From owner-pcp@oss.sgi.com Mon Jan  1 14:20:19 2001
Received:  by oss.sgi.com id <S553794AbRAAWUK>;
	Mon, 1 Jan 2001 14:20:10 -0800
Received: from eik.ii.uib.no ([129.177.16.3]:55202 "EHLO ii.uib.no")
	by oss.sgi.com with ESMTP id <S553790AbRAAWTs>;
	Mon, 1 Jan 2001 14:19:48 -0800
Received: from granbar-192.ii.uib.no (granbar.ii.uib.no) [129.177.192.137] 
	by ii.uib.no with esmtp (Exim 3.03)
	id 14DDJ8-00060o-00 ; Mon, 01 Jan 2001 23:19:58 +0100
Received: (from jfm@localhost)
	by granbar.ii.uib.no (8.9.3+Sun/8.9.3) id XAA15685;
	Mon, 1 Jan 2001 23:19:46 +0100 (MET)
Date:   Mon, 1 Jan 2001 23:19:45 +0100
From:   Jan-Frode Myklebust <janfrode@parallab.uib.no>
To:     "Philip J. Mucci" <mucci@cs.utk.edu>
Cc:     pcp@oss.sgi.com
Subject: Re: PCP + LSF
Message-ID: <20010101231945.A15668@ii.uib.no>
Mail-Followup-To: "Philip J. Mucci" <mucci@cs.utk.edu>, pcp@oss.sgi.com
References: <3A3FA8B3.768CF7A7@sgi.com> <Pine.LNX.4.21.0012201027580.29856-100000@sandpit.melbourne.sgi.com> <20001230132821.A7485@ii.uib.no> <3A4FCCF0.87A54742@cs.utk.edu>
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="TB36FDmn/VVEgNH/"
Content-Disposition: inline
In-Reply-To: <3A4FCCF0.87A54742@cs.utk.edu>; from mucci@cs.utk.edu on Sun, Dec 31, 2000 at 04:18:57PM -0800
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing


--TB36FDmn/VVEgNH/
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Sun, Dec 31, 2000 at 04:18:57PM -0800, Philip J. Mucci wrote:
> Hi Jan-Frode,
> 
> Can you send me the shell script you used?
> 

Ooops, forgot to include it.. here it is. I'm running this every X minute,
to keep the data up to date. Will probably write something better soon, but
this is working fine as a simple demo...


  -jf

--TB36FDmn/VVEgNH/
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="pbspcp.ksh"

#! /bin/ksh -
#
# Feeding PBS data into PCP trace
#
# To install the trace pmda:
#
# cd /var/pcp/pmdas/trace
# ./Install
#

QUEUES="express parallel monster normal"

for QUEUE in $QUEUES ; do
	TQUEUED=`qstat -a  | tail +6 | grep " $QUEUE " | grep " Q " | \
		awk '{print $7}'`
	QUEUED=0
	for TQUEUED in $TQUEUED; do QUEUED=$(($QUEUED + $TQUEUED)); done
	pmtrace -q -v $QUEUED ${QUEUE}-queued

	################################################################

	TRUN=`qstat -a | tail +6 | grep " $QUEUE " | grep " R " | \
		awk '{print $7}'`
	RUN=0
	for TRUN in $TRUN; do RUN=$(($RUN + $TRUN)); done
	pmtrace -q -v $RUN $QUEUE

	################################################################

	THELD=`qstat -a | tail +6 | grep " $QUEUE " | grep " H " | \
		awk '{print $7}'`
	HELD=0
	for THELD in $THELD ; do HELD=$(($HELD + $THELD)); done
	pmtrace -q -v $HELD ${QUEUE}-held

	################################################################

	TWAIT=`qstat -a | tail +6 | grep " $QUEUE " | grep " W " | \
		awk '{print $7}'`
	WAIT=0
	for TWAIT in $TWAIT ; do WAIT=$(($WAIT + $TWAIT)); done
	pmtrace -q -v $WAIT ${QUEUE}-wait

	################################################################
done

--TB36FDmn/VVEgNH/--

From owner-pcp@oss.sgi.com Mon Jan  1 16:10:00 2001
Received:  by oss.sgi.com id <S553798AbRABAJk>;
	Mon, 1 Jan 2001 16:09:40 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:56157 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553795AbRABAJ3>;
	Mon, 1 Jan 2001 16:09:29 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id QAA05481
	for <pcp@oss.sgi.com>; Mon, 1 Jan 2001 16:09:21 -0800 (PST)
	mail_from (nathans@wobbly.melbourne.sgi.com)
Received: from wobbly.melbourne.sgi.com (wobbly.melbourne.sgi.com [134.14.55.135]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id LAA07178; Tue, 2 Jan 2001 11:08:04 +1100
Received: (from nathans@localhost) by wobbly.melbourne.sgi.com (980427.SGI.8.8.8/980728.SGI.AUTOCF) id LAA06724; Tue, 2 Jan 2001 11:08:03 +1100 (EDT)
From:   "Nathan Scott" <nathans@wobbly.melbourne.sgi.com>
Message-Id: <10101021108.ZM6684@wobbly.melbourne.sgi.com>
Date:   Tue, 2 Jan 2001 11:08:02 -0400
In-Reply-To: Jan-Frode Myklebust <janfrode@parallab.uib.no>
        "Re: PCP + LSF" (Dec 30,  1:28pm)
References: <3A3FA8B3.768CF7A7@sgi.com> 
	<Pine.LNX.4.21.0012201027580.29856-100000@sandpit.melbourne.sgi.com> 
	<20001230132821.A7485@ii.uib.no>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To:     Jan-Frode Myklebust <janfrode@parallab.uib.no>, pcp@oss.sgi.com
Subject: Re: PCP + LSF
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

hi,

On Dec 30,  1:28pm, Jan-Frode Myklebust wrote:
> Subject: Re: PCP + LSF
> ...
> I just have one small question.. How can I delete a value ("Hello World")
> exported to the trace.observe metrics? 
> 

there's no way to delete just one instance.  what you could do
though is reset the instance domains (using a command like:
"pmstore trace.control.reset 1") ...


$ pminfo -Tt trace.control.reset

trace.control.reset [clear all tags known to the trace PMDA]
Help:
Storing any value into this metric with pmstore(1) will cause the trace
PMDA to clear all tags for all metrics from its historical buffers and
begin afresh.

This is most useful when the instance domains of the trace metrics have
become cluttered with unwanted instances, and the instance domains need
to be refreshed without restarting pmcd(1).


cheers.

-- 
Nathan

From owner-pcp@oss.sgi.com Wed Jan  3 11:12:27 2001
Received:  by oss.sgi.com id <S553760AbRACTMR>;
	Wed, 3 Jan 2001 11:12:17 -0800
Received: from heffalump.fnal.gov ([131.225.9.20]:62693 "EHLO fnal.gov")
	by oss.sgi.com with ESMTP id <S553646AbRACTLw>;
	Wed, 3 Jan 2001 11:11:52 -0800
Received: from fnal.gov ([131.225.80.75])
 by smtp.fnal.gov (PMDF V6.0-24 #44770) with ESMTP id
 <0G6L00BT1PBR2O@smtp.fnal.gov> for pcp@oss.sgi.com; Wed,
 03 Jan 2001 13:11:51 -0600 (CST)
Date:   Wed, 03 Jan 2001 13:11:51 -0600
From:   Troy Dawson <dawson@fnal.gov>
Subject: lmsensors pmda
To:     PCP Mailing List <pcp@oss.sgi.com>
Message-id: <3A537977.EB35DB0D@fnal.gov>
MIME-version: 1.0
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.16-3smp i686)
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7bit
X-Accept-Language: en
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Howdy,
This might be a little late to ask since I'm about 75% done, but does anyone
have a pmda written for lm-sensors?
Thanks
Troy
-- 
__________________________________________________
Troy Dawson  dawson@fnal.gov  (630)840-6468
Fermilab  ComputingDivision/OSS  SCS Group
__________________________________________________

From owner-pcp@oss.sgi.com Wed Jan  3 14:22:20 2001
Received:  by oss.sgi.com id <S553650AbRACWWK>;
	Wed, 3 Jan 2001 14:22:10 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:18019 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553645AbRACWVt>;
	Wed, 3 Jan 2001 14:21:49 -0800
Received: from sydney.sydney.sgi.com ([134.14.48.2]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id OAA06946
	for <pcp@oss.sgi.com>; Wed, 3 Jan 2001 14:21:39 -0800 (PST)
	mail_from (kaos@melbourne.sgi.com)
Received: from kao2.melbourne.sgi.com by sydney.sydney.sgi.com via ESMTP (950413.SGI.8.6.12/930416.SGI)
	 id JAA11745; Thu, 4 Jan 2001 09:20:00 +1100
X-Mailer: exmh version 2.1.1 10/15/1999
From:   Keith Owens <kaos@melbourne.sgi.com>
To:     Troy Dawson <dawson@fnal.gov>
cc:     PCP Mailing List <pcp@oss.sgi.com>
Subject: Re: lmsensors pmda 
In-reply-to: Your message of "Wed, 03 Jan 2001 13:11:51 MDT."
             <3A537977.EB35DB0D@fnal.gov> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date:   Thu, 04 Jan 2001 09:19:59 +1100
Message-ID: <18241.978560399@kao2.melbourne.sgi.com>
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Wed, 03 Jan 2001 13:11:51 -0600, 
Troy Dawson <dawson@fnal.gov> wrote:
>Howdy,
>This might be a little late to ask since I'm about 75% done, but does anyone
>have a pmda written for lm-sensors?

Not that we know of.  All contributions gratefully accepted :)


From owner-pcp@oss.sgi.com Thu Jan  4 10:30:34 2001
Received:  by oss.sgi.com id <S553848AbRADSaO>;
	Thu, 4 Jan 2001 10:30:14 -0800
Received: from heffalump.fnal.gov ([131.225.9.20]:30895 "EHLO fnal.gov")
	by oss.sgi.com with ESMTP id <S553865AbRADSaF>;
	Thu, 4 Jan 2001 10:30:05 -0800
Received: from fnal.gov ([131.225.80.75])
 by smtp.fnal.gov (PMDF V6.0-24 #44770) with ESMTP id
 <0G6N00JQ3I24ZD@smtp.fnal.gov> for pcp@oss.sgi.com; Thu,
 04 Jan 2001 12:30:04 -0600 (CST)
Date:   Thu, 04 Jan 2001 12:30:03 -0600
From:   Troy Dawson <dawson@fnal.gov>
Subject: lmsensors pmda V0.5
To:     PCP Mailing List <pcp@oss.sgi.com>
Message-id: <3A54C12B.82F4B6AC@fnal.gov>
MIME-version: 1.0
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.16-3smp i686)
Content-type: multipart/mixed; boundary=------------AFFC7F44FB344F5F562B15FF
X-Accept-Language: en
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

This is a multi-part message in MIME format.
--------------AFFC7F44FB344F5F562B15FF
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Howdy All,
I finished my lmsensors pmda.  It works quite well (ok, that's my opinion, but
it defiantly works better than the other pmda's I've written).

The reason it is only version 0.5 instead of 1.0 is because of two reasons.
1) I don't have it set up to do 'instances' (more than one of the same
thing).  So if you have two of the same sensor, such as two lm75's, only one
of them is displayed.  I did that because I've never done instances before,
needed to get this done, and so that was the one corned I cut.  I do believe I
have the code written in such a way as to do instances quite easily.
2) I don't have all of the sensors that lmsensors can monitor, only the ones
that I have on my machines.  This is once again, partly a time reason, partly
not.  It's awfully hard to check your code, when you don't have anything to
check it against.  I also couldn't figure out how they get the values that
they do for the eeprom, but once I do, I could easily put it into the code.

Current Sensors supported
lm75
lm79
w83781d

NOTE: You do have to have lm-sensors working to get this information.  This
isn't something that installs lm-sensors for you.  I just wanted to make that
clear.

Thanks
Troy

-- 
__________________________________________________
Troy Dawson  dawson@fnal.gov  (630)840-6468
Fermilab  ComputingDivision/OSS  SCS Group
__________________________________________________
--------------AFFC7F44FB344F5F562B15FF
Content-Type: application/octet-stream;
 name="lmsensors.0.5.tgz"
Content-Transfer-Encoding: base64
Content-Disposition: attachment;
 filename="lmsensors.0.5.tgz"
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--------------AFFC7F44FB344F5F562B15FF--


From owner-pcp@oss.sgi.com Thu Jan  4 13:02:55 2001
Received:  by oss.sgi.com id <S553721AbRADVCp>;
	Thu, 4 Jan 2001 13:02:45 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:16922 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553700AbRADVCU>;
	Thu, 4 Jan 2001 13:02:20 -0800
Received: from rattle.melbourne.sgi.com ([134.14.55.145]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id NAA08437
	for <pcp@oss.sgi.com>; Thu, 4 Jan 2001 13:00:18 -0800 (PST)
	mail_from (kenmcd@melbourne.sgi.com)
Received: from localhost (kenmcd@localhost)
	by rattle.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id HAA73273;
	Fri, 5 Jan 2001 07:58:58 +1100 (AEDT)
X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs
Date:   Fri, 5 Jan 2001 07:58:58 +1100
From:   Ken McDonell <kenmcd@melbourne.sgi.com>
Reply-To: kenmcd@sgi.com
To:     Troy Dawson <dawson@fnal.gov>
cc:     PCP Mailing List <pcp@oss.sgi.com>
Subject: Re: lmsensors pmda V0.5
In-Reply-To: <3A54C12B.82F4B6AC@fnal.gov>
Message-ID: <Pine.SGI.4.21.0101050753190.4050860-100000@rattle.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
Content-ID: <Pine.SGI.4.21.0101050753192.4050860@rattle.melbourne.sgi.com>
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Thu, 4 Jan 2001, Troy Dawson wrote:
> ...
> I finished my lmsensors pmda.

Cool.

Troy, would you like to incorporate this into the PCP distributions?

The options are

(a) make the gzip archive available in the PCP "contrib" download
    area on oss.sgi.com, or

(b) add the source to the PCP open source release (we have a relatively
    painless takeback agreement to cover contributions like this, which
    merely confirms your ownership of the code and agrees to the
    distribution under the terms and conditions of the original PCP
    release, i.e. GPL or LGPL as appropriate).


From owner-pcp@oss.sgi.com Fri Jan  5 14:05:45 2001
Received:  by oss.sgi.com id <S553719AbRAEWFZ>;
	Fri, 5 Jan 2001 14:05:25 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:22024 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553699AbRAEWFR>; Fri, 5 Jan 2001 14:05:17 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id OAA09287
	for <pcp@oss.sgi.com>; Fri, 5 Jan 2001 14:13:56 -0800 (PST)
	mail_from (kenmcd@melbourne.sgi.com)
From:   kenmcd@melbourne.sgi.com
Received: from [192.82.201.66] ([192.82.201.66]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id JAA07039; Sat, 6 Jan 2001 09:03:51 +1100
Date:   Sat, 6 Jan 2001 09:07:43 +1100 (EST)
Reply-To: kenmcd@melbourne.sgi.com
To:     Alan Bailey <abailey@ncsa.uiuc.edu>
cc:     pcp@oss.sgi.com
Subject: Re: shping
In-Reply-To: <Pine.LNX.4.10.10011091527140.20621-100000@osage.ncsa.uiuc.edu>
Message-ID: <Pine.LNX.4.21.0101060827580.14860-100000@kenj-pc.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Thanks for your feedback Alan.

I am going to start the wheels in motion to move the shping
PMDA from the proprietary PCP RPM (pcp-pro) into the open source
distribution.

We are in the middle of a massive source tree reorganization at
the moment to allow us to release all of the core PCP pieces for
Linux and IRIX from a single source tree ... the shping migration
will happen after this reorganization is complete.

On Thu, 9 Nov 2000, Alan Bailey wrote:

> This is in response to your note about shping in the pcp mailing list
> a while ago.  Here are my answers to your questions..
> 
> > I would be willing to consider the case for moving the shping PMDA to
> > open source if I had some justification from the community ... if this
> > is of interest to you, drop me a note affressing questions like:
> >     - why you want it?
> 
> We are currently looking at using pcp for monitoring of very many aspects
> of linux clustering.  These aspects would include the monitoring of the
> basic processes necessary for the normal operation of the machines.
> shping provides a very good method of doing this, along with the other
> thousand things that pcp can do.
> 
> >     - how would you use it?
> 
> We would monitor processes on a cluster of machines.  The number of
> processes might hover around 10 or so, and the number of machines would be
> very high.
> 
> >     - is any SGI hardware involved?
> 
> No.  Well, execpt NFS mounting of home directories that might be located
> on SGI machines.  These NFS mounts would be monitored with shping.
> 
> >     - would this be a make-or-break issue for you embracing PCP in your
> >       environment?
> 
> I doubt it.  PCP has shown to be useful in many other ways.  However, I
> have people above me that might choose otherwise, and of course I would go
> along with them.  :)
> 
> Thanks,
> Alan
> 
> 
> -- 
>  Alan Bailey
> 




From owner-pcp@oss.sgi.com Fri Jan  5 14:06:26 2001
Received:  by oss.sgi.com id <S553738AbRAEWGQ>;
	Fri, 5 Jan 2001 14:06:16 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:19 "EHLO sgi.com") by oss.sgi.com
	with ESMTP id <S553712AbRAEWGA>; Fri, 5 Jan 2001 14:06:00 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id OAA07888
	for <pcp@oss.sgi.com>; Fri, 5 Jan 2001 14:05:55 -0800 (PST)
	mail_from (kenmcd@melbourne.sgi.com)
From:   kenmcd@melbourne.sgi.com
Received: from [192.82.201.66] ([192.82.201.66]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id JAA07045; Sat, 6 Jan 2001 09:04:31 +1100
Date:   Sat, 6 Jan 2001 09:08:23 +1100 (EST)
Reply-To: kenmcd@melbourne.sgi.com
To:     Alan Bailey <abailey@ncsa.uiuc.edu>
cc:     pcp@oss.sgi.com
Subject: Re: nfs monitoring (also shping)
In-Reply-To: <Pine.LNX.4.10.10011141139230.26061-100000@osage.ncsa.uiuc.edu>
Message-ID: <Pine.LNX.4.21.0101060835310.14860-100000@kenj-pc.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Tue, 14 Nov 2000, Alan Bailey wrote:

> What is the suggested way to monitor nfs mounts?  I want to make sure that
> nfs mounted directories are still mounted and up on the remote host.  I
> couldn't find anything by looking at the nfs.* or nfs3.* metrics.

The nfs*.* metrics will only show NFS client activity NFS server, from
which it is difficult to intuit if NFS mounts are still working.

I'd suggest the best way to monitor this is to use shping (when it
becomes available), and periodically

	- try to read a file that is known to exist on the NFS mounted FS
	- try to write to the NFS mounted FS (assuming you can arrange
	  the protections appropriately)

When the NFS mount goes away, the read/write will either fail or be
timed out by shping ... in either case this is reflected in the value
of shping.status and/or shping.error.

Then use pmie to monitor the shping.status metric.

FYI, here is the help text for the shping metrics.

[root@kenj-pc shping]# pminfo -T shping

shping.status
Help:
As each command is executed, the success or failure is encoded in
shping.status, using the following values:

   -1   PMDA is initializing and command has not been run yet
    0	command completed and exit status was 0
    1	command completed and exit status was non-zero
    2	command was run but terminated by a signal
    3	command was run but did not complete (usually a timeout)
    4   command was not run due to some system error or resource
	availability

shping.error
Help:
As each command is executed, if there is a problem, the error
code or cause is stored in shping.error.

The interpretation of the value for shping.error depends on
shping.status as follows:

    If shping.status is 1 (the command was run but returned a non-zero
    exit status) then shping.error is the exit status.

    If shping.status is 2 (the command was run but was terminated by
    a signal) then shping.error is the signal number.

    If shping.status is 3 (the command did not complete) then
    shping.error is a PCP error codes: see pmerr(1).  Of particular
    relevance is -1008 (PM_ERR_TIMEOUT) when the command failed to
    complete in the time specified by shping.control.timeout.

    If shping.status is 4 (the commands was not run) then shping.error
    is the value of errno.

    Otherwise shping.error will be zero.

shping.cmd
Help:
The text of each sh(1) command run by the shping PMDA.

shping.time.real
Help:
This metric records the elapsed time in milliseconds for the most recent
execution of each command to be run by the shping PMDA.

Care should be used when interpreting the value if the corresponding
value for shping.status is non-zero, as the command may not have run to
completion.  If the command timed out, shping.time.real will be -1.

shping.time.cpu_usr
Help:
This metric records the user mode CPU time in milliseconds for the most
recent execution of each command to be run by the shping PMDA.

Care should be used when interpreting the value if the corresponding
value for shping.status is non-zero, as the command may not have run to
completion.  If the command timed out, shping.time.cpu_usr will be -1.

shping.time.cpu_sys
Help:
This metric records the system mode CPU time in milliseconds for the most
recent execution of each command to be run by the shping PMDA.

Care should be used when interpreting the value if the corresponding
value for shping.status is non-zero, as the command may not have run to
completion.  If the command timed out, shping.time.cpu_sys will be -1.

shping.control.numcmd
Help:
number of commands in the group to be run by the shping PMDA

shping.control.cycles
Help:
number of times the command group has been run by the shping PMDA

shping.control.cycletime
Help:
All commands are run by the shping PMDA are executed one after another
in a group, and the group is run once per "cycle" time.  This metric
reports the cycle time in seconds.

The cycle time may be changed dynamically by modifying this metric
with pmstore(1).

shping.control.timeout
Help:
The number of seconds the shping PMDA is willing to wait before
considering a single command to have timed out and killing it off.

The time out interval may be changed dynamically by modifying this
metric with pmstore(1).

shping.control.debug
Help:
The debug flag for the shping PMDA (see pmdbg(1)).  All trace and
diagnostic files are created in /var/adm/pcplog (unless $PCP_LOGDIR
is sent in the environment, see PMAPI(3)).

The debug flags DBG_TRACE_APPL0 (2048) and DBG_TRACE_APPL1 (4096)
may be used as follows:

DBG_TRACE_APPL0	- additional trace messages associated with the running
		  of each command appear in shping.log

DBG_TRACE_APPL1	- the standard output and standard error of each command
		  is appended to shping.out (instead of the default
		  /dev/null)

The debug flags may be changed dynamically by modifying this
metric with pmstore(1), e.g.
	$ pmstore shping.control.debug 6144
would enable both of the diagnostic traces associated with
DBG_TRACE_APPL0 and DBG_TRACE_APPL1.



From owner-pcp@oss.sgi.com Mon Jan  8 15:11:29 2001
Received:  by oss.sgi.com id <S553736AbRAHXLU>;
	Mon, 8 Jan 2001 15:11:20 -0800
Received: from ex1.ncsa.uiuc.edu ([141.142.2.9]:58084 "EHLO ex1.ncsa.uiuc.edu")
	by oss.sgi.com with ESMTP id <S553705AbRAHXK5>;
	Mon, 8 Jan 2001 15:10:57 -0800
Received: from mx1.ncsa.uiuc.edu (mx1.ncsa.uiuc.edu [141.142.2.8])
	by ex1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f08NAuw03368
	for <pcp@oss.sgi.com>; Mon, 8 Jan 2001 17:10:56 -0600 (CST)
X-Envelope-From: abailey@ncsa.uiuc.edu
X-Envelope-To: <pcp@oss.sgi.com>
Received: from osage.ncsa.uiuc.edu (osage.ncsa.uiuc.edu [141.142.2.56])
	by mx1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f08NAtP26112
	for <pcp@oss.sgi.com>; Mon, 8 Jan 2001 17:10:55 -0600 (CST)
Received: from localhost (abailey@localhost)
	by osage.ncsa.uiuc.edu (8.9.3/8.9.3) with ESMTP id RAA00711
	for <pcp@oss.sgi.com>; Mon, 8 Jan 2001 17:10:55 -0600
Date:   Mon, 8 Jan 2001 17:10:54 -0600 (CST)
From:   Alan Bailey <abailey@ncsa.uiuc.edu>
To:     pcp@oss.sgi.com
Subject: speed
In-Reply-To: <Pine.LNX.4.21.0101060827580.14860-100000@kenj-pc.melbourne.sgi.com>
Message-ID: <Pine.LNX.4.10.10101081645370.338-100000@osage.ncsa.uiuc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

I'm wondering what timing tests were used when developing PCP.  I don't
know if this could be answered by the open source developers, or would
need to be redirected to the main developers.  Basically, one of my
coworkers believes that the response time for PCP is pretty slow, and
believes that a better solution for sending data between hosts could be
made and would be significantly better. [1]

Right now it takes almost exactly 4 seconds to call pminfo on 87 hosts for 
mem.freemem.  I think this is very good.  We will be expanding to a few
hundred hosts in the near future though.  So basically, was any optimizing
done in regards to speed when designing PCP?  If so, is it possible to see
any kind of tests or decisions that were made to show that PCP is truly
optimized for speed?

Thanks,
Alan

[1] - One of the ideas he had was to use pcp to just collect data on the
localhost and store it in some structure.  The query mechanism between
hosts would then ask each individual host for data, and this cached data
would be returned.  This is obviously very similar to a caching PMDA,
which might be an option, but I forgot to mention it to him.  He was also
thinking of using UDP, a hierarchical grabbing scheme, and some other
techniques for getting data.  I don't think it would help much, networks
can only go so fast ;)


From owner-pcp@oss.sgi.com Tue Jan  9 06:38:15 2001
Received:  by oss.sgi.com id <S553776AbRAIOiH>;
	Tue, 9 Jan 2001 06:38:07 -0800
Received: from heffalump.fnal.gov ([131.225.9.20]:361 "EHLO fnal.gov")
	by oss.sgi.com with ESMTP id <S553729AbRAIOhy>;
	Tue, 9 Jan 2001 06:37:54 -0800
Received: from fnal.gov ([131.225.80.75])
 by smtp.fnal.gov (PMDF V6.0-24 #44770) with ESMTP id
 <0G6W00MD5GN4KM@smtp.fnal.gov> for pcp@oss.sgi.com; Tue,
 09 Jan 2001 08:37:53 -0600 (CST)
Date:   Tue, 09 Jan 2001 08:37:52 -0600
From:   Troy Dawson <dawson@fnal.gov>
Subject: Re: speed
To:     Alan Bailey <abailey@ncsa.uiuc.edu>
Cc:     pcp@oss.sgi.com
Message-id: <3A5B2240.A968C913@fnal.gov>
MIME-version: 1.0
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.16-3smp i686)
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7bit
X-Accept-Language: en
References: <Pine.LNX.4.10.10101081645370.338-100000@osage.ncsa.uiuc.edu>
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Howdy,
How are you calling this?  Is it a script that goes one at a time, I'd really
say it's your script, but I could be wrong.

Basically I've found that pcp is quite fast beyond anything else I've tried. 
Just for your info, I collect dozens of stuff from over 250 machines, and the
only slowness I experience is from my scripts that gather things (ie ... I
intentially put a pause in my collections so I don't overwhelm anything).

One thing might be that the time to get alot of data, and the time to get just
one metric, is about the same.  So for me the time/data ratio is quite high,
because I have alot of data.  And for you, the time/data ratio is quite low
because you are only getting one metric.  If that makes sense.

I've also found that doing multiple threads, and having them all get the info
at the same time, speeds things up immensly.  Instead of doing just one at a
time.

Troy

Alan Bailey wrote:
> 
> I'm wondering what timing tests were used when developing PCP.  I don't
> know if this could be answered by the open source developers, or would
> need to be redirected to the main developers.  Basically, one of my
> coworkers believes that the response time for PCP is pretty slow, and
> believes that a better solution for sending data between hosts could be
> made and would be significantly better. [1]
> 
> Right now it takes almost exactly 4 seconds to call pminfo on 87 hosts for
> mem.freemem.  I think this is very good.  We will be expanding to a few
> hundred hosts in the near future though.  So basically, was any optimizing
> done in regards to speed when designing PCP?  If so, is it possible to see
> any kind of tests or decisions that were made to show that PCP is truly
> optimized for speed?
> 
> Thanks,
> Alan
> 
> [1] - One of the ideas he had was to use pcp to just collect data on the
> localhost and store it in some structure.  The query mechanism between
> hosts would then ask each individual host for data, and this cached data
> would be returned.  This is obviously very similar to a caching PMDA,
> which might be an option, but I forgot to mention it to him.  He was also
> thinking of using UDP, a hierarchical grabbing scheme, and some other
> techniques for getting data.  I don't think it would help much, networks
> can only go so fast ;)

-- 
__________________________________________________
Troy Dawson  dawson@fnal.gov  (630)840-6468
Fermilab  ComputingDivision/OSS  SCS Group
__________________________________________________

From owner-pcp@oss.sgi.com Tue Jan  9 08:12:15 2001
Received:  by oss.sgi.com id <S553883AbRAIQMG>;
	Tue, 9 Jan 2001 08:12:06 -0800
Received: from ex1.ncsa.uiuc.edu ([141.142.2.9]:51614 "EHLO ex1.ncsa.uiuc.edu")
	by oss.sgi.com with ESMTP id <S553880AbRAIQL5>;
	Tue, 9 Jan 2001 08:11:57 -0800
Received: from mx1.ncsa.uiuc.edu (mx1.ncsa.uiuc.edu [141.142.2.8])
	by ex1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f09GBuw17783;
	Tue, 9 Jan 2001 10:11:56 -0600 (CST)
X-Envelope-From: abailey@ncsa.uiuc.edu
Received: from osage.ncsa.uiuc.edu (osage.ncsa.uiuc.edu [141.142.2.56])
	by mx1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f09GBtP04444;
	Tue, 9 Jan 2001 10:11:55 -0600 (CST)
Received: from localhost (abailey@localhost)
	by osage.ncsa.uiuc.edu (8.9.3/8.9.3) with ESMTP id KAA10808;
	Tue, 9 Jan 2001 10:11:55 -0600
Date:   Tue, 9 Jan 2001 10:11:55 -0600 (CST)
From:   Alan Bailey <abailey@ncsa.uiuc.edu>
To:     Troy Dawson <dawson@fnal.gov>
cc:     pcp@oss.sgi.com
Subject: Re: speed
In-Reply-To: <3A5B2240.A968C913@fnal.gov>
Message-ID: <Pine.LNX.4.10.10101091010040.10714-100000@osage.ncsa.uiuc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

I am calling things sequentially.  I do suspect that a multithreaded
approach would help.  What language did you write your tools in, and could
I see an example?

Thanks a lot.

Alan

On Tue, 9 Jan 2001, Troy Dawson wrote:

> Howdy,
> How are you calling this?  Is it a script that goes one at a time, I'd really
> say it's your script, but I could be wrong.
> 
> Basically I've found that pcp is quite fast beyond anything else I've tried. 
> Just for your info, I collect dozens of stuff from over 250 machines, and the
> only slowness I experience is from my scripts that gather things (ie ... I
> intentially put a pause in my collections so I don't overwhelm anything).
> 
> One thing might be that the time to get alot of data, and the time to get just
> one metric, is about the same.  So for me the time/data ratio is quite high,
> because I have alot of data.  And for you, the time/data ratio is quite low
> because you are only getting one metric.  If that makes sense.
> 
> I've also found that doing multiple threads, and having them all get the info
> at the same time, speeds things up immensly.  Instead of doing just one at a
> time.
> 
> Troy
> 
> Alan Bailey wrote:
> > 
> > I'm wondering what timing tests were used when developing PCP.  I don't
> > know if this could be answered by the open source developers, or would
> > need to be redirected to the main developers.  Basically, one of my
> > coworkers believes that the response time for PCP is pretty slow, and
> > believes that a better solution for sending data between hosts could be
> > made and would be significantly better. [1]
> > 
> > Right now it takes almost exactly 4 seconds to call pminfo on 87 hosts for
> > mem.freemem.  I think this is very good.  We will be expanding to a few
> > hundred hosts in the near future though.  So basically, was any optimizing
> > done in regards to speed when designing PCP?  If so, is it possible to see
> > any kind of tests or decisions that were made to show that PCP is truly
> > optimized for speed?
> > 
> > Thanks,
> > Alan
> > 
> > [1] - One of the ideas he had was to use pcp to just collect data on the
> > localhost and store it in some structure.  The query mechanism between
> > hosts would then ask each individual host for data, and this cached data
> > would be returned.  This is obviously very similar to a caching PMDA,
> > which might be an option, but I forgot to mention it to him.  He was also
> > thinking of using UDP, a hierarchical grabbing scheme, and some other
> > techniques for getting data.  I don't think it would help much, networks
> > can only go so fast ;)
> 
> -- 
> __________________________________________________
> Troy Dawson  dawson@fnal.gov  (630)840-6468
> Fermilab  ComputingDivision/OSS  SCS Group
> __________________________________________________
> 

-- 
 Alan Bailey


From owner-pcp@oss.sgi.com Tue Jan  9 09:44:25 2001
Received:  by oss.sgi.com id <S553900AbRAIRoP>;
	Tue, 9 Jan 2001 09:44:15 -0800
Received: from heffalump.fnal.gov ([131.225.9.20]:59798 "EHLO fnal.gov")
	by oss.sgi.com with ESMTP id <S553695AbRAIRn7>;
	Tue, 9 Jan 2001 09:43:59 -0800
Received: from fnal.gov ([131.225.80.75])
 by smtp.fnal.gov (PMDF V6.0-24 #44770) with ESMTP id
 <0G6W00A17P9A9M@smtp.fnal.gov> for pcp@oss.sgi.com; Tue,
 09 Jan 2001 11:43:58 -0600 (CST)
Date:   Tue, 09 Jan 2001 11:43:58 -0600
From:   Troy Dawson <dawson@fnal.gov>
Subject: Re: speed
To:     Alan Bailey <abailey@ncsa.uiuc.edu>
Cc:     pcp@oss.sgi.com
Message-id: <3A5B4DDE.361D0BAE@fnal.gov>
MIME-version: 1.0
X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.16-3smp i686)
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7bit
X-Accept-Language: en
References: <Pine.LNX.4.10.10101091010040.10714-100000@osage.ncsa.uiuc.edu>
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Hi Alan,
I write my code in java, and have it do a remote procedure call to pminfo.  I
currently have it doing 10 threads.  There is a thread monitor that only
releases the 10 threads every one second.  I suppose I can make alot more
threads now that I have alot more systems, but I haven't.

If your interested in java code I could send my prototype stuff, but not alot
of folks are, so I didn't want to waste mailbox space if you didn't.

Troy

Alan Bailey wrote:
> 
> I am calling things sequentially.  I do suspect that a multithreaded
> approach would help.  What language did you write your tools in, and could
> I see an example?
> 
> Thanks a lot.
> 
> Alan
> 
-- 
__________________________________________________
Troy Dawson  dawson@fnal.gov  (630)840-6468
Fermilab  ComputingDivision/OSS  SCS Group
__________________________________________________

From owner-pcp@oss.sgi.com Tue Jan  9 10:43:16 2001
Received:  by oss.sgi.com id <S553921AbRAISnG>;
	Tue, 9 Jan 2001 10:43:06 -0800
Received: from ex1.ncsa.uiuc.edu ([141.142.2.9]:52142 "EHLO ex1.ncsa.uiuc.edu")
	by oss.sgi.com with ESMTP id <S553916AbRAISmw>;
	Tue, 9 Jan 2001 10:42:52 -0800
Received: from mx1.ncsa.uiuc.edu (mx1.ncsa.uiuc.edu [141.142.2.8])
	by ex1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f09Igow28201;
	Tue, 9 Jan 2001 12:42:50 -0600 (CST)
X-Envelope-From: abailey@ncsa.uiuc.edu
Received: from osage.ncsa.uiuc.edu (osage.ncsa.uiuc.edu [141.142.2.56])
	by mx1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f09IgnP13570;
	Tue, 9 Jan 2001 12:42:49 -0600 (CST)
Received: from localhost (abailey@localhost)
	by osage.ncsa.uiuc.edu (8.9.3/8.9.3) with ESMTP id MAA12311;
	Tue, 9 Jan 2001 12:42:49 -0600
Date:   Tue, 9 Jan 2001 12:42:48 -0600 (CST)
From:   Alan Bailey <abailey@ncsa.uiuc.edu>
To:     Troy Dawson <dawson@fnal.gov>
cc:     pcp@oss.sgi.com
Subject: Re: speed
In-Reply-To: <3A5B4DDE.361D0BAE@fnal.gov>
Message-ID: <Pine.LNX.4.10.10101091240120.12133-100000@osage.ncsa.uiuc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Thanks, but no thanks.  I barely know java, so it wouldn't be much help.
I will try writing a tool with multiple processes.

Later,
Alan

On Tue, 9 Jan 2001, Troy Dawson wrote:

> Hi Alan,
> I write my code in java, and have it do a remote procedure call to pminfo.  I
> currently have it doing 10 threads.  There is a thread monitor that only
> releases the 10 threads every one second.  I suppose I can make alot more
> threads now that I have alot more systems, but I haven't.
> 
> If your interested in java code I could send my prototype stuff, but not alot
> of folks are, so I didn't want to waste mailbox space if you didn't.
> 
> Troy
> 
> Alan Bailey wrote:
> > 
> > I am calling things sequentially.  I do suspect that a multithreaded
> > approach would help.  What language did you write your tools in, and could
> > I see an example?
> > 
> > Thanks a lot.
> > 
> > Alan
> > 
> -- 
> __________________________________________________
> Troy Dawson  dawson@fnal.gov  (630)840-6468
> Fermilab  ComputingDivision/OSS  SCS Group
> __________________________________________________
> 

-- 
 Alan Bailey


From owner-pcp@oss.sgi.com Tue Jan  9 15:04:37 2001
Received:  by oss.sgi.com id <S553957AbRAIXES>;
	Tue, 9 Jan 2001 15:04:18 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:53527 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553948AbRAIXEB>;
	Tue, 9 Jan 2001 15:04:01 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via SMTP id PAA17566
	for <pcp@oss.sgi.com>; Tue, 9 Jan 2001 15:03:06 -0800 (PST)
	mail_from (markgw@sgi.com)
Received: from sandpit.melbourne.sgi.com (sandpit.melbourne.sgi.com [134.14.55.132]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id KAA05122; Wed, 10 Jan 2001 10:02:36 +1100
Date:   Wed, 10 Jan 2001 10:02:36 +1100 (EST)
From:   Mark Goodwin <markgw@sgi.com>
X-Sender: markgw@sandpit.melbourne.sgi.com
To:     Alan Bailey <abailey@ncsa.uiuc.edu>
cc:     Troy Dawson <dawson@fnal.gov>, pcp@oss.sgi.com
Subject: Re: speed
In-Reply-To: <Pine.LNX.4.10.10101091240120.12133-100000@osage.ncsa.uiuc.edu>
Message-ID: <Pine.LNX.4.21.0101100950100.28572-100000@sandpit.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Tue, 9 Jan 2001, Alan Bailey wrote:

> Thanks, but no thanks.  I barely know java, so it wouldn't be much help.
> I will try writing a tool with multiple processes.

Beware, libpcp is not thread safe (static globals in assorted
important places). However, multiple threads could easily
fork/exec pminfo or pmprobe in parallel as an easy way of
implementing non-blocking (aka asynchronous) fetch. In general,
you'll get better latency with pmprobe compared to pminfo, and
the output is designed to be easier to parse.

Regarding the original question - yes we invested a lot of
effort making the PCP protocols efficient. We tuned the
client<->pmcd IPC layers with sockopts such as TCP_NODELAY,
and profiled and tuned the critical paths thru the libpcp call
stack, eliminated unnecessary system calls, etc. We also used a
a binary protocol - more compact and efficient cmp'd to ascii.

-- Mark



From owner-pcp@oss.sgi.com Tue Jan  9 18:49:48 2001
Received:  by oss.sgi.com id <S553783AbRAJCti>;
	Tue, 9 Jan 2001 18:49:38 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:9005 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553694AbRAJCtU>; Tue, 9 Jan 2001 18:49:20 -0800
Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via SMTP id SAA03657
	for <pcp@oss.sgi.com>; Tue, 9 Jan 2001 18:58:03 -0800 (PST)
	mail_from (kenmcd@melbourne.sgi.com)
From:   kenmcd@melbourne.sgi.com
Received: from [192.82.201.231] ([192.82.201.231]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id NAA06859; Wed, 10 Jan 2001 13:47:50 +1100
Date:   Wed, 10 Jan 2001 13:51:46 +1100 (EST)
Reply-To: kenmcd@melbourne.sgi.com
To:     Alan Bailey <abailey@ncsa.uiuc.edu>
cc:     pcp@oss.sgi.com
Subject: Re: speed
In-Reply-To: <Pine.LNX.4.10.10101081645370.338-100000@osage.ncsa.uiuc.edu>
Message-ID: <Pine.LNX.4.21.0101100710420.18169-100000@kenj-pc.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Mon, 8 Jan 2001, Alan Bailey wrote:

> I'm wondering what timing tests were used when developing PCP.

In the original development we (or at least I) placed huge importance
on performance and overhead.  We had a series of test cases (getting
sysinfo via PCP vs direct system call was one that was used a lot)
and the APIs and protocols were optimized up the wazoo.

Some of the things that fell out of this were:

  + minimal protocol round-trips, and in particular any selection of
    metrics and the common cases for selections of instances can be
    fetched in one round-trip, once the intial setup is complete

  + word aligned fields in messages

  + tight code optimization on the common paths

> I don't
> know if this could be answered by the open source developers, or would
> need to be redirected to the main developers.  Basically, one of my
> coworkers believes that the response time for PCP is pretty slow, and
> believes that a better solution for sending data between hosts could be
> made and would be significantly better. [1]
> 
> Right now it takes almost exactly 4 seconds to call pminfo on 87 hosts for 
> mem.freemem. ...

This is not a very good test case for many common monitoring scenarios.

All of the points below apply to a single host ... you'll just see
87 times the differential cost.

  + it takes one fork/exec pair per fetch, compared to one fork/exec for
    all fetches if one considers pmie or pmchart

  + you need to establish a new monitor-collector context for every fetch,
    compared to one for all fetches

  + you need to send a profile before every fetch, compared to sending
    the profile ahead of the first fetch

  + you need to do a PMNS lookup for every fetch, compared to one for
    all fetches

  + you need to do a descriptor retrtieval for every fetch, compared
    to one for all fetches.

In a real monitoring environment, you should expect one PDU round-trip
per fetch ... your pminfo example uses five round-trips per metric value
fetched.

To see the differences in PDU round-trips, use the in-built PCP diagnostics
(see pmdbg(1)) ... and run

	$ pminfo -Dpdu mem.freemem
	$ pmval -Dpdu mem.freemem

> ... I think this is very good.  We will be expanding to a few
> hundred hosts in the near future though.  So basically, was any optimizing
> done in regards to speed when designing PCP?  If so, is it possible to see
> any kind of tests or decisions that were made to show that PCP is truly
> optimized for speed?
> 
> Thanks,
> Alan
> 
> [1] - One of the ideas he had was to use pcp to just collect data on the
> localhost and store it in some structure.  The query mechanism between
> hosts would then ask each individual host for data, and this cached data
> would be returned.  This is obviously very similar to a caching PMDA,
> which might be an option, but I forgot to mention it to him.  He was also
> thinking of using UDP, a hierarchical grabbing scheme, and some other
> techniques for getting data.  I don't think it would help much, networks
> can only go so fast ;)

I cannot see how this will help the elapsed time.  Cacheing of this form
as been suggested to reduce the collector-side costs when many monitors
are fetching the same information, but it will not help the latency to
get the values to the remote monitor site.

I think the peak rate of fetching metrics from 87 hosts would be much
higher than one every 4 seconds if you were using pmie, for example,
rather than pminfo (based on some other tests we've done and a quick test
on my laptop, I think the lower bound on the cost for local accesses
via pmcd is about 5 msec per fetch for mem.freemem).  And in the limit
it will be a TCP/IP latency constraint that will get in the way, before
than any inefficiencies in the PCP protocols and their implementation.

Another factor to consider is that a deployment of O(100) hosts is not
something that can be controlled or changed in a short period of time
... we view these systems as having a turning circle approaching that
of the Titanic, and under these circumstances, fetch rates in the range
O(10) - O(100) seconds are more than adequate.



From owner-pcp@oss.sgi.com Wed Jan 10 08:33:24 2001
Received:  by oss.sgi.com id <S553834AbRAJQdF>;
	Wed, 10 Jan 2001 08:33:05 -0800
Received: from ex1.ncsa.uiuc.edu ([141.142.2.9]:55424 "EHLO ex1.ncsa.uiuc.edu")
	by oss.sgi.com with ESMTP id <S553831AbRAJQc7>;
	Wed, 10 Jan 2001 08:32:59 -0800
Received: from mx1.ncsa.uiuc.edu (mx1.ncsa.uiuc.edu [141.142.2.8])
	by ex1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f0AGWow01488;
	Wed, 10 Jan 2001 10:32:50 -0600 (CST)
X-Envelope-From: abailey@ncsa.uiuc.edu
Received: from osage.ncsa.uiuc.edu (osage.ncsa.uiuc.edu [141.142.2.56])
	by mx1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f0AGWnP24496;
	Wed, 10 Jan 2001 10:32:49 -0600 (CST)
Received: from localhost (abailey@localhost)
	by osage.ncsa.uiuc.edu (8.9.3/8.9.3) with ESMTP id KAA26086;
	Wed, 10 Jan 2001 10:32:49 -0600
Date:   Wed, 10 Jan 2001 10:32:49 -0600 (CST)
From:   Alan Bailey <abailey@ncsa.uiuc.edu>
To:     Mark Goodwin <markgw@sgi.com>
cc:     pcp@oss.sgi.com
Subject: Re: speed
In-Reply-To: <Pine.LNX.4.21.0101100950100.28572-100000@sandpit.melbourne.sgi.com>
Message-ID: <Pine.LNX.4.10.10101101026260.25987-100000@osage.ncsa.uiuc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Wed, 10 Jan 2001, Mark Goodwin wrote:

> On Tue, 9 Jan 2001, Alan Bailey wrote:
> 
> > Thanks, but no thanks.  I barely know java, so it wouldn't be much help.
> > I will try writing a tool with multiple processes.
> 
> Beware, libpcp is not thread safe (static globals in assorted
> important places). However, multiple threads could easily
> fork/exec pminfo or pmprobe in parallel as an easy way of
> implementing non-blocking (aka asynchronous) fetch. In general,
> you'll get better latency with pmprobe compared to pminfo, and
> the output is designed to be easier to parse.

I see.  I didn't know about pmprobe.  That is much easier to parse. :)
However, I don't like that I would have to call it twice to get both the
instance names and the values, like:

pmprobe -I kernel.all.load ; pmprobe -v kernel.all.load

Thanks for the tip, I'm just ignorant.

> 
> Regarding the original question - yes we invested a lot of
> effort making the PCP protocols efficient. We tuned the
> client<->pmcd IPC layers with sockopts such as TCP_NODELAY,
> and profiled and tuned the critical paths thru the libpcp call
> stack, eliminated unnecessary system calls, etc. We also used a
> a binary protocol - more compact and efficient cmp'd to ascii.

I figured as much.  That's awesome.  I basically wanted this information
to give to my coworker to say "yes, PCP is extremely fast, we're just
using it slowly"

Alan

> 
> -- Mark
> 
> 

-- 
 Alan Bailey


From owner-pcp@oss.sgi.com Wed Jan 10 08:47:54 2001
Received:  by oss.sgi.com id <S553834AbRAJQrf>;
	Wed, 10 Jan 2001 08:47:35 -0800
Received: from ex1.ncsa.uiuc.edu ([141.142.2.9]:48515 "EHLO ex1.ncsa.uiuc.edu")
	by oss.sgi.com with ESMTP id <S553831AbRAJQrI>;
	Wed, 10 Jan 2001 08:47:08 -0800
Received: from mx1.ncsa.uiuc.edu (mx1.ncsa.uiuc.edu [141.142.2.8])
	by ex1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f0AGl6w05220;
	Wed, 10 Jan 2001 10:47:06 -0600 (CST)
X-Envelope-From: abailey@ncsa.uiuc.edu
Received: from osage.ncsa.uiuc.edu (osage.ncsa.uiuc.edu [141.142.2.56])
	by mx1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f0AGl5P28430;
	Wed, 10 Jan 2001 10:47:05 -0600 (CST)
Received: from localhost (abailey@localhost)
	by osage.ncsa.uiuc.edu (8.9.3/8.9.3) with ESMTP id KAA26189;
	Wed, 10 Jan 2001 10:47:05 -0600
Date:   Wed, 10 Jan 2001 10:47:05 -0600 (CST)
From:   Alan Bailey <abailey@ncsa.uiuc.edu>
To:     kenmcd@melbourne.sgi.com
cc:     pcp@oss.sgi.com
Subject: Re: speed
In-Reply-To: <Pine.LNX.4.21.0101100710420.18169-100000@kenj-pc.melbourne.sgi.com>
Message-ID: <Pine.LNX.4.10.10101101032570.25987-100000@osage.ncsa.uiuc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Wed, 10 Jan 2001 kenmcd@melbourne.sgi.com wrote:

> On Mon, 8 Jan 2001, Alan Bailey wrote:
> 
> > I'm wondering what timing tests were used when developing PCP.
> 
> In the original development we (or at least I) placed huge importance
> on performance and overhead.  We had a series of test cases (getting
> sysinfo via PCP vs direct system call was one that was used a lot)
> and the APIs and protocols were optimized up the wazoo.
> 
> Some of the things that fell out of this were:
> 
>   + minimal protocol round-trips, and in particular any selection of
>     metrics and the common cases for selections of instances can be
>     fetched in one round-trip, once the intial setup is complete
> 
>   + word aligned fields in messages
> 
>   + tight code optimization on the common paths

Cool.  Like I said in the previous email, I figured that there was a lot
of time invested, I just wanted proof of it. :)

> 
> > I don't
> > know if this could be answered by the open source developers, or would
> > need to be redirected to the main developers.  Basically, one of my
> > coworkers believes that the response time for PCP is pretty slow, and
> > believes that a better solution for sending data between hosts could be
> > made and would be significantly better. [1]
> > 
> > Right now it takes almost exactly 4 seconds to call pminfo on 87 hosts for 
> > mem.freemem. ...
> 
> This is not a very good test case for many common monitoring scenarios.
> 
> All of the points below apply to a single host ... you'll just see
> 87 times the differential cost.
> 
>   + it takes one fork/exec pair per fetch, compared to one fork/exec for
>     all fetches if one considers pmie or pmchart
> 
>   + you need to establish a new monitor-collector context for every fetch,
>     compared to one for all fetches
> 
>   + you need to send a profile before every fetch, compared to sending
>     the profile ahead of the first fetch
> 
>   + you need to do a PMNS lookup for every fetch, compared to one for
>     all fetches
> 
>   + you need to do a descriptor retrtieval for every fetch, compared
>     to one for all fetches.
> 
> In a real monitoring environment, you should expect one PDU round-trip
> per fetch ... your pminfo example uses five round-trips per metric value
> fetched.

This is true, but the tool i'm creating now is a fairly configurable
program to output information on the status of a cluster.  Then someone
could say "I want to see the load on the nodes 45-135 right now" and run
the command with the correct options.  I'm also adding to the program the
ability to look at archived data, like the load on nodes 45-135 over the
last hour.  So, it can't be optimized in the way you say above, because it
creates a connection for each host whenever the program is run.

> 
> To see the differences in PDU round-trips, use the in-built PCP diagnostics
> (see pmdbg(1)) ... and run
> 
> 	$ pminfo -Dpdu mem.freemem
> 	$ pmval -Dpdu mem.freemem

> 
> > ... I think this is very good.  We will be expanding to a few
> > hundred hosts in the near future though.  So basically, was any optimizing
> > done in regards to speed when designing PCP?  If so, is it possible to see
> > any kind of tests or decisions that were made to show that PCP is truly
> > optimized for speed?
> > 
> > Thanks,
> > Alan
> > 
> > [1] - One of the ideas he had was to use pcp to just collect data on the
> > localhost and store it in some structure.  The query mechanism between
> > hosts would then ask each individual host for data, and this cached data
> > would be returned.  This is obviously very similar to a caching PMDA,
> > which might be an option, but I forgot to mention it to him.  He was also
> > thinking of using UDP, a hierarchical grabbing scheme, and some other
> > techniques for getting data.  I don't think it would help much, networks
> > can only go so fast ;)
> 
> I cannot see how this will help the elapsed time.  Cacheing of this form
> as been suggested to reduce the collector-side costs when many monitors
> are fetching the same information, but it will not help the latency to
> get the values to the remote monitor site.

I know this wouldn't help the network latency, but I thought it might help
the overall latency, since an arbitrary (possible large) amount of code
will be run on the remote host, taking time.  But perhaps it is too
insignificant.

> 
> I think the peak rate of fetching metrics from 87 hosts would be much
> higher than one every 4 seconds if you were using pmie, for example,
> rather than pminfo (based on some other tests we've done and a quick test
> on my laptop, I think the lower bound on the cost for local accesses
> via pmcd is about 5 msec per fetch for mem.freemem).  And in the limit
> it will be a TCP/IP latency constraint that will get in the way, before
> than any inefficiencies in the PCP protocols and their implementation.

It is not 1 host every 4 seconds.  It is 4 seconds to do _all_ 87 hosts.
I do think that this very adequate, but another guy said that something
better could be done.  I now realize that the improvements need to be made
to my code and his code. :)  I've already gotten it down to 1.7 seconds
for 87 hosts by querying each host with a different process.

> 
> Another factor to consider is that a deployment of O(100) hosts is not
> something that can be controlled or changed in a short period of time
> ... we view these systems as having a turning circle approaching that
> of the Titanic, and under these circumstances, fetch rates in the range
> O(10) - O(100) seconds are more than adequate.

I agree that these rates are very adequate.  Great job, PCP developers. :)

Possibly of interest, I have some stuff written that I will release in the
maybe near future.  Two pmdas and a few tools to analyze data.

> 
> 

-- 
 Alan Bailey


From owner-pcp@oss.sgi.com Wed Jan 10 12:35:44 2001
Received:  by oss.sgi.com id <S553842AbRAJUff>;
	Wed, 10 Jan 2001 12:35:35 -0800
Received: from ex1.ncsa.uiuc.edu ([141.142.2.9]:61342 "EHLO ex1.ncsa.uiuc.edu")
	by oss.sgi.com with ESMTP id <S553837AbRAJUfP>;
	Wed, 10 Jan 2001 12:35:15 -0800
Received: from mx1.ncsa.uiuc.edu (mx1.ncsa.uiuc.edu [141.142.2.8])
	by ex1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f0AKZEw29660
	for <pcp@oss.sgi.com>; Wed, 10 Jan 2001 14:35:14 -0600 (CST)
X-Envelope-From: abailey@ncsa.uiuc.edu
X-Envelope-To: <pcp@oss.sgi.com>
Received: from osage.ncsa.uiuc.edu (osage.ncsa.uiuc.edu [141.142.2.56])
	by mx1.ncsa.uiuc.edu (8.11.0/8.11.0) with ESMTP id f0AKZDP24335
	for <pcp@oss.sgi.com>; Wed, 10 Jan 2001 14:35:13 -0600 (CST)
Received: from localhost (abailey@localhost)
	by osage.ncsa.uiuc.edu (8.9.3/8.9.3) with ESMTP id OAA28637
	for <pcp@oss.sgi.com>; Wed, 10 Jan 2001 14:35:13 -0600
Date:   Wed, 10 Jan 2001 14:35:13 -0600 (CST)
From:   Alan Bailey <abailey@ncsa.uiuc.edu>
To:     pcp@oss.sgi.com
Subject: A few questions
In-Reply-To: <Pine.LNX.4.10.10101101032570.25987-100000@osage.ncsa.uiuc.edu>
Message-ID: <Pine.LNX.4.10.10101101421420.28431-100000@osage.ncsa.uiuc.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

A few simple questions...

- I'm using the following command:

> pmafm Latest pmdumplog -S "@whatever"

This is all fine and good.  But say I just want a single metric.  This
doesn't work:

> pmafm Latest pmdumplog -S "@whatever" mem.freemem
pmdumplog: Cannot open archive "mem.freemem": No such file or directory

Obviously, this is because there is no archive on the command line, and
mem.freemem is thought to be the archive.  Is there a way around this, or
is this just the way it is?


- When I run pmlogger_check to start up the pmlogger processes, there is a
considerable gap between each one starting up.  Is this on purpose so that
the pmlogger processes are staggered and don't eat up the machine at the
same time?  Or is there some other reason...


- Is there a way to round the time values in the archives to, say, every
five minutes, so that 13:42:23.045 would become 13:40?  I doubt this,
because I know the code for any arbitrary rounding time would be
nontrivial, and this isn't that necessary.

- Is pminfo the suggested way to get a single value from an archive at a
certain time, and pmdumplog the suggested way to get a range of time
values?

Yup, they all are simple questions.

Thanks,
Alan

-- 
 Alan Bailey


From owner-pcp@oss.sgi.com Sun Jan 14 15:42:42 2001
Received:  by oss.sgi.com id <S553770AbRANXmY>;
	Sun, 14 Jan 2001 15:42:24 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:63579 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553767AbRANXmB>; Sun, 14 Jan 2001 15:42:01 -0800
Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id PAA02320
	for <pcp@oss.sgi.com>; Sun, 14 Jan 2001 15:50:51 -0800 (PST)
	mail_from (max@kuku.melbourne.sgi.com)
Received: from kuku.melbourne.sgi.com (kuku.melbourne.sgi.com [134.14.55.163])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id PAA96988;
	Sun, 14 Jan 2001 15:40:43 -0800 (PST)
Received: (from max@localhost)
	by kuku.melbourne.sgi.com (SGI-8.9.3/8.9.3) id KAA41678;
	Mon, 15 Jan 2001 10:40:42 +1100 (EST)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <14946.14585.839228.985075@kuku.melbourne.sgi.com>
Date:   Mon, 15 Jan 2001 10:40:41 +1100 (EST)
From:   Max Matveev <makc@sgi.com>
To:     Alan Bailey <abailey@ncsa.uiuc.edu>
Cc:     pcp@oss.sgi.com
Subject: Re: A few questions
In-Reply-To: <Pine.LNX.4.10.10101101421420.28431-100000@osage.ncsa.uiuc.edu>
References: <Pine.LNX.4.10.10101101032570.25987-100000@osage.ncsa.uiuc.edu>
	<Pine.LNX.4.10.10101101421420.28431-100000@osage.ncsa.uiuc.edu>
X-Mailer: VM 6.72 under 21.1 (patch 10) "Capitol Reef" XEmacs Lucid
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

>>>>> "AB" == Alan Bailey <abailey@ncsa.uiuc.edu> writes:

AB> A few simple questions...
AB> - I'm using the following command:

AB> > pmafm Latest pmdumplog -S "@whatever"

AB> This is all fine and good.  But say I just want a single metric.  This
AB> doesn't work:

AB> > pmafm Latest pmdumplog -S "@whatever" mem.freemem
AB> pmdumplog: Cannot open archive "mem.freemem": No such file or directory

AB> Obviously, this is because there is no archive on the command line, and
AB> mem.freemem is thought to be the archive.  Is there a way around this, or
AB> is this just the way it is?

That's because pmdumlog is not the best tool for this job. Partially
becuase it's command line options are not the same as "the rest", i.e
it doesn't use -a to specify archives. Try pmval -S "@whatever" with
'repeat'

i.e

pmafm Latest repeat pmval -S "@sometime" mem.freemem

AB> - Is there a way to round the time values in the archives to, say, every
AB> five minutes, so that 13:42:23.045 would become 13:40?  I doubt this,
AB> because I know the code for any arbitrary rounding time would be
AB> nontrivial, and this isn't that necessary.

You're right here. You can achieve similar effect with some extra -S
and -t options to be passed to your pmlogger. It still wouldn't
guaranty exact round time, but it will be close.

AB> - Is pminfo the suggested way to get a single value from an archive at a
AB> certain time, and pmdumplog the suggested way to get a range of time
AB> values?

Pretty much so. You can use pmval but it's more heavy-weight compared
to pminfo and pmprobe doesn't do times, it only does one metric at the
top of the archive.

max

From owner-pcp@oss.sgi.com Tue Jan 16 09:43:45 2001
Received:  by oss.sgi.com id <S553878AbRAPRng>;
	Tue, 16 Jan 2001 09:43:36 -0800
Received: from london.emaginet.com ([63.65.80.130]:59403 "HELO
        london.emaginet.com") by oss.sgi.com with SMTP id <S553876AbRAPRnX>;
	Tue, 16 Jan 2001 09:43:23 -0800
Received: (qmail 16163 invoked from network); 16 Jan 2001 17:35:42 -0000
Received: from unknown (HELO emaginet.com) (172.16.0.26)
  by 172.16.0.1 with SMTP; 16 Jan 2001 17:35:42 -0000
Received: from ecentives.com ([172.16.0.111])
	by emaginet.com (8.9.3/8.9.0) with ESMTP id MAA14163
	for <pcp@oss.sgi.com>; Tue, 16 Jan 2001 12:40:40 -0500
Message-ID: <3A649821.D7A2721F@ecentives.com>
Date:   Tue, 16 Jan 2001 12:51:13 -0600
From:   "Robert K. Hamilton" <rkh@ecentives.com>
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To:     pcp@oss.sgi.com
Subject: Install problems
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Howdy,

I built the pcp source on the following configuration:

H/W: Intel x86
Linux: Red Hat 6.2
PCP: 2.1.10 build 8

When I issue the pminfo command as specified in INSTALL I receive the
following:

[root@tacan pcp-2.1.10]# pminfo -fmdt
PMNS appears to be empty!

When I issue any variation of the pcp command I get the following:

[root@tacan pcp-2.1.10]# pcp
Performance Co-Pilot configuration on tacan.bethesda.emaginet.com:

 platform:
 hardware:
 timezone: Unknown
 licenses: unknown
     pmcd: Version unknown

The following processes are running on the system:

32675 ?        S      0:00 /usr/share/pcp/bin/pmcd
  349 pts/2    S      0:00 pmlogger -P -c config.default 20010116.13.02
  835 pts/2    S      0:00 grep pm

It looks like everything installed okay but I'm not collecting any data.
When I try to use the graphical interface via pcpmon (V1.2.0) I get
can't fetch errors ... 

What's missing?

Thanks!

-B
-- 
Robert K. Hamilton
Senior Systems Engineer
Manager, Test Networks
E-centives, Inc.
6901 Rockledge Drive, 7th Floor
Bethesda, MD 20817
240.333.6178
rkh@e-centives.com
www.e-centives.com/corp
--
e-centives, Inc.
The leading online direct marketing 
infrastructure company

From owner-pcp@oss.sgi.com Tue Jan 16 11:24:28 2001
Received:  by oss.sgi.com id <S553923AbRAPTYS>;
	Tue, 16 Jan 2001 11:24:18 -0800
Received: from london.emaginet.com ([63.65.80.130]:36872 "HELO
        london.emaginet.com") by oss.sgi.com with SMTP id <S553915AbRAPTYK>;
	Tue, 16 Jan 2001 11:24:10 -0800
Received: (qmail 19250 invoked from network); 16 Jan 2001 19:16:25 -0000
Received: from unknown (HELO emaginet.com) (172.16.0.26)
  by 172.16.0.1 with SMTP; 16 Jan 2001 19:16:25 -0000
Received: from ecentives.com ([172.16.0.111])
	by emaginet.com (8.9.3/8.9.0) with ESMTP id OAA18308
	for <pcp@oss.sgi.com>; Tue, 16 Jan 2001 14:21:23 -0500
Message-ID: <3A64AFBC.F7AD77D4@ecentives.com>
Date:   Tue, 16 Jan 2001 14:31:56 -0600
From:   "Robert K. Hamilton" <rkh@ecentives.com>
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To:     pcp@oss.sgi.com
Subject: Re: Install problems
References: <3A649821.D7A2721F@ecentives.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing



I've actually made progress myself. It looks like the problem is name
space related. I found the location of the root file for the default
name space and checked the file. /var/pcp/pmns/root. I saved the initial
root file and copied a file named linux_root to root and started getting
data.

Anything else related to this that might be interesting?

-B

"Robert K. Hamilton" wrote:
> 
> Howdy,
> 
> I built the pcp source on the following configuration:
> 
> H/W: Intel x86
> Linux: Red Hat 6.2
> PCP: 2.1.10 build 8
> 
> When I issue the pminfo command as specified in INSTALL I receive the
> following:
> 
> [root@tacan pcp-2.1.10]# pminfo -fmdt
> PMNS appears to be empty!
> 
> When I issue any variation of the pcp command I get the following:
> 
> [root@tacan pcp-2.1.10]# pcp
> Performance Co-Pilot configuration on tacan.bethesda.emaginet.com:
> 
>  platform:
>  hardware:
>  timezone: Unknown
>  licenses: unknown
>      pmcd: Version unknown
> 
> The following processes are running on the system:
> 
> 32675 ?        S      0:00 /usr/share/pcp/bin/pmcd
>   349 pts/2    S      0:00 pmlogger -P -c config.default 20010116.13.02
>   835 pts/2    S      0:00 grep pm
> 
> It looks like everything installed okay but I'm not collecting any data.
> When I try to use the graphical interface via pcpmon (V1.2.0) I get
> can't fetch errors ...
> 
> What's missing?
> 
> Thanks!
> 
> -B
> --
> Robert K. Hamilton
> Senior Systems Engineer
> Manager, Test Networks
> E-centives, Inc.
> 6901 Rockledge Drive, 7th Floor
> Bethesda, MD 20817
> 240.333.6178
> rkh@e-centives.com
> www.e-centives.com/corp
> --
> e-centives, Inc.
> The leading online direct marketing
> infrastructure company

-- 
Robert K. Hamilton
Senior Systems Engineer
Manager, Test Networks
E-centives, Inc.
6901 Rockledge Drive, 7th Floor
Bethesda, MD 20817
240.333.6178
rkh@e-centives.com
www.e-centives.com/corp
--
e-centives, Inc.
The leading online direct marketing 
infrastructure company

From owner-pcp@oss.sgi.com Tue Jan 16 12:12:49 2001
Received:  by oss.sgi.com id <S553955AbRAPUMi>;
	Tue, 16 Jan 2001 12:12:38 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:2888 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553948AbRAPUMb>;
	Tue, 16 Jan 2001 12:12:31 -0800
Received: from rattle.melbourne.sgi.com ([134.14.55.145]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via ESMTP id MAA02021
	for <pcp@oss.sgi.com>; Tue, 16 Jan 2001 12:12:29 -0800 (PST)
	mail_from (kenmcd@melbourne.sgi.com)
Received: from localhost (kenmcd@localhost)
	by rattle.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id HAA42733;
	Wed, 17 Jan 2001 07:11:02 +1100 (AEDT)
X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs
Date:   Wed, 17 Jan 2001 07:11:01 +1100
From:   Ken McDonell <kenmcd@melbourne.sgi.com>
Reply-To: kenmcd@sgi.com
To:     "Robert K. Hamilton" <rkh@ecentives.com>
cc:     pcp@oss.sgi.com
Subject: Re: Install problems
In-Reply-To: <3A64AFBC.F7AD77D4@ecentives.com>
Message-ID: <Pine.SGI.4.21.0101170701580.5085203-100000@rattle.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Tue, 16 Jan 2001, Robert K. Hamilton wrote:

> 
> 
> I've actually made progress myself. It looks like the problem is name
> space related. I found the location of the root file for the default
> name space and checked the file. /var/pcp/pmns/root. I saved the initial
> root file and copied a file named linux_root to root and started getting
> data.

Indeed ... that's what the PMNS appears to be empty! message means.

When you install RPMS, the following empty file is created
	/var/pcp/pmns/.NeedRebuild

then when you run
	/etc/rc.d/init.d/pcp start

The PNMS (Performance Metrics Name Space) is rebuild using
	cd /var/pcp/pmns
	./Rebuild -du

and you should see a message of the form
	Performance Co-Pilot rebuilding PMNS ...

Here's the interactive sequence to mimic this ...

[root@sherman kenmcd]# touch /var/pcp/pmns/.NeedRebuild
[root@sherman kenmcd]# /etc/rc.d/init.d/pcp start
Waiting for PMCD to terminate ...
Performance Co-Pilot rebuilding PMNS ...
Performance Co-Pilot starting PMCD (logfile is /var/log/pcp/pmcd/pmcd.log) ...
Performance Co-Pilot starting archive loggers ...


My guess would be that in making from source, some part of your
installation missed touching /var/pcp/pmns/.NeedRebuild (since it
is part of the RPM packaging, this is not surprising if you did not
re-make and install RPMs).

> Anything else related to this that might be interesting?

Not that I can think of.

Even though you have it working, I'd recommend doing the

	# touch /var/pcp/pmns/.NeedRebuild
	# /etc/rc.d/init.d/pcp start

to get a kosher PMNS.

> -B
> 
> "Robert K. Hamilton" wrote:
> > 
> > Howdy,
> > 
> > I built the pcp source on the following configuration:
> > 
> > H/W: Intel x86
> > Linux: Red Hat 6.2
> > PCP: 2.1.10 build 8
> > 
> > When I issue the pminfo command as specified in INSTALL I receive the
> > following:
> > 
> > [root@tacan pcp-2.1.10]# pminfo -fmdt
> > PMNS appears to be empty!
> > 
> > When I issue any variation of the pcp command I get the following:
> > 
> > [root@tacan pcp-2.1.10]# pcp
> > Performance Co-Pilot configuration on tacan.bethesda.emaginet.com:
> > 
> >  platform:
> >  hardware:
> >  timezone: Unknown
> >  licenses: unknown
> >      pmcd: Version unknown
> > 
> > The following processes are running on the system:
> > 
> > 32675 ?        S      0:00 /usr/share/pcp/bin/pmcd
> >   349 pts/2    S      0:00 pmlogger -P -c config.default 20010116.13.02
> >   835 pts/2    S      0:00 grep pm
> > 
> > It looks like everything installed okay but I'm not collecting any data.
> > When I try to use the graphical interface via pcpmon (V1.2.0) I get
> > can't fetch errors ...
> > 
> > What's missing?
> > 
> > Thanks!
> > 
> > -B
> > --
> > Robert K. Hamilton
> > Senior Systems Engineer
> > Manager, Test Networks
> > E-centives, Inc.
> > 6901 Rockledge Drive, 7th Floor
> > Bethesda, MD 20817
> > 240.333.6178
> > rkh@e-centives.com
> > www.e-centives.com/corp
> > --
> > e-centives, Inc.
> > The leading online direct marketing
> > infrastructure company
> 
> -- 
> Robert K. Hamilton
> Senior Systems Engineer
> Manager, Test Networks
> E-centives, Inc.
> 6901 Rockledge Drive, 7th Floor
> Bethesda, MD 20817
> 240.333.6178
> rkh@e-centives.com
> www.e-centives.com/corp
> --
> e-centives, Inc.
> The leading online direct marketing 
> infrastructure company
> 


From owner-pcp@oss.sgi.com Tue Jan 16 12:40:58 2001
Received:  by oss.sgi.com id <S553964AbRAPUku>;
	Tue, 16 Jan 2001 12:40:50 -0800
Received: from london.emaginet.com ([63.65.80.130]:63506 "HELO
        london.emaginet.com") by oss.sgi.com with SMTP id <S553961AbRAPUke>;
	Tue, 16 Jan 2001 12:40:34 -0800
Received: (qmail 22166 invoked from network); 16 Jan 2001 20:32:51 -0000
Received: from unknown (HELO emaginet.com) (172.16.0.26)
  by 172.16.0.1 with SMTP; 16 Jan 2001 20:32:51 -0000
Received: from ecentives.com ([172.16.0.111])
	by emaginet.com (8.9.3/8.9.0) with ESMTP id PAA21541;
	Tue, 16 Jan 2001 15:37:49 -0500
Message-ID: <3A64C1A7.6D2AE46@ecentives.com>
Date:   Tue, 16 Jan 2001 15:48:23 -0600
From:   "Robert K. Hamilton" <rkh@ecentives.com>
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To:     kenmcd@sgi.com
CC:     pcp@oss.sgi.com
Subject: Re: Install problems
References: <Pine.SGI.4.21.0101170701580.5085203-100000@rattle.melbourne.sgi.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Ken,

This did the trick. I put the empty root file back and performed the
steps you outlined below. The root file is now populated. 

I'm using PCPMON 1.2.0 and I did notice a lack of predefined expressions
to use. Is there a place where I could find a broader set of predefined
expressions for linux? If not, where is the best place to start reading
up on how to build them (i.e. comprehensive list of metrics etc).

Thanks a bunch.

-B

Ken McDonell wrote:
> 
> On Tue, 16 Jan 2001, Robert K. Hamilton wrote:
> 
> >
> >
> > I've actually made progress myself. It looks like the problem is name
> > space related. I found the location of the root file for the default
> > name space and checked the file. /var/pcp/pmns/root. I saved the initial
> > root file and copied a file named linux_root to root and started getting
> > data.
> 
> Indeed ... that's what the PMNS appears to be empty! message means.
> 
> When you install RPMS, the following empty file is created
>         /var/pcp/pmns/.NeedRebuild
> 
> then when you run
>         /etc/rc.d/init.d/pcp start
> 
> The PNMS (Performance Metrics Name Space) is rebuild using
>         cd /var/pcp/pmns
>         ./Rebuild -du
> 
> and you should see a message of the form
>         Performance Co-Pilot rebuilding PMNS ...
> 
> Here's the interactive sequence to mimic this ...
> 
> [root@sherman kenmcd]# touch /var/pcp/pmns/.NeedRebuild
> [root@sherman kenmcd]# /etc/rc.d/init.d/pcp start
> Waiting for PMCD to terminate ...
> Performance Co-Pilot rebuilding PMNS ...
> Performance Co-Pilot starting PMCD (logfile is /var/log/pcp/pmcd/pmcd.log) ...
> Performance Co-Pilot starting archive loggers ...
> 
> My guess would be that in making from source, some part of your
> installation missed touching /var/pcp/pmns/.NeedRebuild (since it
> is part of the RPM packaging, this is not surprising if you did not
> re-make and install RPMs).
> 
> > Anything else related to this that might be interesting?
> 
> Not that I can think of.
> 
> Even though you have it working, I'd recommend doing the
> 
>         # touch /var/pcp/pmns/.NeedRebuild
>         # /etc/rc.d/init.d/pcp start
> 
> to get a kosher PMNS.
> 
> > -B
> >
> > "Robert K. Hamilton" wrote:
> > >
> > > Howdy,
> > >
> > > I built the pcp source on the following configuration:
> > >
> > > H/W: Intel x86
> > > Linux: Red Hat 6.2
> > > PCP: 2.1.10 build 8
> > >
> > > When I issue the pminfo command as specified in INSTALL I receive the
> > > following:
> > >
> > > [root@tacan pcp-2.1.10]# pminfo -fmdt
> > > PMNS appears to be empty!
> > >
> > > When I issue any variation of the pcp command I get the following:
> > >
> > > [root@tacan pcp-2.1.10]# pcp
> > > Performance Co-Pilot configuration on tacan.bethesda.emaginet.com:
> > >
> > >  platform:
> > >  hardware:
> > >  timezone: Unknown
> > >  licenses: unknown
> > >      pmcd: Version unknown
> > >
> > > The following processes are running on the system:
> > >
> > > 32675 ?        S      0:00 /usr/share/pcp/bin/pmcd
> > >   349 pts/2    S      0:00 pmlogger -P -c config.default 20010116.13.02
> > >   835 pts/2    S      0:00 grep pm
> > >
> > > It looks like everything installed okay but I'm not collecting any data.
> > > When I try to use the graphical interface via pcpmon (V1.2.0) I get
> > > can't fetch errors ...
> > >
> > > What's missing?
> > >
> > > Thanks!
> > >
> > > -B
> > > --
> > > Robert K. Hamilton
> > > Senior Systems Engineer
> > > Manager, Test Networks
> > > E-centives, Inc.
> > > 6901 Rockledge Drive, 7th Floor
> > > Bethesda, MD 20817
> > > 240.333.6178
> > > rkh@e-centives.com
> > > www.e-centives.com/corp
> > > --
> > > e-centives, Inc.
> > > The leading online direct marketing
> > > infrastructure company
> >
> > --
> > Robert K. Hamilton
> > Senior Systems Engineer
> > Manager, Test Networks
> > E-centives, Inc.
> > 6901 Rockledge Drive, 7th Floor
> > Bethesda, MD 20817
> > 240.333.6178
> > rkh@e-centives.com
> > www.e-centives.com/corp
> > --
> > e-centives, Inc.
> > The leading online direct marketing
> > infrastructure company
> >

-- 
Robert K. Hamilton
Senior Systems Engineer
Manager, Test Networks
E-centives, Inc.
6901 Rockledge Drive, 7th Floor
Bethesda, MD 20817
240.333.6178
rkh@e-centives.com
www.e-centives.com/corp
--
e-centives, Inc.
The leading online direct marketing 
infrastructure company

From owner-pcp@oss.sgi.com Sun Jan 21 22:38:19 2001
Received:  by oss.sgi.com id <S553841AbRAVGiJ>;
	Sun, 21 Jan 2001 22:38:09 -0800
Received: from pneumatic-tube.sgi.com ([204.94.214.22]:2881 "EHLO
        pneumatic-tube.sgi.com") by oss.sgi.com with ESMTP
	id <S553839AbRAVGh7>; Sun, 21 Jan 2001 22:37:59 -0800
Received: from cthulhu.engr.sgi.com (gate3-relay.engr.sgi.com [130.62.1.234]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id WAA05560
	for <pcp@oss.sgi.com>; Sun, 21 Jan 2001 22:46:55 -0800 (PST)
	mail_from (max@kuku.melbourne.sgi.com)
Received: from kuku.melbourne.sgi.com (kuku.melbourne.sgi.com [134.14.55.163])
	by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id WAA84984;
	Sun, 21 Jan 2001 22:36:30 -0800 (PST)
Received: (from max@localhost)
	by kuku.melbourne.sgi.com (SGI-8.9.3/8.9.3) id RAA15734;
	Mon, 22 Jan 2001 17:37:20 +1100 (EST)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <14955.54559.901665.43288@kuku.melbourne.sgi.com>
Date:   Mon, 22 Jan 2001 17:37:19 +1100 (EST)
From:   Max Matveev <makc@sgi.com>
To:     "Robert K. Hamilton" <rkh@ecentives.com>
Cc:     pcp@oss.sgi.com
Subject: pcpmon views (Re: Install problems)
In-Reply-To: <3A64C1A7.6D2AE46@ecentives.com>
References: <Pine.SGI.4.21.0101170701580.5085203-100000@rattle.melbourne.sgi.com>
	<3A64C1A7.6D2AE46@ecentives.com>
X-Mailer: VM 6.72 under 21.1 (patch 10) "Capitol Reef" XEmacs Lucid
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing


>>>>> "RH" == Robert K Hamilton <rkh@ecentives.com> writes:

RH> I'm using PCPMON 1.2.0 and I did notice a lack of predefined expressions
RH> to use. Is there a place where I could find a broader set of predefined
RH> expressions for linux? If not, where is the best place to start reading
RH> up on how to build them (i.e. comprehensive list of metrics etc).

You can start with 'pminfo' by itself.

$ pminfo

should dump all the metrics which are available on your local host
(assuming pmcd is running). If memory serves, pcpmon only accept the
metric names at the moment so it should be sufficient to start
crafting pcpmon config files.

I don't know if Michal has some plans to provide "canned" views to
solve the some common problems, but it's definitely worth some
investigation from whoever needs them. Perhaps a repository for pcpmon
configs on oss.sgi.com could be a useful addition. Mark?

max


From owner-pcp@oss.sgi.com Tue Jan 23 12:15:51 2001
Received:  by oss.sgi.com id <S553685AbRAWUPc>;
	Tue, 23 Jan 2001 12:15:32 -0800
Received: from beret.stetson.edu ([147.253.10.47]:65260 "EHLO
        beret.stetson.edu") by oss.sgi.com with ESMTP id <S553682AbRAWUPS>;
	Tue, 23 Jan 2001 12:15:18 -0800
Received: from monky.stetson.edu (monky.stetson.edu [147.253.10.142])
	by beret.stetson.edu (8.11.0/8.11.0) with ESMTP id f0NKEH606677
	for <pcp@oss.sgi.com>; Tue, 23 Jan 2001 15:14:17 -0500 (EST)
Received: from paland by monky.stetson.edu with local (Exim 3.12 #1 (Debian))
	id 14LAmR-0005eg-00
	for <pcp@oss.sgi.com>; Tue, 23 Jan 2001 16:15:07 -0500
Date:   Tue, 23 Jan 2001 16:15:07 -0500
From:   Patrick Aland <paland@stetson.edu>
To:     pcp@oss.sgi.com
Subject: Output format
Message-ID: <20010123161504.K16498@stetson.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Which program will take the pcp data and output it in a certain format?
Mainly I am looking to use our current system which take logs of the form

[column1_name]	[column2_name]	...	[columnx_name]
[column1_data]	[column2_data]	...	[columnx_data]
[column1_data]	[column2_data]	...	[columnx_data]
[column1_data]	[column2_data]	...	[columnx_data]
[column1_data]	[column2_data]	...	[columnx_data]

I think i could use pmie to do this but am not sure. Any help is appreciated. Thanks. 

--Patrick

From owner-pcp@oss.sgi.com Tue Jan 23 15:39:31 2001
Received:  by oss.sgi.com id <S553720AbRAWXjL>;
	Tue, 23 Jan 2001 15:39:11 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:39432 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553708AbRAWXjD>;
	Tue, 23 Jan 2001 15:39:03 -0800
Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id PAA19623
	for <pcp@oss.sgi.com>; Tue, 23 Jan 2001 15:38:03 -0800 (PST)
	mail_from (kenmcd@melbourne.sgi.com)
Received: from localhost (kenmcd@localhost)
	by rattle.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id KAA14347;
	Wed, 24 Jan 2001 10:37:32 +1100 (AEDT)
X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs
Date:   Wed, 24 Jan 2001 10:37:32 +1100
From:   Ken McDonell <kenmcd@melbourne.sgi.com>
Reply-To: kenmcd@sgi.com
To:     Patrick Aland <paland@stetson.edu>
cc:     pcp@oss.sgi.com
Subject: Re: Output format
In-Reply-To: <20010123161504.K16498@stetson.edu>
Message-ID: <Pine.SGI.4.21.0101240849460.408516-200000@rattle.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: MULTIPART/MIXED; BOUNDARY="-2045888623-852029301-980293052=:408516"
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

---2045888623-852029301-980293052=:408516
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Tue, 23 Jan 2001, Patrick Aland wrote:

> Which program will take the pcp data and output it in a certain format?
> Mainly I am looking to use our current system which take logs of the form
> 
> [column1_name]	[column2_name]	...	[columnx_name]
> [column1_data]	[column2_data]	...	[columnx_data]
> [column1_data]	[column2_data]	...	[columnx_data]
> [column1_data]	[column2_data]	...	[columnx_data]
> [column1_data]	[column2_data]	...	[columnx_data]
> 
> I think i could use pmie to do this but am not sure. Any help is appreciated. Thanks. 

depends a little ... can you expand on what columnx_name might be?

If these are instance of a single metric, pmval will do this
out of the box.

If these are different metrics (with or without instances), then
if you have the IRIX or pcp-pro pieces, pmdumptext will do this,
otherwise something like ...

cat <<End-of-File \
| pmie \
| some-awk
column1_name=some expression;
column2_name=some other expression;
column3_name=a third expression;
...
End-of-File

To be more concrete, here's a pmie script ...

load_1 = kernel.all.load #'1 minute';
idle = kernel.all.cpu.idle;
EOL = hinv.ncpu;

The EOL line is a sentinel and special in what follows, so there
are 2 metrics being reported here.

Running pmie -v produces ...

load_1: 3.70
idle: ?
EOL: 12

load_1: 3.55
idle: 8.3
EOL: 12

load_1: 3.20
idle: 9.9
EOL: 12

load_1: 3.04
idle: 9.4
EOL: 12

load_1: 3.22
idle: 8.6
EOL: 12

Now feed this to pmie2col (a quick and dirty awk script, copy attached)
and you get ...

 load_1:   idle:
    3.70       ?
    3.55    8.30
    3.20    9.90
    3.04    9.40
    3.22    8.60


---2045888623-852029301-980293052=:408516
Content-Type: TEXT/PLAIN; charset=US-ASCII; name=pmie2col
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.SGI.4.21.0101241037310.408516@rattle.melbourne.sgi.com>
Content-Description: pmie2col
Content-Disposition: attachment; filename=pmie2col
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---2045888623-852029301-980293052=:408516--

From owner-pcp@oss.sgi.com Wed Jan 24 00:07:25 2001
Received:  by oss.sgi.com id <S553786AbRAXIHP>;
	Wed, 24 Jan 2001 00:07:15 -0800
Received: from [209.209.13.26] ([209.209.13.26]:26890 "EHLO smile.idiom.com")
	by oss.sgi.com with ESMTP id <S553778AbRAXIG4>;
	Wed, 24 Jan 2001 00:06:56 -0800
Received: from mdaxke (adsl-64-169-94-77.dsl.snfc21.pacbell.net [64.169.94.77])
	by smile.idiom.com (8.9.1/8.8.5) with SMTP id AAA18607
	for <pcp@oss.sgi.com>; Wed, 24 Jan 2001 00:06:55 -0800 (PST)
Message-ID: <00e001c085dc$73603c80$0201a8c0@wwc.com>
From:   "Mark D. Anderson" <mda@discerning.com>
To:     <pcp@oss.sgi.com>
Subject: introductory pcp questions
Date:   Wed, 24 Jan 2001 00:05:13 -0800
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4522.1200
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

i just ran into pcp while looking at another project on the oss.sgi page....

i've looked through the web site and through the downloads for some kind
of overview, and couldn't find any. i looked through the powerpoint slides
by ken mcdonnel, and i've read "man PCPIntro" and "man PMAPI".
somewhere between the trees and the earth should be a forest....

the man pages mention a "Programmer's Guide" and a "Tutorial", but
perhaps those are only available as part of the commercial product?

so far as i can make out, the basic architecture consists of:
- console and GUI monitor clients that can subscribe to a real-time feed or an archive feed
- a per-host pmcd daemon that brokers between the clients and agents
- pmda agent daemons that are per-host and per-namespace
- a per-host pmlogger daemons that archive data locally, from pmcd to disk
but i'm not even sure of that much.

here are some of the basic parameters i would have hoped to have found
answers to and didn't, and what i've been able to divine so far:

- license.
http://oss.sgi.com/projects/pcp/license.html indicates this is (mostly) LGPL, while
the downloads indicate it is GPL (even the libraries). This is not a trivial distinction.
- language support.
it appears that agents and clients have to be written in C.
- security
i see no provision for client or server authentication, provision for encryption, integrity checks, etc.
- clocks
i see no provision for clock synchronization.
- query versus notify
i can find no protocol definition, so i can't tell whether monitors must query for particular data,
or whether they can subscribe to asynchronous notifications.
- sampled vs. events
said another way, can a monitor ask for qualitative events (threshold passing), instead of regularly sampled snapshots?
- connection vs. connectionless
i can find no protocol definition, so i can't tell whether it is stateful or not, let alone
what provisions it has for resumption after a connection loss.
nor can i tell whether it is message-per-row, message-per-request, or what.
nor whether the protocol allows pipelining, or multiple asynchronous requests, etc.
nor whether it is the same protocol between monitor, pmcd, pmlogger, and pmda.
- agent-side computation
obviously a monitor can compute anything it likes. 
but can a monitor request that a agent do some server-side computation before sending
the resulting data back, either across measurements (say, changing units or adding together),
or across time (running average, etc.).
- agent-side filtering
similarly, what kinds of filters can a monitor request?
- fast localhost monitoring
is there a shared memory or similar mechanism for monitoring an application's "counters"
without the overhead of tcp/ip communication?
- triggers
i see no indication of what external integrations have been done for actions to be taken
based on various events (paging, email, etc.)
- agent collapsing of requests
if 10 monitors ask agent for the same regularly sampled data, does it measure it 10 times, or just once?
if i ask for both user and system time, will it be smart enough to do this in one operation, not two?
- discovery
can an agent automatically discover a monitor? can a monitor automatically discover an agent?
- metadata
i found some discussion of a name to oid mapping by pmns, but no definition of how a monitor
queries the schema or instances available from a particular agent, nor if there is a way to
get notified of schema evolution, or instance addition/deletion.
nor can i find an explanation of what can be declared about a name besides its type.
- naming
can a value be given a universally unique path identifier such as "host=bar;process=89;thread=98;request_count"?
i suppose this would be part of that protocol definition....
- strict definition of measured values
concepts such as "free memory" and "pages of io" are essentially meaningless without
very strict definitions of what is counted and how. The specifics vary from OS to OS, and
sometimes from OS version to version. It appears there has been no effort in pcp
to alleviated this chaos; the linux pmda just passes on the /proc information in all
its ill-defined glory.
- standards compliance
pcp seems to use no standards, even derivatively, for anything:
transport-level protocol, data and log file formats, metadata representation, 
names of particular measured values, the client or agent apis, or the name to number resolution.

that's all for now :).

thanks...

-mda




From owner-pcp@oss.sgi.com Wed Jan 24 08:41:06 2001
Received:  by oss.sgi.com id <S553806AbRAXQkq>;
	Wed, 24 Jan 2001 08:40:46 -0800
Received: from beret.stetson.edu ([147.253.10.47]:8369 "EHLO beret.stetson.edu")
	by oss.sgi.com with ESMTP id <S553801AbRAXQkd>;
	Wed, 24 Jan 2001 08:40:33 -0800
Received: from monky.stetson.edu (monky.stetson.edu [147.253.10.142])
	by beret.stetson.edu (8.11.0/8.11.0) with ESMTP id f0OGdI610916
	for <pcp@oss.sgi.com>; Wed, 24 Jan 2001 11:39:18 -0500 (EST)
Received: from paland by monky.stetson.edu with local (Exim 3.12 #1 (Debian))
	id 14LTtw-0003Ox-00
	for <pcp@oss.sgi.com>; Wed, 24 Jan 2001 12:40:08 -0500
Date:   Wed, 24 Jan 2001 12:40:08 -0500
From:   Patrick Aland <paland@stetson.edu>
To:     pcp@oss.sgi.com
Subject: No disk.partition information
Message-ID: <20010124124007.O16498@stetson.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5i
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

Is there something special that has to be done to be able to view the disk.all.read_bytes and write_bytes (and disk.partition.read_bytes) doing a:

"pminfo -f disk.all" returns
disk.all.read
    value 1613136

disk.all.write
    value 2094781

disk.all.total
    value 3707917

disk.all.blkread
    value 12905022

disk.all.blkwrite
    value 16758248

disk.all.blktotal
    value 29663270

disk.all.read_bytes
No value(s) available!

disk.all.write_bytes
No value(s) available!

disk.all.total_bytes
No value(s) available!


(and similar for the .partition.*_bytes)

Thanks.

--Patrick

From owner-pcp@oss.sgi.com Wed Jan 24 23:10:21 2001
Received:  by oss.sgi.com id <S553792AbRAYHKL>;
	Wed, 24 Jan 2001 23:10:11 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:41806 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S553645AbRAYHJy>;
	Wed, 24 Jan 2001 23:09:54 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id XAA03120
	for <pcp@oss.sgi.com>; Wed, 24 Jan 2001 23:09:52 -0800 (PST)
	mail_from (markgw@sgi.com)
Received: from sandpit.melbourne.sgi.com (sandpit.melbourne.sgi.com [134.14.55.132]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id SAA13797; Thu, 25 Jan 2001 18:08:19 +1100
Date:   Thu, 25 Jan 2001 18:08:18 +1100 (EST)
From:   Mark Goodwin <markgw@sgi.com>
X-Sender: markgw@sandpit.melbourne.sgi.com
To:     Patrick Aland <paland@stetson.edu>
cc:     pcp@oss.sgi.com
Subject: Re: No disk.partition information
In-Reply-To: <20010124124007.O16498@stetson.edu>
Message-ID: <Pine.LNX.4.21.0101251802410.2717-100000@sandpit.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Wed, 24 Jan 2001, Patrick Aland wrote:
> Is there something special that has to be done to be able to view the
> disk.all.read_bytes and write_bytes (and disk.partition.read_bytes)

The per-partition and {read,write}_bytes metrics are require an
extended version of /proc/partitions. The kernel patch to extend
/proc/partitions with the necessary fields is called the "sard patch",
from Stephen Tweedie. You can get the patch for various kernel versions
from sct's site at ftp://ftp.uk.linux.org:/pub/linux/sct/fs/profiling/

thanks

-- Mark Goodwin
SGI Engineering



From owner-pcp@oss.sgi.com Thu Jan 25 06:42:44 2001
Received:  by oss.sgi.com id <S553895AbRAYOm0>;
	Thu, 25 Jan 2001 06:42:26 -0800
Received: from UMS.MSFC.NASA.GOV ([128.158.96.49]:36625 "EHLO
        ums.msfc.nasa.gov") by oss.sgi.com with ESMTP id <S553846AbRAYOmK>;
	Thu, 25 Jan 2001 06:42:10 -0800
Received: by UMS.MSFC.NASA.GOV with Internet Mail Service (5.5.2650.21)
	id <DPQ0TWK5>; Thu, 25 Jan 2001 08:40:57 -0600
Message-ID: <FCF379336618D3119BDA00C00D00924602E7D17E@UMS.MSFC.NASA.GOV>
From:   "Junkin, Bobby" <Bobby.Junkin@UMS.MSFC.NASA.GOV>
To:     pcp@oss.sgi.com
Subject: Archive processing and reporting
Date:   Thu, 25 Jan 2001 08:40:52 -0600
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C086DC.D3FB0286"
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C086DC.D3FB0286
Content-Type: text/plain

I'm currently using PCP 2.1 in an IRIX 6.5.9 environment.
According to the pmlogger_daily man page, toward the bottom, it indicates
"The script /usr/pcp/bin/pmlogger_daily could be copied and modified to
implement a site-specific procedure for end-of-week and/or end-of-month
management for a set of PCP archives." 
My question is, what specifically "could be modified" in this script to
accomplish an end-of-week and month processing? I currently have
pmlogger_daily running (as specified for daily processing) and using pmsnap
to generate snapshots every 30-minutes. This all works great! However, what
I also would like to implement Daily charts (at say 2 hour intervals for the
past 7 days) and Monthly charts (depicting daily averages for the past 31
days). 
I'm no scripting guru or ace programmer. I've tried setting up the crontab
for pmlogger_daily to only run on Sunday which works fine for one week
(provided the system isn't rebooted or pcp restarted). But I figured someone
in this list might have a better way or be able to provide some more
information. Any help is greatly appreciated.
Thanks,
Bobby Junkin
Lockheed Martin Space Operations
Huntsville AL
bobby.junkin@ums.msfc.nasa.gov


------_=_NextPart_001_01C086DC.D3FB0286
Content-Type: text/html
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2650.12">
<TITLE>Archive processing and reporting</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2 FACE=3D"Arial">I'm currently using PCP 2.1 in an IRIX =
6.5.9 environment.</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">According to the pmlogger_daily man =
page, toward the bottom, it indicates &quot;The script<B> =
/usr/pcp/bin/pmlogger_daily</B> could be copied and modified to =
implement a site-specific procedure for end-of-week and/or end-of-month =
management for a set of PCP archives.&quot; </FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">My question is, what specifically =
&quot;could be modified&quot; in this script to accomplish an =
end-of-week and month processing? I currently have pmlogger_daily =
running (as specified for daily processing) and using pmsnap to =
generate snapshots every 30-minutes. This all works great! However, =
what I also would like to implement Daily charts (at say 2 hour =
intervals for the past 7 days) and Monthly charts (depicting daily =
averages for the past 31 days). </FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">I'm no scripting guru or ace =
programmer. I've tried setting up the crontab for pmlogger_daily to =
only run on Sunday which works fine for one week (provided the system =
isn't rebooted or pcp restarted). But I figured someone in this list =
might have a better way or be able to provide some more information. =
Any help is greatly appreciated.</FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Arial">Thanks,</FONT>
<BR><FONT SIZE=3D2 FACE=3D"Arial">Bobby Junkin<BR>
Lockheed Martin Space Operations<BR>
Huntsville AL<BR>
bobby.junkin@ums.msfc.nasa.gov</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01C086DC.D3FB0286--

From owner-pcp@oss.sgi.com Thu Jan 25 09:27:38 2001
Received:  by oss.sgi.com id <S553981AbRAYR12>;
	Thu, 25 Jan 2001 09:27:28 -0800
Received: from mail.msi.umn.edu ([128.101.31.10]:40640 "EHLO mail.msi.umn.edu")
	by oss.sgi.com with ESMTP id <S553973AbRAYR1O>;
	Thu, 25 Jan 2001 09:27:14 -0800
Received: from squall.msi.umn.edu (IDENT:root@squall.msi.umn.edu [128.101.31.34])
	by mail.msi.umn.edu (8.9.3/8.9.3) with ESMTP id LAA56979
	for <pcp@oss.sgi.com>; Thu, 25 Jan 2001 11:27:13 -0600 (CST)
Received: from localhost (jay@localhost)
	by squall.msi.umn.edu (8.9.3/8.9.3) with ESMTP id LAA28250
	for <pcp@oss.sgi.com>; Thu, 25 Jan 2001 11:27:12 -0600
X-Authentication-Warning: squall.msi.umn.edu: jay owned process doing -bs
Date:   Thu, 25 Jan 2001 11:27:12 -0600 (CST)
From:   Jay Srinivasan <jay@msi.umn.edu>
To:     <pcp@oss.sgi.com>
Subject: myrinet pmda
Message-ID: <Pine.LNX.4.30.0101251119360.28073-100000@squall.msi.umn.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing


Hi,

I see that the ACE docs mention a Myrinet switch PMDA, but this isn't
to be found in the open source distribution of PCP. Is it available
only in the Pro version, can I get it from somewhere else, or will one
have to written again? I'd appreciate some pointers on this since I'm
just getting started on this.

Thanks,

Jay


From owner-pcp@oss.sgi.com Sun Jan 28 16:25:04 2001
Received:  by oss.sgi.com id <S554108AbRA2AYo>;
	Sun, 28 Jan 2001 16:24:44 -0800
Received: from sgi.SGI.COM ([192.48.153.1]:49421 "EHLO sgi.com")
	by oss.sgi.com with ESMTP id <S554105AbRA2AYV>;
	Sun, 28 Jan 2001 16:24:21 -0800
Received: from larry.melbourne.sgi.com ([134.14.52.130]) 
	by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam:
       SGI does not authorize the use of its proprietary
       systems or networks for unsolicited or bulk email
       from the Internet.) 
	via SMTP id QAA00329
	for <pcp@oss.sgi.com>; Sun, 28 Jan 2001 16:24:19 -0800 (PST)
	mail_from (markgw@sgi.com)
Received: from sandpit.melbourne.sgi.com (sandpit.melbourne.sgi.com [134.14.55.132]) by larry.melbourne.sgi.com (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id LAA10332; Mon, 29 Jan 2001 11:22:57 +1100
Date:   Mon, 29 Jan 2001 11:22:57 +1100 (EST)
From:   Mark Goodwin <markgw@sgi.com>
X-Sender: markgw@sandpit.melbourne.sgi.com
To:     Jay Srinivasan <jay@msi.umn.edu>
cc:     pcp@oss.sgi.com
Subject: Re: myrinet pmda
In-Reply-To: <Pine.LNX.4.30.0101251119360.28073-100000@squall.msi.umn.edu>
Message-ID: <Pine.LNX.4.21.0101291112010.26763-100000@sandpit.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

On Thu, 25 Jan 2001, Jay Srinivasan wrote:
> 
> Hi,
> 
> I see that the ACE docs mention a Myrinet switch PMDA, but this isn't
> to be found in the open source distribution of PCP. Is it available
> only in the Pro version, can I get it from somewhere else, or will one
> have to written again? I'd appreciate some pointers on this since I'm
> just getting started on this.

The Myrinet PMDA uses SNMP queries to extract per-port performance stats
from the SNMP server in 16-port Myrinet switches. The 4-port switch does
not have an SNMP server and thus cannot be monitored.

The Myrinet PMDA is part of the pcp-ace RPM, which has RPM dependencies
on the pcp-pro RPM. However, if you have the pcp-ace RPM you can
always rpm --force the install. The Myrinet PMDA should work without
pcp-pro installed, and you could monitor a Myrinet switch using tools
like pmval and pminfo (which are open source), but the GUI monitoring
tools mentioned in the ACE docs will either not work or not be available
at all because pcp-pro is not installed.

Another option is to use snmpget to query the switch directly.

Hope this helps,

-- Mark Goodwin
SGI Engineering


From owner-pcp@oss.sgi.com Mon Jan 29 03:15:37 2001
Received:  by oss.sgi.com id <S553984AbRA2LPS>;
	Mon, 29 Jan 2001 03:15:18 -0800
Received: from deliverator.sgi.com ([204.94.214.10]:63534 "EHLO
        deliverator.sgi.com") by oss.sgi.com with ESMTP id <S553740AbRA2LOs>;
	Mon, 29 Jan 2001 03:14:48 -0800
Received: from rattle.melbourne.sgi.com (rattle.melbourne.sgi.com [134.14.55.145]) by deliverator.sgi.com (980309.SGI.8.8.8-aspam-6.2/980310.SGI-aspam) via ESMTP id DAA29374
	for <pcp@oss.sgi.com>; Mon, 29 Jan 2001 03:13:47 -0800 (PST)
	mail_from (kenmcd@melbourne.sgi.com)
Received: from localhost (kenmcd@localhost)
	by rattle.melbourne.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id WAA72358;
	Mon, 29 Jan 2001 22:13:25 +1100 (AEDT)
X-Authentication-Warning: rattle.melbourne.sgi.com: kenmcd owned process doing -bs
Date:   Mon, 29 Jan 2001 22:13:25 +1100
From:   Ken McDonell <kenmcd@melbourne.sgi.com>
Reply-To: kenmcd@sgi.com
To:     Patrick Aland <paland@stetson.edu>
cc:     pcp@oss.sgi.com
Subject: Re: Output format
In-Reply-To: <20010123224845.A12589@stetson.edu>
Message-ID: <Pine.SGI.4.21.0101241457570.408516-200000@rattle.melbourne.sgi.com>
MIME-Version: 1.0
Content-Type: MULTIPART/MIXED; BOUNDARY="-2045888623-610679772-980766805=:767908"
Sender: owner-pcp@oss.sgi.com
Precedence: bulk
Return-Path: <owner-pcp@oss.sgi.com>
X-Orcpt: rfc822;pcp-outgoing

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

---2045888623-610679772-980766805=:767908
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Tue, 23 Jan 2001, Patrick Aland wrote:

> Sorry to reply again but is there a metric for the time the entry was made?

As I was working on the quick and dirty script I noticed this short
coming, and had a bet with myself on how long it would take someone
else to notice.

I won the bet.

There is now a "-e" option to pmie, and this produces output like ...

load_1 (Wed Jan 24 15:24:13 2001): 0.57
idle (Wed Jan 24 15:24:13 2001): 11.0
EOL (Wed Jan 24 15:24:13 2001): 12


I'll add this in the next open source release, and mail the trivial
patch here in the interim.

To convert pmie2col to recognize this new format and report accordingly
is left as an exercise for the reader.

> > On Wed, Jan 24, 2001 at 10:37:32AM +1100, Ken McDonell wrote:
> > > On Tue, 23 Jan 2001, Patrick Aland wrote:
> > > 
> > > > Which program will take the pcp data and output it in a certain format?
> > > > Mainly I am looking to use our current system which take logs of the form
> > > > 
> > > > [column1_name]	[column2_name]	...	[columnx_name]
> > > > [column1_data]	[column2_data]	...	[columnx_data]
> > > > [column1_data]	[column2_data]	...	[columnx_data]
> > > > [column1_data]	[column2_data]	...	[columnx_data]
> > > > [column1_data]	[column2_data]	...	[columnx_data]
> > > > 
> > > > I think i could use pmie to do this but am not sure. Any help is appreciated. Thanks. 
> > > 
> > > depends a little ... can you expand on what columnx_name might be?
> > > 
> > > If these are instance of a single metric, pmval will do this
> > > out of the box.
> > > 
> > > If these are different metrics (with or without instances), then
> > > if you have the IRIX or pcp-pro pieces, pmdumptext will do this,
> > > otherwise something like ...
> > > 
> > > cat <<End-of-File \
> > > | pmie \
> > > | some-awk
> > > column1_name=some expression;
> > > column2_name=some other expression;
> > > column3_name=a third expression;
> > > ...
> > > End-of-File
> > > 
> > > To be more concrete, here's a pmie script ...
> > > 
> > > load_1 = kernel.all.load #'1 minute';
> > > idle = kernel.all.cpu.idle;
> > > EOL = hinv.ncpu;
> > > 
> > > The EOL line is a sentinel and special in what follows, so there
> > > are 2 metrics being reported here.
> > > 
> > > Running pmie -v produces ...
> > > 
> > > load_1: 3.70
> > > idle: ?
> > > EOL: 12
> > > 
> > > load_1: 3.55
> > > idle: 8.3
> > > EOL: 12
> > > 
> > > load_1: 3.20
> > > idle: 9.9
> > > EOL: 12
> > > 
> > > load_1: 3.04
> > > idle: 9.4
> > > EOL: 12
> > > 
> > > load_1: 3.22
> > > idle: 8.6
> > > EOL: 12
> > > 
> > > Now feed this to pmie2col (a quick and dirty awk script, copy attached)
> > > and you get ...
> > > 
> > >  load_1:   idle:
> > >     3.70       ?
> > >     3.55    8.30
> > >     3.20    9.90
> > >     3.04    9.40
> > >     3.22    8.60
> > > 
> > 
> > Content-Description: pmie2col
> > > #!/bin/sh
> > > #
> > > #  convert pmie
> > > #
> > > 
> > > awk '
> > > NF == 0				{ next }
> > > state == 0 && $1 == "EOL:"	{ print ""; ncol = i; state = 1 }
> > > state == 0			{ printf " %7.7s",$1 }
> > > $1 == "EOL:"			{ for (i = 0; i < ncol; i++) {
> > > 					if (v[i] == "?") printf " %7.7s","?"
> > > 					else printf " %7.2f",v[i]
> > > 					v[i] = "?"
> > > 				  }
> > > 				  print ""
> > > 				  i = 0
> > > 				  next
> > > 				}
> > > 				{ v[i++] = $2 }'
> > 
> > 
> > -- 
> > ------------------------------------------------------------
> >  Patrick Aland                          paland@stetson.edu
> >  Network Administrator                  Voice: 904.822.7217
> >  Stetson University                     Fax: 904.822.7367
> > ------------------------------------------------------------
> 
> -- 
> ------------------------------------------------------------
>  Patrick Aland                          paland@stetson.edu
>  Network Administrator                  Voice: 904.822.7217
>  Stetson University                     Fax: 904.822.7367
> ------------------------------------------------------------
> 


---2045888623-610679772-980766805=:767908
Content-Type: TEXT/PLAIN; charset=US-ASCII; name=patch
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.SGI.4.21.0101292213250.767908@rattle.melbourne.sgi.com>
Content-Description: pmie patch for -e
Content-Disposition: attachment; filename=patch
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---2045888623-610679772-980766805=:767908--