Posts by monMonitor

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

    Currently I am trying to implement the great plugin by mmarodin (thx for this nice job) for notifications with perfdata.


    Unfortunately I am stuck here:

    Code
    1. [2017-12-11 18:27:21 +0100] warning/PluginNotificationTask: Notification command for object 'Windows-Test-SRV!Windows_Load' (PID: 643, arguments: '/etc/icinga2/scripts/mail-service-perfdata-notification.py' 'HOSTADDRESS' 'Windows-Test-SRV' 'HOSTALIAS' 'Windows-Test-SRV' 'HOSTDISPLAYNAME' 'Windows-Test-SRV' 'LONGDATETIME' '2017-12-11 18:27:21 +0100' 'NOTIFICATIONAUTHORNAME' 'XXX' 'NOTIFICATIONCOMMENT' 'test' 'NOTIFICATIONTYPE' 'CUSTOM' 'SERVICEDESC' 'Windows_Load' 'SERVICEDISPLAYNAME' 'Windows_Load' 'SERVICEOUTPUT' 'OK: 4 CPU(s) average usage is: 0.00% ' 'SERVICEPERFDATA' 'cpu0=0.00%;;;0;100 cpu1=0.00%;;;0;100 cpu2=0.00%;;;0;100 cpu3=0.00%;;;0;100 total_cpu_avg=0.00%;0:80;0:99;0;100' 'SERVICESTATE' 'OK' 'USEREMAIL' 'XXX@XXX.XX') terminated with exit code 1, output: Traceback (most recent call last):
    2. File "/etc/icinga2/scripts/mail-service-perfdata-notification.py", line 106, in <module>
    3. (LABEL,DATA) = PERFDATA.split("=")
    4. ValueError: need more than 1 value to unpack

    Lines 98 to 106 are looking like this:

    Code
    1. if (SERVICEPERFDATA or PANELURL):
    2. HTML += '\n</table><br>'
    3. HTML += '\n<table width=' + WIDTH + '>'
    4. HTML += '\n<tr><th colspan=6 class=perfdata>Performance Data</th></tr>'
    5. if (SERVICEPERFDATA):
    6. HTML += '\n<tr><th>Label</th><th>Last Value</th><th>Warning</th><th>Critical</th><th>Min</th><th>Max</th></tr>'
    7. PERFDATALIST = SERVICEPERFDATA.split(" ")
    8. for PERFDATA in PERFDATALIST:
    9. (LABEL,DATA) = PERFDATA.split("=")

    As far as I understand the Array "PERFDATALIST" should contain this elements:

    Code
    1. cpu0=0.00%;;;0;100
    2. cpu1=0.00%;;;0;100
    3. cpu2=0.00%;;;0;100
    4. cpu3=0.00%;;;0;100
    5. total_cpu_avg=0.00%;0:80;0:99;0;100

    with each of them containing the "=" used for splitting them.


    What do I miss here?

    After upgrading to 2.8.0-1 from 2.7.2-1 today I ran into the following issue trying to follow the documentation:

    Code
    1. mysql -p -u root < /usr/share/icinga2-ido-mysql/schema/upgrade/2.8.0.sql
    2. Enter password:
    3. ERROR 1046 (3D000) at line 18: No database selected


    If that is a known issue and is already answered elsewhere I deeply apologize but after trying to find a solution for abouth two hours I gave up -.-


    Luckily the monitoring, notifications and Grafana are still running it "just" affects IcingaWeb2.


    Further information:




    It does offer both '--verbose' as well as '--debug' which unfortunately don't result in any more information.


    I followed the same lead considering the probable size of the request.


    Still that doesn't make any sense to me at this point as it seems to be machine-independed (Working on most of 2012R2's).


    I'll try to narrow the problem down and keep my progress posted here.


    If anyone have had that problem before I owe you a beer if you help me solving this


    Thx dnsmichi for the contribution so far

    Doing it by using 'snmpget' is working but I assume that the Centreon-Plugin somehow uses 'snmptable' to acquire the values which then results in an UNKNOWN state.

    Code
    1. centreon_plugins.pl --plugin=os::windows::snmp::plugin --hostname 2012srv --snmp-community xxx --mode service --service='SAP***' --critical 1:
    2. UNKNOWN: SNMP Table Request : (genError) A general failure occured
    3. root@MON3SRV:~# snmpget -c xxx -v 2c 2012srv iso.3.6.1.4.1.77.1.2.3.1.1.9.83.65.80.66.86.84.95.53.48
    4. iso.3.6.1.4.1.77.1.2.3.1.1.9.83.65.80.66.86.84.95.53.48 = STRING: "SAP***"

    BTW turning SNMP off on the respective servers results in an SNMP-Timeout error if that is of any use to solve the problem.

    Hey there,


    I got a problem monitoring Windows-Service via SNMP.


    To do that I am using Centreon-Plugins which are working just fine.


    Some days ago some of the 2012R2 throwing this error:


    Code
    1. UNKNOWN: SNMP Table Request : (genError) A general failure occured


    and I have no idea what that means.


    We tried several things around the Servers:


    Disabling Firewalls

    Disabling AV

    Removing/Adding SNMP-Service

    VM/Non-VM Systems

    Different Patch States (they are all on the same patch level)


    Funny thing is that none of the above mentioned seems to have any impact on my issue.


    Other SNMP-Request are just working fine (Monitoring of CPU, RAM, Storage, etc...)


    I am using the following command on the command line:


    Code
    1. centreon_plugins.pl --plugin=os::windows::snmp::plugin --hostname 2012srv --snmp-community xxx --mode service --service='Windows Time' --critical 1
    2. :

    Try to see first if the object still exists for Director using:


    Code
    1. icingacli director service exists 'check_email_age_eGain'


    if so just delete it using 'delete' in exchange for 'exists'


    Be aware that after that you need to create the object again as you deleted it.


    Even if this isn't the exact answer to your question it should solve your problem.

    Code
    [root@eusbnoc01 director]# icingacli director config deploy
    PHP Fatal error: Allowed memory size of 1073741824 bytes exhausted (tried to allocate 130968 bytes) in /usr/share/icingaweb2/modules/director/library/Director/Objects/IcingaObject.php on line 158

    Fatal error: Allowed memory size of 1073741824 bytes exhausted (tried to allocate 130968 bytes) in /usr/share/icingaweb2/modules/director/library/Director/Objects/IcingaObject.php on line 158


    Well this problem is of different nature though

    I personally surpassed the indivudial Host Object Service threshold problem by using variables applied to Host-Templates.

    Using the same variables in the Services that I apply to the Host-Template gives me the flexibility to apply individual thresholds to an applicable Host.


    Still I am not sure how well this solution is thought through but it is working in my case.


    The only downside I found so far is that you won't see the actual threshold values on Icingaweb2 rather than the variable names.



    Ich poste mal für dich zum Abgleich eine entsprechende Konfiguration die bei mir läuft.



    Service-Template:


    Service:


    Host-Template:

    Code
    1. template Host "T-Netzwerk-HTTP-Check" { import "T-generic-internet"
    2. }


    Host:


    Code
    1. object Host "Deine-Website" {
    2. import "T-Netzwerk-HTTP-Check"
    3. display_name = "Deine-Website"
    4. address = "internet.de"
    5. icon_image = "internet.png"
    6. groups = [ "G-Netzwerk-Internet" ]
    7. }


    Tipp: Wenn du einen Host mit die IP oder dem Domain-Namen der Website die du monitoren möchtest, als "Hostadresse" zuweist, kannst du einfach die Variable $address$ benutzen.

    Dann musst du auch keine eigenen Felder erstellen.

    Ich habe das Problem bei mir wie folgt gelöst.


    Code
    1. command = [ PluginDir + "/check_by_ssh" ]


    An der Stelle im Director die Argumente für sich wiederholende Parameter mitreichen.

    Sieht bei mir in etwa so aus:



    Code
    1. command = [
    2. PluginDir + "/thirdparty/centreon-plugins/centreon_plugins.pl",
    3. "--filter=sensor,supply.8",
    4. "--filter=sensor,supply.6",
    5. "--filter=sensor,supply.9"
    6. ]


    Vielleicht nicht schön, aber erfüllt seinen Zweck.