An Apache Spark-based analytics platform optimized for Azure.
Hi Cardio Bricks,
This issue occurs due to a backend inconsistency in the Azure Databricks resource provider, where the workspace gets stuck in a Deleting state and the system deny assignment on the managed resource group (MRG) is not released. Because of this, the workspace shows a conflicting state (Succeeded / Deleting), deletion requests return "ApplianceBeingDeleted", and the managed resource group cannot be removed due to DenyAssignmentAuthorizationFailed, which also keeps dependent resources locked.
To resolve this, first ensure that there are no active clusters or running jobs associated with the workspace if you still have access. Then retry deleting the workspace using Azure CLI:
Bash
az databricks workspace delete \
--resource-group <WorkspaceRG> \
--name <WorkspaceName> \
--yes
The delete operation is idempotent, so re-running it allows Azure to continue and complete any partially executed backend deletion process. The managed resource group should not be deleted manually because it is system-managed and protected by deny assignments until the workspace deletion is fully completed.
This is a transient platform state during deletion, and retrying the workspace delete operation will allow Azure to reconcile the state and automatically complete the cleanup of the managed resource group and its dependent resources.
Reference Links: Azure Databricks workspace deletion Deny assignments in Azure Delete resource group troubleshooting
Hope this helps. If you have any follow-up questions, please let me know. I would be happy to help.
Please do not forget to "up-vote" wherever the information provided helps you, as this can be beneficial to other community members.