Github user gatorsmile commented on the issue:
https://github.com/apache/spark/pull/21158
The default value introduced by SPARK-22479 was already part of Apache
Spark 2.3. It is hard to say this is a regression, since URIs could contain the
access keys and usernames could have potentially personal identifiable
information. Thus, it also makes sense to block them. Although the original
JIRA is for JDBC, we also make it more secure. If users do not want to redact
them, they can change the default. I am not feeling comfortable to not redact
them in the current situation, since we already did it in Spark 2.3. All the
security issues could be serious. Thus, I think we should make it unchanged.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
