home *** CD-ROM | disk | FTP | other *** search
- From: davidb@hpuadla.aus.hp.com (David Booth)
- Date: Wed, 9 Sep 1992 11:36:43 GMT
- Subject: Re: Mysterious load on HP9000/730
- Message-ID: <1992Sep9.113643.126@hparc0.aus.hp.com>
- Organization: Adelaide, Australia, Smallest HP DownUnder
- Path: sparky!uunet!cis.ohio-state.edu!zaphod.mps.ohio-state.edu!usc!elroy.jpl.nasa.gov!sdd.hp.com!scd.hp.com!hplextra!hpcc05!hparc0.aus.hp.com!davidb
- Newsgroups: comp.sys.hp
- References: <1992Sep3.064444.21989@yc.estec.esa.nl>
- Sender: news@hparc0.aus.hp.com (News Adminstrator)
- X-Newsreader: Tin 1.1.3 PL5
- Lines: 25
-
- Duncan Gibson (duncan@yc.estec.esa.nl) wrote:
- [stuff deleted]
- : constant load of 1.0 (yesterday it was 2.0 and later on there was a case
- : of 3.0) and this lasts for several (tens of) minutes. However, when I
- [stuff deleted]
- : This isn't a problem, per se, but I'm very curious about it. Has anyone
- :
- I suspect this might come from the way laoding is measured. Does your tool
- use the athena load-widget, or something like that? Odd programs may cause
- strange effects on the load measurement:
-
- eg: the program "kalied" that draws kaleidescopes on an X11 display -
- it uses under 2% of my machine (a 9000/360), yet registers on
- a number of commands (eg uptime, xload etc) as a constant load of "1".
- Perhaps someone is running a similar situation? (I have seen tar(1) do
- this as well).
- --
- #include <standard_disclaimer>
- All opinions in this posting are mine, with no endorsement by my employer.
- ---
- David Booth Phone: +61 8 366-5126 Fax : +61 8 332-0020
- Inet : davidb@hpuadla.aus.hp.com DESK: David BOOTH / HP9061
- Hewlett-Packard Australia A.C.N. 004 394 763
- 191 Fullarton Rd.,
- Dulwich, SA, 5065, AUSTRALIA.
-