Path: utzoo!utgpu!cunews!bnrgate!brtph3!brchh104!brchs1!bnr.ca!rice.edu!sun-spots-request
From: cel!ag@relay.eu.net (Andrew R. Gray)
Newsgroups: comp.sys.sun
Subject: Re: Artificially high load average
Keywords: Miscellaneous
Message-ID: <1416@brchh104.bnr.ca>
Date: 23 Jan 91 17:52:24 GMT
Sender: news@brchh104.bnr.ca
Organization: Sun-Spots
Lines: 13
Approved: Sun-Spots@rice.edu
X-Refs:  Original: v10n24
X-Sun-Spots-Digest: Volume 10, Issue 29, message 1
X-Note: Submissions: sun-spots@rice.edu, Admin: sun-spots-request@rice.edu

In article <1352@brchh104.bnr.ca> jon@netlabs.com (Jonathan Biggar) writes:

>On the fileserver, I have noticed that even when the entire network is
>idle, I often see the load average sit at a very constant level of 3 or 4.
>I have watched the ethernet and I don't see any traffic which could cause
>this, and top reports that only one process is in the running state at any
>time, and the system is running at 98-99% idle.

I have started noticing this on two Sun-3/280's, each with 3 diskless
clients after we upgraded them to 4.1. The frantic (untraceable) activity
continues for up to 30 minutes after they actually cease doing any real
work. Is this some form on garbage collection that's carried out when the
system isn't doing any real work ?