Please clarify..
This is a tricky problem to explain clearly; it involves drawing deep and careful distinctions in the use of terms like "application". Specifically, you have to understand the behavior of Java class loaders in an application server environment. In most situations, the current Derby mechanisms for preventing accidental dual-boot of a database work very well, but in some app server environments they can be fooled, and you can accidentally have two copies of the Derby engine servicing the same database, and then the software does not work properly. For a deep and detailed explanation of what's going on, start here: https://issues.apache.org/jira/browse/DERBY-700 Your questions are important and completely legitimate and you should continue to ask them, and refine your understanding of how the Derby software works. I'm confident that you will be able to configure the software to work adequately for your environment, once you fully understand how the software works, and the various configuration choices at your disposal. thanks, bryan
