I'm curious ... If you type "sleep 1" at the command prompt, does it come
back in one second? On some of my installs, it returns in ~45 seconds!
I'm curious what you see when you try this ...
Scott C. Lemon
-----Original Message-----
From: darwin-x86-bounces+scott.lemon=email@hidden
[mailto:darwin-x86-bounces+scott.lemon=email@hidden]
On Behalf Of Andrew Gallatin
Sent: Friday, October 07, 2005 12:39 PM
To: email@hidden
Subject: incorrect CPU usage reported by top
I have a dual amd64 running darwin 8.0.1. The output from
top seems a bit strange.
When the system is idle:
Processes: 26 total, 2 running, 24 sleeping... 80 threads
14:31:39
Load Avg: 0.01, 0.03, 0.00 CPU usage: 0.0% user, 0.0% sys, 100.0% idle
SharedLibs: num = 0, resident = 0B code, 0B data, 0B LinkEdit
MemRegions: num = 1164, resident = 11.0M + 0B private, 14.8M shared
PhysMem: 26.5M wired, 17.8M active, 24.1M inactive, 68.4M used, 442M free
VM: 1.47G + 256M 3765(0) pageins, 0(0) pageouts
PID COMMAND %CPU TIME #TH #PRTS #MREGS RPRVT RSHRD RSIZE VSIZE
117 ntpd 257.4% 2:44:47 1 11 30 248K 2.56M 760K 29.0M
200 top 86.3% 1:09.05 1 18 58 1012K 5.13M 3.42M 95.3M
0 kernel_tas 42.9% 1:48:38 35 2 0 0K 0K 24.3M 401M
48 mDNSRespon 42.9% 47:49.09 4 50 61 772K 6.87M 3.80M 46.9M
<...>
And when something really is using CPU:
Processes: 31 total, 4 running, 27 sleeping... 85 threads
14:34:18
Load Avg: 0.31, 0.08, 0.02 CPU usage: 0.2% user, 73.5% sys, 26.3% idle
SharedLibs: num = 0, resident = 0B code, 0B data, 0B LinkEdit
MemRegions: num = 1413, resident = 12.3M + 0B private, 15.1M shared
PhysMem: 27.1M wired, 18.9M active, 25.3M inactive, 71.4M used, 439M free
VM: 1.82G + 256M 3765(0) pageins, 0(0) pageouts
PID COMMAND %CPU TIME #TH #PRTS #MREGS RPRVT RSHRD RSIZE VSIZE
230 netperf224 36245% 94:54:53 1 15 25 288K 2.03M 1.06M 86.3M
231 netserver2 33504% 87:02:06 1 10 26 220K 2.09M 568K 84.6M
0 kernel_tas 1744.% 1:56:09 35 2 0 0K 0K 24.8M 402M
117 ntpd 342.9% 2:49:43 1 11 30 248K 2.56M 760K 29.0M
It looks like some stats clock is running horribly fast or slow, so cpu
usage is off by a few orders of magnitude. But the system seems to keep
decent time.
Any ideas what is wrong?
Thanks,
Drew
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Darwin-x86 mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
ns.com
This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Darwin-x86 mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden