On Jul 17, 2006, at 3:21 PM, Jacek Laskowski wrote:
On 7/17/06, David Blevins <[EMAIL PROTECTED]> wrote:
Give running the test in fork mode a try and see if that doesn't just
fix it -- i think the openejb-core pom does it. If not, post a
stack trace and I'll see if I can't make sense of it.
Take a look at the following command line snippets and please explain
how it could work fine for other tests? It can't be.
[...]
java.net.MalformedURLException: no protocol: and
at java.net.URL.<init>(URL.java:567)
at java.net.URL.<init>(URL.java:464)
at java.net.URL.<init>(URL.java:413)
at sun.rmi.server.LoaderHandler.pathToURLs
(LoaderHandler.java:747)
at sun.rmi.server.LoaderHandler.loadClass(LoaderHandler.java:
147)
at java.rmi.server.RMIClassLoader$2.loadClass
(RMIClassLoader.java:620)
at java.rmi.server.RMIClassLoader.loadClass
(RMIClassLoader.java:247)
at sun.rmi.server.MarshalInputStream.resolveClass
(MarshalInputStream.java:197)
[...]
at java.rmi.MarshalledObject.get(MarshalledObject.java:135)
I see. The sun RMIClassLoader is totally broken. We had this
problem in the 2.x codebase as well.
I switched it to just use ByteArray{Input,Output}Stream and Object
{Input,Output}Stream instead of MarshalledObject.
Should work fine on windows now.
-David
