Date: Thu, 01 Jan 1998 13:54:35 +0000 From: Brian Somers <brian@awfulhak.org> To: dg@root.com Cc: Brian Somers <brian@awfulhak.org>, freebsd-hackers@FreeBSD.ORG Subject: Re: load averages never decrease to 0 (FreeBSD 2.1.6) Message-ID: <199801011354.NAA17928@awfulhak.demon.co.uk> In-Reply-To: Your message of "Wed, 31 Dec 1997 12:02:38 PST." <199712312002.MAA26043@implode.root.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> >[.....]
> >> I have my laptop running 60 ppp processes - all idle - for the last
> >> couple of days. The load average tapers out at exactly 1.0. It's a
> >> P120 w/ 40Mb of memory running 2.2-stable from about 3 days ago.
> >[.....]
> >
> >I take that back. It's now running at exactly 5.0 ! Nothing else
> >(except the 60 ppps) is running ?????????
>
> The code that calculates the load average is very simple and can be found
> in /sys/vm/vm_meter.c. You should be able to take the output of 'ps' and
> determine why it is that the code is getting it wrong.
It's not really that simple, is it ? For example, updatepri() in
kern_synch.c adjusts p_slptime based on the load average....
All of the ppp processes are ``Ss'' (sleeping session leaders),
and always blocked in `select' according to ps -l.
> -DG
>
> David Greenman
> Core-team/Principal Architect, The FreeBSD Project
--
Brian <brian@Awfulhak.org>, <brian@FreeBSD.org>, <brian@OpenBSD.org>
<http://www.Awfulhak.org>;
Don't _EVER_ lose your sense of humour....
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199801011354.NAA17928>