Table Access Control without High-Concurrency and with R

Data Governance

Join discussions on data governance practices, compliance, and security within the Databricks Community. Exchange strategies and insights to ensure data integrity and regulatory compliance.

As we are starting to build our Lakehouse solution on Databricks, we need ACLs to be active. So far I have found two options:

via UI or terraform: create a high-concurrency cluster and enable table access control for python and SQL. In terraform this would look like this:

spark_conf = {
    "spark.databricks.cluster.profile": "serverless",
    "spark.databricks.repl.allowedLanguages": "python,sql",
    "spark.databricks.acl.dfAclsEnabled": "true"
}

only in terraform: create a "standard" cluster with enabled table access control for python, SQL and R. In terraform the code for config is as followed:

spark_conf = {
    "spark.databricks.repl.allowedLanguages" : "python,sql,r",
    "spark.databricks.acl.dfAclsEnabled" : true,
}

We have tested that this second option only allows me to see tables that I have been granted access to. Do I miss something in the documentation? Is it the correct way to deploy a single-user cluster with ACLs?