You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by "Sai Hemanth Gantasala (Jira)" <ji...@apache.org> on 2021/08/31 04:28:00 UTC
[jira] [Created] (HIVE-25490) Table object should be authorized
with owner info in the get_partitions() api in
Sai Hemanth Gantasala created HIVE-25490:
--------------------------------------------
Summary: Table object should be authorized with owner info in the get_partitions() api in
Key: HIVE-25490
URL: https://issues.apache.org/jira/browse/HIVE-25490
Project: Hive
Issue Type: Bug
Components: Hive, Standalone Metastore
Reporter: Sai Hemanth Gantasala
Assignee: Sai Hemanth Gantasala
HiveMetaStore#get_partitions() api is currently authorizing against table name. Instead, the table object should be authorized so that it also has table_owner information in the table object.
Currently, a user from spark-shell running these commands (in a rangerized environment):
> spark.sql( " create database 791237_db1 " ).show(false)
> spark.sql( " CREATE EXTERNAL TABLE IF NOT EXISTS 791237_db1.t1(cal_dt timestamp) PARTITIONED BY (year string) stored as parquet location '/791237/791237_db1' " ).show(false)
> spark.sql( " select * from 791237_db1.t1 " ).show(false)
ERROR metadata.Hive: NoSuchObjectException(message:Table t1 does not exist)
Even though the user is the owner of the table, but the same user cannot query the table. This should be addressed.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)