Well, that would be nice... i get the following error:
17:50:27,173 INFO [EntityContainer] Creating
17:50:27,183 WARN [EntityContainer] *** EJB 'WebKundeEJB' deployed as CLUSTERED
but not a single clustered-invoker is bound to container ***
17:50:27,183 INFO [EntityInstancePool] Creating
17:50:27,183 INFO [EntityInstancePool] Created
17:50:27,193 INFO [EntityContainer] Created
17:50:27,193 INFO [EntityContainer] Creating
Farming is working, HA-JNDI, too.
The jboss.xml looks like
<configuration-name>Clustered Stateless SessionBean</configuration-name>
I tried a false configuration-name, then jboss throws an exception. When I leave the line blank, I get the same message as above... When I insert the whole standardjboss.xml file into my jboss.xml, I get the same error message,too.
We had the same problem yesterday with our EntityBeans. We had to give them another configuration-name. After that everything worked fine.
I saw that you already have a "Clustered" configuration name for your Bean, but perhaps you could try some of the ones being offered in standard-jboss.xml.
Thank you so far.... but it does not work yet.
I thought I gave it the right configuration-name? In standardjboss.xml there is the following part:
<container-name>Clustered Stateless SessionBean</container-name>
If I use any name (one that i made up myself without configuring it) jboss throws an exception (name not found or something like it)... as far as i can see, the invoker-proxy-binding-name is the invoker, and this one here is clustered...
why does jboss then tell me that no clustered invoker is bound?
You don't need to include a specific configuration name in your jboss.xml. And even if you would, you don't want to have a clustered session bean if you're deploying an entity.
Also check that your cluster-service.xml deployed correctly.
the cluster works (at least I believe it works, since it load-balances), so the cluster-service is deployed, or should be.
I have entity beans and session beans in one file (the output I pasted is a little missleading here I guess) - maybe this is a problem?
Except the error message I gave, no other errors are beeing shown on start up.
I removed the configuration-name, but get the same error message again.
Any help is still appreciated...
Your HA-RMI invoker MBean is deployed as part of the cluster-service.xml, check the log and see this service started succesfully.
The log says it is ok.
I lookup the session beans via 192.168.99.2:1100, 192.168.99.1:1100 from a different maching, so I believe HA-RMI is working...
If I deploy on one machine only, the requests are handled there, even if I do a lookup on the other machine. My conclusion is, HA-RMI and Load Balancing is working. (A multi-threaded test showed that load is balanced).
Then what's NOT working?
See my first post for the error message i get...
the beans are not deployed as clustered. I only have the same bean on every server.
Which means: If during a transaction one server dies, the transaction is lost...
My understanding of bean clustering is, that in the case above, the other server would continue the transaction - so the complete state of every clustered bean is replicated through the cluster.
Take a statefull session bean for example, the other server should continue with the same state as the server which died.
This is not working as far as I can see from the error message.
Well I don't know how to help you, if it can't find the invoker then it can't find it.
The state replication works out of the box in default 'all' configuration on all cases I've tried it.
Well, it does not.
Anyway, thank you for your help.