Server lifecycle improvement
Whilst testing the 1.8.9-SNAPSHOT build I could see that the elasticsearch module was starting too late and missed some incoming events. This was due to a wrongly chosen SmatLifecycle phase.
In the end I check all modules again and fixed also a problem with the Spring Boot context startup (see ServerStartup
).
Finally to improve the server shutdown I refactored the AbstractBufferedCacheListener
and reduced also the default pull frequency for buffered listeners to 5 seconds (see CacheProperties
).
Overall this should stabilize the server for the upcoming release.