Quantcast
Channel: Monitoring-Portal Feed
Viewing all 1338 articles
Browse latest View live

USV Netzwerkkarte: CS121 seltsames SNMP Problem

$
0
0
Hallo,

ich monitore zwei Riella-USP USV-Anlagen die je eine Generex CS121 Netzwerkkarte haben mit OMD 1.30 und check_mk 1.2.6p12.

Wenn ich in check_mk mit WATO einen neuen Host anlege wird sofort erkannt dass SNMP verfügbar ist, egal welche community ich angebe!
sysDescr: CS121 v 5.18.7
sysContact: xxxxx
sysName: USV1
sysLocation: xxxxx
(die richtigen Angaben hab ich durch xxxx ersetzt)

Egal welche Community ich angebe, diese Infos werden korrekt ausgelesen! (in allen 3 SNMP Fenstern in WATO, also SNMPv1, SNMPv2c und SNMPv2c (without Bulkwalk))
Aber leider hab ich dann immer einen Fehler bei der check_mk disvovery Funktion erhalten. Ich kann den Fehler leider nicht mehr reproduzieren aber da stand glaube ich
"UNKNOWN - SNMP Error on check_mk Timeout: No Response from xxx.xxx.xxx.xxx ..." (xxx = IP)

Jedenfalls ging dann nur PING und dieser Fehler beim Discovery-Service.

Nach langem Recherchieren und Probieren habe ich jetzt in WATO bei den Host-Einstellungen die Art des Agenten von "SNMP" auf "Alte SNMP Geräte(mit Version 1)" gestellt und sofort wurden vom Dicovery Check 16 neue Services erkannt!

Ist doch irgendwie sehr seltsam? Die CS121 unterstützt sogr SNMPv3!
Und bei der zweiten USV habe ich spaßeshalber die Communiy absichtlich falsch eingestellt, da wurden zwar auch diese Services erkannt nachdem ichs auf SNMPv1 gestellt habe, allerdings blieben die permanent auf "Pending" stehen. Erst als ich die Community dann richtig eingestellt habe wurden alle Werte richtig ausgelesen.

Die CS121 Netzwerkkarte ist in vielen USV's im Einsatz, hat evtl jemand Ähnliches festgestellt ?

Master Verbindung zum Client

$
0
0
Hallo @all,

ich habe meinen Icinga 2 Monitoring Master bei mir zu Hause auf meinem Homeserver.
Normalerweise verbinden sich die Clients ja zum Master, aber kann man den Spieß auf für einen speziellen Endpoint umdrehen? Ich würde gerne, dass mein Master sich zu meinem Root-Server verbindet, da andersrum die Verbindung wegen der privaten IP Adressen nicht funktioniert.

Danke,
~velo

Wagner Titanus Brandfrühsterkennung Überwachung

$
0
0
Hallo,

möchte mit check_mk (1.2.6p12) eine Wagner Titanus "Racksense 1HE" und "Microsense" überwachen. (Brandfrühsterkennung, Raucherkennung für Serverschränke etc ) Auf Anhieb hat mir die Discovery Funktion leider nur SNMP Info und Uptime ausgespuckt.

Es gibt aber in check_mk diverse checks für Titanus "TopSense" (hier in der Liste, bei "WAGNER Group" https://mathias-kettner.de/checkmk_check_catalogue.html)


Kann ich irgendwie überprüfen ob sich evtl damit auch etwas auslesen lässt oder wäre das dann bereits automatisch erfolgt?

Oder hat jemand auch solche Geräte bei sich im Monitoring und weiß wie ich an mehr Daten komme?

Using Nagastamon with Icinga Web 2

$
0
0
I'm a n00b currently setting up an Icinga instance, the sysadmin has used Nagastamon (https://nagstamon.ifw-dresden.de/) for some time and essentially treats it as the frontend to Icinga.

I have installed Icinga 2 and Icinga Web 2, however it doesn't look like Icinga Web 2 exposes a 'Monitor CGI' directory like the original Icinga web interface does, e.g.:

http://<hostname>/icinga/cgi-bin

Is there something equivalent in the Icinga 2 site installation? Am I missing something here?

Thanks

Wrong undertanding a config architecture

$
0
0
Hello everyone.
I have some questions about config architecture. Does someone can show a simple config for localhost monitoring?

As i understand in hosts.conf i need to describe all hosts with options, and choose at least one template like "generic-host". But in template "generic-host" i can see a check_command - hostalive. And where is description of "hostalive".

I can't uderstand what config is main and includes all another. Please try to explain this to me.


Thank you very much.

zero footprint - checking service on windows with snmp

$
0
0
Hi,

Is there a way to monitor specifik services on a windows server by snmp?

//Regards KB-always

NSClient + Nagios

$
0
0
Hallo,

wir haben bei uns Nagios (mit Centreon als GUI) im Einsazu und möchten nun unsere Citrix Lizenzen überwachen. Dafür habe wir eine batch Datei erstellt welche auch Grundsätzlich funktioniert. Das Problem ist diese batch Datei nun in NSClient entsprechend einzubinden damit Nagios/Centreon den Status abfragen kann. Das ganze wird derzeitig versucht per nrpe zu realisieren.
Bevor wir unser eigenes Script einbauen versuchen wir nun das mitgeliferte check_ok Script abzufragen. Leider bekommen wir dabei immer eine Fehlermeldung:

Quellcode

1
2
root@proxy:~# /usr/lib/nagios/plugins/check_nrpe -H <IP> -c citrix_free_lic
No handler for command: citrix_free_lic


In der nsclient.ini sind externe Scripts erlaubt, und wir haben das Script wie folgt verlinkt:

Quellcode

1
2
3
4
5
6
CheckExternalScripts = 1
/settings/external scripts/wrappings]

; BATCH FILE WRAPPING - 
;;bat = scripts\\%SCRIPT% %ARGS%
citrix_free_lic = scripts\check_ok.bat


Generell scheint NRPE ja zu laufen denn ein anderer Check läuft ohne Probleme:

Quellcode

1
2
root@proxy:~# /usr/lib/nagios/plugins/check_nrpe -H <IP> -c checkServiceState -a CheckAll "exclude=Acronis VSS Provider" "exclude=SysmonLog" "exclude=VSS" "exclude=clr_optimization_v4.0.30319_32" "exclude=clr_optimization_v2.0.50727_32" "exclude=clr_optimization_v4.0.30319_64" "exclude=ShellHWDetection" "exclude=sppsvc" "exclude=TrustedInstaller" exclude="RemoteRegistry"
OK: All services are in their appropriate state.


Was haben wir übersehen bzw. falsch gemacht?

ORA-12154 - TNS Fehler mit check_oracle

$
0
0
Hallo zusammen,

ich habe aktuell Probleme mit der Oracle/check_oracle.
Kurz zu meinen Setup:

Ubuntu 15.04 (x64)
Icinga2 r2.3.11-1
Oracle Instant Client 12.1 (sqlplus,devel,basic)

Mit der cl:

Quellcode

1
/usr/lib/nagios/plugins/check_oracle --login --sid meineDB

bekomme ich immer folgende Meldung:

CRITICAL - ORA-12154: TNS:could not resolve the connect identifier specified

Meine /etc/profile.d/oracle.sh sieht folgend aus:

Quellcode

1
2
3
export ORACLE_HOME=/usr/lib/oracle/12.1/client64
export TNS_ADMIN=/usr/lib/oracle/12.1/client64/network/admin
export PATH=$PATH:$ORACLE_HOME/bin


Eine tnsnames.ora ist unter /usr/lib/oracle/12.1/client64/network/admin/tnsnames.ora vorhanden und es gibt einen Eintrag für "meineDB"

Habe ich etwas übersehen ich kann mir gerade den TNS Fehler nicht erklären.

Neuer check in check_mk : no such checktype

$
0
0
Ich habe einen vorhandenen check im Ordner "~/mysite/share/check_mk/checks." kopiert, ein paar OIDs geändert und wollte den check jetzt ausprobieren.
Der check ist ursprünglich für ein anderes Gerät vom selben Hersteller, daher sind die OIDs kaum anders und ich hoffe dass auch die checks so funktionieren.

Wenn ich aber den check ausprobieren will mit;
check_mk -v --checks wagner_titanus_microsens -I Microsens
Kommt die Meldung:
-> Failed: No such check type 'wagner_titanus_microsens'

wagner_titanus_microsens ist der Dateiname, die Datei ist auch vorhanden! Andere checks kann ich damit aufrufen.

Muss ich beim abspeichern der Datei irgendwas beachten?
Die Besitzer und Berechtigungen habe ich genau so eingestellt wie die anderen checks. Daran liegts also nicht.
Achja bevor ichs vergesse: check_mk Version 1.2.6p12 mit OMD 1.30

Event Log leer

$
0
0
Hallo,

im Thruk bekomme ich keine Event Log Einträge angezeigt.
Muss in irgendeiner Config der Pfad zur Logdatei angegeben werden?

Folgende Versionen haben wir im Einsatz:
Thruk: Version 2.02
Icinga r2.3.11-1

Gruß
Patrick

[Multisite] Ansicht von Hosts einschraenken

$
0
0
Ich habe eine Multisite Installation mit 2 check_mk Instanzen an unserem Standort und 1 Instanz an einem anderen Firmenstandort. Die externe Instanz kann auf eine Instanz bei uns zugreifen, um Hosts und Services zu sehen, die für Sie ebenfalls von Interesse sind (ESX Hosts, Serverfarm). Leider bekomme ich es nicht hin, dass ihnen alle anderen Hosts und Services, die sie nicht interessieren, nicht angezeigt werden. Konfiguration sieht so aus (nur ein Ausschnitt):

Zitat


[multisite.mk]
users = [ "omdadmin" ]

[main.mk]
host_contactgroups += [
( 'ext-group', [ "srvext" ], ALL_HOSTS ),
( 'int-group', [ "win" ], ALL_HOSTS ),
( 'int-group', [ "sol" ], ALL_HOSTS ),

]
service_contactgroups += [
( 'ext-group', [ "srvext" ], ALL_HOSTS, '' ),
( 'int-group', [ "win" ], ALL_HOSTS ),
( 'int-group', [ "sol" ], ALL_HOSTS ),

]

contactgroup_members["ext-group"] = [ "omdadmin" ]
contactgroup_members["int-group"] = [ "nagiosuser" ]

all_hosts = [
'win2k|tcp|snmp|snmpv2c|win|srvext',
'linux|tcp|snmp|snmpv2c|srvext',

'solaris|tcp|snmp|snmpv2c|sol',
]

Nach meinen Erwartungen sollten die User der Gruppe "int-group" die Hosts win2k und solaris sehen (wegen Tags win und sol), aber nicht den Host linux (da kein passendes Tag vorhanden). Das funktioniert auch! Die Gruppe "ext-group" sollte aber nur die Hosts win2k und linux sehen (wegen des Tags srvext), aber nicht den Host solaris (da kein passendes Tag vorhanden). Tatsächlich sieht die Gruppe "ext-group" aber alle Hosts, obwohl ich das eben nicht will. Wo liegt der (Denk)fehler?

Frühere Checkwerte mit einbeziehen

$
0
0
Hallo Monitoring-Portal-Community,

schön, dass ihr hier aktiv seid.
Ich bin Neuling und habe trotzdem bereits mit Hilfe der Check_MK-Doku einen agent based check schreiben können.

Im Moment ändert sich der Status auf WARN, wenn ein Schwellwert überschritten wird.
if WERT > warn:
return 1, ...


Ich würde das aber gerne insofern ändern, dass er geändert wird, wenn alle Werte, von vor einer Stunde (die letzten 60 checks) kleiner/gleich dem Wert des aktuellen Checks sind.
if WERT[-60:now] >= WERT:
return 1, ...
Also, wenn z.B. der Speicher in der letzten Stunde stets monoton wachsend war.

Wie man sicher erkennen kann, weiß ich nicht, wie man an die alten Checks herankommt.
Ist das überhaupt möglich? Wenn ja, wie komme ich an sie heran?

Ich bin gerne dazu bereit Dokus o.ä. zu studieren, allerdings müsste ich sie erst einmal finden.
Hier im Forum habe ich etwas von Trendanalyse gesehen, aber das hat mir (noch) nicht weitergeholfen.
Das liegt sicher nicht zuletzt daran, dass ich die Materia noch nicht so in der Tiefe erschlossen habe.

Ich bin für jeden Tipp dankbar.
Gruß :)

Gadget PHP + Javascript

$
0
0
Dear Forum
Gadget doing great with PHP script, however I found not executing JavaScript.
me doing wrong or not possible? appreciate your comments.
<?PHP
echo "<script>document.write('Java in PHP')</script>";
?>
Regards

Importing configuration files from Icinga 1.x to 2.x

$
0
0
Hello






I
am building an Icinga 2 server, and my plan was to import the configuration files
that were used in our Icinga 1.x server to hopefully


facilitate
the configuration of the new system. The files in question are the following:



"checkcommands.cfg


hostgroups.cfg


misccommands.cfg


notifications.cfg


printer.cfg


contactgroups.cfg


services.cfg


contacts.cfg


hosts.cfg


switch.cfg


templates.cfg


timeperiods.cfg


hostextinfo.cfg


windows.cfg


commands.cfg


localhost.cfg


cgiauth.cfg


htpasswd.users


icinga.cfg


ido2db.cfg


cgi.cfg


idomod.cfg


resource.cfg
"


I
had several questions concerning this:





1. Are these files still used
in the Icinga 2 release?


2. Is simply placing the above
files in the appropriate locations sufficient to complete the import, or is
there another


method or utility that can do this?





Thanks.

Komisches Verhalten nach Serverupgrade beim Checken mit LM-Sensors

$
0
0
Hallo Leute,

ich hab hier ein sehr seltsames Problem. Ich hab meine Proxmoxserver von Debian7 auf 8 upgegradet. Nun funktionieren die NRPEchecks vom Nagiosserver (Gentoo) aus mit LM-Sensors nicht mehr richtig. Nagiosversion habe ich hier die 4.1.1.

Ich prüfe vom Server aus die Temperatur des Netzteils:

Quellcode

1
2
./check_nrpe -H meinhost.local -c check_sensors_auxtin
LM_SENSORS UNKNOWN - Error while executing /usr/sbin/hddtemp -n /dev/sda:


Super, jetzt möchte der Befehl die Festplatte SDA checken, wie witzig ist das denn? Den das nette im check_sensors_auxtin auf dem Host steht folgendes drinnen:

Quellcode

1
command[check_sensors_auxtin]=/usr/lib/nagios/plugins/check_lm_sensors  -h "AUXTIN"=35,40


Und wenn ich es direkt auf dem Host ausführe, sieht es so aus:

Quellcode

1
2
/usr/lib/nagios/plugins/check_lm_sensors  -h "AUXTIN"=35,40
LM_SENSORS OK - AUXTIN=23.5|AUXTIN=23.5;35;40;;


Also passt's! Ich habe dann auch einige Checks gemacht ob wohl die richtige Zeile verwendet wird. Und ja sie wird verwendet. Und nein, nirgends in ganz /etc kommt "/usr/sbin/hddtemp -n /dev/sda" vor. Also was könnte hier nicht passen, woher nimmt check_nrpe die Info.

lg und Danke

Advice on installing Nagios (OMD) on a Raspberry Pi

$
0
0
Hello Forum,

I've found this blog that gives instructions on running OMD on a Raspberry Pi - http://www.vdsar.net/raspberry-pi-open-m…/#comment-28511.

Would anyone care to share with me what setup I would need in order to make this work? What size SD card should I consider using if running OMD from a Pi SD card. Would I kill the SD card with all the writes that will occur from monitoring my network?

Any advice you can provide would be appreciated.

Thanks.

check_mk: how to use 'Tomcat number of threads in thread pool' plugin

$
0
0
Hallo,

I'm new to check_mk and the complete monitoring stuff. Last night I installed check_mk successfully :) .


So I can monitor 2 external hosts and their standard services. Both hosts have a tomcat running and I want to include the tomcats into my monitoring.


On catalog of check plugins of check_mk I find one page titled 'Tomcat number of threads in thread pool'. But I have no idea, how to use the information on the page. :(

https://mathias-kettner.de/checkmk_check…metrics.tp.html

peter
Dortmund

Check command "apt" is checking different host, but why?

$
0
0
Hi all,

i am a newbe - please be keen.

I added some hosts (13 virtual machines). The hostalive checks for all of these hosts are fine (as expected).
Now i try to add a "apt" check to my atlassian confluence virtual machine ("vm_confluence").

object Host "vm_confluence" {
import "generic-host"
address = "192.168.1.133"
vars.os = "Linux OS"
vars.sla = "24x7"
vars.production = "yes"
}
object Service "aptmytest" {
host_name = "vm_confluence"
check_command = "apt"
}

After realoading the config i can see this host and two services of it ("ping4" and "aptmytest"). Ok.
But "apt" is returning:

"APT OK: 0 packages available for upgrade (0 critical updates). "

This is true for "vm_icinga" but not for "vm_confluence" (should be "66"). I guess icinga is checking vm_icinga instead of vm_confluence.

What am i doing wrong? Where can i see what "aptmytest" is doing (command line execution and output)?

Best regards
Daniela

Cluster Probleme - Error while sending JSON-RPC message

$
0
0
Hallo,

ich habe ein dringliches Problem. Und zwar habe ich vor einiger Zeit meinen ICINGA Server gewechelt und dabei vergessen die CA mit zu kopieren. Da ich die *.crt Files hatte konnte es damals weiter laufen. Entsprechend konnten nur keine neuen Zertifikate angelegt werden.

Also habe ich heute via
icinga2 pki new-ca
eine neue angelegt und das Zertifikate für den Master erstellt. Dann auch mal testweise für einzelne Nodes.

Leider bekomme ich seit dem folgende Meldungen

Quellcode

1
2
3
4
5
6
7
8
9
[2015-11-01 23:28:00 +0100] information/ApiListener: New client connection for identity 'icingabieg'
[2015-11-01 23:28:00 +0100] critical/ApiListener: Client TLS handshake failed.
Context:
        (0) Handling new API client connection

[2015-11-01 23:28:00 +0100] information/ApiListener: Syncing global zone 'global-templates'.
[2015-11-01 23:28:00 +0100] warning/ApiClient: Error while sending JSON-RPC message for identity 'icingabieg'
[2015-11-01 23:28:00 +0100] warning/ApiClient: API client disconnected for identity 'icingabieg'
[2015-11-01 23:28:00 +0100] warning/ApiListener: Removing API client for endpoint 'icingabieg'. 0 API clients left.


An sich ja recht eindeutig, dass er Probleme mit dem Zertifikat hat. Allerdings ist ein openssl verify erfolgreich

Quellcode

1
2
openssl verify -verbose -CAfile ca.crt icingabieg.crt
icingabieg.crt: OK


Der Name ist in den jeweiligen constants.conf und zones.conf richtig geschrieben.

Auf dem Client steht im Log

Quellcode

1
2
3
[2015-11-01 23:39:18 +0100] warning/ApiClient: Error while sending JSON-RPC message for identity 'master'
[2015-11-01 23:39:18 +0100] warning/ApiClient: API client disconnected for identity 'master'
[2015-11-01 23:39:18 +0100] warning/ApiListener: Removing API client for endpoint 'master'. 0 API clients left.


Eingesetzt wird die neuste stable ICINGA Version

Quellcode

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
icinga2 -V
icinga2 - The Icinga 2 network monitoring daemon (version: r2.3.11-1)

Copyright (c) 2012-2015 Icinga Development Team (https://www.icinga.org)
License GPLv2+: GNU GPL version 2 or later <http://gnu.org/licenses/gpl2.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

Application information:
  Installation root: /usr
  Sysconf directory: /etc
  Run directory: /run
  Local state directory: /var
  Package data directory: /usr/share/icinga2
  State path: /var/lib/icinga2/icinga2.state
  Objects path: /var/cache/icinga2/icinga2.debug
  Vars path: /var/cache/icinga2/icinga2.vars
  PID path: /run/icinga2/icinga2.pid
  Application type: icinga/IcingaApplication

System information:
  Operating system: Linux
  Operating system version: 3.16.0-4-amd64
  Architecture: x86_64
  Distribution: Debian GNU/Linux 8.2 (jessie)


Ich hoffe ihr könnt mir da weiter helfen. Grüße

Icinga Web 2: Kein Update mehr seit Mai

$
0
0
Hallo @all,

die Debian Jessie Repos von http://packages.icinga.org/debian/ sind das letzte mal im Mai geupdated worden.
Könnt ihr das bitte nachziehen, damit ich auch die neue Icinga Web 2 Version einspielen kann?

Danke,
~velo
Viewing all 1338 articles
Browse latest View live