Azure SQL Managed instance delete failed

Question

Your Managed Instance is in deletion process after 24 hours still in progress why?

Step 1/1
Request validation
Failed

Creating managed instance
Requested parameters
GP / Gen5 / 4vCores / 32GB

Step 2/2: Virtual cluster deletion
In Progress

Executed delete command on 1 selected items
Succeeded: 0, Failed: 1, Canceled: 0.
Error details
managedsqlinstance: Subscription '...............' does not have the server 'managedsqlinstance'. (Code: ServerNotInSubscription)

Answer 1

Welcome to the Microsoft QnA forums Shahid and Happy New year

This issue is familiar and I also get error from time time to time when I delete Managed Instance. There are a few thing you can do

(1) Try again in "the right order"

There is a dependencies between objects related to the managed instance and many times the issue is with specific object but since other object depended on that one, then you cannot delete them as well.

Recommendation! when you create managed instance in most cases, I recommend to use special separate Resource Group. This makes it simpler to manage and delete the managed instance when needed since we know that everything in the resource group related to the MI and should be deleted - we can try to delete the entire resource group and let the Azure system manage the order of the deletion .

Try to delete one object at a time and check the detailed error information if you get error. This will inform you if there was a different entity which this object depended on -> use this information to delete the depended object first.

(2) Open a support ticket from your Azure subscription

Usually support ticket require a support plan, except for issues related to billing for example. This issue related to billing since while the MI exists you still pay for it, so issue with deleting related to payment which you do not succeed to end. Therefore, you can open a free support ticket (choose category billing) even without a support plan.

Ask the team to delete the specific MI

* By the way, It happened to me just a few days ago (again) and option one solve the issue but took multiple hours