Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 18 Jan 2005 20:57:07 -0500
From:      "The Jetman" <jetman@mycbc.com>
To:        "FreeBSD Net" <freebsd-net@freebsd.org>
Subject:   Re: Network Accounting
Message-ID:  <001801c4fdca$37bc1270$7300a8c0@EAGLE>

next in thread | raw e-mail | index | archive | help

I've been following this little thread and was curious about how 
my own solution compares w/ the others discussed.

####    x.y.z.14
${ipfw} add pipe 7 ip from x.y.z.14 to any
${ipfw} pipe 7 config bw 1024Kbit/s queue 50
${ipfw} add pipe 8 ip from any to x.y.z.14
${ipfw} pipe 8 config bw 1024Kbit/s queue 50
####

I setup a series of rules for each of a real IPs, as shown above, to 
cap bwidth usage and to provide a series of byte counters that could be 
captured hourly.  This box was a bridge box bet the client's internal net 
and their T1.  My CRON job would sit on the working side of a pipe from 
the 'ipfw -a list' command, then parse each rule for the inbound then 
outbound byte/packet count.  Each inbound and outbount count was then 
inserted into a SQL UPDATE stmt for each IP.

I inquire bet I've seen a nbr of other approaches and was wondering are 
all of these approaches equally valid, but I'm esp curious if my approach 
makes sense.

TIA.  Later....Jet


===============  From the desk of Jethro Wright, III  ================
+      Beer is proof that God loves us and wants us to be happy.     -
==============================================  Benjamin Franklin  ===



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?001801c4fdca$37bc1270$7300a8c0>