With iReports 3.0.0 standalone application (not the plugin), I've managed to make it work doing the stated below: Replace, in the iReports "lib" folder, the following JARs by the ones in the HIBERNATE distribution you're using - don't bother doing this if you're not using HIBERNATE, of course: EJB3 PERSISTENCE, JAVASSIST, HIBERNATE VALIDATOR, ANTLR, COMMONS COLLECTION, DOM4J, COMMONS LOGGING, EHCACHE, JTA, LOG4J, XML-APIS, HIBERNATE3, HIBERNATE COMMONS ANNOTATIONS, HIBERNATE ANNOTATIONS and HIBERNATE ENTITY MANAGER. Generate a JAR for your ORM EJB project (not for the WEB one) replacing, in PERSISTENCE.XML, "transaction-type=JTA" by "transaction-type=RESOURCE_LOCAL" and <jta-data-source> by <non-jta-data-source>. This JAR should be copied elsewhere, because your project won't run this way. After doing this JAR generation, you should undo these changes and recompile. Define the iReports CLASSPATH to: all JDBC drivers + any accessory JARs you have (if you have them) + the resource bundle folder of your web project (if you use resource bundles) + the JAR built on step 2 + all JARs relevant to your application server (in the case of GLASSFISH, they are APPSERV_RT, APPSERV_ADMIN, APPSERV_DEPLOYMENT_CLIENT, APPSERV_EXT, APPSERV_EE, J2EE, JAVAEE and install/applications/jmsra/ imgjmsra.jar). Make sure you JASPERREPORTS library, in Netbeans, include iReports.jar and barbecueXXX.jar (if you intend to print barcodes). Setup a properly formatted jndi.properties file, in you <JRE>/lib folder, so that it's possible to find your data sources from the JNDI names specified in the PERSISTENCE.XML. Keep your JNDI service running (just keep your container running when you're testing the reports).After doing all this, I was able to develop reports using iReports and EJB/QL queries.
