From owner-freebsd-smp@FreeBSD.ORG Mon Apr 23 15:33:35 2007 Return-Path: X-Original-To: freebsd-smp@freebsd.org Delivered-To: freebsd-smp@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id D78D116A402 for ; Mon, 23 Apr 2007 15:33:35 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from elvis.mu.org (elvis.mu.org [192.203.228.196]) by mx1.freebsd.org (Postfix) with ESMTP id C2CE213C468 for ; Mon, 23 Apr 2007 15:33:35 +0000 (UTC) (envelope-from kris@obsecurity.org) Received: from obsecurity.dyndns.org (elvis.mu.org [192.203.228.196]) by elvis.mu.org (Postfix) with ESMTP id A43931A4DC1; Mon, 23 Apr 2007 08:33:57 -0700 (PDT) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id 524545138E; Mon, 23 Apr 2007 11:33:34 -0400 (EDT) Date: Mon, 23 Apr 2007 11:33:34 -0400 From: Kris Kennaway To: Paul Message-ID: <20070423153334.GA530@xor.obsecurity.org> References: <20070422221426.74CF83D7@fep9.cogeco.net> <20070422224157.GA63390@xor.obsecurity.org> <20070423010224.C16984@chrishome.localnet> <20070423004105.7984082E@fep7.cogeco.net> <00a601c78549$6b480fa0$b6db87d4@multiplay.co.uk> <20070423135619.D6FB4495@fep9.cogeco.net> <009c01c785b6$c8ac36e0$b6db87d4@multiplay.co.uk> <20070423145954.8244094A@fep1.cogeco.net> <00b301c785b9$0bc9b180$b6db87d4@multiplay.co.uk> <20070423152703.715922D1@fep2.cogeco.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070423152703.715922D1@fep2.cogeco.net> User-Agent: Mutt/1.4.2.2i Cc: freebsd-smp@freebsd.org, Steven Hartland Subject: Re: System Cpu Between 50-70% and need to find out why X-BeenThere: freebsd-smp@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: FreeBSD SMP implementation group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 23 Apr 2007 15:33:35 -0000 On Mon, Apr 23, 2007 at 11:27:17AM -0400, Paul wrote: > Hi Steve, > > When I shut off everything except apache with very low load and > qpopper it is still maxed out. That does not look like very low load to me. On the contrary, your stats show that 30-40 running processes are being scheduled for running on the 4 CPUs at any one time. Under those circumstances you should expect to see 100% CPU usage (anything less means your system is not being completely used). Can you please confirm what you are using for your diagnosis that there is something wrong with the system performance? Kris > Here is another look with the vm setting off. > > I don't always see the idle: cpu in the list as it changes > constantly. I have a hunch this is disk related but I am not sure. I > include two snapshots below > > Thanks, > > Paul > > last pid: 18967; load averages: 29.00, 44.28, > 46.39 up > 0+10:16:42 11:17:00 > 308 processes: 41 running, 239 sleeping, 7 zombie, 21 waiting > CPU states: 13.3% user, 0.0% nice, 74.5% system, 12.2% interrupt, 0.0% > idle > Mem: 1204M Active, 5678M Inact, 381M Wired, 20K Cache, 214M Buf, 8398M Free > Swap: 8192M Total, 8192M Free > > PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU > COMMAND > 11 root 1 171 52 0K 16K RUN 2 93:59 > 22.80% idle: cpu2 > 10 root 1 171 52 0K 16K RUN 3 116:20 > 22.46% idle: cpu3 > 12 root 1 171 52 0K 16K RUN 1 101:36 > 22.36% idle: cpu1 > 13 root 1 171 52 0K 16K RUN 0 123:32 > 22.22% idle: cpu0 > 18944 root 1 132 0 15944K 5332K RUN 1 0:01 > 14.99% perl5.8.8 > 18922 root 1 131 0 15648K 3196K RUN 2 0:01 12.45% imapd > 18640 root 1 127 0 7484K 2792K CPU1 0 0:10 11.86% top > 18952 root 1 131 0 10536K 1412K RUN 1 0:00 11.30% > qpopper > 18894 user1 1 127 0 10740K 2948K CPU3 3 0:02 10.21% > qpopper > 18845 user2 1 -4 0 10740K 2944K getblk 0 0:04 10.14% > qpopper > 18871 user3 1 126 0 10740K 3008K CPU0 0 0:03 9.48% > qpopper > 18920 root 1 129 0 15648K 3196K RUN 3 0:01 8.68% imapd > 17491 user4 1 123 0 7864K 3168K select 2 0:44 8.54% top > 14 root 1 -32 -151 0K 16K > WAIT 0 59:32 7.52% swi4: clock sio > 18939 root 1 130 0 10740K 2940K RUN 2 0:01 7.10% > qpopper > 18923 user5 1 128 0 10740K 2948K RUN 2 0:01 7.03% > qpopper > 48 root 1 -4 0 0K 16K ufs 2 28:03 6.74% > syncer > 18953 root 1 130 0 10544K 2140K RUN 0 0:00 6.46% > qpopper > 18935 root 1 130 0 10740K 2944K RUN 2 0:01 6.21% > qpopper > 18941 user6 1 130 0 10740K 3008K RUN 2 0:01 6.07% > qpopper > 18956 root 1 131 0 6084K 860K RUN 3 0:00 5.95% > qpopper > 16 root 1 -44 -163 0K 16K > WAIT 0 52:09 5.71% swi1: net > 18940 user7 1 129 0 10740K 2944K RUN 0 0:00 5.62% > qpopper > 18934 root 1 130 0 10740K 2940K RUN 1 0:00 5.47% > qpopper > 18954 root 1 130 0 10532K 2104K RUN 0 0:00 5.38% > qpopper > 18949 root 1 130 0 10576K 1424K RUN 0 0:00 5.07% > qpopper > 18965 root 1 132 0 5844K 1536K RUN 1 0:00 5.00% inetd > > > last pid: 20588; load averages: 47.61, 36.13, > 39.78 up > 0+10:24:00 11:24:18 > 531 processes: 93 running, 413 sleeping, 19 zombie, 6 lock > CPU states: 19.1% user, 0.0% nice, 74.8% system, 6.1% interrupt, 0.0% > idle > Mem: 1590M Active, 5795M Inact, 404M Wired, 20K Cache, 214M Buf, 7872M Free > Swap: 8192M Total, 8192M Free > > PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU > COMMAND > 1375 vscan 3 98 0 65632K 52772K ucond 0 52:18 17.63% clamd > 1184 bind 1 125 0 63620K 60408K select 2 60:11 9.42% named > 19776 root 1 126 0 8096K 3408K CPU2 0 0:23 7.57% top > 20429 user1 1 127 0 10740K 3008K RUN 0 0:03 6.98% > qpopper > 20424 user2 1 127 0 10740K 3008K RUN 1 0:03 6.81% > qpopper > 20395 user3 1 127 0 10740K 2944K RUN 2 0:03 6.81% > qpopper > 20442 user4 1 127 0 10740K 2944K RUN 0 0:03 6.75% > qpopper > 17491 user5 1 118 0 8184K 3488K RUN 0 1:08 6.74% top > 20391 user6 1 127 0 10768K 2972K RUN 2 0:03 6.59% > qpopper > 19135 vscan 1 128 0 126M 85504K select > 2 0:20 5.22% perl5.8.8 > 19136 vscan 1 127 0 124M 83900K > RUN 2 0:20 4.69% perl5.8.8 > 20476 root 1 127 0 15644K 3188K RUN 3 0:01 4.40% imapd > 20478 user7 1 127 0 15648K 3232K RUN 3 0:01 4.40% imapd > 20475 root 1 127 0 10024K 4284K RUN 2 0:01 4.33% > sendmail > 20139 user8 1 127 0 15724K 3288K RUN 2 0:05 4.07% imapd > 20499 user9 1 127 0 10740K 2944K RUN 0 0:01 4.00% > qpopper > 19134 vscan 1 125 0 127M 86256K select > 3 0:21 3.91% perl5.8.8 > 20304 user10 1 4 0 10740K 2948K sbwait 2 0:04 3.86% > qpopper > 19133 vscan 1 -4 0 123M 83372K > RUN 2 0:20 3.86% perl5.8.8 > > > > > At 11:07 AM 23/04/2007, you wrote: > >But when disabled does the stats in top show a different picture which > >might identify the app / component which is causing so much vm work? > > > > Steve > > _______________________________________________ > freebsd-smp@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-smp > To unsubscribe, send any mail to "freebsd-smp-unsubscribe@freebsd.org" >