[mrtg] NT stats gathering problem

[Mike Holley-]

Hi there,

I am trying to setup MRTG to monitor the CPU utilization for one of our NT
boxes. MRTG is working fine on our Linux box for monitoring our router
throughputs, but when I attempt to use the following config file along with
MRTG to poll the NT box for  its CPU Util. It errors out. The following is
the config file that I am using along with the errors generated. I know that
the config file mentions that it is for MRTG for NT, so another question I
have is whether or not the config files are portable between platforms.
Thanks in advance for any help.

The following is the content of ntcpu.cfg that is used for polling the NT
box for CPU utilization

####################################################################
# MRTG WINDOWS NT CONFIG FILE
# EDM268:  CPU #1 Usage (%)
# Researched and configured (including compile into NT environment)
# by Garth K. Williams (1998)
# Edmonton, AB (403) 469-4341
# Original Date: 05April1998
# Lost ALOT of sleep, but couldn't have done it without:
# MRTG (WHAT A GREAT PROGRAM!)
# PERL (SEE ABOVE)
# GD (DITTO)
# DJGPP (DITTO, AND ALSO HARD TO BELIEVE IT'S FREE) 
####################################################################
workDir: /home/httpd/html/web/nt/phoenix
IconDir: /home/http/html/web

Title[^]: CPU #1 Usage (%) on System:

Background[192.168.50.251_CPU1]: #9090909
Target[192.168.50.251_CPU1]:1.3.6.1.4.1.311.1.1.3.1.1.2.1.3.0:public at 192.168
.50.251
MaxBytes[192.168.50.251_CPU1]: 100
Title[192.168.50.251_CPU1]: Phoenix
PageTop[192.168.50.251_CPU1]: <H2>CPU #1 Usage (%) on System: Phoenix</H2>
options[192.168.50.251_CPU1]: gauge, nopercent, absolute, transparent
Unscaled[192.168.50.251_CPU1]: dwmy
#Supress[192.168.50.251_CPU1]:dwmy
XSize[192.168.50.251_CPU1]: 380
YSize[192.168.50.251_CPU1]: 100
YLegend[192.168.50.251_CPU1]: CPU Usage (%)
ShortLegend[192.168.50.251_CPU1]: %
Legend1[192.168.50.251_CPU1]: CPU use in percent
#Legend2[192.168.50.251_CPU1]:
#Legend3[192.168.50.251_CPU1]:
#Legend4[192.168.50.251_CPU1]:
LegendI[192.168.50.251_CPU1]: CPU:&nbsp

Errors received are as follows : 

SNMPGET problems for ifOutOctets sysUptime sysName on public at 192.168.50.251
<mailto:public at 192.168.50.251> 
SNMPGET:Failed to reach target:
"1.3.6.1.4.1.311.1.1.3.1.1.2.1.3.0:public at 192.168.50.251. I tried multiple
times!

If I use SNMPGET to query OID : 1.3.6.1.4.1.311.1.1.3.1.1.2.1.3.0 it returns
the value without any problems.

It looks like MRTG is still attempting to query the OIDs for the router
traffic stats. Are there modifications to the MRTG perl script that need to
be made in order to do all of the extra data collection, such as the CPU,
hard drive, etc.. monitoring? I thought it was simply the implementation of
the custom config file that was needed in this case.

Michael D. Holley, CNE
IT Technician
Delta Controls Inc.
mholley at deltacontrols.com

--
* To unsubscribe from the mrtg mailing list, send a message with the
  subject: unsubscribe to mrtg-request at list.ee.ethz.ch
* The mailing list archive is at http://www.ee.ethz.ch/~slist/mrtg