Yes, of course, I configured host check for each device too. By the way,
since I have configured dependencies, don't come any host alert
notification (up or down), but host dependencies doesn't include parts,
that can cause this behaviour. For example, there is myHost, and two
related dependencies (the full configuration is based on this pattern):

define servicedependency{
        host_name                       myHost
        service_description             pingv4
        dependent_host_name             myHost
        dependent_service_description   ospf_neighbor_count,
bgp_neighbor_count
        notification_failure_criteria   w,u,c
}

define hostdependency{
        host_name                       myHost
        dependent_host_name             l2vpn-myHost, l3vpn-myHost
        notification_failure_criteria   d,u
}

The goal of service dependency is to repress all of service notifications
for device when it is down - I thought, if I will check pingv4 service and
it fails, none of other services will notify.
On the other hand, host dependency is only for repress hosts alerts for
related l2vpn and l3vpn hosts - it is working.


Thanks in advance!


Gergo



2016-07-14 14:32 GMT+02:00 Brian O'Neill <[email protected]>:

> You say the host went down...do you have a host check as well? If the host
> check fails, a service check won't notify, at least in the default
> configuration I believe. You should get a host down notification (if
> configured) and that's it.
>
>
> On 7/14/2016 8:27 AM, Szabó Gergő wrote:
>
>> Hi Michael!
>>
>> Sorry, but I can't publish these config files due to company's privacy. :(
>> There are no any special configuration, Till I didn't expand
>> configuration with service dependency, we got a lot of notifications
>> from monitored devices, but now the problem is that we don't get if is
>> necessary. I have searched for this case on the internet, how it should
>> to use, but when I compared with existing (and previously mentioned
>> test) configuration, there were no differences in logic, also by Nagios
>> books the config seems like it should. I have been using and configuring
>> Icinga for many years, so I think, that it may be a bug or something
>> else, which is occuring only at large networks (there are ~1000 hosts,
>> ~3000 services and ~10000 host/service dependencies).
>>
>> Perhaps... any idea, that what can cause this abnormal behaviour?
>>
>> Thanks for reply!
>>
>>
>> Gergo
>>
>> 2016-07-13 19:17 GMT+02:00 Michael Friedrich
>> <[email protected] <mailto:[email protected]>>:
>>
>>
>>     Am 13.07.2016 um 17:37 schrieb Szabó Gergő:
>>
>>         Hi all!
>>
>>         I have previously configured some service dependencies as it seems
>>         below (for example):
>>
>>         define servicedependency{
>>                 host_name                       myHost
>>                 service_description             pingv4
>>                 dependent_host_name             myHost
>>                 dependent_service_description   service1, service2
>>                 notification_failure_criteria   w,u,c
>>         }
>>
>>         The problem is that today the myHost went down for some minutes
>> and
>>         didn't come any notification related to myHost reachability. If I
>>         searched for notifications in Icinga, there were all of related
>>         notifications in the Alert history.
>>         So, the main problem is, that if myHost fails on pingv4 service
>> too,
>>         doesn't come any notification about it (of course neither of
>>         dependent
>>         services).
>>
>>         Last month I tested similar configuration in Icinga and it
>> worked. I
>>         don't know, where is the configuration wrong (if it is anywhere).
>>
>>         Did anybody hear about similar Icinga bug?
>>
>>
>>     Can you provide some more configuration items and logs to allow the
>>     reader reproduce the problem more easily? :)
>>
>>     Kind regards,
>>     Michael
>>
>>
>>     --
>>     Michael Friedrich, DI (FH)
>>     Senior Developer
>>
>>     NETWAYS GmbH | Deutschherrnstr. 15-19 | D-90429 Nuernberg
>>     Tel: +49 911 92885-0 <tel:%2B49%20911%2092885-0> | Fax: +49 911
>>     92885-77 <tel:%2B49%20911%2092885-77>
>>     CEO: Julian Hein, Bernd Erk | AG Nuernberg HRB18461
>>     http://www.netways.de | [email protected]
>>     <mailto:[email protected]>
>>
>>     ** OSBConf 2016 - September - osbconf.org <http://osbconf.org> **
>>     ** OSMC 2016 - November - netways.de/osmc <http://netways.de/osmc> **
>>     _______________________________________________
>>     icinga-users mailing list
>>     [email protected] <mailto:[email protected]>
>>     https://lists.icinga.org/mailman/listinfo/icinga-users
>>
>>
>>
>>
>> _______________________________________________
>> icinga-users mailing list
>> [email protected]
>> https://lists.icinga.org/mailman/listinfo/icinga-users
>>
>> _______________________________________________
> icinga-users mailing list
> [email protected]
> https://lists.icinga.org/mailman/listinfo/icinga-users
>
_______________________________________________
icinga-users mailing list
[email protected]
https://lists.icinga.org/mailman/listinfo/icinga-users

Reply via email to