Dear [Team / Support / Apache Spark Community], I hope this message finds you well.
I'm reaching out to inquire about the support for *user impersonation* in the *Spark Thrift Server* across different versions of Apache Spark, specifically from *Spark 1.x through Spark 4.x*. We are currently evaluating the use of Spark Thrift Server in a multi-user environment, and one of our key requirements is the ability to have queries executed under the identity of the requesting user (i.e., impersonation). This is critical for ensuring proper access control and audit logging on underlying systems such as HDFS and YARN. Could you kindly confirm the following: 1. Does Spark Thrift Server support user impersonation in Spark 1.x, 2.x, 3.x, and 4.x? 2. If supported, is it enabled by default, or does it require specific configuration? We would greatly appreciate your guidance or any relevant documentation references that could help us configure this properly. Thank you very much for your support. Best regards, Allen Chu Email : cddac1...@gmail.com
