Posts by log1c

    You are right, I forgot them in my previous post.

    Adding them doesn't change the output though :(

    Code
    1. root@ST-MON2-RZ:/etc/icinga2/conf.d# icingacli businessprocess process check "Citrix-Umgebung" --details --colors --state-type hard
    2. Business Process OK: Citrix-Umgebung



    edit:

    running the same command with sudo -u nagios give me the error message:

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

    I don't understand why. The nagios user is the default user for icinga2 when installed on Ubuntu. I also added the user to the icingaweb2 group.

    Thanks for the screenshot!

    I got it to work now, though I don't get the fancy colors and detailed output.

    Code
    1. root@ST-MON2-RZ:~# icingacli businessprocess process check Citrix-Umgebung
    2. Business Process OK: Citrix-Umgebung
    3. root@ST-MON2-RZ:~# icingacli businessprocess process check "LAN-OL" --config "Core Infrastruktur OL"
    4. Business Process OK: LAN-OL

    So the "node" (1st argument after "check") is the root of the business process tree, which then can have several sub-nodes.

    I mostly need the --config because I have some node names multiple times in different BPs.


    Will need to adjust some of my processes, so that they have one single root node.


    Both my check commands still return ERROR: There is no such module or command: 'businessprocess'

    Hi dnsmichi,


    could you show me your BP and the syntax of your successful query?

    Maybe I can rebuild your bp or apply it to my setup and understand what the icingacli businessprocess expects as "node" and "config".


    Thanks for your help!



    #EDIT:

    I previously had the service check bound to the wrong host, so that the check told me, that icingacli could not be found (duh)

    Now I got the check to "work", kind of...


    I tried the "original" check command icingacli-businessprocess as well as created a new one "businessprocess".

    Both commands get the two variables icingacli_businessprocess_process and icingacli_businessprocess_config.


    The new command looks like this (due to the deprecated warning with icingacli businessprocess check process:

    This is the applied service:

    Code
    1. apply Service "BusinessProcess-Citrix" {
    2. import "BusinessProcess-Check"
    3. assign where match("st-mon2-rz*", host.display_name)
    4. vars.icingacli_businessprocess_config = "Citrix"
    5. vars.icingacli_businessprocess_process = "Citrix-Umgebung"
    6. import DirectorOverrideTemplate
    7. }


    And the surprising output is (no matter which of the two commands I use:

    Code
    1. ERROR: There is no such module or command: 'businessprocess'
    2. USAGE: icingacli [module] <command> [action] [options]
    3. [...]

    What the hell?! :D

    cat /etc/icingaweb2/modules/monitoring/backends.ini

    Code
    1. [icinga]
    2. type = "ido"
    3. resource = "icinga_ido"


    cat /etc/icingaweb2/resources.ini

    Sadly it doesn't...

    Code
    1. sudo -u nagios /usr/bin/icingacli businessprocess check --config "E-Mail" process "E-Mail-Versand und Empfang"
    2. ERROR: Icinga\Exception\ConfigurationError in /usr/share/icingaweb2/modules/monitoring/library/Monitoring/Backend/MonitoringBackend.php:181 with message: No backend has been configured

    also tried:

    Code
    1. sudo -u nagios /usr/bin/icingacli businessprocess process check --config "E-Mail" "E-Mail-Versand und Empfang"
    2. ERROR: Icinga\Exception\ConfigurationError in /usr/share/icingaweb2/modules/monitoring/library/Monitoring/Backend/MonitoringBackend.php:181 with message: No backend has been configured

    I don't get why this check tells me that no backend is configured when I clearly have one, because I have a working monitoring system?! Or do I need to configure something else additionally?

    Hi fugstrolch,


    thanks for your input!


    E-Mail is on of my business processes.

    In this process I configured the node "E-Mail-Versand und Empfang" and this has four "sub-nodes" BYOD, Clearswift, Exchange-Sevrer, Kemp-Loadmaster which than hold the specific hosts.


    If I run your line, this is the outcome:

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


    I find that quite strange, as the monitoring itself works, and a backend is configured.

    I can't get it to work.

    Apart from that the service check tells me that "usr/bin/icingacli" can't be found: execvpe(/usr/bin/icingacli) failed: No such file or directory though it's there and i can use it from the command line, the bigger problem is, that I can't figure out what input the check/cli command expects.


    List of my business processes:

    sudo -u nagios /usr/bin/icingacli businessprocess process list

    What command exactly do I need to check the status of one of these?


    I tried sudo -u nagios /usr/bin/icingacli businessprocess process check --process E-Mail and get and ERROR: Exception in /usr/share/icingaweb2/modules/businessprocess/library/Businessprocess/BpConfig.php:518 with message: The node "" doesn't exist

    What is expected for the argument "node"? Is it an Icinga2 node (icinga2 node list?)?

    It doesn't seem to be a node from the business process.


    See attached image and following code for email bp config.


    Tried this:


    Code
    1. sudo -u nagios /usr/bin/icingacli businessprocess process check --process E-Mail "E-Mail-Versand und Empfang"
    2. ERROR: Exception in /usr/share/icingaweb2/modules/businessprocess/library/Businessprocess/BpConfig.php:518 with message: The node "E-Mail-Versand und Empfang" doesn't exist


    Can you point me in the right direction?

    Hi all,


    is it possible to send notifications if a business process gets critical/fails?

    I could not find any option for this in Director, maybe it's possible via Icinga2 directly?


    I'm using Icinga2 2.6.2, Icinga Director 1.3.1 and BusinessProcess 2.1.


    Best regards,


    Logic

    D.h. du hast dem Check-Plugin keine OID zum überprüfen mitgegeben


    Beispiel eines Checks: ./check_snmp -H <hostadresse> -o .1.3.6.1.2.1.1.3.0 -C <snmp_community -P <snmp_version> --label "UPTIME"


    Du solltest im Director unter "Kommandos" auch das Kommando "snmp" haben (mit einem "Pin" marktiert und nicht änderbar). Sollte so aussehen:


    Hier fügst du jetzt die zu verwendeten Felder hinzu, bspw:

    snmp_community
    snmp_crit
    snmp_label
    snmp_oid
    snmp_units
    snmp_version
    snmp_warn


    Dann hast du diese Felder in deinem Service angezeigt und kannst sie mit den richtigen Werten füllen.

    snmp_oid mit der zu prüfenden OID, snmp_community mit der SNMP Community usw.

    Hi all,


    I'm currently using Icinga2 2.6 with Icinga Director 1.3 and would like to reproduce dependencies in my monitoring environment, somehow.

    I know that the Director already has a feature request opened for this, but I'd still like to ask if this is already possible by a work-around?


    I have seen this thread Icingaweb2 Director und Parent Hosts were TomGelf posted a screenshot where there are dependencies listed in the config and the module "fileshipper" is mentioned. Are these dependencies solely configured via external files and imported with fileshipper?


    Or is there a way to configure the dependencies by hand (via normal icinga2 config?).

    What would happen if I do it this way or even edit the config written by the Director? Would these changes be preserved or do they get lost when the Director deploys a new config?

    Or would it even crash my config?


    I hope someone can get me started on how to implement dependencies for my current setup.



    Greetings and have a nice weekend!

    You seem to have active checks disabled and passive checks are not provided because the target is down.

    Active checks are enabled now. The problem with this host is, that it can't be pinged, so the host itself has a passive check that is always OK as host check.

    As long as the server (Azure machine) is online and NSClient is running, the checks are OK and get updates. In case the server goes down or can't reach the monitoring host the check should switch to a CRTICAL state given by the dummy command.

    Thanks for your answers!

    That goes for active checks.

    But with passive (only) checks that are late, the last known state is displayed, the red marker is turned on and the time increases.

    An active check is not triggered then.

    So the "next update" time is not directly linked to the threshold for the check result freshness mentioned in the documentation?

    enable_active_checks = true is required.

    Will try and give an update on the outcome.


    edit:

    Stopped the NSClient on the remote host and the service is still "OK" (see picture), though there was no update on the check for 30 minutes.

    Hi all,


    I'm having problems with some passive services, that get their input via NSCA from a Windows Host (NSClient++).


    As long as NSClient or NSCA-Server is running, everything is working fine. The checks get their input (e.g CPU usage or Mem usage) and get updated.


    The service looks like this:


    Now to my problem:

    If I stop NSClient or NSCA Server (to simulate that the host is down) the checks don't switch to CRITICAL state after some time.

    The "last update" time counts up, the dot gets red, the "next update" time goes "negative". (see picture)


    If I understand this (check-result-freshness) correctly:

    1. the "next update" time for my check should be 5 minutes (time when a check result is expected). Instead it always says 10 minutes, when a check result was received.

    2. If the "next update" time reaches 0 the check command should be triggered and set the service to CRITICAL

    Is that correct?


    NSCA Client on the Windows host is configured to schedule the checks every 4 minutes.


    Can anyone help?


    Best regards,


    Logic

    Es funktioniert jetzt, mit folgender Config:


    zones.conf auf dem Satellit:


    Endpoint-Objekt entfernt, endpoint für Zone "Satellit" von "nodeName" in "st-isr04-dev" umbenannt


    Anschließend Zone und Endpounkt im Director wieder konfiguriert.


    Problem ist scheinbar, dass der Director/oder icinga2 die Zonen nicht synct?!

    Aber es kann doch nicht sein, dass ich die Konfig, die mir der Node Wizard angelegt hat, ändern muss und diese Änderungen anschließend in den Director wieder einfüge?!


    Den Endpoint habe ich jetzt testweise im Director entfernt, so dass icinga2 auf dem Satellit wieder startet. Bringt aber, wie zu erwarten, nix.