[mrtg] Re: MRTG - Windows 2000 SNMP!?!

[Meissner Sebastian-asm037]

Hi Colin,

I'm monitoring the disk space (and other things) of a W2k-Server and found
some problems:
The values of special OID's (here: the disk space OID's) are not updated
until you restart the snmp-server. Microsoft knows about that problem and
will hopefully deliver a patch.
That information won't solve your problem (sorry ;-)), but perhaps your
problem isn't MRTG but the W2k-snmp-server too...

Sebastian Meissner


-----Original Message-----
From: Colin Horsington [mailto:ColinH at flow.com.au]
Sent: Donnerstag, 29. Juni 2000 03:53
To: 'mrtg at list.ee.ethz.ch'
Subject: [mrtg] MRTG - Windows 2000 SNMP!?!


MRTG Users,

Has anyone tried to collect the CPU load on a DUAL CPU Windows 2000 server?

The problem I am facing is:

%(xx)-(0)-(yy) -=> snmpwalk 10.0.0.1 public .1.3.6.1.2.1.25.3.3.1.2.1
~
host.hrDevice.hrProcessorTable.hrProcessorEntry.hrProcessorLoad.2 = 1

%(xx)-(0)-(yy) -=> snmpwalk 10.0.0.1 public .1.3.6.1.2.1.25.3.3.1.2
~
host.hrDevice.hrProcessorTable.hrProcessorEntry.hrProcessorLoad.1 = 58
host.hrDevice.hrProcessorTable.hrProcessorEntry.hrProcessorLoad.2 = 1

%(xx)-(0)-(yy) -=> snmpwalk 10.0.0.1 public .1.3.6.1.2.1.25.3.3.1.2.0
~
host.hrDevice.hrProcessorTable.hrProcessorEntry.hrProcessorLoad.1 = 58

%(xx)-(0)-(yy) -=> snmpwalk 10.0.0.1 public .1.3.6.1.2.1.25.3.3.1.2.1
~
host.hrDevice.hrProcessorTable.hrProcessorEntry.hrProcessorLoad.2 = 1

%(xx)-(0)-(yy) -=> snmpwalk 10.0.0.1 public .1.3.6.1.2.1.25.3.3.1.2.2
~
host.hrDevice.hrNetworkTable.hrNetworkEntry.hrNetworkIfIndex.3 = 1

Note: that Microsoft have moved where there system information is in 2000.

When viewing the above commands (performed on Linux - Debian) doing a walk
presents two processors .1 & .2
snmpwalk or snmp get on these oids does not return the proper results
instead you have to walk/get on .0 & .1

Also using the Oid in MRTG of .1.3.6.1.2.1.25.3.3.1.2.1 returns the proper
CPU load for the CPU (CPU 0 on NT),
whilst using Oid the following Oid on MRTG (.1.3.6.1.2.1.25.3.3.1.2.1 )
returns the following error:

C:\MRTG\run>perl c:\mrtg\run\mrtg cpu1.cfg
SNMP Error:
Received SNMP response with error code
  error status: noSuchName
  index 1 (OID: 1.3.6.1.2.1.4.20.1.2.0.1.3.6.1.2.1.25.3.3.1.2.0)
SNMPv1_Session (remote host: "ggggggggggggg" [10.0.0.1].161
                  community: "public"
                 request ID: 1758447544
                PDU bufsize: 8000 bytes
                    timeout: 2s
                    retries: 5
                    backoff: 1)
SNMPGET Problem for ipAdEntIfIndex..1.3.6.1.2.1.25.3.3.1.2.0 on
public at AU-WEB-00
05
SNMP Error:
Received SNMP response with error code
  error status: noSuchName
  index 1 (OID: 1.3.6.1.2.1.2.2.1.10)
SNMPv1_Session (remote host: "gggggggggggggg" [10.0.0.1].161
                  community: "public"
                 request ID: 1758447545
                PDU bufsize: 8000 bytes
                    timeout: 2s
                    retries: 5
                    backoff: 1)
SNMPGET Problem for ifInOctets. ifInOctets. sysUptime sysName on
public at AU-WEB-0
005
SNMPGET: Failed to reach target:
".1.3.6.1.2.1.25.3.3.1.2.0:public at gggggggggggggg".
 I tried multiple times! public at ggggggggggggg

I also recieve a similar error using Oid ".1.3.6.1.2.1.25.3.3.1.2.2". Has
anyone done any work
with Windows 2000 and MRTG in monitoring such things as CPU load (dual quad)
Web stats and Pagefiles.

(ifInOctets. ifInOctets works fine via the cfgmaker perl script for 2000)

Any help would be gratefully appreciated.

Colin.
Flow Communications