Hi all,
I had encountered this problem,In fact,the implementation of RMI references is
simple,it lacked the capacity to respond to some exception.The Spring rmi
client is a good job(org\springframework\remoting\rmi\RmiProxyFactoryBean,you
can find more details in class RmiClientInterceptor),it provides a
refreshStubOnConnectFailure property,that indicate whether to refresh the RMI
stub on connect failure. (If a cached RMI stub throws an RMI exception that
indicates a remote connect failure, a fresh proxy will be fetched and the
invocation will be retried.) The lookupStubOnStartup and cacheStub are also
helpful properties,you can learn more from the spring source code.
Need of special note is if we use the spring rmi client to connect the rim
service that is exposed by Tuscany RMI Service,although,the property
refreshStubOnConnectFailure of RmiProxyFactoryBean class set to be true,we will
still encounter an exception.
You should modify invoke method of the RmiClientInterceptor class to do with
UndeclaredThrowableException,
Details is :
public Object invoke(MethodInvocation invocation) throws Throwable {
Remote stub = getStub();
try {
return doInvoke(invocation, stub);
} catch (RemoteConnectFailureException ex) {
return handleRemoteConnectFailure(invocation, ex);
} catch (RemoteException ex) {
if (isConnectFailure(ex)) {
return handleRemoteConnectFailure(invocation, ex);
} else {
throw ex;
}
} catch (UndeclaredThrowableException ex) {//
return handleRemoteConnectFailure(invocation, ex);
}
}
The tuscany rmi service and reference may some improvement,Spring related
implementation is a good reference .Thank you
2011-03-16
fzhong
发件人： Millies, Sebastian
发送时间： 2011-03-15 20:43:34
收件人： [email protected]
抄送：
主题： Tuscany 1.6 RMI bug: ConnectException after component restart
Hello there,
I believe the following is a bug in the Tuscany 1.6 RMI binding:
When I have a network of components connected by RMI references,
then restarting a component will cause a java.net.ConnectException
in all dependent components on the next remote method call.
I suspect some kind of connection factory caches out-of-date information.
If it's a bug, it's a serious one, because it goes right against one main
reason for using distributed components in the first place.
Example:
ServerComponent exposes service "Server" with an RMI binding on port 8777.
ClientComponent exposes service "Client" with an RMI binding on port 8666
and has a reference to the service "Server".
Tester is a non-SCA Java class that exercises the Client service over RMI.
Everything works fine until the ServerComponent Java process is stopped
and re-started. The tester will then fail, because the client cannot
re-establish the connection to the server.
I have attached a zip-file with the example. Steps to reproduce the problem:
Run ServerLauncher
Run ClientLauncher
Run ClientTest
Stop process in which server is running
Re-Run ServerLauncher
Re-Run ClientTest
Has anyone else encountered this? Is it really a bug in Tuscany or in some
other component? Is there a workaround? Should I open a JIRA?
-- Sebastian
