nullpointerexception in GC about codenameone

[shai-almog]

Might be related to my work on #3726

[ddyer0]

i agree. if the finalizer is new that would explain how it was unnoticed. I'm still tracking the issue, but I haven't seen
any directly bad behavior that corresponds to this other than in the debugger.

[ddyer0]

The issue I’m tracking is compatible with the hypotheses that this npe leaves the gc disabled, and subsequently any process that becomes blocked for gc is effectively dead.

Additional evidence: The error I'm tracking first occurred in my logs in a build I made on Sep 15

I hacked my use of sockets to prevent my SocketInputStream and SocketOutputStream instances from being
garbage collected, and that seems to have fixed it.

I don't have a good model how this could have wedged the garbage collector without causing a more
immediate problem - do you ? - It seem to me that there should have been some immediate effect rather
than leave it to victimize some random process sometime in the future in this mysterious way.