Cannot delete Azure Backup restore point due to DiskRestorePointUsedByCustomer

Dmytro Hureshidze 0 Reputation points

Hi, all! Compute RP reports active SAS handle, but diskRestorePoint resource does not exist in ARM. This is an Azure platform inconsistency and requires backend cleanup.

The VM restore point is stuck in Failed state because Compute RP reports an active SAS handle (DiskRestorePointUsedByCustomer), but the corresponding diskRestorePoint does not exist as an ARM resource. The diskRestorePoint ID exists only in restore point metadata, but ARM returns NoRegisteredProviderFound for the resource. ARM cannot call EndGetAccess because the disk restore point is missing. This is an inconsistent state between ARM and Compute RP and requires backend cleanup.

Subscription: <PII Removed>

Resource group: <PII Removed>

Restore point: <PII Removed>

Missing disk restore point: <PII Removed>

Region: westus2

The portal won't let me submit a support ticket for technical issues without paying. But I just finished a hands-on lesson in the Azure Administration course and, as usual, deleted the entire resource group at the end of the lesson. And now I'm stuck...

  1. Dmytro Hureshidze 0 Reputation points

    @Sina Salam please help me, if you can escalate this issue to the support team. This is the same problem as in the case described here

  2. Siva shunmugam Nadessin 10,895 Reputation points • Microsoft External Staff • Moderator

    Dmytro Hureshidze, Kindly reply to our private message for us to create support ticket?

Sign in to comment

2 answers

  1. Siva shunmugam Nadessin 10,895 Reputation points • Microsoft External Staff • Moderator

    Hello Dmytro Hureshidze,

    Based on the information you provided below steps helped to resolve the issue.

    1.Releasing the stale SAS (access) handle in the Compute service.

    2.Internally completing the access cleanup (EndGetAccess equivalent).

    3.Cleaning up orphaned restore point metadata.

    4.Synchronizing state between Azure Resource Manager (ARM) and Compute.

    5.Removing the affected restore point / restore point collection.

    If the assistance was helpful, kindly “Accept Answer” & up-vote this can be beneficial to other community members.

    1. Siva shunmugam Nadessin 10,895 Reputation points • Microsoft External Staff • Moderator

      Hello Dmytro Hureshidze, kindly 👁 210246-screenshot-2021-12-10-121802.png
      this can be beneficial to other community members.

    Sign in to comment

  2. AI answer

    The error DiskRestorePointUsedByCustomer occurs when there is an active Shared Access Signature (SAS) on a disk restore point. A restore point cannot be deleted until all active SAS handles are ended.

    In this scenario, the platform believes a SAS is still active on an underlying disk restore point, but the corresponding disk restore point resource is no longer accessible through ARM. This leaves the restore point in a stuck/failed state and prevents normal cleanup.

    From the available guidance, the only supported remediation paths for restore point/backup cleanup are:

    1. Ensure there are no locks on the restore point resource group
      1. In the Azure portal, go to All resources and open the restore point collection resource group (for Azure Backup, this is typically AzureBackupRG_<region>_1, which matches the resource group in this case).
      2. Under Settings > Locks, delete any locks on that resource group.
    2. Trigger cleanup of restore point collections After locks are removed, use one of the supported cleanup methods:
      • On-demand backup cleanup path (if the VM and backup configuration still exist):
        1. Trigger an on-demand backup of the protected VM.
        2. The first on-demand backup may fail, but it initiates automatic cleanup of restore points in the restore point collection.
        3. After a few hours, retry the scheduled or on-demand backup.
      • Manual cleanup from the Azure portal (if the restore point collection is still present):
        1. In the Azure portal, go to All resources and open the resource group AzureBackupRG_<region>_1.
        2. In the resource group Overview, select Show hidden types.
        3. Locate the restore point collections with names like AzureBackupRG_<VMName>_<number>.
        4. Select the relevant restore point collection and choose Delete.
        5. If deletion times out, retry the delete operation two or three times until it succeeds.

    If the restore point collection or disk restore point resources are missing or inconsistent (for example, ARM reports NoRegisteredProviderFound for the disk restore point while Compute still reports an active SAS), this is a platform-side inconsistency. The documented remediation for move/backup issues caused by restore point collections is to stop backup and delete the restore point collections; when that is not possible due to inconsistent state, the remaining resolution is backend cleanup by Azure Support.

    Because the portal is not allowing technical support ticket creation on this subscription, the supported options are:

    • Verify again that the restore point collection resource group AzureBackupRG_westus2_1 exists, has no locks, and that all hidden restore point collection resources inside it have been deleted as described above.
    • If the restore point collection cannot be deleted or the error persists even after repeated delete attempts, this falls under the category of platform inconsistency that requires backend cleanup by Microsoft. In that case, use one of the public support channels referenced in the documentation (Azure forums on Microsoft Q&A or Stack Overflow, or @AzureSupport on Twitter) to request that the restore point and its phantom SAS handle be cleaned up for the specified subscription, resource group, and restore point name.

    References:

    AI-generated content may be incorrect. Read our transparency notes for more information.
Sign in to answer

Your answer