How to remove “Unsupported Cluster Configuration” in SCVMM 2012

Recently I had a incident in one of the end user’s environment. 

The scenario is as follow :
The storage LUNs where the Virtual Machines resides had several Hard Disks failed.  There were not enough hotspare disks assigned to it.  The only way is to replace the faulty Hard Disks in the storage and then reconfigure it again.

Once it was done, the LUN was presented back to the Hosts and Failover Cluster recognize the LUN and then VMs were configured.

However, in the System Center Virtual Machine Manager (SCVMM 2012) has an issue.  It contained two identical records.  One is displayed as Running and another as Unsupported Cluster Configuration

Steps Taken

• Ensure the VM is running in the Cluster
• Launch the Virtual Machine Manager Command Shell in the Server that is running the SCVMM 2012
  (right-click run as administrator)
  
• Key in the command as follow in the VMM Command Shell that was launched :
  
  

  Get-SCVirtualMachine | where { $_.Name -EQ "Duplicate-ComputerName"} | fl name, status

  
  Note : Replace the Pink with your computer name that you want to remove from the SCVMM.   Remember the rest of the commands stay, including the quotes.
•  It will display something like the screen below :
  
  
• Next is to execute the command to  remove from the SCVMM database
  

  Get-SCVirtualMachine | where { $_.Name -EQ "Duplicate-ComputerName"} | Remove-SCVirtualMachine –Force

  
  Note : Replace the Pink with the computer name that you want to remove from the SCVMM.  The rest of the commands stays inclusive the quotes.  This is going to remove from the SCVMM DB but not removing the VM.  Remember the VM needs to be up and running.

• In the SCVMM, the record will be removed and you need to do a refresh to once again get the information from the Cluster. 
  Right click on the Cluster in the VMM and click refresh
  
  
  

After the SCVMM has completed the job (getting the information from the Cluster), the Unsupported Cluster Configuration is no longer in the record.  The one that is healthy is displayed in the cluster information.

A thank you to Aidan Finn and Law.

keywords : error 2604, unsupported cluster configuration, system center virtual machine manager, scvmm hyper-v, duplicate computer name in scvmm 2012, how to remove computer from scvmm 

Cluster Disk in Failover Cluster Failed to Start (File Server)

Recently one of the customer had an upgrade in their firmware of their storage.  Their setup is as follow :

2 Hosts in Windows 2012 Server.  Using a shared storage and it’s setup with Windows 2012 Failover Cluster.

After both the hosts were rebooted, the Cluster Disk that was assigned as FileServer in the Failover Cluster failed to start.  From what I know it’s because both the hosts are assuming they are the owner of that particular Disk.  So both are locked to the resources.

I clicked on the “more details” to check it displayed as :

Further probing into the Disk Management and found that the Disk is in RAW. Before this it was in NTFS format.

Next I go into the shared Storage Management (which is the Dell Storage Manager in this scenario), tried to repair but in vain.  It displayed as follow :

After going through the logs as earlier on, found that the resource was locked due to both the hosts are assume they are the rightful owner and both are trying to write at the same time.  In a Windows cluster environment, there’s only one owner at any one time.

Therefore I launched a PowerShell ISE in Administrator mode.

Execute a command to clear the Cluster Disk Reservation on both the hosts

In my scenario as above, in the Disk Management it shows as Disk 6, I execute the command of :

Clear-ClusterDiskReservation –disk 6

 Note : change only the one in Pink according to the info from disk management

Resource of the command : https://technet.microsoft.com/en-us/library/ee461016.aspx 

After that, I started the FileServer and no issue.  The format was in NTFS and content was intact.

In the Failover cluster of the File Server :

Hopefully this helps.

keywords : cluster disk reserve, error 219, dell storage manager, disk in RAW format, incorrect function, cannot start up cluster disk file server, the requested resource is in use, persistent reservation

SCDPM – Protecting SQL 2012 & above.

To protect SQL Server 2012 or SQL Server 2014, the following must be done.

• Login to SQL Server Management Studio

• Add the system account NTAuthority\System to the Sysadmin group on SQL Server
• DPM cannot protect databases that are stored on remote SMB stores
• Availability group replicas are configured as read-only
• When you perform an alternate location recovery for a partially contained database, you must ensure that the target SQL instance has the Contained Databases feature enabled

Hope the above illustration helps

 

 

Azure–Powershell (Updated)

Recently the PowerShell for Azure, there’s an update that you can download from the GitHub

The Link is located in here.

You need to have the Azure SDK installed first.  The installation file can be download from this link.

Hope this helps.

keywords : Azure, powershell, azure command-line,

Hyper-V VSS Writer State : [7] Failed Last error : Timed out

Had an issue of the Hyper-V VSS when I typed to check.

vssadmin list writers

In the event viewer I got as the following :

\

I want to get the VSS back to be stabilize but not wanting to restart the server, therefore I restart a service.

Service to restart : Microsoft Virtual Machine Management

Keyword :

Volume Shadow Copy Service error: Unexpected error querying for the IVssWriterCallback interface. hr = 0x80070005, Access is denied.. This is often caused by incorrect security settings in either the writer or requestor process.Operation:  Gathering Writer Data Context:  Writer Class Id: {e8132975-6f93-4464-a53e-1050253ae220}  Writer Name: System Writer  Writer Instance ID: {bc497fa6-6627-47fb-a005-ac94fa83904e}Error : 8194 vssadmin list writer Hyper-V State 7 timed out

SCDPM–Job Failed to Cancel or Hung or Stuck BUT Cannot be remove

Environment
SCDPM 2012 SP1

It’s been hours the job has been cancelled but couldn’t be remove from the DPM job.  Can’t execute the next job due to the job is still in progress to cancel.   However, found the issue is with the SQL Database that is holding the job from being release. 

In the SCDPM console, it displays as below :

The following steps were carried out :

• Launch the SQL Server Management Studio (<right-click> run as Administrator)

Once it’s launched, ensure DPM DB is connected

Copy the script below :

USE DPMDB BEGIN TRAN -- mark replica as invalid if there was some operation happening on that replica UPDATE tbl_PRM_LogicalREplica SET Validity = 1 -- Invalid WHERE OwnerTaskIdLock IS NOT NULL AND Validity <> 5 AND -- ProtectionStopped Validity <> 6 -- Inactive -- Release all the locks held UPDATE tbl_PRM_LogicalREplica SET OwnerTaskIdLock = null, Status=8 if (select COUNT(name) from tbl_AM_Agent where Name like 'DPM RA v2%') > 0 begin exec sp_executesql N'UPDATE tbl_RM_ShadowCopy SET ArchivetaskId = NULL, RecoveryJobId = NULL' end UPDATE tbl_ARM_Datasource SET Status = 0, OwnerLockId = NULL DELETE tbl_RM_DatasourceServerlock DELETE tbl_RM_ShadowCopyLocks -- Set All running tasks and jobs to failed UPDATE tbl_TE_TaskTrail SET ExecutionState = 3, LastStateName = 'Failure', StoppedDateTime = GetUtcDate() WHERE ExecutionState NOT IN (2,3) UPDATE tbl_JM_JobTrail SET JobState= 'Failed', EndDateTime = GetUtcDate() WHERE jobstate= 'Execute' OR jobstate= 'Retire' -- unreserve resources held UPDATE tbl_MM_Global_Media SET ReservationLevel = 0, ReservationOwnerMMId = null UPDATE tbl_MM_Global_Drive SET ReservationLevel = 0, ReservationOwnerMMId = null UPDATE tbl_MM_Global_IEPortResource SET ReservationLevel = 0, ReservationOwnerMMId = null COMMIT TRAN

 

• Click on New Query
• Paste the Script into the Query Column
• Click Execute

• Check on the OUTPUT column below
• Close the SCDPM console and open it back.  It should clear the status and be able to execute whatever job that is outstanding.

Hope this helps.   Works on SCDPM 2012 R2 too. 

Resource of the script is from here.

Thank you.

keywords : SCDPM, Data protection manager 2012 SP1, SCDPM 2012 SP1, SCDPM 2012 R2, Job cancel but hang, cancel job failed

PowerShell–Get-VM command is not recognize or missing

Currently I’m supporting one of the customer, found out that the PowerShell is in version 1.0 in the SCVMM.  Therefore, I tried to execute a script using the get-vm but it says as not recognize.

The message was :

PS C:\Users\Administrator\Desktop> Get-VM The term 'Get-VM' is not recognized as the name of a cmdlet, function, script file, or operable program. Check the spelling of the name, or if a path was included, verify that the path is correct and try again.

As a workaround, I’ve execute the command as below :

Import-Module virtualmachinemanager

Then the get-vm command works.

keywords : SCVMM, get-vm missing, get-vm command not recognize, powershell command not recognize.