Re: phoenix.Activation seems to deadlock when ActivationGroup fails to start

OK, I understand what's going on.  There are basically two possible
outcomes, totally dependent on how threads get scheduled.  From your
perspective, both outcomes are undesirable.  The root issue is that
if a group has an object registered with restart set to true, and
the group VM exits, phoenix/rmid will attempt restart, and that will
continue forever, there's no policy implemented to limit such restarts.
The one thing that thing that does limit it is a deadlock situation,
which is what you're hitting.  Phoenix/rmid will limit the number of
concurrent VM execs, and that code is not anticipating the kind of
recursive restart that's taking place when group creation fails.
When the groupThrottle limit is reached in this situation, a deadlock
can result.  You can work around the deadlock by making the groupThrottle
config entry large, but that will put you back into perpetual restart.

A workaround you might consider is to restructure so that group creation
doesn't fail, but object activation fails instead.

- Bob

--------------------------------------------------------------------------
Getting Started:     http://www.jini.org/wiki/Category:Getting_Started
Community Web Site:  http://jini.org
jini-users Archive:  http://archives.java.sun.com/archives/jini-users.html
Unsubscribing:       email "signoff JINI-USERS"  to [email protected]