home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!sun-barr!ames!lll-winken!phoenix.ocf.llnl.gov!terry
- From: terry@phoenix.ocf.llnl.gov (Terry Heidelberg x24154)
- Newsgroups: comp.unix.aix
- Subject: Re: Monitor 1.05.1 (was Re: Uptime problem in Aix 3.2)
- Message-ID: <133786@lll-winken.LLNL.GOV>
- Date: 19 Aug 92 00:47:55 GMT
- References: <1992Aug11.170432.420@mintaka.lcs.mit.edu> <JMAKI.92Aug12160253@veivi.hut.fi>
- Sender: usenet@lll-winken.LLNL.GOV
- Organization: LLNL
- Lines: 32
- Nntp-Posting-Host: phoenix.ocf.llnl.gov
-
- In article <JMAKI.92Aug12160253@veivi.hut.fi>, jmaki@veivi.hut.fi (Jussi Maki)
- writes:
- |>
- |> I'm sorry, but monitor 1.05.1 will show wrong uptime values when used
- |> with AIX version 3.2. This is due the modifications that IBM has made to their
- |> system. If anybody knows the real story of how sysinfo.runque and
- |> sysinfo.runocc values (described in /usr/include/sys/sysinfo.h) are
- |> handled in aix3.2 versus aix3.1 I would be very glad to know it. Those
- |> variables are used to calculate load average values. The documention
- |> with include files don't seem to be up to date.
- |>
- |> I have tried to modify load average calculations to work with aix3.2
- |> but there might still be some bugs with those modifications. I'll
- |> make the beta test version available from
- |> ftp.funet.fi:pub/unix/AIX/RS6000/monitor-1.06beta.tar.Z with anonymous
- |> ftp. I don't guarantee that it will work since I've been very busy
- |> with other projects here.
- |>
- |> About that real memory value, I don't know the correct vay to do
- |> it in AIX3.?. My way has worked with 300 and 500 series. We don't have
- |> any RS/6000 model 220's here so perhaps somebody who has one could
- |> try to peek the way to look the real memory value. (Or IBM could
- |> tell us the correct way to do it?). If there would be an "AIX
- |> Internals manual" life would be lot easier with this.
-
- Or better yet, maybe IBM could take over the maintenance of 'monitor',
- include it in the base software, and see to it that it works right on
- each release? It sure beats the hell out of PS for watching the
- dynamic process behavior of a system.
-
- How do IBM software developers observe the dynamic process behavior of
- AIX, anyhow? By installing a version of 'monitor'?
-