This was the exact issue that we had (see the nistnet archives Vol. 18
No. 1) that
never got addressed through the list; through a circuitous number of
events
(including actually getting it to work without dupes on a different pc)
one
of our hardware/networking gurus figured out that it was related to how
newer
motherboards handle the irq8 interrupt so it appeared to be an issue on
some motherboards and not others. He said that he had figured out a fix
for it but I don't know what it was.
Sorry I can't provide more details than that but I've moved onto another
project right now and don't have any.
Good luck .... N
------------------------------------------------------------------------
---
Date: Thu, 13 Mar 2008 16:31:29 -0800 (PST)
From: David Morris <[EMAIL PROTECTED]>
Subject: Re: [nistnet] NISTNET ippt->func compilation issues and exact
duplicate packets
To: "Karl A. Nyberg" <[EMAIL PROTECTED]>
Cc: nistnet@antd.nist.gov
Message-ID: <[EMAIL PROTECTED]>
Content-Type: TEXT/PLAIN; charset=US-ASCII
I did a simple test ... traffic was all a Skype UDP call ... I couldn't
detect any massive duplication. I use --bandwidth and --drd but not
--delay.
On Wed, 12 Mar 2008, Karl A. Nyberg wrote:
> I'm getting exact duplicates of every packet (reviewing with ethereal)
> out of NISTNET while running. My non-IP address arguments are
>
> --bandwidth=250000 --delay=300 --drd 0 100 90
>
> I thought that maybe the little hack to get over the ippt->func
missing a fourth parameter by adding a NULL at the end might be the
culprit and found at
http://portal.miun.se/~necu9500/dccp/NistNetPatch.html a
suggestion to use "dev" as the fourth parameter instead. No
improvement...
>
> I have confirmed that the culprit is NISTNET as the packet trace
immediately
clears up when NISTNET is shut down.
>
> Has anybody else seen this?
>
> -- Karl --
>
> Karl A. Nyberg
> http://karl.nyberg.net
> 703-406-4161
>
_______________________________________________
nistnet mailing list
nistnet@antd.nist.gov
http://www-x.antd.nist.gov/mailman/listinfo/nistnet
