Posts by log1c

    there is also an argument for object_type:

    Code
    1. icingacli director service create generic-service --object_type template --max_check_attempts '5' --check_interval '1m' --retry_interval '30s'
    2. icingacli director service create load --object_type template --imports "generic-service" --check_command load
    3. icingacli director service create load --json "{\"object_type\":\"apply\",\"imports\":[\"load\"],\"assign_filter\":\"host.name=%22${HOSTNAME}%22\"}"

    If I see this correct, you created a service, but it is not bound to anything, so it does not show at any host.


    here is how a service template and an applied service look like:


    so I'd say, you need to create an apply rule that imports your service template and assign it to a host.

    Can you be more precise as to what you mean by

    Quote

    .is there any way to find some built in check command like the nagios plugins.

    In the command list inside Icinga Director all commands displayed with a "pin" in the front are pre-defined.

    The are ready to use, you only need to configure:

    1. the needed fields for the arguments

    2. the corresponding plugin script in /usr/lib/nagios/plugins

    3. a service template that uses the check command

    Do you have the scripts that these commands need in the respective folder?

    Default for "PluginDir" is "/usr/lib/nagios/plugins"

    So you will need your "check_updates" script in this folder, as well as any other script you want to execute.


    Concerning your command:

    Why the skip_key for your "-c" argument?


    to get more scripts in you PluginDir you can install these packages (example for Ubuntu):

    Code
    1. monitoring-plugins - Erweiterungen für zu Nagios kompatible Überwachungssysteme
    2. monitoring-plugins-basic - Plugins for nagios compatible monitoring systems (basic)
    3. monitoring-plugins-common - Common files for plugins for nagios compatible monitoring
    4. monitoring-plugins-standard - Plugins for nagios compatible monitoring systems (standard)
    5. nagios-snmp-plugins - SNMP Plugins for nagios

    Hi all,


    I'm having a question(or some :D) concerning an Icinga2 setup with satellites in different time zones.

    The system has a master and a satellite in time zone Europe/Berlin, one satellite in the US and one in South Korea (don't know the exact location of those two).

    As of know I have set up ntp on all of these machines and configured it to "Europe/Berlin".

    The reasons I did this are:

    1. the customer has disabled Hyper-V time sync for the guest machines

    2. before the ntp setup, checks were often overdue


    The disabled Hyper-V time sync also leads to the problem, that Icinga2 starts with the "random time" the Hyper-V injects (e.g. two months in the future).

    To work around this I added the ntp service to the requirements in the icinga2 system.d script.


    Is setting all sattellites to the same time zone the correct way?

    I have read, that IcingaWeb2 takes the time zone from the browser and converts timestamps to UTC internally. Can this and the ntp settings combined, be a problem when someone from, lets say, South Africa works with IcingaWeb2.


    Version:

    I'm not sure, to be honest, as I can only work on the system, when the customer lets me :D

    Last update of Icinga2 and IcingaWeb2 I did was around March '17


    If I can provied anything else, pls tell me.


    Thanks and greetings :)

    I haven't configured anything knowingly.


    SELinux isn't installed,

    AppArmor gives me this:


    Code: ls -lah /etc/icingaweb2/modules/businessprocess
    1. ls -lah /etc/icingaweb2/modules/businessprocess
    2. insgesamt 12K
    3. drwxr-sr-x 3 www-data icingaweb2 4,0K Mär 2 09:19 .
    4. drwxrwsr-x 6 root icingaweb2 4,0K Mär 2 09:19 ..
    5. drwxr-sr-x 2 www-data icingaweb2 4,0K Apr 27 13:35 processes




    Checkresults for removed "o" permissions:

    ERROR: Icinga\Exception\ConfigurationError in /usr/share/icingaweb2/modules/monitoring/library/Monitoring/Backend/MonitoringBackend.php:181 with message: No backend has been configured

    Your hostname has "agent" in it, also you configured your check to "Run on Agent".

    Have you installed the Icinga2 Agent on this host? (Icinga2 Documentation on Master/Satellites/Clients(Agents), Director Documentation on Agents)


    I assume you want to monitoring memory usage via snmp?

    Then you don't need the option "Run on Agent" = yes


    Have you configured all needed arguments for the plugin? Please post a screenshot of the "Preview" tab of the check command, the service template and your host.


    I recommend reading the Director documentation and/or this very good (but german) documentation: Icinga Director HowTo by django (use Google Translator to translate to english, it works pretty good)

    1. Your check command has to be a "Plugin Check Command" (Type)

    2. In your service you are using a different check command.

    To use your own memory check command that you created (check_abdouMemory) you need to set this check command in your service template (Check Command = Check_abdouMemory).

    Did you install pnp4nagios via packages or manually?

    Are the paths in the pnp module settings correct? (Configuration->Modules->pnp->Config)


    Afaik if you run Ubuntu 16.04 you need to install pnp4nagios manually and need to change some things compared to an Ubuntu 14.04 install:

    (path might differ)

    /usr/local/pnp4nagios/etc/config_local.php

    $views[]*

    comment those lines

    I'll push this up again.


    I had the time to set up a fresh Icinga2 install.

    Ubuntu 16.04.2 with all current versions:

    Icinga2 r2.6.3-1

    businessprocess 2.1.0 (git)

    director 1.3.1(git)

    Icingaweb2 2.4.1


    I also watched the latest webinar on the topic (Icinga2 BP Webinar) and made sure I set up the check command accordingly.

    One note on this: In the webinar the deprecated 'check process' instead of 'process check' is still used.


    My check command (with switched postitions for check and process (though that makes no difference)


    The output by my monitoring check:

    Permissions on /usr/bin/icingacli

    -rwxr-xr-x 1 root root 218 Jan 18 15:27 icingacli*

    Permissions on /etc/icingaweb2

    drwxrwsr-- 4 root icingaweb2 4096 Feb 14 16:34 icingaweb2/

    had to add chmod o+r -R, otherwise the monitoring check would say that /etc/icingaweb2 is not accessible, even if nagios user was added to icingaweb2 group:icingaweb2:x:121:www-data,nagios


    What I find most curious though, if I run the checks on console as nagios user, it works flawlessly...

    Code
    1. root@st-isr01-dev:/etc# sudo -u nagios icingacli businessprocess process check --config Icinga2 Icinga2-Server
    2. Business Process OK: Icinga2-Server
    3. nagios@st-isr01-dev:/etc$ icingacli businessprocess process check --config Icinga2 Icinga2-Server
    4. Business Process OK: Icinga2-Server



    Any ideas what could be wrong in my setups?