  3.0 0 0

Can you tell what is the meaning of above text highlighted 
  in bold. Also I saw that in the first letter + or - being used.
[Nangunoori, 
  Srikanth] + and - indicates that the packet is queued and dequeued 
  respectively. 

On running ns hw2.tcl, an output file 
  out.nam and out.tr is produced. On running sequence [PLAY], this file (out.tr) 
  does not get updated. Is it because everytime we run the same seq again, the 
  output is the same.
[Nangunoori, Srikanth] Pl. correct me if I'm wrong on 
  this one. Everytime you replay, nam wont simulate all the 
  actions again instead it just shows the sequence of actions created by the ns 
  output. So thats why you will see the same data everytime you 
  replay.

--Vivek



------_=_NextPart_001_01C1DB43.14398410--

From s_nangunoori@trillium.com Wed Apr  3 11:11:08 2002
Received: from hermes.fm.intel.com (fmr01.intel.com [192.55.52.18])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g33JB7m22023
	for ; Wed, 3 Apr 2002 11:11:07 -0800 (PST)
Received: from talaria.fm.intel.com (talaria.fm.intel.com [10.1.192.39])
	by hermes.fm.intel.com (8.11.6/8.11.6/d: outer.mc,v 1.37 2002/04/01 17:50:55 root Exp $) with ESMTP id g33JCPJ21640
	for ; Wed, 3 Apr 2002 19:12:25 GMT
Received: from fmsmsxvs042.fm.intel.com (fmsmsxv042-1.fm.intel.com [132.233.48.110])
	by talaria.fm.intel.com (8.11.6/8.11.6/d: inner.mc,v 1.15 2002/04/01 17:51:48 root Exp $) with SMTP id g33JCL406683
	for ; Wed, 3 Apr 2002 19:12:21 GMT
Received: from fmsmsx26.fm.intel.com ([132.233.42.26])
 by fmsmsxvs042.fm.intel.com (NAVGW 2.5.1.16) with SMTP id M2002040311145127098
 for ; Wed, 03 Apr 2002 11:14:51 -0800
Received: by fmsmsx26.fm.intel.com with Internet Mail Service (5.5.2653.19)
	id ; Wed, 3 Apr 2002 11:11:05 -0800
Message-ID: 
From: "Nangunoori, Srikanth" 
To: "'csci551-talk@mailman.isi.edu'" 
Date: Wed, 3 Apr 2002 11:11:00 -0800 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Csci551-talk] HW#2 queries
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,

I've following queries regarding the HW#2 questions:

1> For question 2a - Do we need mention the numerical value or the
expression representing some inequality?

2> For questions 3c and 3e "What action did TCP mean" - I didnt exactly
understood the question. Can someone clarify on that? My doubt is do we need
to mention the algorithm used by TCP in that particular situation or do we
need to explain the complete sequence?


Regards,
Srikanth


From hussain@ISI.EDU Wed Apr  3 11:41:05 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g33Jf5m00548
	for ; Wed, 3 Apr 2002 11:41:05 -0800 (PST)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g33Jf2W29042;
	Wed, 3 Apr 2002 11:41:02 -0800 (PST)
Date: Wed, 3 Apr 2002 11:41:02 -0800 (PST)
From: Alefiya Hussain 
To: Vivek Khanna 
cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] HW2: Q3(c) query
In-Reply-To: <5.1.0.14.2.20020403110259.04607608@illyana>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 


On Wed, 3 Apr 2002, Vivek Khanna wrote:

> Another one. This is one of the lines in output
> 
> r 1.010032 0 1 tcp 40 ------- 0 0.0 3.0 0 0
    (time)       (pro)            (s) (d)   (id)
> 
> Can you tell what is the meaning of above text highlighted in bold. Also I 
> saw that in the first letter + or - being used.

The first column is the operation the trace file is reporting ie: 
enque operations (indicated by ``+''),  deque
operations (indicated by ``-''), receive events (indicated by ``r''),
and one drop event (idicated by "d"). 

Some of the other important fields have been indicated above. Please refer
to the ns manual of more information. 
http://www.isi.edu/nsnam/ns/doc/index.html


> 
> On running ns hw2.tcl, an output file out.nam and out.tr is produced. On 
> running sequence [PLAY], this file (out.tr) does not get updated. Is it 
> because everytime we run the same seq again, the output is the same.

Since the hw2 simulation is deterministic, the same script if run twice
will produce the same output ( simulations is repeatable). The trace files will
be the same. 

i hope this helps, 
alefiya 





From hussain@ISI.EDU Wed Apr  3 11:46:24 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g33JkOm02272
	for ; Wed, 3 Apr 2002 11:46:24 -0800 (PST)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g33JkIC01660;
	Wed, 3 Apr 2002 11:46:18 -0800 (PST)
Date: Wed, 3 Apr 2002 11:46:18 -0800 (PST)
From: Alefiya Hussain 
To: John Heidemann 
cc: kaiyuxia , csci551-talk@ISI.EDU
In-Reply-To: <200204031753.g33Hrbl02846@dash.isi.edu>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Re: problem with class web
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

The hints page can be reached at 
http://www.isi.edu/~hussain/cs551_hints.html

The settings required to run ns and nam in your accounts is present in the
second paragraph on the page.
I have updated the page since last week by adding the error
commonly experienced by students. 

alefiya 

On Wed, 3 Apr 2002, John Heidemann wrote:

> On Tue, 02 Apr 2002 23:24:06 PST, kaiyuxia wrote: 
> >Hi, Prof. & TA,
> >
> >I tried to download the latest version of homework2, yet it still contains
> >the same kind of typo like Fle, Confge etc.
> 
> Thank you for catching this.  There was a font encoding problem
> that should now be fixed.
> 
> >For TA's clarification page,  the syntax of ns and nam are not there.
> >(Very strange thing, I did remember seeing it last week. )
> 
> Alefiya, can you check on this part?
> 
>    -John Heidemann
> 
> >
> >Could you please have a check of both files?
> >
> >
> >Thanks very much,
> >
> >
> >
> 


From gina@lanl.gov Wed Apr  3 12:38:59 2002
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 g33Kcxm23529
	for ; Wed, 3 Apr 2002 12:38:59 -0800 (PST)
Received: from merilahna.lanl.gov (splat.lanl.gov [128.165.17.254])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g33Kcwp17054;
	Wed, 3 Apr 2002 12:38:58 -0800 (PST)
Received: by merilahna.lanl.gov (Postfix, from userid 1000)
	id 3996A4BA8A; Wed,  3 Apr 2002 13:28:00 -0700 (MST)
Received: from localhost (localhost [127.0.0.1])
	by merilahna.lanl.gov (Postfix) with ESMTP
	id 3448843A84; Wed,  3 Apr 2002 13:28:00 -0700 (MST)
Date: Wed, 3 Apr 2002 13:28:00 -0700 (MST)
From: Gina Fisk 
To: Alefiya Hussain 
Cc: csci551-talk@ISI.EDU
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] H2 submission
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Just a minor question:

In the assignment it says to use 'submit' on aludra with:

... -tag hw1 ...

Since this is hw2, I am assuming this is incorrect and you want us to
submit to 

... -tag hw2 ...

???

Gina
_______________________________________
Gina Fisk 
Los Alamos National Laboratory
CCS-1 Advanced Computing Group, MS D451
Los Alamos, New Mexico  87545
(work) 505-667-6769, (fax) 505-665-4934


From hussain@ISI.EDU Wed Apr  3 12:44:13 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g33KiDm24924
	for ; Wed, 3 Apr 2002 12:44:13 -0800 (PST)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g33KiC408161;
	Wed, 3 Apr 2002 12:44:12 -0800 (PST)
Date: Wed, 3 Apr 2002 12:44:11 -0800 (PST)
From: Alefiya Hussain 
To: Gina Fisk 
cc: csci551-talk@ISI.EDU
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Re: H2 submission
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Thanks Gina,

Thats correct. Please submit homework 2 using "-tag hw2" 

alefiya 

On Wed, 3 Apr 2002, Gina Fisk wrote:

> 
> Just a minor question:
> 
> In the assignment it says to use 'submit' on aludra with:
> 
> ... -tag hw1 ...
> 
> Since this is hw2, I am assuming this is incorrect and you want us to
> submit to 
> 
> ... -tag hw2 ...
> 
> ???
> 
> Gina
> _______________________________________
> Gina Fisk 
> Los Alamos National Laboratory
> CCS-1 Advanced Computing Group, MS D451
> Los Alamos, New Mexico  87545
> (work) 505-667-6769, (fax) 505-665-4934
> 


From johnh@ISI.EDU Wed Apr  3 14:22:18 2002
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 g33MMIm01225
	for ; Wed, 3 Apr 2002 14:22:18 -0800 (PST)
Received: from dash.isi.edu (dash.isi.edu [128.9.160.184])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g33MMHp02483;
	Wed, 3 Apr 2002 14:22:17 -0800 (PST)
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g33MM2O05245;
	Wed, 3 Apr 2002 14:22:03 -0800
Message-Id: <200204032222.g33MM2O05245@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: Gina Fisk 
Cc: Alefiya Hussain , csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] H2 submission 
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Wed, 03 Apr 2002 14:22:02 -0800
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Wed, 03 Apr 2002 13:28:00 MST, Gina Fisk wrote: 
>
>Just a minor question:
>
>In the assignment it says to use 'submit' on aludra with:
>
>... -tag hw1 ...
>
>Since this is hw2, I am assuming this is incorrect and you want us to
>submit to 
>
>... -tag hw2 ...
>
>???

Yes, definitely.  Thanks for catching that obvious bug.

   -John Heidemann

From renuka_kumar@hotmail.com Thu Apr  4 15:19:37 2002
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 g34NJbm26007
	for ; Thu, 4 Apr 2002 15:19:37 -0800 (PST)
Received: from hotmail.com (f263.law10.hotmail.com [64.4.14.138])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g34NJbp01369
	for ; Thu, 4 Apr 2002 15:19:37 -0800 (PST)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Thu, 4 Apr 2002 15:19:31 -0800
Received: from 128.125.30.31 by lw10fd.law10.hotmail.msn.com with HTTP;
	Thu, 04 Apr 2002 23:19:31 GMT
X-Originating-IP: [128.125.30.31]
From: "renuka kumar" 
To: csci551-talk@ISI.EDU
Date: Fri, 05 Apr 2002 04:49:31 +0530
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: 
X-OriginalArrivalTime: 04 Apr 2002 23:19:31.0784 (UTC) FILETIME=[2CE4A480:01C1DC2F]
Subject: [Csci551-talk] Trace file
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,

There something i found weird in the Trace file generated , The Acks for the 
packet generated at Node 3 , does not seem to maintain the Seq No. of the 
packet properly (i dont mean the Unique Packet Identifier). Did anybody else 
observe the sam error , or am i missing something ?

thanx
renuka

_________________________________________________________________
Chat with friends online, try MSN Messenger: http://messenger.msn.com


From johnh@ISI.EDU Sun Apr  7 01:00:47 2002
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 g3790lm22265
	for ; Sun, 7 Apr 2002 01:00:47 -0800 (PST)
Received: from albatross.prod.itd.earthlink.net (albatross.mail.pas.earthlink.net [207.217.120.120])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3790kp17643
	for ; Sun, 7 Apr 2002 01:00:46 -0800 (PST)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by albatross.prod.itd.earthlink.net with esmtp (Exim 3.33 #1)
	id 16u8XW-0000KG-00
	for csci551-talk@isi.edu; Sun, 07 Apr 2002 01:00:46 -0800
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3790Q027713
	for ; Sun, 7 Apr 2002 01:00:26 -0800
Message-Id: <200204070900.g3790Q027713@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: csci551-talk@ISI.EDU
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Sun, 07 Apr 2002 01:00:26 -0800
From: John Heidemann 
Subject: [Csci551-talk] project b updated
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Important changes to project B have been added to the class web page.

- Several *additions* to the  required output for stage 4.
	(Make sure you check them out.)

- Stage 5 reduced in scope.

- Sample input changed, and partial sample output added.


In addition, the sample solution to Project A
has changed slightly:

- the Makefile was updated along the lines of Alefiya's changes

- some comments have been added to linkstate.cc

   -John Heidemann


From akapoor@usc.edu Sun Apr  7 13:29:13 2002
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 g37KTDm14368
	for ; Sun, 7 Apr 2002 13:29:13 -0700 (PDT)
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g37KTCp16540;
	Sun, 7 Apr 2002 13:29:12 -0700 (PDT)
Received: from aludra.usc.edu (akapoor@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id NAA04314; Sun, 7 Apr 2002 13:29:28 -0700 (PDT)
Received: from localhost (akapoor@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g37KUCR16544; Sun, 7 Apr 2002 13:30:12 -0700 (PDT)
Date: Sun, 7 Apr 2002 13:30:11 -0700 (PDT)
From: akapoor 
To: John Heidemann 
cc: 
Subject: Re: [Csci551-talk] project b updated
In-Reply-To: <200204070900.g3790Q027713@dash.isi.edu>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Professor,
Could you please provide the distribution of grading as well.
As in what part carries what wieghtage.

Thank you
Aditi


On Sun, 7 Apr 2002, John Heidemann wrote:

>
> Important changes to project B have been added to the class web page.
>
> - Several *additions* to the  required output for stage 4.
> 	(Make sure you check them out.)
>
> - Stage 5 reduced in scope.
>
> - Sample input changed, and partial sample output added.
>
>
> In addition, the sample solution to Project A
> has changed slightly:
>
> - the Makefile was updated along the lines of Alefiya's changes
>
> - some comments have been added to linkstate.cc
>
>    -John Heidemann
>


From samdadiy@usc.edu Sun Apr  7 19:09:01 2002
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 g38290m08589
	for ; Sun, 7 Apr 2002 19:09:00 -0700 (PDT)
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g38290p07523
	for ; Sun, 7 Apr 2002 19:09:00 -0700 (PDT)
Received: from sal-sun021.usc.edu (samdadiy@sal-sun021.usc.edu [128.125.5.91])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id TAA03144 for ; Sun, 7 Apr 2002 19:09:14 -0700 (PDT)
Received: from localhost (samdadiy@localhost)
	by sal-sun021.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g38297Q04425 for ; Sun, 7 Apr 2002 19:09:07 -0700 (PDT)
Date: Sun, 7 Apr 2002 19:09:07 -0700 (PDT)
From: Parag Samdadiya 
To: 
In-Reply-To: <200204070900.g3790Q027713@dash.isi.edu>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Makefile of sample implementation
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi TA,

I tried to make the sample implementation.
But instead of using g++ as compiler as specified
in the Makefile somehow it is using CC
Hence I am getting this errors.

If anyone has faced same problem and know the prob/solution
let me know

Thanks,
Parag

CC    -c  gen.cc
"gen.hh", line 29: Error: Could not open include file .
"gen.hh", line 30: Error: Could not open include file .
"gen.hh", line 51: Error: Type name expected instead of "bool".
"gen.hh", line 51: Error: No storage class or type for this declaration.
"gen.hh", line 51: Error: "," expected instead of "operator".
"gen.hh", line 52: Error: Use ";" to terminate declarations.
"gen.hh", line 52: Error: Type name expected instead of "bool".
"gen.hh", line 52: Error: No storage class or type for this declaration.
"gen.hh", line 52: Error: Multiple declaration for bool.
"gen.hh", line 52: Error: "," expected instead of "operator".


From hussain@ISI.EDU Mon Apr  8 09:30:20 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g38GUJm22408
	for ; Mon, 8 Apr 2002 09:30:19 -0700 (PDT)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g38GUJX14654;
	Mon, 8 Apr 2002 09:30:19 -0700 (PDT)
Date: Mon, 8 Apr 2002 09:30:19 -0700 (PDT)
From: Alefiya Hussain 
To: Parag Samdadiya 
cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Makefile of sample implementation
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Set the default compiler to g++ 

you can use the following line in the makefile
CCC = g++ 

alefiya 



On Sun, 7 Apr 2002, Parag Samdadiya wrote:

> 
> Hi TA,
> 
> I tried to make the sample implementation.
> But instead of using g++ as compiler as specified
> in the Makefile somehow it is using CC
> Hence I am getting this errors.
> 
> If anyone has faced same problem and know the prob/solution
> let me know
> 
> Thanks,
> Parag
> 
> CC    -c  gen.cc
> "gen.hh", line 29: Error: Could not open include file .
> "gen.hh", line 30: Error: Could not open include file .
> "gen.hh", line 51: Error: Type name expected instead of "bool".
> "gen.hh", line 51: Error: No storage class or type for this declaration.
> "gen.hh", line 51: Error: "," expected instead of "operator".
> "gen.hh", line 52: Error: Use ";" to terminate declarations.
> "gen.hh", line 52: Error: Type name expected instead of "bool".
> "gen.hh", line 52: Error: No storage class or type for this declaration.
> "gen.hh", line 52: Error: Multiple declaration for bool.
> "gen.hh", line 52: Error: "," expected instead of "operator".
> 


From hussain@ISI.EDU Mon Apr  8 13:00:19 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g38K0Jm03545
	for ; Mon, 8 Apr 2002 13:00:19 -0700 (PDT)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g38K0Jw21596;
	Mon, 8 Apr 2002 13:00:19 -0700 (PDT)
Date: Mon, 8 Apr 2002 13:00:19 -0700 (PDT)
From: Alefiya Hussain 
To: csci551-talk@ISI.EDU
cc: John Heidemann 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Hints Webpage updated
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

FYI: 
The hints page located at http://www.isi.edu/~hussain/cs551_hints.html has
been updated to include the following information: 

	- The testcases used to grade project A are now available 
	- Project B can be run on aludra/nunki( limited to a max of 15
processes) using the process name: cs551P2 
	- Sample output for the manager.conf available 

all the best
alefiya 


From tipu@usc.edu Mon Apr  8 23:58:36 2002
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g396wam27086
	for ; Mon, 8 Apr 2002 23:58:36 -0700 (PDT)
Received: from aludra.usc.edu (tipu@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id XAA11920 for ; Mon, 8 Apr 2002 23:58:50 -0700 (PDT)
Received: from localhost (tipu@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g396xaH23669 for ; Mon, 8 Apr 2002 23:59:36 -0700 (PDT)
Date: Mon, 8 Apr 2002 23:59:36 -0700 (PDT)
From: tipu 
To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Office hours for Project A grading
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

hi,

If u have any doubts about ur grade in project A, u are welcome to come at
SAL on Wednesday. I will be there from 10 to 12 in the morning.

Quamrul Hasan Tipu
Graduate Student
Dept. of Computer Science
USC.


From gina@lanl.gov Tue Apr  9 07:18:45 2002
Received: from mailrelay2.lanl.gov (mailrelay2.lanl.gov [128.165.4.103])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g39EIjm18465
	for ; Tue, 9 Apr 2002 07:18:45 -0700 (PDT)
Received: from merilahna.lanl.gov (localhost.localdomain [127.0.0.1])
	by mailrelay2.lanl.gov (8.11.6/8.11.6/(ccn-5)) with ESMTP id g39EIiC28196
	for ; Tue, 9 Apr 2002 08:18:44 -0600
Received: by merilahna.lanl.gov (Postfix, from userid 1000)
	id 7C3444BBCF; Tue,  9 Apr 2002 08:18:08 -0600 (MDT)
Received: from localhost (localhost [127.0.0.1])
	by merilahna.lanl.gov (Postfix) with ESMTP
	id 0444D43A84; Tue,  9 Apr 2002 08:18:07 -0600 (MDT)
Date: Tue, 9 Apr 2002 08:18:06 -0600 (MDT)
From: Gina Fisk 
To: tipu 
Cc: csci551-talk@mailman.isi.edu
Subject: Re: [Csci551-talk] Office hours for Project A grading
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Tipu-

I live in New Mexico, so I can't come to your office hours.   The next
time I am scheduled to be in Los Angeles is May 15th, which is after class
is over.

Gina

On Mon, 8 Apr 2002, tipu wrote:

> hi,
> 
> If u have any doubts about ur grade in project A, u are welcome to come at
> SAL on Wednesday. I will be there from 10 to 12 in the morning.
> 
> Quamrul Hasan Tipu
> Graduate Student
> Dept. of Computer Science
> USC.
> 

_______________________________________
Gina Fisk 
Los Alamos National Laboratory
CCS-1 Advanced Computing Group, MS D451
Los Alamos, New Mexico  87545
(work) 505-667-6769, (fax) 505-665-4934


From johnh@ISI.EDU Tue Apr  9 09:20:43 2002
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 g39GKhm25716
	for ; Tue, 9 Apr 2002 09:20:43 -0700 (PDT)
Received: from dash.isi.edu (dash.isi.edu [128.9.160.184])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g39GKhp08785
	for ; Tue, 9 Apr 2002 09:20:43 -0700 (PDT)
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g39GKHo26982;
	Tue, 9 Apr 2002 09:20:17 -0700
Message-Id: <200204091620.g39GKHo26982@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: Gina Fisk 
Cc: tipu , csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Office hours for Project A grading 
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Tue, 09 Apr 2002 09:20:17 -0700
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Tipu, can you post the phone number of the office you'll be in?
(As well as the actual office number :-)

Gina and other remote students---please realize that while the class
has several remote students, it has about 15x more local students, so
we sometimes forget to explicilty mention that remote students are
encouraged to call in during office hours if they have a question.

   -John Heidemann


On Tue, 09 Apr 2002 08:18:06 MDT, Gina Fisk wrote: 
>I live in New Mexico, so I can't come to your office hours.   The next
>time I am scheduled to be in Los Angeles is May 15th, which is after class
>is over.
>
>On Mon, 8 Apr 2002, tipu wrote:
>
>> hi,
>> 
>> If u have any doubts about ur grade in project A, u are welcome to come at
>> SAL on Wednesday. I will be there from 10 to 12 in the morning.
>> 
>> Quamrul Hasan Tipu
>> Graduate Student
>> Dept. of Computer Science
>> USC.
>> 
>
>_______________________________________
>Gina Fisk 
>Los Alamos National Laboratory
>CCS-1 Advanced Computing Group, MS D451
>Los Alamos, New Mexico  87545
>(work) 505-667-6769, (fax) 505-665-4934


From tipu@usc.edu Wed Apr 10 00:47:30 2002
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 g3A7lTm11686
	for ; Wed, 10 Apr 2002 00:47:30 -0700 (PDT)
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3A7lTp17275;
	Wed, 10 Apr 2002 00:47:29 -0700 (PDT)
Received: from aludra.usc.edu (tipu@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id AAA22504; Wed, 10 Apr 2002 00:47:42 -0700 (PDT)
Received: from localhost (tipu@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3A7mUe04719; Wed, 10 Apr 2002 00:48:30 -0700 (PDT)
Date: Wed, 10 Apr 2002 00:48:30 -0700 (PDT)
From: tipu 
To: John Heidemann 
cc: Gina Fisk , 
Subject: Re: [Csci551-talk] Office hours for Project A grading 
In-Reply-To: <200204091620.g39GKHo26982@dash.isi.edu>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Dear sir,

I will in SAL user room. The phone number is 213-740-4440.

Regards,

Quamrul Hasan Tipu
Graduate Student
Dept. of Computer Science
USC.

On Tue, 9 Apr 2002, John Heidemann wrote:

>
> Tipu, can you post the phone number of the office you'll be in?
> (As well as the actual office number :-)
>
> Gina and other remote students---please realize that while the class
> has several remote students, it has about 15x more local students, so
> we sometimes forget to explicilty mention that remote students are
> encouraged to call in during office hours if they have a question.
>
>    -John Heidemann
>
>
> On Tue, 09 Apr 2002 08:18:06 MDT, Gina Fisk wrote:
> >I live in New Mexico, so I can't come to your office hours.   The next
> >time I am scheduled to be in Los Angeles is May 15th, which is after class
> >is over.
> >
> >On Mon, 8 Apr 2002, tipu wrote:
> >
> >> hi,
> >>
> >> If u have any doubts about ur grade in project A, u are welcome to come at
> >> SAL on Wednesday. I will be there from 10 to 12 in the morning.
> >>
> >> Quamrul Hasan Tipu
> >> Graduate Student
> >> Dept. of Computer Science
> >> USC.
> >>
> >
> >_______________________________________
> >Gina Fisk
> >Los Alamos National Laboratory
> >CCS-1 Advanced Computing Group, MS D451
> >Los Alamos, New Mexico  87545
> >(work) 505-667-6769, (fax) 505-665-4934
>


From ubaidd@yahoo.com Sat Apr 13 05:32:12 2002
Received: from web9101.mail.yahoo.com (web9101.mail.yahoo.com [216.136.128.238])
	by gamma.isi.edu (8.11.6/8.11.2) with SMTP id g3DCWCm11053
	for ; Sat, 13 Apr 2002 05:32:12 -0700 (PDT)
Message-ID: <20020413123210.20366.qmail@web9101.mail.yahoo.com>
Received: from [128.125.5.113] by web9101.mail.yahoo.com via HTTP; Sat, 13 Apr 2002 05:32:10 PDT
Date: Sat, 13 Apr 2002 05:32:10 -0700 (PDT)
From: ubaid dhiyan 
To: csci551-talk@mailman.isi.edu
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: [Csci551-talk] terminating condition
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

hi,
for project b, what is the terminating condition for
stages 4 and 5? the getline function returns 0 at the
end of file so there is no way to distinguish between
an end of file and a blank line using getline() (or is
there?) 
i'm currently proceeding on the premise that if there
is a blank line when the parser is reading for stage 4
or stage 5, then the parsing is over. please let me
know if a different implementation is required.

thanks and take care,
ubaid.

__________________________________________________
Do You Yahoo!?
Yahoo! Tax Center - online filing with TurboTax
http://taxes.yahoo.com/

From vkhanna@qualcomm.com Sat Apr 13 15:14:05 2002
Received: from illyana.qualcomm.com (illyana.qualcomm.com [129.46.65.100])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3DME4m07527
	for ; Sat, 13 Apr 2002 15:14:05 -0700 (PDT)
Received: from VKHANNA.qualcomm.com (vkhanna.qualcomm.com [129.46.235.13])
	by illyana.qualcomm.com (8.12.1/8.12.1/1.0) with ESMTP id g3DME2Wg000688;
	Sat, 13 Apr 2002 15:14:02 -0700 (PDT)
Message-Id: <5.1.0.14.2.20020413150757.02ce2db0@illyana>
X-Sender: vkhanna@illyana
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Sat, 13 Apr 2002 15:14:07 -0700
To: csci551-talk@mailman.isi.edu
From: Vivek Khanna 
Cc: vivekkha@usc.edu
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Csci551-talk] Project 1 sample soln does not compile
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

I tried running Project 1 sample solution and I get the following errors on 
running "make all"

g++    -c  gen.cc
g++    -c  main.cc
g++    -c  master.cc
g++    -c  router.cc
g++    -c  msg.cc
g++ -g -Wall -c
g++: No input files
*** Error code 1
make: Fatal error: Command failed for target `timers.o'


Then I tried running "make all" first in timers driectory and I got this error.

g++ -I. -g -gstabs+ -fno-inline  -c tools.cc
g++ -I. -g -gstabs+ -fno-inline  -c timers.cc
g++ -I. -g -gstabs+ -fno-inline  -c test-app.cc
In file included from test-app.cc:23:
test-app.hh:35: parse error before `{'
test-app.hh:42: parse error before `}'
test-app.hh:44: parse error before `{'
test-app.hh:49: parse error before `}'
test-app.cc:38: invalid use of undefined type `class TestTimer1'
test-app.hh:35: forward declaration of `class TestTimer1'
test-app.cc: In method `int TestTimer1::expire(long int)':
test-app.cc:39: `p_' undeclared (first use this function)
test-app.cc:39: (Each undeclared identifier is reported only once
test-app.cc:39: for each function it appears in.)
test-app.cc: At top level:
test-app.cc:44: invalid use of undefined type `class TestTimer2'
test-app.hh:44: forward declaration of `class TestTimer2'
test-app.cc: In method `void TestApp::start()':
test-app.cc:81: no matching function for call to `Timers::ExecuteTimer ()'
test-app.cc: In method `TestApp::TestApp()':
test-app.cc:106: `TimerCallbacks' undeclared (first use this function)
test-app.cc:106: `tcb' undeclared (first use this function)
*** Error code 1
make: Fatal error: Command failed for target `test-app.o'


From samdadiy@usc.edu Sat Apr 13 17:14:58 2002
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3E0Evm25994
	for ; Sat, 13 Apr 2002 17:14:57 -0700 (PDT)
Received: from sal-sun023.usc.edu (sal-sun023.usc.edu [128.125.5.93])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id RAA11696 for ; Sat, 13 Apr 2002 17:15:08 -0700 (PDT)
Received: from localhost (samdadiy@localhost)
	by sal-sun023.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3E0FMW16626 for ; Sat, 13 Apr 2002 17:15:22 -0700 (PDT)
Date: Sat, 13 Apr 2002 17:15:22 -0700 (PDT)
From: Parag Samdadiya 
To: 
In-Reply-To: <5.1.0.14.2.20020413150757.02ce2db0@illyana>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Dynamic port
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,

I am facing problem of finding out the dynamically assigned port number.
Before bind, the sockaddr_in.sin_port should be set to 0.
So the bind will dynamically assign an port.

But how do I retrieve the assigned port number ??

Try to access the sockaddr_in.sin_port after the bind call has value 0
only.

Here is my code.

  my_addr.sin_family = AF_INET;
  my_addr.sin_port   = htons(0);
  my_addr.sin_addr.s_addr = INADDR_ANY;
  memset(&(my_addr.sin_zero), '\0', 8 );

  if( bind(my_udpSockfd, (struct sockaddr *)&my_addr,
                        sizeof(struct sockaddr)) == -1 ) {
      perror("router udp bind");
      exit(1);
  }

  my_udpPort = my_addr.sin_port;
  cout<<"Router "<; Sat, 13 Apr 2002 18:18:03 -0700 (PDT)
Received: from cic-mail.lanl.gov (localhost.localdomain [127.0.0.1])
	by mailrelay2.lanl.gov (8.11.6/8.11.6/(ccn-5)) with ESMTP id g3E1I2C30391;
	Sat, 13 Apr 2002 19:18:02 -0600
Received: from localhost (gina@localhost)
	by cic-mail.lanl.gov (8.11.6/8.11.6/(ccn-5)) with ESMTP id g3E1I2a06353;
	Sat, 13 Apr 2002 19:18:02 -0600
Date: Sat, 13 Apr 2002 19:18:02 -0600 (MDT)
From: Gina Fisk 
To: Parag Samdadiya 
cc: csci551-talk@mailman.isi.edu
Subject: Re: [Csci551-talk] Dynamic port
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

You want to use getsockname(). 

Here is the info from the man page:

NAME
       getsockname - get socket name

SYNOPSIS
       #include 

       int   getsockname(int   s  ,  struct  sockaddr  *  name  ,
       socklen_t * namelen )

DESCRIPTION
       Getsockname returns the current  name  for  the  specified
       socket.   The  namelen  parameter should be initialized to
       indicate the amount of  space  pointed  to  by  name.   On
       return  it  contains  the actual size of the name returned
       (in bytes).

RETURN VALUE
       On success, zero is returned.  On error, -1  is  returned,
       and errno is set appropriately.

I hope this helps...

Gina


On Sat, 13 Apr 2002, Parag Samdadiya wrote:

> 
> Hi,
> 
> I am facing problem of finding out the dynamically assigned port number.
> Before bind, the sockaddr_in.sin_port should be set to 0.
> So the bind will dynamically assign an port.
> 
> But how do I retrieve the assigned port number ??
> 
> Try to access the sockaddr_in.sin_port after the bind call has value 0
> only.
> 
> Here is my code.
> 
>   my_addr.sin_family = AF_INET;
>   my_addr.sin_port   = htons(0);
>   my_addr.sin_addr.s_addr = INADDR_ANY;
>   memset(&(my_addr.sin_zero), '\0', 8 );
> 
>   if( bind(my_udpSockfd, (struct sockaddr *)&my_addr,
>                         sizeof(struct sockaddr)) == -1 ) {
>       perror("router udp bind");
>       exit(1);
>   }
> 
>   my_udpPort = my_addr.sin_port;
>   cout<<"Router "< << " : " << my_udpPort << endl;
> 
> 
> However the value is 0 when printed. Thus how do I find out what
> port my bind function picked ( or the OS assigned )
> 
> Parag
> 

-- 
***********************************************************
** Gina M. Fisk 		     work: (505)667-6769 **               
** Los Alamos National Laboratory     fax: (505)665-4934 **
** CCS-1, MS D451		     home: (505)662-5662 ** 
** Los Alamos, NM  87545	   e-mail: gina@lanl.gov **
***********************************************************


From srangwal@usc.edu Sat Apr 13 18:20:37 2002
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3E1Kbm05431
	for ; Sat, 13 Apr 2002 18:20:37 -0700 (PDT)
Received: from aludra.usc.edu (srangwal@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id SAA05200; Sat, 13 Apr 2002 18:20:47 -0700 (PDT)
Received: from localhost (srangwal@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3E1Le313881; Sat, 13 Apr 2002 18:21:40 -0700 (PDT)
Date: Sat, 13 Apr 2002 18:21:39 -0700 (PDT)
From: Sumit Rangwala 
To: Parag Samdadiya 
cc: 
Subject: Re: [Csci551-talk] Dynamic port
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 


man getsockname


Sumit



On Sat, 13 Apr 2002, Parag Samdadiya wrote:

>
> Hi,
>
> I am facing problem of finding out the dynamically assigned port number.
> Before bind, the sockaddr_in.sin_port should be set to 0.
> So the bind will dynamically assign an port.
>
> But how do I retrieve the assigned port number ??
>
> Try to access the sockaddr_in.sin_port after the bind call has value 0
> only.
>
> Here is my code.
>
>   my_addr.sin_family = AF_INET;
>   my_addr.sin_port   = htons(0);
>   my_addr.sin_addr.s_addr = INADDR_ANY;
>   memset(&(my_addr.sin_zero), '\0', 8 );
>
>   if( bind(my_udpSockfd, (struct sockaddr *)&my_addr,
>                         sizeof(struct sockaddr)) == -1 ) {
>       perror("router udp bind");
>       exit(1);
>   }
>
>   my_udpPort = my_addr.sin_port;
>   cout<<"Router "< << " : " << my_udpPort << endl;
>
>
> However the value is 0 when printed. Thus how do I find out what
> port my bind function picked ( or the OS assigned )
>
> Parag
>


From bhat@usc.edu Sat Apr 13 19:01:49 2002
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3E21mm11807
	for ; Sat, 13 Apr 2002 19:01:48 -0700 (PDT)
Received: from aludra.usc.edu (bhat@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id TAA18692; Sat, 13 Apr 2002 19:01:58 -0700 (PDT)
Received: from localhost (bhat@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3E22pO13058; Sat, 13 Apr 2002 19:02:51 -0700 (PDT)
Date: Sat, 13 Apr 2002 19:02:50 -0700 (PDT)
From: Gajanan Bhat 
To: Parag Samdadiya 
cc: 
Subject: Re: [Csci551-talk] Dynamic port
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

hey parag,

use getsockname() to retrieve the entire structure and then get the port
number from that structure..

gaj


------------------------------------------------------------------------------
Gajanan Bhat
Masters in Electrical Engineering
University of Southern California, Los Angeles
------------------------------------------------------------------------------

On Sat, 13 Apr 2002, Parag Samdadiya wrote:

>
> Hi,
>
> I am facing problem of finding out the dynamically assigned port number.
> Before bind, the sockaddr_in.sin_port should be set to 0.
> So the bind will dynamically assign an port.
>
> But how do I retrieve the assigned port number ??
>
> Try to access the sockaddr_in.sin_port after the bind call has value 0
> only.
>
> Here is my code.
>
>   my_addr.sin_family = AF_INET;
>   my_addr.sin_port   = htons(0);
>   my_addr.sin_addr.s_addr = INADDR_ANY;
>   memset(&(my_addr.sin_zero), '\0', 8 );
>
>   if( bind(my_udpSockfd, (struct sockaddr *)&my_addr,
>                         sizeof(struct sockaddr)) == -1 ) {
>       perror("router udp bind");
>       exit(1);
>   }
>
>   my_udpPort = my_addr.sin_port;
>   cout<<"Router "< << " : " << my_udpPort << endl;
>
>
> However the value is 0 when printed. Thus how do I find out what
> port my bind function picked ( or the OS assigned )
>
> Parag
>


From anshulpa@usc.edu Sun Apr 14 18:29:01 2002
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3F1T1m11264
	for ; Sun, 14 Apr 2002 18:29:01 -0700 (PDT)
Received: from aludra.usc.edu (anshulpa@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id SAA15306; Sun, 14 Apr 2002 18:29:11 -0700 (PDT)
Received: from localhost (anshulpa@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3F1U5N24677; Sun, 14 Apr 2002 18:30:05 -0700 (PDT)
Date: Sun, 14 Apr 2002 18:30:04 -0700 (PDT)
From: anshulpa 
To: 
cc: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] proj 2
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,

I have a question regarding modifications for link up and down. I was not
able to interpret the output for the sample test case. Can you please
explain me the steps , what happens when there are
following statements

w 10
d 1 2
u 1 4
w 30

and what output has to be generated.

I am interpreting it as - 1. When the first statement w 10 is encountered
, I will wait for that much time.

2. for the sequence of all up link and down link messages.. i will pass it
to routers.

3. When manager encounters a w 30 it will send a GO signal , that will
make all the routers start exchanging information

4. the new links will be treated as down at first instance and first
message(ack or advertisement) from them will make  that link up, as
we did in
first project

and one more question - will the first wait time in the input large enough
for everything to converge at teh first instance.

Thanks,
Anshul.


From hussain@ISI.EDU Sun Apr 14 19:03:22 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3F23Lm17277
	for ; Sun, 14 Apr 2002 19:03:21 -0700 (PDT)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g3F23LY21834;
	Sun, 14 Apr 2002 19:03:21 -0700 (PDT)
Date: Sun, 14 Apr 2002 19:03:21 -0700 (PDT)
From: Alefiya Hussain 
Reply-To: Alefiya Hussain 
To: anshulpa 
cc: csci551@usc.edu, csci551-talk@ISI.EDU
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Re: proj 2
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 


On Sun, 14 Apr 2002, anshulpa wrote:

> Hi,
> 
> I have a question regarding modifications for link up and down. I was not
> able to interpret the output for the sample test case. Can you please
> explain me the steps , what happens when there are
> following statements
> 
> w 10
> d 1 2
> u 1 4
> w 30
> 
> and what output has to be generated.
> 
> I am interpreting it as - 
>1. When the first statement w 10 is encountered
> , I will wait for that much time.
>

The I should mean the manager, not routers

> 2. for the sequence of all up link and down link messages.. i will pass it
> to routers.

thats correct

> 3. When manager encounters a w 30 it will send a GO signal , that will
> make all the routers start exchanging information

do you really need this synchronization GO signal ?? 
 
> 4. the new links will be treated as down at first instance and first
> message(ack or advertisement) from them will make  that link up, as
> we did in
> first project

thats correct 

> will the first wait time in the input large enough
> for everything to converge at teh first instance.

yes, it will be large enough to allow convergence. But please ensure that
your implementation does not have large timeout values when not needed. 

all the best
alefiya 


From johnh@ISI.EDU Sun Apr 14 19:17:26 2002
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 g3F2HPm19544
	for ; Sun, 14 Apr 2002 19:17:25 -0700 (PDT)
Received: from hawk.mail.pas.earthlink.net (hawk.mail.pas.earthlink.net [207.217.120.22])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3F2HPp03829
	for ; Sun, 14 Apr 2002 19:17:25 -0700 (PDT)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by hawk.mail.pas.earthlink.net with esmtp (Exim 3.33 #1)
	id 16ww3W-00065M-00; Sun, 14 Apr 2002 19:17:22 -0700
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3F2Gke01222;
	Sun, 14 Apr 2002 19:16:46 -0700
Message-Id: <200204150216.g3F2Gke01222@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: anshulpa 
Cc: csci551@usc.edu, csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] proj 2 
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Sun, 14 Apr 2002 19:16:46 -0700
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Sun, 14 Apr 2002 18:30:04 PDT, anshulpa wrote: 
>Hi,
>
>I have a question regarding modifications for link up and down. I was not
>able to interpret the output for the sample test case. Can you please
>explain me the steps , what happens when there are
>following statements
>
>w 10
>d 1 2
>u 1 4
>w 30
>
>and what output has to be generated.
>
>I am interpreting it as - 1. When the first statement w 10 is encountered
>, I will wait for that much time.
>
>2. for the sequence of all up link and down link messages.. i will pass it
>to routers.

So far so good.

>3. When manager encounters a w 30 it will send a GO signal , that will
>make all the routers start exchanging information

Hmmm... where did you get this "go signal" from?  It's not mentioned
in the assignment, nor in the routing papers we've read.

The way LS routing is supposed to work is that a router sends an LSA
message when any of its links change state.  I recommend you implement
something more along these lines.

>
>4. the new links will be treated as down at first instance and first
>message(ack or advertisement) from them will make  that link up, as
>we did in
>first project

I'm not sure what you mean by this statement.

>
>and one more question - will the first wait time in the input large enough
>for everything to converge at teh first instance.

The first wait time will be at least 30s.
I'm assuming everyones LSA code can converge in that amount of time,
as that allows up to 6 timeouts.  Given that there's no loss in stages
4 and 5, that should be about 6 more than are needed :-)

   -John Heidemann


>
>Thanks,
>Anshul.


From bhat@usc.edu Sun Apr 14 22:53:53 2002
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 g3F5rqm21253
	for ; Sun, 14 Apr 2002 22:53:52 -0700 (PDT)
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3F5rqp09085;
	Sun, 14 Apr 2002 22:53:52 -0700 (PDT)
Received: from sal-sun079.usc.edu (sal-sun079.usc.edu [128.125.5.179])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id WAA29921; Sun, 14 Apr 2002 22:54:03 -0700 (PDT)
Received: from localhost (bhat@localhost)
	by sal-sun079.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3F5sFk23586; Sun, 14 Apr 2002 22:54:15 -0700 (PDT)
Date: Sun, 14 Apr 2002 22:54:15 -0700 (PDT)
From: Gajanan Bhat 
To: Alefiya Hussain 
cc: anshulpa , , 
Subject: Re: [Csci551-talk] Re: proj 2
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

hi...

i think anshul is right as far as the "go" signal is concerned...
cos we cant really assume that the routing tables will converge between
a successive up and down line in the config file...
if u look at the sample output given on the website, u will see that
even there b4 u start printing the routing table u first mention that
a link is down and a link is up, after which the routing tables appear in
the output...

ideally, this shoudlnt happen....whenever a link goes up or down new
routing tables have to be generated....but since in our project the
routing tables wont converge within the time it takes to parse successive
up and down link lines, i suppose we can assume that the "w" character in
the config file can be thought of as a "go" signal to the routers to start
generating the routing tables...


gajanan


------------------------------------------------------------------------------
Gajanan Bhat
Masters in Electrical Engineering
University of Southern California, Los Angeles
------------------------------------------------------------------------------

On Sun, 14 Apr 2002, Alefiya Hussain wrote:

>
>
> On Sun, 14 Apr 2002, anshulpa wrote:
>
> > Hi,
> >
> > I have a question regarding modifications for link up and down. I was not
> > able to interpret the output for the sample test case. Can you please
> > explain me the steps , what happens when there are
> > following statements
> >
> > w 10
> > d 1 2
> > u 1 4
> > w 30
> >
> > and what output has to be generated.
> >
> > I am interpreting it as -
> >1. When the first statement w 10 is encountered
> > , I will wait for that much time.
> >
>
> The I should mean the manager, not routers
>
> > 2. for the sequence of all up link and down link messages.. i will pass it
> > to routers.
>
> thats correct
>
> > 3. When manager encounters a w 30 it will send a GO signal , that will
> > make all the routers start exchanging information
>
> do you really need this synchronization GO signal ??
>
> > 4. the new links will be treated as down at first instance and first
> > message(ack or advertisement) from them will make  that link up, as
> > we did in
> > first project
>
> thats correct
>
> > will the first wait time in the input large enough
> > for everything to converge at teh first instance.
>
> yes, it will be large enough to allow convergence. But please ensure that
> your implementation does not have large timeout values when not needed.
>
> all the best
> alefiya
>


From kulshres@usc.edu Tue Apr 16 20:00:04 2002
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3H304m19855
	for ; Tue, 16 Apr 2002 20:00:04 -0700 (PDT)
Received: from sal-sun059.usc.edu (sal-sun059.usc.edu [128.125.5.158])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id UAA23374 for ; Tue, 16 Apr 2002 20:00:14 -0700 (PDT)
Received: from localhost (kulshres@localhost)
	by sal-sun059.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3H30tQ14644 for ; Tue, 16 Apr 2002 20:00:55 -0700 (PDT)
Date: Tue, 16 Apr 2002 20:00:55 -0700 (PDT)
From: kulshres 
To: 
In-Reply-To: <200204151905.g3FJ52m19542@gamma.isi.edu>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] (no subject)
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,
Can anyone tell me how can we remove the problem of "Broken Pipe" which
comes due to SIGPIPE signal. I am getting this problem while writing the
information from .conf file to the routers.

Thanks,
Jitendra.


From ganglu@usc.edu Wed Apr 17 01:36:23 2002
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 g3H8aNm09300
	for ; Wed, 17 Apr 2002 01:36:23 -0700 (PDT)
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3H8aNp25236
	for ; Wed, 17 Apr 2002 01:36:23 -0700 (PDT)
Received: from aludra.usc.edu (ganglu@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id BAA24995 for ; Wed, 17 Apr 2002 01:36:33 -0700 (PDT)
Received: from localhost (ganglu@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3H8bRt28090 for ; Wed, 17 Apr 2002 01:37:27 -0700 (PDT)
Date: Wed, 17 Apr 2002 01:37:27 -0700 (PDT)
From: ganglu 
To: 
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] how to reject duplicat RREQ
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,

By the DSR paper, a node will discard duplicate RREQ.

If a node receive a RREQ with longer hops first, then it will
discard the same RREQ even it has shorter hops.
This could happen in the project because the process schedule.
So there maybe a chance that the path found is not
a shortest hop path.

Another question, the calfification 12-apr-02 says:
You should generate a "new RREP" message for each RREQ message you get.

But if we apply the "drop duplicate RREQ", the node would only
generate a RREP for the first RREQ it received.

thanks,

Gang Lu


From tipu@usc.edu Wed Apr 17 02:34:23 2002
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3H9YNm17727
	for ; Wed, 17 Apr 2002 02:34:23 -0700 (PDT)
Received: from aludra.usc.edu (tipu@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id CAA28648 for ; Wed, 17 Apr 2002 02:34:33 -0700 (PDT)
Received: from localhost (tipu@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3H9ZS327383 for ; Wed, 17 Apr 2002 02:35:28 -0700 (PDT)
Date: Wed, 17 Apr 2002 02:35:28 -0700 (PDT)
From: tipu 
To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Office hour for grader...
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

I will be in the TA's office from 2:30pm to 3:30pm on 17th April,
Wednesday..

If u have any doubt regarding project A or hw2, u r welcomed.

Quamrul Hasan Tipu
Graduate Student
Dept. of Computer Science
USC.


From srangwal@usc.edu Wed Apr 17 14:41:38 2002
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3HLfZm27398
	for ; Wed, 17 Apr 2002 14:41:37 -0700 (PDT)
Received: from aludra.usc.edu (srangwal@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id OAA25729; Wed, 17 Apr 2002 14:41:43 -0700 (PDT)
Received: from localhost (srangwal@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3HLgYX08947; Wed, 17 Apr 2002 14:42:34 -0700 (PDT)
Date: Wed, 17 Apr 2002 14:42:34 -0700 (PDT)
From: Sumit Rangwala 
To: kulshres 
cc: 
Subject: Re: [Csci551-talk] (no subject
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

In our project this occurs when one write to a socket that had been
closed by the peer. Just make sure that the socket is  not closed before
all the communication is over.


Sumit


On Tue, 16 Apr 2002, kulshres wrote:

> Hi,
> Can anyone tell me how can we remove the problem of "Broken Pipe" which
> comes due to SIGPIPE signal. I am getting this problem while writing the
> information from .conf file to the routers.
>
> Thanks,
> Jitendra.
>


From kaiyuxia@usc.edu Wed Apr 17 21:37:56 2002
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3I4bum10489
	for ; Wed, 17 Apr 2002 21:37:56 -0700 (PDT)
Received: from aludra.usc.edu (kaiyuxia@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id VAA21641 for ; Wed, 17 Apr 2002 21:38:03 -0700 (PDT)
Received: from localhost (kaiyuxia@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3I4cqL00535 for ; Wed, 17 Apr 2002 21:38:52 -0700 (PDT)
Date: Wed, 17 Apr 2002 21:38:52 -0700 (PDT)
From: kaiyuxia 
To: 
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Question about Proj B
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi, Prof. & TA,

In the sample conf file, there are the following messages:
************************
# now change some stuff
d 1 3
u 1 4
************************
In the sample output file, there are corresponding lines:
************************
---
1 10.1.0.0/16 1 0
2 10.2.0.0/16 2 1
3 192.168.123.0/24 3 1
4 192.168.101.0/24 2 2
5 192.168.200.128/25 3 2
---
link down to 3

link up to 4

*************************

It looks like that router 1 gets two messages (a link-down and a link-up)
at the same time. Can we assume that messages between two adjacent "w n"
will  be sent and received at the same time?  ( I hope we can make
this assumption, as it does in the sample. It makes things simple :) )

Thanks and best regards,

Kaiyu Xiao




From bhat@usc.edu Thu Apr 18 09:17:26 2002
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3IGHPm11216
	for ; Thu, 18 Apr 2002 09:17:25 -0700 (PDT)
Received: from aludra.usc.edu (bhat@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id JAA22496; Thu, 18 Apr 2002 09:17:33 -0700 (PDT)
Received: from localhost (bhat@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3IGIUl18257; Thu, 18 Apr 2002 09:18:30 -0700 (PDT)
Date: Thu, 18 Apr 2002 09:18:30 -0700 (PDT)
From: Gajanan Bhat 
To: kaiyuxia 
cc: 
Subject: Re: [Csci551-talk] Question about Proj B
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

hi...

what i have assumed is that for each up link or down link the concerned
routers send LSA packets.....so my o/p is a bit different in the sense
that i first get a link up statement in my o/p file and then a routing
table entry/entries and then a link down statement and then sum more
routing table entries....

can i make this assumption??

gajanan

------------------------------------------------------------------------------
Gajanan Bhat
Masters in Electrical Engineering
University of Southern California, Los Angeles
------------------------------------------------------------------------------

On Wed, 17 Apr 2002, kaiyuxia wrote:

> Hi, Prof. & TA,
>
> In the sample conf file, there are the following messages:
> ************************
> # now change some stuff
> d 1 3
> u 1 4
> ************************
> In the sample output file, there are corresponding lines:
> ************************
> ---
> 1 10.1.0.0/16 1 0
> 2 10.2.0.0/16 2 1
> 3 192.168.123.0/24 3 1
> 4 192.168.101.0/24 2 2
> 5 192.168.200.128/25 3 2
> ---
> link down to 3
>
> link up to 4
>
> *************************
>
> It looks like that router 1 gets two messages (a link-down and a link-up)
> at the same time. Can we assume that messages between two adjacent "w n"
> will  be sent and received at the same time?  ( I hope we can make
> this assumption, as it does in the sample. It makes things simple :) )
>
> Thanks and best regards,
>
> Kaiyu Xiao
>
>
>


From hussain@ISI.EDU Thu Apr 18 09:21:23 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3IGLNm12996
	for ; Thu, 18 Apr 2002 09:21:23 -0700 (PDT)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g3IGLK018675;
	Thu, 18 Apr 2002 09:21:20 -0700 (PDT)
Date: Thu, 18 Apr 2002 09:21:20 -0700 (PDT)
From: Alefiya Hussain 
To: kaiyuxia 
cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Question about Proj B
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 


On Wed, 17 Apr 2002, kaiyuxia wrote:

> Hi, Prof. & TA,
> 
> In the sample conf file, there are the following messages:
> ************************
> # now change some stuff
> d 1 3
> u 1 4
> ************************
> In the sample output file, there are corresponding lines:
> ************************
> ---
> 1 10.1.0.0/16 1 0
> 2 10.2.0.0/16 2 1
> 3 192.168.123.0/24 3 1
> 4 192.168.101.0/24 2 2
> 5 192.168.200.128/25 3 2
> ---
> link down to 3
> 
> link up to 4
> 
> *************************
> 
> It looks like that router 1 gets two messages (a link-down and a link-up)
> at the same time. Can we assume that messages between two adjacent "w n"
> will  be sent and received at the same time?  ( I hope we can make
> this assumption, as it does in the sample. It makes things simple :) )
> 
> Thanks and best regards,
> 
> Kaiyu Xiao
> 

There was a discussion regarding this same observation last week on the
mailing list. I would recommend you refer to it. 

However i will repeat the main conclusion here. 

"There is no  requirement that the link-up link-down lines be adjacent.
I don't think it's good to add a  signal.
Just show the transient results your routers get." 
			   -John Heidemann


all the best
alefiya 


From renuka_kumar@hotmail.com Fri Apr 19 01:08:36 2002
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 g3J88am05177
	for ; Fri, 19 Apr 2002 01:08:36 -0700 (PDT)
Received: from hotmail.com (f214.law10.hotmail.com [64.4.15.214])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3J88Zp11347
	for ; Fri, 19 Apr 2002 01:08:35 -0700 (PDT)
Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC;
	 Fri, 19 Apr 2002 01:08:30 -0700
Received: from 172.191.240.133 by lw10fd.law10.hotmail.msn.com with HTTP;
	Fri, 19 Apr 2002 08:08:30 GMT
X-Originating-IP: [172.191.240.133]
From: "renuka kumar" 
To: csci551@usc.edu
Cc: csci551-talk@ISI.EDU
Date: Fri, 19 Apr 2002 13:38:30 +0530
Mime-Version: 1.0
Content-Type: text/plain; format=flowed
Message-ID: 
X-OriginalArrivalTime: 19 Apr 2002 08:08:30.0751 (UTC) FILETIME=[64932AF0:01C1E779]
Subject: [Csci551-talk] partitions ?
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 


Hi,

suppose u have a topology like
1 2
1 5
2 5
2 3
3 4
if the link between 3 and 4 goes down, 4 get cut off from the network .
is it required to handle such partition, can it be assumed that there always 
an alternative path to a router. in this case , how would routers 1 and 5 
come to know about the partition ??
can anybody help me approach this problem ?

thanks,
renuka


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


From qcao@usc.edu Fri Apr 19 07:15:03 2002
Received: from usc.edu (root@usc.edu [128.125.253.136])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3JEF3m00080
	for ; Fri, 19 Apr 2002 07:15:03 -0700 (PDT)
Received: from nunki.usc.edu (qcao@nunki.usc.edu [128.125.5.168])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id HAA11456 for ; Fri, 19 Apr 2002 07:15:11 -0700 (PDT)
Received: from localhost (qcao@localhost)
	by nunki.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3JEFlh25639 for ; Fri, 19 Apr 2002 07:15:47 -0700 (PDT)
Date: Fri, 19 Apr 2002 07:15:47 -0700 (PDT)
From: qcao 
To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] can't compine proj1a-sol
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi,

I tried to play with professor's code. But I couldn't compine(make) it,
seems like some library problem. Anybody met the same problem?

That's what I got:

nunki.usc.edu(78): make
CC    -c  gen.cc
"gen.hh", line 29: Error: Could not open include file .
"gen.hh", line 30: Error: Could not open include file .
"gen.hh", line 51: Error: Type name expected instead of "bool".
"gen.hh", line 51: Error: No storage class or type for this declaration.
"gen.hh", line 51: Error: "," expected instead of "operator".
"gen.hh", line 52: Error: Use ";" to terminate declarations.
"gen.hh", line 52: Error: Type name expected instead of "bool".
"gen.hh", line 52: Error: No storage class or type for this declaration.
"gen.hh", line 52: Error: Multiple declaration for bool.
"gen.hh", line 52: Error: "," expected instead of "operator".
"gen.hh", line 58: Error: Use ";" to terminate declarations.
"gen.hh", line 58: Warning (Anachronism): Class friends require an
explicit "class".
"gen.hh", line 58: Note: Type "CC -migration" for more on anachronisms.
"gen.hh", line 58: Error: "," expected instead of "&".
"gen.hh", line 89: Error: Type name expected instead of "bool".
"gen.hh", line 89: Error: No storage class or type for this declaration.
"gen.hh", line 89: Error: "," expected instead of "operator".
"gen.hh", line 91: Error: Use ";" to terminate declarations.
"gen.hh", line 95: Error: Type name expected instead of "bool".
"gen.hh", line 95: Error: No storage class or type for this declaration.
"gen.hh", line 95: Error: "," expected instead of "operator".
"gen.hh", line 101: Error: Type name expected instead of "list".
"gen.hh", line 101: Error: No storage class or type for this declaration.
"gen.hh", line 101: Error: "," expected instead of "internal_routes_".
"gen.hh", line 102: Error: Use ";" to terminate declarations.
"gen.hh", line 102: Error: Type name expected instead of "list".
"gen.hh", line 102: Error: No storage class or type for this declaration.
Compilation aborted, too many Error messages.
*** Error code 1



From hussain@ISI.EDU Fri Apr 19 07:39:55 2002
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 g3JEdsm05526
	for ; Fri, 19 Apr 2002 07:39:54 -0700 (PDT)
Received: from mat.isi.edu (mat.isi.edu [128.9.160.109])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3JEdsp29413
	for ; Fri, 19 Apr 2002 07:39:54 -0700 (PDT)
Received: from localhost (hussain@localhost)
	by mat.isi.edu (8.11.6/8.11.6) with ESMTP id g3JEdoP13523;
	Fri, 19 Apr 2002 07:39:50 -0700
Date: Fri, 19 Apr 2002 07:39:50 -0700 (PDT)
From: Alefiya Hussain 
To: renuka kumar 
cc: , 
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Re: partitions ?
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Fri, 19 Apr 2002, renuka kumar wrote:

> 
> 
> Hi,
> 
> suppose u have a topology like
> 1 2
> 1 5
> 2 5
> 2 3
> 3 4
> if the link between 3 and 4 goes down, 4 get cut off from the network .
> is it required to handle such partition, can it be assumed that there always 
> an alternative path to a router. in this case , how would routers 1 and 5 
> come to know about the partition ??
> can anybody help me approach this problem ?
> 
> thanks,
> renuka
> 


We are not required to handle partitioned networks. 

alefiya


From hussain@ISI.EDU Fri Apr 19 07:43:24 2002
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 g3JEhOm05762
	for ; Fri, 19 Apr 2002 07:43:24 -0700 (PDT)
Received: from mat.isi.edu (mat.isi.edu [128.9.160.109])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3JEhNp00249
	for ; Fri, 19 Apr 2002 07:43:23 -0700 (PDT)
Received: from localhost (hussain@localhost)
	by mat.isi.edu (8.11.6/8.11.6) with ESMTP id g3JEhNE13533;
	Fri, 19 Apr 2002 07:43:23 -0700
Date: Fri, 19 Apr 2002 07:43:23 -0700 (PDT)
From: Alefiya Hussain 
To: qcao 
cc: 
Subject: Re: [Csci551-talk] can't compine proj1a-sol
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Have you used the Makefile given on the CS551 hints webpage ? 

alefiya 

On Fri, 19 Apr 2002, qcao wrote:

> Hi,
> 
> I tried to play with professor's code. But I couldn't compine(make) it,
> seems like some library problem. Anybody met the same problem?
> 
> That's what I got:
> 
> nunki.usc.edu(78): make
> CC    -c  gen.cc
> "gen.hh", line 29: Error: Could not open include file .
> "gen.hh", line 30: Error: Could not open include file .
> "gen.hh", line 51: Error: Type name expected instead of "bool".
> "gen.hh", line 51: Error: No storage class or type for this declaration.
> "gen.hh", line 51: Error: "," expected instead of "operator".
> "gen.hh", line 52: Error: Use ";" to terminate declarations.
> "gen.hh", line 52: Error: Type name expected instead of "bool".
> "gen.hh", line 52: Error: No storage class or type for this declaration.
> "gen.hh", line 52: Error: Multiple declaration for bool.
> "gen.hh", line 52: Error: "," expected instead of "operator".
> "gen.hh", line 58: Error: Use ";" to terminate declarations.
> "gen.hh", line 58: Warning (Anachronism): Class friends require an
> explicit "class".
> "gen.hh", line 58: Note: Type "CC -migration" for more on anachronisms.
> "gen.hh", line 58: Error: "," expected instead of "&".
> "gen.hh", line 89: Error: Type name expected instead of "bool".
> "gen.hh", line 89: Error: No storage class or type for this declaration.
> "gen.hh", line 89: Error: "," expected instead of "operator".
> "gen.hh", line 91: Error: Use ";" to terminate declarations.
> "gen.hh", line 95: Error: Type name expected instead of "bool".
> "gen.hh", line 95: Error: No storage class or type for this declaration.
> "gen.hh", line 95: Error: "," expected instead of "operator".
> "gen.hh", line 101: Error: Type name expected instead of "list".
> "gen.hh", line 101: Error: No storage class or type for this declaration.
> "gen.hh", line 101: Error: "," expected instead of "internal_routes_".
> "gen.hh", line 102: Error: Use ";" to terminate declarations.
> "gen.hh", line 102: Error: Type name expected instead of "list".
> "gen.hh", line 102: Error: No storage class or type for this declaration.
> Compilation aborted, too many Error messages.
> *** Error code 1
> 
> 


From johnh@ISI.EDU Fri Apr 19 23:04:02 2002
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 g3K642m06732
	for ; Fri, 19 Apr 2002 23:04:02 -0700 (PDT)
Received: from pintail.mail.pas.earthlink.net (pintail.mail.pas.earthlink.net [207.217.120.122])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3K641p01929
	for ; Fri, 19 Apr 2002 23:04:01 -0700 (PDT)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by pintail.mail.pas.earthlink.net with esmtp (Exim 3.33 #2)
	id 16ynyb-0002t8-00; Fri, 19 Apr 2002 23:04:01 -0700
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3K63AN14605;
	Fri, 19 Apr 2002 23:03:19 -0700
Message-Id: <200204200603.g3K63AN14605@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: ubaid dhiyan 
Cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] terminating condition 
In-reply-to: <20020413123210.20366.qmail@web9101.mail.yahoo.com> 
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Fri, 19 Apr 2002 23:03:10 -0700
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Sat, 13 Apr 2002 05:32:10 PDT, ubaid dhiyan wrote: 
>hi,
>for project b, what is the terminating condition for
>stages 4 and 5? the getline function returns 0 at the
>end of file so there is no way to distinguish between
>an end of file and a blank line using getline() (or is
>there?) 
>i'm currently proceeding on the premise that if there
>is a blank line when the parser is reading for stage 4
>or stage 5, then the parsing is over. please let me
>know if a different implementation is required.

The master should determine there are no more lines by detecting
end-of-file on the input.

Blank lines are not part of the spec, so your assumption probably
won't cause you grief, but you're better off detecting real
end-of-file.

If you're referring to the C++ streams getline function, then you're
right it doesn't detect eof.  However, there is another function that
DOES detect eof that you should consider.

   -John Heidemann

From johnh@ISI.EDU Fri Apr 19 23:18:34 2002
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 g3K6IYm08718
	for ; Fri, 19 Apr 2002 23:18:34 -0700 (PDT)
Received: from avocet.prod.itd.earthlink.net (avocet.mail.pas.earthlink.net [207.217.120.50])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3K6IXp04629;
	Fri, 19 Apr 2002 23:18:33 -0700 (PDT)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by avocet.prod.itd.earthlink.net with esmtp (Exim 3.33 #2)
	id 16yoCf-00072F-00; Fri, 19 Apr 2002 23:18:33 -0700
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3K6HTI14650;
	Fri, 19 Apr 2002 23:17:29 -0700
Message-Id: <200204200617.g3K6HTI14650@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: Gajanan Bhat 
Cc: Alefiya Hussain , anshulpa ,
   csci551@usc.edu, csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Re: proj 2 
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Fri, 19 Apr 2002 23:17:29 -0700
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Sun, 14 Apr 2002 22:54:15 PDT, Gajanan Bhat wrote: 
>hi...
>
>i think anshul is right as far as the "go" signal is concerned...
>cos we cant really assume that the routing tables will converge between
>a successive up and down line in the config file...
>if u look at the sample output given on the website, u will see that
>even there b4 u start printing the routing table u first mention that
>a link is down and a link is up, after which the routing tables appear in
>the output...
>
>ideally, this shoudlnt happen....whenever a link goes up or down new
>routing tables have to be generated....but since in our project the
>routing tables wont converge within the time it takes to parse successive
>up and down link lines, i suppose we can assume that the "w" character in
>the config file can be thought of as a "go" signal to the routers to start
>generating the routing tables...

No, what you're saying about a go signal is NOT what's required by the
project specs.

To clarify:

There's a difference between when routing tables *converge* and when
they should be printed.

You should print a new routing table every time it's generated, i.e.,
after reciept of each LSA mesage.

The routing tables converge when they stop changing, i.e., when there
are no more LSA messages.

You should NOT wait for the tables to converge before printing them,
this will be INCORRECT output (since you're explicilty required to
print out intermediate stages).


To clarify what Project B IS looking for, wrt the sample output:

There is *no requirement* to make
the link down and link up messages appear consecuitively as in the
above sample output.  It is acceptable (and, in fact, preferrable) to
have a new routing table between these messages if that's convenient
in your implementation.  (For my implement there is a 1ms delay before
computing a routing table, allowing these messages to appear
sequentially.)  *Either* output is acceptable.  (Similarly, you
may get routing tables if your LSA messages are processed in different
orders.  This kind of minor variation is fine.)


   -John Heidemann

From johnh@ISI.EDU Fri Apr 19 23:53:00 2002
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 g3K6qxm13687
	for ; Fri, 19 Apr 2002 23:52:59 -0700 (PDT)
Received: from albatross.prod.itd.earthlink.net (albatross.mail.pas.earthlink.net [207.217.120.120])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3K6qxp11637;
	Fri, 19 Apr 2002 23:52:59 -0700 (PDT)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by albatross.prod.itd.earthlink.net with esmtp (Exim 3.33 #2)
	id 16yojy-0001l0-00; Fri, 19 Apr 2002 23:52:59 -0700
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3K6qGM14919;
	Fri, 19 Apr 2002 23:52:16 -0700
Message-Id: <200204200652.g3K6qGM14919@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: Alefiya Hussain 
Cc: kaiyuxia , csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Question about Proj B 
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Fri, 19 Apr 2002 23:52:16 -0700
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Thu, 18 Apr 2002 09:21:20 PDT, Alefiya Hussain wrote: 
>
>
>On Wed, 17 Apr 2002, kaiyuxia wrote:
>
>> Hi, Prof. & TA,
>> 
>> In the sample conf file, there are the following messages:
>> ************************
>> # now change some stuff
>> d 1 3
>> u 1 4
>> ************************
>> In the sample output file, there are corresponding lines:
>> ************************
>> ---
>> 1 10.1.0.0/16 1 0
>> 2 10.2.0.0/16 2 1
>> 3 192.168.123.0/24 3 1
>> 4 192.168.101.0/24 2 2
>> 5 192.168.200.128/25 3 2
>> ---
>> link down to 3
>> 
>> link up to 4
>> 
>> *************************
>> 
>> It looks like that router 1 gets two messages (a link-down and a link-up)
>> at the same time. Can we assume that messages between two adjacent "w n"
>> will  be sent and received at the same time?  ( I hope we can make
>> this assumption, as it does in the sample. It makes things simple :) )
>> 
>> Thanks and best regards,
>> 
>> Kaiyu Xiao
>> 
>
>There was a discussion regarding this same observation last week on the
>mailing list. I would recommend you refer to it. 
>
>However i will repeat the main conclusion here. 
>
>"There is no  requirement that the link-up link-down lines be adjacent.
>I don't think it's good to add a  signal.
>Just show the transient results your routers get." 
>			   -John Heidemann

And more specifically to your question:
Rather than assume the messages arrive at the same time,
I'd prefer that you treat them separately
and print out a new routing table between them.
I believe that that will also make things simple,
but a better kind of simple :-)

   -John Heidemann

From johnh@ISI.EDU Fri Apr 19 23:54:18 2002
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 g3K6sIm14077
	for ; Fri, 19 Apr 2002 23:54:18 -0700 (PDT)
Received: from hawk.mail.pas.earthlink.net (hawk.mail.pas.earthlink.net [207.217.120.22])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3K6sIp12299
	for ; Fri, 19 Apr 2002 23:54:18 -0700 (PDT)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by hawk.mail.pas.earthlink.net with esmtp (Exim 3.33 #2)
	id 16yolF-0003QW-00; Fri, 19 Apr 2002 23:54:17 -0700
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3K6rZg14938;
	Fri, 19 Apr 2002 23:53:35 -0700
Message-Id: <200204200653.g3K6rZg14938@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: Gajanan Bhat 
Cc: kaiyuxia , csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Question about Proj B 
In-reply-to:  
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Fri, 19 Apr 2002 23:53:34 -0700
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Thu, 18 Apr 2002 09:18:30 PDT, Gajanan Bhat wrote: 
>hi...
>
>what i have assumed is that for each up link or down link the concerned
>routers send LSA packets.....so my o/p is a bit different in the sense
>that i first get a link up statement in my o/p file and then a routing
>table entry/entries and then a link down statement and then sum more
>routing table entries....
>
>can i make this assumption??

Yes you can make this assumption.

In fact, I prefer that simplification to the one suggested below
(making the messages happen "at the same time").

   -John Heidemann

>
>gajanan
>
>------------------------------------------------------------------------------
>Gajanan Bhat
>Masters in Electrical Engineering
>University of Southern California, Los Angeles
>------------------------------------------------------------------------------
>
>On Wed, 17 Apr 2002, kaiyuxia wrote:
>
>> Hi, Prof. & TA,
>>
>> In the sample conf file, there are the following messages:
>> ************************
>> # now change some stuff
>> d 1 3
>> u 1 4
>> ************************
>> In the sample output file, there are corresponding lines:
>> ************************
>> ---
>> 1 10.1.0.0/16 1 0
>> 2 10.2.0.0/16 2 1
>> 3 192.168.123.0/24 3 1
>> 4 192.168.101.0/24 2 2
>> 5 192.168.200.128/25 3 2
>> ---
>> link down to 3
>>
>> link up to 4
>>
>> *************************
>>
>> It looks like that router 1 gets two messages (a link-down and a link-up)
>> at the same time. Can we assume that messages between two adjacent "w n"
>> will  be sent and received at the same time?  ( I hope we can make
>> this assumption, as it does in the sample. It makes things simple :) )
>>
>> Thanks and best regards,
>>
>> Kaiyu Xiao
>>
>>
>>


From johnh@ISI.EDU Sat Apr 20 00:05:23 2002
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 g3K75Nm15497
	for ; Sat, 20 Apr 2002 00:05:23 -0700 (PDT)
Received: from swan.prod.itd.earthlink.net (swan.mail.pas.earthlink.net [207.217.120.123])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3K75Mp14753
	for ; Sat, 20 Apr 2002 00:05:22 -0700 (PDT)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by swan.prod.itd.earthlink.net with esmtp (Exim 3.33 #2)
	id 16yovy-0006sb-00
	for csci551-talk@isi.edu; Sat, 20 Apr 2002 00:05:22 -0700
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3K74eh15082
	for ; Sat, 20 Apr 2002 00:04:40 -0700
Message-Id: <200204200704.g3K74eh15082@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: csci551-talk@ISI.EDU
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Sat, 20 Apr 2002 00:04:40 -0700
From: John Heidemann 
Subject: [Csci551-talk] project b clarifications and hw3 on the web
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Another (and final!) round of clarifications for Project B was posted
to the web page.

There is one addition: the README file must contain a "what works"
section.  Hopefully this will just by "all" for most people.
See the updated Project spec for details.

In addition, HW3 is now on the web.

   -John Heidemann

From bhat@usc.edu Sat Apr 20 01:54:01 2002
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 g3K8s0m03316
	for ; Sat, 20 Apr 2002 01:54:00 -0700 (PDT)
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3K8s0p07844;
	Sat, 20 Apr 2002 01:54:00 -0700 (PDT)
Received: from sal-sun078.usc.edu (sal-sun078.usc.edu [128.125.5.178])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id BAA16253; Sat, 20 Apr 2002 01:54:06 -0700 (PDT)
Received: from localhost (bhat@localhost)
	by sal-sun078.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3K8sC319859; Sat, 20 Apr 2002 01:54:12 -0700 (PDT)
Date: Sat, 20 Apr 2002 01:54:12 -0700 (PDT)
From: Gajanan Bhat 
To: Alefiya Hussain 
cc: 
Subject: Re: [Csci551-talk] Clarification.
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

hi...

i m quoting this from the website...


"When the router that originated the RREQ receives its RREP it should
output the line ``received
RREP for host h1 to host h2: a,b,c,...z''. Where a...z is the same forward
path from h1's router to
h2's router.

When a router originates a data packet it should include the line
``sending data from host h1 to
host h2'' where h1 and h2 are the IP addresses.

When a router terminates a data packet it should print the line ``received
data from host h1 to host
h2 via SR a,b,c,...z'' Where a...z is the set of routers recorded as the
data packet traveled. "


do we have to write these messages in our output files??
cos the sample output doesnt have these lines....


gajanan




From hussain@ISI.EDU Sat Apr 20 13:12:12 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3KKCCm18366
	for ; Sat, 20 Apr 2002 13:12:12 -0700 (PDT)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g3KKCBV02363;
	Sat, 20 Apr 2002 13:12:11 -0700 (PDT)
Date: Sat, 20 Apr 2002 13:12:11 -0700 (PDT)
From: Alefiya Hussain 
To: Gajanan Bhat 
cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Clarification.
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 


On Sat, 20 Apr 2002, Gajanan Bhat wrote:

> 
> hi...
> 
> i m quoting this from the website...
> 
> 
> "When the router that originated the RREQ receives its RREP it should
> output the line ``received
> RREP for host h1 to host h2: a,b,c,...z''. Where a...z is the same forward
> path from h1's router to
> h2's router.
> 
> When a router originates a data packet it should include the line
> ``sending data from host h1 to
> host h2'' where h1 and h2 are the IP addresses.
> 
> When a router terminates a data packet it should print the line ``received
> data from host h1 to host
> h2 via SR a,b,c,...z'' Where a...z is the set of routers recorded as the
> data packet traveled. "
> 
> 
> do we have to write these messages in our output files??
> cos the sample output doesnt have these lines....
> 
> 
> gajanan
> 
> 
> 

Hi Gajanan,

The project has been updated since then. These are no longer project
requirements. please refer to the current specification. It is much
simpler.

alefiya


From johnh@ISI.EDU Sat Apr 20 15:29:03 2002
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 g3KMT2m08092
	for ; Sat, 20 Apr 2002 15:29:02 -0700 (PDT)
Received: from hawk.mail.pas.earthlink.net (hawk.mail.pas.earthlink.net [207.217.120.22])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3KMSvp21180
	for ; Sat, 20 Apr 2002 15:28:57 -0700 (PDT)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by hawk.mail.pas.earthlink.net with esmtp (Exim 3.33 #2)
	id 16z3Ll-0006JK-00
	for csci551-talk@isi.edu; Sat, 20 Apr 2002 15:28:57 -0700
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3KMSDu17698
	for ; Sat, 20 Apr 2002 15:28:13 -0700
Message-Id: <200204202228.g3KMSDu17698@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: csci551-talk@ISI.EDU
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Sat, 20 Apr 2002 15:28:13 -0700
From: John Heidemann 
Subject: [Csci551-talk] bug in HTML version (not PDF) of project B spec fixed
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

There was a bug in the HTML version (not the PDF version) of Project B
that indcated that one must actually send RREP messages.  Actual RREP
messages do not need to be sent.  This text has been fixed in both
versions of Project B now.

If you were looking at the HTML version, please check out the current
fixed version.

   -John Heidemann
    

From kaiyuxia@usc.edu Sun Apr 21 11:20:00 2002
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 g3LIK0m26669
	for ; Sun, 21 Apr 2002 11:20:00 -0700 (PDT)
Received: from usc.edu (root@usc.edu [128.125.253.136] (may be forged))
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3LIJxp10079;
	Sun, 21 Apr 2002 11:19:59 -0700 (PDT)
Received: from aludra.usc.edu (kaiyuxia@aludra.usc.edu [128.125.253.184])
	by usc.edu (8.9.3.1/8.9.3/usc) with ESMTP
	id LAA24489; Sun, 21 Apr 2002 11:20:06 -0700 (PDT)
Received: from localhost (kaiyuxia@localhost)
	by aludra.usc.edu (8.10.1/8.10.1/usc) with ESMTP
	id g3LIL7S06236; Sun, 21 Apr 2002 11:21:07 -0700 (PDT)
Date: Sun, 21 Apr 2002 11:21:06 -0700 (PDT)
From: kaiyuxia 
To: John Heidemann 
cc: Alefiya Hussain , 
In-Reply-To: <200204200652.g3K6qGM14919@dash.isi.edu>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Subject: [Csci551-talk] Several errors in the sample output of Proj B
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Hi, Prof. & TA,

I think the following results are wrong in the sample output file:
(assume we are using the same manager.conf file and have the same event:
original connection:
1 2
1 3
2 4
3 4
3 5
4 5
events:
d 1 3
u 1 4

So finally we get something like:
1 2
1 4
2 4
3 4
3 5
4 5
)

For s4-2.out,
3 192.168.123.0/24 1 2
wrong because now 1 has no connection with 3.
Should be: 3 192.168.123.0/24 4 2

For s4-3.out:
1 10.1.0.0/16 1 1
wrong because 3 has no direct connection with 3 now.
Should be:  1 10.1.0.0/16        4 2

For s4-4.out:
1 10.1.0.0/16           2 2
wrong because 4 now has direct connection with 4.
Should be: 1 10.1.0.0/16        1 1

By the way, if we treat the link down and up events as independent, (as
one of your previous email suggested ), the LSA packets number might be
different than the sample output. Because:
When router 1 & 3 got the link down event, they do not know that the link
between 1 & 4 is established (which will happen later). So router 1 will
not be able to flood one LSA (triggered by the link down event) to router
4.

The actual number of LSA packets may flunctuate depending how quickly the
link up events between 1 & 4 is received by 1 & 4. Hopefully this will be
taken into account when project B is graded.

Thanks and best regards,

Kaiyu


From hussain@ISI.EDU Mon Apr 22 07:43:13 2002
Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3MEhDm16637
	for ; Mon, 22 Apr 2002 07:43:13 -0700 (PDT)
Received: from localhost by boreas.isi.edu (8.11.6/8.11.2) with SMTP id g3MEh7g27868;
	Mon, 22 Apr 2002 07:43:07 -0700 (PDT)
Date: Mon, 22 Apr 2002 07:43:07 -0700 (PDT)
From: Alefiya Hussain 
To: kaiyuxia 
cc: John Heidemann , csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Several errors in the sample output of Proj B
In-Reply-To: 
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 


On Sun, 21 Apr 2002, kaiyuxia wrote:

> Hi, Prof. & TA,
> 
> I think the following results are wrong in the sample output file:
> (assume we are using the same manager.conf file and have the same event:
> original connection:
> 1 2
> 1 3
> 2 4
> 3 4
> 3 5
> 4 5
> events:
> d 1 3
> u 1 4
> 
> So finally we get something like:
> 1 2
> 1 4
> 2 4
> 3 4
> 3 5
> 4 5
> )
> 
> For s4-2.out,
> 3 192.168.123.0/24 1 2
> wrong because now 1 has no connection with 3.
> Should be: 3 192.168.123.0/24 4 2
> 
> For s4-3.out:
> 1 10.1.0.0/16 1 1
> wrong because 3 has no direct connection with 3 now.
> Should be:  1 10.1.0.0/16        4 2
> 
> For s4-4.out:
> 1 10.1.0.0/16           2 2
> wrong because 4 now has direct connection with 4.
> Should be: 1 10.1.0.0/16        1 1

Thanks Kaiyu , that was a bug in the output for the links. 

I have corrected it. 



> 
> By the way, if we treat the link down and up events as independent, (as
> one of your previous email suggested ), the LSA packets number might be
> different than the sample output. Because:
> When router 1 & 3 got the link down event, they do not know that the link
> between 1 & 4 is established (which will happen later). So router 1 will
> not be able to flood one LSA (triggered by the link down event) to router
> 4.
> 
> The actual number of LSA packets may flunctuate depending how quickly the
> link up events between 1 & 4 is received by 1 & 4. Hopefully this will be
> taken into account when project B is graded.

The number of LSA messages will fluctutate depending on the
implementation. Thats number beasically checkes that LSAs are not looping
unnecessarily 

alefiya 


From vkhanna@qualcomm.com Mon Apr 22 18:29:48 2002
Received: from illyana.qualcomm.com (illyana.qualcomm.com [129.46.65.100])
	by gamma.isi.edu (8.11.6/8.11.2) with ESMTP id g3N1Tkm15419
	for ; Mon, 22 Apr 2002 18:29:48 -0700 (PDT)
Received: from VKHANNA.qualcomm.com (vkhanna.qualcomm.com [129.46.235.13])
	by illyana.qualcomm.com (8.12.1/8.12.1/1.0) with ESMTP id g3N1ThWM005076
	for ; Mon, 22 Apr 2002 18:29:44 -0700 (PDT)
Message-Id: <5.1.0.14.2.20020421182639.043d60f0@illyana>
X-Sender: vkhanna@illyana
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Sun, 21 Apr 2002 18:29:41 -0700
To: csci551-talk@mailman.isi.edu
From: Vivek Khanna 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Subject: [Csci551-talk] Project 1 soln does not run
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

Project1 soln does not run. I tried running all the sample files given at 
web site and I get the same ERROR.
"contentful files in config files after all sections parsed"

Here is the sample file (Test case 1) that I am using. Infact, this happens 
for all the test cases.

5
0
1       2
1       3
1       5
2       5
3       4
4       5
0       0
10.1.1.0        16
10.1.2.0        17
10.1.3.0        18
10.1.4.0        24
10.1.5.0        16
0.0.0.0         0
0       0       0.0.0.0         0       0



From johnh@ISI.EDU Mon Apr 22 20:52:57 2002
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 g3N3qvm09982
	for ; Mon, 22 Apr 2002 20:52:57 -0700 (PDT)
Received: from albatross.prod.itd.earthlink.net (albatross.mail.pas.earthlink.net [207.217.120.120])
	by tnt.isi.edu (8.11.6/8.11.2) with ESMTP id g3N3qup17415
	for ; Mon, 22 Apr 2002 20:52:56 -0700 (PDT)
Received: from user-112ujfa.biz.mindspring.com ([66.47.77.234] helo=dash.isi.edu)
	by albatross.prod.itd.earthlink.net with esmtp (Exim 3.33 #2)
	id 16zrMO-0002UP-00; Mon, 22 Apr 2002 20:52:56 -0700
Received: from dash.isi.edu (johnh@localhost)
	by dash.isi.edu (8.11.6/8.11.6) with ESMTP id g3N3q8029731;
	Mon, 22 Apr 2002 20:52:08 -0700
Message-Id: <200204230352.g3N3q8029731@dash.isi.edu>
X-url: http://www.isi.edu/~johnh/
To: Vivek Khanna 
Cc: csci551-talk@ISI.EDU
Subject: Re: [Csci551-talk] Project 1 soln does not run 
In-reply-to: <5.1.0.14.2.20020421182639.043d60f0@illyana> 
MIME-Version: 1.0 (generated by SEMI 1.14.3 - "Ushinoya")
Content-Type: text/plain; charset=US-ASCII
Date: Mon, 22 Apr 2002 20:52:08 -0700
From: John Heidemann 
Sender: csci551-talk-admin@mailman.isi.edu
Errors-To: csci551-talk-admin@mailman.isi.edu
X-BeenThere: csci551-talk@mailman.isi.edu
X-Mailman-Version: 2.0.5
Precedence: bulk
List-Help: 
List-Archive: 

On Sun, 21 Apr 2002 18:29:41 PDT, Vivek Khanna wrote: 
>Project1 soln does not run. I tried running all the sample files given at 
>web site and I get the same ERROR.
>"contentful files in config files after all sections parsed"
>
>Here is the sample file (Test case 1) that I am using. Infact, this happens 
>for all the test cases.
>
>5
>0
>1       2
>1       3
>1       5
>2       5
>3       4
>4       5
>0       0
>10.1.1.0        16
>10.1.2.0        17
>10.1.3.0        18
>10.1.4.0        24
>10.1.5.0        16
>0.0.0.0         0
>0