i think this information is stored in the workspace. we always use the latest workspace before we run headless rad.

I can access the server name from JNDI using:

String name = (String) initCtx.lookup("servername");

Is there a better way?

that error message doesn't make sense if edu.auth.CustomUserRegistry implements com.ibm.websphere.security.UserRegistry. have you tried using the file user registry example? that was very easy to setup and that's how i got my custom registry to work.

I am getting the following exception intermittently while using the IBM JMS client.

10 217 1/2/07 10:24:00:055 EST com.ibm.ws.sib.comms.server.ObjectStoreFullException com.ibm.ws.sib.comms.server.clientsupport.StaticCATProducer.rcvCreateProducerSes... 3-008-0002
11 218 1/2/07 10:26:00:051 EST com.ibm.ws.sib.comms.server.ObjectStoreFullException com.ibm.ws.sib.comms.server.ConversationState.extendObjectTable 5-004-0001

The code which i used to send the JMS messages is as follows:-
Session session =
connection.createSession(false, Session.AUTO_ACKNOWLEDGE);
Destination dest = (Destination) ctx.lookup(JMS_ALERT_EMAIL_QUEUE);
MessageProducer producer = session.createProducer(dest);

TextMessage message = null;
String alrtId = null;
try {
for(Iterator itr = alertIds.iterator();itr.hasNext();){
message = session.createTextMessage();
alrtId = (String)itr.next();

I know the docs and infocenters all say that ffdc is really for use by IBM tech staff, but it looks like there might be some juicy nuggets in there that would help resolve issues, and perhaps issues that I don't even know I have. For example, on start up of a clean server with no errors in the systemOut, I still see entries in this directory even though I would expect to see nothing.

Is there some guide on how to use ffdc for the rest of us?

Hi there,

after updating from was 6.0.2.3 to was 6.0.2.15 the cluster is no
longer accessible to a remote client.
However, the single servers in the cluster can be accessed. The
following error (SystemOut.log) occurs when client attempts to access
the cluster:

Context implementation: com.ibm.ws.naming.jndicos.CNContextImpl
Context method: lookupExt
Context name: cell01/clusters/app-prod
Target name: cell/clusters/app-prod
Other data: ""
Exception stack trace...

As a new user of WAS, I attempted to instal a new test application. The following JAR file was created which contains:

0 Tue Jan 02 16:15:12 GMT 2007 META-INF/
71 Tue Jan 02 16:15:12 GMT 2007 META-INF/MANIFEST.MF
1986 Tue Jan 02 13:42:36 GMT 2007 HelloServlet1.class
93 Tue Jan 02 13:45:54 GMT 2007
HelloServletsTextMessages_fr_FR.properties
83 Tue Jan 02 13:46:32 GMT 2007
HelloServletsTextMessages_en_US.properties

It does not apear to be corrupt.

The Local file system radio button was selected and the full path was entered to the new application.

On clicking Next the following error message was displayed:
---
The EAR file might be corrupt or incomplete.
AppDeploymentException: [ADMA0157E: The system cannot find any valid enterprise archive (EAR) wrapper extensions or an extension did not run successfully to set a new EAR file path.]
---

Just to find out more about this error, I checked the remote file system and entered the full path to the JAR file there, this gave the error message:

"The specified file is not an EAR, WAR, JAR, or SAR archive."

It can be seen from the above listing that it is a JAR file. I have attached the JAR file in case you have any doubts.

I would be interested to hear if anyone has got any ideas about what is causing this error or how to solve it.

i've got an astonishing exception on WAS 6

Indeed, when a simple remove on an ejb entity included in a transaction made by a Session Stateless, I've got :

[1/3/07 0:30:50:785 CET] 0000002f ExceptionUtil E CNTR0020E: EJB threw an unexpected (non-declared) exception during invocation of method "remove" on bean "BeanId(test#testEJB.jar#DocumentTypes, 10121...