----
Zack,
a nice tool to capture and visualzize SNMP data is MRTG, which was initially
developed to track router traffic statistics.
Find below the MRTG file that we set up for monitoring the length of the run
queue (which you know from the output or the "w" command).
In this configuration, pmgrd is addressed as a "subserver" of snmpd via port
161.
Actually you could do a "getmany <host> <community> 1" to see the tremendous
amount of information that a Tru64 UNIX hosts is offering (which might also
be
a security concern).
Performance Manager is offering its data via iso(1) org(3) dec (24) pm (11)
pmPrivate(1), which means under 1.3.24.11.1:
(from /usr/opt/pm/data/pm-mib)
...
dec OBJECT IDENTIFIER ::= { iso org(3) 24 }
pm OBJECT IDENTIFIER ::= { dec 11 } -- DEC Performance
Metrics Registry
pmPrivate OBJECT IDENTIFIER ::= { pm 1 } -- PM private root
...
By the way, a complete rewrite of the MRTG package is under way. See
http://ee-staff.ethz.ch/~oetiker/webtools/rrdtool/ for details.
Regards,
Thomas Meyer, EDV-Systemadministrator
email: Thomas.Meyer_at_dla-marbach.de
smail: Deutsches Literaturarchiv Marbach
Schillerhoehe 8-10
D-71672 Marbach
Telefon: +49-7144-848-141
Fax: ...-387
WWW: http://www.dla-marbach.de
#---------------------------------------------------------------------
# Deutsches Literaturarchiv Marbach
#---------------------------------------------------------------------
# $Source: /usr/local/mrtg/cfg-files/RCS/dlanserv.cfg,v $
# $Id: dlanserv.cfg,v 1.3 1999/07/16 10:05:32 meyert Exp $
# Aufruf:
# Parameter:
# Zweck: cfg-File fuer MRTG
# Umgebung: Tru64-UNIX mit MRTG
# Siehe auch:
# Bemerkungen: RCS-Pflege zur Zeit ueber dlanserv.cfg: dataserv.cfg etc.
# wird daraus abgeleitet
# $Log: dlanserv.cfg,v $
# Revision 1.3 1999/07/16 10:05:32 meyert
# Umstellung auf 30s-Werte statt 5s-Werte, da durch mrtg
# der 5s-Wert auf DLANSERV immer hoch ist
#
# Revision 1.2 1999/07/16 09:51:08 meyert
# Gewinnung der Werte direkt ueber Nennung der SNMP-OID
#
# Revision 1.1 1999/07/16 09:09:12 meyert
# Initial revision
#
#---------------------------------------------------------------------
# Workdir specifies where the logfiles and the webpages should be created.
WorkDir: /home/owner-www/www-intra/dla-intern/mrtg/dlanserv
# How often do you call mrtg? The default is 5 minutes. If you call it
# less often, you should specify it here. This does two things:
# * the generated HTML page does contain the right information about
# the calling interval ...
# * a META header in the generated HTML page will instruct caches
# about the time to live of this page .....
Interval: 5
# With this switch mrtg will generate .meta files for CERN and Apache
# servers which contain Expiration tags for the html and gif files. The
# *.meta files will be created in the same directory as the other files,
# so you might have to set ``MetaDir .'' in your srm.conf file for this
# to work
WriteExpires: Yes
# If you want to keep the mrtg icons in some place other than the
# working directory, use the IconDir variable to give its url.
IconDir: /dla-intern/mrtg/icons/
#
# Default-Werte fuer alle Targets festlegen
#
Options[_]: growright, transparent
kilo[_]: 1024
background[_]: #FAEBD7
Directory[pmAoCpuAvgRunQLen]: pmAoCpuAvgRunQLen
Target[pmAoCpuAvgRunQLen]:
1.3.24.11.1.2.4.2.0&1.3.24.11.1.2.4.3.0:public_at_dlanserv
MaxBytes1[pmAoCpuAvgRunQLen]: 9999
MaxBytes2[pmAoCpuAvgRunQLen]: 9999
Options[pmAoCpuAvgRunQLen]: gauge, growright, transparent, nopercent
kMG[pmAoCpuAvgRunQLen]:
WithPeak[pmAoCpuAvgRunQLen]: ym
Title[pmAoCpuAvgRunQLen]: DLANSERV pmAoCpuAvgRunQLen30Sec /
pmAoCpuAvgRunQLen60Sec
YLegend[pmAoCpuAvgRunQLen]: Count
ShortLegend[pmAoCpuAvgRunQLen]: /100
Legend1[pmAoCpuAvgRunQLen]: average length of the run queue over the last 30
seconds
Legend2[pmAoCpuAvgRunQLen]: average length of the run queue over the last 60
seconds
Legend3[pmAoCpuAvgRunQLen]: maximum average length of the run queue over the
last 30 seconds
Legend4[pmAoCpuAvgRunQLen]: maximum average length of the run queue over the
last 60 seconds
LegendI[pmAoCpuAvgRunQLen]: 30s RunQLen
LegendO[pmAoCpuAvgRunQLen]: 60s RunQLen
PageTop[pmAoCpuAvgRunQLen]:
<H1>DLANSERV: Length of Run Queue * 100</H1>
pmAoCpuAvgRunQLen30Sec (One hundred times the average length of the run
queue
over the last 30 seconds) +
pmAoCpuAvgRunQLen60Sec (One hundred times the average length of the run
queue
over the last 60 seconds)
---
Zack
> > 3) Is there a better way to view SNMP on Digital UNIX 4.0d that actually
> > keeps archives and can make performance reporting possible? - I hope so.
I just saw the summary and thought this might be useful too...
Mrtg is a tool for plotting SNMP parameters against time, initially it
was intended for charting router traffic, but it can handle just about
and SNMP value provided you have a MIB for it. You compile it, configure
it, add to cron and forget about it, just look at the pretty pictures.
http://www.ee.ethz.ch/~oetiker/webtools/mrtg/mrtg.html
Simon
--
Simon Greaves voice: (+679) 212114
Computer Centre fax: (+679) 304089
The University of the South Pacific email: Simon.Greaves_at_usp.ac.fj
Suva, Fiji
Please check out Collect:
ftp://ftp.digital.com/pub/DEC/collect
It gathers statistics in a very unobtrusive way (unlike Performance
Manager). These can then be displayed later via a Motif GUI, or
can be fed to Excel for appropriate reporting to Management.
It comes in a setld installable kit, and uses Perl (which is also
available in a setld installable kit).
Regards,
Ram Rao
Received on Tue Oct 05 1999 - 20:36:20 NZDT
This archive was generated by hypermail 2.4.0 : Wed Nov 08 2023 - 11:53:39 NZDT