No announcement yet.


  • Filter
  • Time
  • Show
Clear All
new posts

  • SNMP Help

    Hello to everyone. I'm new to Nagios and Centreon world and I'm trying to configure it for the first time. My need is to monitor some hardware parameters like array status of the physical machines that I have in a datacenter (they are all HP servers, from G5 generation to G7).

    I have 6 ESXi servers, some Linux machines (RedHat) and some windows 2000 and 2003. I'm a bit confused about OID-Mib and other things involving SNMP.

    For test I enabled this protocol on a win2003 server and a ESXi and I added hosts to the Centreon machined I deployed. I succeed in monitoring some standard parameters like disk usage, memory, cpu (so I understood that SNMP was working properly)...then I tryed to understand how to configure custom SNMP traps for monitoring other hardware parameters of my servers but I'm a bit confused on how to do it. I tried to create some snmp traps: I found some OID codes corresponding to parameters I want to monitor and I checked them from web interface and also by cmd using the command:

    "$USER1$/check_centreon_snmp_value -H $HOSTADDRESS$ -C $ARG1$ -v $ARG2$ -o $ARG3$ -w $ARG4$ -c $ARG5$"

    passing some parameters like: !public!1!!30!40 and IP address of my machine as $HOSTADDRESS$.

    Some OID returns some values (like temperature or so) but other that should correspond to Array Status returns "No such name, error status...".

    Now: how can I make the system "interpret" this codes? I tried to make some custom SNMP traps but I don't know how to find trap names. The guide says to insert the names specified in mib file but where I can find it?

    To summarize my doubts and triyng to solve my problems I ask you:

    How the SNMP system works? (traps, Mibs, OID?)

    I have to import Mibs packages on Centreon to interpret OIDs? How can I do it? (I found some hp MIB packages but when I try to import them from Centreon web interface under Configuration-Services-Mibs i get some errors).

    After importing them what should I make? I think that I would find some new SNMP traps corrisponding to them.

    Then I shoud create a service using the command check_centreon_snmp_value with newly imported OID as parameter? I have to link the same SNMP trap also in "relations" tab under service?

    The last question: how this package work and can it solve all my problems? How can I import and use it for monitoring array status? Is there any step by step guide that I can follow? (consider my inexperience ;-) )!

    Thanks in advance

  • #2
    There are a lot of OIDs to get in order to get the HP hw status, and that's very hard to do that with the check_centreon_snmp_value plugin.

    An excellent HP hardware plugin has been created by Gerhard
    The only requirement is a recent version of the HP Agents (PSP) on all servers to be monitored.

    If you want to have a better overview on SNMP, just google it..

    Hope it helps.


    • #3
      Thank u! I understood much more reading posts around and I was trying to implement check_snmp_cpqHealth Version 1.4.

      The HP agents you are saying exists for ESXi, windows and linux? They are difficult to install?


      • #4
        Hi Mante80
        check on HP website, this is for Proliant DL380 G6 s&taskId=135&prodTypeId=15351&prodSeriesId=38840 82

        Installation is not so difficult, have a look at the documentation. It should be the same for G5 platforms.
        Check on the related release notes if you need to upgrade also some firmware before installing PSP.
        Leave native ESX/ESXi drivers, do not install FC HBAs or NICs drivers.

        Hope it helps