Hi - I have an app that launches a cluster on the fly, does
processing, and then shutdown down each node. With 1.9.4.8, I am
seeing exceptions thrown when shutting down, as shown here:
http://pastebin.com/A8q1E2QV
Basically, when each node detects that the jobs are finished, they
call Hazelcast.getLifecycleService().shutdown(). Nodes know when they
are done because they check, as follows:
while (Hazelcast.getSet(Sets.COLLAPSED_END.toString()).size() !=
files.size()) {
Thread.sleep(100)
}
Is this the right way to shutdown nodes? Is this some sort of race
condition which causes the exception to the thrown?
Thanks,
Chris