strange behaviour on FreeBSD 7.1

[Jonathan Chen]

2010/11/18 Коньков Евгений <kes-kes at yandex.ru>:
> Hi.
>
> Sometimes system goes to this situation: 0% idle and no processes take CPU time
>
> #top -SIHP
> last pid: 62813;  load averages:  4.17,  3.64,  2.16   up 28+06:44:02  20:41:41
> 155 processes: 7 running, 129 sleeping, 19 waiting
> CPU: 99.3% user,  0.0% nice,  0.7% system,  0.0% interrupt,  0.0% idle
> Mem: 177M Active, 27M Inact, 124M Wired, 13M Cache, 60M Buf, 148M Free
> Swap: 2048M Total, 51M Used, 1997M Free, 2% Inuse
>
>  PID USERNAME   PRI NICE   SIZE    RES STATE    TIME   WCPU COMMAND
>
> #top
> last pid: 62852;  load averages:  4.10,  3.67,  2.22   up 28+06:44:36  20:42:15
> 172 processes: 4 running, 168 sleeping
> CPU: 99.6% user,  0.0% nice,  0.4% system,  0.0% interrupt,  0.0% idle
> Mem: 203M Active, 27M Inact, 125M Wired, 13M Cache, 60M Buf, 121M Free
> Swap: 2048M Total, 51M Used, 1997M Free, 2% Inuse
>
>  PID USERNAME    THR PRI NICE   SIZE    RES STATE    TIME   WCPU COMMAND
> 62817 root         36  -8    0 29696K 23276K piperd   0:00  7.62% perl5.8.8

If you look at the "last pid" between the 2 top-output snippets, you
can see that approx 40 processes came and went in-between. This
indiciates that you probably have some script running that's spawning
a large number of short-lived processes.
-- 
Jonathan Chen <jonc at chen.org.nz>