Nagios Support Forum

I am having trouble pulling process/cpu information for a specific process. On our Linux server we have two processes that have the same name but use different arguments. We are using the check_snmp_process_wizard.pl plug-in/script and the following syntax to identify each process:

./check_snmp_process_wizard.pl -H 192.168.2.101 --v2c -C Public -f -A -n '/opt/test/bin/check/test/test -l /opt/test/bin/test/config -c test.xml' -u 195,200

This works great for up/down and memory usage but returns No data for CPU (1 lines(s)):UNKNOWN when using the CPU check.

I tried editing the my $delta_of_time_to_make_average in the script but it doesn't seem to make a difference. Is there any way to obtain CPU data for a process this way?

This plugin is going to store some information in /tmp/tmp_Nagios_proc.*

It needs to run for several times (exactly the same command) before it can make the calculation., if it cannot it will return the (1 lines(s)):UNKNOWN you describe.

On the second run you should see
(2 lines(s)):UNKNOWN

You should be able to open the /tmp/tmp_Nagios_proc.* files and see the data it has collected.

Once you have several data points and the delta has been reached you should get the cpu percentage.

As I mentioned before, the command must be exactly the same or it will change the tmp filename and it will start over

Thanks for your help. I was able to narrorw it down to the $o_desc variable in the script/plug-in. It couldn't write the file as the arguments used to define the service were outside the range for what a file could be named. I modified the script, editited the command.cfg and it now writes the file and produces the correct result.