A nagios plugin to check/query values provided by the local
collectd daemon. It's implemented as a drop-in
replacement for collectd-nagios understanding/passing all of its command
line arguments.
If you already use collectd and
nagios you'll most probably use the
collectd-nagios
utility to query values from your local collectd daemon. In this case you (most
probably) also are annoyed by the completely useless service output line
provided by this utility. Most people can't see the context in something like
CRITICAL: 1 critical, 0 warning, 0 ok. This is where check_collectd will
come in handy.
check_collectd does exactly what collectd-nagios does but allows you to
specify Perl sprintf format strings for its output.
While -f specifies the format string for the service status OK, the -F
switch specifies the format string used for service status Warning and
Critical. The Unknown service status is outputted as returned from
collectd-nagios.
If provided, check_collectd will call printf with the appropriate format
string providing the service status string as first argument. The following
arguments depend on the specified collectd-nagios consolidation function.
If the consolidation function is none the following arguments are the values
(in most cases floats/doubles) found in the performance data field. If one of
the other consolidation functions is used (either average, sum or
percentage), the respective value (also a float/double in most cases) is
passed as the second argument following the performance data values as in the
case of none. Feel free to use or not use all or any values in your format
string.
To see some example nagios commands refer to examples.md.
collectd-nagios must be in the user's PATH which is the default if
collectd was installed via package managers. Queries using collectd-nagios
must be working (e.g. you configured the
unixsock
plugin correctly). Although, check_collectd was tested using collectd-5.8.1
and perl-5.20.3, it most probably also works with older versions of both tools.