locked
Storage Spaces - unmark hard drive as "retired" after SATA controller failure RRS feed

  • Question

  • Hello,

    I am having a problem with Storage Spaces - it won't start recovering the mirror set.
    The mirror set got degraded because one of SATA ports on the controller went bad (I/O errors with any hard drive on that port).
    After the hard drive is reconnected to another port, Storage Spaces still shows the drive as 'Retired'.

    I can unmark the disk by running Set-PhysicalDisk -UniqueID id -Usage AutoSelect

    Then, I can access a Simple (non-mirrored) thin-provisioned virtual disk located on that physical disk, and there are no I/O errors in Event Log, or any other errors.
    But after I attempt to initiate Mirror set repair (Repair-VirtualDisk -FriendlyName mirror), the physical disk becomes 'Retired' again, although no I/O errors have occured.

    Seems that somewhere in the metadata there is incorrect information about actual disk health state, and I need a way to reset it.

    Thank you.

    Saturday, October 13, 2012 10:53 AM

Answers

  • Thank you for reply.

    Figured out a workaround which solved the problem, however, it is somewhat crude (use at your own risk!):
    1. If there are any non-redundant (simple) virtual disks in the storage pool, back up the data somewhere (they have to be deleted and recreated from scratch)
    2. Disable the spaceport driver (sc config spaceport start= disabled), so you can see the hidden GPT partitions of type 'e75caf8f-f680-4cee-afa3-b001e56efc2d'.
    3. Reboot.
    4. Wipe the affected physical hard disk using diskpart (use clean command)
    5. Enable the spaceport driver (sc config spaceport start= boot) and reboot
    6. Now you can add the affected hard disk yet again
    7. At this point - the affected hard disk is listed twice - one in the 'Retired' state, the other is 'Healthy'.
    8. Initiate the repair of the affected mirror virtual disks, and wait for completion
    9. Now you can delete the 'Retired' disk.

    ----
    All that said, it would be great to have an option to un-retire a hard disk in place, without going through all that hassle.
    Set-PhysicalDisk -UniqueID id -Usage AutoSelect does not really work (the disk becomes Retired again when starting repair).
    Remove-PhysicalDisk also fails with 'Not supported' error (unless you do the above workaround).

    Thank you.


    Monday, October 15, 2012 9:04 PM

All replies


  • Hello,
     
    Thank you for your question.

    I am currently looking into this issue and will give you an update as soon as possible.
     
    Thank you for your understanding and support.


    TechNet Subscriber Support in forum |If you have any feedback on our support, please contact tnmff@microsoft.com.

    Monday, October 15, 2012 9:14 AM
  • Thank you for reply.

    Figured out a workaround which solved the problem, however, it is somewhat crude (use at your own risk!):
    1. If there are any non-redundant (simple) virtual disks in the storage pool, back up the data somewhere (they have to be deleted and recreated from scratch)
    2. Disable the spaceport driver (sc config spaceport start= disabled), so you can see the hidden GPT partitions of type 'e75caf8f-f680-4cee-afa3-b001e56efc2d'.
    3. Reboot.
    4. Wipe the affected physical hard disk using diskpart (use clean command)
    5. Enable the spaceport driver (sc config spaceport start= boot) and reboot
    6. Now you can add the affected hard disk yet again
    7. At this point - the affected hard disk is listed twice - one in the 'Retired' state, the other is 'Healthy'.
    8. Initiate the repair of the affected mirror virtual disks, and wait for completion
    9. Now you can delete the 'Retired' disk.

    ----
    All that said, it would be great to have an option to un-retire a hard disk in place, without going through all that hassle.
    Set-PhysicalDisk -UniqueID id -Usage AutoSelect does not really work (the disk becomes Retired again when starting repair).
    Remove-PhysicalDisk also fails with 'Not supported' error (unless you do the above workaround).

    Thank you.


    Monday, October 15, 2012 9:04 PM
  • I had the same thing happening to me today. No problem with the disk, it just got booted from the Storage Pool and marked as retired with no option to remove it because of degraded Virtual Disks which could not be repaired because of the retired disk. Nice catch 22.

    After trying different things and looking on the net for few hours I was just about to pull the drive out and wipe it in another machine when I came across your post and I just wanted to say thank you! You saved me some time, I wish I had found your post earlier.

    Wednesday, October 31, 2012 5:41 PM
  • I have the same problem too.  Nothing seems to be wrong with the disk when I remove it and re-format it, but it gets "retired" again within a couple days of me adding it back to the storage pool.  Does anyone know what can cause drives to become "retired"?
    Sunday, March 3, 2013 11:50 PM
  • Thank you for this; it just helped me with exactly the same problem.

    I'd just like to add, however, that at step 4 I was unsure if I should just clean the GPT partition or the whole drive. I opted to clean the whole drive, which worked, but would it have been possible to save time with repairing the storage pool by only clean the first partition?

    Wednesday, October 23, 2013 11:47 AM
  • I am having virtually the same problem. I am trying to remove a physical disk from a virtual disk, and having retired it find that it cannot be removed because this virtual disk is not in a healthy state. Note that the virtual disk contains 6 physical disks has 3.5 GB allocated of a 13.6 TB.)  Yet repair virtual disk completes successfully.

    I believe my problem of the same as that of this thread.

    Now for my point. It has been over a year since since the moderator stated "I am currently looking into this issue..." without a satisfactory reply. This is frustrating!




    • Edited by Mparisi Wednesday, December 11, 2013 5:21 PM
    Wednesday, December 11, 2013 5:19 PM
  • This solution worked perfectly for me.

    RPP

    Sunday, September 21, 2014 1:59 PM
  • hi,

    this doesnt appear to work on windows 10.

    I followed the steps and I indeed am given the option to add the cleaned disk at 6. however when i select it i get an error 'cant prepare drives clos all applications that are accessing the drive and then try again'

    I've closed everything down nd confirmed onResource Monitor that there is no activity on the disk.

    Anyone know how to solve this one?

    Tuesday, September 1, 2015 1:16 PM
  • The command in question is simply "reset-physicaldisk".

    Monday, April 22, 2019 10:04 PM
  • I had the same problem last week on windows 10 with 6x4tb parity configuration, 7 years after original post. Sad to see no progress on microsoft's part. Reset physicaldisk simply gave an unknown error and failed, as all the other normal methods i tried. This workaround worked like a charm. Although two jobs for repair and regeneration, which i can not fathom why there are two, are still ongoing resulting in currently degraded performance, disk is back to normal state. You are a life saver!
    Saturday, September 28, 2019 2:49 PM