Ask Your Question

Revision history [back]

How do I impersonate a Hive user over JDBC?

SDC supports the use of HDFS User and impersonation for writing to HDFS as a user other than the SDC service user sdc. Is there a similar concept for accessing the Hive Metastore? There is no _Hive User_ exposed in the SDC UI for the Metastore processor or destination. The JDBC URL seems to be used to pass a Hive principal when the Hadoop environment is Kerberized, but it seems that the SDC service user is still used and needs permissions to access the Hive Metastore and perform DDL operations.