ramon, ragator, flows and networks.

Peter Van Epp vanepp at sfu.ca
Fri May 4 14:25:34 EDT 2001


	It can be done with perl running off ra output. Although I'm currently
only using nets to identify scans rather than traffic (which I do by IP address)
the network (assuming class C subnet sizes) is broken off and it wouldn't be 
a problem to sort traffic by subnet.
	For instance this report (traffic and traffic by port) could be 
modified to be traffic by destination subnet easily:


142.58.101.24   total traffic: 328,174,671
           142.58.101.24    192.75.241.11   2049               0               0

           142.58.101.24     192.75.241.3   2049               0               0

           142.58.101.24    192.75.241.53  49153               0               0

           142.58.101.24     192.75.241.7   1524               0               0

           142.58.101.24    192.75.241.75  49257               0               0

	so this would become a single line of 

	142.58.101.24	   192.75.241				0 	0

Peter Van Epp / Operations and Technical Support 
Simon Fraser University, Burnaby, B.C. Canada

> 
> Hi,
> 
> I have an interest in figuring out, per subnet, the top "talkers" to
> other subnets.  In order to do some network provisioning, we're looking
> to find the most active (in terms of traffic sent/received) networks
> that are talking to certain specific subnets of ours.
> 
> Is this something that argus can do?  It seems like the rough plumbing
> for it is definitely there, and in the case of ragator perhaps that's
> the exact tool I need, but I'm not sure how to best go about creating a
> flow model that generates the type of data I'm looking for.
> 
> My goal is to have a breakdown so that I get something similar to ramon
> output, as so:
> 
> 988847907     ip 192.168.15.0/24                 0        3689339   0 221692479   INT
> 988847926     ip 192.168.37.0/24                 94475    0 101755360    0           INT
> 988847926     ip 10.20.10.0/24                   0        94444     0 101753470   INT
> 
> With each of those addresses being networks sending or receiving data to
> certain target networks of ours.
> 
> Any hints would be appreciated!
> 
> Scott
> 
> 



More information about the argus mailing list