I am running CentOS 5.1, nagios 3.0.1 with nagios-plugins-1.4.11. Though the plugin I am referring to, I got off the Exchange nagiosexchange.org/cgi-bin/p … 1.html;d=1
It is a simple bash script that uses uptime and awk to spit out the uptime and compare it to the warn/critical levels. This is intended to be used for the localhost.
This plugin worked just fine on my test machine which was Fedora Core 7.
In fact, the plugin works fine from the command line. I can run it as the nagios user and it works just fine. But on the website it comes back as UNKNOWN status with stderr = 65280 : in the Status Information box.
I havent been able to find this error online.
I dont see why the script returns an error on the website but not the command line. The permissions are 755, owner nagios:nagios, same as all the other plugins.
Anyone have any thoughts on how to get this to report to the website correctly?
usually, when you have this kind of errors, it comes from one of the 2 following causes:
envirronment: you may have certain envirronment variables set when your are logged in, but the nagios user don’t have the same. ie: when your are logged in, the command “uptime” is known thanks to the “PATH” variable, but Nagios doesn’t have it.
the current path. It is almost the same problem: when using command line, your are located in a certain directory. Nagios won’t have the same. As far as I know, it is usually located in the / directory… which means it won’t have access to the same executables.
Yeah I am leaning towards “something else”. My user and the nagios user have the same paths, and all of the same env vars setup. The script runs fine as the nagios user on the command line.
It must be something quirky with this plugin, as all the other plugins for the localhost work fine.
I have tripled check the service’s definition and command it runs, they are correct.
