Hello,

I have something strange with my shinken deployment. I think it can be a bug with the escalation but perhaps it's the way it should work.

I have services with escalation (normal services) and services without escalation (passive services checked every 24 hours). With the normal services (with escalation) all seems to be fine but with the passive services (without escalation) I have a problem :

My service "r1backup" is configured like that :
Code:
define service {
        service_description          r1backup
        check_command             passive_alert!r1backup
        hostgroup_name            Net4all - R1 Backup
        contact_groups            OnCall
        use                  Service Passif 24h
}
define service {
        name                 Service Passif 24h
        register               0
        max_check_attempts          1
        check_interval            10
        retry_interval            2
        first_notification_delay       0
        notification_interval         1440
        notification_options         u,c,r
        active_checks_enabled         0
        passive_checks_enabled        1
        notifications_enabled         1
        check_freshness            1
        freshness_threshold          93600
        check_period             24x7
        notification_period          workhours
}
So it send notifications once a day (notification_interval 1440) to the contactgroup "OnCall" without any escalation.

The host is conifuged like that :
Code:
define host {
        host_name               db1.domain.com
        alias                 192.168.1.10
        address                192.168.1.1
        action_url              /pnp4nagios/index.php/graph?host=$HOSTNAME$&srv=_HOST_' target='main' class='tips' rel='/pnp4nagios/index.php/popup?host=$HOSTNAME$&srv=_HOST_
        icon_image_alt            Windows
        icon_image              win40.gif
        statusmap_image            win40.gd2
        check_command             check-host-alive
        check_command             check-host-alive
        parents                sw_9_hotel
        use                  Elib - 24x7
}
define host {
        name                 Elib - 24x7
        register               0
        max_check_attempts          4
        check_interval            5
        retry_interval            1
        notification_interval         30
        notification_options         d,u,r,f
        active_checks_enabled         1
        passive_checks_enabled        0
        notifications_enabled         1
        check_freshness            0
        check_period             24x7
        notification_period          24x7
        contact_groups            OnCall,Elib
        escalations              Elib_Backup
}
So if the host is down or unreachable it send the first notification to "Oncall" and "Elib" and the next to the escalation group "Elib_Backup".

The escalation is configured this way :
Code:
define escalation {
        escalation_name            ELib_Backup
        first_notification_time        30
        last_notification_time        0
        contact_groups            OnCall,Backup,Elib
}

Now my problem. If the service "r1backup" has an issue it send me correctly the first notification :
Code:
db1.domain.com	r1backup	CRITICAL	2012-11-05 12:27:09	user-Oncall	notify-service-by-email
It should wait 1440 minutes before sending the next notification but no, it send the notification by using the escalation settings of the host :
Code:
db1.domain.com	r1backup	CRITICAL	2012-11-05 12:57:09	user-Backup	notify-service-by-email	
db1.domain.com	r1backup	CRITICAL	2012-11-05 12:57:09	user-Elib	notify-service-by-email	
db1.domain.com	r1backup	CRITICAL	2012-11-05 12:57:09	user-Oncall	notify-service-by-email
After that notification it stop sending messages and wait one day before sending the next one.

I have the 1.2.2 version of shinken.

I hope my explanation can be understand Sorry for my bad english...

Thank you very much for your help. If someone can help me to understand what is wrong

With my best regards