Feeds:
Posts
Comments

Archive for October, 2011

Problem

We wanted to enable VMware EVC on our cluster (cluster 1) but we couldn’t enable the EVC , because in order to be able to enable EVC – all virtual machines must be powered off , and our vCenter server as a virtual machine, was standing in our way.

Solution

  • Take one ESX server from cluster 1 and insert it into maintenance mode.
  • Remove ESX server from cluster 1 .
  • Create a new cluster (cluster 2)
  • Add the ESX server which you have just removed from cluster 1 to cluster 2 (make sure not to change anything in the storage configuration in that ESX server – the ESX server has access to the old cluster LUNs)
  • Use Live migration (vMotion) in order to migrate vCenter server from cluster 1 to cluster 2

             

  • Power off all virtual machines in cluster 1.
  • Enable EVC on cluster 1 ( stand on the cluster > Edit Settings > VMware EVC > Change > Enable EVC for Intel Hosts)

             

              

  • Connect using the vSphere client directly to the ESX host in cluster 2.
    • Stand on vCenter Server virtual machine.
    • Right click > Edit Settings > Options > General Options > Virtual Machine Configuration File.
    • Write down the full path of the vCenter server virtual machine.
  • Connect using SSH to the ESX host in cluster 2.
    • Navigate to the vCenter path (which you’ve found out in the paragraph above)
    • Copy the full path (including *.vmx file).
    • From the SSH, power off vCenter server. Using the following command : vmware-cmd “Virtual_Mchine_Full_Path.vmx”  stop
    • In order to check if the vcenter server has powered off, run the following command: vmware-cmd “Virtual_Mchine_Full_Path.vmx” getstate
    • After you’ll see that the server is powered off , unregister the virtual machine from the ESX host in cluster 2 , using the following command:
      • vmware-cmd –s unregister “Virtual_Mchine_Full_Path.vmx”

                                         

  • Afterwards connect using SSH to another ESX server in cluster 1.
    • Register vCenter server in that ESX server using the following command:
      • vmware-cmd –s register “Virtual_Mchine_Full_Path.vmx”

                               

    • Make sure that the vCenter server has successfully registered in the new ESX server, and is currently powered off using the following command:
      • vmware-cmd “Virtual_Mchine_Full_Path.vmx” getstate
    • Power on vCenter server virtual machine using the following command:
      • vmware-cmd “Virtual_Mchine_Full_Path.vmx” start
    • Answer vCenter server virtual machine’s question.
      • vmware-cmd “Virtual_Mchine_Full_Path.vmx” answer
    • Create a new identifier by choosing 0
  • Enter the ESX which exist in cluster 2 into maintenance mode.
  • Remove the ESX from the cluster.
  • Add the ESX server to cluster 1.
  • Delete cluster 2.

Fixed! You have just configured EVC in cluster 1, and returned vCenter server virtual machine back to that cluster 1.

If you like you can read about it in the following link:

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1013111

Yaron & Idit.

Read Full Post »

Problem

I connected an external DVD, but I couldn’t use it because I received the following error message: “is not accessible , Incorrect function”  (see photo below)

Solution

Go to device manager and disabled the default CD-ROM device.

Problem solved 🙂

Yaron.

Read Full Post »

Problem

Our file server was suffered from lack of CPU resources, and it reached up to a point that users were not able to login to the file server, and us (as administrators) were not able to do anything on the server in order to troubleshoot it.

Our file server was installed on Windows 2008 storage server x64 edition and connected through iSCSi to Iomega storage system (ix12-300r).

As a solution we’ve installed the file server on a new server (Dell PowerEdge R610) , only this time we’ve decided to install on the server Windows 2008 R2 (and not Windows 2008 storage server) . Afterwards we’ve connected the file server’s storage back through iSCSi initiator, but when we’ve tried to access certain files, we’ve received the following error: “The file cannot be accessed by the system”.

First we‘ve thought that this might be a permission problem, so we’ve done the following:

  • Took ownership on the file/folder
  • We’ve granted Full permissions on that file/folder to our domain user
  • We’ve granted Full permissions on the share itself

But none of these actions worked, and we’ve still received the same error message.

Solution

After a lot of nights without any sleep, we’ve found a solution. The problem was as a result of a service called: “Single Instance Store Groveler”

The Groveler knows to search for and consolidate duplicate files on the volume and as a result will save disk space.

After a test that we’ve done, we saw that after installing Windows 2008 storage server x64 , the Role “File Server” is installed by default and it contains the “Single Instance Storage” Role service (see the photo below) as opposed to Windows 2008 R2 that after installation does not install any role by default, and when we added the role “File server” it didn’t contain that service. “Single Instance Storage” is only available on Windows 2008 storage server.  

The consequence of installing windows 2008 R2 was that windows was not familiar with files which were stored in a hidden/inaccessible  folder called “SIS Common Store” which contains all the source files of the “Single Instance” storage role. (see photo below)

The solution was to reinstall the server with Windows 2008 storage server, and connect the iSCSi disks once again, while  windows was able to access “SIS Common Store” folder, and “Single Instance Store Groveler” was  running – all data returned to be accessible.

You can find additional information regarding “Single Instance Store Groveler “ in the following link: http://technet.microsoft.com/en-us/library/cc978292.aspx

Idit & Yaron 🙂

Read Full Post »