Checkmk 2.0 is here! See what's new.

Check manual page of job

Job Monitoring
Distribution official part of Check_MK
License GPL
Supported Agents Linux
This check monitors state and performance information of any linux program call, for example regular running cronjobs.

The check uses information provided by the wrapper program mk-job. This program is shipped with the linux agent and installed to /usr/bin. mk-job is a wrapper which is called instead of the program. For example, if you have a command line nightly-backup >/dev/null which gets executed by a cronjob every night, you can change the command line to mk-job backup nightly-backup >/dev/null to let mk-job collect information about the job during runtime. In this expression, the string backup is the identifier of the job to be executed. It must be a unique identifier for this job on each host. When the job is finished, mk-job writes the collected data to /var/lib/check_mk_agent/job/. The agent sends all these data to the Checkmk server.

The check is CRITICAL if the exit code of the job is not 0, or if warning or critical limits for the job age have been reached. Other mappings of exit code to check state can be defined in the check parameters.

Limits can be configured with WATO.

This check is also cluster-aware and worst or best state is configurable. Default is worst state.

Item

The identifier of the job defined by the first argument to {mk-job}.

Discovery

One service is created for each job.