none
Cluster Manager test fails on Storage Section

    Pregunta

  • Running the test for cluster failover and I am not sure if I can proceed to create the cluster with the error that I have.  I understand that there is no failover for the local machine's C: Drive (boot), all other tests pass.  Or is there a way to remove the local drive from the test or from mpio?

    Storage
    Name Result Description
    List All Disks Success
    List Potential Cluster Disks Success
    Validate Disk Access Latency Success
    Validate Disk Arbitration Success
    Validate Disk Failover Failed
    Validate File System Canceled
    Validate Microsoft MPIO-based disks Success
    Validate Multiple Arbitration Success
    Validate SCSI device Vital Product Data (VPD) Success
    Validate SCSI-3 Persistent Reservation Success
    Validate Simultaneous Failover Canceled

    Here is the detailed section:

    List All Disks

      List all disks visible to one or more nodes (including non-cluster disks).
      Prepare storage for testing
      Preparing storage for testing on node tluxhs1.tluxvm.orl.com
      Preparing storage for testing on node tluxhs2.tluxvm.orl.com

      tluxhs1.tluxvm.orl.com

      Getting information on PhysicalDrive 0 from node tluxhs1.tluxvm.orl.com
      Getting information on PhysicalDrive 1 from node tluxhs1.tluxvm.orl.com
      Getting information on PhysicalDrive 2 from node tluxhs1.tluxvm.orl.com
      Getting information on PhysicalDrive 3 from node tluxhs1.tluxvm.orl.com
      Getting information on PhysicalDrive 4 from node tluxhs1.tluxvm.orl.com
      Getting information on PhysicalDrive 5 from node tluxhs1.tluxvm.orl.com
      Getting information on PhysicalDrive 6 from node tluxhs1.tluxvm.orl.com
      Disk Number Disk Identifier Disk Bus Type Disk Stack Type Disk Address (PORT:PATH:TID:LUN) Adapter Description Eligible for Validation Disk Characteristics
      PhysicalDrive0 fceb8fb7 Bus Type SAS Stor Port 2:0:0:0 Dell SAS 6/iR Integrated Controller False Disk is a boot volume. Disk is a system volume. Disk is used for paging files. Disk is used for memory dump files. Disk is on the system bus. Disk partition style is MBR. Disk partition type is BASIC.
      PhysicalDrive1 871348a8 Bus Type iSCSI Stor Port 3:0:0:0 Microsoft Multi-Path Bus Driver True Disk partition style is MBR. Disk partition type is BASIC. Disk is Microsoft MPIO based disk
      PhysicalDrive2 871348bc Bus Type iSCSI Stor Port 3:0:3:1 Microsoft Multi-Path Bus Driver True Disk partition style is MBR. Disk partition type is BASIC. Disk is Microsoft MPIO based disk
      PhysicalDrive3 871348a4 Bus Type iSCSI Stor Port 3:0:3:2 Microsoft Multi-Path Bus Driver True Disk partition style is MBR. Disk partition type is BASIC. Disk is Microsoft MPIO based disk
      PhysicalDrive4 a9429ada Bus Type iSCSI Stor Port 3:0:3:3 Microsoft Multi-Path Bus Driver True Disk partition style is MBR. Disk partition type is BASIC. Disk is Microsoft MPIO based disk
      PhysicalDrive5 87134890 Bus Type iSCSI Stor Port 3:0:0:4 Microsoft Multi-Path Bus Driver True Disk partition style is MBR. Disk partition type is BASIC. Disk is Microsoft MPIO based disk
      PhysicalDrive6 <Unknown> Bus Type iSCSI Stor Port 3:0:2:31 Microsoft Multi-Path Bus Driver False Disk partition style is RAW. Disk is Microsoft MPIO based disk

     

    Any help would be greatly appreciated.

    Robert Ramos

    miércoles, 04 de noviembre de 2009 20:27

Respuestas

  • Actually, it says "Cluster disk 0" which is different than Disk 0.

    Under the "List Potential Cluster Disks" section, it should have a section similar to the following:


    Disks eligible for cluster validation: Disks will be referred to by the following cluster disk numbers in subsequent tests
    Cluster disk 0 has disk identifier f2891a82
    Cluster disk 1 has disk identifier f289197d
    Cluster disk 2 has disk identifier f289197c
    Cluster disk 3 has disk identifier f289197f
    Cluster disk 4 has disk identifier f289197e


    You would use the identifier value listed here to line up the Disk Identifier value from the List All Disks section of the report to figure out which physicaldrive number cluster is using for Cluster disk 0.

    Basically, this report is stating that the node is having issues with the volume on the disk identified as Cluster Disk 0. You might try re-formatting this disk to see if this helps. If this doesn't help, you might consider contacting your storage vendor for further support.

    Visit my blog about multi-site clustering - http://msmvps.com/blogs/jtoner
    jueves, 05 de noviembre de 2009 18:57

Todas las respuestas

  • Cluster knows about your boot drive...in the "Eligible for Validation" column it lists PhysicalDrive0 as not eligible for validation, so it would not test this disk.

    The validation error appears to be occurring during the failover testing. Post the "Validate Disk Failover" section of the validation test.

     


    Visit my blog about multi-site clustering - http://msmvps.com/blogs/jtoner
    miércoles, 04 de noviembre de 2009 23:07
  • Thanks for the help.  Here is the "Validate Disk Failover" section.  It refers to disk 0, which I thought should be exempt.

    Validate Disk Failover

      Validate that a disk can fail over successfully with data intact.
      Getting partition table for cluster disk 0 from node tluxhs1.tluxvm.orl.com
      Arbitrating for cluster disk 0 from node tluxhs1.tluxvm.orl.com
      Bringing online cluster disk 0 on node tluxhs1.tluxvm.orl.com
      Writing file data on cluster disk 0 partition 1
      Failed to write file data on cluster disk 0 partition 1, failure reason: The disk structure is corrupted and unreadable.
      Failed to write file data on cluster disk 0 partition 1, failure reason: The disk structure is corrupted and unreadable.
    jueves, 05 de noviembre de 2009 14:30
  • Actually, it says "Cluster disk 0" which is different than Disk 0.

    Under the "List Potential Cluster Disks" section, it should have a section similar to the following:


    Disks eligible for cluster validation: Disks will be referred to by the following cluster disk numbers in subsequent tests
    Cluster disk 0 has disk identifier f2891a82
    Cluster disk 1 has disk identifier f289197d
    Cluster disk 2 has disk identifier f289197c
    Cluster disk 3 has disk identifier f289197f
    Cluster disk 4 has disk identifier f289197e


    You would use the identifier value listed here to line up the Disk Identifier value from the List All Disks section of the report to figure out which physicaldrive number cluster is using for Cluster disk 0.

    Basically, this report is stating that the node is having issues with the volume on the disk identified as Cluster Disk 0. You might try re-formatting this disk to see if this helps. If this doesn't help, you might consider contacting your storage vendor for further support.

    Visit my blog about multi-site clustering - http://msmvps.com/blogs/jtoner
    jueves, 05 de noviembre de 2009 18:57
  • This guided me down the right path.  Thanks.  Paid close attention to the message stating the disk structure is corrupt or unreadable.  Newly formatted hard drives with a VHD on them but for whatever reason there was corruption.  I ran chkdsk and found directory structure issues and orphaned files.  I can't believe it.  Directory structure is simple...

    Hyper-V/Tluxhs1 or Hyper-V/Tluxhs2

    and whatever Hyper-V Manager puts in there for the configuration.  As far as files the VHD and again whatever Hyper-V Manager puts in there.

    All good after the chkdsk /f.
    • Propuesto como respuesta thflores23 miércoles, 13 de marzo de 2013 14:24
    viernes, 06 de noviembre de 2009 16:43
  • Hi, this is the same case that happens to me. This answer save me. Thanks very much.

    Capa
    sábado, 28 de noviembre de 2009 15:38
  • Glad I could help...feel free to click on that little green arrow above "Vote As Helpful" if this answer was indeed helpful ;-)
    Visit my blog about multi-site clustering - http://msmvps.com/blogs/jtoner
    lunes, 30 de noviembre de 2009 17:04
  • That was helpful.........
    viernes, 05 de marzo de 2010 7:27
  • This totally fixed my issues as well. Who would have thought that chkdsk would be the fix :)

    Thanks

    Rick

    miércoles, 22 de septiembre de 2010 19:23
  • Hiya

    We have this problem as well, but i'm not sure where to run chkdsk. Is it inside the VM that has the physical disk assigned to it, or do I need to remove the storage from the cluster so it's available in Disk Management on the host server, then assign it a drive letter and run chkdsk against this drive on the host?

    Thanks

    sábado, 09 de octubre de 2010 7:26
  • thanks much!  New cluster with NetApp iscsi, thought I was in trouble!  All good now.

    John

    sábado, 23 de octubre de 2010 1:37
  • I too had this problem, however, when I ran chkdsk on the Cluster Disk and Partition it did not return any errors.  I re-ran chkdsk with the "/f" switch and again it didn't return any errors but when I re-ran the Validation Test the volume was validated without error.

    These were newly provisioned, freshly formatted volumes that I presented to the Hyper-V nodes after initial installation.  The volumes had not been used or mounted for any other purpose.  Does anyone know why Failover Cluster Services and the Validation Report would give this error and what chkdsk does to correct it?

    jueves, 21 de abril de 2011 16:56
  • Solution for this and all other Hyper-V Clustering, CSV and Live Migration related problems, are documented here:
    http://zahirshahblog.com/2012/01/19/hyper-v-live-migrati-on-is-not-working-the-disk-structure-i-s-corrupted-and-unre-adable-cluster-res-ource-virtual-machi-ne-vm-in-clustered-service-or-applicat-ion-vm-failed-vi-rtual-machine/
    Zahir Hussain Shah | MVP - Exchange Server | Senior Infrastructure Consultant - Messaging | My blog: http://zahirshahblog.com ]If my answer fixes your problem, mark them as answer, so it will help others to find a solution]
    jueves, 19 de enero de 2012 9:58
  • I just ran into this error when installing a SQL Cluster using mount points. Using JohnToner's troubleshooting tips, I reformatted and ran chkdsk against the drive. Re-ran and got the same error on another drive. I ended up reformatting every single disk and ran the validation wizard again with success. Saved me from making this day even longer.
    lunes, 16 de abril de 2012 23:21
  • Thanks a lot! That was a trick!
    jueves, 05 de septiembre de 2013 14:22
  • This guided me down the right path.  Thanks.  Paid close attention to the message stating the disk structure is corrupt or unreadable.  Newly formatted hard drives with a VHD on them but for whatever reason there was corruption.  I ran chkdsk and found directory structure issues and orphaned files.  I can't believe it.  Directory structure is simple...

    Hyper-V/Tluxhs1 or Hyper-V/Tluxhs2

    and whatever Hyper-V Manager puts in there for the configuration.  As far as files the VHD and again whatever Hyper-V Manager puts in there.

    All good after the chkdsk /f.

    While not necessary, to make the resolution easier for some users reading this, one way to overkill the issue with successful results was run chkdsk /f (as admin) against all of the drives on our server except for the c:\ (boot) drive. Instant pass after this.
    • Propuesto como respuesta Octavio-Admin viernes, 21 de febrero de 2014 18:12
    viernes, 21 de febrero de 2014 18:12
  • I concur with this method. This worked out very well and helped get past this stage. Thanks!
    viernes, 21 de febrero de 2014 18:17