From owner-freebsd-net@FreeBSD.ORG Fri Sep 26 16:58:52 2008 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 09FC11065687 for ; Fri, 26 Sep 2008 16:58:52 +0000 (UTC) (envelope-from 000.fbsd@quip.cz) Received: from elsa.codelab.cz (elsa.codelab.cz [91.103.162.4]) by mx1.freebsd.org (Postfix) with ESMTP id BAD638FC08 for ; Fri, 26 Sep 2008 16:58:51 +0000 (UTC) (envelope-from 000.fbsd@quip.cz) Received: from localhost (localhost.codelab.cz [127.0.0.1]) by elsa.codelab.cz (Postfix) with ESMTP id 6DC6219E027; Fri, 26 Sep 2008 18:40:31 +0200 (CEST) Received: from [192.168.1.2] (r5bb235.net.upc.cz [86.49.61.235]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by elsa.codelab.cz (Postfix) with ESMTPSA id 4061519E02E; Fri, 26 Sep 2008 18:40:29 +0200 (CEST) Message-ID: <48DD109E.9040900@quip.cz> Date: Fri, 26 Sep 2008 18:41:02 +0200 From: Miroslav Lachman <000.fbsd@quip.cz> User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.12) Gecko/20050915 X-Accept-Language: cz, cs, en, en-us MIME-Version: 1.0 To: Steve Bertrand References: <48DCEC3E.70303@ibctech.ca> In-Reply-To: <48DCEC3E.70303@ibctech.ca> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Cc: FreeBSD Net Subject: Re: netstat byte/bit confusion X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 26 Sep 2008 16:58:52 -0000 Steve Bertrand wrote: > Hey all, > > I'm experiencing conflicting information on throughput numbers when > comparing information garnered via MRTG on a 1000Mbps HP Procurve, and > netstat -h -w1 on a server connected to the switch. > > What I want to know is if netstat in the below case is actually > displaying the info in bits, even though it is telling me the result is > in bytes: > > amanda# netstat -h -w1 > input (Total) output > packets errs bytes packets errs bytes colls > 109K 0 90M 200 0 14K 0 > 104K 0 88M 201 0 14K 0 > ^C > > On a different server running MRTG (FBSD 7.0-STABLE), I have configured > it to display in bits/s, and it is showing ~90Mbps. > > Which one is accurate? I'm trying to evaluate the difference between > Cisco Cat 29xx 100Mb switches and this ProCurve GigE 2848 switch. So > far, my results are that the 100Mb Cisco can peak and sustain a 98Mbps > throughput. > > The Procurve, unless MRTG is wrong, and netstat output should be 90M*8, > I'm far less than impressed. > > ...or could it be that MRTG is broken/capped at ~100Mbps calculations? > > Thanks for any insight, If you are using MRTG with SNMP info from switch, be aware of 32 bit counters. If you have high network load on the switch, the counter overflows between two five minutes intervals of MRTG and then MRTG shows inacurate results. (for example 60Mbps instead of 250Mbps peak) Some devices have 64 bit counters and MRTG can be configured to use these. netstat is in bytes AFAIK. Miroslav Lachman