Github user skonto commented on the issue:
https://github.com/apache/spark/pull/21398
@vanzin @cloud-fan. Going back to the facts. The fix/work-around in
Cloudera's distro obviously exists because the distro integrates with Sentry
anyway. Removing the location did fix the Spark Job at the customer side. To
re-phrase the last request from @cloud-fan, if we remove the explicit location
what are the consequences, when do you need the explicit location for managed
tables? Can this be a default behavior in general?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
