Re: Deadlock in ModuleClassLoader

This appears to be an older version of the Felix framework, right? Makes it difficult to know if the issue still effects the latest release since there were definitely changes in this area. If you could verify that, it would be good. If it is related to JVM internal class loader locking, then t

Re: Deadlock in ModuleClassLoader

iceMix > 4.3. > -- View this message in context: http://old.nabble.com/Deadlock-in-ModuleClassLoader-tp31266713p33127356.html Sent from the Apache Felix - Users mailing list archive at Nabble.com. - To unsubscribe, e

Re: Deadlock in ModuleClassLoader

y test that the JVM flags "XX:+UnlockDiagnosticVMOptions -XX:+UnsyncloadClass" work reliably as a workaround. metatech -- View this message in context: http://old.nabble.com/Deadlock-in-ModuleClassLoader-tp31266713p33127027.html Sent from the Apache Felix - Users mailing list archive at

Re: Deadlock in ModuleClassLoader

Thanks for your reply, the page you point me to seems actually to be related to my problem. I'm going to test the workaround and see if the deadlock disappears. Regards -- View this message in context: http://old.nabble.com/Deadlock-in-ModuleClassLoader-tp31266713p31274765.html Sent from t

Re: Deadlock in ModuleClassLoader

I am fairly certain that you are experiencing the long-standing issue where the JVM is too aggressive in its locking of class loaders. The Felix framework doesn't hold class loader locks while class loader. The point where the threads are blocking are just simple checks to see if the class has

Deadlock in ModuleClassLoader

leImpl.java:734) at org.apache.felix.framework.ModuleImpl.access$400(ModuleImpl.java:71) at org.apache.felix.framework.ModuleImpl$ModuleClassLoader.loadClass(ModuleImpl.java:1768) at java.lang.ClassLoader.loadClass(ClassLoader.java:248) at java.lang.Class.forName0(Native Method)