Top output isn't quite right

[Jonathan Glaschke]

On Sat, Jul 16, 2005 at 01:54:56PM -0400, Louis LeBlanc wrote:
> I'm confused about the output from top in 5.4-RELEASE-p4.  The gkrellm
> CPU monitor is telling me I'm running at 95% load while updating the
> portsdb, but top tells me everything is taking 0.0% of the CPU.
> 
> It may have something to do with the CPU HT being enabled, but I don't
> know.  The system has one physical CPU, but with HT, some processes
> run on CPU 0 while others run on CPU 1.
> 
> Any idea how to get top to get the actual (or virtual) CPU loads?
> 
> Here is the output I'm seeing:
> 
> last pid: 82084;  load averages:  2.10,  1.88,  1.17 up 9+17:06:23  13:47:44
> 131 processes: 3 running, 128 sleeping
> CPU states: 73.3% user,  0.0% nice, 23.5% system,  0.0% interrupt, 3.1% idle
> Mem: 296M Active, 477M Inact, 148M Wired, 27M Cache, 111M Buf, 44M Free
> Swap: 2048M Total, 70M Used, 1978M Free, 3% Inuse
> 
>   PID USERNAME  PRI NICE   SIZE    RES STATE  C   TIME   WCPU    CPU COMMAND
> 81981 root        8    0  1524K  1368K wait   1   0:00  3.00%  0.15% make
>  1024 root       96    0   181M   167M select 0  86:36  0.10%  0.10% Xorg
> 81993 root        8    0   816K   652K wait   0   0:00  1.00%  0.05% make
>  1067 userid     20    0 15016K  8040K kserel 0 241:36  0.00%  0.00% gkrellm
>   628 www        20    0   233M 11852K kserel 0  12:00  0.00%  0.00% java
>   250 root       96    0  3428K  1208K select 0   6:29  0.00%  0.00% ppp
>   804 mysql      20    0 56568K  1728K kserel 0   3:43  0.00%  0.00% mysqld
>   401 _pflogd   -58    0  1592K   580K bpf    0   1:42  0.00%  0.00% pflogd
> 76688 bind       20    0  7004K  3648K kserel 0   1:10  0.00%  0.00% named
> 12360 root        4    0 36548K 34824K accept 1   0:55  0.00%  0.00% perl
> 12358 root        4    0 37232K 35204K accept 1   0:55  0.00%  0.00% perl
> 12361 root        4    0 43136K 40080K accept 1   0:54  0.00%  0.00% perl
> 41249 userid     96    0  3452K  1648K select 0   0:37  0.00%  0.00% xscreensaver
>  1050 userid     96    0  3384K  1740K select 0   0:34  0.00%  0.00% fvwm2
>   893 root       96    0  5448K   640K select 0   0:29  0.00%  0.00% nmbd
> 14699 root        4    0  3628K   392K select 0   0:23  0.00%  0.00% master
>   651 root       96    0 21852K  2192K select 0   0:20  0.00%  0.00% httpd
>   310 root       96    0  1360K   280K select 0   0:17  0.00%  0.00% syslogd
> 64422 root       96    0  2436K  1544K select 0   0:13  0.00%  0.00% fam
>   339 root       96    0  1508K   584K select 0   0:12  0.00%  0.00% rpcbind
> 
> -- 
> Louis LeBlanc                          FreeBSD-at-keyslapper-DOT-net
> Fully Funded Hobbyist,                   KeySlapper Extrordinaire :)
> Please send off-list email to:         leblanc at keyslapper d.t net
> Key fingerprint = C5E7 4762 F071 CE3B ED51  4FB8 AF85 A2FE 80C8 D9A2
> 
> Corollary:
>   If you are given a take-home exam, you will forget where you live.

Hello,

CPU states: 73.3% user,  0.0% nice, 23.5% system,  0.0% interrupt, 3.1% idle

That means you have 96,8% CPU load.
The load shown for each process is "WCPU", that is the weighted cpu percentage,
as you can read in "man top".

Jonathan

-- 
 | /"\   ASCII Ribbon   | Jonathan Glaschke - Lorenz-Görtz-Straße 71,
 | \ / Campaign Against | 41238 Mönchengladbach, Tel: 02166-265876
 |  X    HTML In Mail   | Mobil: 0162-3390789, ICQ: 231021883
 | / \     And News     | http://jonathan-glaschke.de/