Icinga 2 client windows - warning and critical levels not set and not work

This forum was archived to /woltlab and is now in read-only mode.
  • Hello all

    I have a problem.

    I installed the icinga client on version 2.7.1 of Icing on version 2.4.1 in OS Windows

    After installing the client on Icingi side, I have processes such as disk, processor etc.

    Unfortunately, critical conditions and warnings do not work for them.

    In Performance data and Custom Variables I have something like this.


    Although I tried all of the above solutions to launch critical and warning options but nothing helped.


    1. apply Service "procs" { import "generic-service"
    2. check_command = "procs"
    3. assign where host.name == NodeName
    4. vars.procs_warning = 400
    5.     vars.procs_critical = 500
    6. }


    1. object CheckCommand "procs" {
    2. command = [ PluginDir + "/check_procs" ]
    3. ....
    4. vars.procs_traditional = false
    5.      vars.procs_nokthreads = false
    6.      vars.procs_warning = 250
    7.      vars.procs_critical = 400
    8. }

    Do you have any idea how to solve this problem?

  • That screen with "dummy" as check_command does not match your service apply rule. Furthermore you're not using the config sync, and neither command endpoint. That given example will be executed by the master node only, nothing relates to your client here.

  • Thank you for your answer,

    The point is that I just installed Icing 2 Client on the server and in settings I did nothing.

    In the /etc/icinga2/repository.d/hosts/host.xx.local/ directory I have these files:

    disk C.conf, disk.conf, icinga.conf, load.conf, ping4.conf, ping6.conf, procs.conf, swap.conf, users.conf.

    FoFor procs.conf and other files are such settings:

    1. object Service "procs" {
    2. import "satellite-service"
    3. check_command = "dummy"
    4. host_name = "host.xx.local"
    5. zone = "hots.xx.local"
    6. vars.procs_critical = 500
    7. vars.procs_warning = 400
    8. }

    Here I also added parameters for critical and warning but that also did not help.

  • Oh, you are using the deprecated "bottom up" client mode where configs are synced from the client to the master.

    Well, that explains it. The configuration modification on the master is never synced to the client again. The client still thinks that it does not have any threshold applied, you'll need to edit it over there.

    This is one reason for deprecating and removing this client mode (2.8 fully removes it). It does not work reliably and you cannot manage the configuration on the master node, or modify configuration attributes.

    I'd suggest to migrate your client mode to "top down", the easiest one is by using command endpoint.