Hi,

I currently have a setup where I have numerous machines collecting
NetFlow data. Circumstances during the initial deployment caused me to
collect and process on a per site basis. I now wish to bring this all
together and have a centalized collection point and I am just curious
on how people deal with this.

There is 15 sites in total and I would like to see them all report
back to a sinlgle machine. What I cant figure out is how the directory
structure will affect the scripts that I use to process the flows as I
would still like to keep the processing the same.

As an example:

1) I use Joe Loiacono's FlowViewer and I would like to keep each site
seperate. (I think this will already handle this through the device
drop down selection) I am not sure how it differentiates between sites
though, if it is all done within the same directory this wont work for
me because of (2).

2) I have flow-capture run a shell script on each new dump. This
script updates RRD's and generates reports based on the data.

I currently call flow-capture like:

flow-capture -n300 -R flow_sql.sh -w/usr/home/flow/Flows 0/127.0.0.1/2055

Will I have to run a seperate instance of flow-capture for each site?
and if so how do I get the directory structure to look like

Flows/site1
Flows/site2
..

or, is there another way to do this. Running 15 instances of
flow-capture sounds a little silly.

Any input would be appreciated.

Thanks.

Paul.
_______________________________________________
Flow-tools mailing list
[EMAIL PROTECTED]
http://mailman.splintered.net/mailman/listinfo/flow-tools

Reply via email to