Re: [ns] Flowmonitor Trace - missing data

["J�rgen, Ols�n" <thejorge@hotmail.com>]

Redhat 5.? + ns2.1b6  No good combination?!
--------------------------------------------
Yes, I found the solution. There is nothing wrong with the code.
The problem was with the ns-installation/OS version.

We were running the Linux, Redhat 5.? installation and it appears to me that 
ns-2.1b doesn't work correctly under this.
Installing the same distribution on a SunOs 5.6 system resulted in a working 
program. Reinstalling ns on my Redhat 5.? installation still gave the same 
errors in the trace file whereas upgrading to Redhat 6.1 and reinstalling 
resulted in a working system.

Hope this helps some of you.
		Regards, Jorgen


>From: "Petra E. J. Napitu" <pejna7@zoom.co.uk>
>To: "Olsen Jorgen" <thejorge@hotmail.com>
>Subject: Re: [ns] Flowmonitor Trace - missing data
>Date: Tue, 9 May 2000 01:04:26 +0100
>
>Hi ...
>
>How you got the answer to your problem ... ? I wonder if you can share the
>solution.
>
>Thank you kindly,
>Bonatua B. V. Napitu
>Flat C, 15-19 Upper Montagu St.
>London W1H-1RR. United Kingdom
>+44 (0)20 7724 9068
>----------------------------------------------------------------------
>
>----- Original Message -----
>From: J�rgen, Ols�n <>
>To: <ns-users@mash.cs.berkeley.edu>
>Sent: Thursday, April 27, 2000 10:23 AM
>Subject: [ns] Flowmonitor Trace - missing data
>
>
> > Hello,  I use flowmonitor to gather information about flows. The monitor
>has
> > been initiated this way:
> >
> >   set flow_stat [open flow.stat w]
> >   set flowmon_ [$ns makeflowmon Fid]
> >   $ns attach-fmon [$ns link $receiver $final_receiver] $flowmon_ 0
> >   set_flowMonitor $flowmon_ $flow_stat
> >
> >   proc set_flowMonitor {flowmon output_chan} {
> >       $flowmon attach $output_chan
> >       return $flowmon
> >   }
> >
> > Thereafter I dump the flowinfo once every 0.1 seconds using the follwing
> > piece of code:
> >
> >   proc dump_flow_info {}  {
> >       global ns flowmon_
> >       $flowmon_ dump
> >       $ns after 0.1 "dump_flow_info"
> >   }
> >
> >
> > Looking in my trace file flow.stat I have lines looking like this:
> >    1.000 1 0 2 1 0 6 qd qd 156 0 qd qd 156000 0 0 0 341 0
> >    1.000 2 0 2 2 1 6 qd qd 159 0 qd qd 159000 0 0 0 341 0
> >    1.000 3 0 2 3 2 6 qd qd 26 0 qd qd 26000 0 0 0 341 0
> >
> > Some of the columns doesn't have proper values but instead are marked
> > "qd". Does anybody have any clue why this is happening?
> >
> > I get the same sort of result if I run the test file
> > test-suite-simple.tcl, suggesting that maybe I am not doing anything
> > wrong?? test-suite-simple.tcl plots the following information to the
> > screen:
> >
> > koltrast$ ns test-suite-simple.tcl stats
> > ...
> > ...
> > tcp 0 total_packets_acked 335
> > tcp 1 total_packets_acked 87
> > fid: 0 per-link total_drops 13
> > fid: 0 per-link total_packets qd
> > fid: 0 per-link total_bytes qd
> > fid: 1 per-link total_drops 6
> > fid: 1 per-link total_packets qd
> > fid: 1 per-link total_bytes qd
> > aggregate per-link total_drops 19
> > aggregate per-link total_packets qd
> >
> >
> >         Any suggestions would be greatly appreciated!
> >
> >                 Regards, Jorgen
> >
> >
> >
> >
> >
> > ________________________________________________________________________
> > Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
> >
> >
>
>
>
>
>
>
>
>
>

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com