Head First EJB

Chapter 10. Exceptions in EJB: When beans go bad

Expect the unexpected. Despite your best efforts, things can go wrong. Terribly, tragically, wrong. You need to protect yourself. Others depend on you. You can’t let your entire program collapse, just because one bean in the family throws an exception. The application must go on. You can’t prevent tragedy, but you can prepare for it. You need to know what is and is not recoverable, and who is responsible when a problem occurs. Should the Bean Provider tell the Container there’s still hope? Should the Container tell the client to try again? Or should the Container cut its losses, kill the bean, and let the Sys Admin sort it out later...

Objectives

Exceptions

Official:		What it really means:
12.1	Identify correct and incorrect statements or examples about exception handling in EJB.	You have to know that application exceptions are things the client expects, and might recover from, while system exceptions are for things the client can’t recover from. You need to know the five standard EJB application exceptions, and all of the common system exceptions. You have to know exactly how the Container behaves when the bean throws an application exception vs. a system exception, and you have to know the difference between exceptions for local clients and exceptions for remote clients.
12.2	Given a list of responsibilities related to exceptions, identify ...