Given that rivers is soon to be deprecated I was wondering what the best system would be to move data from a SQL instance onto an elasticsearch instance. As it stands not only do I need to load the initial data that is already in the server, but from there keep the data flowing as new things are added onto the SQL instance.
Most of the solutions out there today are based on rivers which is soon to be deprecated from elasticsearch and hence no longer a viable solution to the problem.
The data on our SQL is often times not at all flat and requires manipulation to make it into a elasticsearch store-able document.