Writing data using Azure Synapse Dedicated SQL Pool Connector for Apache Spark

Hi scirqueira 

We're following the same approach. I have one doubt about user's permission. Do we need to grant ADMINISTER DATABASE BULK OPERATIONS permissions to user. Because we're passing spark.conf.set("spark.synapse.runAsMsi", "true") or running notebook with enable Managed Identity while manual run (it's in development phase) and while running it's throwing error.

When we're using managed identity, why this error is coming in manual run. It should run on the behalf of the managed identity or it will run on the behalf of user as it manual run? and synapse managed identity is by-default user in all dedicated pools. Do we need to grant ADMINISTER DATABASE BULK OPERATIONS permission to managed identity? I believe identity should have sufficient permissions.