From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ted Kaczmarek Subject: Re: per-connection byte counts Date: Mon, 19 Jan 2004 18:59:18 -0500 Sender: netfilter-admin@lists.netfilter.org Message-ID: <1074556757.20675.11.camel@tarkus> References: <20040118140903.GA3351@glympton.airtime.co.uk> <16394.59800.443846.342711@fisica.ufpr.br> <000e01c3de21$28bf0440$0300a8c0@indoaksesbno> <200401190031.50360.Antony@Soft-Solutions.co.uk> <002001c3de26$4b970da0$0300a8c0@indoaksesbno> Mime-Version: 1.0 Content-Transfer-Encoding: 7BIT Return-path: In-reply-to: <002001c3de26$4b970da0$0300a8c0@indoaksesbno> Errors-To: netfilter-admin@lists.netfilter.org List-Help: List-Post: List-Subscribe: , List-Id: List-Unsubscribe: , List-Archive: Content-Type: text/plain; charset="us-ascii" To: bino-psn Cc: netfilter@lists.netfilter.org Ntop does a wonderful job of this, and according to what I read, the latest version will allow you to store to a file . I mostly use it to collect sflow data, but have run it on our nms as a local agent as well. Ted On Sun, 2004-01-18 at 19:50, bino-psn wrote: > ----- Original Message ----- > From: "Antony Stone" > To: > Sent: Monday, January 19, 2004 7:31 AM > Subject: Re: per-connection byte counts > > > > If this package provides the information you need, why not just run it on > the > > netfilter machine? What's the need for a separate box? > > No problem Sir. > I just want to split the Monitoring-function from the Packet-Forwarding > function. > I just don't want a fail of one function savere the other. > I never mix complex function in a single box, unless they need tobe. > my small corporate network is build from a bunch of used classic pentium > class PC, with small RAM .. and WISP-Dist in it. > Let's say that the network is in a peak state ... super high traffic. It > will eat more CPU of the monitoring function. I don't want this to kick-down > the packet-forward function. That why I put it in another box. > > Sincerely > -bino- > >