Frequently Asked Questions - Windows File System iDataAgent

Table of Contents

General

Can I create multiple subclients and backupsets to organize the data?

You can create different subclients to organize the data on a client computer. The data that is backed up by any default subclient is the complement of the data that is backed up by its sibling subclients. In effect, the data associated with a default subclient is implicitly determined by the user defined subclients within the same backup set. 

For example: A client computer called onyx has only one drive, c:\ which has the following folders which you want to backup:
\international\forecasts
\international\hr
\domestic\hr

Using backup sets, you can establish different subclient groups, which will enable you to back up and restore all of the data relating only to international operations or all of the data relating only to human resource operations. You can organize the data as follows:

Backup Sets Subclient Subclient Content Storage Policy
BK1 default C:\ Storage Policy A
  International \international Storage Policy B
BK2 default C:\ Storage Policy C
  HR \domestic\hr

\international\hr

Storage Policy D

Note that while the default subclients in BK1 and BK2 have the same name (i.e., default), they back up different data subsets.

How do I backup the Quorum data in a cluster setting?

The Windows File System iDataAgent on a physical node of the cluster includes the quorum data during a system state backup. It is essential for quorum data to be backed up, as it includes recovery logs as well as cluster configuration and state data. To back up the quorum data on a regular basis, you must ensure that a data protection operation is scheduled for a Windows File System iDataAgent subclient on the physical nodes, with the Backup System State option selected; a system state backup will automatically include quorum data.

Why are my snapshots missing from the backup?

When backing up a snapshot of a client computer, ensure that the operating system of the backup server is either same or higher version than the client computer.

Why am I not able to see the newly added data in the backed up files?

It is recommended to perform a full backup after changing or adding and subclient content to ensure that the newly added or modified data is completely backed up.

Why does my backup and recovery display inconsistent data?

Ensure that the volume, network or the user quota for an object are not set lower than the amount already used.

Do I need to install QSnap while performing the full system restore?

Depending upon the backup from which you are restoring the system, install the QSnap either before the full system restore or after the full system restore. The following table explains the scenarios to reinstall the QSnap:

Backup

Install QSnap

QSnap Enabled Backup Install QSnap before performing the full system restore.
QSnap was installed on the source computer after the backup was performed.
  • Install the Windows File System iDataAgent and then perform the full system restore. Then install QSnap after the client is restored and has rebooted.
  • If you are not rebuilding the operating system, uninstall QSnap, restore the registry, and then reinstall QSnap.

How to backup the source data for a Directory Junction Point?

By default, the directory junction point gets backed up but the source data (data linked to junction point) is not backed up. This prevents undesired backup and restore of system created directory junction points. To back up the source data of a junction point that is not part of a subclient contents, the source data must be explicitly added to a subclient's content.

When does a non-full backup automatically get converted to a full backup?

A non-full backup is automatically converted to a full backup in the following situations:
  • First backup of the subclient.
  • Re-associating a subclient to another storage policy.
  • Promote a secondary storage policy copy that is not synchronized with a primary copy (for all the subclients of a storage policy).
  • If a backup job within the most recent backup cycle is pruned or disabled from a primary copy.
  • If you switch from a SnapProtect backup to a traditional backup or vice versa.
  • CommCell Migration operation.

Do we stage components in the Job Results folder during the system state restore?

During the system state restore some components are staged in the job results folder and then get deleted once the restore of that component is complete. If a file needs to be moved during reboot, the file will not be deleted from the job results folder until the computer reboots.

Backup operation is slow. How do I troubleshoot the slowness?

If you are experiencing performance issues during backup, you can troubleshoot them by enabling logging of performance details in the log files. These performance counters contain information that help in resolving the performance related issues during backups. For more information on viewing log files, see View the Log Files of an Active Job.

Performance counters printed in the log files include:

File Open Time Time taken to open a file.
Disk Read Time Time taken to read the file from a disk.
Pipeline Allocation Time Time taken to assign or distribute the network buffers on the client computer.
Pipeline Write Time Time taken to write data in to the network buffers on the client computer.

The clBackup log displays the time taken by a particular process. For example, a log snippet may look as follows:

ID=File Open time, Bytes Read = 0, Total time = 0.011359, Average = 0.011359, Samples = 1
ID=Disk Read time, Bytes Read = 20, Total time = 0.013465, Average = 0.001417, Samples = 1
ID=Pipeline Allocation time, Bytes Read = 0, Total time = 0.000010, Average = 0.000010, Samples = 1
ID=Pipeline Write time, Bytes Read = 1160, Total time = 0.000004, Average = 282.587336, Samples = 1

Recommendations:

  • If the time taken by Read operation is high, use the DiskReadWrite utility to verify the disk read performance.
  • If multiple reads are enabled on the disk, verify that the hardware you use supports multiple reads else enabling multiple reads might degrade the disk read performance.
  • You can use Raid disks or disks that allow multiple simultaneous reads which improves the performance of the backup job.
  • If the backups are slow due to network or media problems, use the testport, DiskReadWrite and TapeTool utilities to troubleshoot the slowness.
  • If the media is a tape, use the TapeTool utility. If the media is a disk, use the cvdiskperf utility in the write mode.

    For more information about all the utilities, refer to Tools and Utilities.

Why does the backup job complete after the Scan phase?

If no files from the subclient content are added or changed since the last backup, the backup job will complete after the scan phase. The backup and archive index phases will not be performed for such backup jobs.

However, if you want the backup job to go through all phases irrespective of whether the subclient content has been changed or not, enable the bRunAllJobPhases additional settings on the computer where the File System iDataAgent has been installed.

To view the jobs that were marked complete after the scan phase, select Include Media Deleted Jobs in the Selection tab of the Job Summary Report.

Can I backup the data which is encrypted using the McAfee Endpoint Encryption?

Yes. You can perform the backup and restore of the data which is encrypted using the McAfee Endpoint Encryption.

Why are the number of objects shown higher than the number of objects actually backed up?

The total number of objects shown will be more than the actual number of objects backed up because empty folders with no files/content are also considered during the count.

For example, if you have 10 files and 1 empty folder under the root directory, the total number of objects backed up would be shown as 11 as the empty folder was also counted as an object backed up.

Which Retention Criteria Takes The Maximum Precedence?

You can  specify the time period for which you want to retain the deleted files in the backup cycle. This enables you to restore the deleted files from the latest backup cycle and avoid point-in-time restores. This is useful to view (and restore) the deleted files from clients (like, laptop clients) that have backup cycles with frequent & numerous incremental backups after a full backup.

By default, when you delete any file from the source, they will be removed from the backup cycle during the next backup. You can specify the retention period for each subclient. When you run the synthetic full backup at the end of specified retention period, the deleted files will be removed from the backup cycle.

The retention criteria set at the storage policy level takes a higher precedence over the one set at the subclient level.

Example 1:

  1. A subclient named <subclient 1> is associated to a storage policy with a retention value set for 4 days and 2 cycles.
  2. A retention value of Keep For 5 days is set in the Retention tab of <subclient 1> properties dialog box.
  3. On Day 1, perform a full backup of <subclient 1>.
  4. On Day 2, run an incremental backup.
  5. On Day 3, delete a file manually.
  6. On Day 4, perform a full backup. You will find that the deleted file has been dropped in this backup cycle.
  7. However, the deleted file will be available in the media for a period of 4 days and up to 2 backup cycles according to storage policy copy retention. After 4 days and 2 backup cycles, the data will be pruned from the media.

Example 2:

  1. A subclient named <subclient 1> is associated to a storage policy with a retention value set for 4 days and 2 cycles.
  2. A retention value of Keep For 5 days is set in the Retention tab of <subclient 1> properties dialog box.
  3. On Day 1, perform a full backup of <subclient 1> .
  4. On Day 2, perform an incremental backup.
  5. On Day 3, delete a file manually during an Incremental backup.
  6. On Day 4, run a synthetic full backup. Since the deleted file does not meet the retention criteria set at the subclient level, the file is carried forward.
  7. On Day 5, perform an incremental backup.
  8. On Day 8, perform a synthetic full backup. Since this job is run after five days of deletion (exceeds the retention criteria set at the subclient level), you will find that the deleted file will be dropped backup cycle.

Why do I see HSM Recaller Services running on my computer when I have not enabled archiving?

HSM Recaller Services are installed as part of the Windows File System iDataAgent installation. It works differently with the following agents.

Windows File System iDataAgent

Simpana OnePass Agent for Windows

Driverless Simpana OnePass Agent for Windows

You can backup and restore data by installing Windows File System iDataAgent on your computer.

HSM Recaller Services get installed as part of the installation process.

For more information, see

Getting Started - Windows File System - Install

You can archive and restore/recall data residing in your computer by installing Windows File System iDataAgent and Driver for File Recalls.

HSM Recaller Services aids the driver in triggering recall of the archived files.

For more information, see

Getting Started - Deployment - Simpana OnePass Agent for Windows

You can archive and restore/recall data residing in your computer by installing Windows File System iDataAgent and enabling the Explorer Plugin on your client computer.

When you enable the archiving option on your client, you will be able to archive your data.

HSM Recaller Services aides in triggering recall of these archived files.

For more information about the installation process, see

Getting Started - Deployment - Driverless Simpana OnePass Agent for Windows File System

For more information about enabling the Explorer Plugin, see

Getting Started - Explorer Plugin - Driverless Simpana OnePass Agent for Windows File System

Can I set the archive attribute for files?

No. You cannot set the archive attribute for files. During a file system backup, there may be a message logged in clbackup.log indicating you are unable to turn off the archive attribute after the backup operation. This happens because the user does not have the rights to write to this folder (e.g., C:\Program Files\).

How can I collect information about file access time and file modified time?

Enable Catalog additional file and system attributes check box in the Advanced Options tab of the Subclient Properties dialog box before performing a backup. This collects the following information about files, volumes, and directories within the subclient:

  • The name and location of each file.
  • The size of each file.
  • The time at which it was last modified.
  • The time at which the file was last accessed.
  • The time at which the file was created.
  • The name of the primary user of the file.
  • The name of the primary user group of the file.

For more information, see Configuring File Level Information for a Subclient.

Can I perform backup of subclients whose content is non-existent?

If all the contents included in your subclient are non-existent, then the backup of that subclient fails.

Why is IIS (Internet Information Server) not restored during a full system recovery?

When an operating system is re-installed , IIS is not installed again. After a full system restore, IIS would fail to start or open in Internet Services Manager or Computer Management because the old keys are not replaced by the new keys.

Uninstall and Re-install IIS to resolve this issue.

Can we filter SIS common store from backup?

Yes. You can add SIS (Single Instance Storage) common store to be filtered from the backup job. However, if SIS links are found in any of the files, then that file residing in the common store is backed up. SIS links do not fail during backup since they are automatically detected.

Does SnapProtect modify last access time on files?

Yes. If you use Change Journal scan method, then the access time of files are modified during the scan phase. If you are backing up an existing snapshot, then the last access time will not be modified. Also, if you switch to Recursive Scan method, last access time of files are not modified during the scan phase.

Does modified time of a file get backed up? If yes, are the files restored with the same time?

Yes. The last modified time of a file is backed up and the file is restored with the same time it had before the backup was performed. However, the last modified time of a parent directory may not be restored properly. This mainly depends on the order in which the files and directories are restored.

In Change Journal scan method, files and directories are backed up randomly based on the time in which the file was last modified.

In Recursive Scan method, files inside a directory are backed up first and then the directory is backed up. The files and directories are then restored in the same order in which the files were backed up.

Does a full system restore override all the contents on the server and restore the server to the state it was before the backup?

Yes. A full system restore would override all the contents on the server and restore the server to the state it was before the backup. Most of the registry keys are restored in the same state it was before the backup. Existing files are replaced only if the Unconditional overwrite option is selected from the Restore Options for All Selected Items dialog box. However, some system files may require a reboot.

Prior to performing a full system restore, make sure that the latest security, server image and service pack updates are installed.

Why am I unable to backup or restore subclients with UNC paths as content?

  1. Verify that the user account given has access to perform the backup or restore operation. Backing up or restoring to an admin share would require administrator rights.
  2. Identify the Job ID having issues in clbackup.log or clrestore.log. Use net helpmsg to identify the error code. For example, when you run this command, you may notice the following in the logs:

    584 27c 10/04 17:53:05 50640 CVArchive::ReadBuffer() - PL_FS_OPEN_AFILE(2, 4, 259821) 584 27c 10/04 17:58:59 50640 FsRestoreNTObject::handleError() - Write data error = 5 net helpmsg 5 Access is denied.

Why are my incremental backups same size as that of full backup?

Incremental backups may be the same size as that of a full backup due to any one of the following reasons:

  1. A third party software or antivirus scanner installed on your system is modifying the files every time they are backed up. In this process, these files are picked up in the subsequent backups increasing the size of incremental backups.
  2. Oracle Fusion Middleware (OFM) is configured on your computer with incorrect settings. This could prevent the directory change rename after the full backup. This could also result in synthetic full backup failure.
  3. Backups are performed from a Snap (hardware snap, QSnap, VSS snap) using recursive scan method with Check archive bit during backups option enabled. Selecting Check archive bit during backups picks up all the files from the snap for backup. FAT volumes and volumes with mount points that do not have an associated volume letter would also be picked up for backup.

    As a workaround, change the scan method to Change Journal or disable Check archive bit during backups option in the Subclient Properties dialog box.

Can we filter the Page File from backup?

Yes. You can filter the Page File from backup. For instructions, see Filtering Files from a Backup.

Can we back up only those files that have been recently modified or created?

Yes. You can configure backups only for the data that has been modified or created after a specified date. This option is useful if you need to filter out data that was modified before a certain date.

For more details, refer to Configuring Backups for Recently Modified or Created Data.

How will I view the files that were skipped during backup?

Certain files found during scan phase might be missing during backup. By default, the backup job skips these missed files during backup so that these files are not counted as failed objects. However, if you do not want these missing files to be skipped during backup, enable nDoNotSkipMissingFilesOnBackup additional setting. Additionally, these missing files will be counted as failed objects.

WFS0020: Can I restore files with special characters in their names?

No. You cannot directly restore a file or a folder that has special characters like square brackets([ ]) in their names. You can restore the parent directory in which these files or folders reside.

WFS0021: Why do encrypted files take longer time to backup?

Encrypted files take longer time to backup due to the processes used to back them up. Encrypted data has to communicate with the software that was used to encrypt the files. This communication happens throughout the backup process. Hence, the backup software has to provide a call back function in a different thread for these encrypted files to be backed up thereby slowing down the backup process.

WFS0022: Why am I prompted to reboot the computer after performing a restore operation?

After a file system or a System State restore, you may be prompted to reboot the computer to which the data was restored. This is because the data that was attempted to be restored was in use or locked by another process.

In such cases, the following is observed:

  1. The file is restored into the same directory. For example, filename.txt would be restored as ~filename.txt~ in the same directory.
  2. An entry for this file is created in the PendingFileRenameOperations key in the registry.

At the end of the restore operation, the restore process will report to the CommServe that the system must be rebooted which in turn would display message in the CommCell Console to reboot the computer.

When the server is rebooted, Microsoft Windows operating system will overwrite the locked files with the temporary files created in the PendingFileRenameOperations registry key.

You may be prompted for a reboot while restoring System State components and the restore operation is not considered complete until the server is rebooted.

For more information  on PendingFileRenameOperations, see Microsoft KB 181345.

For more information on a third party tool made by Microsoft that will show the pending files to be renamed, see Microsoft Windows Sysinternals.

WFS0023: What are the critical and non-critical system state components on Windows operating system?

Critical Components

Critical components are components that are essential and necessary for the operating system to come back alive after the restore of system state.

If a critical component fails during the backup of system state, the backup job fails with an error code as the backed up data is not sufficient for a restore of the operating system.

  • Cluster Database
  • Registry
  • System Info Archive
  • System Protected Files
  • System Volume

    You may filter critical system state components from a backup operation. However, you cannot restore the server to the state in which it was at the time of backup thereby leading to a incomplete restore of the server and system state.

Non-Critical Components

Non-Critical components are components that are part of system state but are not essential for the operating system to come back alive after the restore of system state.

If a non-critical component fails during the backup of system state, the backup job completes with an error code as the backed up data is sufficient for a restore of the operating system. However, the non-critical components do not get restored.

  • Active directory application Mode (ADAM)
  • BITS service
  • Certificate services
  • COM+ database
  • Content indexing catalogs
  • DDR explicit content
  • DFS replication services
  • DHCP
  • Disk quotas
  • Event logs
  • File Server Resource Manager (FSRM)
  • IIS
  • Metadata
  • NPS database
  • Offline files
  • One touch files
  • Remote storage database
  • RSM
  • Terminal service licensing
  • UDDI database
  • WMI database
  • WINS

For more information on system state, see Microsoft Articles on Backing Up and Restoring System State, Microsoft Backing Up and Restoring System State in Windows Server 2003 R2 and Windows Server 2003 SP1, and Backing Up and Restoring the Failover Cluster Configuration Using VSS.

Scan Methods

How does the Optimized Scan enhance the scanning capabilities of an iDataAgent?

When you enable the Optimized Scan for a File System, a meta database is created. The meta database contains information about the files in the associated volume. This Optimized service constantly updates the databases to track file additions, volume additions, content update to files, etc. Journals from the File System Filter (FSF) driver are used to keep track of the updates to each meta database.

What is the default location of the Metadata database?

On each volume, the metadata database is created at the following location:

System Volume Information\NetApp\Data Classification

It is named as dc_<instance_number>.db

For example: On the C drive, it would be C:\System Volume Information\NetApp\Data Classification\dc_<instance_number>.db

You cannot relocate the database.

How to turn off the Optimized Scan?

Use the nStartDataClassifier additional setting to enable or disable the Optimized Scan services on a client computer as follows:
  1. From the CommCell Browser, navigate to Client Computers.
  2. Right-click the <Client> in which you want to add the additional setting, and then click Properties.
  3. In the Client Computer Properties dialog box, click Advanced.
  4. In the Advanced Client Properties dialog box, click the Additional Settings tab.
  5. Click Add.
  6. In the Add Additional Settings on Windows Client dialog box, the Name box, type nStartDataClassifier.

    The Category and Type details are automatically populated.

  7. In the Value box, type 1 to enable or 0 to disable the data classification monitoring for all the volumes.
  8. Click OK.

If you want to disable Optimized Scan for selected volumes, use the VolumeFilters additional setting to provide the list of volumes as follows:

  1. From the CommCell Browser, navigate to Client Computers.
  2. Right-click the <Client> in which you want to add the additional setting, and then click Properties.
  3. In the Client Computer Properties dialog box, click Advanced.
  4. In the Advanced Client Properties dialog box, click the Additional Settings tab.
  5. Click Add.
  1. In the Add Additional Settings on Windows Client dialog box, the Name box, type VolumeFilters.

    The Category and Type details are automatically populated.

  2. In the Value box, type Drive Letter for the volume that you want to filter.
  3. Click OK.

How much hard disk space is consumed by the Metadata database?

The meta databases created by Optimized scan method usually consume about 5% of the total space on the hard disk. Depending on the type of data and folder layout, the metafiles may consume additional space.

Does the backup include all the source files associated with a hard link?

If a subclient contains a hard link and the selected scan method is Change Journal, then only the first source file is included in the backup. However, if you select Optimized or Recursive Scan, all the source files associated with a hard link are included in the backup.

Can I change the scan method for a subclient?

Yes. You can change the scan method for a subclient. Before changing the scan method for a subclient, consider the following scenario:

  • If the scan method is optimized for FAT Volumes, UNC Paths, mount points to free spaces and you change to any other scan method, the next incremental backup for the subclient will become a full backup.
  • If the scan method is optimized for Whole Volume or Whole Volume (using Optimized Scan) and you change to the scan method optimized for FAT Volumes, UNC Paths, mount points to free spaces, the backup cycle will continue as it is.

VSS

How much disk space is required on a volume for the VSS shadow copy?

During a VSS backup, the free space needed on a volume for the shadow copy is related to the duration of the backup and the amount of changes being made to the data objects (files, databases) that are being backed up. VSS snapshots capture and copy disk blocks that are being changed by other processes while the backup for a particular object is running. If the backup runs longer, or there is a lot of disk activity (files or databases being changed). This causes VSS to copy more data blocks and therefore, use more disk space.

Why does the differential or incremental backup becomes a full backup when VSS is enabled?

If you Use Scan Method Optimized for FAT Volumes, UNC Path, mount point to free spaces and the select the Check archive bit during backups option, then all backups by VSS on subclients will be converted to full backups. If you want to perform incremental or differential backup, you must use the Use Scan Method Optimized for Whole Volumes. If you want to Use Scan Method Optimized for FAT Volumes, UNC Path, mount point to free spaces scan, do not select the Check archive bit during backups option.

Can I use VSS to backup files on a virtual server?

Use of VSS on virtual servers controls the backup method for file system data only, not system state components; system state data is backed up by the physical node(s), not by the virtual server(s). Note that the physical cluster node will automatically filter all shared disk resources, and the virtual server will automatically filter all physical disk resources from backup operations.

When I am performing backup of a data, what happens to the system state component?

When you want to backup only data, ensure that the Backup System State check box is not selected on the Content tab of the Subclient Properties dialog box. Now when you perform the backup for any such subclient, the VSS will filter any system state component from the backup data. VSS will be used for this operation even when Use VSS option is not selected on the General tab of the of the Subclient Properties dialog box.

Lync Server

Can I backup Response File Group Configuration on a Lync server?

No. Response File Group Configuration does not get backed up when you perform the Lync Server backup. For more information about backing up Response File Group Configuration, refer to http://technet.microsoft.com/en-us/library/hh202170.aspx.

Can I use Restore by Job option to restore the Lync server data?

No. The Restore by Job option is not supported for the Lync Server backup jobs. You must perform the browse and restore operation to restore the Lync server data.