Created 07-05-2019 06:37 PM
Hi,
trying to run queries on hive managed tables in hive version3, but ı m getting errors. from my understanding, hive catalog and spark catalog was separated in hortonworks (HDP) 3 version+.
My question is - hive, spark catalog separation is just valid for HDP3+ ? or is it related with directly hive version3?
wondering that because the only way seems to use hortonwork hivewarehouseconnector for now, so if there is any way to run queries on hive managed tables in hdp3+ cluster, could you share?
Versions;
spark 2.3
HDP 3.1
Hive 3.1.x
Thanks.
Created 07-08-2019 07:28 PM
@Baris Akgun this is a feature of hive 3. The integration between spark and hive in hive 3 is solely through the connector. There are a number feature-rich reasons why this is the case.
Created 07-09-2019 07:08 AM
Thanks, so Hivewarehouseconnector is kind of unique solution if I try to connect Hive3 from spark. Am I correct? or is there any other way to connect hive3 from spark?
Thanks.