We're updating the issue view to help you get more done.Learn more

Spark Adapter permission checks at the View level

On the "unprivileged" Spark Adapter we don't check permissions at the View level, we check the permission for the underlying table because we are giving the user application full access to all the data stored in the table.

This works fine on normal queries, we check the permission at the View level.

Status

Assignee

Daniel Gómez Ferro

Reporter

Daniel Gómez Ferro

Test Plan

1. Create view on kerberized cluster with splice.authentication.token.enabled = true 2. Give permission to access view to a user (not base table) 3. Compare execution of df() vs internalDf() accessing the view 4. Compare execution of sqlContext.read.options(internalOptions).splicemachine with and without JDBC_INTERNAL_QUERIES set

Fix Build

1826

Time estimate

24h

Components

Sprint

None

Fix versions

Priority

Critical