What's using all my CPU cycles
-
So here is where my basic BSD knowledege is badly letting me down. :(
Today I logged into my box to check the logs and find that the CPU is running almost 100% all the time. This is up from 5% at most before 5.30pm today. I have only upgraded to 2.0.1 4 days ago and have been having some trouble with a package but anyway I fired up top -SH and:
[2.0.1-RELEASE][root@pfsense.fire.box]/root(4): top -SH last pid: 28083; load averages: 2.99, 2.83, 1.40 up 0+00:05:22 23:30:31 116 processes: 5 running, 95 sleeping, 16 waiting CPU: 40.4% user, 0.0% nice, 59.6% system, 0.0% interrupt, 0.0% idle Mem: 86M Active, 18M Inact, 50M Wired, 1060K Cache, 46M Buf, 330M Free Swap: PID USERNAME PRI NICE SIZE RES STATE TIME WCPU COMMAND 3686 root 76 0 43356K 16348K piperd 0:08 4.98% php 4445 root 76 0 44380K 19496K ppwait 0:08 3.96% php 4146 root 76 0 44380K 20300K lockf 0:08 2.98% php 3510 root 76 0 44380K 17980K piperd 0:08 1.95% php 0 root -16 0 0K 88K sched 1:59 0.00% {swapper} 10 root 171 ki31 0K 8K RUN 0:24 0.00% idle 11 root -64 - 0K 128K WAIT 0:07 0.00% {irq14: ata0} 4 root -8 - 0K 8K - 0:02 0.00% g_down 1571 root 44 0 6588K 4400K kqread 0:01 0.00% lighttpd 11 root -32 - 0K 128K WAIT 0:01 0.00% {swi4: clock} 11 root -24 - 0K 128K WAIT 0:00 0.00% {swi6: task queue} 11 root -68 - 0K 128K WAIT 0:00 0.00% {irq18: em0 ath0+} 23243 root 64 20 47452K 17184K nanslp 0:00 0.00% php 22313 root 64 20 47452K 17220K nanslp 0:00 0.00% php 32258 root 64 20 47452K 17220K nanslp 0:00 0.00% php 59818 root 64 20 47452K 17220K nanslp 0:00 0.00% php 5983 root 64 20 47452K 17220K nanslp 0:00 0.00% php
ps aux doesn't show anything either. How do I find out what's using my CPU cycles?
Steve
-
check to make sure this isnt checked:
-
Thanks for the reply, that was my first thought also.
Checked it, polling is not enabled and never has been. Would be odd that it spontaneously started anyway.Steve
-
Is it constant right now?
From a console- Restart webconfigurator.
-
Ah, good suggestion but I have sorted it now.
I'm not sure it would have worked though since I twice restarted to whole box (you can see the gaps in the RRD graph) and that just made it worse!
It turns out I had left another computer on with multiple copies of the dashboard open that I had forgotten about. Interesting that the connection survived a reboot.
However would you not expect to be able to see the system load using top -SH? :-
Is there some other tool I could use in future to diagnose something like this?Steve