Dear, I am new to Livy but succeeded in setting it up for a Spark cluster.
It seems however it is not possible to process statements concurrently, even with the Livy session using Spark's fair scheduler. For sql queries the Thriftserver integration of Livy appears to allow concurrent processing from beeline. After running maven with Pthriftserver, it works fine. However, to create a Thriftserver session a POST request to the thriftserver/sessions is not allowed (405 status code) in my running livy server. Do I resolve this by means of enabling authentication? Another plausible reason is that POST is not implemented for thrift server sessions. But then I don't understand how to create a session using Scala code. The example code ( https://github.com/apache/incubator-livy/blob/4d8a912699683b973eee76d4e91447d769a0cb0d/thriftserver/server/src/test/scala/org/apache/livy/thriftserver/ThriftServerSuites.scala#L448) specified in a comment of the JIRA issue for the thriftserver issue, exceeds my knowledge of scala. How do I run this on the node with the livy server as a separate process? Best Eddy
