PERMISSION_DENIED: Cannot access Spark Connect. when trying to run serverless databricks connect

[ivan7256]

I am not able to run a file as "run as workflow" nor "run with databricks connect" when I choose serverless run on my paid account.  However I can perform this action in my free edition account . 

See error : 

pyspark.errors.exceptions.connect.SparkConnectGrpcException: PERMISSION_DENIED: Cannot access Spark Connect. (requestId=e6437897-ac03-4c06-b7a6-5b25ee676d03)

[RobTheCoder]

I'm having the same problem. I'm admin on both the account and the workspace. I get this with `databricks-connect test` too. I've tried updating my databricks-connect version to `==16.4.*` which seems to be the latest version with serverless support.

One thing I have noticed is that even though I can use serverless in the web UI for notebooks and SQL warehouse, I don't see the serverless option under feature enablement in the account settings. I suspect my account is misconfigured so I'm contacting support about that.

[RobTheCoder]

I think adding this to my databricks.yml file helped. I probably did have a permissions problem and just granting these permissions over the whole of the bundle solved my issue.

permissions:
- group_name: my-admin-group
  level: CAN_MANAGE
- group_name: my-user-group
  level: CAN_VIEW

[SP_6721]

Hi @ivan7256 ,

This might be because serverless compute isn't enabled for workflows in your paid workspace.