QCG-Notification Nagios Probe

The Nagios probe for QCG-Notification tests the service availability and usability. The tested notification scenario is typical and includes all the basic interactions within the system. The list prestented below describes the particularities of the alghoritm:

  1. Create PullPoint
  2. Subscribe the PullPoint to receive specific type of notifications.
  3. Send a notification of a given type.
  4. Get the message stored in the PullPoint and chek if it matches the notification previously sent.
  5. Unsubscribe.
  6. Destroy the PullPoint.

Installation from RPMS

  • If you have not done this yet, install EPEL, PL-Grid and QCG repositories:  instruction.
  • Install probe:
    yum install qcg-ntf-nagios-probe

Installation from Sources

Get the QCG-Notification Nagios Probe package from Downloads page.

After unpacking the package simple run the ./install.sh script giving as the first argument directory where the probes should be installed to. E.g.:

  tar -xzf qcg-ntf-nagios-3.0.tar.gz
  cd qcg-ntf-nagios-3.0
  ./install.sh /opt/qcg-ntf-nagios

In order to use the probe installed in this way the QCG-Notification client must be available on your system. You need to install two packages: QCG-Core and QCG-Notification (see QCG-Notification installation). If you install the packages directly to the same directory as the probe (e.g. /opt/qcg-ntf-nagios) you should be able to use the probe without any changes. Alternatively, if your QCG-Core and QCG-Notification will be installed to the different location, you can change the QCG_NTF_HOME and QCG_CORE_HOME variables in the check_qcg_ntf script to point to the correct place.


./check_qcg_ntf (-l url | -H hostname -p port) -t timeout [-v 0-3]
-l url - QCG Notification url
-H hostname - QCG Notification host              
-p port - QCG Notification port
-t timeout - test timout given in seconds
-v 0-3 - verbosity (default: 0)


./check_qcg_ntf -H grass1.man.poznan.pl -p 19011 -t 60

Exit Codes

  • STATUS_OK (0) - Test finished successfully.
  • STATUS_WARNING (1) - Test finished with exit code different than 0, Test did not finish within given timeout.
  • STATUS_CRITICAL (2) - Test failed.
  • STATUS_UNKNOWN (3) - The probe internal or configuration error.