Duplicate notifications #10623
Description
From time to time, maybe about once per week, i see Icinga sending out two notifications for services going into the critical state.
The logfile for one of that occasions state:
[2025-11-06 03:15:10 +0100] information/Notification: Sending 'Problem' notification '<servicename redacted>' for user '<username redacted>'
[2025-11-06 03:15:10 +0100] information/Notification: Sending reminder 'Problem' notification '<servicename redacted>' for user '<username redacted>'
[2025-11-06 03:15:10 +0100] information/Notification: Completed sending 'Problem' notification '<servicename redacted>' for checkable '<servicename redacted>' and user '<username redacted>' using command '::Notify_Service-redacted'.
[2025-11-06 03:15:10 +0100] information/Notification: Completed sending 'Problem' notification '<servicename redacted>' for checkable '<servicename redacted>' and user '<username redacted>' using command '::Notify_Service-redacted'.
So, in the same second, Icinga sends out the initial notification and a reminder.
The definition of the notification is as follows:
apply Notification "service_notify_via_redacted" to Service {
command = "::Notify_Service-redacted"
interval = 0s
period = "7x24"
assign where --redacted--
states = [ Critical, OK, Unknown, Warning ]
types = [ Problem, Recovery ]
users = [ "redacted" ]
}
The icinga2 installation has a lot of services to check and sends out notifications quite frequently. Normally, they are not duplicated.
The documentation says about the interval:
The notification interval (in seconds). This interval is used for active notifications. Defaults to 30 minutes. If set to 0, re-notifications are disabled.
Does Icinga have an issue with "0s" instead of "0"? I am using the director for configuring Icinga and the interval is set to "0", but rendered as "0s".
I am using Icinga2 version v2.15.0-51-g0a08fcc0e