Retry control in the FlowEngine (FlowEngine notifications over Grid with retries support)
This functionality is available from version 2.1.0-gradius onwards.
Introduction
FlowEngine flows are often used to process the insertions in Ontologies with an engine flow. When the FlowEngine was stopped by any reason, these notifications failed to reach the flows, so developers needed to manage this process to not lose them.
In this version, a new functionality has been added, that allows the user to configure whether they want to have retries in the notification process, as well as assigning a TTL to them.
How to use it
The following properties have been added to the notification node:
By default, all checks will be disabled, but selecting “Retry notification after error” will allow to activate retries. In that case, the check “Discard notification retry after elapsed time” will also be displayed. This property allows you to create a current TTL in which the notification will be retried in case of error. After that time, if the notification has not been delivered, it will be discarded.
If the first check is selected, the node will have two outputs: The first one will show the notifications that have managed to be sent in the first try, and the second one will be for notifications that have had at least one retry.
Regardless of the retry policy you choose, the output notification message will have 2 new fields that can be useful:
retriedNotification: Boolean value that tell whether the notification has been retried at some point.
originalNotificationTimestamp: Original timestamp of when the notification order was created.