Reading Time: 2 minutes

As written in the VMware Data Recovery (VDR) as a backup solution post, VDR could be a possible backup solution for small environment (if there are big budget constraints).

But some issues are really quite difficult to handle with this appliance and the biggest one is the integrity of CIFS NAS destinations.

If you cannot avoid them and for some reason you cannot limit them to 0.5 TB for each destination, you may have some integrity check errors (sometimes).

Most of them could depend by some damaged restore points, in this case you can view them in the restore tab and choose to delete. Usually after a new integrity check (when also a reclaim occurs) those restore points will be fixed. But in some cases this trick does not work, or other errors are the reason of a integrity check fail.

In this case the only solution is follow the KB 1015901 (VMware Data Recovery backups fail with the error: failed, error -2241 (Catalog File invalid/damaged)) to recreate the catalog files:

  1. Verify that there are no operations running on the appliance.
  2. Stop the datarecovery service. See  Restarting the datarecovery service on a VMware Data Recovery appliance (1036768)
       service datarecovery stop
  3. Delete the following files in the dedupe store mount point:
    • /{dedupe mount point}/VMwareDataRecovery/BackupStore.cat
    • /{dedupe mount point}/VMwareDataRecovery/BackupStore.cat.bak
      Note: The restore points are temporarily deleted.
  4. Start the datarecovery service. A restart of the VDR virtual machine will similarly restart the service.
    service datarecovery start
  5. Run an integrity check on the destination directory to recreate the catalog files. The integrity check rebuilds the BackupStore.cat files, and repopulate VDR with all restore points.
Reading Time: 3 minutes

In the previous post I’ve esplain how to install XenServer 6.0 (but can works also with previous versions) in a VM, using VMware products like Workstation 8 or vSphere 5.

To build a XenServer Pool (that is basically the same of a VMware Cluster) you need at least two hosts and a shared storage (and of course a shared network). I suggest to do not clone the second host from the fist because XenServer works with several UUID and in this way you will clone also the host UUID and XenCenter will be not able to identify which is the original and the cloned host. There are also several issue with the networking (the original MAC address remain in the bridged interface and you have to put the new MAC in the sysconfig/network-scripts config file).

After you have build the new host you can create a shared in two different way: using an iSCSI target or using the “guest cluster” configuration (if you work with vSphere).

continue reading…

Reading Time: 3 minutes

As written in a previous post, Citrix with XenServer 6 is trying to fill the gap from VMware vSphere… and most features (included distributed virtual switches) are now common also in this type of solution. Of course features are only one of the possible criteria to select an hypervisor and a virtualization solution.

To make some practice or just to have a look at the product, it is possible run it in a VM on VMware products, both Workstation 8 (and probably also 7) and vSphere 5 can be used.

continue reading…

Reading Time: 2 minutes

During the Partner Exchange (PEX) of last year, VMware has announced a certifications path for the Enterprise Desktop Virtualization area (for more detail see this page). At the event of this year other there was some update on this path.

After his beta period, the VCP5-DT exam is now available, as also described in the official page. To become a VCP5-DT you have to:

  1. Be a VMware Certified Professional on vSphere 5 (VCP5)
  2. Pass the VCP5-DT exam

No requirement for a VCA certification, as was also modified in the VCP4-DT. No advantages for who is already a VCP4-DT: no discount and simplified path.

The VCP5 is quite strange, because View 5 can run also on vSphere 4.x, so IMHO a simple VCP4 or VCP5 could be a better requirement.

But there are also some good changes: the price is finally the same as a usual VCP exam (when I took the VCP4-DT it was more expensive) and, most important, same testing center of VCP (and not of VCAP).

And what’s happen with other DT certifications? There will exist a VCA5-DT? Sound strange there is no rumors about it, but also for VCAP5 the DCD beta was released before the DCA.

But more important… where is the VCAP-DT certification announced one year ago? Still no news about it… at this point seem sure that there will not exist a VCAP4-DT… let’s see if at least the VCAP5-DT will born before View 6.

Reading Time: < 1 minute

For the vExpert 2012 program there are some big changes, starting from the new paths:

  • Evangelist Path: The Evangelist Path includes book authors, bloggers, tool builders, public speakers, and other IT professionals who share their knowledge and passion with others with the leverage of a personal public platform to reach many people. Employees of VMware can also apply via the Evangelist pathway.
  • Customer Path: The Customer Path is for internal evangelists and community leaders from VMware customer organizations. They have contributed to success stories, customer references, or public interviews and talks, or were active community contributors, such as VMUG leaders.
  • VPN (VMware Partner Network) Path: The VPN Path is for employees of our partner companies who lead with passion and by example, who are committed to continuous learning and to making their technical knowledge and expertise available to many. This can take shape of event participation, video, IP generation, as well as public speaking engagements.

Although there will be three explicit paths, there will be only a single vExpert designation. The exact requirements for becoming a vExpert in each path are in applications (open on February the 15th). You will find all the details at vmware.com/go/vexpert2012

Reading Time: 2 minutes

As written in the previous post related to the upgrade to vSphere 5, the old VCB framework can still be used also with latest version, but is a not supported configuration (I’ve made the test only with a Windows Server 2008 R2 for the VCB proxy, but I suppose that works also with previous OS). Of course the SAN transport mode could not work (or not if you use VMFS5), but the NBD works as on the old vSphere 4.x.

The only issue that I get (but honestly I’ve make only few tests) is on VM with Windows Server 2008 or 2008 R2, both with virtual hardware 7 or 8. The error message is quite strange and generic:

The workaround is to set the disk.EnableUUIDparameter in the virtual machine settings to False by performing the following steps:

  1. Power off the Windows Server 2008 virtual machine through vSphere Client.
  2. Click Settings > Options >General > Configuration Parameters.
  3. Set disk.EnableUUID to False.
  4. Save the settings.
  5. Power on the virtual machine.

Note that the option is exactly the opposite that the default one and also is not the same that you may find in VM backup troubleshooting for most backup programs (for example see the VDR note).

By the way, if you have at least the Essential Plus license you can use VDR instead of VCB. Or could be the right time to choose a real backup program.

Reading Time: 2 minutes

Some days ago I found a strange issue with vSphere 5 where some VMs was not movable with Storage vMotion with a strange error: The method is disabled by ‘SYMC-INCR dd-mm-yyyy hh:mm’.

From the name SYMC my first though went on Symantec Backup Exec (the 2010 R3 version) and may on some corrupted snapshots: but the VM file where good, and there was nothing to be consolidate.

I’ve looked in the VMware KB and I found the KB 2008957 (Storage vMotion fails with the error: The method is disabled by ‘SYMC-INCR dd-mm-yyyy hh:mm’) with the symptoms and an a possible explanation: seems that Backup Exec (but probably also NetBackup) add an information (_DP-VEPA_BACKUP_RESTORE_LOCK_) to the VM before the backup that it will removed after the job. This info seems stored in the vCenter Database.

In fact, the suggested workaround for issue is:

  1. Power off the virtual machine.
  2. In the vSphere Client, right-click the virtual machine and click Remove from Inventory.
  3. Browse to the datastore where the virtual machine resides.
  4. Right-click the .vmx file of the virtual machine and click Add to Inventory. This gives the virtual machine a new ID.
  5. Power on the virtual machine and retry the Storage vMotion.

Unfortunately is not a good solution because it temporally stop the VM. In some cases (but not in all) a full backup could fix the issue.

But probably the best solution is fix it in the vCenter Database. The KB 1021265 (Manually enable vSphere solutions) explain how works the extensible framework and probably by removing the entries in the related table the issue can be fixed.

© 2025-2011 vInfrastructure Blog | Disclaimer & Copyright