I have a Corba Servant which works if I access it by corbaloc.
Currently I am running client and server at the same machine.
xx.xx.xx.xx = my machines ip adress
I registered the Servant at JBoss Naming service and got no errors but one warning at runtime.
2004-12-06 15:38:48,382 INFO [org.jboss.naming.NamingService] Starting jnp server
2004-12-06 15:38:48,664 INFO [org.jboss.naming.NamingService] Started jnpPort=1099, rmiPort=1098, backlog=50, bindAddress=/0.0.0.0, Client SocketFactory=null, Server SocketFactory=org.jboss.net.sockets.DefaultSocketFactory@ad093076
2004-12-06 15:38:48,679 WARN [org.jboss.naming.NamingService] Context.PROVIDER_URL in server jndi.properties, url=xx.xx.xx.xx:10992004-12-06 15:38:48,695 INFO [org.jboss.naming.NamingService] Listening on port 1099
2004-12-06 15:38:48,726 INFO [org.jboss.naming.NamingService] Started jboss:service=Naming
When I try to access a Server method at the client the following
Jacorb Traces appears at the Server:
ClientConnectionManager: created new conn to target 0.0.0.0:3528
The Client trace is at this time:
Trying to connect to xx.xx.xx.xx:3528 ]
[ 06:12:2004:15:39:27:742> Connected to xx.xx.xx.x:3528 ]
looking up "CmdLogger2"
[ 06:12:2004:15:39:27:804> Delegate.getReference with POA < empty> ]
[ 06:12:2004:15:39:27:992> ConnectionManager: created new conn to target 0.0.0.0:3528 ]
[ 06:12:2004:15:39:27:992> (Pool)[0/1] creating new thread ]
[ 06:12:2004:15:39:27:992> Successfully negotiated Codesets. Using ISO 8859-1 as TCS and UTF-16 as TCSW ]
[ 06:12:2004:15:39:28:007> (Pool)[1/2] added idle thread ]
[ 06:12:2004:15:39:28:226> (Pool)[0/2] removed idle thread (job scheduled) ]
[ 06:12:2004:15:39:28:242> Trying to connect to 0.0.0.0:3528 ]
[ 06:12:2004:15:39:28:242> Retrying to connect to 0.0.0.0:3528 ]
[ 06:12:2004:15:39:28:742> Retrying to connect to 0.0.0.0:3528 ]
org.omg.CORBA.TRANSIENT: Retries exceeded, couldn't reconnect to 0.0.0.0:3528 minor code: 0 completed: No
Does anybody know what could be the reason for jacorb to use 0.0.0.0
instead of my machines adress ?
Not sure if this is an ejb3 problem or not but...
Using version 4.01RC1, when I run the all configuration, I don't get any of the services declared in the /farm/cluster-examples-services.xml file working. I'm new to farming etc, so I may be wrong here, but I've renamed the file eg, "test-service.xml" and taken out all comments - even though all the services look as if they are uncommented anyway! - and then I begin to see the services activated. Still getting exceptions though.
any advice? - anyone see this working? I've successfully deployed ejb3 files so this might affect it, alternatively it may just be a 4.0.1RC1 issue
any help much appreciated,
I am real new to JBoss, but most IP based application will connect on 0.0.0.0 so that they are listening on all IP addresses that the host has open. Normally if you want an IP application to listen on a specific address only, you must configure the application with the specific address.
The log shows you specific address, most likely because something is using your hostname to figure out what you IP address is. Your hostname will reslove to your IP address.
If you notice in the startup there is a parm that has "bindAddress=/0.0.0.0", that means to open up the port on 0.0.0.0, which means every IP address on your host.
Normally every host has at least two IP address, one is the loopback address (127.0.0.1) and then at least one address for your NIC. A single NIC could have multiple addresses, or you could have mutliple NIC's.
Use the following when starting the jboss server:
run -c "your configuration" --host="your host name"
It works for JBoss 3.2.x and for JBoss 4.0
I don't know why but if I build and run my client from eclipse it works without changing the JBoss configuration.
Building and running the same client with JBuilder doesn't work.
So I will use eclipse for this in the future.