I have an issue when trying to migrate our event processing to Tracking Processors. Until now we have used async subscribing processors.
We have a setup with two databases for our events, with historic events in a larger database (around 140.000.000 events), and a second database with the latest events (arond 50.000.000 events today). Events for any given aggregate can be stored in either the first, second or both databases, and these are connected with SequenceEventStorageEngine. Both datasources use JdbcEventStorageEngine.
When I setup a local equivalent for testing, the TrackingEventProcessor goes into a loop where it will re-read and apply all events from the second (current) event database indefinately. Events from the first database seems to be treated only once, but the second datastore goes into a loop re-applying events.
Is this a known issue? I could configure only the second event database as source for the Tracking Processor, but I was hoping to be able to avoid this and reuse the already configured EventStore.
I created a testcase showing the problem and pushed to Github. Hopefully someone with a deeper understanding of tracking processors and sequenced eventstreams can help on this. https://github.com/AxonFramework/AxonFramework/issues/1491.
Any update on when the next release will be?
We’re looking forward to migrating away from subscribing processors - currently we’re vulnerable to application restarts (when scaling and replicating) in kubernetes and have “lost” some events in subscribing processors…
We typically don’t give dates for any of our releases.
However, I can work on gutfeels in messages like these I think
Pretty confident the 4.4.3 release of framework could be done next week.
Stay tuned!
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.