none
Cluster Disk Driver Letter Fiasco RRS feed

  • Question

  • I have a 4 node cluster and my Cluster Disk on node1 is assigned a drive letter D:

    If I fail over my node1 to node2, the Drive Letter from Node1 gets assigned to Node2 and the Node2 Drive letter D: disappears

    I am bit concerned, if i have something running on Node2 on Drive D: that will get lost?

    I know in clustering u have to resreve the Drive Letters so that no other node uses those drive letter; but whats the solution to this in 2019; I dont want to use CSV



    Thanks

    SV


    • Edited by vai2000 Monday, September 9, 2019 5:02 PM
    Monday, September 9, 2019 5:01 PM

All replies

  • Hi,

    Thanks for your question.

    For your concern << if i have something running on Node2 on Drive D: that will get lost? >>

    Based on my experience, the data on this drive wouldn't get lost unless you remove this drive from the cluster or unmount it.

    Did this cluster disk assign to a clustered role?

    We can refer to this blog , and check the FCM GUI, or use the powershells "get-disk" to show the disk info.

    https://www.altaro.com/hyper-v/failover-cluster-manager/nodes-storage/

    The cluster automatically determines the Resource Name by using the text “Cluster Disk” and incrementing a number. Disk Info helps you to identify what is being connected, as it does not read volume information such as labels. The signature can also be used to identify the disk; it’s retrievable by using Get-Disk. When adding several disks at once that are of equal size, be certain to match them when accessing this screen as it will not be so readily available after being attached to the cluster. Check the box(es) for the disk(s) you’d like to add and click OK. Each disk should then appear in the center pane:

    The next item in the Disks sub-node’s context menu is Move Available Storage. Its sub-options are the same for virtual machine migrations: Best Possible Node and Select Node. These items operate only on standard cluster disks; quorum disks and Cluster Shared Volumes are unaffected. Every single cluster disk is moved if possible.

    The remaining options in this node are the standard ViewRefresh, and Help items which work as they do elsewhere in Failover Cluster Manager.

    Besides, may I confirm that you onlined this disk on the destination node you failover? We can check it on the disk management of the node.

    Hope above information can help you.

    Highly appreciate your effort and time. If you have any question or concern, please feel free to let me know.

    Best regards,

    Michael


    Please remember to mark the replies as an answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com


    Tuesday, September 10, 2019 9:08 AM
    Moderator
  • Well, what I saw after failover was node2 had a Drive D with a different LUN, that got assigned as \\GUID in diskmgmt and Drive D from Node1 got Mounted as Drive D on node2.

    SV

    Tuesday, September 10, 2019 2:57 PM
  • Hi,

    Thanks for your reply.

    May I confirm and check if there's another disk used Drive Letter D: on the local storage of the node 2?

    Due to when the cluster disk failover, if a disk with a same drive letter on the other node, probably it happened your issue. We suggest to name a different drive letter with the local disks of the nodes.

    Meanwhile, please check your cluster nodes to patch the latest update. Please refer to this docs the latest update for Windows server 1903 version.

    https://support.microsoft.com/en-us/help/4515384/windows-10-update-kb4515384 

    Hope this helps.

    If you have any question or concern, please feel free to let me know.

    Best regards,

    Michael


    Please remember to mark the replies as an answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

    Wednesday, September 11, 2019 11:29 AM
    Moderator
  • Hi,

    Just checking in to see if the information provided was helpful. Please let us know if you would like further assistance.

    Best Regards,

    Michael


    Please remember to mark the replies as an answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

    Thursday, September 12, 2019 9:44 AM
    Moderator
  • But I will be running out of letters if I start assigning Drive Letters, I have 17 LUNs

    SV

    Wednesday, September 18, 2019 4:30 PM
  • Could you provide more information about your configuration?  It sounds like you do not have the hosts configured the same before you configured your cluster.  You state you are "a bit concerned, if i have something running on Node2 on Drive D: that will get lost?".  You should not have such a configuration.  You should not have a drive D: configured on one node of a cluster that is not configured on another node of the cluster.

    Did you run the cluster validation for your configuration?


    tim

    Thursday, September 19, 2019 1:01 PM
  • Tim, I was speculating that scenario, as of now I know the cluster drives are RESERVED Letters & two hosts can't have the SAME drive letters

    But what if I have 20+ Luns, I will Run out of Drive letters

    Thanks


    SV

    Thursday, September 19, 2019 4:08 PM
  • "what if I have 20+ Luns, I will Run out of Drive letters"

    You don't need to assign drive letters to every volume.  Only if the application requires a drive letter to you have to use a drive letter.  Most applications can make use of the volume GUID or mount points.  Both have been supported in clusters for many years.


    tim

    Friday, September 20, 2019 2:06 PM
  • Hi,

    Just want to confirm the current situations.

    Please feel free to let me know if you need further assistance.

    Best regards,

    Michael


    Please remember to mark the replies as an answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com


    Thursday, September 26, 2019 10:35 AM
    Moderator
  • Hi,

     

    Was your issue resolved?

     

    If you resolved it using our solution, please "mark it as answer" to help other community members find the helpful reply quickly.

    If you resolve it using your own solution, please share your experience and solution here. It will be very beneficial for other community members who have similar questions.

    If no, please reply and tell us the current situation in order to provide further help.

     

    Best Regards,

    Michael


    Please remember to mark the replies as an answers if they help.
    If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

    Monday, October 7, 2019 8:07 AM
    Moderator