Frequently Asked Questions - Virtual Server Agent for VMware
- How can I calculate the number of licenses required for VMware?
- How does the virtualization client perform automatic load balancing of backup jobs?
- Why is the backup throughput very low during the full backup?
- Can I back up fault tolerance virtual machines using the Virtual Server Agent?
- Why is an incremental backup converted to a full backup after an in-place restore?
- How do I manually install a new version of VDDK on a VSA proxy?
- How do backup and restore operations handle independent/RDM disks?
- How do backup and restore operations handle virtual RDM Disks?
- Using the Virtual Server iDataAgent in HotAdd mode configurations
- What happens if a virtual machine is deleted in the vCenter and a new virtual machine with the same name is created?
- How is white space handled in virtual machine capacity licensing?
- What are the requirements to enable thin provisioned disk restores?
- How can I limit VSA user access to specific resource pools or ESX hosts?
- Are Storage Spaces supported for backups and restores?
- How is capacity licensing calculated for virtual machines?
- Why is my backup data showing volume numbers instead of drive letters?
- Do backups take a snapshot of virtual machine memory?
Generate a Virtual Machine Infrastructure Report to get a summary of protected and unprotected virtual machines that you can use as a basis for requesting capacity licenses or socket licenses for VM protection.
If you are performing a full backup of virtual machines which have thin provisioned disks on NFS datastore, the backup throughput may become very low. VMware does not support the retrieving allocated blocks on NFS volume. Therefore, if a NFS datastore has a thin provisioned disk, the software reads the complete disk during the full backup. This reduces the backup throughput during the full backup. In case of incremental backups, software uses Change Block Tracking (CBT) and thus reads and backs up only the changed data.
For more information, refer to “Changed Block tracking on Virtual disks" section in the following document:
If the Thin provisioned disk is on VMFS volume, the software reads and backs up only the allocated part of the disk.
You can back up fault tolerant virtual machines that meet the following requirements:
- Configured as fault tolerant in the vSphere Web client
- Hosted on ESX 6.x or later
- VM hardware version 11 or later
For earlier versions of ESX or VM hardware, use an in-guest agent.
For more information about backing up fault tolerance virtual machines, refer to http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&docType=kc&externalId=1016619&sliceId=1&docTypeID=DT_KB_1_1&dialogID=381700304&stateId=1 0 381710452
After performing an In-place restore, the first run backup job will always be converted to a Full backup. The system assumes that it is a newly created virtual machine and hence defaults to a full backup.
Version 6.0.2 of the VDDK is automatically installed when you install the Virtual Server Agent.
VDDK files should never be placed in the VDDK installation in the SnapProtect base directory.
To install a newer version of VDDK on a 64-bit computer, perform the following steps:
- Create a new VDDK install folder on the proxy computer (for example, C:\Program Files\VMware\VMware Virtual Disk Development Kit\).
This folder structure already exists if a previous manual 64-bit installation has been performed.
- Download the VDDK package from the VMware download site and extract the files. Copy the bin and lib folders to the new VDDK install folder on the proxy computer.
- Create the following registry key:
HKEY_LOCAL_MACHINE\SOFTWARE\VMware, Inc.\VMware Virtual Disk Development Kit
- Under that registry key, create the InstallPath string and set the value as the VDDK install folder you created for the VDDK executables (for example, C:\Program Files\VMware\VMware Virtual Disk Development Kit\).
- Go to the bin folder under the new VDDK install folder and run the file vstor2install.bat file to install the file-level driver.
For more information see Virtual Disk Development Kit 6.0.2 Release Notes.
If a virtual machine undergoing a backup job includes independent disks or physical RDMs, those disks will be skipped.
If a subclient contains virtual machines with independent disks/physical RDMs, the backup job will always complete with the status "Completed w/ one or more errors". However, if you configure the IgnoreUnsupportedDisks additional setting on the proxy computer, the backup job will complete successfully.
If the Unconditionally overwrite VM with the same name option is used when restoring a virtual machine that has independent disks, the independent disks and their VMDKs are removed from the datastore and are not restored.
Virtual RDMs are protected by the backup job (but not during SnapProtect backup). However at the time of restore, the data is restored as a regular VMDK on a datastore. A virtual RDM is not re-created and the data is not restored to the virtual RDM’s device.
- When deploying the Virtual Server Agent, install the software on a datastore with the largest VMFS block size. This is necessary to ensure that the Virtual Server Agent can mount and back up virtual machines residing on all datastores.
- Helper virtual machines are not required for HotAdd Virtual Server Agent servers using VADP.
What happens if a virtual machine is deleted in the vCenter and a new virtual machine with the same name is created?
In this scenario, even though the name of the virtual machine is same, the new virtual machine will have different GUID/UUID. Therefore, it is recommended to perform following steps:
- In the CommCell console, delete the virtual machine client from the list of client computers
- Perform the backup of the virtual machine. A client will get created again for the virtual machine.
If you do not perform these steps, a new client will get created for the virtual machine with name as follows: <original_virtual_machine_name>_1.
When capacity licensing is used, the used space as reported by the guest OS for the virtual machine is reported against licensed capacity. For example, if the guest OS shows 20GB used on a 100GB disk, 20GB is what counts against the licensed capacity. This is the case even if a different figure is reported for VM disks on the datastore, and is not affected by deduplication or compression.
Disk level backups can use allocated block tracking, which is part of VMware Changed Block Tracking (CBT). Allocated block tracking identifies portions of the virtual machine disk that have not been used, so those portions can be skipped during backup or restore. In 10.0, disk restores can reclaim unused space even if allocated block tracking is not enabled.
During disk restore operations, writes of empty data are automatically discarded if the target disk is thin or eager zeroed. This enables thin provisioned disks to be restored to and from NFS datastores or for other VMDKs. Depending on the state of the data inside the virtual machine, it may also be possible for the restore to discard empty data that was previously allocated. The restore recovers all space that is reported in use (not zeroed out), so the restored VMDK can be larger than the size that was reported by the guest OS.
You can define users and provide them with role-based privileges in vCenter at any desired level: vCenter, datacenter, ESX host, resource pool, or virtual machine.
- On the vCenter server, define a local user who should have access to a specific level.
- In vCenter, define a role with the required permissions.
- At the desired level in vCenter, add permissions for the user and role.
- In the CommCell, add the user information to the Virtual Server instance properties.
For detailed steps, see Add a Custom User with Limited Scope.
Storage Spaces is a feature in Windows 8 and Windows Server 2012 that enables drives to be grouped into storage pools to appear as virtual drives in Windows File Explorer.
You can use the Virtual Server Agent to back up virtual machines and disks to Storage Spaces; but metadata collection is not supported for Storage Spaces. To retrieve guest files, restore the full virtual machine or virtual machine disk files.
Virtual machine backup job data for capacity licensing usage is calculated as follows:
- For streaming backups and backup copy jobs, the capacity usage is based on the guest size for all virtual machines being backed up.
- For SnapProtect backups, the capacity usage is based on the application size for the virtual machines being backed up.
In the License Summary Report, the Job Size column lists guest sizes for virtual machines included in full or synthetic full streaming backups (including backup copy jobs) and application sizes for snapshots.
Capacity usage is summarized as follows:
- Virtual machine streaming backups are included in the Backup license count (depending on the storage policy configuration).
- Backup copy operations are included in the Backup license count.
- Archived VMs are included in the Archive license count.
- SnapProtect backups are included in the Snapshot count.
When the same virtual machine is included in multiple backup jobs, only one of the jobs counts against capacity usage. If a virtual machine is included in multiple subclients, the latest backup provides the size included in overall capacity usage. If a virtual machine is included in both backups and archiving jobs, the guest size for the VM counts toward the Backup license.
Logical volume manager (LVM) metadata processing for volumes encrypted using BitLocker is currently not supported. Decrypting contents of such volumes may not be feasible during backup because decryption requires a recovery password or a decryption key. Because metadata collection for the volume fails, the reported guest size for virtual machines with encrypted volumes may be incorrect and a file-level browse operation for the encrypted volume cannot display file information.
When capacity licensing is used, the used space as reported by the guest OS for the virtual machine is reported against licensed capacity. For example, if the guest OS shows 20 GB used on a 100 GB disk, 20 GB is what counts against the licensed capacity. This is the case even if a different figure is reported for VM disks on the datastore, and is not affected by deduplication or compression.
For more information about virtual machine sizes, see Size Measures for Virtual Machines.
Verifying Backup Job Data for Virtual Machines
To review backup information for virtual machines, generate a VM Backup Report.
Alternatively, you can log in to the CommServe host using qlogin and run the following stored procedure on the CommServe database:
qoperation execscript -sn QS_CLAGetVSADetails –cs commserve_host_name –file file_name -format csv
The resulting output shows the size of the latest backup job for each virtual machine. The size value is the guest size for the VM, or the used space if guest size is unavailable; the size value for all VMs is used in the overall capacity licensing calculation.
The output also shows which license category each VM counts against, and provides instance, backup set, subclient, and job information for each VM.
If hard disk filtering was used for the backup and the drive that contains the operating system was excluded from the backup, the backup is unable to obtain information about drive letters. You can still browse and restore backed up files on the volumes that were backed up.
Virtual Server Agent backups do not snapshot VM memory during backups.
If the File System and Application Consistent option is configured in the Backup Options tab for the subclient, the guest file system and applications are quiesced before taking a hardware snapshot for the backup to ensure data consistency.