Page 1 of 1

Defining metric type for SNMP polls using check_snmp

Posted: Thu Mar 28, 2013 3:44 pm
by Admin_UCOP
Howdy experts!

probably a very simple question here. I set up an SNMP query for a DNS appliance to check the processor loads. It does return the correct values for all 4 CPU's with an output that looks like this:

SNMP OK - A B C D (where ABCD are the CPU utilization values in %). But it does not say that it is measured in % and the graph shows just a number. How do I define within my query that the value is % and so that it will reflect it on the performance graph. My current command format is:

check_snmp -o "OID-1 OID-2 OID-3 OID-4" -C communitystring -P 2c -w 75, 75, 75, 75 -c 90, 90, 90, 90

Thanks a lot in advance!!

Re: Defining metric type for SNMP polls using check_snmp

Posted: Thu Mar 28, 2013 4:34 pm
by Admin_UCOP
Folks,

i have an update. I was able to get the units set using the '-u' option. But it is not updated on the performance graphs. Can somebody suggest how to set the units on the performance graphs?

I changed the main command to:

snmp_check -o "OID-1 OID-2 OID-3 OID-4" -D , -u %,%,%,% -C communitystring -P 2c -w 75,75,75,75 -c 90,90,90,90

thanks

Re: Defining metric type for SNMP polls using check_snmp

Posted: Thu Mar 28, 2013 4:39 pm
by nscott
This is unfortunately outside the ability of check_snmp.

A workaround for you would be to write a wrapper script for check_snmp or search the Nagios Exchange for plugins for your specific device. Please let me know if you have questions about either of these options.

Re: Defining metric type for SNMP polls using check_snmp

Posted: Thu Mar 28, 2013 5:24 pm
by Admin_UCOP
nscott - thanks for confirming this. I will go ahead and create a new wrapper script for this...i know how to do this. Just wanted to see if there was a shortcut.