|
From: Chuck W. <ch...@ma...> - 2006-12-15 06:36:09
|
Hi All,
I've got a strange anomaly in an application running in Windows under
the wrapper and am wondering if the wrapper might be in any way
complicit. The issue is that the application hangs registering
finalizers inside Object.clone() for 15 minutes. The class whose
instance is being cloned has one trivial one-line finalize() method.
There are 4 threads doing cloning the same instance at the same time.
Has anybody seen anything like this? Does wrapper interact with java in
some way that might alter finalizer management in Object.clone()?
The bottom of the stack trace, the part inside java, looks like this:
java.lang.ref.Finalizer.add(Unknown Source)
java.lang.ref.Finalizer.<init>(Unknown Source)
java.lang.ref.Finalizer.register(Unknown Source)
java.lang.Object.clone(Native Method)
Each thread doing this sits there for 15 minutes. Normally this same
operation is near-instantaneous. The problem happens sporadically,
sometimes happening but usually not on precisely the same clone() operation.
Thanks for any suggestions or ideas,
Chuck
|