Configuring alert engine
Processing data (particularly radar data) for alert detection can be system intensive. In the case of a complex alert configuration, the system may sometimes be unable to keep up. When this occurs, the delay from the time when data arrives until the time when you are notified will start to increase.
When the alert engine detects that there is too much delay in alerts from some combination of area of interest and event, the alert engine clears all of the pending data for that area/event combination. As a result, some data gets dropped, but the system is able to catch up to real-time. The timeout period before data is dropped is adjustable. The default is 2 minutes.
To configure parameters related to alert engine:
- Go to the /etc/vaisala/radarsw/webapp directory.
- Open the application.yaml file.
-
Adjust these parameters:
threads: Defines the number of threads to make available for alert processing.flushTimeOut: Defines the timeout period (the period after which data is dropped), in seconds.
