Databricks Workspace Delete struck for couple of days

Question

I deleted my Databricks Workspace, but the process didn't complete. It still shows as deleting after couple of days. I tried the CLI ways also but it says the workspace is deleted. I'm not able to delete or redeploy the Databricks workspace. Also cannot delete the resource group created for this workspace.

I guess that the delete process is struck. How do I resolve this problem?

Answer 1

Hi Gom Kat,
It sounds like your Databricks workspace delete operation is stuck in the “Deleting” state, and because of system-level locks on the managed resource group (MRG), you can’t clear it out or remove the resource group manually. Here’s how you can unblock things:

1. Double-check the workspace deletion status • Run: az databricks workspace show

   --resource-group <YourRG>   
   
   --name <YourWorkspace>  
   

   • Look at the ProvisioningState value. If it’s still Deleting, Azure is still cleaning up behind the scenes. Sometimes this takes longer than the typical 5–10 minutes if there are dependencies or transient backend delays.
2. Use force-deletion to clean up the workspace catalog + MRG If the normal delete path has stalled, you can force the operation to remove the workspace catalog and managed resource group in one shot. • Azure CLI: az databricks workspace delete

   --resource-group <YourRG>   
   
   --name <YourWorkspace>   
   
   --force-deletion  
   

   • PowerShell: Remove-AzDatabricksWorkspace

   -ResourceGroupName <YourRG>   
   
   -Name <YourWorkspace>   
   
   -ForceDeletion  
   

3. If you’re tearing down the entire RG and want to ensure the workspace goes away, you can delete the RG with a force-deletion type for Databricks: az group delete --name <YourRG> --force-deletion-types Microsoft.Databricks/workspaces
4. Wait or escalate • After running the force-deletion, give it a few minutes, then re-run az databricks workspace show to confirm the workspace and its MRG are gone. • If it’s still stuck beyond 30 minutes, open a support ticket—Azure Support can look into any backend locks or hanging operations that your subscription can’t clear on its own.

Hope one of these steps helps you clear out the stuck delete and lets you either redeploy or fully remove the RG. Good luck!

Reference documentation:

• Diagnose and resolve issues with workspace deletion

https://learn.microsoft.com/azure/databricks/admin/workspace/delete-workspace

• Azure CLI delete workspace (including --force-deletion)

https://learn.microsoft.com/cli/azure/databricks/workspace?view=azure-cli-latest#az-databricks-workspace-delete

• Resolve DeploymentBeingDeleted for Microsoft.Databricks workspaces

https://learn.microsoft.com/azure/azure-resource-manager/management/monitor-deployments#deploymentbeingdeleted

• Force-delete resource group with Databricks workspace

https://learn.microsoft.com/azure/databricks/admin/workspace/delete-workspace#force-delete-the-workspace-catalog

Note: This content was drafted with the help of an AI system. Please verify the information before relying on it for decision-making.

Answer 2

Azure Databricks workspace deletion in Azure is soft for 7 days. During this period the workspace appears as deleted from Databricks/CLI perspective, but the Azure resource can remain in a Deleting/stuck state and block redeployment or resource group deletion.

Key points and actions:

1. Understand the retention behavior
   • When a workspace is deleted, Azure Databricks puts the workspace into a soft-deleted state for 7 days before permanent deletion.
   • Azure Databricks–managed metadata (notebooks, jobs, configs) may be recoverable during these 7 days, but Azure resources (VMs, managed disks, networking) are cleaned up immediately and cannot be restored.
   • Treat workspace deletion as permanent and do not expect a guaranteed restore.
2. Why the workspace appears stuck
   • From the Databricks side, the workspace is considered deleted or soft-deleted.
   • From the Azure Resource Manager side, the resource can remain in a Deleting or failed state and block:
     • Recreating a workspace with the same name in the same region/subscription.
     • Deleting the resource group that still “contains” the stuck workspace resource.
3. What can be done by self-service
   • There is no documented self-service “force delete” for a stuck Databricks workspace resource beyond the normal delete operation in the Azure portal or via ARM/CLI.
   • If the workspace deployment had failed originally, the guidance is to delete the failed workspace and then recreate it; however, in this case the delete is already stuck, so this path is exhausted.
4. Recommended resolution path
   • Because the resource is stuck in a backend state and blocks both deletion and redeployment, this requires intervention from Azure support/Databricks backend teams.
   • Create an Azure support request from the Azure portal targeting the Azure Databricks resource and describe:
     • The workspace name, subscription, region, and resource group.
     • That the workspace shows as Deleting for several days and blocks both redeployment and resource group deletion.
   • The backend team can clean up the stuck workspace resource so that:
     • The delete operation completes.
     • The resource group can be deleted.
     • The workspace name can be reused if needed.
5. Preventive recommendations
   • Before deleting a workspace, back up notebooks, jobs, and export any data stored in DBFS that needs to be retained.
   • Verify no production workloads depend on the workspace.

---

References:

• Delete a workspace
• Delete a workspace – Workspace deletion retention policy
• Deploy a workspace using the Azure Portal
• Deploy a workspace with PowerShell
• Deploy a workspace with an ARM template
• workspace command group
• Sign up for Azure Databricks for free
• account workspaces command group
• Accidentally Deleted Resource Group - Microsoft Q&A