The EJB loadbalancing and failover works with a client proxy and a server communication.
The server communicate a shutdown or detect a full stop (e.g. complete hanging or looong full GC).
A crash, e.g. JVM cored, is also detected.
It depends on the situation how long it takes.
Fact is
- all Tx on the chrased server are not commited
- the client proxy might hung for a few millis and try the next server of its list
- the next server provide the new cluster view without the crashed server.
So there is no measurable time for such failover in best case and only a few millis in worst.