A customer noticed that, on his configuration with replicated HA, the swap memory of the backup cluster was filled out quickly,  Digging into the memory usage, it appears that some org.apache.activemq.artemis.core.paging.impl.LargeMessage objects still remain on the heap.

Is there some way to speed evacuation of these objects from the live heap after they are copied over? 

This ticket is very similar to https://issues.redhat.com/browse/ENTMQBR-9058