check_oes_stat plugin for Nagios

0 Likes
check_oes_stat script is an attempt to do for OES2/Linux servers a portion of what mrtgext.nlm did for NetWare servers. In conjunction with Nagios and NRPE, this script can return statistics on NCP Connections, NCP uptime, NCP Open Files, and DS Version.

You need to have a functional nagios installation and nrpe on the target server. Rainer Brunold has an excellent article, "Nagios 3.0 - A Extensible Host and Service Monitoring" at http://www.novell.com/coolsolutions/feature/19807.html that describes how to get it all setup and configured.

For this script, simply download and install in your plugins directory on the server to be monitored. Then edit the nrpe.cfg file, adding the following lines:


command[check_oes_edir_ver]=sudo /usr/lib/nagios/plugins/check_oes_stat -v DSVER
command[check_oes_conns]=sudo /usr/lib/nagios/plugins/check_oes_stat -v CONNS
command[check_oes_uptime]=sudo /usr/lib/nagios/plugins/check_oes_stat -v UPTIME
command[check_oes_ofiles]=sudo /usr/lib/nagios/plugins/check_oes_stat -v OFILES


This script uses the 'ncpcon' and 'ndsstat', which can only be run as the root user, hence the use of 'sudo' above. So you must also add to your sudoers file (using 'visudo') the following line (assumes nrpe is running as user nagios):


nagios ALL=NOPASSWD:/usr/lib/nagios/plugins/


Then on your nagios server just add the services, something like:

define service{
use oes-service
service_description NCP Connections
check_command check_nrpe!check_oes_conns
service_groups oesservices
hostgroup_name oesservers
}

define service{
use oes-service
service_description NCP Open files
check_command check_nrpe!check_oes_ofiles
service_groups oesservices
hostgroup_name oesservers
}

etc.


Limitations: this tool doesn't accept values for 'warning' or 'critical' levels like most other nagios plugins, but the output is well suited for graphing with an add-on like nagiosgraph. (See Rainer Brunold's article "Nagios 3.0 Extension - NagiosGraph" at http://www.novell.com/coolsolutions/feature/19843.html)

To graph NCP connections using nagiosgraph, add the following to your map file:

# Service type: NCP connections (OES2 servers)
# output: NCP Conn Slots Used = 415, NotLoggedIn Conns = 264, Licensed Conns = 34
# perfdata: ncpconns=415;264;34
/output:NCP Conn Slots Used = (\d ), NotLoggedIn Conns = (\d ), Licensed Conns = (\d )/
and push @s, [ ncpconns,
[ slots_used, GAUGE, $1 ],
[ nli_cons, GAUGE, $2 ],
[ licensed, GAUGE, $3 ],
];


You could similarly chart Open Files as well.

Labels:

How To-Best Practice
Collateral
Comment List
  • I see that in your service definitions, you're using a template called oes-service. Would you be kind enough to share what that looks like? I'm happy to create my own, but I'd like to compare notes. Thanks!
  • in reply to MigrationDeletedUser
    The utils.sh is included in a nagios distribution, which is where this plugin was intended to be run. But it can run stand-alone; the primary things that it's getting from utils.sh are the following definitions:

    STATE_OK=0
    STATE_WARNING=1
    STATE_CRITICAL=2
    STATE_UNKNOWN=3
    STATE_DEPENDENT=4

  • Hi,
    this script tries to include antoher scipt, that doesnt exists:
    . ${PROGPATH}/utils.sh


    Please explain what this utils.sh should be for

    Axel
Related
Recommended