Well, I did not come across any physical documentation from Cisco on this,
but here are my Findings:
I was observing the duplication of frames when running a port SPAN on a
Catalyst 6509. The way I set up the port span was as such: "set span 4,5
3/8", which basically is saying SPAN ALL PORTS on VLAN 4 and 5 and puts
those frames on port 3/8 (where my sniffer is).

It turns out that by doing this I am capturing BOTH INGRESS and EGRESS
frames, and that is why I see duplicate frames. 
For Example, Router A wants to send Frame 1 to Router B. Both Routers have
interfaces on VLAN 4, and are both plugged physically in the same Catalyst.
When I run the SPAN port for VLAN 4, The SPAN port will receive Frame 1 sent
by Router A, then will also receive the Frame 1 again when it is sent from
it's Catalyst port to Router B. 

So by observing this, I have to be more selective in my SPAN logic. I now
choose to SPAN individual ports instead of the entire VLAN, and that keeps
the duplicate of frames from occurring. Hope this helps.

Jeffery Kraus
Data Services Engineer
773.216.3179 (cell)
224.653.3720 (office)
224.653.3766 (fax)


-----Original Message-----
From: Jacky Buyck [mailto:[EMAIL PROTECTED]
Sent: Saturday, December 27, 2003 6:26 AM
To: 'Kraus, Jeffery'
Subject: RE : [tcpdump-workers] Observing duplicate frame captures in
TCPDU MP


Hi.
I'm really interested by your problem because I've encounter the same
one but with Nortel Switches.
What Nortel wasn't abble to give us correct answer to this so if you
have generic informations from CISCO I'm interrested.

Thanks.
JB


-----Message d'origine-----
De : [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] De la part de
Kraus, Jeffery
Envoyé : mercredi 24 décembre 2003 16:41
À : 'George Bakos'
Cc : '[EMAIL PROTECTED]'
Objet : RE: [tcpdump-workers] Observing duplicate frame captures in
TCPDU MP


In this scenario it is not 802.11. I have my Redhat 9 Box running
TCPDUMP connected to a Cisco Catalyst 6509 using 10/100 CAT5e Ethernet
on a port Spanning 2 VLANs. The Multiples frames are identical (every
byte, MAC, IP,
etc...) Although the timestamps are slightly different (off by less than
a millisecond or so).

It definitely seems to be an issue with the Cisco Span Port....I have
connected a PC running Sniffer Pro and IRIS Sniffers, and are seeing the
same thing. I will investigate this further with Cisco. Thank you all
for your help.


-----Original Message-----
From: George Bakos [mailto:[EMAIL PROTECTED]
Sent: Tuesday, December 23, 2003 11:57 PM
To: Kraus, Jeffery
Cc: '[EMAIL PROTECTED]'
Subject: Re: [tcpdump-workers] Observing duplicate frame captures in
TCPDU MP


This is normal behaviour for managed wireless networks, where the frame
is encapsulated in 802.11 both to and from the WAP. If this is a copper
or fiber net, are you certain you aren't seeing the effects of a funny
bridge/VLAN/routing environment? Are the multiples being reported with
identical timestamps? How about src MAC addresses?

g

On Tue, 23 Dec 2003 09:43:56 -0600
"Kraus, Jeffery" <[EMAIL PROTECTED]> wrote:

> The machine is Redhat 9, and it is just receiving frames from the 
> network. It does not have an IP address bound to the adaptor so it 
> should not be generating any frames itself.
> 
> Here is the Kernal details:
> uname -a
> Linux usc-schaum-sniff 2.4.20-8 #1 Thu Mar 13 17:54:28 EST 2003 i686 
> i686 i386 GNU/Linux
> 
> 
> 
> -----Original Message-----
> From: Guy Harris [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 19, 2003 6:06 PM
> To: Kraus, Jeffery
> Cc: '[EMAIL PROTECTED]'
> Subject: Re: [tcpdump-workers] Observing duplicate frame captures in 
> TCPDUMP
> 
> 
> 
> On Dec 19, 2003, at 2:41 PM, Kraus, Jeffery wrote:
> 
> > Whenever I run captures I always get every packet displayed twice. I
> > have
> > seen numerous emails regarding this issue, but no real fix. I am 
> > currently
> > using eth4 as the capture interface and I do not have an IP address 
> > bound to
> > it.
> 
> On what OS are you running this?
> 
> Is the machine running tcpdump sending or receiving those packets, or
> is it just passively capturing other machines' traffic on a network?
> -
> This is the TCPDUMP workers list. It is archived at
> http://www.tcpdump.org/lists/workers/index.html
> To unsubscribe use
mailto:[EMAIL PROTECTED]


-- 
George Bakos
Institute for Security Technology Studies - IRIA
Dartmouth College
[EMAIL PROTECTED]
603.646.0665 -voice
603.646.0666 -fax
-
This is the TCPDUMP workers list. It is archived at
http://www.tcpdump.org/lists/workers/index.html
To unsubscribe use
mailto:[EMAIL PROTECTED]
-
This is the TCPDUMP workers list. It is archived at
http://www.tcpdump.org/lists/workers/index.html
To unsubscribe use mailto:[EMAIL PROTECTED]

Reply via email to