Current-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: sleep sleeping too long

"Jeremy C. Reed" <> writes:

>> For instance, heavy I/O load.
>> Step by step explanation:
>>    date     - starts "date", prints time, exits
>>    sleep 3  - starts "sleep", sleeps for 3 sec.
>>    date     - waits for I/O (may wait any time, 5 sec isn't unusually long),
>>               starts "date", prints time, exits
> I don't see that in my case. (And system is very responsive.)
> Better example:
> tx:work$ time sleep 3 ; time sleep 5 ; time sleep 10
>     6.81s real     0.00s user     0.01s system
>     8.52s real     0.00s user     0.00s system
>    25.84s real     0.01s user     0.00s system

"time" works the same way, you cannot see the delay when "sleep" is loaded.

When no swap involved, the system stays responsive even under heavy disk I/O.
(I don't deny, that there're other causes of such behaviour.)


Home | Main Index | Thread Index | Old Index