Not sure if it is a bug, however the server startup fails after a restart (redeploy). This is not really stable and robust.

If you register a namespace with the same URI but with a different prefix, the old namespace gets replaced by the new one (thus removed). When the old namespace is in use by a nodetype, the server fails to start. A data wipe is neccesary then.

Reproducible in unit test using file system, directory gets cleaned up after every tearDown.