Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 22 Jan 2009 00:05:51 +0200
From:      KES <kes-kes@yandex.ru>
To:        Lowell Gilbert <freebsd-questions-local@be-well.ilk.org>
Cc:        freebsd-questions@freebsd.org
Subject:   Re[2]: 'top' shows wrong CPU usage
Message-ID:  <1312688589.20090122000551@yandex.ru>
In-Reply-To: <44k58os79x.fsf@be-well.ilk.org>
References:  <571280828.20090121220816@yandex.ru> <44k58os79x.fsf@be-well.ilk.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Здравствуйте, Lowell.

Вы писали 21 января 2009 г., 23:01:30:

LG> KES <kes-kes@yandex.ru> writes:

>> top shows often nonsense in CPU usage of a process, but totals are OK
>> and it seems that WCPU and CPU has no differences in results

LG> Why do you think it's nonsense?  It's averaged over a different time
LG> period than the "CPU:" line in the header, so I wouldn't expect it to
LG> match...

When I gzip some big files I see that 0% idle, but I can not obtain
from process list which process get all CPU.
I top -S for WCPU and top -S -C for CPU.
In both cases results are *useless*!

>> top -S
>> last pid: 66182;  load averages:  2.51,  2.15,  2.03             up 10+23:40:14  22:05:41
>> 798 processes: 6 running, 772 sleeping, 1 zombie, 18 waiting, 1 lock
>> CPU:  4.4% user,  0.0% nice, 14.8% system, 16.7% interrupt, 64.0% idle
>> Mem: 264M Active, 60M Inact, 147M Wired, 6968K Cache, 60M Buf, 9888K Free
>> Swap: 2048M Total, 1903M Used, 145M Free, 92% Inuse
>>
>>   PID USERNAME    THR PRI NICE   SIZE    RES STATE    TIME   WCPU COMMAND
>>    11 root          1 171 ki31     0K     8K RUN    104.1H 88.48% idle: cpu0
>> 66178 firebird      1  49    0 23120K  5828K select   0:00  1.37% fb_inet_server
>> 66159 firebird      1  48    0 23120K  5760K select   0:01  1.17% fb_inet_server
>>  5156 root          1  44    0  9024K   544K select  57:39  0.68% snmpd
>> 66182 root          1  44    0  4556K  2608K RUN      0:00  0.68% top
>> 66147 root          1   8    0  3124K   840K nanslp   0:00  0.59% monitord
>> 66138 firebird      1  44    0 23120K  5736K select   0:01  0.49% fb_inet_server
>> 75745 www           1  44    0 24628K  9500K select   5:05  0.29% python2.5
>> 66180 firebird      1  46    0 23120K  5852K select   0:00  0.10% fb_inet_server
>>
>>
>> #top -S -C
>> last pid: 66209;  load averages:  2.13,  2.10,  2.02             up 10+23:41:07  22:06:34
>> 814 processes: 6 running, 788 sleeping, 1 zombie, 18 waiting, 1 lock
>> CPU:  9.3% user,  0.0% nice, 13.4% system, 12.8% interrupt, 64.5% idle
>> Mem: 269M Active, 56M Inact, 148M Wired, 12M Cache, 60M Buf, 3700K Free
>> Swap: 2048M Total, 1903M Used, 145M Free, 92% Inuse
>>
>>   PID USERNAME    THR PRI NICE   SIZE    RES STATE    TIME    CPU COMMAND
>>    11 root          1 171 ki31     0K     8K RUN    104.1H 92.29% idle: cpu0
>> 66138 firebird      1  49    0 23120K  5556K select   0:01  1.46% fb_inet_server
>> 66180 firebird      1  44    0 23120K  5612K select   0:01  0.59% fb_inet_server
>> 66209 root          1  44    0  4556K  2556K RUN      0:00  0.59% top
>> 66179 firebird      1  44    0 23120K  5624K select   0:01  0.49% fb_inet_server
>>  5156 root          1  44    0  9024K   544K select  57:39  0.39% snmpd
>> 66147 root          1   8    0  3124K   840K nanslp   0:01  0.39% monitord
>> 66178 firebird      1  44    0 23120K  5584K select   0:01  0.20% fb_inet_server
>>    12 root          1 -44    -     0K     8K WAIT   126.8H  0.00% swi1: net
>>    42 root          1 -68    -     0K     8K -      219:53  0.00% dummynet




-- 
С уважением,
 KES                          mailto:kes-kes@yandex.ru




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1312688589.20090122000551>