We are seeing the same problem, coincidentally also with 3.0.8 but with 1.3.1_05 JVM. In our case we are actually running out of memory because too many of these threads build up.
I've also had a hard time tracking down sufficient information about the nature of these RenewClean threads. We definitely have a case where they just *won't* go away.
Any information is greatly appreciated as this is devastating to our application!
I am still trying to resolve this issue myself, the only place I have found that mentions this behaviour is on an old sun forum archive (http://archives.java.sun.com/cgi-bin/wa?A1=ind0210&L=rmi-users#26) , that may help, there are a number of discussion regarding thread leaks in RMI. One small hint I have found is that it seems to be remote objects that have callbacks that are particulary prone to this issue.