Posts by Ruven.Graf

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

    Hello everyone,

    at the moment I am having some trouble with Plugins that source their information by using SNMPWALK.

    Here is the problem:

    Whenever I execute some kind of SNMPWALK command it always shows this:

    After that it return the expected snmpwalk resullt that is then further used in the plugin to create output.

    So what basically happens is that SNMPWALK tries to load all those mibs with every call and I am sure it didn't do that before.

    Is there any way to keep it from trying to read in those modules? I have not yet found a way to tell SNMP not to try include those modules.

    Thanks in advance.

    Greetings Ruven

    Hello everyone,

    I have a little question that I couldn't find a solution for when I googled.

    I am recently trying to automate the configuration of an icinga2-client.

    Everything seems to work fine, but I have a problem that stops me from finishing this thing.

    When I run my config scripts all the conf-Files in Icinga2 seem just fine, but in Icingaweb2 I can still only see:


    Endpoint ''" does not accept commands.

    When I run the node wizard manually it works like a charm.

    So hers is the point, the node wizard asks me to:

    - Accept config from master? [y/N]:

    - Accept commands from master? [y/N]:

    Both times "no" is the default, but I actually want it to do so.

    So the question is: Where do I set this if I don't/can't do it in the node wizard?

    Tl;dr: Where can I configure my Client to accept conf and commands from master? (Not using the node wizard?)

    thanks in advance,

    kind regards

    Hello everyone,

    I have a little question on the automated authentication for nagvis.

    I am using Icinga2 and have recently embedded the NagVis module.

    As authentication method I have used the ".htaccess" method from the NagVis-Site.



    A second way to "disable" the authentication is setting the variable REMOTE_USER via .htaccess file. Place a file called .htaccess with the following content in the NagVis root directory:

    1. SetEnv REMOTE_USER nagiosusername

    But I did use another Username, of course.

    I set the rights to this "public" user accordingly, so the users can work with nagvis, but here's my problem:

    As an admin who from time to time wants to create a new map I need to get admin access somehow.

    Since I am not really experienced with webservers in general I wondered if I can just set a link on the NagVis login mask and call it from my root index.html...

    That's just one scenario I imagined, any other way that works is fine, too.

    What would the best way to realize this this be?



    This is driving me crazy, I just can't find the reason why this doesnt work...

    Its throwing me out at this point:

    Is there any way to check if the cookie path works? Do I have to activate it somehow?

    Thsi is again how it looks... as it should in my opinion.

    Another thing, in the doc it say that I have to cofigure "ndomy_1" as backend, that doesn't really work for either, but thats a minor problem I guess since I can't even get in.

    Oh, and can someone move this thread to the modules category?

    // Edit: falls jemand nicht auf English antworten möchte, ihr könnt auch gern auf Deutsch oder Französisch antworten, Hauptsache ich krieg dieses Problem gelöst.

    This is what my zones.conf looks like:

    And this is what the Kickstart Wizard looks like when im tyring to ... well... kickstart it

    The password is inserted correctly before i click "import"

    // Edit: falls jemand nicht auf English antworten möchte, ihr könnt auch gern auf Deutsch oder Französisch antworten, Hauptsache ich krieg
    dieses Problem gelöst.

    Hello everyone,

    I just tried to install and activate the icinga2-Director-Module.

    Everything has worked fine so far, until I wanted to run the import in the Kickstart Wizard.

    It tells me: "Failed to load icinga_zone "master""

    Everything I have found on the web tells me that this was a known bug in version 1.0.0, but I am on 1.3.1.

    Is anyone by any chance familiar with this problem?



    Another thing i have noticed, is that when i call the servers "root-page": http://FQDN

    It shows me this:

    And it doesn't matter which one im opening, they all promt a login except (who wouldve guessed) nagvis.

    It says: "not authenticated"

    What I also tried, was to log on from a newly installed computer to avoid any kind of stored cookies, but there was no success either...

    same message.

    This really has got to do something with the cookie path, doesn't it?

    What I have also noticed, is that the nagvis directory is the only one with different permissions on it in the /usr/local/ dir:

    Everything in it belongs to www-data (the apache user)

    Hello everyone,

    as I was to trying to set up NagVis in Icingaweb2 I encountered a problem.

    When I try to enter NagVis it tells me "not authenticated".

    When configuring NagViz I sticked closely to the specified README:

    I have already discovered threads with people having the same problem but they are usually able to fix it, by doing nothing more than adding the infamous "[cookie]"-Section to their icingaweb2 config.ini.

    I am a little lost at the moment, because I am not sure what else I can try. I have already deleted the cookies and tried it on several browsers, but there is still something wrong obviously.

    Maybe someone does have an example Config he would like to share or something similar...

    If you need any further information, let me know.

    Thanks in advance,

    Ruven Graf

    So this is how the debug.log looks, parameters are missing as guessed:

    1. [2017-01-18 09:30:30 +0100] notice/Process: Running command 'C:\Windows\System32\cscript.exe //NoLogo "C:\Program Files\ICINGA2\/sbin\check_folder_size.vbs"': PID 3932

    So I knew where to look and I have found the problem and it was of course a pretty obvious one.

    The checkcommand on the Icinga-master was different then the one found on the windows host... thats it.

    Works now.

    Thank you for your help.

    Have a nice day.

    Good Morning,

    thanks for the answers.

    Concerning the weird looking parameter definition, this actually works.

    Still, I changed it accordingly to the way you have mentioned to cut out all possible reasons for my problem, unfortunately it didn't change anything.

    This is how it looks now:

    Just to get a little bit more into detail:

    My script checks the size of a directory/folder, in order to do that it needs 3 parameters:

    1. The folder to check

    2. the warn-threshold

    3. the critical-threshold

    If one of those 3 is missing the script will return "UNKNOWN"

    The script:

    The other to specified parameters are modifiers for the cscript.exe that actually have nothing to do with the vbs itself:

    1. //Nologo changes the output of the script so it only show whats necessary.

    2. Is the path to the actual scriptfile, self explaining.

    So it seems like the modifiers are used correctly because the script is succesfully executed and has "nologo".

    My Problem is now that I dont have the slightest idea why it doesn't use the other 3 parameters and I am not even sure if it does even use one or two of them.

    But what I found out is that it can't use all 3 parameters, because I included the output of the input array of the script and it returns an outofbounds-error, so I am positive that icinga is obviously not passing over the three parameters.

    The easiest way out would probably be to find a plugin that does what I want that is an .exe file.

    Hello everyone,

    today I encountered a little problem concerning the definition of a new checkcommand.

    The plugin I wanted to use is a vbs. The plugin is being used on a Windows-Host.

    So I tested the VBS itself in the CMD and everything was fine, the plugin did what I was looking for.

    I then proceeded with writing the checkcommand for the plugin which looks something like this:

    (skip_key was used because the script doesn't have any letters in front of the parameters.)

    Then I got on with the according service:

    1. object Service "checkdir_C_Temp" {
    2.   import "generic-service"
    3.   check_command = "filesystemcheck"
    4.   vars.dirsize_dir = "C:\\temp"
    5.   vars.dirsize_crit = "6500"
    6.   vars.dirsize_warn = "6000"
    7.   command_endpoint = "myServer"
    8.   host_name = "myServer"
    9. }

    Ok, so far so good.

    First things first, the mere execution of the plugin is working fine. It delivers a result, the problem is, the result is always "UNKNOWN" which appears when the plugin doesn't have all the parameters specified when executed.

    It seems like the parameters are being skipped and the service doesn't hand them over to the command.

    Is there anything I might have missed?

    thanks in advance.


    erstmal danke für die schnelle Antwort.

    In der Tat ist es so, dass wenn ich die Config unter "/etc/icinga2/conf.d..." auskommentiere, es funktioniert.

    Unter dem anderen Verzeichnis bringts ja auch herzlich wenig, da das ja bei jedem Start neu gezogen wird.

    Hab den Fehler aber gefunden,

    ich habe mein eigenes Notofication-Command gemacht, dabei natürlich als Ausgang die commands.conf aus der Standard-Config übernommen und schlauerweise das Teil zwei mal kopiert, somit wurde das objekt wirklich redefined, da das ja dann ein identisches File war.

    Danke für deine Hilfe!

    Hallo Foren-Gemeinde,

    ich habe seit ein paar Tagen ein Problem mit Icinga2 und kam bis jetzt nicht auf eine sinnvolle Lösung.

    Und zwar kann ich Icinga2 nicht mehr reloaden, restarten oder eben starten.

    Beim start bringt er mit in "/var/log/icinga2/startup.log" folgenden Fehler:

    Code: startup.log
    1. information/cli: Icinga application loader (version: r2.5.4-1)
    2. information/cli: Loading configuration file(s).
    3. critical/config: Error: Object 'mail-host-notification' of type 'NotificationCommand' re-defined: in /var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf: 3:1-3:51; previous definition: in /etc/icinga2/conf.d/commands.conf: 3:1-3:51
    4. Location: in /var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf: 3:1-3:51
    5. /var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf(1): /* Command objects */
    6. /var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf(2):
    7. /var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf(3): object NotificationCommand "mail-host-notification" {
    8. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    9. /var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf(4): import "plugin-notification-command"
    10. /var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf(5):

    Ich nehme mal an, dass das an dem automatischen Verteilen der config liegt, wie ich es bei der Client-Installation angehakt habe. (Der Client hat übrigens Ubuntu 14.04, Windows Clients haben das Problem nicht.)

    Der Witz ist jedoch, dass das die definitionen der objekte die er anmault eigentlich identisch sind, sieht nicht so aus als sei da etwas editiert worden.

    Was ich auch getestet habe ist folgendes:

    Zuerst löscht man die Datei unter "/var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf" und startet dann Icinga dann geht alles.

    Das Problem ist nur, das die Datei nach dem Start wieder da ist.

    Was ich mich hauptsächlich Frage ist, wie kommt die Config in "/var/lib/icinga2/api/zones/global/_etc/checkcommands/commands.conf" zustande?

    Kleiner Edit:

    Mein Master Server hat die selbe Datei und auch Ubuntu 14.04, der scheint jedoch kein Problem damit zu haben, da geht alles...

    Vielleicht könnt ihr mir ja ein bisschen aushelfen.

    Vielen Dank schonmal im Voraus.

    Viele Grüße

    Also, ich habs...

    ich hatte mich mit Herrn Wahl kurzgeschlossen und wir konnten ein paar Sachen ausschließen.

    Woran es letztendlich lag kann ich nicht sagen, es hat einfach plötzlich getan, also kanns ja nur ein blöder kleiner Fehler gewesen sein.

    Aber was ich weiß bzw. was ich getan habe, was hilfreich sein könnte:

    Ich habe für den Icinga-Server den FQDN benutzt und habe ihm gesagt er soll KEINE Cookie authentifizierung nutzen, das wars dann auch schon.

    Vielleicht hilft das ja jemand.

    Wie immer danke für die Hilfe.

    Viele Grüße