Date: Sun, 22 Apr 2007 20:41:18 -0400 From: Paul <bsdlist@cogeco.ca> To: freebsd-smp@freebsd.org Subject: Re: System Cpu Between 50-70% and need to find out why Message-ID: <20070423004105.7984082E@fep7.cogeco.net> In-Reply-To: <20070423010224.C16984@chrishome.localnet> References: <20070422221426.74CF83D7@fep9.cogeco.net> <20070422224157.GA63390@xor.obsecurity.org> <20070423010224.C16984@chrishome.localnet>
next in thread | previous in thread | raw e-mail | index | archive | help
Dear Gentlemen, Thank you very much for your help so far. I was certainly looking for the "S" option. When you are tired these things are easy to miss. Here is an output of the system with the md driver removed and everything is on the hard drive for now. The system usage is still really high. Does anyone have any suggestions based on the output below? Thanks Paul last pid: 36756; load averages: 8.74, 9.74, 10.40 up 0+16:08:41 20:28:14 279 processes: 15 running, 241 sleeping, 21 waiting, 2 lock CPU states: 11.9% user, 0.4% nice, 80.6% system, 6.9% interrupt, 0.2% idle Mem: 1732M Active, 5146M Inact, 354M Wired, 48K Cache, 214M Buf, 8432M Free Swap: 8192M Total, 8192M Free PID USERNAME THR PRI NICE SIZE RES STATE C TIME CPU COMMAND 44 root 1 171 52 0K 16K CPU2 3 67:25 60.40% pagezero 13 root 1 171 52 0K 16K RUN 0 94:38 17.09% idle: cpu0 10 root 1 171 52 0K 16K RUN 3 92:38 15.19% idle: cpu3 36391 vscan 1 20 0 121M 80732K lockf 0 0:04 12.35% perl5.8.8 36705 user1 1 131 0 15792K 3352K RUN 3 0:03 11.67% imapd 36301 vscan 1 20 0 122M 81896K lockf 1 0:09 10.79% perl5.8.8 36180 vscan 1 20 0 125M 84988K lockf 1 0:17 10.01% perl5.8.8 36266 vscan 1 20 0 126M 84964K lockf 0 0:10 8.89% perl5.8.8 14 root 1 -44 -163 0K 16K WAIT 3 99:21 7.57% swi1: net 15 root 1 -32 -151 0K 16K WAIT 0 93:54 7.47% swi4: clock sio 1211 bind 1 101 0 74372K 71180K select 0 108:51 7.37% named 12 root 1 171 52 0K 16K RUN 1 71:18 6.93% idle: cpu1 36731 user2 1 130 0 15648K 3228K CPU3 0 0:02 6.54% imapd 94452 nobody 1 20 0 144M 38836K lockf 1 0:17 5.76% httpd 35798 vscan 1 20 0 127M 86464K lockf 0 0:25 5.08% perl5.8.8 11 root 1 171 52 0K 16K RUN 2 61:38 4.64% idle: cpu2 36405 root 1 100 0 7416K 2680K CPU0 0 0:02 3.61% top 94286 nobody 1 98 0 142M 36928K select 0 0:11 3.32% httpd 36337 user3 1 4 0 10768K 2968K sbwait 3 0:03 3.08% qpopper 47 root 1 20 0 0K 16K syncer 0 50:27 2.83% syncer 94448 nobody 1 20 0 148M 43092K lockf 1 0:45 2.69% httpd 24 root 1 -68 -187 0K 16K WAIT 3 31:51 2.64% irq17: em0 25 root 1 -68 -187 0K 16K WAIT 0 47:22 2.59% irq18: em1 em2 3554 vscan 2 99 0 104M 93616K ucond 3 11:59 2.39% clamd 5028 nobody 1 98 0 144M 38840K select 0 0:11 2.34% httpd 7187 root 1 131 0 9656K 4016K RUN 3 3:59 2.25% sendmail 36179 vscan 1 103 0 126M 85340K select 0 0:12 2.20% perl5.8.8 1582 root 1 130 0 5868K 1536K allpro 3 9:36 1.32% inetd 45382 root 1 97 0 3688K 1308K select 0 8:30 1.32% syslogd >>>My system is being killed by a system cpu always between 50-70% on >>>Freebsd 6.2 with SMP enabled. Does anyone know of a quick tool to be >>>able to tell me what is making up that 50-70% for the System Cup Usage. >>Uh you had the right idea but forgot to paste the relevant bits ;-) >> >>top *is* the tool for working out what processes are using your CPU. >>Although your high load average suggests that your system is just >>heavily loaded and it is expected that it will be using 100% CPU. >Maybe Paul havn't found "S" in top? > >Enter "S" when runing top (to show system processes) and you might >find out more about whats going on. > > > /Chris
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070423004105.7984082E>