Re: runaway shell (tcsh) processes and load average

[Niklas Hallqvist <niklas@appli.se>]

> Date: Wed, 3 Mar 1999 01:32:43 -0500 (EST)
> From: John Nitis <john@interscape.com>

> The other question I had relates to the load average.  It's constantly
> between 0.10 - 0.20 even when there's nothing running to bring the load
> average up that high.  Is there anything that may be causing this on
> OpenBSD/sparc? (110mhz sparc4 with 64mb ram) `top' shows:

There's often a misconception of what load average really means.  Load
average is the average queue length of the runnable processes at some
sampling events.  Depending on the interrelationship of the sample
rate and other events in the system the load average can give what can
be regarded as weird results.

I don't see 0.1 - 0.2 in LAV as a high value.  It meant that every
fifth to tenth sample a process was runnable.  Surely you have a fair
amount of background processes, and some of them are actually runnable
once in a while.  It seems you are more interested in the CPU% than in
the load on the scheduler.  If so, that is what you should watch.

Niklas