Hi Oliver,
The number of segments you define at start up, is called the "initialSegmentCount".
This field will only be used to deduce the initial number of segments if no segments are present.
Thus, when the Tracking Event Processor starts up for the first time.
So, if you want to scale the number of segments after the application has run, you will have to use the split and merge functionality.
Although indeed unfortunate, an Axon Framework application on it's own could never deduce this, as a single application is not aware of any other application running the same Tracking Event Processor.
Hence why the API of split and merge is implemented and usable through the Axon Server dashboard, as it serves as the "man in the middle" which can coordinate these operations.
As such the automatic scaling of segments will very likely eventually end up in Axon Server.
Within a single Axon application however, I don't think this will be introduced due to the aforementioned reasons.
Trusting this clarifies things for you Oliver!
Cheers,
Steven
PS. It is important to note that this mailing list will be discontinued as specified in this
thread.
The thread also specifies where to look further for help when it comes to Axon as soon as this mailing list is closed.