From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthias Saou Subject: Re: Wrong network usage reported by /proc Date: Mon, 4 May 2009 21:11:51 +0200 Message-ID: <20090504211151.74622f29@python3.es.egwn.lan> References: <20090504171408.3e13822c@python3.es.egwn.lan> <49FF2BB2.4030700@cosmosbay.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: linux-kernel@vger.kernel.org, Linux Netdev List To: Eric Dumazet Return-path: In-Reply-To: <49FF2BB2.4030700@cosmosbay.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Eric Dumazet wrote : > Matthias Saou a =E9crit : > > Hi, > >=20 > > I'm posting here as a last resort. I've got lots of heavily used RH= EL5 > > servers (2.6.18 based) that are reporting all sorts of impossible > > network usage values through /proc, leading to unrealistic snmp/cac= ti > > graphs where the outgoing bandwidth used it higher than the physica= l > > interface's maximum speed. > >=20 > > For some details and a test script which compares values from /proc > > with values from tcpdump : > > https://bugzilla.redhat.com/show_bug.cgi?id=3D489541 > >=20 > > The values collected using tcpdump always seem realistic and match = the > > values seen on the remote network equipments. So my obvious conclus= ion > > (but possibly wrong given my limited knowledge) is that something i= s > > wrong in the kernel, since it's the one exposing the /proc interfac= e. > >=20 > > I've reproduced what seems to be the same problem on recent kernels= , > > including the 2.6.27.21-170.2.56.fc10.x86_64 I'm running right now.= The > > simple python script available here allows to see it quite easily : > > https://www.redhat.com/archives/rhelv5-list/2009-February/msg00166.= html > >=20 > > * I run the script on my Workstation, I have an FTP server enabled > > * I download a DVD ISO from a remote workstation : The values matc= h > > * I start ping floods from remote workstations : The values report= ed > > by /proc are much higher than the ones reported by tcpdump. I us= ed > > "ping -s 500 -f myworkstation" from two remote workstations > >=20 > > If there's anything flawed in my debugging, I'd love to have someon= e > > point it out to me. TIA to anyone willing to have a look. > >=20 > > Matthias > >=20 >=20 > I could not reproduce this here... what kind of NIC are you using on > affected systems ? Some ethernet drivers report stats from card itsel= f, > and I remember seeing some strange stats on some hardware, but I cann= ot > remember which one it was (we were reading NULL values instead of > real ones, once in a while, maybe it was a firmware issue...) My workstation has a Broadcom BCM5752 (tg3 module). The servers which are most affected have Intel 82571EB (e1000e). But the issue is that with /proc, the values are a lot _higher_ than with tcpdump, and the tcpdump values seem to be the correct ones. Matthias --=20 Clean custom Red Hat Linux rpm packages : http://freshrpms.net/ =46edora release 10 (Cambridge) - Linux kernel 2.6.27.21-170.2.56.fc10.x86_64 Load : 2.20 0.88 0.42