Closed
Description
Coming from this discussion: #848 (comment)
Currently Spring Data Elasticsearch repositories create the index where their entities should be stored in when the repository implementation is instantiated. This leads to calls to Elasticsearch to check for the existence, to create and store a mapping even when the repository might not be needed in that applicaiton run.
It should be possible to change this behaviour so that this initalization call is done when the first method from the repository is called. Both the current and the new behaviour should be possible, this should be configurable.